1. Introduction
Plastics are an essential part of modern economies and have been extensively used in a variety of fields, such as food packaging, electronics, aerospace, and more. However, the disposal of solid wastes in which plastics are included has proven to be a challenging endeavor. The issue of solid waste disposal has turned into an urgent environmental problem worldwide, especially petrol-derived plastics wastes. Therefore, governments and researchers worldwide have intensified their efforts to develop novel biocomposites in their quest to replace conventional plastic manufacturing with biocomposites [
1].
Biocomposites are defined as biocompatible and/or eco-friendly composites, which can consist of organic polymers such as polysaccharides and proteins [
2]. Such materials are important research topics in modern science and technology because they can help industries achieve a more carbon-neutral production and replace the use of non-renewable resources, like oil and gas, with renewable resources such as biomass [
3]. Studies in the field of biocomposites have focused mainly on fibers such as hemp [
4], flax [
5], sisal [
6], and bamboo [
7]. Nevertheless, researchers have kept identifying new sources of potential natural fiber reinforcement that can improve the mechanical and physical properties of biocomposites, such as
Abelmoschus esculentus [
8],
Grewia tilifolia [
9],
Sterculia urens [
10], and
Prosopis juliflora [
11]. Following this trend, fibers extracted from the
Musa paradisiaca (plantain) tree are highlighted as a viable composite reinforcement due to its favorable physical and mechanical characteristics [
12]. Chamorro et al. [
13] successfully tested the possibility of incorporating
Musa paradisiaca fibers into thermosetting polymeric resins such as polyester and epoxy resins.
Exploring different composite matrices, organic biodegradable macromolecules, like proteins and polysaccharides, are becoming ever more attractive to researchers, due to their ability to form bio-based polymers. Even though macromolecules are found abundantly in most organic solid wastes in landfills and other trash disposal facilities, they continue to be left to decompose or even incinerate instead of using these wastes’ high potential as a source of bio-based polymers for the manufacturing industry [
14]. Among these macromolecules, starch has been implemented in several studies [
15,
16,
17,
18] as the main component in the fabrication of biodegradable films and coating, indicating the high interest of researchers in implementing starch-based biopolymers, also known as thermoplastic starch, for product manufacturing. Its biodegradability, high commercial availability, and easy processing make starch very attractive for polymer production. Due to starch being mainly present in the edible part of the fruit, recent studies began to evaluate the possibility of using fruit-peel waste flour as a source of starch. Such flours contain not only starch but also proteins, lipids, fibers, and other polysaccharides [
16,
19,
20]. The interest in using starch-based biopolymers, in combination with natural fiber reinforcements, lies in the possibility of creating added value products by using agroindustrial wastes.
The banana fruit and its varieties, such as plantain, are a renewable source from which both starch and natural fibers can be extracted. Unfortunately, it is used solely as a food crop, meaning that most of the residues of its life cycle (peels and pseudostem) have no current use in the industry [
21], making finding new applications a research challenge. In particular, plantain as a source of starch for bio-based polymer production has been studied in the last decade. F.M. Pelissari et al. [
19] compared the physical, thermal, mechanical, and morphological properties of both plantain starch-based and plantain flour-based polymer films, highlighting the potential of flour-based polymer films as a greener and cheaper alternative to the use of pure starch for manufacturing, as well as championing the tenets of a circular economy.
Even though the use of natural fibers as composite reinforcement and the development of bio-based matrices has been evaluated, the integration of natural fibers and starch bio-based matrices has not been deeply explored in literature. Therefore, the production of biocomposites combining plantain flour bio-based polymeric matrices and plantain natural fibers could combine both fields in the manufacturing of added-value products. This new approach in the development of biocomposites would assess the possibility of implementing most of the wastes of a single crop’s production (plantain fibers) and consumption (plantain peels) stages.
Hence, this work presents the development and characterization of a bio-composite film of plantain flour thermoplastic starch (TPF) reinforced with plantain tree fibers (PF). The chemical, thermal, mechanical, and morphological characterization of TPF, untreated PF, and the composite material (TPF/PF) are shown through chemical composition analysis, Fourier-transformed infrared analysis (FTIR), thermos-gravimetric analysis (TGA), tensile tests, and scanning electronic microscopy (SEM).
2. Materials and Methods
2.1. Materials
Unripe plantain bananas of the variety “Harton” were acquired during the 2021 harvest in local markets in Bogota, Colombia, and used to obtain the plantain flour. This flour is extracted to produce TPF, in combination with industrial-grade glycerin (87% w/w) provided by Roda Químicos S.A.S. Plantain fiber bundles with a diameter between 94–189 μm were provided by farmers from San Agustín, Huila, Colombia, to implement the fiber as reinforcement. This fiber was previously cut and combed manually by the providers.
2.2. Chemical Composition of Plantain Fibers
A chemical composition analysis of the plantain fiber was carried out following TAPPI methods. Ash, lignin, cellulose, and extractive contents were determined. A minimum of 3 samples were used in each test.
2.2.1. Ash Content
This test was performed following the TAPPI T211 Standard Method [
22]. For this procedure, a 1 g dry sample was exposed to a three-step drying cycle to avoid flames during the experiment. First, the sample was heated at 105 °C for 1 h, then heated at 250 °C for another hour, and finally at 575 °C for 4 h. This procedure guarantees good carbonization and the calcination of the sample. For this test, a VIBRA HT224R analytical scale and a Thermolyne 62700 furnace were used. The ash content was calculated as the recovered ash mass (m
ash) divided by the initial sample mass (m
sample), as explained in Equation (1).
2.2.2. Lignin Content
This test was performed following the TAPPI T222 Standard Method [
23]. A 1 g dry sample was treated with 72% (
w/
w) sulfuric acid for 2 h. The sample was then macerated and stirred using a glass rod to dissolve extractive contents entirely during this process. Next, water was added to the solution until a 3% sulfuric acid mixture was obtained. Afterward, the solution was heated for 4 h to maintain constant volume to get lignin to coagulate. Lignin content was filtered from the solution via glass vacuum filtration, washed with water at 90 °C, dried in a Memmert UNE 200 forced convection oven at 105 °C, cooled, and finally weighed. The lignin content was calculated as the ratio of recovered lignin mass (m
lig) to the initial sample mass (m
sample), as seen in Equation (2).
2.2.3. Cellulose Content
This test was performed following the TAPPI T203 Standard Method [
24]. A 1.5 g sample was extracted using 17.5% sodium hydroxide solution at 25 °C. Alpha (α), beta (β), and gamma (γ) cellulose were determined via oxidation with potassium dichromate; therefore a two-step oxidation was carried out to determine the content of each cellulose type correctly. In the first oxidation, beta and gamma cellulose content was separated from α-cellulose content because α-cellulose is insoluble in sodium hydroxide. α-cellulose content was then determined as the difference between the initial specimen (100%) and the undissolved fraction, as seen in Equation (3).
V
1 corresponds to the sample titration volume; V
2 corresponds to the blank titration volumen; N is the normality of the titrate (ferrous amonium); A is the volume of the sample used, and W is the dry weigth of the fiber sample. Subsequently, γ-cellulose was recovered in the second oxidation of the solution, leaving behind the β-cellulose precipitated content. γ-cellulose content was determined using Equation (4).
where V
3 corresponds to the sample titration volume after filtering the precipitated β-cellulose, and V
4 corresponds to the blank titration volume. Finally, β-cellulose content was determined by the difference between the first and second oxidations, see Equation (5).
2.2.4. Extractive Content
This test was performed following the TAPPI T204 Standard Method [
25]. A 4 g sample was oven-dried at 105 °C and milled to obtain a 0.4 mm granule size. Using an ethanol-benzene mixture (1:2,
v/
v), the sample was boiled at 75 ± °C in a Soxhlet apparatus for 5 h. Afterward, the solvent mixture was evaporated, and the residue was weighed. Extractive content was calculated as the ratio of the extract dry weight (m
ext) to the initial sample mass (m
sample), as seen in Equation (6).
2.3. Film and Composite Preparation
Plantain flour was extracted from plantain peels as a preliminary step to prepare the TPF films. Peels were cut into small pieces, and the endocarp was separated from the cellulose-rich outer peel. Then, the starch-rich endocarp was dried for 90 min at 105 °C and subsequently ground using a Pulversisette 19 Universal Cutting Mill to obtain a fine flour powder. Finally, the flour was sieved using a 100 µm sieve, achieving homogeneous particle sizes and removing unwanted cellulose residue.
Films were prepared by casting method [
15,
16,
19], a process in which a film-forming suspension (FFS) is applied over a surface and dried until a film is obtained. The FFS was obtained by homogenizing an aqueous solution of 4%
w/
w flour employing mechanical stirring at a constant temperature between 70 °C and 80 °C for 30 min to dissolve the starch content of the flour adequately. Industrial-grade glycerin was added at 2%
w/
w to the solution and processed at the same conditions for 15 more minutes. The produced suspension was poured into a mold to obtain a constant thickness film. The suspension was dried in a Memmert UNE 200 forced convection oven at 50 °C for 1.5 h to achieve a fast initial plasticization. Finally, the molds were left to dry approximately for 72 h at 20 °C and 50% relative humidity. The final product is a film, which acts as the matrix for the composite.
The composite films were obtained using the casting method described previously. PF was placed and fixed in a unidirectional orientation on the mold. Approximately 60 individual fibers were placed at a separation distance of 1 mm to permeate the fibers thoroughly. FFS was added to the previously conditioned mold with the fibers in place and left to dry for approximately 72 h at 20 °C and 50% relative humidity. The resulting film contained 6.54% w/w fiber content.
2.4. Physical Characterization
The thickness of each specimen was measured using an Ono Sokki GS-332 Linear Sensor Gauge. To obtain an average thickness, five different measures, each 30.5 mm apart, were taken along the length of the specimen. The density of TPF and TPF/PF was determined as the ratio between the mass and volume. The moisture content of each specimen was determined by a Precisa XM 60 Thermobalance and then averaged, following the ASTM D4442 standard procedure [
26]. A minimum of 10 samples of each material were used for this test.
2.5. Fourier-Transformed Infrared Spectroscopy (FTIR)
An FTIR test was carried out to analyze the functional groups present in the TPF, PF, and TPF/PF. For this test, all specimens were ground and mixed with KBr in a ratio of 100:1. The mixture was later homogenized and pressed at 9 MPa for 30 s to obtain a pellet. The wavelength range analyzed was between 4000 cm−1 to 400 cm−1. A Thermo Nicolet 380 FT-IR was used for these tests.
2.6. Thermo-Gravimetric Analysis (TGA)
Thermo-gravimetric analysis (TGA) tests for the TPF, PF, and TPF/PF, were carried out using a TA Instruments Q600 analyzer and following the ASTM E1131 [
27] standard procedure. Samples were heated from room temperature (25 °C) to 600 °C at a rate of 10 °C/min in a nitrogen atmosphere.
2.7. Mechanical Characterization
Tensile properties of the TPF and TPF/PF were determined by an Instrom 3367 Universal Testing System following the D882-10 ASTM standard [
28]. Rectangular specimens, each 25 mm wide and 152.4 mm long, were tested. The specimens were cut utilizing die-cutting. The initial crosshead speed and grip separation were set at 12.5 mm/min and 125 mm, respectively.
Additionally, tensile properties of PF were determined by an Instrom 3367 Universal Testing System according to the C1557 ASTM standard [
29]. A 50 mm long gauge length with a 10 mm/min crosshead speed where used in this test.
A minimum of 10 specimens for each material were used for each test.
2.8. Scanning Electron Microscopy (SEM)
PF and the fractured surfaces of TPF/PF samples were observed by FE-MEB LYRA 3 TESCAN (SEM). Before the analysis, specimens were coated with a thin gold layer, increasing sample conductivity, using a Desk® IV apparatus.
2.9. Statistical Analysis
An ANOVA one-way test was carried out to determine if the physical and mechanical properties of TPF and TPF/PF presented a significant difference. A
p-value lower than 0.05 (confidence level of 95%) was considered statistically significant [
30]. To perform the statistical analysis, Minitab 18 Statistical Software was used.
4. Conclusions
In this study, a biocomposite from agroindustrial plantain waste with attractive mechanical properties and eco-friendly character was developed, contributing to the industrial conversion towards a circular economy.
The overall physical characterization shows that both the plantain flour-based biopolymer film (TPF) and plantain fibers (PF) are low-density and low-moisture materials when compared to synthetic polymer resins and other natural fibers, respectively. As a result, the TPF/PF composite has the advantage of being a lightweight, renewable and biodegradable material.
The chemical composition and spectral analysis of PF showed a high amount of α-cellulose and low amounts of lignin. These characteristics make PF a good candidate for reinforcing composite materials. In addition, the FTIR spectra of the TPF matrix and TPF/PF composite revealed the presence of OH and CH2 groups and amide I and III groups, which are found primarily in polysaccharides and proteins, respectively.
With regards to thermal degradation, it has been determined that, for TPF matrices, it starts at 130 °C, while PF fibers start degrading at 240 °C. The TPF/PF composite displayed the behavior of its constituent materials; therefore its processing temperature is set to below 130 °C to avoid unwanted degradation of the material.
The mechanical properties of both TPF and PF proved to be comparable to other starch bio-based resins and natural fibers, respectively. When combined into TPF/PF, composite films displayed a superior performance in comparison to TPF films, showing plantain fibers as effective reinforcement. SEM analysis showed a good bonding between TPF and PF; and fiber fracture, pull-out, and matrix breakage were determined to be the main failure mechanisms in the TPF/PF composite. The rough surface of the fibers created mechanical interlocking with the matrix, resulting in enhanced mechanical characteristics. Nevertheless, a waxy layer was identified on the fiber’s surface, reducing its adhesion to the matrix and causing gaps between the fiber and matrix. Chemical treatment of the fibers would improve this issue.