No distinguishable melt crystallization peak was detected for neat PLA, PLA/EVA80, and PLA/ EVA80/1wt% TiO2 samples while cooling from a melt even at 10 K min−1, indicating that PLA chain crystallization is entirely impeded at this rate. However, a melt crystallization peak was observed for PLA/1 wt% TiO2 and occurred at 97.6 °C. These results are consistent with TiO2 nanoparticles, which have been reported to be an efficient nucleating agent for PLA9,11,38. On subsequent heating, an exothermic cold crystallization peak is detected for PLA-based samples, as shown in Fig. 1. Figure 1-a shows that the cold crystallization peak of PLA became sharper and shifted to a lower value with the incorporation of EVA 80 into the PLA matrix. This can be elucidated by the fact that EVA80 made PLA macromolecular chains more mobile, generating favorable conditions for PLA crystallization, as reported before15,24,39,40. Another possible reason is that a heterogeneous structure between PLA and EVA80 accelerates the PLA’s crystallization process, as Tien et al.41 mentioned. Figure 1-b shows the DSC second heating run of PLA: EVA80:1 wt% TiO2 The inclusion of 1 wt% TiO2 in PLA/EVA 80 caused the cold crystallization peak to shift to a lower temperature, as illustrated in Fig. 1-b and Table 1. Moreover, the cold crystallization peak became sharper, narrower, and stronger than PLAEVA 80, as shown in Fig. 1b. These results indicated that TiO2 nanoparticles act as an efficient nucleating agent. A narrow, sharp, single melting peak was observed for PLA: 1 wt% TiO2 and PLA: EVA 80: 1 wt% TiO2. EVA 80 content and TiO2 nanoparticles slightly influenced the maximum melting temperature, as shown in Fig. 1. A similar trend was reported for PLA/nano-bio composites42 and PLA/PHO/Talc6. The heat of crystallization depends on the thermal history of the polymer. It is also vital to know the percent of crystallinity since it substantially affects the physical properties of the polymer. The degree of crystallinity of PLA, PLA/EVA80, and PLA/EVA80/1 wt% TiO2 composites was determined using Eq. (1) and presented in Table 1.Figure 1DSC second heating curves for (a) PLA: EVA80; (b) PLA: EVA80:TiO2.Table 1 Thermal parameters of PLA/EV80 and PLA/EVA80/1 wt.% TiO2 blends from DSC (cool and second heating) curves.The presence of TiO2 and EVA 80 significantly increases the degree of crystallinity of PLA, as shown in Table 1.Modulated DSC runs were conducted to investigate the miscibility between PLA and EVA 80, as shown in Fig. 2. It is well known that glass transition temperature is accompanied by a step change in the ‘reversing heat capacity and a peak in the phase signal. The values of glass transition temperatures of pure PLA and pure EVA 80 are 60 °C and 0 °C, respectively, as shown in Fig. 2 and Table 2. Two glass transition temperatures are observed for PLA: EVA80: 1 wt% TiO2. The position of the glass transition temperature of PLA changes slightly and shifts to a lower value (around 4 °C) with the addition of EVA80, indicating the plasticization effect of EVA80. The depression of cold crystallization temperature and a slight lowering of PLA’s modulated glass transition temperature may indicate the strong compatibility of PLA and EVA80. Several authors have reported that the vinyl acetate content influenced the compatibility and phase morphology of the PLA/EVA blends in the random copolymers25,29,43. The phase separation between the two immiscible components has been reported for EVA, with VA content of 18–70%. However, EVA with VA content 85 and 90 have been reported to be entirely miscible with PLA29,33.Figure 2TMDSC curves in the glass transition region of the PLA: EVA80: 1 wt% TiO2; (a) reversing Cp; (b) phase angle.Table 2 Glass transition temperature from modulated DSC, Peak is taken from a maximum of heat flow phase (rad).Scanning electron microscopyFigure 3 illustrates SEM images for PLA/1 wt%TiO2 and PLA/EVA80/1 wt% TiO2.Figure 3SEM micrographs for PLA: EVA80: 1 wt% TiO2.SEM images reveal that the spherical shape of TiO2 nanoparticles is randomly distributed in the PLA matrix. The fracture surface of PLA:TiO2 (100:0:1) shows the features of a very brittle material. The phase morphology of a polymer blend is mainly influenced by its miscibility. The roughness of the surface decreases as the EVA 80 content. No discrete domains of EVA 80 and many elongated fibrils were observed, especially for 50 wt% EVA 80. The elongated shape was formed as a result of the soft EVA 80 phase flowing into the PLA hard phase during the sample’s fabrication. These results reflect the compatibility of PLA and EVA80 in the blend. It is worth noting that PLA and EVA became thermodynamically miscible when the vinyl acetate content in copolymer was increased to 85 wt%, and no phase separation occurs in compatible blends like PLA/PVAC and PLA/EVA 90 as reported in Refs.27,33. The presence of EVA’s polar vinyl acetate groups to interact with PLA enough accounts for PLA and EVA’s compatibility, as reported in Ref.43, which confirms the shift of glass transition temperature of the PLA phase in the blend.Polarized optical microscopyFigure 4 shows the size and morphology of PLA spherulites after isothermal crystallization at 125 °C for 90 min. When comparing the PLA/EVA 80 with the neat PLA, adding EVA 80 decreased spherulite size and increased their number. PLA’s spherulitic structure and spherulitic boundaries in the blends are not as sharp as in the neat PLA. When the EVA80 content increases, it can be seen that the spherulite morphology changes significantly. This is because the copolymer enters intraspherulitic regions, causing the crystal superstructure to lose perfection (non-birefringent zones). This has been seen in several blends containing partially miscible or immiscible components44. The spherulite texture in blends with a lower EVA80 content (10 wt%) is almost identical to neat PLA. However, in higher EVA80 concentrations (30 wt% and 50 wt%), the texture mainly became irregular and open with diffused borders, showing that copolymer molecules were partially included into the amorphous zones between the crystalline lamellae of PLA and incorporated into the spherulites. The findings showed that EVA 80 inclusion could increase the nucleation density of PLA and a strong indication of compatibility between PLA and EVA 80. Therefore, one can expect an improvement in mechanical properties upon adding EVA80. Hoch et al. reported that the blown film of a blend of PLA with EVA 80 (80:20) exhibited better mechanical properties than that of immiscible PLA/EVA 6045. It was previously mentioned that TiO2 acted as a nucleation site for PLA11,38. In the case of PLA/EVA 80/1 wt% TiO2, increasing crystal nuclei results in smaller and denser spherulites.Figure 4polarized optical micrograph for PLA: EVA80:TiO2 after isothermal crystallized at 125 °C for 90 min.Figure 5 displays the TG and DTG thermograms of neat PLA, EVA 80, PLA/1 wt% TiO2 and PLA/EVA 80 loaded with 1 wt% TiO2. Neat PLA showed only one loss decomposition stage within the 300–350 °C temperature range ascribed to the chain scission of the PLA into several products, including lactides, cyclic oligomers, acetaldehydes, carbon monoxides, carbon dioxide, and acrylic acids46,47. As summarized in the data in Table 3, the onset decomposition (Tonset) of this loss was found to be at 267 °C with a temperature of maximum mass loss rate (Tp) at 341 °C. For PLA/nano-TiO2 thermal stability enhancement occurred, as evidenced by the increase of Tonset by 10 °C which agrees with the previous report10. Native EVA 80 exhibits two stages of decomposition; the first one started at 314 °C with a maximum of 346 due to deacetylation of the pendant acetate groups of EVA, while the second stage of decomposition with a maximum temperature at 461 is attributed to thermal scission of the backbone methylenic chain48,49. For PLA/ EVA80/ nano-TiO2 blends, the TGA profile exhibits two distinct stages of thermal decomposition. The initial mass loss observed between temperatures of 300–380 °C is attributed to the disintegration of the PLA component and the deacetylation of EVA. The second mass loss took place between 380 and 460 due to the decomposition of the main chain scission of the main backbone of the polyethylene and deacetylated EVA phases and releasing unsaturated butene and ethylene compounds in gaseous form. A comparison, based on the values of onset and maximum decomposition temperatures, was made between the thermograms of PLA/EVA80/TiO2 composites and neat PLA (Table 3), indicating that incorporating EVA80 and TiO2 nanoparticles improved the thermal stability of PLA. Previous reports have observed the same trend7,12.Figure 5TGA (a) and Drev. Mass loss (b) curves of PLA: EVA80: 1 wt% TiO2 with a heating rate of 10 °C/min.Table 3 TGA results of PLA/EVA80/TiO2 composites.Non-isothermal cold crystallization kineticsFigure 6 displays the DSC thermograms of neat PLA, PLA/1 wt% TiO2 and PLA/EVA 80/TiO2 for a second heating scan at different heating speeds. Exothermic cold crystallization peaks appeared in the curves at temperatures between 104 and 123 °C for neat PLA and 100–116 °C for PLA/1 wt% TiO2 and 95–106 for PLA/EVA80/1 wt% TiO2. When EVA80 or TiO2 are added to PLA, a distinct, sharper, narrower cold crystallization peak occurs during the heating scan. Figure 6 and Table 4 show that the cold crystallization temperature of neat PLA shifts to a lower temperature range as the heating rate is reduced. It demonstrated that increased crystallinity could be attained with lower heating rates (Table 4) and suggested that the lower heating rate enhanced crystallization by providing an appropriate crystallization time.Figure 6non-isothermal DSC different heating curves of PLA: EVA80: 1 wt% TiO2.Table 4 Modified Avrami plot for PLA: EVA80:1 wt.% TiO2.The same trend was also observed for PLA/EVA80/TiO2 blends. The cold crystallization enthalpy for neat PLA strongly depends on the heating rate. However, cold crystallization enthalpy for PLA/EVA 80/1 wt% TiO2 slightly depends on the heating scan, as shown in Table 3. The relative crystallinity X(T) as a function of crystallization temperature T would be obtained by integrating \(\left(\frac{\partial H}{\partial t}\right)\) in the specified crystallization temperature range, as shown in Eq. (3).$$\text{X}\left(\text{T}\right)= \frac{{\int }_{{T}_{0}}^{T}\left(\frac{\partial H}{\partial t}\right)\partial t}{{\int }_{{T}_{0}}^{{T}_{\infty }}\left(\frac{\partial H}{\partial t}\right)\partial t}$$
(2)
\({T}_{0}\), \(T\), and \({T}_{\infty }\) are the starting, arbitrary, and final crystallization temperatures. It is possible to translate the heating-induced crystallization temperatures to crystallization times based on Eq. (3)$$t= \frac{\left|T-{T}_{0}\right|}{\varphi }$$
(3)
where \(\varphi\) is the heating rate.The relationship between the relative degree of crystallinity x(t) and the crystallization time t at different heating rates is shown in Fig. 7.Figure 7Relative crystallinity, \(\text{X}\left(\text{t}\right)\) versus crystallization time for non-isothermally crystallized PLA: EVA80: 1 wt% TiO2 composites at different cooling rates, dot lines refer to t0.5.The time taken for 50% of total relative crystallinity is known as the crystallization half-time (\({t}_{0.5})\). (\({t}_{0.5})\) is evaluated from Fig. 7, and its value is inserted in Table 4. \({t}_{0.5}\) is inversely proportional to the crystallization rate of polymers. Table 2 shows that the value of \({t}_{0.5}\) decreases with the increase in the EVA80 content and the presence of TiO2 nanoparticles.The Crystallization Rate Parameter (CRP) is utilized to compare non-isothermal crystallization rates quantitatively. The assessment is conducted by analyzing the gradient of a linear graph created by plotting the inverse of \({(t}_{0.5})\) against the rate of heating, as seen in Fig. 8. Straight lines were achieved in this investigation for neat PLA and its composites, with a regression coefficient of 0.95. The value of CRP was evaluated and recorded in Table 4. A higher slope value suggests a more rapid rate of crystallization. The CRP value indicates that the rate of crystallization increases as the EVA 80 content increases. The correlation between CRP and EVA 80 levels indicates the compatibility between PLA and EVA 80.Figure 8\(\frac{1}{{\text{t}}_{0.5}}\) versus a heating rate for PLA:EVA80: 1 TiO2 composites.It was reported that a completely immiscible blend is slightly CRP-dependent on blend composition50.Jeziorny’s modified Avrami equation51 can be employed to analyze the kinetics of non-isothermal crystallization in crystalline polymers6,49,52,53$$\text{log}[-\text{ln}\left(1-X\left(t\right)\right]=\text{log}{Z}_{t}+n \,log(t)$$
(4)
where X(t) represents the relative crystallinity at time t. The Avrami exponent and crystallization rate constant are n and \({Z}_{t}\) respectively. Because temperature fluctuates constantly during non-isothermal crystallization, the constants Z and n have different physical meanings than isothermal crystallization. According to Jeziorny’s proposal, \({Z}_{t}\) needs to be modified by the heating rate as follows$${Z}_{c}= \frac{\text{log}{Z}_{t}}{\varphi }$$
(5)
Figure 9 shows the modified Avrami plot for PLA samples for x(t) in the range of 0.2–0.8. For all PLA samples in Table 4, \({Z}_{c}\) rises as the heating rate increases, whereas t1/2 exhibits the reverse pattern. \({Z}_{c}\) is in the order: PLA/EVA 80/1 wt% TiO2 > PLA/1 wt% TiO2 > PLA. The typical Avrami exponent n values are 2.53 for pure PLA, 3.48 for PLA with 1 wt% TiO2 composite, and 3.7 for PLA/EVA80 with 1 wt% TiO2, indicating three-dimensional (3-D)6,52,54Figure 9log(− ln(1 − xt)) versus log(t) for PLA: EVA 80: 1 wt% TiO2.The MO approach, which combines Avrami and Ozawa models, is frequently utilized in research to evaluate non-isothermal crystallization:$$\text{ln}\left(\varnothing \right)=Ln(F\left(T)\right)- \alpha \text{ln}(t)$$
(6)
Mo variables related to heating rate, \(\varnothing\), and crystallization temperature. The plot of \(\text{ln}(\varphi )\) against \(ln (\text{t)}\) at a given relative crystallinity \(X(t)\) is plotted in Fig. 10, where the slope and intercept of the curve are \(\alpha\) and \(\text{lnF(T)}\), respectively.Figure 10\(\ln \left( \emptyset \right)\) versus \(\text{ln}(\text{t})\) for different \(\text{X}\left(\text{t}\right)\) of PLA: EVA 80: 1 wt% TiO2 composites.All data give a straight line from \(\text{X}(\text{t})=0.2\)–0.8, with a regression of 0.999 (as shown in Fig. 10; indicating that the MO approach could describe the crystallization behavior of the PLA and PLA/EVA 80/1 wt% Ti O2 composites well. A similar pattern was observed for PHB/EVA6050. Table 5 displays the values of “F(T)” and α for PLA and its composites. As a general trend, an increase in the degree of crystallinity shifts \({\text{the}}\) \(\text{F(T)}\) values to higher values. PLA/EVA 80/TiO2 composites \(\text{F(T)}\) values are lower than that of neat PLA. This suggests an enhanced effect of EVA80 and TiO2 on the crystallization behavior of PLA. However, when the concentration is considered at a given \(\text{X(t)}\), the dependence of \(\text{F(T)}\) values of PLA/EVA80/1 wt% TiO2 on concentration is weak due to slight miscibility between EVA 80 and PLA. Table 5 demonstrates the slope values. \(\text{(b)}\) are almost constant for each composite’s composition.
Table 5 MO approach parameters for PLA/EVA80/1 wt% TiO2 nanocomposites.An effective activation energyBy using Friedman’s differential isoconversional approach, as described in Eq. (8), to the non-isothermal cold crystallization curve, it is possible to determine an effective activation energy \({E}_{X(t)}\) for PLA and its composites.$${\left[\frac{\partial \mathit{ln}(X(t)/t)}{{\partial T}^{-1}}\right]}_{x(t)}= -\frac{{E}_{X\left(t\right)}}{R}$$
(7)
\(T\) is the temperature and \(\mathit{ln}(X(t)/t)\) is the instantaneous crystallization for a certain relative crystallinity. For a given relative crystallinity, the slope of a linear plot of \(\mathit{ln}(X(t)/t)\) as a function of 1/T is equal to \({E}_{X(t)}\), where \({E}_{X(t)}= -slope*R\). It is worth noting that the combined activation energy of the crystal growth and nucleation processes makes up effective activation energy. The effective activation energies of neat PLA and its composites at each relative crystallinity are positive, as seen in Fig. 11. This circumstance demonstrated that the crystallization rate increased as the crystallization temperature decreased6,52,53. Vyazovkin and Dranca et al.55found that a positive effective activation energy is obtained during non-isothermal crystallization when heating from the glassy state in the crystallization temperature zone above the glass transition temperature but below the region of maximum crystallization rate. Effective activation energy values dropped as relative crystallinity increased X(t), as seen in Fig. 11a. PLA/PEVA80/TiO2 exhibits higher effective activation energies than neat PLA. Reports6,52 have indicated a comparable pattern with PLA blends and composites. Figure 9b plot shows the relationship between effective activation energy and average temperature. The average temperature is computed at each X(t). Figure 11b shows crystallization occurred for PLA/EVA80/TiO2 nanocomposites at lower temperatures. This result is consistent with the DSC outcomes in Fig. 1 and Table 1. A similar trend has been reported for poly(butylene adipate-co-terephthalate)/treated calcium carbonate54, PLA/talc56,57,58, PLA/PHO/talc6Figure 11An effective activation energy as a function of relative crystallinity and average crystallization temperature.In future work, it’s important to explore different real processing techniques, such as 3D printing or molding, to study the effectiveness of these composites in practical applications.
Synergistic effect of TiO2 nanoparticles and poly (ethylene-co-vinyl acetate) on the morphology and crystallization behavior of polylactic acid
