*4.2. Lab-Scaled Composites Results*

The results of mechanical tests and MFR are summarized in Table 3. From the point of view of tensile tests, the powders' addition makes the material more brittle with a decrement of both stress and elongation at break. The HSPs addition, on the other hand, significantly increases the elastic modulus compared to pure PLA. This is a common trend [43] and it is due to the introduction of fillers having higher elastic modulus than pure matrix. In general, a decrement of the mechanical properties increasing the HSPs amount can be observed; however, for H0210, the tensile decrement is less marked than HM200 and the impact resistance is not worsened with respect to pure PLA.

**Table 3.** Mechanical and MFR results of lab-scaled composites with different amounts of H0210 and HM200 HSP.


The better mechanical response achieved with H0210 can be attributed to several factors. First of all, H0210 possesses a greater diameter distribution that represents a more efficient obstacle towards the crack that advances during the Charpy test, compared to

HM200 with a finer diameter distribution [36]. Furthermore, residual moisture content must also be considered because it also affects the final mechanical response. H0210, under the same drying conditions, lost a greater amount of moisture that could potentially degrade the PLA (it must be considered that in this first lab-scale step, no venting for the humidity stripping is present in the mini-extruder). Finally, the filler/matrix adhesion must also be considered. It is well known from the literature that natural fibers have poor adhesion with PLA [44]. However, comparing HM200 and H0210, the fillers with higher grain sizes will have greater adhesion (the stress able to cause the fiber detachment is in fact a function of various parameters, including the aspect ratio) [45]. The different adhesion is also confirmed by the B parameter obtained as the slope of the Pukanszky's plot (Figure 6). A decrement of the B value (from 1.41 for H0210 to 0.56 for HM200) can be observed, indicating a worsening of the matrix/filler adhesion.

**Figure 6.** Pukanszky's plot for PLA-HSPs composites.

Observing in addition the experimental values of composites tensile strength (Figure 7), a different interaction between H0210 and HM200 with the PLA matrix can be observed. The experimental data in fact, lies between the upper and lower bound; however, HM200 are much closer to the Nicolais and Nicodemo lower-bound equation indicating a weaker adhesion respect to H0210 that are closer to the upper bound. The particles with smaller size have a great tendency to agglomerate, causing greater weakening of the matrix.

**Figure 7.** Comparison between the experimental composite strength and the values predicted according to the upper and lower bound equations.

The SEM images reported in Figure 8 confirm the prediction of the analytical models and of the mechanical results obtained. A better adhesion is registered for H0210, respect to HM200. In particular, for H0210, it can be observed that at 20 wt.% (Figure 8a), the

fillers are fairly well distributed, and few agglomerations can be observed with 30 wt.% of HSP (Figure 8b). At 40 wt.% however, a greater agglomeration tendency, due to the greater HSPs amount introduced, is registered. The agglomerates are also less adherent to the PLA matrix and in Figure 8c, holes due to the detachment of these agglomerates are clearly visible; the presence of agglomerates is also responsible for the marked drop down of the mechanical properties recorded for the PLA\_40\_H0210 composite. HM200 show worse adhesion and already at 20 wt.%, voids can be observed due to HSPs' detachment (Figure 8d). However, for all the compositions (Figure 8d–f), a HSPs detachment can be recorded, which is very marked compared to H0210, confirming the results of mechanical tests and analytical models adopted.

**Figure 8.** SEM micrographs of the fractured Charpy surface of: (**a**) PLA\_20\_H0210, (**b**) PLA\_30\_H0210, (**c**) PLA\_40\_H0210, (**d**) PLA\_20\_HM200, (**e**) PLA\_30\_HM200, (**f**) PLA\_40\_HM200.

Regarding the MFR values, it can be observed that the viscosity increased (so the fluidity decreased) on adding the HSPs (Figure 9). However, this occurred only for H0210. For HM200 the MFR values are higher if compared to those obtained with H0210, but no trend with the HSPs amount was detected. These results are in agreement with those reported by Song et al. [28] and with the injection pressure, reported in Table 2, where the injection pressure was increased with the H0210 content while it was decreased by increasing the HM200 content. These MFR trends can be attributable to the probable partial PLA hydrolysis caused by the filler moisture that is greater for HM200 due to its larger surface area and thus humidity content. However, as reported in literature, the general decrement of stress at break is mainly correlated to the poor interfacial adhesion of lignocellulosic fiber with the biopolymeric matrix [45,46].

**Figure 9.** Trend of Melt Flow Rate (MFR) as a function of HSP content.

From a thermal point of view, a decrement of both melting temperature and glass transition temperature caused by the addition of HSPs can be observed (from Table 4); this decrement seems to be correlated to the HSP content. However, the HSP typologies also affect the melting temperature and glass transition temperature differently with a decrement that is more marked with HM200. This behavior can be correlated to the different granulometry between H0210 and HM200. HM200 have a higher surface area than H0210 and it adsorbs more moisture that can lead to decrease in the average molecular weight (resulting in a decrement of the glass transition and melting temperatures). The HSPs' addition increases the crystallinity of the PLA, causing a shift of the cold crystallization temperature towards lower temperatures. HSP seems to act as a nucleating agent, providing heterogeneous nucleation sites similar to other systems filled with natural fibers [27,47,48]. In particular, as the HM200 are finer and more homogeneous with a tighter diameter distribution curve, they are more effective in crystallizing the PLA, when compared to their H0210 counterparts.
