3.1. Structure and physical properties of foams
In this research work, the results were discussed with respect to the influence of G + W and W/G amounts on microstructural and physical properties of Fx-y foams, to find an optimized formulation for TPS/PS blend foam. Fabricated Fx-y samples with various amount of plasticizers, x, at different y ratios were shown in Fig. 1.
SEM images of TPS/PS (40/60) blend foams with various plasticizer contents of G + W and ratios of W/G are demonstrated in Fig. 2. As seen in the figure, using lower amount of plasticizer at the TPS preparation helps to achieve a foam with better morphology. Moreover, increasing y from 1/2 to 2/1, caused a better distribution for the foams cellules at the lower amount of x. Therefore, increasing and decreasing the contents of glycerol and water in the plasticizer combination, respectively, damaged the cells structure and led to a further cell collapse. Since, in spite of the plasticization role, evaporation of water has favorable influence on foaming in the extruder process, from blowing aspect. While, glycerol with a much higher boiling point [34] than the process temperature, acted as a plasticizer only, and caused the reduction of blend melt viscosity. The increment of plasticizer content (x) led to the collapse of foam cells. The worst effect of plasticizer on the blend foam belonged to F23-y and F25-y samples, which significant cell wall rupture was observed in them. It can be attributed to lower melt strength of the Fx-y blend foams compared to the enhancement of the gas pressure within extruder barrel, due to the increase of water vapors. This issue causes rupturing the wall of created bubbles during the foaming process, and subsequently the foam collapse. Accordingly, the effect of raising y had an opposite influence on the cells structure in the F23-y and F25-y samples, due to the surplus made internal pressure by water vapors, and followed by rupturing the cells wall. Because, TPS with these levels of plasticization could not endure the bubble internal pressure. Therefore, cell wall rupture occurred for F23-y and F25-y samples, particularly at higher y. In addition, enhancing the plasticizer amount to x = 29% changed the aforementioned variation trend, due to further reduction of crystallinity degree and hydrogen bonding between the starch chains compared to x = 23 and 25%. In fact, this issue increased the melt strength [24] of the F29-y samples that helped saving the created cells, subsequently, this issue obliged the cells to endure even higher amount of water vapors at further y values.
According to Fig. 2, the cellular structure of Fx-y samples did not have any regularity, and included of large and small bubbles. In order to create a deep insight into the plasticizer effect on the microstructural properties of Fx-y, first, the cells diameter of each Fx-y sample was obtained via using SEM images. After that, the probability density function (PDF) of the cells diameters was computed for all the foam samples and then were shown in Fig. 3. As discussed currently, increasing the x and y contents in F21-y and F29-y had a favorable effect on the cell size, with taller and slender distributions at less amounts of d (see Fig. 3). Whereas, this behavior was completely vice versa in the F23-y and F25-y samples.
The achieved cell characteristics of each Fx-y sample are listed in Table 3. As seen in the table, the calculated results of cell size (d) and standard deviation (σ) have a good conformity with Fig. 2 and Fig. 3, and verifies the mentioned discussions on the irregularity of foam cellules, clearly. According to the results, increasing the loading content of water in the plasticizer combination led to enhancing the cell density in F21-y and F29-y. Whereas, the behavior was inverse in the F23-y and F25-y samples. Besides, cell wall thickness did not have a specified variation. Moreover, the VER values of F21-y and F29-y samples nearly had an irrespective behavior to y, while, its evolution was uptrend in F23-y and F25-y. Since, the cell rupture occurred in the foams with x = 23 and 25% leading to increasing the cell size significantly (see Figs. 2 and 3). This issue raised the volume of mentioned foams compared to the foams with x = 21 and 29%.
Table 3
Cell characteristics of Fx-y blend foams.
Sample
|
d±σ
(mm)
|
VER
|
Nc
(cell/mm3)
|
δ
(mm)
|
F21-1/2
|
0.283 ± 0.097
|
1.990
|
2.195×105
|
0.118
|
F21-2/3
|
0.339 ± 0.152
|
1.708
|
1.064×105
|
0.187
|
F21-3/2
|
0.197 ± 0.108
|
1.587
|
4.838×105
|
0.127
|
F21-2/1
|
0.197 ± 0.073
|
1.296
|
2.991×105
|
0.215
|
F23-1/2
|
0.873 ± 0.401
|
1.696
|
0.061×105
|
0.489
|
F23-2/3
|
0.577 ± 0.334
|
3.065
|
0.350×105
|
0.126
|
F23-3/2
|
0.402 ± 0.222
|
5.253
|
1.246×105
|
0.044
|
F23-2/1
|
0.678 ± 0.396
|
6.427
|
0.271×105
|
0.059
|
F25-1/2
|
0.119 ± 0.051
|
1.489
|
19.500×105
|
0.088
|
F25-2/3
|
0.230 ± 0.072
|
1.755
|
3.537×105
|
0.120
|
F25-3/2
|
0.323 ± 0.184
|
2.981
|
1.972×105
|
0.073
|
F25-2/1
|
0.491 ± 0.212
|
5.167
|
0.681×105
|
0.055
|
F29-1/2
|
0.254 ± 0.095
|
1.580
|
2.242×105
|
0.165
|
F29-2/3
|
0.326 ± 0.196
|
1.535
|
1.006×105
|
0.225
|
F29-3/2
|
0.196 ± 0.093
|
1.950
|
6.472×105
|
0.084
|
F29-2/1
|
0.124 ± 0.062
|
1.745
|
22.397×105
|
0.065
|
To create a better view into the density variation of Fx-y foams with changing the plasticizer contents and combinations, foam density at various x and water percentage in plasticizer combination (z) were reported in Fig. 4. As shown in Fig. 4a, density of F21-y and F29-y have a smooth change with increasing the z value, while this change is rapidly in F23-y and F25-y. This issue is due to the uniform distribution, and integration of bubbles in the F21,29-y and F23,25-y foams, respectively. Therefore, a main factor of the decrement of foam density of F23,25-y, was the cell rupture and followed by agglomeration of the bubbles. Additionally, Fig. 4b shows the density change versus plasticizer content at a same z. As shown, after engendering a minimum in the curve, the foam density increased with increasing x at diverse amount of z. higher amount of water content in plasticizer combination caused a lower value for ρf.
3.2. Water absorption property
TPS is a water soluble polymer due to its numerous hydroxyl groups [29], which leads to moisture attack and followed by significant changes in its physical and mechanical properties [36]. Moreover, the hydrophilicity reduces the TPS tendency to its hydrophobic counterpart, like PS, in polymer blending [37] Accordingly, this issue limits the TPS usage in industrial applications. In this work, we considered the effect of different plasticizer contents and combinations on the moisture absorption of TPS/PS blend foam with the ratio of 40/60. The obtained MA versus z and x for each Fx-y sample was reported in Fig. 5a and b, respectively. As exhibited in the figure, MA value reduced with increasing the z amount at each x, and as well as raising the x content led to the reduction and then enhancement of MA at each z. Factually, the increment of z had an extreme effect to increase the hydrophobicity of the blend foam. Because, rising the water content in plasticizer leading to reduction of the intermolecular hydrogen bonding between the starch chains via frustrating the chains' OH groups [38]. In contrast, increasing glycerol in the plasticizer combination have an undesirable effect on the hydrophilicity reduction of the blend foam, due to its high water tendency [39].
3.3. Statistical data analysis
In order to find an optimized formulation for TPS/PS blend foam from plasticizer point of view, 4×4 full factorial design was used. This consideration was carried out to achieve a blend foam with the most desirable microstructural and moisture absorption properties. Design of experiment (DOE) and analysis of variance (ANOVA) were applied via utilizing MINITAB® 18 [40]. The plasticizer contents of x = 21, 23, 25, 29%, and water percentage in plasticizer combination of z = 33.3, 40, 60, 66.7%, were evaluated as the independent factors, cell diameter and water absorption amounts were also defined as the responses in DOE. Factually, all the microstructural properties and hydrophobicity behaviors of the blend foams are the function of these two responses. The responses were investigated in five replications. After removing of non-significance items, the fitted models provided by the mentioned software for average cell diameter, da, and water absorption, MA. The results were reported in Table 4. The coefficients in each equation demonstrates the effect of the related term on that property. Positive and negative coefficients of the linear models indicate a synergistic and decreasing impact on the responses, respectively. In fact, the goal of optimization was to minimize cell diameter and foam ability to absorb moisture. Because, a TPS/PS blend foam with these characterizations leads to the best microstructural and hydrophobicity properties (see Sect. 3.1).
Table 4
Linear model for average cell diameter and moisture absorption of the TPS/PS blend foams with the ratio of 40/60. x and z are the plasticizer content and water percentage in the plasticizer combination.
Response
|
Linear model
|
Cell diameter
|
d = 0.3483–0.09431 x [21] + 0.2752 x [23]- 0.05756 x [25] - 0.1233 x [29]+ 0.02494 z [33.3] + 0.01969 z [40.0]- 0.06881 z [60.0] + 0.02419 z [66.7]+ 0.004062 x×z [21,33.3]+ 0.06531 x×z [21,40.0] + 0.01181 x×z [21,60.0] - 0.08119 x×z [21,66.7] + 0.1886 x×z [23,33.3] - 0.06619 x×z [23,40.0] - 0.1527 x×z [23,60.0] + 0.03031 x×z [23 66.7] - 0.1967 x×z [25,33.3] - 0.08044 x×z [25,40.0] + 0.1011 x×z [25 60.0] + 0.1761 x×z [25,66.7] + 0.004063 x×z [29,33.3] + 0.08131 x×z [29,40.0] + 0.03981 x×z [29,60.0] - 0.1252 x×z [29,66.7]
|
Moisture absorption
|
MA = 4.6250–0.012 x [21] - 0.640 x [23] + 0.112 x [25]
+ 0.540 x [29] + 0.735 z [33.3]+ 0.613 z [40.0] - 0.600 z [60.0]- 0.748 z [66.7]
|
Further, desirability function approach (D) was used at the optimization process of the results. It is based on transforming d and MA into a desirability values, the combination of individual responses into a composite function and then its optimization [41]. The D values were so close to 1, which indicates the efficient settings to obtain desirable results for d and MA [42]. Finally, x = 21% and z = 66.7% (y = 2/1) was obtained from optimization process (see Fig. 6) to produce the optimum level of microstructural and hydrophobicity properties for TPS/PS blend foam.
3.4. Effect of PS content on the TPS/PS foam properties
To investigate the impact of TPS loading content on the microstructural, physical and moisture absorption properties of the TPS/PS blend foams, the optimized TPS (x = 21% and z = 66.7% or y = 2/1) was used. For this purpose, TPS/PS foam with the PS percentage of 0, 20, 40, 60, 80 and 100% was supplied via presented methodology in Sect. 2.2.2. The TPS/PS blend foam with different amount of PS was symbolized with Fr, in which r is the PS percentage. The created foam with different r was shown in Fig. 7.
To create a deep understanding into the impact of TPS content on the microstructural properties of the blend foam, SEM images were taken from the Fr samples. Moreover, cells diameter was estimated from the SEMs and their PDF was drawn for each sample. The SEM and PDF achievements were reported in Fig. 8 together.
As shown in Fig. 8, the blend foam at the absence of PS had a good cell morphology and uniform distribution of cell diameter. While, increasing the PS content led to widen the PDF curve and creation of the cell wall rupture, and followed by joining the bubbles together and then creation of very large cellules. This trend was inverse after r > 50%. As seen in the figure, TPS/PS with the r value of 80% has a good cell size, lower irregularity, and also probability density function with a slender distribution at a less amount of d. F100 (pure PS) had the slenderest PDF distribution and the furthest amount of probability at the least cell diameter, which was distinct already, due to the PS ability to foam [44]. To quantitate the cell and foams properties, d, Nc, , ρf and MA amounts were listed in Table 5. According to the table, F20 has efficient foam from i.e. cellular, physical, moisture absorption properties. This formulation has higher expansion ratio than TPS and PS with about half moisture absorption compared to TPS. This combination can be a good alternative candidate for pure PS foam packaging.
Table 5
Cell and foam characteristics for each Fr sample.
Sample
|
d±σ
(mm)
|
Nc
(cell/mm3)
|
δ
(mm)
|
VER
|
MA
(%)
|
F0
|
0.169 ± 0.063
|
17.90×105
|
0.012
|
7.36
|
9.77
|
F20
|
0.352 ± 0.247
|
2.10×105
|
0.015
|
12.17
|
5.02
|
F40
|
0.883 ± 0.569
|
0.11×105
|
0.0994
|
5.18
|
4.83
|
F60
|
0.197 ± 0.073
|
2.991×105
|
0.215
|
1.29
|
4.15
|
F80
|
0.426 ± 0.219
|
0.66×105
|
0.166
|
2.07
|
2.11
|
F100
|
0.112 ± 0.258
|
30.96×105
|
0.057
|
1.77
|
0
|
TPS had a negative effect on the cell growth and its diameter, which were randomly distributed. In fact, it can be attributed to the PS good response to the blowing agent in comparison with starch. In addition, the good melt strength of PS prevented the cell rupture, while this issue was vice versa in the TPS media.