Printer Friendly

Effects of a diisocyanate compatibilizer on the properties of citric acid modified thermoplastic starch/poly(lactic acid) blends.


Awareness of the environmental problems caused by petroleum-based plastic materials over the last few decades has pushed the polymer researchers to produce eco-friendly plastics. The biodegradable polymers derived from renewable resources are attractive substitutes of those petroleum-based plastics in order to prevent the growth of the solid waste pollution. Poly(lactic acid) (PLA) and starch shine out individually or in the blend form in this emerging field.

PLA, a promising synthetic biodegradable polymer, is derived from agricultural resources through bioconversion and polymerization. It has comparable stiffness, strength, and gas permeability to synthetic polymers obtained from fossil-fuels such as polyethylene, polystyrene, polysthyrene terephthalate) and so on. On the other hand, the applications of PLA are limited due to its brittleness, low toughness, and relatively high price (1). The lower price and completely biodegradable nature of starch makes it possible to be used as filler in PLA. (2). However, native starch is not processable like thermoplastics because of the strong intermolecular and intramolecular hydrogen bonds between starch molecules (3). In the presence of water or other plasticizers such as glycerol or vegetable oils, starch can be converted into thermoplastic starch (TPS) (4). Many efforts have been made to develop a fully biodegradable product from completely renewable resources by blending starch with PLA with an economical way (5-11).

The major problem with PLA/TPS blend system is the poor interfacial interaction between hydrophilic starch granules and hydrophobic PLA matrix (12). Gelatinization of starch is a good method to enhance the interfacial affinity (3). In this method, starch is gelatinized in order to disintegrate granules and overcome the strong interaction of starch molecules in the presence of water and/or other plasticizers, to improve the dispersion in another polymer matrix. However PLA, as a hydrophobic synthetic biopolyester, can undergo hydrolytic depoly-merization in the presence of water at processing temperatures, which results in poor mechanical properties (13). Other commonly applied method reported in literature is the reactive compatibilization to improve the interfacial interactions between PLA and TPS using a reactive component in the blend. Some of the examples of reactive components used to compatibilize the PLA and TSP are peroxide derivatives (14), amylose-g-PLA (14), and maleic anhydride-g-PLA (15-17). In addition to that, highly reactive difunctional isocyanates were successfully used to compatibilize TPS and PLA, since isocyanate groups can interact with--OH groups present on both starch and PLA, and -- OH groups of PLA by forming urethane linkage (18-21).

Plasticization of starch can be obtained by destruction of the granules in the presence of plasticizers by the help of heat and shearing. The commonly used plasticizer for starch is polyols, such as glycerol (22), ethylene glycol (23), and sorbitol (24). It is reported that glycerol plasticized TPS can recrystallize (retrogradate) during storage. In order to prevent retrogradation, some amide-based co-plasticizers such as formamide and acetamide were used (25). However, these chemicals are toxic, and would not be allowed in many food-contact and biomedical applications. Citric acid (CA), already approved by FDA for its use in foods and food contact applications was utilized as a coplasticiser in TPS, since it makes stronger H-bonding with starch as compared with glycerol, and as a result of this, it provides ageing resistant ability even at very low contents (26). On the other hand, CA can also react with starch molecule to form ester bonds, which results in the formation of new carbonyl and ester groups on the starch molecule that can behave like an internal plasticizer of compatibilizer (27).

The CA co-plasticized TPS/PLA blends were previously investigated by some researchers (4), (28). Wang et al. dealt with the influence of CA addition to the glycerol plasticized starch/PLA system (4). The rheological investigations showed that CA decreased the viscosity of both TPS and of TPS/PLA blends, which resulted in a better dispersion by decreasing in the interfacial tension between TPS and PLA, as also proved by scanning electron microscopy (SEM). The better dispersion led to enhanced mechanical properties. In other study, the effect of water on the properties of CA modified TPS/PLA blends were investigated (28). It was shown that when 10 wt% water was used as plasticizer in TPS/PLA blend, the fluidity and plasticization of TPS could be increased dramatically, but the serious depolymerization of PLA and starch deteriorated the mechanical properties of TPS/ PLA blend.

The aim of this study is to investigate the combinatory effects of CA modification and a diisocyanate compatibilizer on the properties of PLA/glycerol plasticized-TPS blends as a function of CA and phenylene diisocyanate (PDI) content. To the best of our knowledge, there is not any published study found in the literature investigating the combinatorial effects of CA and PDI on the properties of PLA/TPS blends. PLA/TPS blends were prepared by melt compounding. A two-step processing was applied in order to prepare blends. At the first step, dried and glycerol-plasticized TPS were processed with different amount of CA in a laboratory scale twin-screw compounder. Then at the second step, these TPSs were blended with PLA at different ratios in the presence of different amounts of PDI. The chemical structures, mechanical, morphological, and thermal properties of blends were investigated.



PLA (2002 D) was purchased from Natureworks Company. Native corn starch was kindly obtained from Cargill, Turkey. Glycerol and CA were purchased from Merck Chemicals. 1,4- PDI, the compatibilizer, was obtained from Sigma-Aldrich.


Sample Preparation. Corn starch was dried at 135 [degrees]C for 4 h under vacuum prior to processing. Glycerol and CA were blended with the dried starch in a laboratory scale co-rotating twin-screw micro-compounder (DSM Xplore 15 ml Microcompounder). The ratio of corn starch/glycerol was kept constant at 70%/30% by weight and the ratio of CA was varied as 0%, 1%, and 5% by weight. The screw speed was 100 rpm and the barrel temperature was 150[degrees]C during compounding. At the end of mixing period of 3 min, the molten TPS was pelletized. The composition of three different types of TPSs was given in Table 1.

TABLE 1. Sample codes and compositions of different TPS.

Sample code  Dried corn starch (wt%)  Glycerol (wt%)  CA (wt%)

TPSO                              70              30         0
TPS1                              70              30         1
TPS5                              70              30         5

PLA was dried at 80[degrees]C for 12 h under vacuum before blending with TPS. TPS and PLA were blended in the twin-screw micro-compounder to obtain blends containing 10-40% TPS by weight. PDI was added to the blends at different ratios of 0.5%, 1%, and 2% by weight. The operating conditions of the micro-compounder were 100 rpm and barrel temperature is 180[degrees]C. At the end of mixing period of 3 min under [N.sub.2] atmosphere, the molten blends were directly injection molded using the transfer cylinder of DSM Xplore 10 ml injection molding machine to obtain impact and tensile test specimens. The injection and holding pressure was set to 10 bars. Melt temperature and mould temperatures were 180 and 30[degrees]C, respectively.

The samples were labeled according to CA content in TPS type, PLA/TPS ratio, and PDI content. For example, "40TPS5_60PLA_0,5PDI" means that the matrix formed by 40 wt% TPS and 60 wt% PLA. TPS used in the blend involves 5 wt% CA and the PDI content is 0.5 wt%.

Characterization. The IR spectra of TPS types and PLA/TPS blends were obtained with a Perkin Elmer Spectrum 100 instrument equipped with ATR unit. Each sample was scanned for 16 times.

The morphology of TPSs and PLA/TPS blends was analyzed by using a low voltage SEM (JEOL JSM-6335F FEG). Injection molded samples for SEM studies were cryogenically fractured in liquid nitrogen. All samples were sputter coated with gold prior to observation. Before the SEM analysis, in order to clearly observe the dispersed phase of TPS, a selective dissolution of TPS was performed with 1 M HCI. The SEM micrographs were processed with an image analysis software (ImageJ Version 1.36) in order to obtain particle size distribution of TPS granules. At least 300 particles were analyzed for each blend to obtain the size distribution and the number average particle size.

Differential scanning calorimeter (DSC) analysis was performed using a Mettler Toledo DSC1 Star System between 0 and 200[degrees]C at a heating rate of 10[degrees]C/min under nitrogen atmosphere. Glass transition temperature, crystallization temperature, and melting temperature values were obtained from DSC thermograms.

The vertical force measurements during compounding were conducted to compare the change in melt viscosity of the blends. The barrel of the laboratory compounder is positioned on a lever, which swivels around a stationary axis and counter balanced by a load-cell at the other end. The load-cell is typically 10 KN in range and measures the axial force exerted by the barrel opposing the pushing forces imposed by the screws toward the bottom while the melt is pumped through the re-circulation channel or die. This force is an indication of the melt viscosity of the material at the given temperature and the screw speed.

Charpy impact strength was determined from notched specimens using a Ceast Resil Impactor according to ISO 180. At least ten samples were tested from each batch and the average values were reported with the standard deviation.

The tensile tests of the samples were conducted with a Lloyd Instruments LRX Plus Machine according to ISO 527. Tensile strength, elastic modulus and elongation at break values were measured. The crosshead speed was 10 mm/min.


Fourier Transformation Infrared Spectroscopy of PLA/TPS Blends

Figure 1a-e shows the Fourier transformation infrared spectroscopy (FTIR) spectra of PLA/TPS blends containing different CA and PDI content. The broad peak seen in Fig. lb is the O--H stretching vibration peak resulting from the -- OH groups of starch [29, 301. It can be seen that the intensity of this peak decreased as CA content increased because of the esterification reaction between starch and CA, which consumes the -OH groups of starch. This reaction was discussed in detailed by Ma et al. (31). It was stated that when CA was heated, it dehydrated to yield an anhydride, which could react with starch to form a starch-citrate derivatives. With the addition of PDI, the intensity of this -- OH peak declined dramatically as compared with the decrease occurred in the presence of CA. This shows that PDI interacted with -- OH groups. The reaction of PDI with PLA and/or starch can result in different product such as, (i) PLA-grafted-starch that yields branching due to the polyfunctionality of starch, (ii) chain-extended of PLA, and (iii) chain-extended starch, also yields branching or crosslinking of starch. In Fig. lc, the peak seen at 2275 [cm.sup.-1], attributed to the -- N=C=O asymmetric stretching vibration (32), is only found in the FTIR spectrum of pure PDI. This peak could not be observed on the FTIR spectra of blends. This also supports that the -- N=C=O groups of PDI consumed during the melt blending. In Fig. id, the new peak at 1577 [cm.sup.-1]], attributed to the -- N--H--stretching vibration of urethane linkage (32) is only found in the FTIR spectra of blends that included PDI. Moreover, the intensity of this peak increased as PDI content increased, as seen in Fig. le. The FTIR results pointed out that a urethane linkage occurred between PDI and hydroxyl groups of either TPS or PLA during melt blending.


Figure 2 presents the typical morphology of cryofractured surfaces of PLA/TPS blends containing 40 wt% TPS and different amounts of CA and PDI. In the lack of neither CA nor PDI in the system (40TPS0_60PLA_OPDI), starch granules did not disperse homogeneously and agglomerated in the matrix by forming coarse two-phase morphology with larger dispersed particles size of starch. The interface between PLA and starch was clearly visible that indicated a very poor interaction. Most of the starch particles were surrounded by a dark ring, which occurred due to the separation of the two-phase during fracture. All these observations point out a poor interfacial adhesion between TPS and PLA because of the difference in hydrophobicity. In the case of only PDI addition at content of 1 wt% (40TPS0_60PLA_ 1PDI), starch granules became more spherical but the dispersed particle size was still large. The starch particles separated from PLA matrix as a result of poor adhesion. When only CA modification was carried out at 5 wt% (40TPS5_60PLA_OPDI), the granular structure of starch was destroyed. It is tough that CA promoted the fragmentation and dissolution of starch granules by acid-hydrolysis, so that the degree of penetration of glycerol between starch chains increased and the degree of plasticization of starch granules increased. The plasticization promoted the lowering of viscosity, and as a result, the morphology of blend became more homogeneous. In addition, the adhesion seems better with respect to previous cases.

As CA modification and PDI addition were carried out together (40TPS5_60PLA_ 1 PDI), the morphology of PLA/TPS blend seemed more homogenous. The size of starch phase decreased and the interface seemed better in comparison with other cases. It is thought that the combination of CA and PDI helped the destruction of granular structure of starch and the interfacial modification by chemical interactions shown by FTIR studies.

In order to clearly observe the change in dispersed particle size of starch by SEM analysis, the starch phase was selectively dissolved by 1 M HCI solution from the cryo-fractured surfaces of the blends. SEM micrographs of selectively dissolved PLA/TPS blends were represented in Fig. 3. The black holes were left after the removal of TPS phase. In order to obtain the particle size distribution an image analysis software (Image J) was used to count and measure the diameter of the phases. As it can be seen that in the absence of either CA or PDI, very large particles of TPS was obtained. Moreover, the addition of PDI alone did not significantly alter the morphology, but the combination of CA and PDI together resulted in a finer morphology. Figure 4 shows the TPS particle size distribution for 40TPS_60PLA blend system with respect to CA and/or PDI content. It is seen from the plot that in the absence of CA, (40TPS0_60PLA_OPDI and 40TPS0_60PLA_IPDI), the size distribution of TPS was broad and ranging from I to 10 microns with an average of approximately 4 microns. When only CA was added, the distribution curve became narrower with an average particle size of approximately 3 microns. The most remarkable change in particle size of TPS was obtained in the case of combination of CA modification and PDI addition. In this case, the distribution curve is the narrowest. The particle size of more than 70% of starch granules lied between 1 and 3 pm with an average of approximately 2 microns. This also proves that the combination of CA and PDI resulted in finer blend morphology.

Thermal Properties

Tables 2 and 3 show the glass transition ([T.sub.s]), cold crystallization ([]), and melting temperatures ([T.sub.m]) of PLA phase in PLA/TPS blends for 10% TPS and 40% TPS, respectively.

The [T.sub.g] of pure PLA was 60-C, and the [T.sub.g] of TPS types was decreasing from 35 to -25[degrees]C with increasing CA content (not shown here). As a general trend, the [T.sub.g] of PLA phase in the blends decreased as PDI content increased from 0 to 2%, independently from TPS and CA content. This result indicated that PDI molecules acted as compatibilizer in the system resulting a possible urethane linkage to form TPS-g-PLA copolymers and as it was known from the literature that in immiscible blends, compatibilization results in shifting of [T.sub.g] values of components toward each other due to the formation of copolymers at the interface. As a general trend, at different PDI loadings, the [T.sub.g] of PLA phase of 10 wt% TPS containing blends exhibited lower values than that of 40 wt% TPS containing blends with the increasing CA content (Tables 2 and 3). It can be speculated based on this observation that PDI mostly consumed between TPS molecules rather than between PLA-TPS. Due to the acid hydrolysis effect of CA in TPS, many--OH groups may possibly form on the TPS (as shown by FTIR) that can act as the reactive sites for PDI. As a result of this, PDI can most probably interact with TPS molecules to form TPS-TPS branching or cross-linking rather than in the reaction forming PLA-g-TPS. As a consequence, in 40% TPS blends, the added amount of PDI is not enough to be consumed in compatibilization reaction, because it is already consumed in TPS. Therefore, in the lack of insufficient compatibilization in 40% TPS, the [T.sub.g] of the PLA phase did not shifted to lower temperatures.

TABLE 2. Thermal transition temperatures of PLA phase in the
blends including 10% TPS.

Sample                     Glass             Cold      Melting
                      transition  crystallization  temperature
                     temperature      temperature   [T.sub.m],
                      [T.sub.g],      [],  [degrees]C)
                     [degrees]C)      [degrees]C)

10TPS0_90PLA_0PDI          57.85           109.50       152.74

10TPS0_90PLA_0.5PDI        56.14           102.04       151.90

10TPS0_90PLA_1PDI          55.16            99.78       151.88

10TPS1_90PLA_0PDI          58.97           111.44       152.65

10TPS1_90PLA_0.5PDI        56.44           102.99       152.17

10TPS1_90PLA_1PDI          55.99           102.41       152.22

10TPS1_90PLA_2PDI          55.14           101.69       152.16

10TPS5_90PLA_0PDI          58.05           112.19       151.94

10TPS5_90PLA_0.5PDI        56.22           105.00       151.35

10TPS5_90PLA_1PDI          54.46           102.63       151.33

10TPS5_90PLA_2PDI          54.11           101.95       151.37

TABLE 3. Thermal transition temperatures of PLA phase in the
blends including 40% TPS.

Sample                     Glass             Cold      Melting
                      transition  crystallization  temperature
                     temperature      temperature   [T.sub.m],
                      [T.sub.g],      [],  [degrees]C)
                     [degrees]C)      [degrees]C)

40TPS0_60PLA_0PDI          59.63           112.82       151.42

40TPS0_60PLA_0.5PDI        57.66           103.99       150.73

40TPS0_60PLA_1PDI          56.23           103.39       150.70

40TPS0_60PLA_2PDI          55.64           102.00       150.32

4OTPSI_6OPLA_0PDI          59.81           113.36       151.30

40TPS1_60PLA_0.5PDI        56.56           106.14       150.54

40TPS1_60PLA_1PDI          56.38           106.00       150.65

40TPS1_60PLA_2PDI          55.37           105.70       150.88

40TPS5_60PLA_0PDI          59.27           119.92       151.42

40TPS5_60PLA_0.5PDI        58.90           113.33       151.45

40TPS5_60PLA_1PDI          57.57           112.95       151.59

40TPS5_60PLA_2PDI          57.10           111.94       151.91

The cold crystallization temperature ([]) of pure PLA was found as 121.0[degrees]C. As a general trend, the [] of PLA phase in blends decreased as PDI content increased. It means that due to the improvement in interfacial interactions in the presence of PDI, the nucleating efficiency of dispersed starch phase improved since necessary conformations in order to crystallize PLA may be possible. The [] of PLA phase in the blends increased as TPS content increased. This can be attributed to the increasing particle size of TPS, which result in a reduction of the interfacial area where crystallization starts. In addition to that, PLA phase was diluted as the TPS content increased, therefore it may be more difficult to diffuse to nucleate the crystals for PLA chains as TPS content increased.

The melting temperature ([T.sub.m]) of pure PLA is 152[degrees]C. As shown in Tables 2 and 3 that the [T.sub.m], of PLA in blends decreased approximately 1-2[degrees]C with the addition of PDI. This phenomenon is known as melting point depression, which happens in the presence of the compatibility in immiscible, crystallizable polymer blends (33-35). This also supports the compatibilization phenomena as PDI content increased. In addition, the decreasing lamellar thickness of the PLA can lead to decreasing melting point, as also observed by the decrease in [] the blends.

Vertical Forces

Steady state vertical force measured during the compounding period is related to the melt viscosity of the polymer blend under same screw speed and temperature as explained in the experimental part in detail. In addition to that one should keep in mind that vertical force is an indication of the melt viscosity that can be used to compare the rheological characteristics of the melts and cannot be directly converted into the viscosity units due the complex geometrical shape of the compounder.

As seen in Fig. 5a, when TPS content was 10 wt%, the vertical forces of blends were not changed significantly as PDI content increased. On the contrary, as shown in Fig. 5b, when TPS content was 40 wt%, the vertical forces of blends increased as PDI content increased. This can be explained by considering the increasing amount of hydroxyl groups with increasing TPS content in the blends; since, hydroxyl groups are reactive sites for PDI, the possibility of occurring PLA-g-TPS or TPS-g-TPS increases, hence the probable branching, chain extension or cross-linking enhanced the average molecular weight and as a consequence, the melt viscosity of the blend increases. Substantially, all these reactions are expected to be occurred also when TPS content is 10 wt%, however, may be due to the low content of TPS, the effect could not be observed.

Impact Properties

The variation of the impact strengths of the selected PLA/TPS blends are shown in Fig. 6a and b. One should note that the impact strength of pure PLA was measured as 9.93 kJ/[m.sup.2]. The impact strength of the blends was found to be higher than neat-PLA regardless of CA or PDI addition. As a general trend, in 10% TPS containing blends (Fig. 6a), increasing CA resulted in an increase in the impact strength, since the addition of CA to the TPS decreased its glass transition temperature and hence the impact energy absorption capacity of the dispersed phase. At given CA contents, maxima were obtained for 0.5% PDI addition in 10% TPS containing blends. Beyond this value, the impact strengths gradually decreased, however they still stayed above the incompatibilized blends (i.e. no PDI). This improved impact strengths by the addition of PDI can also be related to the decreasing TPS phase size. It is known that as the particle size of the dispersed phase reduced, the impact strength of the blends increased (36), (37). A similar trend is obtained in 40% TPS containing blends; however, in this case, the maxima were obtained at 1% PDI in the compatibilized blends. This can be attributed to the requirement of higher amount of compatibilizer as the amount of TPS increased.

In the absence of CA, the maximum impact strength of 12.79 kJ/[m.sup.2] was achieved at a PDI content of 0.5 wt% for 10% TPS containing blends. The improvement with respect to neat PLA was about 30%. On the other hand, the addition of 5% CA to this system results in the highest impact strength of 15.03 kJ/[m.sup.2], corresponding to 50% improvement with respect to neat PLA.

Tensile Properties

The variation of the tensile strengths of the selected PLA/TPS blends are shown in Fig. 7a and b. All the tensile strength values were obtained at the yield point. As the tensile strength of pure PLA that was measured as 65 MPa (not shown on the graphs) is compared with the PLA/TPS blends, it is seen that the blends have lower tensile strength than neat PLA. It was reported in the literature that the tensile strength of TPSs are much lower than the pure PLA (4), (15). For that reason, the tensile strength of all the blends were found to be lower than that of PLA, as can be associated to the "rule of mixtures" by assuming a very good interfacial adhesion.

The tensile strength of blends varied with respect to PDI, CA and TPS concentration in the system. It is seen from Fig. 7a that addition of 0.5% PDI to the 10% TPS containing system resulted in a remarkable improvement in the tensile strength of the blends, but further addition slightly increased the tensile strength for any given CA concentration. The enhanced tensile strength in the presence of PDI can be explained by the retarded premature crack formation phenomena. In the presence of poor interfacial interaction with two-phase polymer systems, cracks formed at the interface between phases in the early stage of tensile testing, which causes the failure of the system. However, if the interfacial adhesion is better, the crack formation is retarded and the sample can withstand higher tensile stresses. In the current case, PDI improved the interfacial adhesion between PLA and TPS in the presence of CA. The variation of the tensile strength with respect to CA content was not significant when TPS content was 10% in the blend.

The tensile strength of 40%TPS/PLA blends was found to be lower than that of 10%TPS/PLA blends due to the higher TPS content, as shown in Fig. 7b. As in the case of 10 wt% TPS system, the tensile strength of the blends improved with the addition of PDI. Different than 10% TPS, the remarkable increase in strength values were obtained at higher PDI loading level possibly due to the requirement of higher amount of compatibilizer to stabilize the PLA/TPS system when TPS amount was increased to 40%. The effect of CA addition is significant in this particular case. In the absence of PDI, the increasing CA content reduced the tensile strength of the blends; however, the addition of PDI to this system improved the tensile strength of the PLA/TPS system. This improvement is highly significant in the case of 5% CA. These results were in harmony with the fact that molecular weight of starch molecules decreased due to CA, yielding a high number of hydroxyl groups, which acted later on as the active sites for interfacial interactions by the help of PDI between two phases.

Figure 8 shows the variation of the elastic modulus of 40% TPS containing PLA/TPS blends as the function of PDI and CA content. It should be noted that the elastic modulus of PLA was found as 1270 MPa. In the first look to Fig. 8, obviously it is seen that the elastic modulus of blends diminished as TPS content increased from 10 to 40% when Fig. 8a and b is compared. The reason of this general observation can be explained by the "rule of mixtures", since TPS's intrinsic modulus was lower than that of PLA; therefore, increasing amount in the blend reduced the elastic modulus of the system.

In the case of 10 wt% TPS, the effect of CA was not significant, but the modulus of the blends increased with the addition of 0.5% PDI. Further addition of PDI did not make a remarkable change in the modulus of the system. When 40% TPS containing blend is considered, it is seen that the modulus decreased with the increasing CA content regardless of PD1 concentration. The addition of PDI, to the PLA/TPS blend system enhanced the elastic modulus. The change in the elastic modulus of the blends can not be directly associated with the interfacial interactions since modulus is measured at very low strains where the interfacial adhesion does not play any role, but the modulus depends on the molecular weight of the system. Here the change in the modulus could be attributed to the molecular weight increase due to possible co-polymerization reactions between PLA and TPS, however it needs further investigations.

Elongation at break values of PLA/TPS blends was shown in Fig. 9a and b as a function of PDI and CA content. It is seen that the elongation at break values of blends increases with the CA content independently from TPS content. When TPS amount was 10%, the jump in elongation at break values of the blends were obtained at 0.5% PDI loading level in the presence of CA. Further addition did not alter the elongation values significantly. In 40 %TPS containing blends, a steady increase in all CA concentrations was obtained with the addition of PDI with in the error limits. Especially in 5% CA loaded PLA/TPS systems, highly tough blends were obtained. This finding is well correlated by the impact test results in terms of toughness of the blends.

Thermogravimetric Analysis

Figure 10 represents the TGA. TPS and some selected 40% TPS containing TPS/PLA blends to compare the effects of CA and PDI addition. It is seen from the TGA curves that PLA has the higher thermal stability in comparison to TPS and PLA/TPS blends. The thermal degradation onset temperature of neat PLA is approximately 290[degrees]C and degradation completes around 400[degrees]C leaving 3% residue. The thermal degradation observed around 100C in TPS is due to the evaporation of the moisture. The weight loss continuing to the degradation onset temperature in TPS can be associated to the volatility of the glycerol, which was the plasticizer. The degradation onset temperature for TPS is around 280[degrees]C and degradation completes around 310[degrees]C by leaving approximately 20% residue. This residual char is the highest among the others. The TGA curves of TPS/PLA blends exhibit one-step degradation. The thermal stability of PLA/TPS blends in the absence of either CA or PDI (40TPS0_60PLA_OPDI) is seen to be improved in comparison to neat TPS, but still lower than the neat PLA. This improvement can be observed from the increased thermal decomposition onset temperature. The %-char residue is between PLA and TPS. This improvement can be associated to the higher thermal stability of PLA chains in comparison to the TPS. By the addition of CA to the PLA/TPS blend, the thermal stability is enhanced possibly due to the improved intermolecular interaction between TPS chains and PLA-TPS chains as a result of improved physical and/or chemical bonding. Further improvement is observed when 2%PDI is added to this blend due to the same reason.


PLA and CA modified or unmodified TPS blends were prepared by melt compounding in order to observe the effect of a diisocyanate-based compatibilizer on the physical properties of the blends. FTIR results showed that PDI acted as a compatibilizer by interacting with hydroxyl groups of PLA and/or starch by forming urethane linkages. It was revealed from SEM micrographs that when only CA modification or PDI addition was carried out, the particle size distribution of TPS particles and the homogeneity of blend did not change remarkably; but on the other hand, when CA modification and PDI addition were carried out together, the homogeneity of the blends, the particle size distribution of TPS phase and the interfacial interactions between two phases were improved. This improvement also affected the mechanical properties of blend, especially the impact strength and elongation at break. Mechanical test results indicated that the toughness of PLA can be improved without too much loosing the strength and modulus by the help of the combinatory use of CA and PDI. The thermal properties such as [T.sub.g] and [T.sub.m] revealed from DSC analysis were in line with the morphological structure of the blends by suggesting the compatibilization phenomena in the presence of PDI and CA together. The incorporation of both CA and PDI improved the thermal stability of the blends. As a general conclusion, the combinatory use of PDI and CA may be utilized to obtain tougher PLA/TPS blends-based materials to overcome the brittleness problem.


The authors are thankful to Cargil Company for providing the corn-starch; and Prof. Dr. Y. Menceloglu from Sabanci University for allowing us to use the impact tester in his labs.

Correspondence to: Guralp Ozkoc; e-mail:

DOI 10.1002/pen.23478

Published online in Wiley Online Library (

[c] 2013 Society of Plastics Engineers


(1.) R. Auras, L. Lim, S.E.M. Selke, and H. Tsuji, Poly(lactic acid)-Synthesis, Structures, Properties, Processing and Application, John Wiley & Sons, Inc., London (2010).

(2.) L.P.B. Janssen and L. Moscicki, Thermoplastic Starch-A Green Material for Various Industries, Wiley-VCH, London (2009).

(3.) R. Shi, Z. Zhang, Q. Liu, Y. Han, L. Zhang, D. Chen, and W. Tian, Carbohydr. Polym., 69, 748 (2007).

(4.) N. Wang, J. Yu, P.R. Chang, and X. Ma, Starch, 59, 409 (2007).

(5.) R.L. Shogren, W.M. Doane, D. Garlotta, J.W. Lawton, and J.L. Willett, Polym. Dc grad. Stab., 79, 405 (2003).

(6.) T.Y. Ke and X.Z. Sun, J. Polym. Environ., 11, 7 (2003).

(7.) N. Wang, Y. Jiugao, and M. Xiaofei, Polym. Inter., 56, 1440 (2007).

(8.) H. Wang, X.Z. Sun, and P. Seib, J. Polym. Environ., 10, 133 (2002).

(9.) J.F. Zhang and X.Z. Sun, Biomacromolecules, 5, 1446 (2004).

(10.) C. S Wu, Macromol. Biosci., 5, 352 (2005).

(11.) L. Chen, X.Y. Qiu, Z.G. Xie, Z.K. Hong, J.R. Sun, X.S. Chen, and X.B. Jing, Carbohydr. Polym., 65, 75 (2006).

(12.) 0. Martin and L. Ave'rous, Polymer, 42, 6209 (2001).

(13.) N. Wang, J. Yu, and X. Ma, Polym. Intern., 56, 1440 (2007).

(14.) T. Ke and X. Sun, J. Appl. Polym. Sci., 81, 3069 (2001).

(15.) E. Schwach, J. Six, and L. Averous, J. Polym. Environ., 16, 286 (2008).

(16.) P. Dubois and R. Narayan, Macromol. Symp., 198, 233 (2003).

(17.) H. Li and M.A. Huneault, J. Appl. Polym. Sci., 119, 2439 (2011).

(18.) H. Li and M.A. Huneault, J. Appl. Polym. Sci., 122, 134 (2011).

(19.) T. Key and X. Sun, J. Appl. Polym. Sci., 88, 2947 (2003).

(20.) H. Wang, X. Sun, and P. Seib, J. Appl. Polym. Sci., 82, 1761 (2001).

(21.) L. Yu, K. Dean, Q. Yuan, L. Chen, and X. Zhang, J. Appl. Polym. Sci., 103, 812 (2007).

(22.) C.L. Jun, J. Polym. Environ., 8, 33 (2000).

(23.) M.L. Fishman, D.R. Coffin, R.P. Konstance, and C.I. Onwulata, Carbohydr. Polym., 41, 317 (2000).

(24.) A.L.M. Smits, M. Wubbenhorst, P.H. Kruiskamp, J.J.G. van Soest, J.F.G. Vliegenthart, and J. van Turnhout, J. Phys. Chem. B, 105, 5630 (2011).

(25.) L. Wang, R.L. Shogren, and C. Carriere, Polym. Eng. Sci., 40, 499 (2000).

(26.) X.F. Ma and J.G. Yu, Starch, 56, 545 (2004).

(27.) J.G. Yu, N. Wang, and X.F. Ma, Starch/Starke, 57, 494 (2005).

(28.) R. Shi, Z. Zhang, Q. Liu, Y. Han, L. Zhang, D. Chen, and W. Tian, Carbohydr. Polym., 69, 748 (2007).

(29.) N. Wang, X. Zhang, N. Han, and J. Fang, J. Thermoplast. Comp. Mater., 23, 19 (2010).

(30.) K.M. Ardakani, A.H. Navarchian, and F. Sadeghi, Carbohydr. Polym., 79, 547 (2010).

(31.) X. Ma, P.R. Chang, J. Yu, and M. Stumborg, Carbohydr. Polym., 75, 1 (2009).

(32.) V.H. Orozco, W. Brostow, W. Chonkaew, and B.L. Lopez, Macromol. Symp., 277, 69 (2009).

(33.) M. Oaten and M.N.R. Choudhury, Macromolecules, 38, 6392 (2005).

(34.) B.P. Rimt and J.P. Runt, Macromolecules, 17, 1520 (1984).

(35.) L.B. Robeson, Polym. Blends A Comp. Rev., Hanser, Munich (2007).

(36.) S. Park and S. Choc, Macromol. Res., 13, 4 (2005).

(37.) W. Jiang, H. Liang, J. Zhang, D. He, and B. Jiang, J. Appl. Polym. Sci., 58, 537 (1995).

Serkan Karagoz, Guralp Ozkoc

Department of Chemical Engineering, Kocaeli University, 41380 lzmit, Kocaeli, Turkey
COPYRIGHT 2013 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Karagoz, Serkan; Ozkoc, Guralp
Publication:Polymer Engineering and Science
Article Type:Report
Geographic Code:7TURK
Date:Oct 1, 2013
Previous Article:Real-time diagnosis of gas-assisted hot embossing process by ultrasound.
Next Article:Direct functionalization with 3,5-substituted benzoic acids of multiwalled carbon nanotube/epoxy composites.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters