Obtention and Characterization of Bio-Based Composite PBAT/PLA Active Trays for Fresh Food Packaging

TL;DR

This study creates sustainable, antimicrobial food packaging using a bio-based composite material that can help reduce plastic waste and preserve fresh produce.

Contribution

The paper introduces a novel bio-based composite with antimicrobial properties for fresh food packaging using PBAT, PLA, and menthol.

Findings

A formulation with 85/15 PLA/PBAT ratio and 3% plantain microfibers showed good mechanical properties.

Adding 5% menthol provided fungistatic activity against Botrytis cinerea.

The composite sheets and trays exhibited suitable tensile and compressive strengths for packaging applications.

Abstract

Currently, the packaging sector must continue developing more sustainable systems to reduce the high quantities of single-use plastic waste generated. This study evaluated the production and characterization of bio-based composite trays with antimicrobial activity. Different formulations of polybutylene adipate co-terephthalate (PBAT) and polylactic acid (PLA) with polyethylene glycol (PEG) as plasticizer and citric acid as a compatibilizer/crosslinker were evaluated, in addition to the inclusion of plantain microfibers (PFs), TiO2, and menthol as reinforcing and antimicrobial agents, respectively. The mixtures were subjected to pellet extrusion (165/175/185/190 °C and 60 rpm) and then to flat sheet extrusion (at 185/190/195/205 °C and 60 rpm), besides calendering (at 3.5–6.0 rpm). A single-screw extruder was used in both cases. The obtained sheets (0.317 ± 0.040 mm thick and 17 cm wide) were molded into 12.5 × 11.0 × 3.5 cm trays in a thermoforming machine (at 325 °C and vacuum pressure). For the resulting composite sheets and trays, measurements of mechanical strength, moisture absorption, barrier (WVTR), transmittance, and color were performed. FT-IR, DSC, TGA, SEM, and in vitro antimicrobial tests were also conducted. Based on these tests, an initial formulation with an 85/15 (w/w) PLA/PBAT ratio was defined, which was then reinforced with 3% (w/w) PF. Furthermore, the inclusion of 5% (w/w) menthol in the composite led to fungistatic activity against Botrytis cinerea, also resulting in homogeneous sheets (tensile strength 24.137 ± 1.439 MPa) and trays (compressive strength 0.113 ± 0.010 MPa). These findings can be applied to the packaging and preservation of perishable produce.

Figures

Full text

1. Introduction

The production of food packaging has been growing steadily in recent years, driven by the increased production and marketing volume of these products [1]. Many food preservation packages, such as those for fresh produce, are single-use and discarded after the product is consumed, generating significant waste and pollution [2]. In rigid and semi-rigid food packaging, petrochemical plastics such as polyethylene terephthalate (PET) and polypropylene (PP) currently dominate the market due to their good mechanical and barrier properties, low packaging/product interactions, and ease of molding and processing [3,4]. However, these materials are poorly biodegradable and mostly end up in landfills and oceans for decades [5,6].

For several years, progress has been made in obtaining and applying bio-based packaging from different natural and/or biodegradable materials [7,8]. Many of these alternatives are already commercially available, and their development has been aimed at replacing traditional petrochemical plastics, trying to match their mechanical, optical, and barrier characteristics while maintaining a high degree of biodegradability/compostability [9,10]. In the case of rigid/semi-rigid packaging, the most commonly used biomaterials are polylactic acid (PLA), polybutylene adipate terephthalate (PBAT), and, of course, cellulose-based cardboard [11,12,13].

For bio-based packaging to have similar or improved characteristics compared to traditional PP and PET packaging, techniques such as compounding mixing plastics with complementary properties are often used, as well as composite formation, with appropriate reinforcements in the form of nano- or microparticles, and/or chemical modifications to alter the properties of the polymers, e.g., making them less sensitive to moisture [14,15]. In the case of composites, Reis et al. [16] obtained trays composed of thermoplastic starch/PLA reinforced with beeswax by using flat-sheet extrusion, calendering, and thermocompression, achieving improved mechanical and moisture-barrier properties. Hernandez-Garcia et al. [17] developed bi-layer trays with paper and bio-based polybutylene succinate (PBS) and polybutylene succinate-co-adipate (PBSA), achieving improved gas/aroma barrier properties, as well as enhanced mechanical resistance. Vorawongsagul et al. [18] also evaluated the production of a foamed composite based on PLA/PBS/cellulose fiber for use in rigid packaging subjected to high temperatures. For their part, Sung et al. [19] developed a bio-nanocomposite film using PLA as the base material and cellulose nanocrystals (CNC) as the reinforcing agent, obtaining good barrier properties. Silva et al. [20] developed biodegradable films based on wheat flour and PBAT, with varying amounts of malt bagasse, resulting in lower hydrophilicity, reduced elongation at break and tensile strength, and higher thermal stability.

A line of research of increasing interest is the development of biobased packaging systems with active properties for food preservation. In this case, a significant part of the proposed solutions is focused on microbial control, modulation of deterioration factors such as humidity and O_2_, and preservation of the product’s quality properties [21,22]. Of particular interest in fresh produce is the control of deteriorative microorganisms, such as Botrytis cinerea, a necrotrophic fungus known as gray mold that causes tissue rot in a wide range of foods, including tomato, blueberry, and cape gooseberry [23,24,25].

Active materials can be incorporated standalone into the packaging system in the form of sachets, pads, or tags, or added directly into polymeric matrices to form composites [26]. Among the substances commonly used as antimicrobial agents, those with controlled-release mechanisms, typically via evaporation from the delivery matrix, are particularly noteworthy. These include natural extracts, essential oils [27,28], and their active components such as menthol, limonene, cinnamaldehyde, carvacrol, and thymol [29,30,31]. On the other hand, solid micro- and nanoparticle materials with static antimicrobial activity have also been widely used, including metal oxides such as TiO_2_ and ZnO, as well as amino polysaccharides such as chitosan [32,33,34].

Regarding controlled-release active compounds, López-Gómez et al. [35] evaluated the inclusion of different essential oils (EOs) encapsulated in β-cyclodextrin in cardboard packaging for grapes, nectarines, and lettuces, thereby extending shelf life. Amiri et al. [36], on the other hand, used a starch-based microcellular foam loaded with peppermint essential oil, whose active compound is menthol, resulting in reduced growth of mesophilic bacteria, molds, and yeasts in fresh strawberries and a decrease in the loss of antioxidant activity in this fruit. On the other hand, regarding the use of materials with static antimicrobial activity, Bodaghi [37] evaluated the use of low-density polyethylene nanocomposite films containing clay (Cloisite 20A) and TiO_2_ nanoparticles in the preservation of tomatoes stored at 4 °C, obtaining a decrease in metabolism, a reduction in ethylene production, and greater maintenance of quality properties. Nevado-Velasquez et al. [38], for their part, used Ag-doped TiO_2_ nanoparticles for the postharvest preservation of ‘Hass’ avocados under refrigeration and ambient conditions, obtaining a reduction in the deteriorative activity of Colletotrichum gloeosporioides and a decrease in fruit metabolism.

While stand-alone elements such as sachets are easier to manufacture and have higher capacities for loading preservatives or removing unwanted compounds from the packaging system, the provision of active materials directly within the packages as composites facilitates logistics in the packing process and eliminates additional parts in the system that may be inadvertently ingested by consumers or be rejected by them [39,40].

When creating the composite, it is necessary to consider how incorporating the active materials impacts packaging characteristics such as its mechanical strength, color/transparency, and gas and liquid permeation, for example. The maximum limit for incorporating active materials is defined by the largest possible quantity that results in effective preservation without compromising the packaging performance [41,42]. Therefore, it is advisable to consider optimizing the system by appropriately selecting the active materials according to the product to be preserved.

Research into biobased active composites for food packaging has been more focused on the development of flexible coatings and films, especially using techniques such as solvent casting [43,44]. However, some studies are available on the obtention of rigid/semi-rigid packaging with preservative activity by directly incorporating the active substances into the polymer matrix. Gonon et al. [45] evaluated the production of PLA trays by extrusion and thermoforming, incorporating carvacrol and cinnamaldehyde, and obtained a package with acceptable mechanical properties and the ability to inhibit fungal growth. Srisa et al. [46], for their part, evaluated the incorporation of EDTA, ethyl maltol, and ethyl lauroyl arginate into PLA sheets obtained by extrusion and the resulting antimicrobial and antioxidant activity for the preservation of meat products. In any case, further research is needed on the production of biobased composites to develop trays with antimicrobial/preservative capabilities and their specialized application in perishable food preservation.

Considering the above, this study aimed to develop and characterize semi-rigid composite trays based on PLA/PBAT and reinforced with powdered plantain fibers (PFs), TiO_2_ particles, and menthol. Initially, the fiber procurement process was adjusted, and then the processing conditions for developing the composite trays by extrusion and thermoforming were defined. Finally, physicochemical characterization and antimicrobial activity tests were performed on the resulting trays in order to determine their application in the packaging of fresh produce.

2. Materials and Methods

2.1. Materials and Reagents

The biodegradable polymers used as matrix materials were Polylactic acid (PLA) Ingeo^TM^ 2003-D grade (NatureWorks Co., Ltd., Plymouth, MN, USA) and Polybutylene adipate-co-terephthalate (PBAT) KHB21AP11 MIF-3 (Kanghui New Material Technology Co., Ltd., Yingkou, China). Regarding additives, polyethylene glycol 6000 (Indorama Ventures Ltd., São Paulo, Brazil) was used as a plasticizer, Span 80 (Sigma-Aldrich, St. Louis, MO, USA) as an emulsifier, and powdered citric acid (Shandong Ensign Industry Co., Ltd., Weifang, China) as a compatibilizer/crosslinker. Titanium dioxide (Disproquímica S.A.S., Bogotá, Colombia) and menthol (Disproquímica S.A.S.) were evaluated as antimicrobial active agents, both ground to a particle size of 100 mesh (<0.149 mm).

For the plantain fiber extraction process, plantain pseudostems were collected from a local farm in the municipality of Villavicencio (Department of Meta, Colombia). The samples were washed, disinfected, cut into 1 cm-thick slices, dried in a convection oven at 60 °C until a moisture content of >10%, and finally ground in a high-speed blade mill to obtain a dry powder. Likewise, the following chemicals were used for this process: potassium hydroxide (87% w/w, Sigma-Aldrich), sodium chlorite (NaClO_2_), technical grade (80% w/w, Disproquímica S.A.S.), and glacial acetic acid (95% w/v, Sigma-Aldrich).

2.2. Fiber Extraction

For the extraction process of plantain fibers (PFs), the procedures described by Pelissari et al. [47] and Mishra et al. [48] were used with adjustments (Figure 1). First, the powdered plantain pseudostem was treated with 5% NaOH (w/v) at a 1:15 solid-to-liquid ratio. The solution was mechanically stirred at 40 °C for 4 h, then filtered and washed several times with distilled water until the pH was neutral. Next, a bleaching treatment was performed with NaOCl_2_ 1% (w/v) at a 1:5 solid to liquid ratio (adjusting the pH to 5 by using acetic acid). The resulting solution was kept under mechanical stirring at 70 °C for 1 h, and then it was filtered and washed several times with distilled water until the pH reached neutrality. From this last step, the wet fibers were recovered and then dried in a forced convection oven at 60 °C for 24 h. Finally, the fibers were milled and sieved to obtain particle sizes of 40 mesh (<0.420 mm) and 100 mesh (<0.149 mm) in order to evaluate these sizes in the tests for obtaining composite sheets and trays, described in the following sections. In any case, during the milling and sieving process, the larger fibers were ground to the sizes desired for evaluation.

The plantain fiber yield (FY, % w/w) was calculated by using the formula given below:

$$[eqn]$$

where M_1_ is the initial weight of the dried and ground plantain pseudostem, and M_2_ refers to the final weight of PF after they have been extracted, ground, and sieved.

The moisture content of the obtained fibers was determined using a halogen-light moisture analyzer HR73 (Mettler Toledo Inc., Columbus, OH, USA), and their mid-infrared spectrum was analyzed to identify relevant functional groups using a Fourier Transform Infrared Spectrophotometer FI/IR-4700 (JASCO Corporation, Tokyo, Japan) equipped with a horizontal single-reflection diamond ATR accessory. The spectra were scanned over a range of 4000–400 cm^−1^ with a spectral resolution of 4 cm^−1^ in 20 scans at room temperature, and the absorbance peaks in the range of 4000–500 cm^−1^ were observed using the flat-tip ATR accessory.

2.3. Formulations for Obtaining Composite Trays

The evaluation of different formulations of composite sheets and trays was conducted in a three-stage successive process: (i) first, blends of PLA/PBAT and additives were tested until a stable and homogeneous thermoformable sheet formulation was achieved; (ii) subsequently, the addition of different percentages of reinforcing microfibers (PF) was evaluated, while maintaining the homogeneity and stability of the thermoformable sheet; and (iii) finally, the addition of different percentages of TiO_2_ and menthol to the composite was evaluated while keeping the sheet homogeneous. Characterization tests were performed at all stages, and the most consistent formulation was selected at the end according to its physicochemical characteristics and antimicrobial activity, as shown in Figure 2.

Initially, PLA and PBAT pellets were mixed in four different compositions: 55/45, 70/30, 85/15, and 93/7 (PLA/PBAT) to assess the stability of the mixture during the molding process and to obtain a semi-rigid material [49]. The additives were incorporated as 2% (w/w) of the formulation (0.2% citric acid and 1.8% PEG) for each of the mixtures evaluated, similar to that proposed by Wasti et al. [50]. After performing the extrusion and thermoforming processes described below (Section 2.4 and Section 2.5), the formulation with the highest homogeneity and the best mechanical and moisture barrier properties (Section 2.6) was selected for the inclusion of the PF in the next stage.

In the second stage, the incorporation of plantain fibers (PFs) into the PLA/PBAT matrix was evaluated to determine their reinforcing capacity, specifically the improvement of the mechanical properties of the molded sheet and tray. The PFs were mixed into the selected formulation of the previous phase, at three concentrations (1, 3, and 5% w/w), similar to what was proposed in the study by Ibrahim et al. [15], and two particle sizes (40 mesh, <0.420 mm, and 100 mesh, <0.149 mm). Span 80 was added as a dispersing agent for the fibers at 0.5% (w/w) of the total mixture. The formulations evaluated are shown in Table 1.

The most favorable percentage of PF inclusion was determined by verifying the stability of the mixture during extrusion and thermoforming processes, and by performing measurements of mechanical, optical, and water vapor transmission rate (WVTR) as described in the property characterization section (Section 2.6). Similarly, comparisons were made with the mixture without fibers to determine the changes resulting from their inclusion.

Once the percentage and size of the PF were determined in the composite, the third stage involved defining the percentages of antimicrobial substances to be included. For this case, it was decided to use a combination of an antimicrobial agent with static activity, such as Titanium Dioxide (TiO_2_) [51], and a controlled-release preservative, such as menthol [24]. Six formulations were evaluated: (i) 0.5% TiO_2_, (ii) 1% TiO_2_, (iii) 1% menthol, (iv) 2% menthol, (v) 0.5% TiO_2_ + 1% menthol, and (vi) 5% menthol (Table 2).

The most favorable formulation was selected based on its mechanical, optical, and barrier properties, as well as its antimicrobial capabilities. Similarly, for specific samples, additional FT-IR, DSC, TGA, and SEM tests were performed. These measurements are described in more detail in Section 2.6 and Section 2.7, which cover property characterization and antimicrobial activity, respectively.

2.4. Mixing, Extrusion, and Calendering

To obtain the composite system for the different formulations, the polymer pellets (PLA and PBAT) were first dry-mixed with the powder additives (citric acid and PEG), using a homogenizer. Then, the antimicrobial agents (TiO_2_ and menthol) were added to this mixture according to the established formulation. Next, Span 80 was added, and finally, the plantain fibers were added. All components were mixed until a homogeneous mixture was obtained for the extrusion.

The extrusion stage was divided into two phases: (i) strands/pellet production and (ii) sheet production. Strand/pellet production was achieved using a Bimek model B40 single-screw extruder (Bimek Ltd., Bogotá, Colombia), with an L/D ratio of 28 and equipped with a die for producing cylindrical strands of 8 mm diameter. The temperature profile was set at 165/175/185 °C for the three heating zones in the barrel and 190 °C for the die, with a screw speed of 60 rpm. After extrusion, the strands were cooled by forced air and then cut into pellets of 5 mm length and 3 mm diameter.

The pellets obtained in the first phase were used to feed a second extruder of the same model and brand as the one used in the previous stage, but equipped with a die for obtaining flat sheets. The temperature profile and screw speed were set at 185/190/195 °C for the barrel and 205 °C for the die, setting an extruder screw speed of 60 rpm. The flat sheet die used was 21.5 cm long and had a lip opening of 1.04 mm. After extruding, calendering was performed in a machine equipped with water-cooled rollers, operating at a rotation speed of 3.0–5.0 rpm and a separation of 0.5 mm between the rollers, to produce homogeneous sheets with a thickness of 0.3–0.4 mm. The rotation speed of the rollers and the pulling speed were adjusted in such a way that sheets of similar thickness to a typical commercial PET or PLA tray were obtained.

2.5. Thermoforming Process

The resulting extruded sheet was cut into 26 × 30 cm rectangles for thermoforming semi-rigid trays using a Verpacken SN 954 vacuum thermoforming machine (Verpacken Ltd., Bogotá, Colombia). The process involved placing the sheet in the thermoforming frame and then heating it for one minute at an average temperature of 325 °C. The heated sheet was then brought into contact with the mold, and vacuum pressure (76 kPa) was applied for 40 s, allowing the sheet to take the shape of a semi-rigid rectangular tray (12.1 cm long × 10.4 cm wide × 3.2 cm high). Finally, the thermoformed piece was air-cooled for 30 s to solidify its shape, and the excess material was trimmed off after removing it from the mold.

2.6. Characterization Tests

As previously mentioned, to select the optimal PLA/PBAT base formulation, including the appropriate percentage and particle size of PF, as well as the addition of TiO_2_ and menthol, several physicochemical tests were performed as described below:

Thickness: For all samples of extruded sheets and thermoformed trays, the thickness was measured using a Mitutoyo^®^ Digimatic IP65 digital micrometer (Mitutoyo, Kawasaki, Japan), taking measurements at three different points on the sample and reporting the average value.

Mechanical properties: The tensile strength of the samples was evaluated using a Lonroy universal testing machine model LR C-001 (Dongguan Lonroy Equipment Co., Ltd., Dongguan, China). For this purpose, 2 × 10 cm flat sheet specimens were clamped in jaws and then separated at a constant speed of 3 mm s^−1^ until the material failed, following the ASTM D638-14 standard [52]. The peak force (N), tensile strength (MPa), elongation at break (%), and Young’s modulus (MPa) were reported.

The compressive strength of the thermoformed trays was determined using the same machine, in accordance with the ASTM D695-23 standard [53]. Each sample was placed in its normal position (opening facing upwards) between two flat plates 15 cm in diameter and then compressed at a rate of 5 mm s^−1^ until it reached 50% of its original height. The compressive strength (MPa) was reported for this test.

Moisture barrier: The water vapor transmission rate (WVTR) was determined using a gravimetric method in a controlled environment at 20 °C and 60% relative humidity [54]. Distilled water (6.00 ± 0.05 g) was added to a cylindrical aluminum cell (35 mm in diameter and permeation area of 0.00096 m^2^). Subsequently, the film was placed over the cell, and the cell was hermetically sealed so that water vapor could only permeate through the film. The WVTR (g m^−2^ d^−1^) and the permeability coefficient, Q_H2O_ (g mm m^−2^ d^−1^ atm^−1^), were recorded.

Optical Properties: Luminous transmittance (%) was estimated from the reflectance value, which was measured using a 3nh model YS3020 spectrophotometer (Shenzhen 3nh Technology Co., Ltd., Shenzhen, China) for a wavelength of 600 nm. “Daylight 65” was considered the standard illuminant with a 2° observer.

Yellowness Index (YI) and White Index (WI): This was calculated by using the formula given below [55]:

$$[eqn]$$ $$[eqn]$$

where L* is the lightness and a* and b* are the chromatic coordinates of the CIELAB color space.

Water absorption: The water absorption test was performed with square samples (20 × 20 mm) taken from each formulation and dried at 60 °C for 72 h; after this time, the initial weight of the samples was recorded, and they were stored in an incubator at 20 °C and 60% RH. After 24 h, the weight of the sheets was measured again, and the water adsorption was determined as the weight increase of the sheets relative to their initial weight.

Fourier transform infrared (FT-IR) spectroscopy: Similar to the analysis of the PF samples described above, FT-IR analysis was performed for all the sheet and tray samples by determining the absorbance spectra for each sample.

Thermal properties: The thermal stability of selected samples was evaluated by differential scanning calorimetry (DSC) using a TA Instruments Discovery series DSC25 calorimeter (TA Instruments, Newcastle, DE, USA), with a heat rate of 10 °C min^−1^, and a temperature range of −40 to 400 °C. Nitrogen was used as a purge gas for all the tests, with a constant flow rate of 40 mL min^−1^. Differential scanning calorimetry (DSC) was conducted to detect thermal events, such as phase transitions, crystallization, or other endothermic and exothermic reaction-derived events. Likewise, a thermogravimetric analysis (TGA) was also performed on a Discovery TGA550 thermal analyzer (TA Instruments, USA) with a heat rate of 15 °C min^−1^, and a temperature range of 30 to 950 °C. Nitrogen was used as a purge gas for all the tests, with a constant flow rate of 25 mL min^−1^. This test was performed to measure the mass loss of the material as a function of increasing temperature, allowing the identification of events related to moisture desorption and material decomposition.

Scanning electron microscopy (SEM): The surface morphology of selected sheets was evaluated using a scanning electron microscope Dual Beam System Tescan Lyra 3 (Tescan, Brno, Czech Republic) at an operating voltage of 5.0 kV with 90 s of emission. The surface of the samples was previously treated by applying a thin (circa 10 nm) gold sputter-coating. Some samples were cut with a fine blade to allow for cross-sectional imaging.

2.7. Antimicrobial Activity Tests

The packaging samples were evaluated for antifungal activity against Botrytis cinerea from strains provided by the Food Microbiology Laboratory (Universidad Nacional de Colombia, Bogotá, Colombia). For this purpose, an inoculum was prepared by mixing viable colonies of B. cinerea with physiological saline solution, adjusting the turbidity of the mixture to match a McFarland turbidity standard of 0.5 (10^6^ colony-forming units—CFU per ml). Subsequently, 0.500 mL of this solution was inoculated into Petri dishes containing PDA culture medium (Merck KGaA, Darmstadt, Germany). At this point, the packaging samples were positioned in the form of 1.5 cm diameter circles in the middle of the Petri dish, and the dishes were incubated for up to 4 days at 23 °C [56], with observations of B. cinerea growth being performed daily. A comparative analysis was conducted to examine the growth rate for the different formulations evaluated.

2.8. Statistical Analysis

For the characterization of the different tray formulations, an analysis of variance (ANOVA) was performed with a 95% significance level (α = 0.05) to determine the existence of significant differences in the measured properties. Significant differences between the mean values were determined using Tukey’s HSD test using Excel^®^ version 2511 (Microsoft Corporation, Redmond, WA, USA).

3. Results and Discussion

3.1. Obtention and Characterization of Plantain Microfibers (PF)

After the extraction process with alkaline and bleaching treatments, rigid, brittle, and thin fiber sheets were obtained after the forced convection drying process. The resulting sheets exhibited a whitish color and were composed of clusters of long, thin fibers intertwined with each other, with a moisture content of 8.65 ± 0.75% (w/w). After milling and sieving (first to 20 mesh, then to 40 mesh, and finally to 100 mesh), shorter microfibers with a particle size <0.149 mm were obtained. Microscopic analysis (at 50× magnification) revealed that the fibers have an elongated, filamentous, tubular structure, with a few amorphous structures around them corresponding to remnants of hemicellulose, lignin, or non-cellulosic compounds that were not completely removed during the extraction process. According to Ardila et al. [57], alkaline treatments are responsible for breaking the bonds between cellulose, hemicellulose, and lignin, enabling the extraction of the former. Throughout this process, the fibers gradually changed from dark brown to whitish due to the alkaline treatment and bleaching. This can be explained by the oxidation of the lignin’s chromophore components during bleaching without damaging the fiber structure [58,59]. The percentage yield obtained after the extraction process was 9.23% (Equation (1)), counting from the dried and ground plantain pseudostem to the final 100-mesh plantain microfibers (PF). This yield value is comparable to those obtained by Elhrari et al. [60], who evaluated a 4% (w/v) alkaline extraction method with NaOH and a 1.7% (w/v) bleaching method with NaClO_2_ for different plant resources, with yield values between 12 and 68%. The differences can be attributed to the several plant sources used in that study.

Supplementary Figure S1 shows the mid-infrared spectrum for the fibers obtained. Peaks are evidenced corresponding to the presence of hydroxyl groups associated with water absorbed by cellulose or hemicellulose (3318 cm^−1^), aliphatic C-H stretching of the glycosidic chain (2898 cm^−1^), presence of carbonyls of ester or acetyl groups characteristic of hemicelluloses and associated compounds (1735 cm^−1^), CH and CH_2_ deformations of the anhydroglucose unit of both lignin and cellulose (1426, 1369, 1318 cm^−1^) and presence of β-1.4 glycosidic bonds corresponding to the C-O-C bonds of the pyranose rings (1156, 1025, 899 cm^−1^). The higher intensity of the peak corresponding to the wavelength of 1025 cm^−1^ suggests the predominant presence of a polysaccharide structure typical of cellulose, compared to the lower vibration in the regions corresponding to lignin (1618 cm^−1^) and hemicellulose (1735 cm^−1^). Accordingly, it can be inferred that the fibers obtained have a high cellulose content after the removal of the other components through the chemical treatments used. The spectrum shown in Figure S1 is similar to that obtained by other authors for cellulose fibers extracted from agricultural waste [61,62].

3.2. Obtention of Compounded PLA/PBAT Sheets and Trays

In the first-phase extrusion tests, the amount of PBAT in the mixture was inversely related to the stiffness of the sheets obtained after calendering. PLA/PBAT ratios lower than 55/45 were not included because higher amounts of PBAT resulted in extruded sheets that were too flexible and thermoformed trays that were too flimsy. In general, for the different PLA/PBAT ratios evaluated, the dispersion and homogeneity of the strands and the extruded sheets were acceptable. However, this uniformity was higher when the amount of PBAT was reduced, with ratios of 85/15 and 93/7, achieving higher surface homogeneity. This is similar to that observed by Li et al. [63] for mixtures in this PBAT range, where its dispersion in the PLA matrix is uniform. For the 93/7 ratio, the resulting sheet was more rigid during calendering and thermoforming, which can be explained by the potential of PBAT to promote PLA crystallization during sheet cooling in calendering, acting as a nucleating agent at low concentrations [64]. In all cases, opaque sheets with a pearly white tone were obtained. Extrusion temperatures were adjusted in preliminary tests until the values mentioned in Section 2.4 were defined. Higher temperatures led to excessive softness in both strand and sheet extrusion. Likewise, lower temperatures hindered proper material output and resulted in poor homogeneity and continuity problems (hole formation, air pockets, and significant thickness variations). Regarding the additives, the specified amounts of PEG and citric acid were satisfactory, resulting in adequate mechanical strength, flexibility, and sheet homogeneity during extrusion and calendering, given their plasticizing (PEG) and compatibilizing (both) capabilities [65]. Increasing the PEG content made the mixture too fluid during extrusion and resulted in greater sheet elongation during calendering, while higher amounts of citric acid led to less dispersion and discontinuous output of the mixture through the extruder’s die. Lower amounts or the absence of these additives resulted in lower dispersion and homogeneity during extrusion and calendering in preliminary tests.

After obtaining the extruded sheets and thermoformed PLA/PBAT trays, the mechanical, barrier, and optical characteristics shown in Table 3 were determined. Regarding the extruded sheets, a direct relationship was observed between stiffness and PLA content, indicating that increasing the proportion of this polymer provides greater strength and lower ductility [66]. The formulation with a PLA/PBAT ratio of 85/15 exhibited the best mechanical performance, with the highest tensile strength (49.06 MPa) and the lowest elongation at break (2.30%). This combination of properties allows for obtaining sheets with good tensile strength and minimal deformation. Similarly, Lyu and Han [67] found that PLA sheets exhibited higher stiffness and less flexibility than those made of PBAT. Regarding the compression tests of the thermoformed trays, higher maximum forces and compressive stress were obtained as the proportion of PLA in the mixture increased, with the highest values for the 93/7 formulation. Nonetheless, no significant differences for compressive strength were found between the formulations evaluated.

Regarding barrier properties, it was observed that increasing the proportion of PLA in the mixture decreased the WVTR, which is consistent with the findings reported by Wang et al. [68], who demonstrated that a higher PLA proportion in a PLA/BAT mixture increases the moisture barrier. This behavior may be because PBAT is more flexible and less crystalline than PLA, creating more pathways for water vapor to permeate [62]. Likewise, a similar trend was observed in moisture absorption, with higher values for formulations with a higher PBAT proportion. However, a higher moisture content was also observed for the formulations with more PLA content, which may be because the resulting blend can still hold or absorb more moisture overall because of the spaces left by PLA’s brittleness and the flexibility of PBAT, which allows for more space in the structure to be occupied by water molecules [69]. On the other hand, the sheet’s transmittance increased with increasing the proportion of PLA in the mixture, and the whiteness index decreased, which corresponds to the difference in light barrier and opacity between PLA and PBAT and to light scattering due to the dispersion of PBAT in the PLA matrix in separate phases [70].

In the overall analysis, the 85/15 mixture proved to have the best balance between homogeneity, mechanical properties, and barrier properties, standing out especially for its higher tensile strength and lower moisture absorption. Therefore, subsequent tests were conducted based on this PLA/PBAT ratio.

3.3. Sheets and Trays with Plantain Microfiber (PF) Inclusion

After defining the most favorable PBAT/PLA ratio, different inclusion percentages of PF at two particle sizes were evaluated (Table 1). For these formulations, a constant concentration of 0.5% (w/w) Span 80 was added to improve fiber dispersion in the mixture [71].

In the first extrusion, strands with varying degrees of heterogeneity and light brown tones were obtained, depending on the amount of fibers incorporated into the mixture. In all cases, a lower degree of flowability was observed compared to the base formulation with an 85/15 PLA/PBAT ratio (Control in Table 1). The same was observed for the sheet extrusion, where all the formulations with fiber inclusions had continuous but lower flowability compared to the control formulation. For the 1% and 3% (w/w) PF formulations, adequate dispersion of the mixture was observed, although some holes were also evidenced, especially for formulations with 40-mesh fiber (<0.400 mm). These homogeneity deficiencies increased with the fiber percentage, and for the 5% fiber formulation, the sheet homogeneity decreased considerably, with agglomerations and large holes observed, in addition to localized thickness differences due to variations in the extruder die’s exit flow. The above is similar to what was reported by Haafiz et al. [72], where the increased inclusion of cellulose microfibers resulted in agglomerations and irregular protrusions in PLA sheets due to the low compatibility between these fibers and the PLA matrix. Given this, it was not possible to obtain thermoformable sheets from the PF 5-40 and PF 5-100 formulations, so no physicochemical characterization of these formulations was conducted.

To maintain a uniform sheet thickness between 0.41 and 0.44 mm (Table 4) for the different formulations, it was necessary to increase the calendering speed, starting at 6.0 rpm for the PLA/PBAT base formulation (Control) and up to 6.5 rpm for the PF 3-100 and PF 3-40 formulations. This can be explained, as mentioned previously, by the reduction in the fluidity and increase in viscosity of the extruded mixture due to the fiber inclusion, which made it necessary to compensate by increasing the roller speed to avoid the formation of thicker sheets [73]. In thermoforming, stable trays without breakage were obtained for the PF 1-40, 1-100, 3-40, and 3-100 formulations, as well as the fiber-free PLA/PBAT control. However, the homogeneity of the trays with the smaller particle size PF was higher compared with the formulations with 40 mesh (<400 mm) microfibers. In this case, the surface of the sheets and thermoformed trays appears rougher, again due to the lesser dispersion of these larger fibers in the polymer matrix, resulting in more agglomeration.

Regarding the inclusion of fibers, a reduction in tensile strength and Young’s modulus was identified for the PF-included formulations presented in Table 4 compared to the control base mixture. This reduction in tensile strength was less pronounced for the PF 3-100 formulation, where no significant differences were observed. Likewise, an increase in the flexibility of the PF sheets was evidenced, with higher elongation at break, except for the PF 3-40 formulation. This can be explained by the contribution to tensile strength that including the microfiber reinforcement can provide, considering the fibers are in the appropriate proportion and particle size [19]. Very large particle sizes do not allow for good interaction with the polymer matrix, leading to weaker sheets due to stress concentration points and poor adhesion, while very small particle concentrations have insufficient coverage and do not allow for generalized interaction and reinforcement of the polymer matrix [74]. Regarding the compression tests of the trays, a significant increase in peak force and compressive strength was observed for the PF 3-40 and 3-100 formulations (compared to the control and the formulations with 1% PF), going from 0.007–0.009 MPa to 0.009–0.013 MPa. This can be attributed to the fiber’s contribution in bridging cracks, improving stress transfer, and enhancing overall stiffness [75].

For WVTR and moisture absorption, a significant increase was observed in the measured values for the PF formulations compared to the control formulation, rising from 2 to 4.9–7.2 g m^−2^ d^−1^ at 20 °C and 60% RH. This can be attributed to the higher affinity and hydrophilicity of the fibers compared to PLA and PBAT, forming pathways for water uptake by creating hydrogen bonds with water molecules [76]. This may also be related to the effective dispersion of the fibers within the polymer matrix. In the case of the PF 1-40 and PF 3-40 formulations, with their larger fibers, these are more exposed on the sheet surface, leading to greater moisture absorption and permeation compared to the PF 3-100 formulation, for example, which has smaller and better-dispersed particles, resulting in a lower WVTR value (4.984 g m^−2^ d^−1^) compared to the other formulations.

In general, transmittance values increased slightly with decreasing particle size and increasing plantain microfibers (PFs), although no significant differences were observed. This can be explained by the fact that the inclusion of smaller fibers results in less light scattering and more light transmission through the sheet, although there are no significant differences with the base mixture without PF, as already mentioned. Regarding the yellowness index, it was observed that with a higher fiber content, there is a transition from a slightly brownish white to more yellow tones, which is more pronounced at the 100-mesh particle size due to the better dispersion within the PLA/PBAT matrix.

As a result of this phase of incorporating plantain microfibers into the PLA/PBAT matrix (85/15), the PF 3-100 formulation exhibited the overall best characteristics in terms of homogeneity, dispersion, mechanical strength, barrier properties, and moisture absorption. Therefore, this mixture was used as the base for incorporating antimicrobial active substances.

3.4. Sheets and Trays with Inclusion of Menthol and TiO2

In this stage, the active materials were added according to the experimental design in Table 2, initially evaluating their effect on the structure of the extruded sheet and the thermoformed trays. TiO_2_ levels of 2% (w/w) or higher caused losses in flow continuity through the sheet extruder and reduced the sheet’s mechanical strength during calendering. Regarding menthol, partial phase separation in the dispersion began to be observed at 3–5%, with an initial very fluid flow of part of the mixture, possibly rich in menthol, followed by a slower flow of a more viscous material. For 3% and 5% menthol in the formulation, this phenomenon was brief, while for higher concentrations evaluated, it lasted longer, with greater differences in flowability during extrusion. This can be attributed to an excess of menthol that cannot be dispersed in the composite. Similarly, when attempting to test higher menthol concentrations than in formulation M5 (Table 2), the extruded sheet exhibited large, repeated holes attributable to possible evaporation of the excess menthol, so no further work was attempted with this formulation. In the thermoforming process, on the other hand, stable trays with acceptable homogeneity were obtained for all the formulations evaluated (See Supplementary Figure S2).

As shown in Table 5, extruded sheets with thicknesses between 0.358 and 0.386 mm were obtained, with no significant differences between the formulas containing menthol, TiO_2_, and the control PF (fibers only). In this case, the calendering speed had to be increased from 6.5 rpm with only 3% PF (control PF) to 7.0 rpm to maintain a roughly constant thickness and compensate for the inclusion of the active substances, which resulted in less fluidity of the mixture at the extrusion die exit. Regarding mechanical properties, the inclusion of antimicrobial compounds showed a reduction in the mechanical strength of the sheets, reaching minimum tensile strength values of 18.03 MPa for formulation T0.5M1 compared to 47.42 MPa for the control PF sheet. The higher inclusion of menthol in the sheets also generated a decrease in tensile strength, with values of 30.84, 27.77, and 21.94 MPa for formulations M1, M2, and M5, respectively. The inclusion of these agents reduces the interactions between the polymer chains, causing a loss of rigidity and resistance in the sheets [77]. Scaffaro et al. [78] reported that the inclusion of rigid fillers such as clays in films made with biodegradable polymers helped to increase the elastic modulus; however, when adding an antimicrobial component such as carvacrol with a lower melting point compared to the matrix polymers, this property decreased. Regarding the compression test, similar or higher values of peak force and compressive strength were evidenced for the formulations with the inclusion of only TiO_2_ (T0.5 and T1) compared to the control PF (only fibers), while there was a reduction in these mechanical parameters for the formulations with the inclusion of menthol reaching values lower than 100 N of peak force and 0.006 MPa of compressive strength.

On the other hand, the inclusion of antimicrobial agents influenced the water vapor transmission and water absorption properties. There was an increase in the WVTR by including these components compared to the sheets that did not have antimicrobial agents (control PF), going from 4.98 g m^−2^ d^−1^ to 5.42–7.90 g m^−2^ d^−1^. The highest value was observed for formulation T1 with 1% TiO_2_, possibly due to the lack of dispersion and compatibility with the polymer matrix at this concentration. This corresponds to that reported by Noori et al. [79], who showed that the inclusion of plant essential oil components in PLA films decreased their barrier properties, making the material more permeable. As for sheets with menthol, the greater water absorption can also be attributed to the presence of the hydroxyl group in its chemical structure, which allows it to form hydrogen bonds with water, promoting higher moisture retention [80].

The transmittance values do not show a defined trend with respect to the content of antimicrobial agents, with a slight increase in transmittance being observed for the formulations M2, T0.5M1, and M5 with menthol compared to the control PF, although there is no direct relationship between the concentration of active substance and the first. Likewise, a reduction in transmittance was observed for the T0.5 and T1 formulations (27.170% and 25.065%) compared to the control PF (28.819%), which may be due to the greater light scattering caused by the TiO_2_ in the PLA/PBAT mixture [81]. Regarding the Yellowness index, no clear trend was observed either, although formulations M2, T0.5M1, and M5 recorded more negative values than the control PF, indicating lower yellow tones.

3.5. Spectroscopy Analysis, SEM, and Thermal Behavior

Figure 3 shows the infrared spectra for the base formulations: PLA/PBAT 85/15 (control in Table 1), PF 3-100 (control PF in Table 2), T0.5M1, and M5. In all spectra, vibrations are evidenced in the regions corresponding to the symmetric and asymmetric CH_3_/CH_2_ stretching (2997–2946 cm^−1^) and the characteristic vibrations of the ester groups (1750 cm^−1^), indicating the preservation of the chemical structure of the PLA/PBAT system. In this case, any degradation in the polyester matrix, mainly of PLA, would be identifiable within the band range (1755 cm^−1^), but this does not appear to be the case for the different formulations evaluated when observing Figure 3 [82]. Additionally, in formulation M5, higher vibrations are evidenced in the bands 2917 and 2849 cm^−1^ compared to the formulations without including antimicrobials. This intensification may be associated with C-H stretching, a characteristic behavior for pure menthol [83], which indicates the presence of this antimicrobial in sheet M5. In fact, similar peaks are observed for the pure menthol spectrum analyzed. On the other hand, formulations with 3% banana fibers (control PF, M5, T0.5M1) presented a peak around 1081 cm^−1^ attributed to C–O stretching and characteristic of the cellulose spectrum, indicating the presence of this polymer in the sheets [84], due to the inclusion of PF. Finally, the intensification of the band at 734 cm^−1^, together with the increase in absorbance in the C–O/C–O–C region (1270–1048 cm^−1^), points to physical changes in the packing and organization of the polymer chains, probably induced by the presence of PF or TiO_2_.

Figure 4 shows the SEM micrographs of the morphology of selected extruded sheets. Figure 4a shows a smooth surface, and the PLA-PBAT transition is identifiable by the formation of longitudinal channels. In Figure 4d, a cross-section shows some agglomeration points, but overall, a good dispersion between the two polymers, indicating acceptable compatibility with PEG, as previously mentioned. In the composite samples, plantain fibers 200 to 350 µm in length (Figure 4b,c) and 10 to 40 µm in diameter (Figure 4f) can be identified. These are distributed on the surface (Figure 4b,c) as well as within the sheets, as evidenced in the cross-section micrographs (Figure 4d–f). The PFs are identifiable by their elongated shape, dispersed within the PLA/PBAT polymer matrix, with no agglomerations observed within the sheet. Although the compatibility of the fibers with the PLA/PBAT matrix is poor due to polarity differences, resulting in weak interfacial bonding [15], the addition of PEG helped overcome this deficiency and improve the interfacial bond between the fibers and the polymer matrix, enhancing interfacial adhesion by softening and increasing the mobility of the cellulose, thus improving compatibility with the PLA/PBAT [85]. Citric acid also contributes to improved interfacial adhesion of the composite, acting as a compatibilizer and crosslinking agent, leading to stronger bonds between the fibers and the polyester matrix (PLA/PBAT) [86]. Likewise, the addition of Span 80 to the mixture allowed for adequate dispersion of the different components in the extruded sheet by acting as an additional compatibilizer [87].

Figure 4b,e show the sheets with the inclusion of TiO_2_, with some particles dispersed both on the surface and within the sheet. The size difference between the TiO_2_ particles and the microfibers is also evidenced, with the former being much smaller (20–30 µm in diameter with a granular shape). Figure 4c,f show points of crystallized menthol on the surface. No menthol dispersion is observed within the polymer matrix, suggesting that some of the menthol is exuded to the outside of the sheet due to its lower melting point, higher mobility, and volatility compared to the other components of the composite [88].

In general, the results obtained regarding the dispersion of the reinforcing components in the composite sheets can be compared with other studies such as that of Olonisakin et al. [89] who evaluated the addition of epoxidized soybean oil and myristic acid as an agent to improve the compatibility and properties of the PLA/PBAT matrix and bamboo fibers, showing good results in that regard.

Figure 5 shows the DSC and TGA curves for selected formulations. In Figure 5a, the DSC thermograms are shown, identifying in the three samples (from left to right) an initial drop and negative peak associated with the glass transition temperature (T_g_), a second positive peak, possibly associated with the cold crystallization temperature (T_cc_) corresponding to PLA [90], and a third negative peak associated with the melting point (T_m_) of the mixture. A fourth negative peak, only seen in the control PF and M5 (252.86 and 249.16 °C, respectively), could correspond to the boiling point (T_b_) of the PEG. Finally, for all three samples, a broad negative peak is observed at the end corresponding to T_d_ (total degradation of the mixture). Regarding the effect of microfibers on the thermal properties of the composite, a reduction in the temperatures associated with thermal events (T_g_, T_cc_, T_m_, and T_d_) specifically is observed for the PLA/PBAT compound (Control 85/15). The inclusion of plantain microfibers (PF control) and menthol (M5) leads to a decrease in T_g_, being 55.21 °C and 44.22 °C, respectively, compared to the PLA/PBAT compound, whose T_g_ was 59.84 °C. On the other hand, the degradation temperature was 370.35 °C, and the inclusion of 3% PF and menthol decreased the T_d_ to 347.03 °C and 349.39 °C, respectively. These results are similar to those reported by Kostenko et al. [82], who demonstrated a reduction in the peak temperature of the PLA/PBAT matrix when APTES (3-Aminopropyl triethoxysilane) functionalized clays were included. In both cases, this reduction in thermal stability can be attributed to a disruption of the polymeric matrix, a weak interface, and accelerated degradation by the inclusion of PF and menthol. This effect has been evidenced by Srisa et al. [46] in which the inclusion of antioxidant agents in polymers such as PLA reduces their thermal stability and T_g_.

Figure 5b shows the TGA results, revealing a very low mass loss rate of just over 1% for the control sample containing only PLA/PBAT until the onset of material decomposition at 320.47 °C. This corresponds to the higher PLA proportion in the formulation, as determined in other studies [91]. For the PF control sample (with 3% PF 100 mesh), a slightly higher loss rate is observed, with a cumulative mass loss of circa 2% at the onset of degradation at 303.11 °C. For this formulation and for sample M5, a slight decrease in the slope of the weight curve is observed between 70 and 90 °C, which may be associated with moisture evaporation in these samples. For formulation M5, a second drop in slope is observed between 130 and 150 °C, which could be attributed to menthol evaporation in the composite. This is also associated with a more pronounced mass loss compared to the other samples analyzed, with a loss of approximately 6% at the onset of the degradation temperature at 307.03 °C. These results correspond with the DSC thermograms, where the inclusion of the PF and menthol leads to lower thermal stability of the composite.

3.6. In Vitro Antimicrobial Activity Test

Figure 6 shows the in vitro growth of B. cinerea for the treatments specified in Table 2, including the antifungal agents (menthol and TiO_2_) as well as the baseline control of PLA/PBAT 85/15 without PF, over 4 days. In all cases, B. cinerea growth was observed in the PDA medium, but differences were noted in the growth rate and the growth on the composite sheet samples themselves. No microbial growth was observed for any of the samples until day 2. By day 3, growth began to appear around the sheets, although less pronounced in the samples containing menthol (M1, M2, and M5). For the control sample PLA/PBAT 85/15 (without PF), much more pronounced growth was observed by day 3. At this point, colonies began to form on the edges of the sheets in all formulations except M5, and very few in M2. On day 4, abundant growth was present on the edges of all samples except formulation M5, where growth was much less pronounced. Likewise, fungal colonies (dark spots) appeared on the surface of the sheet itself in all samples except M5, with a higher number in the control samples, the PF control, and those containing TiO_2_.

The inclusion of menthol in the composite resulted in a lower growth rate of B. cinerea in the PDA medium around the sheet, although a clearly defined inhibition zone was not observed. This can be because the concentrations tested were not high enough to diffuse in this medium. Nonetheless, partial inhibition of fungal growth on the sheet surface was observed for the samples with menthol, at least until day 3 for formulations M2 and M5, and beyond day 4 for sample M5. These results are consistent with the study by Castellanos-González et al. [24], who identified a reduction in the incidence of spoilage caused by B. cinerea and an increase in shelf life for blueberries packaged in microperforated trays with starch-based sheets adhered to the packaging and loaded with menthol, thus extending the shelf life of this fruit. Regarding TiO_2_, it did not prove to be as effective in this study, which can be attributed mainly to the particle size evaluated and the need for activation of the material by UV light or visible light to increase its effectiveness [92]. Another possibility is doping with other components to increase the antimicrobial activity of TiO_2_.

The tests performed showed that it is possible to impart antifungal activity to PLA/PBAT composite sheets by including 5% (w/w) of menthol, slowing the in vitro growth rate of Botrytis cinerea in the evaluated PDA medium, and exhibiting fungistatic activity that inhibited growth on its surface. In vivo studies are necessary to establish the antifungal efficacy of trays reinforced with PF and including menthol in perishable foods susceptible to spoilage by this fungus, and thus to determine the potential application of the composite packaging in extending the shelf life of these products under typical storage conditions.

4. Conclusions

Based on sheet extrusion and thermoforming tests, it was possible to establish a formulation based on a mixture of PLA and PBAT, along with active materials and reinforcements. Process conditions (temperature and speed) were adjusted to obtain stable and homogeneous sheets and trays with a size and shape similar to commercial petrochemical plastics.

The inclusion of up to 3% (w/w—mesh 100) of previously extracted plantain microfibers (PF) resulted in a stable and homogeneous composite from which sheets and fibers with superior mechanical strength were obtained compared to the base PLA/PBAT formulation. A higher PF percentage resulted in a loss of homogeneity and lower mechanical strength during the extrusion and calendering process of the sheets.

In the incorporation and evaluation tests of the active compounds, menthol led to a higher decrease in the growth of Botrytis cinerea compared to formulations with TiO_2_ and with only PF. The greatest inhibitory effect for the in vitro tests was obtained with a 5% (w/w) menthol concentration. Higher percentages led to a significant loss of mixture stability and poor dispersion during extrusion.

The findings are promising for the packaging of perishable products affected by spoilage caused by B. cinerea, such as perishable foods like berries, which are typically packaged in semi-rigid, non-biodegradable trays made of PET or PP. This opens the possibility of conducting further in vivo analyses to determine the interaction of the active composite trays with these products and their effect on the quality properties.