Recyclability Analysis of Starch Thermoplastic/Almond Shell Biocomposite

This article is focused on studying the effect of the reprocessing cycles on the mechanical, thermal, and aesthetic properties of a biocomposite. This process is based on starch thermoplastic polymer (TPS) filled with 20 wt% almond shell powder (ASP) and epoxidized linseed oil (ELO) as a compatibilizing additive. To do so, the biocomposite was prepared in a twin-screw extruder, molded by injection, and characterized in terms of its mechanical, thermal, and visual properties (according to CieLab) and the melt flow index (MFI). The analyses carried out were tensile, flexural, Charpy impact tests, differential scanning calorimetry (DSC), thermogravimetric analysis (TGA). The effects of the reprocessing were also studied for the biodegradable unfilled TPS polymer. The results showed that TPS and TPS/ASP biocomposite suffer changes progressively on the properties studied after each reprocessing cycle. Furthermore, it was observed that the addition of ASP intensified these effects regarding TPS. However, in spite of the progressive degradation in both cases, it is technically feasible to reprocess the material at least three times without needing to incorporate virgin material.


Introduction
Recently, there has been a considerable interest in the development of biobased polymers to decrease dependency on petroleum-based polymers due to environmental concerns. Biopolymers derived from renewable resources have a wide range of applications in different industries due to their specific characteristics. As a result of researches and advances in R&D, the properties of these materials have been improved, and this ushered in new markets ranging from packaging, food service, consumer electronics, automotive, agriculture/horticulture, and toys to textiles. Although packaging, either rigid or flexible, remains the largest field of application for these materials, with almost 53 percent (1.14 million tons) of the total bioplastics market in 2019 [1]. Despite this interest, there are serious doubts about the behavior of these plastics during processing. One of them is the possibility of reprocessing the material generated by defective production or cut parts (sprues, distribution channels, etc.).
Traditionally, the reprocessing of the thermoplastic materials led to a significant deterioration of the material properties. This deterioration was the result of the decrease in molecular weight caused by the breakage of the polymeric chains that occurs when the material is subjected to a high shear process. On the other hand, loss of mechanical and thermal properties as well as discolorations are some of the most common degradation problems in polymers as a result of the recycling process. Recycled polymers must possess a set of minimum performance characteristics to meet specific requirements after the recycling process. Therefore, it is of great interest to know the impact of successive parts. The most optimal performance was attained for biocomposites with ELO or ESBO between 10 and 20 phr.
Taking as a reference one of the above-mentioned optimal formulations, the main objective of this present study is to study the influence of reprocessing cycles on the properties of TPS and TPS/ASP biocomposite in terms of their mechanical and thermal properties and changes in morphology, visual appearance, and melt flow index.
This study shows the degradation of TPS and TPS/ASP and establishes possible strategies for the reuse of the discarded parts generated during the manufacturing process with these materials.

Materials
For this study, a commercially available starch-based polymer, Mater-Bi ® EI51N0 of Novamont, was studied. This bio-based and biodegradable polymer has a melt flow index (MFI) of 19 g/10 min (190 • C/2.16 kg) and a density of 1240 kg/m 3 (data provided by Novamont). This grade was selected for its properties, which are quite similar to polypropylene (PP), of high use in injection molding. Table 1 shows some properties of the as-received material. Almond shell powder (ASP) with particle size between 0.05 and 0.125 mm provided by Hermen Europe, S.L. (Spain) was used for this study (Figure 1a). It consisted of a mixture of different almond shell varieties. Figure 1b shows that the predominant particle size range was about 0.08-0.125 mm. Table 2 shows the content of fixed carbon, volatile matter, humidity, ash, and the chemical composition of the main components of the used almond shell: hemicellulose, lignin, and cellulose, determined by thermogravimetric analysis (TGA) in previous work [10].  The compatibilization of starch polymer and ASP was carried out with epoxidized linseed oil (ELO), supplied by Traquisa S.L. (Spain). Some of its properties are shown in Table 3. Before processing, TPS and ASP were dried separately for 4 h at 60 • C and 24 h at 105 • C, respectively, in a SLW 115STD INOX air-circulating oven from POL-EKO (WodzisławŚląski, Poland) to minimize its moisture and to avoid hydrolytic reactions [11].
A composite of starch-based polymer with ASP was developed using a BERSTORFF ZE26 × 44D-BASIC co-rotating twin-screw extruder (26:44 L/D). Before feeding the material into the extruder, a manual pre-mixing of the different components, Mater-Bi ® EI51N0/ASP/ELO, was carried out and fed into the extruder through the main hopper. The ratio of TPS/ASP was set at 20 wt% since this content has shown balanced mechanical properties and appealing aesthetics in a previous study dealing with TPS biocomposites [12]. ELO was added at 10 parts per hundred resin (phr) of biocomposite. The temperature profile was set as follows: 130-185-185-185-185 • C (from feeding zone to die). The rotating speed was 80 rpm. The extruded material was finally pelletized using an air knife.

Injection Molding
Testing samples were molded using an injection molding machine Ergotech 110-430h/310V from DEMAG (Demag, Germany). The injection conditions used to develop recycled TPS and biocomposite TPS/AASP test specimens are listed in Table 4. After each injection cycle, the material was milled and dried at 60 • C for 6 h. Then, TPS was processed a total of four cycles and TPS/ASP biocomposite six cycles. The difference in the number of processing cycles between TPS and TPS/ASP was due to the poor quality of the specimens after TPS-4. TPS-5 presented injection defects, such as shrinkage. The nomenclature of recycled TPS and TPS/ASP biocomposite is given in Table 5. Specimens were conditioned at a temperature of 23 • C and relative humidity of 50% for at least 16 h before testing.

Color Measurements
The influence of the reprocessing cycles on the color of the TPS and TPS/SP biocomposite specimens was recorded with a CR-200 Chroma Meter from KONICA MINOLTA (Tokyo, Japan). Moreover, the color indexes (L*, a* and b*) were measured according to the following criteria: L* = 0, darkness; L* = 100, lightness; + a* = red, − a* = Green and + b* = yellow, − b* = blue. From these coordinates, it was possible to determine the color difference associated with this space. The color variation, ∆E * ab , was obtained by the following Equation (1) and compared with the color coordinates of the neat TPS (TPS-1) and TPS/ASP (TPS/ASP-1).

Mechanical Properties
The mechanical properties of TPS and TPS/ASP biocomposites were determined to study the capacity of their reprocessability.
Tensile and flexural tests were performed using an 6025 universal testing machine from INSTRON (Canton, Massachusetts, USA) with 5 kN power sensors. The tensile test was performed according to ISO 527 standard, using a crosshead speed of 1 mm/min for Young's Modulus determination and 5 mm/min for tensile and elongation at break determination. The extensometer used was MTS 634.11F-54. Recorded values included ultimate tensile strength (UTS), Young's modulus, and strain at break. A total of 5 specimens from each material were tested using standardized samples 1A (dogbone).
Impact tests were performed with a Resil 5.5 impact testing device (CEAST RE-SILIMPACTOR) with a 1 Joule hammer. Then, test samples were cut and tested according to ISO 179 standard. A total of 10 sample tests from each material was tested.

Scanning Electron Microscopy (SEM)
The impact fracture surface obtained after a Charpy impact test was analyzed using a Jeol JSM-840 SEM system. Samples were gold-coated before analysis, and the energy of the electron beam was 20 kV.

Thermal Analysis
The main thermal degradation parameters of biocomposites, initial degradation temperature (T onset ), and maximum mass loss rate temperature (Tmax) were studied by TGA using a TA Instrument Q500 thermogravimetric analyzer from TA INSTRUMENTS (Delaware (New Castle, DE, USA)). Then, those samples with an average weight between 8 and 10 mg were placed in standard platinum crucibles of 70 µL. In this case, biocomposites were subjected to the following temperature program: from 30 to 600 • C under nitrogen (N 2 ) atmosphere at a rate of 10 • C/min, and from 600 to 1000 • C under oxygen (O 2 ) atmosphere at a rate of 10 • C/min with a purge gas flow of 10 mL/min.
Later on, thermal transitions of developed biocomposites were studied by differential scanning calorimetry (DSC) in a DSC Q200 calorimeter from TA INSTRUMENTS (New Castle, DE, USA). Then, those samples with an average weight of 8-10 mg were placed in standard aluminum crucibles and subjected to a three-step program that consisted of an initial heating cycle from 0 • C to 250 • C at a rate of 10 • C/min to remove thermal history, followed by cooling to −20 • C at a rate of 5 • C/min, and a second heating cycle to 350 • C at a rate of 10 • C/min. All the tests were run in an N 2 atmosphere with a constant flow of 50 mL/min. Finally, thermal transitions were determined: temperature (T c ) and enthalpy (∆H c ) of crystallization after the cooling cycle and temperature (T f ) and enthalpy of fusion (∆H f ) after the second heating.

Melt Flow Index
The determination of the melt flow index (MFI) was carried out according to ISO 1133 standard with a MFI TWELVINDEX equipment from ASTFAAR (Milan, Italy) at 190 • C and 2.16 kg. The cut time between two consecutive measurements was 15 s.

Visual Aspect
The appearance of the injected specimens of TPS and TPS/ASP after the different reprocessing cycles are gathered in Figure 2a,b, respectively. Table 6 summarizes the color indexes (L*, a* and b*), and the color variation measured by ∆E * ab , with respect to TPS and TPS/ASP biocomposite with only the first processing cycle (TPS-1 and TPS/ASP-1, respectively). The reprocessing cycles had more effect on the change in color of TPS/ASP biocomposite than in the unfilled TPS. As expected, the L * value decreased (white to black) in both cases. a* and b* coordinates changed progressively after each reprocessing cycle. Moreover, after the second processing cycle, the obtained ∆E * ab value of TPS was 1.00 (TPS-2). After the third reprocessing cycle, the value slightly increased to 1.91 (TPS-3); only an experienced observer could notice the difference in color [13]. However, in the case of TPS/ASP, after the second processing cycle, the value obtained of ∆E * ab was 3.96 (TPS/ASP-2), which resulted in samples whose change in color could be noticed by an observer (3.5 < ∆E * ab < 5). The color variation clearly showed an increasing tendency in reprocessing, as expected due to degradation [13]. The ∆E * ab was more remarkable and noticeable for TPS/ASP biocomposite.  Table 7 shows the main thermal degradation parameters, initial degradation temperature (T onset ), and maximum mass loss rate temperature (T max1 and T max2 ) obtained by TGA. Figures 3 and 4 show the TGA curves of TPS and TPS/ASP biocomposite, respectively, obtained after different reprocessing cycles. Furthermore, all samples presented a two-step process because the thermogravimetric analysis was carried out under an N 2 and O 2 atmosphere. When it comes to the comparative TGA curves of TPS samples obtained after different reprocessing cycles, those suggested no significant changes in the thermal degradation parameters since the curves overlapped. TPS-1 presented moderate thermal stability with T onset , and T max1 of 326.1 • C and 362.1, respectively. It is noticeable that, after the fourth injection cycle, the values remained almost invariable. This indicated that the reprocessing cycles did not have a significant effect on thermal degradation at the current processing temperatures. Regarding the TGA thermograms of TPS/ASP, the addition of ASP reduced the thermal stability of the biocomposite because almond shells degrade faster than polymer matrix [10,12], and T onset and T max1 moved towards a lower temperature. T max2 remained practically unchanged. However, in a previous study, which is in the process of publication, it was found that the addition of 10 phr ELO had a positive effect on the thermal stability of this type of biocomposite, increasing T onset and T max1 by 6 • C: TPS/ASP-1 biocomposite presented a T onset of 315.0 • C and T max1 of 346.0 • C. After the second injection cycle, T onset decreased by 9 • C and continued to decrease progressively after each reprocessing cycle. Nevertheless, the T max1 remained almost invariable until the fifth injection cycle (340.6 • C). Table 7. Thermal properties of the injection-molded samples of TPS and TPS/ASP biocomposite subjected to different reprocessing cycles obtained by thermogravimetric analysis (TGA).  To obtain the main thermal transition of the material a differential scanning calorimetry was used. Figures 5 and 6 show the DSC curves achieved from the cooling and second heating scans of TPS and TPS/ASP biocomposite samples, respectively, obtained after different reprocessing cycles. In addition, Table 8 summarizes the most relevant thermal parameters obtained from the cooling and second heating of the samples subjected to different reprocessing cycles. Bastioli et al. [14] described Novamont's starch-based technology as a process that could destroy amylose and amylopectin from starch. Thus, the main endothermic (T m1 y T m2 ) and exothermic (T c ) peaks were related to the melting and the cold crystallization of the crystalline structure of the synthetic biodegradable polymer presented in Mater-Bi, respectively. After the first injection run (TPS-1), TPS presented a bimodal endothermic peak (Figure 5b). The first one, smaller, at around 161.4 • C (T m1 ), and the second one, more pronounced, at 168.5 • C (T m2 ). These are associated with the fusion of the crystalline structure as a result of chain scission caused by different resistance to heat. The TPS-1 thermogram corresponding to the cooling scan presented a unique main exothermic peak around 109.5 • C (Figure 5a). As can be observed, after different reprocessing cycles, the thermal transitions of TPS, such as T c , T m , and their respective enthalpy, did not present significant changes. The addition of 20 wt% produced a slight reduction in all parameters. After the first injection run, TPS/ASP-1 also presented a bimodal endothermic peak. The melting point temperature, T m1 , decreased from 161.4 • C to 157.1 • C and the T m2 from 168.5 • C to 166.4 • C. Regarding the normalized melting enthalpy (∆H m ), it decreased from 23.8 to 17.2 J/g. These results indicated that the addition of almond shell to the starch-based polymer decreases the crystallization of the molecular chains [12]. After the second injection cycle (TPS/ASP-2), T m1 and T m2 decreased 2-3 • C compared to TPS/ASP-1 and continued to decrease progressively after each reprocessing cycle until 144.6 • C and 162.7 • C (Figure 6b), respectively. The decrease in the values may be attributed to the higher mobility of the polymer chains as a result of the reduction in the molecular weight during the recycling process. In addition, it was noticed that the crystallization temperature (T c ) slightly decreased upon adding ASP, thus indicating that the presence of ASP made crystallization start later, compared to as-received TPS. Furthermore, it was noticed that the crystallization temperature and enthalpy of TPS/ASP decreased progressively after each reprocessing cycle (Figure 6b). This indicated a loss of crystalline structure and a major difficulty in the crystallization process (i.e., starting at a lower temperature) as the number of processing cycles was increased [15].     Table 8. Thermal properties of the injection-molded samples of TPS and TPS/ASP biocomposite subjected to different reprocessing cycles obtained by differential scanning calorimetry (DSC).  Table 9 shows a summary of the tensile properties of both TPS and TPS/ASP biocomposite after each reprocessing cycle. The TPS-1 presented Young's modulus of 1658 MPa and tensile strength of 39.4 MPa. As can be seen in Table 9, the number of reprocessing cycles influenced the tensile properties, decreasing the values progressively. In particular, after the third processing cycle, Young's modulus significantly decreased to 1184 MPa. Regarding the tensile and elongation at break, changes were more noticeable after the fourth processing. That indicated that the material was less ductile. unfilled TPS. The addition of ELO increased the flexibility of the biocomposite, reaching values of Young's modulus even lower than those of the as-received TPS. Thus, the tensile and elongation of the TPS/ASP biocomposite decreased due to its progressive lack of capacity to sustain deformation. When it comes to the effect of reprocessing TPS/ASP biocomposite in tensile properties, it did not have a significant effect on Young's modulus, which indicated similar rigidity. The rest of the parameters (tensile and elongation strength/break) kept similar values to TPS/ASP-1 until the fourth reprocessing.  Figure 7 shows the evolution of impact strength in both the reprocessed TPS and TPS/ASP biocomposite. The impact tests of TPS were carried out with notched samples because the material presented a high strength and did not break, even after the applied reprocessing cycles. The as-received TPS (TPS-1) presented an impact strength of 6.0 kJ/m 2 . This value barely decreased after the second injection cycle (TPS-2). After the third and fourth injection cycle, a slight reduction in energy absorption capacity was attained, reaching values around 4.5 kJ/m 2 . The reprocessing cycles did not have a significant effect on the impact strength. This behavior was observed in other grades of starch-based polymer (Solanyl, [16]). The addition of 20 wt% of ASP drastically affected the impact strength. While the TPS samples needed notching to do the test, the TPS/ASP biocomposite did not. The TPS/ASP biocomposite showed an impact strength of 14.13 kJ/m 2 . After the second reprocessing cycle, the impact strength decreased until 12.26 kJ/m 2 . Additionally, an evident reduction in energy absorption capacity was attained when increasing the reprocessing cycles, reaching values down to 9.70 kJ/m 2 (TPS/ASP-6). This behavior was observed in PLA systems with cellulose fibers [17][18][19]. Finally, this loss in the impact strength can be attributed to the polymer degradation occurring during each thermo-mechanical recycle. M ), tensile strength at break (σ R ), elongation at break ( unfilled TPS. The addition of ELO increased the values of Young's modulus even lower than tho and elongation of the TPS/ASP biocomposite de pacity to sustain deformation. When it comes t composite in tensile properties, it did not have which indicated similar rigidity. The rest of strength/break) kept similar values to TPS/ASP-  Figure 7 shows the evolution of impact st TPS/ASP biocomposite. The impact tests of TPS because the material presented a high strength reprocessing cycles. The as-received TPS (TPS kJ/m 2 . This value barely decreased after the seco and fourth injection cycle, a slight reduction in reaching values around 4.5 kJ/m 2 . The reprocess on the impact strength. This behavior was obser mer (Solanyl, [16]). The addition of 20 wt% of AS While the TPS samples needed notching to do th The TPS/ASP biocomposite showed an impact reprocessing cycle, the impact strength decrease dent reduction in energy absorption capacity w cessing cycles, reaching values down to 9.70 k served in PLA systems with cellulose fibers strength can be attributed to the polymer degra chanical recycle. R ), of TPS and TPS/ASP biocomposite subjected to different reprocessing cycles.

Samples
Young's Modulus (MPa) σ M (MPa) maximum strength, tensile, and elongation at brea be observed in Table 8, Young's modulus of TPS/ unfilled TPS. The addition of ELO increased the fl values of Young's modulus even lower than those o and elongation of the TPS/ASP biocomposite decre pacity to sustain deformation. When it comes to th composite in tensile properties, it did not have a s which indicated similar rigidity. The rest of the strength/break) kept similar values to TPS/ASP-1 u  Figure 7 shows the evolution of impact stren TPS/ASP biocomposite. The impact tests of TPS w because the material presented a high strength and reprocessing cycles. The as-received TPS (TPS-1) kJ/m 2 . This value barely decreased after the second and fourth injection cycle, a slight reduction in ene reaching values around 4.5 kJ/m 2 . The reprocessing on the impact strength. This behavior was observe mer (Solanyl, [16]). The addition of 20 wt% of ASP d While the TPS samples needed notching to do the t The TPS/ASP biocomposite showed an impact str reprocessing cycle, the impact strength decreased dent reduction in energy absorption capacity wa cessing cycles, reaching values down to 9.70 kJ/m served in PLA systems with cellulose fibers [17 strength can be attributed to the polymer degradat chanical recycle.

M (%)
σ R (MPa) maximum strength, tensil be observed in Table 8, Y unfilled TPS. The additio values of Young's modulu and elongation of the TPS pacity to sustain deforma composite in tensile prop which indicated similar strength/break) kept simil  Figure 7 shows the e TPS/ASP biocomposite. T because the material pres reprocessing cycles. The kJ/m 2 . This value barely d and fourth injection cycle reaching values around 4. on the impact strength. Th mer (Solanyl, [16]). The ad While the TPS samples ne The TPS/ASP biocomposi reprocessing cycle, the im dent reduction in energy cessing cycles, reaching v served in PLA systems strength can be attributed chanical recycle. In a previous study, the addition of 20 wt% of ASP produced an increment in Young's Modulus and a reduction in the rest of the tensile parameters (tensile and elongation at maximum strength, tensile, and elongation at break) [12]. However, in this work, as can be observed in Table 8, Young's modulus of TPS/ASP biocomposite was lower than the unfilled TPS. The addition of ELO increased the flexibility of the biocomposite, reaching values of Young's modulus even lower than those of the as-received TPS. Thus, the tensile and elongation of the TPS/ASP biocomposite decreased due to its progressive lack of capacity to sustain deformation. When it comes to the effect of reprocessing TPS/ASP biocomposite in tensile properties, it did not have a significant effect on Young's modulus, which indicated similar rigidity. The rest of the parameters (tensile and elongation strength/break) kept similar values to TPS/ASP-1 until the fourth reprocessing. Figure 7 shows the evolution of impact strength in both the reprocessed TPS and TPS/ASP biocomposite. The impact tests of TPS were carried out with notched samples because the material presented a high strength and did not break, even after the applied reprocessing cycles. The as-received TPS (TPS-1) presented an impact strength of 6.0 kJ/m 2 . This value barely decreased after the second injection cycle (TPS-2). After the third and fourth injection cycle, a slight reduction in energy absorption capacity was attained, reaching values around 4.5 kJ/m 2 . The reprocessing cycles did not have a significant effect on the impact strength. This behavior was observed in other grades of starch-based polymer (Solanyl, [16]). The addition of 20 wt% of ASP drastically affected the impact strength. While the TPS samples needed notching to do the test, the TPS/ASP biocomposite did not. The TPS/ASP biocomposite showed an impact strength of 14.13 kJ/m 2 . After the second reprocessing cycle, the impact strength decreased until 12.26 kJ/m 2 . Additionally, an evident reduction in energy absorption capacity was attained when increasing the reprocessing cycles, reaching values down to 9.70 kJ/m 2 (TPS/ASP-6). This behavior was observed in PLA systems with cellulose fibers [17][18][19]. Finally, this loss in the impact strength can be attributed to the polymer degradation occurring during each thermo-mechanical recycle.

R (%)
The fracture of the surface after impact was studied by SEM analysis. TPS surface after the first injection cycle (Figure 8a) showed less roughness than after the fourth injection cycle (TPS/ASP-4) (Figure 8b). This increase in the surface roughness can be directly attributed to the degradation of the TPS matrix. SEM micrographs of the impact fracture surfaces of TPS/ASP biocomposite gave qualitative information about the dispersion of almond shell particles. The fracture surfaces of TPS/ASP after the first injection cycle (Figure 8c) showed that the almond shell was homogeneously distributed in the polymer matrix. As with the TPS surface, it could be observed that the surface roughness and the imperfections on the impact fracture surface increased after the fourth injection cycle (TPS/ASP-4) (Figure 8d).

Melt Flow Index
It is a well-known fact that polymer chain breaking due to polymer degradation results in reduced molecular weight and increased flowability with temperature [20]. MFI is a simple test that can potentially offer detailed information about the degradation a polymer material has undergone [5,20]. Figure 9 shows the evolution of the MFI values after different reprocessing cycles. As can be seen, the MFI suffered a drastic increment in the value in both TPS and TPS/ASP biocomposite from the second injection process (TPS-2 and TPS/ASP-2). This effect was observed in other grades of commercial Mater-Bi and Mater-Bi YI014U/C, in which MFI increased from 9 to 16 g/10 min after the second processing [7]. TPS/ASP-1 biocomposite presented an MFI value higher than TPS-1, probably

Melt Flow Index
It is a well-known fact that polymer chain breaking due to polymer degradation results in reduced molecular weight and increased flowability with temperature [20]. MFI is a simple test that can potentially offer detailed information about the degradation a polymer material has undergone [5,20]. Figure 9 shows the evolution of the MFI values after different reprocessing cycles. As can be seen, the MFI suffered a drastic increment in the value in both TPS and TPS/ASP biocomposite from the second injection process (TPS-2 and TPS/ASP-2). This effect was observed in other grades of commercial Mater-Bi and Mater-Bi YI014U/C, in which MFI increased from 9 to 16 g/10 min after the second processing [7]. TPS/ASP-1 biocomposite presented an MFI value higher than TPS-1, probably because TPS/ASP was previously subjected to a thermal process to obtain the pellets. However, despite the evident degradation of the material according to the MFI values obtained, this was not reflected in the decrease in the mechanical and thermal properties of the material.

Conclusions
This work describes the influence of the recycling, by injection molding reprocessing, of TPS and TPS/ASP biocomposite samples. To do so, the visual aspect, mechanical and thermal properties, and melt flow index were studied. TPS was processed four times and TPS/ASP biocomposite six times.
Experimental results revealed that reprocessing cycles had more effect on the TPS/ASP biocomposite. The change in color of TPS was 2.9 after all four cycles, meaning that only an experienced observer could notice the difference in color. However, after only two injection cycles, the change in color of TPS/ASP biocomposite was higher than unfilled TPS, and an observer could notice the change in color (∆E * ab > 5). Regarding the thermal properties, TGA and DSC analyses showed that after four injection cycles of TPS (TPS-4), the values remained almost invariable compared to TPS-1. This indicated that the reprocessing cycles did not have a significant effect on thermal degradation and main thermal transitions. On the other hand, in the case of TPS/ASP, in the second injection cycle, a significant reduction in T onset (9 • C) was produced and to a lesser extent in the melting temperature (T m1 and T m2 decreased in 2-3 • C respect to TPS/ASP-1). The values continued to decrease progressively after each reprocessing cycle.
Moreover, it was observed that the reprocessing cycles had an influence on the mechanical properties of the TPS after the third processing cycle. In particular, Young's modulus was reduced, indicating a loss of rigidity. Furthermore, after the fourth processing (TPS-4), the material lost ductility. In regard to the TPS/ASP biocomposite, the reprocessing did not have significant effects on tensile properties. The results maintained similar values to TPS/ASP-1 until the sixth cycle. As for impact strength, a slight reduction in energy absorption capacity was obtained after the second processing cycle (TPS-2), and no significant variations on the impact strength were observed with more cycles. However, TPS/ASP biocomposite showed an evident reduction in energy absorption capacity after the second processing cycle and continued when increasing reprocessing cycles. This confirmed the negative effect of biocomposite reprocessing on toughness.
Finally, the MFI showed a noticeable increase in both TPS and TPS/ASP biocomposites. Despite this, it was not reflected in such a drastic decrease in the mechanical and thermal properties of the material.
As a general conclusion, this study revealed that the TPS/ASP biocomposite is more sensitive than TPS to recycling regarding mechanical and thermal properties and visual aspects. From an industrial point of view, TPS could be reprocessed at least four times, as shown in this work, without the need of adding a virgin material. The TPS/ASP biocomposite could be recycled up to two or three cycles, but the impact strength has to be taken into account if it is a critical property for the product considered as it is the most affected property.

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

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