Thermal, morphological, and structural characterization of starch-based bio-polymers for melt spinnability

2.2.1 Plasticization and extrusion of starch

Conversion of starch into thermoplastic material was done using a co-rotating twin screw melt extruder available in the Institute for Circular Economy of Bio:Polymers at Hof University (ibp), Germany. The plasticizer glycerol was used for the plasticization of starch to convert it into thermoplastic material. The proportion of glycerol used was 28.6% w/w (starch dry mass basis). Starch powder and plasticizer of predetermined proportion were premixed and kept for 1 day before extrusion to enhance the diffusion of plasticizer into starch molecules. In case of TPS prepared with wax addition, predetermined amount of starch and Licocare^® RBW 102 wax powder (1.5% based on starch mass) were dry-mixed, then a predetermined amount of glycerol was added, and mixing took place. The premixing was done using a kitchen mixer (BOSCH MUM4, Germany) where the predetermined amount of starch and plasticizer was mixed for about 1 h using a medium speed level of the mixer.

The melt extrusion was done on co-rotating twin screw extruder from Lab Tech Engineering Co., Ltd, Thailand, having a screw length-to-diameter ratio (L/D) of 44 and a screw diameter of 20 mm.

The premixed starch and glycerol with or without wax is fed into the hopper of the twin-screw extruder and subsequently melt-extruded. The melt-extruded strands were air cooled, cut, and pelletized using the associated granulator on the twin screw extruder.

2.2.2 Blending of TPS with PLA

The TPS pellets were dried for 6 h at 60°C by using a LUXOR CAS compressed air dryer (Motan Holding GmbH, Konstanz, Germany) before melt mixing with PLA. The PLA pellets were also dried before melt mixing for 6 h at 80°C using the above-mentioned dryer. Following the drying process, TPS/PLA pellets with a mass ratio of 70/30, respectively, were fed into the hopper of the twin-screw extruder and were melt-extruded. Finally, the extrudates from the blends were air-cooled, pelletized, and used for melt spinning as well as for further analyses.

Figure 1 shows an illustration of the process used for melt blending of TPS and PLA compounds.

Figure 1

Illustration of melt blending process of the compounds.

2.2.3 Blend proportions and extrusion parameters for neat TPS and TPS/PLA compounds on twin screw melt extruder

Blend proportions, recipe preparation, and process parameters for TPS neat and TPS/PLA blend preparation on twin screw extruder are given in Table 1.

The samples are coded as neat TPS, which represents TPS from native starch and glycerol, TPS/PLA, which represents a blend of TPS/PLA (70/30), and TPSw/PLA, which represents TPS/PLA (70/30) blends where TPS is prepared with wax addition.

2.2.4 Melt spinning

Since melt spinning does not require a solvent, as is the case with wet spinning and electro-spinning, it is considered an environmental friendly method of fiber spinning (33).

The current study is aimed at melt spinning of fibers based on native tapioca starch. The melt spinning of TPS-based compounds which are prepared by melt extrusion using a twin-screw extruder was done on the monofilament piston type melt spinning device available at Leibniz-Institut fuer Polymerforschung e.V. (Dresden), Germany.

The piston-type melt spinning device has a single piston barrel and is equipped with a diameter (D) of the capillary hole of 0.3 mm, and the ratio of the capillary length L to the capillary diameter (L/D) being 2. The pre-dried polymer samples (for 12 h at 40°C) were filled and melted in the heated cylinder with varying temperatures for the different sample compounds. Subsequently, the melted materials were forced by a piston through a single-hole die. A winder was used to collect the melt as-spun monofilaments with varying take-up speeds.

The material composition and parameters for melt spinning of fibers are given in Table 2.

2.3.1 Thermal characterization

Thermal characterization of the samples was done with the aim of finding thermal transitions occurring and also finding out the thermal stability and degradation behavior of the compounds.

Thermal characterization of TPS-based compounds was done using the DSC 214 Polyma instrument from NETZSCH-Gerätebau GmbH and the thermogravimetric analysis (TGA) TG 209 F3 instrument also from NETZSCH-Gerätebau GmbH, Germany. TGA was carried out to evaluate the degradation characteristics of the TPS samples and TPS/PLA blend samples. For DSC measurement, a sample mass of 6–10 mg was used with a nitrogen gas atmosphere with a gas flow rate of 40 ml·min⁻¹. The samples were heated at a constant heating rate of 10 K·min⁻¹ within a temperature range of −80°C to 200°C for the first heating cycle followed by a cooling cycle from 200°C to −80°C and a second heating cycle from −80°C to 200°C.

In case of TGA, sample mass from 6 to 10 mg was used for measurement and a nitrogen gas atmosphere with a gas flow rate of 20 ml·min⁻¹ was adjusted. The samples were heated at a constant heating rate of 10 K·min⁻¹ in a temperature range of 35–500°C.

2.3.2 Morphology characterization

2.3.3 MFI measurement

MFI or melt flow rate (MFR) is a measure of a given polymers’ flow characteristics also known as the rheological properties in the molten state under a known pressure and temperature. In another way, it is a measure of the ease of flow of thermoplastic polymer melt. It can also be defined as the weight of polymer in grams flowing in 10 min through a die of a specific diameter with the application of pressure by a predetermined weight at a given temperature.

The MFI measurement was conducted on an MFI device (Karg Industrietechnik) according to the ISO 1133 standard test method using an applied load of 5 kg and a temperature of 160°C with a preheating time of 300 s.

2.3.4 Structural characterization

3.2.1 Flow characteristics

The MFR or MFI has important practical significance for polymer processing and their field of applications.

The MFR measurement results for pure TPS and TPS/PLA blends and neat commercial PLA polymers are given along with their error values in Figure 3.

Figure 3

MFR values for compounds prepared.

The results from the MFR measurement in Figure 3 show that neat TPS has an MFR value of 19.5 g per 10 min indicating the conversion of native starch into melt-flowing material with relatively sufficient melt flow character. Starch during heat treatment can undergo a transition from crystalline to amorphous structure and subsequent recrystallization when cooled and during storage. The addition of glycerol as plasticizer may have reduced starch crystallinity and intermolecular forces of starch improving mobility of starch and facilitating its conversion into amorphous material. At room temperature, glycerol interacts mainly with the amorphous regions of the starch (27). The interaction between starch and glycerol is shown in Figure 4. The relatively high MFR value of the TPSw/PLA (70/30) blend as compared to neat TPS, TPS/PLA (70/30) blends, and neat PLA may be due to the plasticization effect of glycerol on to PLA chains reducing intermolecular force in between PLA chains because of better interaction between TPS and PLA in the presence of wax, which in turn may indicate its effect as dispersion aid. Waxes have different physicochemical properties according to their sources and chemical composition enhancing the properties of wax-based composites and blends (42). The presence of interactions between components can alter the viscoelastic behavior of the PLA component because of the plasticization effect of wax on PLA resulting in further flexibility of the blends and favoring spinnability during melt spinning as well (43).

Figure 4

Interaction between starch and glycerol (re-produced with permission by (25), copyright 2024, John Wiley and Sons).

In the case of TPS/PLA (70/30) blends, the formation of physical network structure between TPS and PLA with large microstructure size due to relatively poor miscibility between the blends may have reduced the free volume and, hence, reduced the flow property due to higher interphase interaction which is also consistent with morphology analysis of the blends. In this study, higher content of TPS in the blends, i.e., 70/30 TPS/PLA, was used with the intention of evaluating melt spinnability of blends with higher TPS content. The proposed interaction between starch and glycerol during thermo-plasticization is given in Figure 4.

3.2.2 Morphology of pellets used for melt spinning

According to Ribeiro et al. (44), native starch exists in granular form during AFM observation without granule destruction. The AFM 3D image (Figure 5) for native cassava/tapioca starch as observed by them indicates the existence of large starch granules during AFM surface observation and during SEM observation according to others.

Figure 5

Native cassava starch surface SEM image (left) and AFM 3D image (right) (44,45).

The results for microscopic images of pressed film samples obtained from AFM for the neat TPS, TPS/PLA (70/30 w/w), and TPSw/PLA (70/30 w/w) samples, as defined in Table 1, are shown in Figure 6.

Figure 6

AFM images (size 2.5 µm²) of (i) neat TPS, (ii) TPS/PLA (70/30), and (iii) TPSw/PLA (70/30); (a) phase image, (b) amplitude, and (c) topography.

During thermo-plasticization of starch, granular starch is converted into a homogeneous material, TPS, comprising a hard-elastic network and soft amorphous regions.

Figures 6(i, a) and 7(i, b) and (i, c) show the results of the phase image, amplitude, and topography, respectively, for neat TPS as observed from AFM image. As we can see from the images, granular crystalline starch is converted to homogeneous amorphous material comprising starch-rich and plasticizer-rich regions (13) as visible with darker and brighter phase areas. The previous research by Werner et al. (46) reported that the darker phase regions indicate soft regions and, therefore, in the phase image of Figure 6(i, a), the dark regions indicate plasticizer-rich regions that are softer.

Along starch transformation, the granular structure is disrupted, releasing mainly amylose chains because amylopectin has more intra- and intermolecular hydrogen bonds and less surface area, making it difficult to disrupt (47). Several physicochemical processes take place during starch thermo-plasticization such as fragmentation, de-structuration, plasticization, and melting. Consequently, TPS is constituted of a highly plasticized amorphous phase and a crystalline region associated with the remaining starch granule as reported by Castillo et al. (13).

In this research work, as indicated by the arrows in the phase and amplitude images in Figures 6(i, a) and 7(i, b), some starch granules are not fully de-structured, and they remain as granular crystalline structures in microscopic images indicating agreement with the above-mentioned previous researcher’s reports. The topography in Figure 6(i, c) also indicates rough surface of pressed films indicating a similar result to that of amplitude and phase images.

It is important to highlight that optimal processing conditions allow obtaining a high proportion of the TPS plasticized phase. More plasticization due to the dispersion effect provides a more plasticized phase and reduces the crystalline phase, associated with remaining or unplasticized granules, hence improving the performance of the resulting TPS in later stages.

The AFM image (size 2.5 μm²) results for TPS/PLA (70/30) blends are shown in Figure 6(ii).

The phase and amplitude images in Figure 6(ii, a) and (ii, b) indicate that there is a kind of phase separation between TPS and PLA phases. The PLA exists as a dispersed phase in TPS matrix with a large microstructure phase size indicating poor miscibility among the blends due to their thermodynamically immiscible nature. Here, the TPS-rich matrix in which PLA-rich domains appear randomly dispersed is evidenced by larger particle size, indicating a lack of miscibility. According to previous research works, existence of large particle domains in the blends indicates poor interphase interaction and poor miscibility between blends (48). The topography image in Figure 6(ii, c) shows surface roughness more evidenced in certain regions. This indicates that low miscibility among the components generates a microstructure with little cohesion between the phases. This poor miscibility and weak cohesion may be the reason for the weak mechanical property and poor performance of polymer pellets during melt spinning. These results are in accordance with previous AFM studies on Mater-Bi starch bioplastic (49) and others (46).

The AFM image results for TPSw/PLA (70/30) are shown in Figure 6(iii).

In the AFM phase image of Figure 6(iii, a), we can observe that there is relatively a homogeneous distribution of blend components with less microstructure size. The PLA is distributed as a dispersed phase with fine microstructure size with difficult to identify it as it is a minor phase, which may be due to relatively better dispersion (50) among the blends as compared to TPS/PLA (70/30). The finely dispersed and uniformly distributed PLA domains in the TPS matrix may give better cohesion and hence better overall properties which was also evidenced by relatively better melt spinnability of this compound. More interestingly, the amplitude image in Figure 6(iii, b) does not clearly show any large edge height difference, which was clearly visible in other blends, confirming a good distribution in this blend. The topography image in Figure 6(iii, c) also shows a smooth surface as compared to the other blends. This relatively better dispersion between the compounds could be due to wax addition during TPS preparation as a dispersion aid; however, the wax is not clearly visible in the AFM images observed due to its very low content of 1.5%. The higher cohesion achieved between phases in TPSw/PLA (70/30) resulted in relatively better melt spinnability of this compound as seen from melt spun fiber samples.

3.2.3 Thermal characterization

DSC has been widely used to characterize the phase transition of starch, such as gelatinization, retrogradation, and glass transition (T _g). The thermal transitions like glass transition T _g and melting temperature T _m of neat TPS and TPS/PLA blends were determined using DSC.

The DSC first and second heating curve for native starch is given in Figure 7.

Figure 7

DSC, first, and second heating curve for native tapioca starch,.

From the diagram in Figure 7, we can see that the first heating DSC curve for native tapioca starch shows a broad endothermic reaction which begins with onset temperature T _onset around 72°C and reaches its maximum peak value T _p at 123°C. This transition is attributed to the gelatinization of starch, which is in accord with the literature (41,51). The gelatinization enthalpy was found to be 200.4 J·g⁻¹ with broad area. The gelatinization process was facilitated by the presence of moisture in starch granules. This result agrees with previous work by Zuo et al. (52) who have found the maximum transitional peak to be 129°C, which is nearly similar to the result obtained in the present work. The second heating curve of native starch does not show any thermal transitions indicating its amorphous behavior after gelatinization during the first heating which is an expected result as the starch is converted to thermoplastic material possessing amorphous nature. In addition, T _g of fully gelatinized starch is difficult to determine using DSC analysis because the heat capacity change is quite low at the T _g and this result agrees with previous works (53).

During the gelatinization process, semi-crystalline starch granules undergo a change of state from an ordered to a completely or partially disordered structure (54). According to Ning et al., a combination of DSC with X-ray diffraction and small-angle X-ray scattering has shown that residual crystallinity and lamellar organization still remain after the gelatinization process (38).

The gelatinization of starch is not a sharp transition in the DSC curve. Instead, it appears as a large endothermic reaction due to the polydispersity of the granules (55) and a similar result is obtained in this work as well.

The DSC second heating curves for neat TPS, neat PLA, TPS/PLA, and TPSw/PLA blends are plotted in Figure 8. The corresponding characteristic thermal values like glass transition, crystallization, and melting temperatures are shown in Table 4.

Figure 8

DSC, second heating curves for neat TPS, neat PLA, and TPS/PLA (70/30) and TPSw/PLA (70/30).

From the DSC curve for neat TPS in Figure 8, a small peak with an onset temperature at −0.5°C and ending at the temperature of about 30.5°C is observed which may indicate the glass transition range of TPS polymer. Glass transition T _g is the temperature at which bounded amylose and amylopectin were loosened to lead a significant movement of the starch molecules (29). This agrees with the work from other researchers for example, according to work by Bergo et al. (56), the glass transition temperature (T _g) of cassava starch films was decreased from 131.9°C to 42.1°C, when the glycerol content was increased from 0 to 45% w/w, respectively, and nearly similar result is obtained in our case. This may be an indication of the successful conversion of native starch into amorphous thermoplastic material.

During gelatinization, the plasticizer molecules eliminate starch–starch interactions and replace them with starch-plasticizer interactions, increasing the mobility of polymeric chains (57). There is no clear melting peak observed in our result, which also is in accord with work from other researchers for TPS polymers (29). The transformation of granular starch is influenced by process conditions such as temperature and plasticizer content (57).

Form the DSC curve for TPS/PLA 70/30, we can see that there are three endothermic peaks observed at a temperature of around 61.3°C, 155.6°C, and 161°C. The first peak at 61°C is attributed to the glass transition temperature T _g of the amorphous phase of PLA in the blend. The endothermic peaks at 155.6°C and 161°C are associated with the melting of TPS and PLA phases in the blend indicating phase separation. However, these values are lower than neat PLA melting temperature T _m of 165°C and T _g of 64°C (from first heating curve not shown). This may be due to constraints induced by starch in crystallization of PLA, leading to thinner lamellae than the ones produced by pure PLA (51). The melting temperature of neat PLA from the second heating curve was found to be 161°C indicating reduction in T _m as compared to the first heating curve and this is related to slow crystallization behavior of PLA during cooling. The lowering in T _g and T _m in case of TPS/PLA (70/30) may be a result of the plasticizer glycerol increasing segmental mobility in the polymeric chains as a result of interaction between TPS and PLA during melt blending. The exothermic peak found at the temperature of 113.4°C refers to the cold crystallization temperature of the blend polymer, indicating that TPS can facilitate PLA crystallization since this crystallization peak was not observed in the case of neat PLA.

From the DSC curve for TPSw/PLA 70/30, we can see four endothermic peaks observed at a temperature of 62.5°C, 81°C, 154.7°C, and 160.2°C. The first endothermic peak at 62.5°C corresponds to the T _g of the blend where the amorphous phase of PLA is dominating (3) and this value is higher than T _g for TPS/PLA (70/70) blend, assuring better miscibility in this case. The second endothermic peak at around 81°C is attributed to the melting of Licocare^® RBW 102 wax used in TPS preparation. The overlapping peaks at 154.7°C and 160.2°C are attributed to melting of the crystals associated with starch and PLA phase, respectively. These melting temperature values are lower than the T _m of pure PLA which is 165°C. This may be due to constraints induced by starch in the crystallization of PLA, leading to thinner lamellae than the ones produced by pure PLA (51). The slightly lower T _m and slightly higher T _g values of TPSw/PLA 70/30 as compared to TPS/PLA 70/30 indicate relatively better dispersion in this case. The exothermic peak found at the temperature of about 111.5°C refers to cold crystallization temperature of the blend polymers. The reduction in cold crystallization temperature in this case may be due to an increase in chain mobility associated with better TPS chain interaction with PLA increasing its rate of crystallization.

However, the presence of double melting peaks may indicate phase separation and poor miscibility between polymers in all blends, which may also be reason for poor melt spinnability associated with all blend polymers.

The results of thermal degradation and thermal stability for thermoplastic starch TPS from tapioca starch and blends of TPS/PLA are presented in plot curves of weight loss (%) of granulates as a function of temperature as shown in the diagrams of Figure 9 and Table 5. The onset of degradation T _onset (°C), maximum degradation temperature or peak temperature T _max (°C), 5% mass loss T _5%, and 80% mass loss T _80% are given in Table 5.

Figure 9

TGA (a) and DTG (b) curve for neat TPS, neat PLA, TPS/PLA (70/30), and TPSw/PLA (70/30).

From the TGA thermogram for neat TPS (Figure 9a), we can see a gradual weight loss with temperature rising from room temperature up to 123°C and this is mainly due to moisture/loosely bound water removal (1,29,58).

The weight loss between 123°C and 296°C continues probably due to glycerol volatilization (reaching the step of starch decomposition which begins at about 296°C). The decomposition of starch occurs at the temperature of about 296°C as indicated by onset temperature and ends at the temperature of about 334°C. This phenomenon might be ascribed to the removal of hydrogen functional groups, degradation, and de-polymerization of the starch carbon chains in accordance with Tarique et al. (29). Furthermore, TGA analysis of neat TPS showed a gradual weight loss with the single-step polymer chain degradation and this means that TPS exists as a single phase (41). This finding is consistent with the AFM result from Figure 6i, which indicated only some crystalline starch granules without phase separation. The T _5% and T _80% mass loss for neat TPS occurred at temperature around 215°C and 334.7°C, respectively. The residual mass at a temperature of 497°C of the analyzed TPS was found to be 11.42%. The residual mass might be inorganic impurities or organic matters that could stand high temperatures. The result obtained by TGA in this study is in accordance with previous works (29,40,59,60).

From the TGA curve for TPS/PLA (70/30) in Figure 9a, we can see that there is a three-step degradation associated with mass loss with the temperature of nearly up to 295°C, from 295°C to nearly 352°C and above 352°C. The first mass loss in the temperature range of up to 296°C occurs due to removal of plasticizer, water, and other volatile compounds, and it was also observed during melt spinning of these polymers. The degradation of polymer starts at a temperature of around 295°C as can be seen from onset temperature T _onset and ends at maximum temperature of 352°C. This indicates relatively good thermal stability of TPS/PLA blends as compared to neat TPS where maximum degradation occurred at a temperature of 334°C. The T _5% and T _80% mass loss occurred at a temperature of 265 and 416.7°C, respectively, which is a higher temperature as compared to neat TPS for the same amount of mass loss. The residual mass at a temperature of 497°C of the analyzed TPS/PLA (70/30) was found to be 18.3%. As compared to neat TPS, TPS/PLA (70/30) blend polymer pellets have higher thermal stability.

From the TGA curve for TPSw/PLA (70/30) blend (Figure 9a), we can see that the degradation of polymer begins at a temperature of around 296°C and ends at maximum temperature of about 345°C. The data reveal that the onset temperature for TPSw/PLA (70/30) compounds is nearly similar to the neat TPS and TPS/PLA (70/30), but maximum degradation which is at (345°C) occurs at a temperature lower than that of TPS/PLA (70/30) (352°C) and higher temperature than that for neat TPS (333°C). The T _5% and T _80% mass loss occurred at a temperature of 193 and 371°C, respectively. This value is higher than the values for neat TPS and lower than the values for TPS/PLA (70/30). The addition of wax as dispersion agent and lubricant reduced the thermal stability of polymers as miscibility is slightly better and the TPS phase is dominating.

Thermal stability of TPS is highly improved by the addition of PLA as it can be seen from all blend compounds analyzed.

Table 5 shows TGA thermal parameters for onset of degradation T _onset, temperature for 5% mass loss (T _5%), temperature for 80% mass loss (T _80%), and residual mass for neat TPS, PLA, and TPS/PLA blend compounds.

Generally, from the TGA curves for neat TPS, PLA, TPS/PLA (70/30), and TPSw/PLA (70/30) we can observe that addition of PLA improved thermal stability of TPS and thermal degradation for blends starts nearly at the same temperature, which is associated to starch phase degradation and the degradation behavior is dominated by TPS phase due to its high proportion in the case of blends. However, the thermal stability of TPS/PLA blends is better than that of neat TPS which is probably due to higher thermal stability behavior of PLA. The data also reveal that slightly better interaction between TPS and PLA can influence the thermal stability and maximum degradation temperature of TPS/PLA blends.

3.2.4 Structural characterization

In this study, the ATR spectra were used to investigate the possible differences between spectra of starch and starch-based blends, which could indicate a change in chemical structure upon blending and plasticization.

The ATR spectra for native tapioca starch, neat PLA, neat TPS, TPS/PLA (70/30), and TPSw/PLA (70/30) are shown in Figure 10.

Figure 10

ATR spectra for native starch, neat TPS, neat PLA, TPS/PLA (70/30), and TPSw/PLA (70/30).

From the spectra for native starch, we can see that there is a broad peak at wavenumber 3,282 cm⁻¹, which is attributed to O–H stretching (39,61,62). Other associated peaks observed here can be referred from Table 3 which indicates the peaks and associated bonds for starch. These results of native starch structural characterizations are used for comparative analysis of peak changes in starch after plasticization and blending with PLA.

The disruption of granular organization during starch thermal extrusion induces modifications in morphological characteristics, crystalline architecture, and chemical structure of native starch (13).

The ATR spectrum for neat TPS presents the typical absorption bands corresponding to the functional groups of starch and glycerol. The broad peak occurring at a wavenumber of 3,315 cm⁻¹ is attributed to O–H stretching resulting from starch and glycerol, and this peak is shifted to higher wavenumbers as compared to the corresponding peak of native starch at 3,282 cm⁻¹ indicating the effect of starch plasticization.

According to Paluch et al. (63), the shift of characteristic bands in TPS compared to native starch observed demonstrates the formation of more stable bonds and stronger interactions in the TPS between the –OH groups of the starch and plasticizer, compared to the bonds between OH groups in native starch chains. The interaction between starch and plasticizer can also be seen from peak shift of C–H stretching from 2,889 cm⁻¹ (in native starch) to 2,927 cm⁻¹ in TPS due to plasticization effect from glycerol. The small peak at 1,645 cm⁻¹ is attributed to O–H bending associated with bound water or hydroxyl groups in starch (64).

The peak bands at 1,150 cm⁻¹ are attributed to C–O–C asymmetric stretching and the peak band at 1,016 cm⁻¹ is attributed to C–O stretching of starch molecules.

The representative bands of the glycerol plasticizer were displayed at 2,884 cm⁻¹ (C–H) and 1,430 cm⁻¹ corresponding to weak signal of CH₂ symmetric deformation associated with glycerol.

According to Abdullah et al. (39), who studied different types of starches (corn, cassava, potato), all starches can possess OH, C–H, C–O–C, and C–O functional groups in wavenumber range of 3,300–3,600, ∼2,900, ∼1,150, and 1,000–1,100 cm⁻¹, respectively, and peak at 926 cm⁻¹ which corresponds to the C–O stretching of pyranose rings or C–O–C ring vibration of carbohydrate (38,39,40).

From the ATR spectra for TPS/PLA (70/30) in Figure 10, we can see that there is a wavenumber shift toward higher values in the region of O–H and C–H stretching, in case of TPS/PLA blends due to interaction between TPS and PLA. The new peak at 1,753 cm⁻¹ is attributed to carbonyl (C═O) stretching in the PLA phase as this peak is not found in case of neat TPS and found in neat PLA at 1,747 cm⁻¹. Another peak shift compared to neat TPS spectra, is the peak at 1,647 cm⁻¹ which was at 1,645 cm⁻¹ in neat TPS corresponding to O–H bending. The C–H bending peak shifted toward a lower wavenumber value of 1,361 cm⁻¹ in case of TPS/PLA as compared to neat TPS which is 1,366 cm⁻¹ and the C–O stretching peak shifted toward a higher value from 1,016 cm⁻¹ in neat TPS to 1,018 cm⁻¹ in TPS/PLA indicating an interaction between the blends.

From the ATR spectra for TPSw/PLA (70/30) in Figure 10 we can see there is peak shift at C–H region to lower value (2,920 cm⁻¹) compared to TPS/PLA (70/30) indicating interaction or miscibility between TPS and PLA due dominance of TPS phase because of its higher proportion in the blend. This may be due to the addition of wax, which acted as dispersion agent facilitating miscibility between the blends. The high intensity peak at 1,751 cm⁻¹ is attributed to carbonyl stretching in PLA phase and it shows downward shift from 1,753 cm⁻¹ in case of TPS/PLA (70/30), indicating relatively better interaction between TPS and PLA in the presence of wax. Another peak that appeared at 1,646 cm⁻¹ corresponds to O–H bending and it shifted to lower values as compared to neat TPS/PLA (70/30). The peak band at 1,079 cm⁻¹ is attributed to C–O–C asymmetric stretching in PLA and the band at 1,020 cm⁻¹ is attributed to C–O stretching. The C–O stretching bands shifted to higher value from 1,018 cm⁻¹ in case of TPS/PLA (70/30) to 1,020 cm⁻¹ in case of TPSw/PLA (70/30), indicating better interaction between TPS and PLA, which is also supported by AFM morphology images.

Generally, the FTIR analysis results indicate peak changes of native starch, associated to energy absorptions indicating the effect of starch plasticization and melt blending. The peak shifts due to plasticization and melt blending indicate the interactions between starch-glycerol and TPS-PLA, which is also in alignment with the results of morphology analysis.