Poly ( lactic acid ) / thermoplastic starch sheets : effect of adipate esters on the morphological , mechanical and barrier properties

1Department of Food Science and Technology, Universidade Estadual de Londrina – UEL, Londrina, PR, Brazil 2Department of Food Technology, Universidade Tecnológica Federal do Paraná – UTFPR, Campus Londrina, Londrina, PR, Brazil 3Department of Food Engineering, Universidade Estadual de Ponta Grossa – UEPG, Ponta Grossa, PR, Brazil 4Department of Food Science and Technology, Universidade Federal de Santa Catarina – UFSC, Florianópolis, SC, Brazil


Introduction
The interest in producing plastic materials from natural resources is considerably increasing as the need for the reduction of the amount of plastic waste in the environment becomes urgent [1] .Due to its biodegradability, low cost, and worldwide availability, starch in the form of thermoplastic starch (TPS) has been extensively studied as a main component in the production of biodegradable materials.TPS is usually obtained by destroying the crystalline structure of native starch through an extrusion process in the presence of plasticizers, such as glycerol [2][3][4] .
Unfortunately, TPS-based materials are hygroscopic and have limited performance.To overcome this deficiency, the TPS must be blended with another biodegradable polymer to produce materials for packaging and industrial applications [5,6] .In this context, the blend of TPS with poly (lactic acid) (PLA) is promising because, in addition to being compostable, PLA is produced from a renewable resource.PLA has received much attention as the most innovative alternative to conventional petroleum-based polymers and has been intensively studied due to its environmentally friendly characteristics such as biocompatibility, sustainability, and potentially useful physical and mechanical properties.
The addition of TPS into a PLA matrix can also decrease the material cost and increase its biodegradation rate [5,7,8] .
Blending native starch with PLA could increase its rigidity but simultaneously greatly reduce its elongation at break and impact strength [9] .Many factors are believed to cause the deterioration in mechanical properties of PLA/starch blends.First, PLA has an inherent brittleness as evidenced by its relatively low tensile strain at break, and toughness [10,11] .Another explanation is linked to the inherent incompatibility between the hydrophilic starch and the hydrophobic PLA that affects the interfacial adhesion [5,7,[12][13][14][15] .
It is well known that the mechanical and barrier properties of polymer blends are strongly related to their morphology.The relationship of the mechanical and water vapor barrier properties to the morphology of PLA and starch blends has been studied by other authors [16][17][18] .Blends of TPS and PLA can result in a wide variety of morphologies from a dispersed starch/PLA matrix, a co-continuous starch/PLA structure, to inverted phase morphologies, in which the starch is the continuous phase [19] .
Addition of an adequate plasticizer and compatibilizer can improve the processability and flexibility of PLA, enabling the blending with starch to produce sheets with adequate mechanical properties.Another way to enhance the interfacial affinity is the addition of previously gelatinized starch to the blend.The disintegrated starch granules overcome the strong interactions between the starch molecules and, in the presence of water and other plasticizers, leads to good dispersion [20,21] .
In the previous work adipate esters, compared with citrate esters, improved the mechanical properties of TPS/PLA films produced by blown extrusion [22] and TPS/PLA sheets obtained by calendering extrusion, [23] reducing the elastic modulus and the tensile strength and increasing the elongation at rupture.Adipate esters were also studied by other authors [24,25] to enable the production of flexible PLA films.
The current study focuses on the production of sheets with different proportions of TPS and PLA and plasticized with two different adipate esters.The morphologies of these sheets are correlated with the mechanical and barrier properties.
All of the components (PLA, starch, glycerol, and adipate ester) were manually mixed and processed using a pilot twin-screw extruder (BGM, D-20 model, Brazil) with a screw diameter of 20 mm (L/D = 35).The screw speed was 100 rpm, the feed speed was 30 rpm, and a temperature profile from entrance to exit of 100/180/180/180/180°C was used.
The extruded cylindrical profiles obtained were pelletized and extruded again in the same twin-screw extruder coupled with a calender (AX-Plasticos, Brazil) for sheet production.The temperature profile from entrance to exit was 100/170/170/170/175°C, the distance between the calender rolls was 0.8 mm, and the roll speed was adjusted for each formulation to maintain a continuous process.

Thickness
The film thicknesses were measured using a digital micrometer (Starrett, Brazil) at ten different locations on the films.

Mechanical properties
Tensile tests were performed using a texture analyzer (Stable Micro Systems, TA XTplus, England) based on the ASTM D882-02 standard [26] .The specimens were previously conditioned at 23 ± 2°C and 53 ± 2% for 48 hours, and the tensile strength (MPa), elongation at break (%), and Young's modulus (MPa) were measured.The tests were repeated ten times for each formulation.

Water vapor permeability
The water vapor permeability (WVP) of the sheets was determined gravimetrically, according to the ASTM E96-00 standard [27] .The sheets were previously stored at 23 ± 2°C and 53 ± 2% for 48 hours and then fixed in the 60-mm-diameter cylindrical aluminum cells.The interior of the cell was filled with calcium chloride (0% RH), and the device was stored at 25°C in a desiccator containing a saturated sodium chloride solution (75% RH).
The samples were weighed in semi-analytical balance (Marte, Brazil) every 12 h during the 5 days of testing.The mass gain (m) of the cell was plotted as a function of the time (t).The slope of the line was calculated using linear regression (R 2 >0.99), and the water vapor permeation ratio (WVPR) was calculated using Equation 1: where m/t is the angular coefficient of the curve and A is the sample permeation area (m 2 ).The WVP (g/m.day.Pa) was calculated as follows (Equation 2): where e is the mean sample thickness (m), p is the water vapor saturation pressure at the assay temperature (Pa), RH 1 is the relative humidity of the desiccator and RH 2 is the relative humidity in the interior of the permeation cell.
The test was conducted in duplicate.

Morphological characterization
The sheet fracture morphology was evaluated using scanning electron microscopy (Philips, FEI Quanta 200 model, Japan).The sheets were fractured after immersion in liquid nitrogen and gold coated using a sputter coater (Bal-Tec, SCD-050 model, Balzers, Liechtenstein).All of the samples were examined using an accelerating voltage of 20 kV at a magnification of 400x.
A confocal laser scanning microscope (Olympus LEXT, OLS400 model, Germany) was used to obtain surface images of the sheets.

Statistical analysis
The obtained results were evaluated using the analysis of variance (ANOVA), and the treatment means were compared using Tukey's test at a 5% significance level (p<0.05) with Statistica 7.0 software (STATSOFT, USA).

Thickness
It was not possible to produce sheets with PLA/TPS proportions of 20/80 and 10/90, because the extruded materials were too stiff to be processed through the calender rolls.The other formulations had good processability, and the sheets showed slightly rough surfaces.As the starch amount increased, the color became more yellowish.
The thickness of the PLA/TPS sheets with adipate esters added ranged from 431 to 815 µm (Table 1).Considering the effect of TPS addition, the increase in its concentration conferred thicker sheets for PLA plasticized with DIA.In the same way, evaluating the effect of the plasticizer type, the sheets with DIA were less thick than those with DEA for the same TPS/PLA compositions.This result could be related to the DIA plasticizing effect that enhanced the flexibility and allowed increased stretching of the sheets by the calender rolls.During the calendering extrusion process, the sheet thickness was controlled by the distance between the rolls, the speed of the rolls, and the stretching capacity of the formulation, as observed in our previous study [23] .

Morphological characterization
The fracture micrographs of the PLA/TPS sheets examined using SEM at 400x magnification are presented in Figure 1.For the 60/40 and 50/50 PLA/TPS sheets with DEA and DIA, starch granules with different sizes were dispersed in the PLA matrix.When the concentration was reduced to 30/70 PLA/TPS, PLA was present as a dispersed phase in the TPS matrix, evidencing a phase inversion point.According to the literature, [28] the co-continuity occurs at the phase inversion point and when both polymers are approximately in equal concentration.Furthermore, in the case of starch/PLA blends, co-continuity occurs for a low proportion of starch because the starch phase is less viscous and has a high interfacial tension.Generally, for up to 40/60 PLA/TPS, a dispersed starch phase in the matrix morphology was observed.Below this proportion of PLA, a co-continuous morphology was obtained.A similar behavior was also observed in blends of PLA and starch with varied amylose contents [16] and in blends of PLA and chemically modified corn starch [18] .
In the sheets with higher proportions of PLA, some spaces between the PLA and TPS phases were evident, showing the interface between the polymers.This result could be associated with poor interaction and incompatibility between the starch and PLA because of the difference in hydrophobicity, as observed in other reports [7,[12][13][14]17,18,[29][30][31] .
Confocal laser microscopy (CLM) is usually employed to study the surface morphology of polymeric films and sheets and confer information about the surface topography and heterogeneity.In this study, CLM was used to analyze the surface properties of the PLA/TPS sheets plasticized with DEA or DIA, and the obtained images are shown in Figure 2. Sheets with a higher proportion of PLA had increase surface roughness.This result can be attributed to the starch granules dispersed in the PLA matrix, as observed in the fractured SEM images.The roughness decreased as the PLA concentration decreased due phase inversion, in which TPS was the continuous phase.CLM was also used to evaluate the surface roughness of the starch films treated by plasma [1] and to identify the blend compositions that correspond to the beginning of partial continuity until total continuity of the poly(butylene succinate-co-adipate) (PBSA) and TPS phases [32] .

Mechanical properties
The Figure 3 shows the mechanical properties in terms of the tensile strength (MPa), elongation at rupture (%) and Young´s modulus (MPa) of the PLA/TPS sheets plasticized with adipate esters.
As the TPS concentration increased, the tensile strength and Young´s modulus of all of the sheets significantly decreases.For the elongation at break, only the sheets plasticized with DEA exhibited a significant increase when the TPS concentration increased.The sheets plasticized with DIA maintained this property for all of the tested TPS concentrations.For the blends with PLA as a continuous phase, DIA interacted better with it and aided in the induced crystallization due chain alignment during the tensile test.This result caused tensile energy diffusion and led to plastic deformation of the PLA matrix.The statistically equal mechanical properties for the 30/70 PLA/TPS sheets support these results.Different results were observed in our previous study [23] in which DEA with lower molecular weight promoted greater values of elongation of break than DIA.Xiong et al. [21] and Xiong et al. [33] evaluated the effects of plant oils such as Tung oil anhydride and epoxidized soybean oil as plasticizers in PLA/TPS blends containing approximately 60% of PLA and reported a decrease in the tensile strength and an increase in the elongation at break of the blends.
Comparing the effects of the adipate esters as plasticizers, DIA promoted higher values of the elongation at break and the Young´s modulus; however, the same tensile strength values were observed for the two plasticizers.It is possible to conclude that adipate esters with higher molecular weight promoted materials with greater elongation, as observed by other authors [24] that used polyadipate as a plasticizer for PLA films.
As observed in the SEM and CLM images, the starch granules dispersed in the PLA matrix act as defects and lead to serious stress concentration, which the material easier to fracture [21] .The presence of spaces between the starch and PLA also impaired the load transfer under stress.Therefore, the tensile strength and elongation at break of   the materials are directly linked to interfacial adhesion between the polymers used in the blends.Similar results for PLA/starch blends plasticized with acetyl triethyl citrate were reported [12] .
The mechanical properties of the 30/70 PLA/TPS samples were similar for both plasticizers, suggesting that in the case of a dispersed phase (PLA) surrounded by a continuous phase (TPS), the mechanical properties of these immiscible blends depend on those of the continuous phase because it absorbs the stress and energy when the material is under load [28] .When percolation occurs, a part or total of one phase forms a continuous structure that permeates the entire sample and eventually dominates the properties of the blend.In our case, as the TPS levels increased, the interface properties improved, as observed by MEV.As a result, the tensile strength of the sheets was expected to increase; however, this result did not occur because the TPS has a lower tensile strength and Young's modulus than the PLA.
Comparing the obtained results with other results reported in the literature, lower values of the tensile strength, Young's modulus and elongation at break were obtained in PLA/TPS (30 g PLA / 100 g TPS) films produced by extrusion and thermopressing [29,30] In this case, because TPS was the continuous phase, it was responsible for the mechanical properties of the films, resulting in more fragile materials.PLA/TPS films (20% TPS) with increased resistance were produced by Teixeira et al. [34] due the addition of cassava bagasse.

Water vapor permeability (WVP)
The WVP of the PLA/TPS sheets with added adipate esters ranged from 2.94 x 10 -6 to 6.98 x 10 -6 g day -1 m -1 Pa -1 , as shown in Figure 4.The values obtained are lower than those observed for cassava starch films [35,36] and PLA/TPS films [29] and greater than PLA films with synthetic phenolic antioxidants added [37] .The presence of PLA promoted a hydrophobic character of the material and was responsible for decreasing the WVP because the material permeability is strongly influenced by the hydrophobic or hydrophilic nature of its components.
The increase in the TPS concentration (from 60/40 to 30/70) elevated the WVP because of the hydrophilic character of the starch.Furthermore, at high starch contents, water could easily saturate the surface of the blend, penetrate into these voids and be quickly absorbed by the starch, resulting in higher water permeation [16] .
Comparing the type of adipate ester, the sheets with DIA had lower WVP than sheets with DEA, mostly in the materials with a higher proportion of PLA.To support the discussion of the results, the plasticizing capacity of the adipate esters was calculated by dividing the molecular weight by the hydrogen bond capacity of each component, resulting in 9.9 x 10 3 and 4.7 x 10 3 for DEA and DIA, respectively.DEA presented a higher value and consequently a better plasticizing capacity that increased the intermolecular distance between the chains and the free volume, facilitating the penetration of water molecules into the polymer matrix.This fact and the longer chains of the DIA explain the lower WVP values of the DIA-added PLA/TPS sheets.Similar results for PLA/TPS sheets plasticized with DEA were observed in our previous study [23] .
Finally, the final WVP is a balance between the hydrophilic and hydrophobic character of the blend components and the interface properties.The greater the tension at the interface (less compatible blends), the greater the spaces between the phases, which facilitates water diffusion.

Conclusion
Calendering extrusion is a feasible way to produce PLA/TPS sheets at a pilot scale with great potential to be used in packaging production.Different proportions of PLA and TPS in the blends confer sheets with modified morphological structures (dispersed or co-continuous structures), which significantly affect the mechanical and barrier properties.
Adipate esters are excellent plasticizers for PLA in PLA/TPS sheets.DIA, because of its higher molecular weight, was the better plasticizer for PLA conferring sheets with higher elongation at break, tensile strength, Young´s modulus, as well as reduced WPV.

Figure 3 .
Figure 3. Mechanical properties of PLA/TPS sheets plasticized with DEA and DIA.Different letters represent significant difference (p ≤ 0.05) between the means obtained by Tukey's test.

Figure 4 .
Figure 4. Water vapor permeability of PLA/TPS sheets plasticized with DEA and DIA.Different letters represent significant difference (p ≤ 0.05) between the means obtained by Tukey's test.