Composite materials derived from biodegradable starch polymer and Atriplex halimus fibers

2.1 Fibers

For this study, cellulose-based fibers were extracted from a saltbush (A. halimus) plant, which is an evergreen shrub that grows to 2–3 m in height at a medium rate, is highly resistant to drought and is found in abundance in Mediterranean zones and the Middle East. This plant is known for its remediation of degraded rangelands and salt-affected areas. It is commonly used as a forage plant for sheep and goats in arid areas. In general, researchers are interested about the plant in that context and several studies have been made to determine the chemical composition of its leaves and stems, which varies depending on the type of soil it grows in (14, 15).

In our case, the fibers of A. halimus were used as a reinforcement for the TPS matrix; the stems of the plant were shredded and air-dried for about a week and then oven-dried at 40°C for 12 h. The dried fibers were mixed mechanically in a Moulinex DPA141 food chopper and then sieved to obtain two sizes of cellulose fibers: short fibers (SF) (length 100 μm, diameter 80 μm, aspect ratio 1.25) and long fibers (LF) (length 1938 μm, diameter 430 μm, aspect ratio 4.5). The average length of the fibers was evaluated by microscopic observation.

The chemical composition of the fibers was estimated after various stages of chemical treatment; the α cellulose content was calculated by further treatment of the α cellulose fibers with NaOH. After determining the hollocellulose content using standard test methods (ASTM D1103-60 and ASTM D1104-56), we used the difference between the values of the hollocellulose and the α cellulose content to determine the hemicellulose content; the lignin content was measured according to the ASTM D1106-56 standard. All reagents were supplied by Sigma Aldrich (France). Table 1 presents the chemical composition of different vegetable fibers. Compared to the other fibers, A. halimus fibers showed a low cellulose content and a high hemicellulose and lignin composition.

2.2 Composite preparation

Regular corn starch (15% moisture; supplied by Sigma-Aldrich) and reagent-grade glycerol with 99.5% purity were used to prepare the TPS. TPS/composites containing 0, 5, 10 and 15 wt.% of cellulose fibers added to the matrix were fabricated from the aforesaid prepared fibers. Preliminary experiments illustrated that the optimal glycerol content should be in the range of 20–35%; however, lower and higher glycerol content led to samples that were too brittle or too sticky (8). For this reason, the samples prepared in this work contained starch with 30 wt.% of glycerol.

Corn starch, glycerol and distilled water were mixed and heated to 80°C using a hot plate with continuous stirring for 35 min; when the mixture had completely gelatinized, proper amounts of the cellulose fibers were added and stirred for a further 3 min and then poured into Petri dishes. The cast solution was allowed to dry in open air for several days at room temperature. The following notations were used: H30, non-reinforced TPS; H30-SF5,10,15, TPS composites containing 5, 10 and 15 wt.% of SF added to the matrix; and H30-LF5,10,15, TPS composites containing 5, 10 and 15 wt.% of LF added to the matrix (Table 2).

2.3 Characterization

Polarized light microscopy examinations using an Olympus BX41 microscope with a capture camera (Micro Publisher 3.3 RTV) were carried out on both sides of the cast samples to determine the distribution of the fiber in the matrix and to analyze the dimensions of the fibers.

The mechanical properties of the films were determined from tension tests using a VersaTest AFG100N motorized test stand. The film samples used in the tests were cut with a mold using a press machine to obtain dimensions of 6×24 mm. The materials testing machine was operated at a crosshead speed of 14 mm/min, and the data were averaged over five to eight specimens.

Thermo gravimetric analysis was carried out on a PerkinElmer thermogravimetric analyzer (Pyris 1 TGA, PerkinElmer). The TPS/fiber composites were cut into small pieces of 5–10 mg and then scanned from 40°C to 600°C at a heating rate of 20°C/min in the presence of nitrogen.

The FTIR spectra of different films were analyzed to identify the types of chemical bonds existing in the prepared composites; the powdered samples were blended with potassium bromide, and the IR spectra were recorded with a PerkinElmer spectrometer, with a wave range from 4000 to 450 cm^-1.

The moisture absorption of the TPS composites was determined by placing samples in a sealed glass container under controlled temperature (20°C) and relative humidity (75%). The samples were weighed daily until a constant weight was achieved.

Degradation testing was performed to evaluate the biodegradability of the samples. The soil media used was a mixture of organic hummus and sand that was sieved to remove large plant materials, stones and other inert materials. TPS/composite samples were cut into 30×30-mm squares, weighed for the first time and then each sample was placed in a vessel containing around 100 g of soil. Distilled water was added to each sample for 2 days to bring moisture, and the mixture was placed in an airy place at room temperature. The incubation time was 31 days; the samples were weighed every day to evaluate their weight loss during biodegradation.

3.1 Biocomposite morphology

Simple naked eye examination of the films obtained by the solvent cast method indicated good filler dispersion in the TPS matrix. From Figure 1A, which shows the films made from starch and glycerol, we can see that the granules of the starch completely gelatinized and collapsed because of the temperature. This disruption of the molecular order inside the starch granule is manifested as loss of birefringence (16–18). The successive addition of fibers reduced the optical properties of the samples, but it did not prevent light completely from passing through. Figure 2B and C shows some clear images of the cellulose fibers embedded in the biodegradable matrix. The biocomposites containing SF showed better adhesion with the matrix than the biocomposites containing LF. There were no apparent differences in fiber homogeneity between the top and the bottom side of the biocomposites. However, some of the samples, particularly the biocomposites containing 15% higher filler content, showed the presence of some fiber aggregates, probably deposited by gravity, at the bottom of the cast films. This good interfacial adhesion was attributed to the strong interaction between the fiber and the TPS and to their chemical similarities (8, 11).

Figure 1:

Optical micrographs of the (A) TPS composite (H30), (B) H30-SF10% biocomposite and (C) H30-LF10% biocomposite.

Figure 2:

Stress-strain curves for the composites containing (A) long fibers and (B) short fibers.

3.2 Mechanical analysis

The tensile stress-strain curve for pure TPS and composites containing 5, 10 and 15 wt.% of cellulose fibers is shown in Figure 2. The Young’s modulus and tensile strength of the biocomposites improved with increasing amount of fiber loading.

A comparison of the tensile strength and elongation at break of the composites and pure TPS film showed that the addition of 15 wt.% of SF and LF increased the tensile strength by 92% and 77%, respectively.

This improvement was due to the strong interfacial interaction between starch and fibers, which resulted in good stress transfer (8, 19). In contrast, the sensitive decrease in the elongation at break from 55% to 7% for SF and to 10% for LF was mainly due to the entanglement of the fibers inside the matrix phase, resulting in the decrease in flexibility of the molecular chains of the TPS matrix (2).

The addition of fibers contributed to the reinforcement of the polymeric matrix, and this was reflected in the stiffness of the composites expressed as the areas below the stress-strain curves of TPS and the composites in Figures 3 and 4.

Figure 3:

Effect of the content of long fibers on the mechanical properties of the composites.

Figure 4:

Effect of the content of short fibers on the mechanical properties of the composites.

Noticeably, the composites reinforced with SF exhibited markedly superior mechanical properties when compared to the composites reinforced with LF, especially the considerable increase in the Young’s modulus of the materials. This was due to their small size and good dispersion in the matrix, which resulted in the strong interfacial interaction between the starch and the cellulose (3).

The results obtained from the mechanical tests are summarized in Table 3.

3.3 Thermal properties

TGA results illustrating the thermal degradation of pure TPS and the composites are shown in Figures 5 and 6; they showed that the weight loss decreased gradually with increasing fiber content. The initial mass drop observed at a temperature varying between 50°C and 100°C corresponded to the loss of absorbed water, which was lower in the composites than in the pure matrix.

Figure 5:

TGA scans for TPS composites containing 0, 5, 10 and 15 wt.% of long fibers.

Figure 6:

TGA scans for TPS composites containing 0, 5, 10 and 15 wt.% of short fibers.

The degradation temperatures of biocomposites obtained from derivative thermogravimetric peaks increased progressively with increasing fiber content; the addition of SF and LF increased the degradation temperatures of the composites and the neat TPS film from 352.6°C to 359.2°C and to 357.4°C, respectively.

This can be explained by the higher thermal stability of fibers compared to starch and by the good compatibility between both polysaccharides (8, 20). Moreover, the weight loss decrease with the addition of cellulose fibers is explained by the fact that, at equilibrium, the composites had lower water content when compared to the pure matrices. Hence, the presence of fibers in the matrices decreased the water content inside and the diverse interactions brought by the fibers took the original water site of the TPS matrices (3, 19).

3.4 FTIR analysis

FTIR spectroscopy was used to investigate the interactions between starch and cellulose fibers.

The infrared spectra of pure starch films, cellulose fibers and composite materials are shown in Figure 7. The spectra of the composites have similar features to those of starch films except for some changes due to the interaction with cellulose fibers. The broad band at 3425.2 cm^-1 was the O-H stretching; the band at 2925.4 cm^-1 were associated with C-H stretching; and the peak positions in the range 1639.7 cm^-1 referred to the bound water present in the non-reinforced TPS and composites (21, 22). The wave numbers in the range 1385.1 cm^-1 for the TPS matrix and the biocomposites were designed for O-H bonding (21, 23). The peaks in the range 1157, 1078, 1028 cm^-1 were attributed to C-O-H, C-O and C-O-C stretching in anhydroglucose, respectively (9). The bands located at 924.39, 577.72 cm^-1 was due to the pyranose ring stretching vibration (6, 23).

Figure 7:

FTIR spectra for cellulose fibers, non-reinforced TPS and composites.

Several noticeable changes occurred in the spectra of starch films after the addition of cellulose fibers, and many new peaks appeared in the spectra of reinforced composites. The band at 3425.6 cm^-1 in TPS films slightly shifted to lower wave numbers by the presence of cellulose fibers to reach 3408 cm^-1 in composites spectra; this refers to an increase in intermolecular hydrogen bonding by the addition of cellulose fibers (20). The peaks at 1740.6 cm^-1 that were absent in the TPS spectra appeared clearly in the spectra of the composites, which referred to the C-O stretching intermolecular hydrogen bonds for cellulose (23). The same behavior was mentioned for the peak at 1323 cm^-1 due to the movement of CH₂ in cellulose chain (2, 24). According to the infrared analysis, a relatively good dispersion and compatibility between the cellulose fibers within the film were observed owing to the lower loading level.

3.5 Water uptake

One of the major drawbacks of starch-based materials is their sensitivity to moisture; they can absorb some amount of water from the environmental humidity, as a consequence of which their mechanical properties decrease radically. Figure 8 shows that the addition of long and short cellulose fibers decreased the amount of absorbed water, from 49% for neat TPS and 23% for composites containing 15 wt.% of LF. The equilibrium was reached in only 9 days instead of 15 days for the neat starch. This can be related firstly to the hydrophobic character of fiber in comparison to the hydrophilic property of starch (3) and, secondly, to the presence of strong hydrogen bonding interactions between fiber/fiber and fiber/starch matrix (21).

Figure 8:

Water content as a function of storage time, type and content of fiber.

3.6 Biodegradability

To evaluate the biodegradation of the starchy composites in real conditions, samples of the TPS composites were visually inspected and weighed on the first day; many changes in color and shape were observed. The shape and color of the samples began to change on the fifth day when black spots started appearing, which is a sign of moldiness. After 15 days of incubation, they lost their square shapes and became too brittle. On the 25th day, the samples were already in pieces such that only a few parts could be handled. We could estimate that, after 30 days of incubation, the samples were totally biodegraded in soil (21, 25). The weight of the TPS composites decreased, leading to a total weight loss of more than 90% after 30 days. The rate of degradation was influenced by the type, moisture and temperature of the soil, which varied depending on the operating conditions (26).

Figure 9 shows that the biodegradation rates of reinforced composites were slightly lower than those of neat TPS films; the lowest percentage of weight loss was observed for TPS composites containing 15 wt.% of LF. This is related to the more hydrophobic character of cellulose fibers when compared to starch as explained previously (3).

Figure 9:

Weight loss curves of the TPS film and composites containing 15 wt.% of short and long fibers.