*3.2. Effect of the EO on Mechanichal and Failure Behaviour of PLA/PBSA Blends*

EO was added (at 2 wt.%) to 80–20 and 60–40 blends. These two formulations were selected because they showed the highest elongation at break and Charpy impact resistance values.

From a morphological point of view, it can be observed in Figure 3, that EO acted as a compatibilizer, working as a bridge between the PLA and PBSA phases, leading to a reduction of the interfacial tension and thus leading to better adhesion and dispersion between the two phases [55,92,93].

The interaction created by EO in the PLA–PBSA blends can also be observed in the torque values that for (80–20)J and (60–40)J were higher than the torque value of pure PLA (Figure 1). The MFR, on the other hand, reached the lowest values when EO was added. It is known that the torque increment is related to the molten polymer viscosity increase, which is caused by a molecular weight increment in polyesters [17]. Furthermore, the chain extender produces an increment of the melt viscosity creating interactions between the matrix and the dispersed phase. This marked viscosity variation with respect to the viscosity values recorded for corresponding blends without EO (80–20 and 60–40), led to a significant variation of the injection molding conditions (as can be observed in Table 2) where the mold temperature and the injection pressure were increased.

From a thermal point of view, it can be observed (Figure 5) that the EO addition caused a slight decrement of *Tg* that could be ascribable to the increased compatibility between the two polymers [17]. A slight shift of the PLA cold crystallization temperature was also registered. On the other hand, the introduction of EO limited the chain mobility and the shift of PLA cold crystallization temperature did not occur in the presence of the chain extender. As could be expected, the chain extender depressed severely the PLA crystallization while the PBSA crystallization did not seem to be much influenced by the EO addition (Figure 7).

However, from the results of HDT (Figure 6), it could be detected that the EO addition did not cause a significant variation of the HDT values and it did not significantly affect the final crystallinity of the material. In any case the increase in molecular weight and branching activity of EO on PLA counterbalanced the decrease in crystallinity in this phase, thanks to the increased resistance of the amorphous phase achieved thanks to the formation of new inter-macromolecular bonds.

The enhanced compatibility induced by EO was overwhelmingly reflected in the mechanical properties (Figure 8). EO caused a relevant increment of the mechanical properties. Not only an improvement in elongation at break and Charpy impact resistance occurred, but also a marked increment in stress at break was recorded.

**Figure 8.** Comparison of the main mechanical properties for blends with and without Joncryl.

Essentially, the increase in stress at break and elongation at break may be related to the molecular weight increase, due to the polyester reaction with EO (Figure 2). On the other hand, the toughness increment is more related to the morphology of the system and, as can be expected, better toughness is reached for the (60–40)J blend where a co-continuous structure combined to a better phase compatibility was observed.

To better understand the effective toughness enhancement obtained with the EO addition, the *JIlim* was calculated adopting the elasto-plastic fracture mechanism approach. The *JIlim* values correspond to the energy absorbed by the specimen at the moment of the crack propagation in slow-rate test conditions.

Comparing the *JIlim* values with the "brittle" *G* value of pure PLA (equal to 2.97 kJ/m<sup>2</sup> [94,95]) all blends showed very interesting values (Figure 9). The fracture energy released at the beginning of crack propagation was markedly increased with the EO addition for the (80–20)J blend, while for the 60–40 composition the *JIlim* value remained almost unchanged with the EO addition. Probably, the 60–40 blend being co-continuous, the effect of EO addition was less marked; the co-continuous nature of the blend led it to have a starting high value of *JIlim* and Charpy impact resistance that were not significantly altered by the presence of EO.

**Figure 9.** Comparison between JIlim and Charpy impact resistance values (both in KJ/m2) for blends with and without Joncryl.

The results of dilatometric tensile tests are reported in Figure 10 where the volume variation (calculated by Equation (1)) is reported as function of axial elongation (the tests were carried out until the deformation of the specimens remained homogeneous).

**Figure 10.** Volume strain–strain curves for blends with and without EO.

The characterization of the dilatational response of a material when it is subjected to an applied stress can lead to an appreciated deformation mechanism in the bulk of the material. According to the type of volume strain–strain curve obtained, it is possible to distinguish between: a cavitational response, a dilational response and deviatoric response [96]. It can be observed that except for the 80–20 blend, where a cavitational response is observed, for the other blends a deviatoric response was registered.

As a consequence, for the 80–20 blend, in the large deformation limit, the hydrostatic tensile stress will cause cavitation type mechanisms that will lead to a rapid volume increase. In particular, when the stress approaches the yield stress value, the cavitationtype mechanism will produce voiding that will cause the rapid increase in the volume strain, and the coalescence of these voids will bring to the final rupture of the specimen [60]. The SEM micrographs at the cryo-fractured surface of tensile specimens along the draw direction (Figure 11) confirmed the presence of many small and big elongated voids grown along the tensile direction due to cavitation for the 80–20 blend.

The volume dilatation response for the other blends ((80–20)J, 60–40 and (60–40)J)), displayed a deviatoric behavior; for these materials, when the stress value approaches stress at yield, the material will continue to deform by changing shape and not volume, leading to a shear yielding as the main failure mechanism. The SEM images confirmed that a different deformation mechanism occurred where the material underwent deformation along the draw direction without voiding formation.

Similar to what was observed for the *Jlim* value, the EO addition in the 80–20 mixture, significantly changed the micromechanical behavior of the material. This marked change can probably be ascribed to the capacity of the chain extender to modify the macromolecule structure and create interactions between the matrix and the dispersed phase, modifying the initial interface adhesion and leading to a different micromechanical response.

On the other hand, the effect of EO to the micromechanical behavior of the 60–40 blend could be considered negligible. The dilatometric curves of the 60–40 and (60–40)J blends were very similar with a very slight increment of the volume variation for the (60–40)J blend. In this case the micromechanical behavior was probably ascribable to the different morphology of the 60–40 blend that, being co-continuous, did not show dispersed PBSA particles into the PLA matrix.

**Figure 11.** SEM micrographs made at the surface of tensile specimen cryo-fractured along the draw direction: (**a**) 80–20; (**b**) (80/20)J; (**c**) 60–40; (**d**) (60/40)J.

#### **4. Conclusions**

In this study, PLA/PBSA binary blends with different PBSA contents (from 5 up to 40 wt.%) were investigated from a rheological, thermal and mechanical point of view. To the blends (80–20 and 60–40), that presented a higher value of impact resistance and elongation at break, 2 wt.% of an epoxy oligomer (EO) as compatibilizer was added. The two blends also presented different morphologies: the 80–20 blend possesses a morphology where the PBSA particles are dispersed within the PLA matrix, while the 60–40 blend shows a co-continuous structure.

From the thermal properties point of view, no significant variations were observed caused by the EO addition, whereas, as could be expected, the interaction created by EO produced an increment of the melt viscosity.

However, the most interesting results were observed from the micro and macro mechanical analysis. The mechanical results showed that the Charpy impact resistance value (high speed test) and the *Jlim* (energy absorbed from the specimen at the moment of the crack propagation in slow-rate test conditions) are higher for the 60–40 blends having a co-continuous morphology. However, for this 60/40 blend the EO addition does not significantly alter the mechanical response. In fact, dilatometric tests show that for the 60/40 blend the micromechanical response is of the deviatoric type and the morphology, investigated at the surface of cryo-fractured tensile specimen along the draw direction, was comparable.

On the other hand, the 80/20 blend was significantly affected by the presence of EO. The micromechanical response passed from a cavitational response to a deviatoric one, probably induced by the marked change in the dispersed phase morphology with the EO addition and the higher complexity of macromolecule structure generated by the reactive processing.

The systematic study described in this paper was thus useful for a better knowledge of thermo-mechanical behavior of PLA blends containing up to 40% of PBSA.

Moreover, the EO used in the present paper resulted in being quite useful in effectively modulating melt viscosity and this is necessary not only for adapting the material to several processing methodologies, but also as compatibility enhancer, impacting positively the mechanical performance. However, this additive is not biobased and biodegradable. Hence in the future it could be important to replace it with biobased and biodegradable ones, to grant a full circularity of the material by keeping into account both its origin and final disposal.

**Author Contributions:** Conceptualization, M.-B.C. and L.A.; experimental work, A.V., I.C. and L.A.; theoretical background, L.A. and A.L.; data curation and elaboration, A.V., I.C., L.A. and M.-B.C.; writing—original draft preparation, L.A. and A.V.; writing—review and editing, M.-B.C.; supervision, P.C. and A.L.; project administration and funding acquisition, M.-B.C. and P.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Bio-Based Industries Joint Undertaking under the European Union Horizon 2020 research program (BBI-H2020), BIONTOP project, grant number G.A 837761.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Acknowledgments:** Belen Monje Martinez of AIMPLAS, coordinator of BIONTOP project, is thanked for helpful suggestions regarding the present manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.
