**3. Results and Discussion**

#### *3.1. Fourier Transform Infrared Spectra*

Figure 1 represents the infrared spectra of PLA, MPLA, and MPLA/CSDG composites. The infrared characterization of these three materials at around 2993 and 2940 cm−<sup>1</sup> wavenumbers gave a description of symmetric and asymmetric vibration peaks of −CH− in the PLA structure. Pure PLA and the composite material had an obvious absorption peak at about 2357 cm−<sup>1</sup> , corresponding to the characteristic absorption peak of carbonyl CO2, which may be due to the inevitable air exposure during the process of placing the sample into the detector. The wavenumber at 1430 cm−<sup>1</sup> corresponded to –CH– scissor bending vibration, and that at 1370 cm−<sup>1</sup> referred to the characteristic absorption peak of –CH3–. Absorption peaks at 1185, 1132, 1092, and 1045 cm−<sup>1</sup> denoted <sup>−</sup>C−O<sup>−</sup> stretching vibration peaks. Due to the presence of hydroxyl at the other end of the PLA molecular structure, there would be a weak absorption peak at 3450 cm−<sup>1</sup> , which is not evident (Figure 1). The main reason is that the hydroxyl absorption peak intensity of PLA was relatively small. Figure 1 does not show it clearly, but when biomass fillers that contained a large number of hydroxyl groups were added to the MPLA matrix, the peak position of the hydroxyl groups slightly shifted to the right. This implies that the hydroxyl groups on the surface of the biomass fillers could interact with PLA. That is, the molecular segments of MPLA and CSDG may be bound together. On the other hand, the position of the characteristic carbonyl peak in PLA was about 1757 cm−<sup>1</sup> , and the characteristic peak of carbonyl in the MPLA composite shifted to the right. This also implies that the hydroxyl group on the surface of MPLA could interact with the carbonyl group in PLA. On the basis of the above analysis, along with the research by Wu and Tsou [30] on PLA and rice husk regarding a modification treatment through the use of a coupling agent, the PLA component indicated a behavior similar to the shift of carbonyl characteristic peaks. This interaction could improve the related properties of biomass filler and PLA composite materials. *Polymers* **2021**, *13*, x FOR PEER REVIEW 8 of 25

**Figure 1.** Fourier transform infrared spectra of PLA, MPLA, and MPLA/CSDG.

#### **Figure 1.** Fourier transform infrared spectra of PLA, MPLA, and MPLA/CSDG. *3.2. Data on Mechanical Properties*

cantly higher than those of the PLA/CSDG composites.

*3.2. Data on Mechanical Properties*  Figure 2 and Table 2 provide test results on the influence of different amounts of CSDG on the mechanical properties of PLA/CSDG and MPLA/CSDG composites. Figure 2a depicts tensile strength as a function of the CSDG content, Figure 2b presents data on elongation at break, and Figure 2c plots stress–strain curves. The tensile strength of pure PLA was indicated to be 43.2 MPa. When the content of CSDG was 10%, the tensile Figure 2 and Table 2 provide test results on the influence of different amounts of CSDG on the mechanical properties of PLA/CSDG and MPLA/CSDG composites. Figure 2a depicts tensile strength as a function of the CSDG content, Figure 2b presents data on elongation at break, and Figure 2c plots stress–strain curves. The tensile strength of pure PLA was indicated to be 43.2 MPa. When the content of CSDG was 10%, the tensile strength of the composite material was reduced to 22.15 MPa. At 20% CSDG, the tensile strength increased to 29.34 MPa, which was the highest value. When the amount of added

**Table 2.** Data on mechanical properties of PLA, PLA/CSDG, MPLA, and MPLA/CSDG.

strength of the composite material was reduced to 22.15 MPa. At 20% CSDG, the tensile strength increased to 29.34 MPa, which was the highest value. When the amount of

lower. In the tensile strength data for MPLA/CSDG composites, the tensile strength of the MPLA composite with 10% CSDG was 38.73 MPa. When the CSDG content was increased to 20% and 30%, the tensile strength increased, and the maximum value of 52.68 MPa was reached at the 30% content. However, when 40% CSDG was added, the tensile strength was greatly reduced to 30.2 MPa. Stress–strain curves in Figure 2c indicated a significant difference between PLA and MPLA composites containing 30% CSDG—the tensile strength and elongation at break of the MPLA/CSDG composites were signifi-

**Sample Tensile Strength (MPa) Elongation at Break (%)**  PLA 43.20 ± 1.9 5.08 ± 0.1 PLA/CSDG10 22.18 ± 2.9 2.43 ± 0.3 PLA/CSDG20 29.33 ± 2.8 2.80 ± 0.2 PLA/CSDG30 23.11 ± 1.9 3.05 ± 0.1 PLA/CSDG40 21.32 ± 2.2 2.60 ± 0.06 PLA/CSDG50 18.47 ± 2.1 2.35 ± 0.07 MPLA 37.08 ± 0.7 2.71 ± 0.3 MPLA/CSDG10 38.72 ± 1.9 2.75 ± 0.08 MPLA/CSDG20 41.13 ± 3.3 2.61 ± 0.4 MPLA/CSDG30 52.65 ± 2.0 6.02 ± 0.1 MPLA/CSDG40 30.30 ± 2.0 4.05 ± 0.5

CSDG reached 50%, the tensile strength was significantly reduced to 18.5 MPa. Compared with the overall strength of pure PLA, that of the PLA/CSDG composites was lower. In the tensile strength data for MPLA/CSDG composites, the tensile strength of the MPLA composite with 10% CSDG was 38.73 MPa. When the CSDG content was increased to 20% and 30%, the tensile strength increased, and the maximum value of 52.68 MPa was reached at the 30% content. However, when 40% CSDG was added, the tensile strength was greatly reduced to 30.2 MPa. Stress–strain curves in Figure 2c indicated a significant difference between PLA and MPLA composites containing 30% CSDG—the tensile strength and elongation at break of the MPLA/CSDG composites were significantly higher than those of the PLA/CSDG composites.


**Table 2.** Data on mechanical properties of PLA, PLA/CSDG, MPLA, and MPLA/CSDG.

When CSDG was added to PLA, the elongation at break showed a trend of an initial large decrease, which was followed by a slight increase. The elongation at break reached 3.07% for the sample with 30% CSDG, which was higher than that for the other samples but still lower than that of the pure PLA sample. For the MPLA/CSDG composites, it is apparent that the trend of elongation at break differed much from that for the PLA/CSDG composites. When the content was 30% CSDG, the elongation at break for MPLA/CSDG reached 6.01%, which is the highest among the samples, and it was 1.2 times higher than that for pure PLA.

According to the analysis, the tensile strength and elongation at break of the MPLA/CSDG samples were higher overall than those of PLA/CSDG. These data verify the results from the FTIR analysis, which indicated that the mechanical properties of composite materials became stronger due to the internal interaction. As illustrated by the SEM morphology, the compatibility between CSDG and the PLA matrix became poor when the addition of CSDG was too high, resulting in decreased toughness of the composite material and increased brittleness. In addition, when CSDG increased to a certain level, agglomeration appeared in the film, causing brittle fracture to be more likely to occur [31]. This explanation is consistent with the results obtained by Zhao et al. [32]. In other words, improving the interfacial adhesion is of great significance in enhancing the mechanical properties of composite materials.

#### *3.3. X-ray Diffraction Patterns*

With the help of the XRD patterns shown in Figure 3, PLA, MPLA, and two sets of composite materials were analyzed. Figure 3a is the XRD spectra of PLA/CSDG composites, and Figure 3b is the XRD spectra of MPLA/CSDG composites. Different proportions of CSDG in the composites affected the crystallization and the degree of enhancement in the structure and other properties.

MPLA/CSDG50 23.73 ± 1.8 2.72 ± 0.3

0

**Figure 2.** Mechanical properties of PLA/CSDG and MPLA/CSDG composites: (**a**) tensile strength; (**b**) elongation at break; (**c**) stress–strain curves.

**Figure 3.** X-ray diffraction patterns: (**a**) PLA/CSDG; (**b**) MPLA/CSDG.

**Figure 3.** X-ray diffraction patterns: (**a**) PLA/CSDG; (**b**) MPLA/CSDG.

*3.4. Morphological Images*  Generally, when polymers deform under stress, the presence of fillers would produce concentration effects and cause micro-cracks around the polymer. The contact area of fillers with the polymer is large, and when the filler amount is small, micro-cracks would be generated. However, when the filler amount becomes too large, such micro-cracks would turn into macro-cracks, which should lead to poor mechanical properties [37]. Therefore, the shape characteristics of filler, the dispersion in the polymer matrix, and the adhesion between the filler and the matrix all have an important impact on the mechanical properties of composite materials [38–41]. SEM images of the tensile section of PLA/CSDG composites are shown in Figure 4. The surface of pure PLA (Figure 4a) was dense and uniform. It is observed that the fracture surface of PLA/CSDG composites There are obvious sharp diffraction peaks in PLA at 2θ = 16.52 and 18.87◦ , which refer to (200/110) and (203) planes corresponding to typical α crystals of PLA [33,34]. Throughout the spectral curve, PLA/CSDG with 10% and 30% CSDG had almost no shift in peak, indicating that the crystal morphology was basically unchanged [35]. When the content was 20%, PLA/CSDG characteristic peaks shifted slightly to the right. However, when the CSDG content was 40–50%, the characteristic peak gradually shifted to the left. Evident self-aggregation may be the reason for this phenomenon. Similarly, in MPLA/CSDGG samples, the characteristic peak did not shift when the CSDG content was 10%. When the CSDG content was 20–30%, the characteristic peak gradually shifted to the right [36]. When CSDG promoted the completion of crystallization, and as the addition of CSDG was continued, the characteristic peak of MPLA shifted to the left. According to the Bragg equation: 2dsinθ = nλ, θ increases when nλ does not change, and the value of d decreases, which reduces the interplanar spacing d in a composite material, which may increase the crystallinity and increase the tensile strength. When CSDG was continually

added, most CSDG particles occupied the PLA crystal array, so that θ decreased and d increased, resulting in decreased crystallinity. In other words, too many crystal nuclei may hinder the growth of crystals, which may lead to a decline in the degree of crystallization. exhibited a rough appearance, and the irregular appearance of CSDG particles inter-

*Polymers* **2021**, *13*, x FOR PEER REVIEW 12 of 25

#### *3.4. Morphological Images* spersed and distributed in the section could be clearly observed. At higher CSDG concentrations (30, 40, 50 wt %), it is observed that the aggregates gradually increased, and it

Generally, when polymers deform under stress, the presence of fillers would produce concentration effects and cause micro-cracks around the polymer. The contact area of fillers with the polymer is large, and when the filler amount is small, micro-cracks would be generated. However, when the filler amount becomes too large, such micro-cracks would turn into macro-cracks, which should lead to poor mechanical properties [37]. Therefore, the shape characteristics of filler, the dispersion in the polymer matrix, and the adhesion between the filler and the matrix all have an important impact on the mechanical properties of composite materials [38–41]. SEM images of the tensile section of PLA/CSDG composites are shown in Figure 4. The surface of pure PLA (Figure 4a) was dense and uniform. It is observed that the fracture surface of PLA/CSDG composites exhibited a rough appearance, and the irregular appearance of CSDG particles interspersed and distributed in the section could be clearly observed. At higher CSDG concentrations (30, 40, 50 wt %), it is observed that the aggregates gradually increased, and it can be established that the dispersibility was better when the filler concentration was lower, and the interfacial effect was relatively good. This means that when the amount of CSDG was higher, poor interfacial adhesion would lead to uneven internal structure of the obtained PLA/CSDG composites, which may lead to deterioration of the mechanical properties [42]. can be established that the dispersibility was better when the filler concentration was lower, and the interfacial effect was relatively good. This means that when the amount of CSDG was higher, poor interfacial adhesion would lead to uneven internal structure of the obtained PLA/CSDG composites, which may lead to deterioration of the mechanical properties [42]. Figure 5 illustrates the tensile section morphology of MPLA/CSDG samples with different amounts of CSDG. Although we can see that there are CSDG particles on the sample surface, MPLA/CSDG showed relatively flat morphology (relative to the rough surface of PLA/CSDG samples). Only a small amount of CSDG aggregates appeared in the matrix, and the distribution was relatively uniform. Nearly no gaps existed between the filler and the matrix, except for the sample with 20 wt % CSDG (Figure 5c). A small number of pores are visible, which may cause the elongation at break to be lower than that of the other samples. The relatively compact structure of MPLA/CSDG samples indicated that a dense structure was formed between MPLA and CSDG. In addition, when the CSDG content was 40 wt % (Figure 5e) and 50 wt % (Figure 5f), significant agglomeration occurred, which led to a decrease in the compatibility between CSDG and the MPLA matrix [43].

**Figure 4.** SEM images of PLA and PLA/CSDG composites: (**a**) PLA; (**b**) PLA/CSDG10; (**c**) PLA/CSDG20; (**d**) PLA/CSDG30; (**e**) PLA/CSDG40; (**f**) PLA/CSDG50. **Figure 4.** SEM images of PLA and PLA/CSDG composites: (**a**) PLA; (**b**) PLA/CSDG10; (**c**) PLA/CSDG20; (**d**) PLA/CSDG30; (**e**) PLA/CSDG40; (**f**) PLA/CSDG50.

Figure 5 illustrates the tensile section morphology of MPLA/CSDG samples with different amounts of CSDG. Although we can see that there are CSDG particles on the sample surface, MPLA/CSDG showed relatively flat morphology (relative to the rough surface of PLA/CSDG samples). Only a small amount of CSDG aggregates appeared in the matrix, and the distribution was relatively uniform. Nearly no gaps existed between the filler and the matrix, except for the sample with 20 wt % CSDG (Figure 5c). A small number of pores are visible, which may cause the elongation at break to be lower than that of the other samples. The relatively compact structure of MPLA/CSDG samples indicated that a dense structure was formed between MPLA and CSDG. In addition, when the CSDG

content was 40 wt % (Figure 5e) and 50 wt % (Figure 5f), significant agglomeration occurred, which led to a decrease in the compatibility between CSDG and the MPLA matrix [43]. *Polymers* **2021**, *13*, x FOR PEER REVIEW 13 of 25

**Figure 5.** SEM images of MPLA and MPLA/CSDG composites: (**a**) MPLA; (**b**) MPLA/CSDG10; (**c**) MPLA/CSDG20; (**d**) MPLA/CSDG30; (**e**) MPLA/CSDG40; (**f**) MPLA/CSDG50. **Figure 5.** SEM images of MPLA and MPLA/CSDG composites: (**a**) MPLA; (**b**) MPLA/CSDG10; (**c**) MPLA/CSDG20; (**d**) MPLA/CSDG30; (**e**) MPLA/CSDG40; (**f**) MPLA/CSDG50.

#### *3.5. Thermal Stability Analysis 3.5. Thermal Stability Analysis*

DSC test results are shown in Figure 6, as well as in Table 3. DSC was conducted to determine the glass transition temperature (Tg), melting temperature, recrystallization temperature (Tc), enthalpy of fusion, enthalpy of crystallization, and degree of crystallinity of composite materials. From the cooling curves in Figure 6, PLA and MPLA indicated no crystallization peaks. This is because PLA had very slow crystallization, and the cooling rate of 10 °C/min was too fast, making it too late for PLA to crystallize [44]. With the addition of CSDG, both PLA and MPLA indicated crystallization peaks, which may be attributed to CSDG acting as a nucleating agent for polymers and accelerating the crystallization rate of PLA. From the first heating curves, because the samples were in the initial thermal history, the shape of the curves was irregular, so the sample second heating curves (Figure 7) were considered. In Figure 7, when the CSDG content gradually increased from 0 to 50%, Tg decreased from 68.7 to 58.6 °C, reaching the lowest temperature. The reason for the decrease in Tg may be the presence of many polar groups in CSDG itself. When the number of polar groups in the composite chain exceeded a certain value, the electrostatic repulsion between them exceeded the attractive force, which led to an increase in the distance between molecular chains and a decrease in Tg. At the same time, the overall melting temperature of PLA/CSDG composites during the second heating was slightly lower than that of pure PLA, indicating that the rigidity of the blend was affected to a lesser extent, and the melting temperature was reduced from 170.1 to 168.2 °C (CSDG content increased from 0 to 20%), and it then began to rise (CSDG content, 20–50%). From the analysis, it is believed that when the CSDG content was too much (30%), so that it was not easy for CSDG to disperse in PLA, CSDG produced some subtle aggregation effects, making the crystallization less complete [45,46]. Tc decreased more significantly by 8–17 °C, indicating that CSDG exerted a nucleation effect on PLA [47]. The data for MPLA/CSDG samples showed that Tc and Tg were significantly lower than those for PLA/CSDG. When CSDG was added to PLA, due to the interaction between the DSC test results are shown in Figure 6, as well as in Table 3. DSC was conducted to determine the glass transition temperature (Tg), melting temperature, recrystallization temperature (Tc), enthalpy of fusion, enthalpy of crystallization, and degree of crystallinity of composite materials. From the cooling curves in Figure 6, PLA and MPLA indicated no crystallization peaks. This is because PLA had very slow crystallization, and the cooling rate of 10 ◦C/min was too fast, making it too late for PLA to crystallize [44]. With the addition of CSDG, both PLA and MPLA indicated crystallization peaks, which may be attributed to CSDG acting as a nucleating agent for polymers and accelerating the crystallization rate of PLA. From the first heating curves, because the samples were in the initial thermal history, the shape of the curves was irregular, so the sample second heating curves (Figure 7) were considered. In Figure 7, when the CSDG content gradually increased from 0 to 50%, T<sup>g</sup> decreased from 68.7 to 58.6 ◦C, reaching the lowest temperature. The reason for the decrease in T<sup>g</sup> may be the presence of many polar groups in CSDG itself. When the number of polar groups in the composite chain exceeded a certain value, the electrostatic repulsion between them exceeded the attractive force, which led to an increase in the distance between molecular chains and a decrease in Tg. At the same time, the overall melting temperature of PLA/CSDG composites during the second heating was slightly lower than that of pure PLA, indicating that the rigidity of the blend was affected to a lesser extent, and the melting temperature was reduced from 170.1 to 168.2 ◦C (CSDG content increased from 0 to 20%), and it then began to rise (CSDG content, 20–50%). From the analysis, it is believed that when the CSDG content was too much (30%), so that it was not easy for CSDG to disperse in PLA, CSDG produced some subtle aggregation effects, making the crystallization less complete [45,46]. T<sup>c</sup> decreased more significantly by 8–17 ◦C, indicating that CSDG exerted a nucleation effect on PLA [47]. The data for MPLA/CSDG samples showed that T<sup>c</sup> and T<sup>g</sup> were significantly lower than those for PLA/CSDG. When CSDG was added to PLA, due to the interaction between the hydroxyl groups in the PLA

hydroxyl groups in the PLA matrix and the filler, the degree of restriction in the movement of molecular chains was increased, hindering the movement of the polymer chains [48]. The existence of more functional groups in MPLA improved the molecular bonding

matrix and the filler, the degree of restriction in the movement of molecular chains was increased, hindering the movement of the polymer chains [48]. The existence of more functional groups in MPLA improved the molecular bonding between CSDG and MPLA as well as the uniform dispersion of CSDG in the MPLA matrix. The interaction between the MPLA matrix and the filler improved, free volume in the polymer increased, the degree of hindrance in the movement of molecular chains decreased, and the average chain length between cross-linking points became large, so T<sup>g</sup> decreased. the polymer increased, the degree of hindrance in the movement of molecular chains decreased, and the average chain length between cross-linking points became large, so Tg decreased.

trix. The interaction between the MPLA matrix and the filler improved, free volume in

*Polymers* **2021**, *13*, x FOR PEER REVIEW 14 of 25

**Figure 6.** Cooling curves: (**a**) PLA/CSDG; (**b**) MPLA/CSDG. **Figure 6.** Cooling curves: (**a**) PLA/CSDG; (**b**) MPLA/CSDG.

**Sample** 

**Glass Transition Temperature (℃)** 


**Table 3.** Differential scanning calorimetry data for PLA, PLA/CSDG, MPLA, and MPLA/CSDG.

**Figure 7.** Second heating curves: (**a**) PLA/CSDG; (**b**) MPLA/CSDG.

**Figure 7.** Second heating curves: (**a**) PLA/CSDG; (**b**) MPLA/CSDG.

**Table 3.** Differential scanning calorimetry data for PLA, PLA/CSDG, MPLA, and MPLA/CSDG.

**Enthalpy of Crystalliza-**

**Melting Temperature (℃)** 

**Melting Enthalpy (J/g)** 

**Crystallinity (%)** 

**Recrystallization Temperature (Tc) (℃)** 

PLA/CSDG20 61.1 102.8 16.77 168.2 26.12 34.84 PLA/CSDG30 61.1 102.2 11.82 169 24.6 37.51 PLA/CSDG40 59.9 97.9 7.692 169.5 23.1 40.93 PLA/CSDG50 58.6 94.8 3.198 169.7 19.4 41.43 MPLA 48.8 102.3 21.27 165.7 30.3 32.31 MPLA/SDG10 61.1 105.4 21.92 169 32.3 38.33 MPLA/SDG20 60.6 102.7 15.94 168.7 31.2 41.58 MPLA/SDG30 60.5 99.6 8.66 169.5 26.8 43.78 MPLA/SDG40 59.2 94.2 2.87 169.8 24.5 41.54 MPLA/CSDG50 59.5 93.6 1.51 170.1 20.3 41.22

#### *3.6. Thermogravimetric Analysis*

DTG results for PLA/CSDG and MPLA/CSDG composites are shown in Figure 8. Table 4 lists the thermal degradation data. The thermal stability of composite materials was examined through TGA. PLA/CSDG and MPLA/CSDG composites both exhibited three stages of mass loss. First, mass loss started at around 200 ◦C due to the volatilization of water. The second step involved the loss of PLA at around 350 ◦C. The mass loss at about 450 ◦C in the third step was due to the decomposition of composite materials and modifiers. It can be deduced that the addition of CSDG fibers significantly reduced the initial decomposition temperature of the composite materials. A previous researcher analyzed the thermal stability of PLA/ramie composites and found that the addition of ramie fibers reduced the initial decomposition temperature of the composites [49], which is consistent with the finding from this present experiment. The initial degradation temperature of PLA/CSDG decreased with increase in the CSDG content. The temperature at maximum mass loss of PLA/CSDG decreased slightly compared with that of pure PLA. For MPLA/CSDG composites, the temperature at maximum mass loss decreased more significantly, and the initial degradation temperature also decreased substantially. This indicates that unmodified composites maintained better thermal stability, but the thermal stability of both modified and unmodified composites was lower compared with that of pure PLA. The thermal results agree with the formation of aggregates revealed by SEM images. Similar observations were detected for biopolymer matrices filled with coffee grounds [50] and inorganic clay nanoparticles [51]. In general, the clustering of fillers generates a reduction of the polymer thermal stability.

**Table 4.** Differential thermogravimetric data for PLA, PLA/CSDG, MPLA, and MPLA/CSDG.


#### *3.7. Analysis of Oxygen Barrier Performance*

Figure 9 describes the effect of different amounts of CSDG on the oxygen permeability in PLA/CSDG and MPLA/CSDG composites. The oxygen permeability in pure PLA was 2.626 cm3/m<sup>2</sup> ·d·Pa. With the increase in the CSDG content, the trend in oxygen permeability increased. When the CSDG content was 40%, the oxygen permeability in PLA/CSDG increased to 113.725 cm3/m<sup>2</sup> ·d·Pa. In particular, when the CSDG content was 50%, at which excessive self-aggregation of CSDG occurred and CSDG and PLA had poor compatibility with each other, the composite material exhibited too many defects and low mechanical properties. Therefore, we infer that due to the presence of a hydrophilic substance, more voids and micro-cracks were formed, resulting in the decreased oxygen barrier performance of PLA/CSDG. However, the oxygen barrier properties of MPLA/CSDG composites improved. When the CSDG content was 10%, the oxygen permeability in MPLA/CSDG dropped to 2.241 cm3/m<sup>2</sup> ·d·Pa, which is slightly lower than that in pure PLA film. However, when the CSDG content was greater than 20%, the oxygen permeability started to increase. At the maximum CSDG content of 50%, the oxygen permeability in MPLA/CSDG was 7.551 cm3/m<sup>2</sup> ·d·Pa, which was lower than that in PLA/CSDG (see sub-Figure 9). It

is clear that MPLA had a significant effect on enhancing the oxygen barrier performance. Due to the acid anhydride group in MPLA, the composite material was able to form a relatively tight network structure and connections, leading to a significant decrease in oxygen permeability and a significant improvement in oxygen barrier performance. *Polymers* **2021**, *13*, x FOR PEER REVIEW 17 of 25

**Figure 8.** Differential thermogravimetric curves: (**a**) PLA/CSDG; (**b**) MPLA/CSDG. **Figure 8.** Differential thermogravimetric curves: (**a**) PLA/CSDG; (**b**) MPLA/CSDG.

Figure 9 describes the effect of different amounts of CSDG on the oxygen permeability in PLA/CSDG and MPLA/CSDG composites. The oxygen permeability in pure PLA was 2.626 cm3/m2·d·Pa. With the increase in the CSDG content, the trend in oxygen permeability increased. When the CSDG content was 40%, the oxygen permeability in

*3.7. Analysis of Oxygen Barrier Performance* 

compatibility with each other, the composite material exhibited too many defects and

oxygen barrier performance.

low mechanical properties. Therefore, we infer that due to the presence of a hydrophilic substance, more voids and micro-cracks were formed, resulting in the decreased oxygen barrier performance of PLA/CSDG. However, the oxygen barrier properties of MPLA/CSDG composites improved. When the CSDG content was 10%, the oxygen permeability in MPLA/CSDG dropped to 2.241 cm3/m2.d.Pa, which is slightly lower than that in pure PLA film. However, when the CSDG content was greater than 20%, the oxygen permeability started to increase. At the maximum CSDG content of 50%, the oxygen permeability in MPLA/CSDG was 7.551 cm3/m2.d.Pa, which was lower than that in PLA/CSDG (see sub-figure 9). It is clear that MPLA had a significant effect on enhancing the oxygen barrier performance. Due to the acid anhydride group in MPLA, the composite material was able to form a relatively tight network structure and connections, leading to a significant decrease in oxygen permeability and a significant improvement in

**Figure 9.** Oxygen barrier properties of PLA/CSDG and MPLA/CSDG composites.

#### **Figure 9.** Oxygen barrier properties of PLA/CSDG and MPLA/CSDG composites. *3.8. Analysis of Water Vapor Barrier Properties*

*3.8. Analysis of Water Vapor Barrier Properties*  Figure 10 plots test results about the influence of different amounts of CSDG on the water vapor permeability in PLA/CSDG and MPLA/CSDG composites. Due to the presence of a large amount of hydrophilic –OH in the molecular chain of PLA, the water-blocking performance of pure PLA film was poor. From the test results, the water vapor permeability in pure PLA was 19.64 g/m2/d. When CSDG was added, the water vapor permeability in PLA/CSDG began to increase. At 50% CSDG, the water vapor permeability in the composite film increased to 250.25 g/m2/d. Therefore, we infer that due to the CSDG hydrophilicity, as indicated from the analysis of electron microscope, excessive CSDG easily led to its agglomeration, and a large number of hydrophilic groups were exposed, thereby reducing the water vapor barrier properties of the composite material [52]. For the case of MPLA, especially when the CSDG content was 10%, the water vapor permeability in MPLA/CSDG dropped to 10.81 g/m2/d. When the added CSDG was 20%, the water vapor permeability in the MPLA/CSDG composite was higher Figure 10 plots test results about the influence of different amounts of CSDG on the water vapor permeability in PLA/CSDG and MPLA/CSDG composites. Due to the presence of a large amount of hydrophilic –OH in the molecular chain of PLA, the waterblocking performance of pure PLA film was poor. From the test results, the water vapor permeability in pure PLA was 19.64 g/m2/d. When CSDG was added, the water vapor permeability in PLA/CSDG began to increase. At 50% CSDG, the water vapor permeability in the composite film increased to 250.25 g/m2/d. Therefore, we infer that due to the CSDG hydrophilicity, as indicated from the analysis of electron microscope, excessive CSDG easily led to its agglomeration, and a large number of hydrophilic groups were exposed, thereby reducing the water vapor barrier properties of the composite material [52]. For the case of MPLA, especially when the CSDG content was 10%, the water vapor permeability in MPLA/CSDG dropped to 10.81 g/m2/d. When the added CSDG was 20%, the water vapor permeability in the MPLA/CSDG composite was higher than that in pure PLA film. When CSDG was 50%, the water vapor permeability reached 25.03 g/m2/d. These findings are consistent with the reported barrier properties of PLA-based composites [1]. However, this result for MPLA/CSDG is better than that for the PLA/CSDG composite films. The results showed that the introduction of the acid anhydride group in MPLA improved the adhesion between CSDG and the polymer, significantly improving the water vapor barrier properties of the composite material. However, at the same time, because the CSDG content was too high, agglomeration occurred, resulting in voids in the composite matrix, making the structure loose. This resulted in a decrease in the composite film resistance to water vapor permeability [53,54]. Figure 11 clearly depicts changes in the permeation path of water vapor or oxygen in PLA/CSDG and MPLA/CSDG composites. These findings are consistent with the reported barrier properties of PLA films [55]. However, due to the poor compatibility and poor dispersion of CSDG in the PLA matrix, PLA/CSDG samples failed to form a dense structure and a good barrier effect against water vapor or oxygen. Figure 11b indicates that MPLA/CSDG composites had denser structure, so it was more difficult for water or oxygen to pass through.

oxygen to pass through.

300

oxygen to pass through.

**Figure 10.** Water vapor barrier properties of PLA/CSDG and MPLA/CSDG composites. **Figure 10.** Water vapor barrier properties of PLA/CSDG and MPLA/CSDG composites.

**Figure 10.** Water vapor barrier properties of PLA/CSDG and MPLA/CSDG composites.

than that in pure PLA film. When CSDG was 50%, the water vapor permeability reached 25.03 g/m2/d. These findings are consistent with the reported barrier properties of PLA-based composites [1]. However, this result for MPLA/CSDG is better than that for the PLA/CSDG composite films. The results showed that the introduction of the acid anhydride group in MPLA improved the adhesion between CSDG and the polymer, significantly improving the water vapor barrier properties of the composite material. However, at the same time, because the CSDG content was too high, agglomeration occurred, resulting in voids in the composite matrix, making the structure loose. This resulted in a decrease in the composite film resistance to water vapor permeability [53,54]. Figure 11 clearly depicts changes in the permeation path of water vapor or oxygen in PLA/CSDG and MPLA/CSDG composites. These findings are consistent with the reported barrier properties of PLA films [55]. However, due to the poor compatibility and poor dispersion of CSDG in the PLA matrix, PLA/CSDG samples failed to form a dense structure and a good barrier effect against water vapor or oxygen. Figure 11b indicates that MPLA/CSDG composites had denser structure, so it was more difficult for water or

than that in pure PLA film. When CSDG was 50%, the water vapor permeability reached 25.03 g/m2/d. These findings are consistent with the reported barrier properties of PLA-based composites [1]. However, this result for MPLA/CSDG is better than that for the PLA/CSDG composite films. The results showed that the introduction of the acid anhydride group in MPLA improved the adhesion between CSDG and the polymer, significantly improving the water vapor barrier properties of the composite material. However, at the same time, because the CSDG content was too high, agglomeration occurred, resulting in voids in the composite matrix, making the structure loose. This resulted in a decrease in the composite film resistance to water vapor permeability [53,54]. Figure 11 clearly depicts changes in the permeation path of water vapor or oxygen in PLA/CSDG and MPLA/CSDG composites. These findings are consistent with the reported barrier properties of PLA films [55]. However, due to the poor compatibility and poor dispersion of CSDG in the PLA matrix, PLA/CSDG samples failed to form a dense structure and a good barrier effect against water vapor or oxygen. Figure 11b indicates that MPLA/CSDG composites had denser structure, so it was more difficult for water or

*Polymers* **2021**, *13*, x FOR PEER REVIEW 19 of 25

**Figure 11.** Schematic of molecular permeation path for water vapor or oxygen: (**a**) PLA/CSDG; (**b**) MPLA/CSDG. **Figure 11.** Schematic of molecular permeation path for water vapor or oxygen: (**a**) PLA/CSDG; (**b**) MPLA/CSDG.

#### *3.9. Contact Angle Data 3.9. Contact Angle Data*

bicity of PLA.

65

*3.10. Water Absorption Analysis* 

70

75

80

Contact angle (°)

85

90

95

From the data analysis for PLA/CSDG and MPLA/CSDG composites in Figure 12, the contact angle of composites increased significantly. The overall contact angle of MPLA/CSDG was higher than that of unmodified composites. This can be explained by the reaction between the acid anhydride group in MPLA and –OH in CSDG to form a relatively tight network structure, making the material internal connections closer. The binding force at the interface between CSDG and PLA was enhanced, so that the hydrophobicity was also enhanced. With increase in the CSDG content, PLA/CSDG and MPLA/CSDG samples reached the highest contact angle when CSDG was 30% and 20%, respectively. At higher CSDG content, the contact angle began to decrease. The reason is the occurrence of CSDG agglomeration in the PLA matrix, suggesting that a certain maximum content of CSDG would cause the hydrophobicity of PLA to increase. PLA gave an initial contact angle of 73.5 ± 2.5°, while PLA/CSDG and MPLA/CSDG samples From the data analysis for PLA/CSDG and MPLA/CSDG composites in Figure 12, the contact angle of composites increased significantly. The overall contact angle of MPLA/CSDG was higher than that of unmodified composites. This can be explained by the reaction between the acid anhydride group in MPLA and –OH in CSDG to form a relatively tight network structure, making the material internal connections closer. The binding force at the interface between CSDG and PLA was enhanced, so that the hydrophobicity was also enhanced. With increase in the CSDG content, PLA/CSDG and MPLA/CSDG samples reached the highest contact angle when CSDG was 30% and 20%, respectively. At higher CSDG content, the contact angle began to decrease. The reason is the occurrence of CSDG agglomeration in the PLA matrix, suggesting that a certain maximum content of CSDG would cause the hydrophobicity of PLA to increase. PLA gave an initial contact angle of 73.5 ± 2.5◦ , while PLA/CSDG and MPLA/CSDG samples both

both showed higher contact angles. The maximum contact angle was close to 90° in the

 PLA/CSDG MPLA/CSDG

Test results on water absorption in PLA/CSDG and MPLA/CSDG composite films are indicated in Figure 13. It is clear that the rate of water absorption indicated a rising

0 10 20 30 40 50

CSDG content (wt%)

**Figure 12.** Contact angle of PLA/CSDG and MPLA/CSDG composites.

showed higher contact angles. The maximum contact angle was close to 90◦ in the case of PLA/CSDG and 85◦ for MPLA/CSDG, indicating improvement in the hydrophobicity of PLA. *3.9. Contact Angle Data*  From the data analysis for PLA/CSDG and MPLA/CSDG composites in Figure 12, the contact angle of composites increased significantly. The overall contact angle of

**Figure 11.** Schematic of molecular permeation path for water vapor or oxygen: (**a**) PLA/CSDG; (**b**)

#### *3.10. Water Absorption Analysis* MPLA/CSDG was higher than that of unmodified composites. This can be explained by

MPLA/CSDG.

*Polymers* **2021**, *13*, x FOR PEER REVIEW 20 of 25

Test results on water absorption in PLA/CSDG and MPLA/CSDG composite films are indicated in Figure 13. It is clear that the rate of water absorption indicated a rising trend. This is because CSDG contained cellulose, hemicellulose, and lignin—all of them rich in hydroxyl groups—and therefore, materials with CSDG should have high water absorption. This same phenomenon is also reflected in the research by Wen et al. [56] on vinasse and polyethylene. The water absorption in MPLA/CSDG was lower than that in PLA/CSDG samples, and the increase in water absorption tended to be slow, indicating that MPLA/CSDG was more compact, so the water absorption was relatively poor. These results are consistent with the results of contact angle analysis. However, when the filler content was too high, CSDG would agglomerate, and the bond absorption capacity of hydroxyl groups on the surface of CSDG would increase. This may be the main reason for the increase in water absorption in samples with high CSDG content. the reaction between the acid anhydride group in MPLA and –OH in CSDG to form a relatively tight network structure, making the material internal connections closer. The binding force at the interface between CSDG and PLA was enhanced, so that the hydrophobicity was also enhanced. With increase in the CSDG content, PLA/CSDG and MPLA/CSDG samples reached the highest contact angle when CSDG was 30% and 20%, respectively. At higher CSDG content, the contact angle began to decrease. The reason is the occurrence of CSDG agglomeration in the PLA matrix, suggesting that a certain maximum content of CSDG would cause the hydrophobicity of PLA to increase. PLA gave an initial contact angle of 73.5 ± 2.5°, while PLA/CSDG and MPLA/CSDG samples both showed higher contact angles. The maximum contact angle was close to 90° in the case of PLA/CSDG and 85° for MPLA/CSDG, indicating improvement in the hydrophobicity of PLA.

**Figure 12.** Contact angle of PLA/CSDG and MPLA/CSDG composites.

**Figure 12.** Contact angle of PLA/CSDG and MPLA/CSDG composites.

**Figure 13.** Water uptake for PLA/CSDG and MPLA/CSDG composites.

tion rates of different samples with time are shown. For all samples, the degradation could be roughly divided into two stages: initial stage (I) for the first 60 days, and final

that the sample degradation rate was greatly increased. In other words, the samples exhibited a "self-accelerating effect" during the biodegradation process [57]. The soil medium penetrated into the polymer matrix, causing the polymer molecular chains to relax, the chemical bonds to gradually decompose, the molecular weight to decrease, and the material to gradually degrade into oligomers. After 60 days of degradation, there would be more and more free hydroxyl (−OH) and carboxyl (−COOH) groups that accelerated their internal degradation, and further degradation would lead the oligomer to decompose into small molecules, resulting in increased degradation rate for the composite materials in later stages. It was observed that the degradation rate for MPLA/CSDG after 90 days was slightly lower than that for PLA/CSDG. On one hand, the MPLA composite exhibited relatively good internal compatibility. On the other hand, MPLA had reactive acid anhydride groups that would react with the hydroxyl groups on the surface of CSDG and with the carboxyl groups in the polymer matrix to increase the interfacial force between CSDG and MPLA, making the composite material relatively stable. Although the carboxyl group in the PLA matrix could also react with CSDG, the −COOH

#### **Figure 13.** Water uptake for PLA/CSDG and MPLA/CSDG composites. *3.11. Biodegradation Rates*

*3.11. Biodegradation Rates*  Figure 14 presents a macroanalysis of PLA, MPLA, as well as PLA/CSDG and MPLA/CSDG composites during the process of degradation; changes in the biodegrada-Figure 14 presents a macroanalysis of PLA, MPLA, as well as PLA/CSDG and MPLA/CSDG composites during the process of degradation; changes in the biodegradation rates of different samples with time are shown. For all samples, the degradation could

be roughly divided into two stages: initial stage (I) for the first 60 days, and final stage (II) after 60 days. In stage I, the sample biodegradation rate changed little with time. However, in stage II, changes in a short period of time were more observable, indicating that the sample degradation rate was greatly increased. In other words, the samples exhibited a "self-accelerating effect" during the biodegradation process [57]. The soil medium penetrated into the polymer matrix, causing the polymer molecular chains to relax, the chemical bonds to gradually decompose, the molecular weight to decrease, and the material to gradually degrade into oligomers. After 60 days of degradation, there would be more and more free hydroxyl (−OH) and carboxyl (−COOH) groups that accelerated their internal degradation, and further degradation would lead the oligomer to decompose into small molecules, resulting in increased degradation rate for the composite materials in later stages. It was observed that the degradation rate for MPLA/CSDG after 90 days was slightly lower than that for PLA/CSDG. On one hand, the MPLA composite exhibited relatively good internal compatibility. On the other hand, MPLA had reactive acid anhydride groups that would react with the hydroxyl groups on the surface of CSDG and with the carboxyl groups in the polymer matrix to increase the interfacial force between CSDG and MPLA, making the composite material relatively stable. Although the carboxyl group in the PLA matrix could also react with CSDG, the −COOH group was only at the end of the PLA molecular chain, which could not be compared with the large number of anhydride groups in the side chain of MPLA. *Polymers* **2021**, *13*, x FOR PEER REVIEW 22 of 25 group was only at the end of the PLA molecular chain, which could not be compared with the large number of anhydride groups in the side chain of MPLA.

**Figure 14.** Degradability of (**a**) PLA/CSDG and (**b**) MPLA/CSDG composites.

In this study, CSDG was incorporated as a reinforcing filler, and PLA/CSDG com-

tween CSDG and PLA, MPLA containing acid anhydride groups was adopted as well. Thus, MPLA/CSDG composites were also fabricated. For the two systems (PLA/CSDG and MPLA/CSDG), mechanical properties, hydrophobicity, gas barrier properties, biocompatibility, and thermodynamic properties were measured and comprehensively analyzed. The analysis revealed that the determining parameters were the dispersion of CSDG and the binding force between PLA and CSDG. Results demonstrated that with increasing CSDG content, the mechanical, barrier, hydrophobic, and thermal degradation properties all had corresponding changes that indicated different trends. The addition of MPLA improved the composite performance. Under the conditions considered in this work, the optimal composition of composites was 20–30% CSDG. In this composition

**Figure 14.** Degradability of (**a**) PLA/CSDG and (**b**) MPLA/CSDG composites.

**4. Conclusions** 

### **4. Conclusions**

In this study, CSDG was incorporated as a reinforcing filler, and PLA/CSDG composites were prepared by melt blending. To improve the interfacial bonding force between CSDG and PLA, MPLA containing acid anhydride groups was adopted as well. Thus, MPLA/CSDG composites were also fabricated. For the two systems (PLA/CSDG and MPLA/CSDG), mechanical properties, hydrophobicity, gas barrier properties, biocompatibility, and thermodynamic properties were measured and comprehensively analyzed. The analysis revealed that the determining parameters were the dispersion of CSDG and the binding force between PLA and CSDG. Results demonstrated that with increasing CSDG content, the mechanical, barrier, hydrophobic, and thermal degradation properties all had corresponding changes that indicated different trends. The addition of MPLA improved the composite performance. Under the conditions considered in this work, the optimal composition of composites was 20–30% CSDG. In this composition range, MPLA/CSDG composites had good filler dispersion, enhanced hydrophobicity, improved stability, and high mechanical properties. The analysis from biodegradation experiments pointed out that as the degradation progressed, all the samples exhibited a "self-accelerating effect". As the filler content increased, CSDG began to agglomerate, the brittleness of composites increased, and the barrier performance decreased. At the same time, MPLA/CSDG composites delivered stronger performance than PLA/CSDG composites. As CSDG is generally categorized as waste resources, recycling it not only reduces environmental issues but also expands the application of the natural biomass in environment-friendly functional materials.

**Author Contributions:** Conceptualization, C.-S.W., C.-H.T. and Z.-J.C.; methodology, C.-H.T. and C.-S.W.; software, Z.-J.C., J.G., T.Y., S.C. and R.-Y.W.; validation, C.-S.W. and M.R.D.G.; formal analysis, Z.-J.C., J.G., P.-W.G., Y.L., L.-J.T. and C.-L.Q.; investigation, C.-H.T., J.G., R.-Y.W. and M.-L.T.; resources, M.-L.T., C.-S.W. and C.G.; data curation, J.G., Z.-J.C., T.Y. and C.G.; writing—original draft preparation, C.-H.T. and Z.-J.C.; writing—review and editing, C.-S.W., M.R.D.G. and C.-H.T.; visualization, C.-S.W.; supervision, C.-S.W. and C.-H.T.; project administration, C.-H.T. and C.G.; funding acquisition, C.-S.W., C.-H.T. and C.G. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Sichuan Province Science and Technology Support Program, grant number (2019JDRC0029), by Wuliangye Group Co. Ltd. (CXY2019ZR001) and by the Opening Project of Key Laboratories of Fine Chemicals and Surfactants in Sichuan Provincial Universities (2020JXY04).

**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:** The authors would like to acknowledge the financial support from the following organizations: Wuliangye Group Co. Ltd. (CXY2019ZR001); Sichuan Province Science and Technology Support Program (2019JDRC0029); Zigong City Science and Technology (2017XC16; 2019CXRC01; 2020YGJC13); Opening Project of Material Corrosion and Protection Key Laboratory of Sichuan Province (2017CL03; 2019CL05; 2018CL08; 2018CL07; 2016CL10); Opening Project of Sichuan Province, the Foundation of Introduced Talent of Sichuan University of Science and Engineering (2017RCL31; 2017RCL36; 2017RCL16; 2019RC05; 2019RC07; 2014RC31; 2020RC16); the Opening Project of Key Laboratories of Fine Chemicals and Surfactants in Sichuan Provincial Universities (2020JXY04). Appreciation is also extended to Sichuan Jinxiang Sairui Chemical Co. Ltd.; Apex Nanotek Co. Ltd.; Ratchadapisek Sompote Fund for Postdoctoral Fellowship (Chulalongkorn University).

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