*3.3. Mechanical Properties of the PLA/PBAT/GR Formulations*

Table 3 shows the main values of the mechanical properties, maximum tensile and flexural strength, Young's moduli, elongation at break, impact absorption energy (Charpy), Shore D hardness, and the HDT temperature of each obtained formulation, as well as the neat PLA as a reference.

It was observed that the tensile strength of the PLA/PBAT formulations significantly decreased (*p* < 0.05) by 6.3%, 17%, 23.2%, and 29.1% (compared to PLA/PBAT) when adding 5, 10, 15, and 20 phr of GR, respectively. However, the toughness (calculated as the area under the stress-strain curve) of the formulation with a 5 phr of GR (PLA/PBAT\_5GR) significantly increased (*p* < 0.05) compared to the base formulation PLA/PBAT by 40%, as shown in Figure 3. In Figure 3, it is observed that as the content of GR increased, the toughness (energy per unit volume) of the formulations significantly decreased (*p* < 0.05). In contrast, the drastic increase in the toughness of PBAT by adding 10 phr of GR should be noted. This fact explains that GR acts as a PBAT plasticizer since it increased its elongation and decreased its tensile strength.

On the other hand, it was observed that when adding 5 and 10 phr of GR, Young's modulus (Table 3) did not significantly decrease (*p* > 0.05) compared to PLA/PBAT. However, from contents higher than 10 phr of GR, where GR saturates notoriously in the PBAT domains (as shown by FESEM in Figure 2), the saturations behaved as reinforcements, increasing Young's modulus mean values. Specifically, Young's modulus of the formulations with a 15 phr GR showed a reduction of less than 10% relative to the reference formulation (PLA/PBAT). For contents of 20 phr GR, it showed an increase of the value, being 6.6% higher than the reference. The tendency of the mean values suggests that at low contents, GR acts as a plasticizer of the PBAT component, making the PBAT domains less compatible with the PLA matrix (because PLA is incompatible with GR). At contents higher than the saturation point (at 10 phr of GR), some nano-scale domains were generated. These domains acted as reinforcement, increasing the mean value of the Young's modulus of the materials. It should be noted that GR resin has higher affinity and therefore higher compatibility with PBAT, confirming its ability to plasticize it.


**Property**

PLA/PBAT \_10GR 41.9 ± 0.4 c

PLA/PBAT \_15GR 38.8 ± 2.8 c

PLA/PBAT \_20GR 35.8 ± 4.5 c

**Tensile Strength (MPa)** 

PBAT 13.6 ± 1.4 d 110 ± 40 c

PBAT\_10GR 14.9 ± 1.7 d 80 ± 10 c

differences between formulations (*p* < 0.05).

**Young's Modulus (MPa)** 

**Formulation** 

**Table 3.** Mechanical properties of PLA/PBAT blends with different contents of gum rosin (GR) as additive.

**Flexural Strength (MPa)** 

PLA 65.1 ± 1.7 a 2100 ± 250 a 6.4 ± 1.6 a 108.8 ± 8.8 a 3170 ± 150 a,b \* 34.6 ± 2.8 a 77 ± 1 a 58.0 ± 0.8 a PLA/PBAT 50.5 ± 0.5 b 1680 ± 200 b 16.4 ± 1.2 b 74.9 ± 8.6 b 2720 ± 130 a 5.1 ± 1.4 b 71 ± 1 b 57.8 ± 0.6 a

1510 ± 90 b 3.8 ± 0.4 c,d 29.7 ± 1.1 d 3400 ± 190 b 10.3 ± 1.3 c

487 ± 70 e 6.8 ± 0.5 e 80 ± 10 c

\* PLA sample tested in specimens without notch. a–f Different letters within the same property show statistically significant

**Flexural Modulus (MPa)** 

2530 ± 180 a 9.3 ± 0.7 c

1790 ± 220 a,b 1.7 ±0.4 d 28.3 ± 3.3 d 3020 ± 180 a,b 6.9 ± 0.3 b,c 75 ± 1 d 53.8 ± 0.8 c

It was observed that the tensile strength of the PLA/PBAT formulations significantly decreased (*p* < 0.05) by 6.3%, 17%, 23.2%, and 29.1% (compared to PLA/PBAT) when adding 5, 10, 15, and 20 phr of GR, respectively. However, the toughness (calculated as the area under the stress-strain curve) of the formulation with a 5 phr of GR (PLA/PBAT\_5GR) significantly increased (*p* < 0.05) compared to the base formulation PLA/PBAT by 40%, as shown in Figure 3. In Figure 3, it is observed that as the content of GR increased, the toughness (energy per unit volume) of the formulations significantly decreased (*p* < 0.05). In contrast, the drastic increase in the toughness of PBAT by adding 10 phr of GR should be noted. This fact explains that GR acts as a PBAT plasticizer since it increased its elon-

7.2 ± 0.8 e 60 ± 10 c

**Charpy Impact Energy (KJ/m2)** 

**Hardness (Shore D)** 

No break 41 ± 1 e 36.8 ± 0.4 d

No break 38 ± 1 f

**HDT Temperature (°C)** 

56.6 ± 0.6 a,b

35.6 ± 0.2 d

71 ± 1 b 55.2 ± 0.4 b,c

74 ± 1 d 54.8 ± 0.8 b,c

**Elongation at Break (%)** 

PLA/PBAT \_5GR 47.3 ± 1.2 b 1440 ± 200 b 7.3 ± 1.4 a 67.2 ± 0.8 b 2510 ± 30 a 8.3 ± 1.2 b,c 72 ± 1 c

1430 ± 100 b 5.2 ± 0.8 a,c 48.0 ± 7.8 c

720 ± 15 f

gation and decreased its tensile strength.

**Figure 3.** The toughness of PLA, PBAT, PBAT\_10GR, PLA/PBAT, and PLA/PBAT with 5, 10, 15, and 20 phr of GR resin. a–h Different letters show statistically significant differences between formulations (*p* < 0.05). **Figure 3.** The toughness of PLA, PBAT, PBAT\_10GR, PLA/PBAT, and PLA/PBAT with 5, 10, 15, and 20 phr of GR resin. a–h Different letters show statistically significant differences between formulations (*p* < 0.05).

On the other hand, it was observed that when adding 5 and 10 phr of GR, Young's modulus (Table 3) did not significantly decrease (*p* > 0.05) compared to PLA/PBAT. However, from contents higher than 10 phr of GR, where GR saturates notoriously in the PBAT Table 3 also shows the mechanical properties of PBAT and PBAT\_10GR, the behavior of which can be considered as an indicator of the compatibility between the PBAT and the GR. The significant reduction (*p* < 0.05) in Shore D hardness, which went from 41 to 38 and the reduction in Young's modulus, both in the flexion and in the tensile test, of the PBAT\_10GR formulation compared to neat PBAT, show that the GR exerted a plasticizing effect on the PBAT matrix. Moreover, a significant increase (*p* < 0.05) of 47.6% in elongation was observed, which reinforces the idea of the plasticizing effect. However, the elongation at break of the PLA/PBAT formulations with different GR contents suffered a significant reduction (*p* < 0.05) of 55.5%, 68.3%, 76.8%, and 88.0%, compared to the PLA/PBAT formulation, when adding 5, 10, 15, and 20 phr of GR, respectively. It is well known that the miscibility of PBAT and PLA is partial with some compatibility [22,24,50]. Nevertheless, a phenomenon of coalescence of the PBAT domains was generated with the addition of GR to the PLA/PBAT formulation, creating larger domains and reducing the interaction between PLA and PBAT. This behavior can be explained since the PLA does not assimilate the GR in its matrix, which remains isolated. This phenomenon makes the GR assimilation by the PBAT domains easier, and therefore, the size of these domains increased with the GR content.

In contrast, a significant increase (*p* < 0.05) in the impact energy absorption was observed (Table 3) when incorporating 10 and 15 phr GR into the PLA/PBAT formulation, managing to increase it by 75.5 and 79.2%, respectively. For contents higher than 15 phr GR, the impact energy absorption mean values began to decrease due to saturation of the GR on the PBAT domains, generating a phase separation between the PLA/PBAT matrix and the GR saturations. This effect was corroborated by FESEM analysis (Figure 2f). According to the literature, when the domains of the ductile and dispersed material in a rigid polymeric matrix have a size of 2–5 µm, the energy absorption is maximum. This phenomenon has been demonstrated in a large number of materials, for example in the synthesis of HIPS [51], where polybutadiene forms domains into the PS matrix. At smaller or larger domain sizes, the energy absorption value decreases again. Therefore, the stress concentration generated in materials with poor interaction will depend on the size of the domains of the minority

component, in this case, the PBAT-GR system. In the present study, the toughness of materials, especially at impact, could be controlled and improved thanks to the phobicity between PLA and GR and the affinity between PBAT and GR. If the GR resin showed a good affinity with both polymers (PLA and PBAT), the toughness modification would depend exclusively on the composition of the polymeric components and their interaction.

Finally, the hardness (Shore D) of the studied blends significantly varied (*p* < 0.05) with the incorporation of GR. This property significantly increased for 15 and 20 phr of GR contents. The plasticizing effect of the GR resin on the neat PBAT significantly reduced the hardness (Table 3). The HDT significantly decreased (*p* < 0.05) with increasing GR content, going from 57.8 ◦C for PLA/PBAT to 53.8 ◦C for the formulation with a 20 phr of GR. Therefore, the processability of these materials improved as the GR content increased. *Polymers* **2021**, *13*, 1913 12 of 19

#### *3.4. Thermal and Thermomechanical Properties of the PLA/PBAT/GR Formulations 3.4. Thermal and Thermomechanical Properties of the PLA/PBAT/GR Formulations*

The calorimetric curves of the PLA, PBAT, PBAT\_10GR, PLA/PBAT, and PLA/PBAT formulations containing 5, 10, 15, and 20 phr of GR formulation were obtained by differential scanning calorimetry (DSC analysis) and are reported in Figure 4. In addition, Table 4 shows the main thermal transitions such as the glass transition temperature (Tg), cold crystallization temperature (Tcc), the meting (∆Hm) and crystallization (∆Hcc) enthalpies, and the degree of crystallinity (Xc). The calorimetric curves of the PLA, PBAT, PBAT\_10GR, PLA/PBAT, and PLA/PBAT formulations containing 5, 10, 15, and 20 phr of GR formulation were obtained by differential scanning calorimetry (DSC analysis) and are reported in Figure 4. In addition, Table 4 shows the main thermal transitions such as the glass transition temperature (Tg), cold crystallization temperature (Tcc), the meting (∆Hm) and crystallization (∆Hcc) enthalpies, and the degree of crystallinity (Xc).

**Figure 4.** DSC curves of PLA, PBAT, PBAT\_10GR, and the studied PLA/PBAT formulations with different content of GR. **Figure 4.** DSC curves of PLA, PBAT, PBAT\_10GR, and the studied PLA/PBAT formulations with different content of GR.

**Table 4.** Thermal properties of studied formulations and neat matrixes materials. **Formulation TgPBAT \* (°C) TgPLA (°C) TccPLA (°C) ∆HccPLA (J/g) TmPLA (°C) ∆HmPLA (J/g) Xc PLA (%)**  PLA - 63.2 ± 1.2 a 102.5 ± 0.8 a 26.0 ± 1.5 a 171.7 ± 0.9 a −32.8 ± 1.3 a 7.4 ± 0.9 a PLA/PBAT −33.5 ± 1.1 a 62.3 ± 1.6 a,b 100.8 ± 0.8 a 23.9 ± 1.3 a 170.3 ± 1.1 a −30.5 ± 1.6 a,b 8.9 ± 0.3 a,b PLA/PBAT\_5GR −20.6 ± 0.5 b 61.8 ± 1.3 a,b 101.8 ± 1.5 a 21.6 ± 1.5 b 169.3 ± 1.3 a,b −28.8 ± 1.9 b 9.5 ± 0.8 a,b PLA/PBAT\_10GR −21.3 ± 1.1 b,c 60.9 ± 0.9 a,b,c 107.4 ± 0.6 b 23.9 ± 1.0 a,b 167.5 ± 1.2 b,c −31.7 ± 1.1 a 11.6 ± 1.1 b PLA/PBAT\_15GR −23.1 ± 0.9 c,d 59.2 ± 0.5 b,c 106.5 ± 1.3 b 23.3 ± 0.6 a,b 165.9 ± 0.8 c,d −29.9 ± 0.9 a,b 10.4 ± 0.7 a,b The T<sup>g</sup> related to the PLA component of the blend tended to decrease both when adding PBAT (1 ◦C lower for the PLA/PBAT formulation, due to the partial miscibility between PLA and PBAT) and when adding GR resin. Neat PLA had a T<sup>g</sup> value of 63.2 ◦C, while this value significantly dropped (*p* < 0.05) to 57.9 ◦C for the T<sup>g</sup> of the PLA/PBAT formulation with a 20 phr of GR. This decrease in the T<sup>g</sup> of PLA is due to the saturation of GR, which acts as a lubricant, facilitating the movement of the chains. However, the PBAT\_10GR had a significantly higher T<sup>g</sup> than the T<sup>g</sup> of the neat PBAT, going from −25.9 ◦C to −20.8 ◦C. This increment in the T<sup>g</sup> when GR was additivated to the PBAT could be due to an increase in PBAT crystallinity due to the presence of the GR. It is important to

(°C)

The Tg related to the PLA component of the blend tended to decrease both when adding PBAT (1 °C lower for the PLA/PBAT formulation, due to the partial miscibility between PLA and PBAT) and when adding GR resin. Neat PLA had a Tg value of 63.2 °C, while this value significantly dropped (*p* < 0.05) to 57.9 °C for the Tg of the PLA/PBAT

∆HmPBAT (J/g)

\* Tg PBAT determined by DMA analysis explained below. a–f Different letters show statistically significant differences

PBAT\_10GR −20.8 ± 0.7 b − − − 79.7 ± 1.2 f −19.4 ± 1.2 c

TmPBAT

between formulations (*p* < 0.05).

PLA/PBAT\_20GR −24.2 ± 1.1 d 57.9 ± 1.0 c 106.0 ± 1.0 b 25.1 ± 1.2 a 163.9 ± 0.9 d −30.0 ± 1.0 a,b 8.2 ± 1.0 a,b

mention that the T<sup>g</sup> values of PBAT and PBAT\_10GR formulations were obtained by DMA analysis since this transition is not easily observable by DSC.


**Table 4.** Thermal properties of studied formulations and neat matrixes materials.

\* T<sup>g</sup> PBAT determined by DMA analysis explained below. a–f Different letters show statistically significant differences between formulations (*p* < 0.05).

> The same effect was observed with the T<sup>m</sup> of PLA and PBAT when adding GR, 171.7 ◦C and 110.8 ◦C being the temperatures for neat PLA and neat PBAT, respectively. A significant reduction to 163.9 ◦ C for the PLA/PBAT formulation with a 20 phr of GR and up to 79.7 ◦C for the PBAT\_10GR formulation was achieved. This decrease confirmed the plasticizing effect of the GR resin on the PBAT and the lubricating effect exerted by the saturated GR on the PLA matrix.

> Table 4 also shows the degree of crystallinity (Xc) of the PLA fraction in the blends when adding PBAT and GR. A slight increase (not statistically significant, *p >* 0.05) in X<sup>c</sup> was observed when adding PBAT, since the microdomains of PBAT act as a nucleating agent due to the interaction between PLA and PBAT. When adding 10 phr of GR, a statistically (*p* < 0.05) higher increase was observed, reaching X<sup>c</sup> values of 11.6%. However, as the GR content increased above 10 phr, X<sup>c</sup> decreased (reaching similar values of neat PLA) since the PBAT domains coalesced thanks to the GR and fewer nucleation points were generated from the PLA spherulites.

> By DMA technique, the T<sup>g</sup> and the storage modulus "G" from torsional tests were obtained and the results are reported in Figure 5. Despite the incorporation of PBAT and GR into the PLA matrix, the values of G' (Figure 5a) did not suffer a big change at lower temperatures, which is in good agreement with the Young's modulus trend discussed in the mechanical characterization (Table 3). It was observed that the incorporation of 20% of PBAT generated a partially miscible blend, since T<sup>g</sup> changed from 70.2 ◦C for neat PLA to 67.1 ◦C for the PLA/PBAT, as shown by the δ peaks of Figure 5b. Al-Itry et al. obtained a lower decrease (only 1 ◦C) in T<sup>g</sup> when incorporating 20% PBAT [52]. A higher reduction in T<sup>g</sup> was observed when adding GR resin. Specifically, a value of 65.5 ◦C was obtained for the formulation with 5 phr of GR and 64.8 ◦C for the formulation with 10 phr of GR. This behavior demonstrates the plasticizing effect of GR on the PBAT domains and the lubricant effect on the PLA matrix. The values of T<sup>g</sup> obtained by DMA were slightly higher than those obtained by DSC but with the same trend.

**Figure 5.** (**a**) Storage modulus (G') and (**b**) loss factor (δ) of PLA and PLA/PBAT as references, and representatives PLA/PBAT formulations with 5 and 10 phr of GR. **Figure 5.** (**a**) Storage modulus (G') and (**b**) loss factor (δ) of PLA and PLA/PBAT as references, and representatives PLA/PBAT formulations with 5 and 10 phr of GR.

Figure 6 shows the results of the thermogravimetric analysis (TGA). The T5% reflects that the addition of PBAT to the PLA matrix (PLA/PBAT) did not significantly modify (*p*  > 0.05) the thermal stability of neat PLA. This shows that the interactions between PLA and PBAT were poor. The Tmax of PLA/PBAT showed a lower but not significantly different value than neat PLA (*p* > 0.05). The effect of adding GR was to obtain significantly lower T5% values, which were due to the partial degradation of the GR component [38]. At higher contents of GR, the saturation of the resin directly affected the thermal stability since the components were not interacting with the PLA matrix due to the weak interaction of the PBAT domains (as also shown by FESEM). Figure 6 shows the results of the thermogravimetric analysis (TGA). The T5% reflects that the addition of PBAT to the PLA matrix (PLA/PBAT) did not significantly modify (*p* > 0.05) the thermal stability of neat PLA. This shows that the interactions between PLA and PBAT were poor. The Tmax of PLA/PBAT showed a lower but not significantly different value than neat PLA (*p* > 0.05). The effect of adding GR was to obtain significantly lower T5% values, which were due to the partial degradation of the GR component [38]. At higher contents of GR, the saturation of the resin directly affected the thermal stability since the components were not interacting with the PLA matrix due to the weak interaction of the PBAT domains (as also shown by FESEM).

**Figure 6.** (**a**) TGA and (**b**) DTG curves of PLA and PBAT matrixes, and studied representative blends: PLA/PBAT, PLA/PBAT\_20GR, and PBAT\_10GR. a–f Different letters within the same property show statistically significant differences between formulations (*p* < 0.05). **Figure 6.** (**a**) TGA and (**b**) DTG curves of PLA and PBAT matrixes, and studied representative blends: PLA/PBAT, PLA/PBAT\_20GR, and PBAT\_10GR. a–f Different letters within the same property show statistically significant differences between formulations (*p* < 0.05).
