Bio-based edible packages from cassava (Manihot esculenta) peel starch incorporating amla (Phyllanthus emblica) pomace microparticles for palmyrah fruit leather

Abstract

Environmental consequences arising from the use of petrochemical-based packages have prompted the development of eco-friendly packages. The present study was conducted to develop a bio-based edible package for palmyrah (Borassus flabellifer L.) fruit leather (PFL) and investigate its role as a protective sheath. The films were synthesized with varying starch contents (60%–90% w/w) along with the incorporation of spray-dried microcapsules of amla pomace extract (APE) at varying concentrations (5%–20% w/w), and analyzed for thickness, mechanical, and barrier properties. The results showed a significant impact (p < 0.05) on the properties of the films caused by the differences in the amount of starch and the concentration of microcapsules. The four best films (F8, F11, F12, and F16) were selected for further physicochemical, morphological, thermal, and sensory analyses. From the results of the sensory evaluation, predominantly F8 was selected and optimized as the best fit. F8 and F5 (control) were used to pack PFL, and the packed products were stored at 25 °C for 50 days to observe their efficacy in protecting the properties of the PFL during storage. F8 showed a significant favorable impact (p < 0.05) on the PFL, in terms of better protection against weight loss, and many physicochemical property variations compared to F5. Furthermore, it was observed that F8 exhibits potential antimicrobial effects to reduce microbial growth over 50 days of storage. Overall, it can be concluded that cassava peel starch with the addition of microencapsulated APE can be utilized as an active package for PFL.

---

1. Introduction

To meet the high demand for sustainable food packaging and customer preference for safe and environmentally friendly packaging materials, emerging packaging materials based on edible biopolymers have recently gained significant interest.¹ Edible packaging is a packaging material designed to be consumed with the food it wraps.² A key characteristic feature of edible films is the incorporation of biopolymers that enable the films to possess biodegradable nature, which is absent in conventional petroleum-based synthetic packaging materials. More importantly, the versatility of edible packaging allows it to be adapted to a wide range of food products. Such packaging can be produced as active materials to retard the growth and proliferation of microbes, which is an appreciable attribute for food safety.³

As a current trend, various edible fruit and vegetable waste materials like pomace, seeds, and peels have become major constituents in the production of edible packages due to their excellent matrix-forming capabilities and structural durability.⁴ In this context, cassava peel is an agro-industrial waste produced on a large scale globally, which is mostly discarded without any further use, but contains a significant amount of starch, highlighting its potential to be applied in various applications.⁵ Recent work by Fadhallah et al. further supports the valorisation of cassava peel starch for bioplastic development, demonstrating good mechanical and environmental properties suitable for food packaging.⁶ The starch in cassava peels is made up of amylopectin and amylose at a ratio of over 70 : 30.⁷ The starch found in cassava peels can thus be used as a raw material to produce edible films and coatings. Its high amylopectin content and matrix-forming ability have been shown to contribute to favourable physical properties in bio-based films.⁸ Furthermore, using them as gel-forming agents is advisable because of their low gelatinization temperature, gel stability, and clarity.⁹

The petrochemical-based plastic low-density polyethylene (LDPE) has been widely employed as the primary commercial packaging material for food products, leading to the increasing municipal waste disposal issue and severe environmental impact due to its non-biodegradable nature.¹⁰ There is a growing demand for innovative packages, such as active and intelligent packaging, which act as sustainable and environmentally friendly alternatives for packing food products, including PFL.¹¹ Palmyrah (Borassus flabellifer L.) is extensively cultivated in South and Southeast Asia, particularly in Sri Lanka and India, where its cultivation is deeply connected to the social and economic factors of the population.^12,13 PFL is a traditional fruit-derived product rich in nutrition, despite the fact that it is highly susceptible to poor storage quality due to moisture migration, microbial attack, and structural changes.^14,15 Conventional synthetic polyethylene packaging is widely used to wrap PFL, but it fails to minimize microbial growth and raises environmental concerns.¹⁶

Amla pomace is a residual byproduct obtained from the processing of amla. It contains abundant bioactive compounds, such as polyphenols, flavonoids, tannins and essential oils, which exhibit strong antimicrobial and antioxidant properties.^17,18 Recent studies have investigated its application in the development of food packaging to enhance food safety and prolong shelf life. In one study, incorporating amla pomace into sugarcane bagasse-based paper, with titanium dioxide (TiO₂) nanoparticles, has been shown to provide effective antimicrobial activity against microbes like Escherichia coli, highlighting it as a promising material for sustainable food packaging.¹⁹ However, direct incorporation of extracts into packaging films would not provide the expected outcome due to the instability and rapid degradation of the active compounds. Therefore, microencapsulation via the spray-drying technique would allow stable incorporation into edible films and controlled release of amla pomace extract, strengthening its bioactivity during the storage period.

Concerning the reported antimicrobial effect of amla (Phyllanthus emblica) and the significant starch content in cassava peel, the current study reports the preparation of bio-based edible packaging films from cassava peel starch incorporating amla pomace microparticles into the model food system PFL.

The main objective of this study was to develop and evaluate a bio-based edible packaging film derived from cassava peel starch, incorporating spray-dried microparticles containing amla pomace extract, for the preservation of PFL while minimizing microbial attack. The novelty of this work lies in two aspects: (i) value addition of cassava peel, an underutilized agro-industrial waste, as a sustainable starch source for edible packaging, and (ii) incorporation of amla pomace microparticles as a natural source enriched with antimicrobial and antioxidant properties to enhance the functional performance of the films. Unlike previous studies that focused on cassava starch from tubers^20,21 or synthetic additives for film reinforcement,²² this study demonstrates a circular bioeconomy approach by simultaneously addressing food waste management and the development of an active packaging material tailored for a traditional fruit-based product.

Most edible film studies use cassava tuber starch, tapioca, corn, potato, and rice as film-forming materials,²³ where cassava peel is often discarded as waste. Only a handful of recent studies have explored the potential of cassava peel starch for bioplastic/film formation, which have mainly focused on general packaging and have scarcely explored edible food wrapping development.²⁴

2. Experimental

2.1. Materials

Cassava peel was provided by the local cassava chip producers; amla pomace was collected from local amla juice producers. Furthermore, food-grade gelatin powder (Type B, HiMedia), 100% glycerol, maltodextrin, acetic acid, 96% ethanol, distilled water, Folin–Ciocalteu reagent, gallic acid, DG-18 medium, NaCl, plate count agar medium, and chloramphenicol were also used. All the reagents were of analytical grade and provided by Palmyrah Research Institute (Jaffna, Sri Lanka).

2.2. Apparatus and instruments

A UV-visible spectrophotometer (Thermo Scientific GENESYS 10S UV-VIS, Waltham, USA), a spray dryer (SP-1500), a scanning electron microscope (ZEISS Evo 18, Oberkochen, Germany), a universal testing machine (Testometric M500-50CT, Rochdale, UK), a hot-air oven (Thermo Scientific, model 666, Waltham, USA), an electric grinder, an electric balance (Adventurer Pro, AV64C, Parsippany, USA), an FT-IR spectrometer (Thermo Scientific Nicolet iS20, Madison, USA), a thermogravimetric analyzer (TGA 5500, New Castle, USA), a thermostatic water bath (WNE22), a magnetic stirrer (IKA C-MAG HS7, Staufen, Germany), a desiccator, a laminar flow hood (BSC-1300II A2-X, Jinan, China), a stomacher (Interscience BagMixer 400CC, Saint Nom La Bretèche, France), a colony counter (Galaxy230), a vortex meter (VELP ZX3, Usmate Velate, Italy), an incubator (BIOBASE IN110, Jinan, China), a colorimeter (NR20XE, Shenzhen, China), a pH meter (edge HI2020, Woonsocket), a thickness gauge, a texture analyser (IMADA, Northbrook, USA) and essential consumables like Petri dishes, plastic trays, beakers, etc. were used in this study.

2.3. Methodology

2.3.1. Isolation of starch from cassava peel. Initially, good-quality cassava peels were collected and rinsed using tap water to remove impurities on the surface. Then, the peels were cut into small pieces, and water was added to twice the volume of the peels. After that, they were pulverized together for 5 min using an electric blender. The mixture was then filtered through muslin cloth and allowed to settle for a day. The obtained sedimented starch was subjected to oven drying at 60 °C for 3 h and ground into a powder. Finally, the dried starch was stored in air-tight glass bottles.²⁵

2.3.2. Preparation of amla pomace extract (APE). The amla pomace was washed with distilled water. Then, the sample was kept at 60 °C for 24 hours in a hot air oven. After that, the sample was powdered very finely with an electric grinder and kept again at 60 °C in the hot-air oven for another 24 hours. 10% (w/v) of the powder was added to distilled water, and the extraction process was conducted at 50 °C for 3 hours using a thermostatic water bath. Finally, the supernatant was separated using Whatman no. 1 filter paper.²⁶

2.3.3. Microencapsulation of APE using spray-drying. The film-forming solution was prepared using a reported method in the literature with slight modifications.²⁷ In brief, the APE and 10% (w/v) of maltodextrin had a core/wall material volume ratio of 1 : 2. The solution was homogenized at 25 °C at 300 rpm for 1 hour using a magnetic stirrer. Then, the solution was spray-dried at a flow rate of 8 mm min⁻¹ at the inlet and outlet temperatures of 160 °C and 85 °C, respectively, with a constant pressure of 6.5 bar. Finally, the resulting powder was collected in a glass vial covered with aluminium foil and kept at 4 °C.

2.3.4. Preparation of the edible film. Initially, cassava peel starch (60, 70, 80, and 90% (w/w)) and gelatin (40, 30, 20, and 10% (w/w)) with a total weight of 5.0 g were dissolved in 100.0 mL distilled water using a magnetic stirrer at 80 °C for 4 h. The concentrations of the cassava peel starch and gelatin were selected based on the previous studies.^28,29 After cooling, the microcapsules of APE with concentrations of 5, 10, and 20% (w/w) were added and homogenized at 9000 rpm for 5 min. Next, 2.0 g of glycerol was added to that mixture, and the stirring was continued for 20 min. Then, the resulting film-forming solution was poured onto the sterilized plastic plates and dried at 35 °C in an oven for 24 hours.³⁰ Finally, the films were peeled off. The compositions of the prepared films are presented in Table S1.

2.3.5. Characterization of the prepared films.
2.3.5.1. Thickness. The thickness of the film samples was measured using a thickness gauge at five different locations.³¹
2.3.5.2. Mechanical properties. The mechanical properties, including tensile strength and elongation at break, were analysed using a universal testing machine with a loading capacity of 100 kN. The analysis was conducted by adapting the procedure of Karakuş et al., with some modifications.³² Firstly, the films were sliced into pieces of 20 mm × 80 mm (width × length). For the analysis, the universal testing machine was set up with a testing speed of 10 mm min⁻¹ and an initial gauge length of 50.0 mm.
2.3.5.3. Water vapor transmission rate. The water vapor transmission rate was assessed using the gravimetric technique. A glass container was filled with 10 grams of dry silica gel and securely covered with an edible film. The sealed setup was placed inside a desiccator containing saturated sodium chloride (NaCl) solution, maintaining a relative humidity of 75% at 27 °C. The container was separated from the setup at intervals, and its weight was measured. Finally, the water vapor transmission rate was calculated using eqn (1):³³where W, t and A denote the weight change (g), time (days), and test area (m²), respectively.
2.3.5.4. Light transparency. The light transparency of the films was analysed using a UV-Vis spectrophotometer at a wavelength of 600 nm. At first, the films were sliced into pieces of 3.5 cm × 0.5 cm (length × width). Next, they were placed in a cuvette, and the absorbance values were measured at 600 nm. An empty cuvette was used as the control. The film transparency for visible light was determined at 600 nm. Transparency values were calculated using eqn (2):³⁴where x and T denote the thickness of the film (mm) and the fraction transmission at 600 nm.

The best four films were selected based on the mechanical and barrier properties of 16 films. The following tests were performed for the four films.

2.3.5.5. Moisture content. About 2.0 g of the edible film sample was taken in a previously labelled and dry-weighed Petri dish. The moisture content of the film sample was measured as the difference in sample weight before and after drying at 105 °C until reaching a constant weight. Then the dried sample with a Petri dish was placed in a desiccator to cool down. After that, the total weight of the film sample and the Petri dish was recorded. The moisture content of the edible film samples was calculated using eqn (3):³⁰
2.3.5.6. Water solubility. The film samples were sliced into squares of 4 cm². The sliced samples were weighed and immersed in 50.0 mL distilled water, with periodic stirring for a day. Then the samples were collected, placed in a previously dried and weighed Petri dish, and then the Petri dish was kept in a hot air oven at 105 ± 1 °C until a constant weight was obtained. The Petri dish with the sample was placed in a desiccator to cool down, and finally, the total weight of the Petri dish with the sample was recorded. The water solubility of the edible film samples was calculated using eqn (4):³⁵
2.3.5.7. Water activity. Water activity was measured using a Novasina water activity meter, operating at 25 °C. The measurements were taken in triplicate.³⁶
2.3.5.8. Determination of total phenolic content. 80.0 mg of the film samples were added into tubes containing 5.0 mL of distilled water. Then the tubes were shaken in a shaker at room temperature (RT) for a day at 125 rpm. After a day, the supernatant of the film extract solution was separated.

The total phenolic content of the edible film samples was identified using the Folin–Ciocalteu method. 400.0 μL of the solution obtained from film extract was taken, followed by the addition of 2.5 mL of 1 : 10 diluted Folin–Ciocalteu reagent. Next, 7.5% sodium carbonate was added into the mixture. Then the whole mixture was allowed to sit for a 30-minute incubation period. After incubation, the absorbance value of the samples was measured using a UV-Vis spectrophotometer at a wavelength of 760 nm. The total phenolic content of the film samples was presented as mg GAE per g of the dried film.³⁷

2.3.6. Characterization of cassava peel starch, APE microcapsules, and edible film.
2.3.6.1. Fourier transform infrared (FTIR) analysis. The chemical structures of the obtained cassava peel starch, APE microcapsules, and the synthesized edible films were analysed using an FTIR spectrophotometer. The film samples were pelleted with KBr. The analysis was carried out in the wavenumber range from 500 cm⁻¹ to 4000 cm⁻¹ with a resolution of 4 cm⁻¹ per sample and 64 scans of summation.³⁸
2.3.6.2. Scanning electron microscopy (SEM) analysis. The surface morphology and size of the film were observed using SEM with an accelerating voltage of 15.0 kV at different magnifications. Small pieces of control and sample films were laid on the stub with two-sided carbon tape. Next, they were sputter-coated with gold particles in a thin layer before the observation.³⁹
2.3.6.3. Thermogravimetric analysis (TGA) of film samples. Thermal decomposition and stability were analysed using a thermogravimetric analyser, where approximately 5.0 mg of the sample was analysed in the temperature range of 30–600 °C at a 10 °C min⁻¹ heating rate under a 20 mL min⁻¹ nitrogen atmosphere.⁴⁰

2.3.7. Application of the film on PFL.
2.3.7.1. Preparation of PFL. Fresh palmyrah ripened fruits with no beetle attacks were collected. The fruits were rinsed with running tap water and then washed with hot water. After that, the peels of the palmyrah fruits were removed. The pulp was prepared using 100.0 mL of distilled water per seed and filtered with a cotton cloth. Next, the pulp was poured onto a coconut oil-treated tray and allowed to sun-dry, enclosed with a net. The new pulp was poured as a thin layer on the tray. It was performed five times. Finally, the fruit leather was peeled off from the tray.
2.3.7.2. Selection of the most suitable film by sensory evaluation. The selected four films, with the control film, were subjected to sensory evaluation using ten trained panellists and ten untrained panellists. The response error of panellists was reduced by presenting the samples with a 3-digit code and changing the arrangement of the samples per person. For the scaling, a 5-point hedonic scale was used (5 = like very much; 4 = like slightly; 3 = neither like nor dislike; 2 = dislike slightly; 1 = dislike very much) for the attributes of appearance, aroma, color, taste, texture, and overall acceptance. The evaluation was carried out at RT, while cream crackers and water were provided to the panellists to avoid the same mouth feel and to have oral rinsing, respectively.⁴¹

The best film formulation was selected as the one that received the highest overall acceptance score, followed by further analysis.

All the panellists were selected as healthy adults who volunteered to participate after being informed about the study objectives. As the sensory evaluation did not involve the use of vulnerable populations or invasive testing, and included safe, food-grade samples, ethical clearance was not required. Furthermore, informed consent was obtained from all the participants before the sensory evaluation.

The following analysis was performed at 10-day intervals for the PFL samples packed with the selected film (F8) and control film (F5).

2.3.7.3. Analysis of PFL.
2.3.7.3.1. Weight loss. The percentage weight loss of the PFL samples was determined as the ratio of weight change relative to the initial weight of the sample. The same procedure was performed separately for PFL samples packed with control and test films. The percentage weight loss of PFL samples was calculated using eqn (5):⁴²where M₀ and M₁ denote the weight on the initial day and the weight on every sampling day, respectively.
2.3.7.3.2. Moisture content. 2.0 g of the PFL sample was taken in a previously dried, weighed, and labelled Petri dish. Next, the Petri dish was kept in a hot air oven at 105 ± 1 °C until a constant weight was obtained. Then the Petri dish was taken out and placed in a desiccator to cool down. Finally, the total dry weight of the Petri dish with the PFL sample was recorded. The moisture content for PFL samples packed with both test and control films was calculated using eqn (3).
2.3.7.3.3. pH determination. 5.0 g of the fruit leather sample was homogenized with the addition of 45.0 mL of distilled water for 1 minute. Then, the homogeneous mixture was allowed to stand for about 30 minutes. Finally, the pH of the PFL samples was measured using a calibrated pH meter.³¹
2.3.7.3.4. Water activity. The water activity of PFL samples was measured using the method reported by Fransiska et al.⁴³ To measure the water activity, a benchtop water activity meter was utilized.
2.3.7.3.5. Microbial analysis. Total plate count (SLS 516 part 1:1991): 5.0 g of the PFL sample was homogenized in 45.0 mL of peptone water using a stomacher (dilution factor = 10⁻¹), from which an aliquot of 1.0 mL was transferred into 9.0 mL of peptone water and homogenized properly with a vortex mixer (dilution factor = 10⁻²). Then, 1.0 mL of an aliquot with a 10⁻² dilution factor was transferred to 9.0 mL of distilled water and homogenized with a vortex mixer (dilution factor = 10⁻³). 1.0 mL from each dilution was transferred into the appropriate Petri dish labelled with the date, type of test, sample, and dilution factor. Then, about 20.0 mL of plate count agar medium was poured on the Petri dish and shaken well in clockwise and anticlockwise directions. After that, the plates were allowed to solidify. Then, the plates were incubated at 37 °C for 24 hours. Finally, after 24 hours, the grown colonies were counted using a colony counter. The total plate count in the PFL sample was determined using eqn (6):where , v, and d denote the sum of the counted colonies from two successive dilutions, the volume of the inoculum transferred into a Petri dish, and the dilution at which the 1st count was initiated.

Total yeast and mold count (SLS 516 part 2:1991): 5.0 g of the PFL sample was homogenized in 45.0 mL of peptone water using a stomacher (dilution factor = 10⁻¹), from which an aliquot of 1.0 mL was transferred into a 9.0 mL peptone water-containing test tube and homogenized using a vortex mixer (dilution factor = 10⁻²). Then, a 1.0 mL aliquot from a 10⁻² dilution was transferred into a 9.0 mL peptone water-containing test tube (dilution factor = 10⁻³). About 20.0 mL of yeast and mold medium was poured into a previously labelled Petri dish. After that, the medium was allowed to settle. Then, 0.1 mL of an aliquot of the PFL sample was transferred into a Petri dish containing the medium. The sample was spread with a sterilized spreader. Then, the Petri plates were incubated at 25 °C for 3–5 days. After the incubation, the grown colonies were counted using a colony counter.

The total yeast and mold count was calculated using eqn (7):

where is the total number of colonies from two successive dilutions, n₁ is the number of Petri plates counted in the 1st dilution, n₂ is the number of Petri plates counted in the 2nd dilution, and d is the dilution factor at which the 1st count was initiated.

2.3.7.3.6. Colour index measurement. Colorimetric parameters of the edible film samples were analysed using a handheld colorimeter (NR20XE) with a D65 standard and 10° observer angle. The results were expressed in terms of CIE L (lightness, 0 = dark, 100 = bright), a* (green–red), and b* (blue–yellow) coordinates. The colorimeter was previously calibrated with a white reflector plate. All measurements were conducted in triplicate.⁴⁴
2.3.7.3.7. Texture analysis. A 1.5 cm × 1.5 cm × 1.0 cm (length × width × height) PFL sample was sliced and placed in the middle of the platform of the instrument. The analyser compressed the sample twice with an interval of 20 s, at a 1 mm min⁻¹ compression rate, with 5 mm s⁻¹ pre- and post-speed compression. The analysis was conducted at RT.⁴⁵

2.3.8. Statistical analysis. SPSS version 20 was utilized for the statistical analysis to determine significant differences among treatments at a 5% significance level (p < 0.05) using ANOVA followed by Tukey's multiple range test. Each treatment formulation was prepared in triplicate as three independent film batches, and all measurements were conducted in triplicate. The values are represented as the mean values of triplicate with standard deviation.

3. Results and discussion

3.1. Thickness, mechanical, and barrier properties of the films

The thickness, mechanical, and barrier properties of the prepared edible films are summarized in Table 1.

Table 1 The resulting thickness, mechanical and barrier properties of the developed films

Sample	Thickness (mm)	Mechanical properties		Barrier properties
		Tensile strength (N mm^−2)	Elongation at break (mm)	Water vapor transmission rate (g m^−2 d^−1)	Light transmission rate
Dissimilar lowercase letters within a similar column denote significant differences (p < 0.05) among them. All values are presented in the form of mean ± SD (n = 3).
F1	0.111 ± 0.002^a	0.24 ± 0.02^ab	1.90 ± 0.56^a	3.58 ± 0.16^abc	14.97 ± 0.45^c
F2	0.116 ± 0.003^ab	0.19 ± 0.05^ab	4.83 ± 4.69^ab	4.22 ± 0.15^ef	13.98 ± 0.85^abc
F3	0.121 ± 0.003^abc	0.18 ± 0.01^a	1.73 ± 0.51^a	3.87 ± 0.12^bcde	13.05 ± 0.83^abc
F4	0.126 ± 0.002^abcd	0.15 ± 0.05^a	3.77 ± 0.82^ab	3.96 ± 0.12^bcde	12.04 ± 0.37^a
F5	0.133 ± 0.004^bcd	0.31 ± 0.09^ab	2.52 ± 0.61^a	3.67 ± 0.06^abcd	14.56 ± 0.47^c
F6	0.136 ± 0.007^cd	0.26 ± 0.06^ab	2.30 ± 0.49^a	4.80 ± 0.37^g	14.26 ± 0.75^ab
F7	0.138 ± 0.001^cd	0.21 ± 0.15^ab	4.95 ± 0.91^ab	4.50 ± 0.07^fg	13.78 ± 0.18^abc
F8	0.142 ± 0.006^d	0.19 ± 0.04^ab	6.72 ± 0.51^b	3.67 ± 0.08^bcde	12.34 ± 0.66^ab
F9	0.142 ± 0.173^d	0.38 ± 0.01^b	2.36 ± 0.63^a	3.78 ± 0.13^abcde	13.52 ± 1.68^abc
F10	0.128 ± 0.003^abcd	0.30 ± 0.03^ab	4.38 ± 0.36^ab	3.31 ± 0.16^a	14.97 ± 0.48^c
F11	0.133 ± 0.003^bcd	0.27 ± 0.05^ab	3.70 ± 0.54^ab	3.62 ± 0.12^abc	13.40 ± 0.57^abc
F12	0.134 ± 0.004^cd	0.18 ± 0.02^ab	1.52 ± 0.29^a	3.51 ± 0.07^ab	13.80 ± 0.85^abc
F13	0.136 ± 0.008^cd	0.20 ± 0.08^ab	1.68 ± 0.45^a	4.09 ± 0.20^def	13.86 ± 0.89^abc
F14	0.138 ± 0.001^cd	0.27 ± 0.05^ab	2.65 ± 0.13^a	4.12 ± 0.08^def	13.64 ± 0.22^abc
F15	0.139 ± 0.001^cd	0.23 ± 0.04^ab	2.75 ± 0.53^a	3.98 ± 0.17^cde	13.26 ± 0.12^abc
F16	0.144 ± 0.004^d	0.17 ± 0.12^a	1.75 ± 0.60^a	3.78 ± 0.10^bcde	12.36 ± 0.40^ab

---

The different percentages of starch and microcapsules in various formulations significantly impact the thickness (p < 0.05), ranging from 0.111 to 0.144 mm. The increased addition of 5% to 20% of the concentration of the microcapsules into the starch matrix resulted in a profound increase in the thickness compared to control films (made with 0% microcapsules), except the films incorporated with 80% cassava peel starch (Table S1 and Table 1). The increasing thickness is the reason for the increase in the solid content since the volume of the film-forming solution used was equal within the same percentages of starch treatments with different percentages of microcapsules.⁴⁶ In the case of films prepared with 80% percentage starch, the nonuniform spreading during the casting procedure might have led to the results presented in Table 1.

With respect to tensile strength, there was a statistically significant difference between the film samples with different percentages of starch molecules and different percentages of microcapsules (p < 0.05). Herein, the tensile strength of the films ranged from 0.15 to 0.38 N mm⁻². Increasing the addition of starch concentration up to 80% increased the tensile strength of the control films (0% microcapsules (F1, F5, and F9)). This increasing trend of tensile strength might have resulted from the formation of a stronger matrix with the increasing incorporation of starch, which will be important to support the load.

Resistance to breaking due to stress and stretching indicates that the edible film possesses higher tensile strength.⁵ However, the tensile strength was reduced with the further addition of starch up to 90%. Continuous addition might have resulted in phase separation due to particle agglomeration, resulting in reduced tensile strength.⁴⁷ A similar decreasing trend was also observed in a previous study, in which chicken skin gelation films were formulated with tapioca starch.⁴⁸

Furthermore, the addition of microcapsules has also led to the reduction of the tensile strength of the films (Table S1 and Table 1). This type of observation is likely to take place due to the development of a heterogeneous film structure arising from the addition of amla pomace microcapsules, which will also lead to the weakening of the starch–glycerol interaction.³⁷

There was a significant impact on the elongation at break with varying percentages of starch and microcapsules (p < 0.05). The elongation at break of the synthesized film samples ranged from 1.52 to 6.72 mm. As shown in Table 1, in control films (0% microcapsules), there was an increasing trend observed in elongation at break with the increase of the starch concentration up to 70% (F5), but further addition decreased the elongation at break in F9 and F13 films. Cassava peel starch is a relatively long-chain polysaccharide, and when cross-linked with gelatin, it might result in the relaxation of macromolecules, leading to an increase in elongation at break.⁴⁸ Yet, further addition resulting in a decrease in elongation value might be due to the higher amylose content, leading to increased viscosity in addition to higher total dissolved solids. Therefore, this will make the matrix bond stronger, resulting in a lower elongation at break value.⁵ Nevertheless, there were some fluctuations observed with the addition of different percentages of microcapsules.

Water vapor permeability and the light transmission rate are two main barrier properties that need to be evaluated during the development of packaging materials. Water plays a vital role in the spoilage of food, making the water vapor transmission rate of food packages a considerable characteristic to prevent the transition of moisture between the food matrix and the medium. The water vapor transmission rate illustrates the measurement of the moisture passing through a unit space of material per unit of time.³⁸ It is a favourable characteristic that the water transmission rate should be as low as possible in the case of packaging materials.

The different percentages of starch and microcapsules in different formulations caused a significant impact on the barrier properties (p < 0.05) mentioned above. The water vapor transmission rate of the synthesized films ranged from 3.31 to 8.54 g m⁻² d⁻¹. Table 1 shows that the water vapor transmission rate tends to increase with the addition of starch percentage from 60% to 90% in the control films. Since starch is a hydrophilic biopolymer, consisting of polar groups, this could interact with water molecules that permeate. The water sorption during the permeation process enhances the free volume of the polymer, enabling the increasing mobility of the segments in the polymer chain, resulting from swelling. This increase of mobility would further lead to a higher water vapor transmission rate.³⁶ Based on the results in Table 1, the water vapor transmission rate decreases with increasing microcapsule percentage. This could be due to the almost even distribution of the APE microcapsules, creating a barrier for the passage of water molecules through the package.⁴⁰

As shown in Table 1, the light transmission rate of the film sample varied from 12.04 to 14.97 in samples F1–F16. The control films (F1, F5, F9, and F13) showed the highest light transmission rate, whereas the addition of amla pomace microcapsules led to a significant reduction in the light transmission of the other samples. This scenario was observed in 60, 70, and 90% starch-incorporated films loaded with 5, 10, and 20% of the microcapsules. Due to the thickness variation in the 80% starch incorporated samples resulting from nonuniform spreading, the light transmission rate was varied in the films incorporated with 5, 10, and 20% of microcapsules. The addition of microcapsules could create a barrier for the UV light passing through the film, which leads to a reduction in the light transmission rate. This observation is in accordance with the previous study conducted with a functional starch-based film incorporated with free and microencapsulated spent black tea extract.³⁴ Furthermore, it can also be noticed that increasing the starch percentage decreased the light transmission rate as compared to the film synthesized from 60% starch. Wahidin et al. observed the same trend in the light transmittance with increasing jack and durian seed starch.⁴⁹ This might be due to the high condensation or density of the starch granules within the same volume of the film-forming solution, resulting in the blockage of the light passing through it.⁴⁹

Based on the mechanical and barrier properties, four films were selected out of 16 films for further analysis. They were F8, F11, F12, and F16, which were subjected to various characterization techniques, and further analysis is described below.

3.2. Physicochemical properties

3.2.1. Moisture content. In the case of moisture content, there was a significant increase with the increasing content of starch from 70% (F8) to 90% (F16) and with the increasing percentage of microcapsules from 10% to 20% (F11 to F12), which is shown in Fig. S1. The moisture content of the selected film samples was found to range from 19.09 to 25.61%. Lower moisture content is favourable to minimize microbial growth. A similar increase in the moisture content with the addition of cassava starch films⁵⁰ and cassava-protein⁵¹ films has been observed in previous studies. The higher moisture content might have resulted from the absorbance and holding of water within the interpolymeric space of the matrix by cassava starch and the wall material of amla pomace microcapsules (maltodextrin).⁵²

Maltodextrin is a hygroscopic polymer that can further contribute to the increase in moisture absorption.⁵³ Moreover, the presence of polyphenolic compounds in amla pomace extract may also strengthen the interactions through hydrogen bonding and trigger moisture absorbance, as has been observed for plant extract incorporated films.⁵⁴ These findings suggest that although moisture content increased with starch and microcapsule concentration, the values obtained in this study remain within the range reported for edible starch-based films, confirming its application as an edible food packaging material.

3.2.2. Water solubility. The solubility of films in water is an important property of starch-based films. In some applications, there is a requirement for low-soluble packages to strengthen and be flexible to withstand the shock created during transportation, storage, and handling.⁵⁵ However, some applications require the packaging to be able to release the encapsulated materials to perform a specific activity. There was a significant difference (p-value < 0.05) between the samples with varying formulations of starch and APE microcapsules. The water solubility of the films ranged from 26.33 to 34.57%. Fig. S2 shows that the water solubility increased significantly while increasing the starch content from 70% (F8) to 90% (F16). Furthermore, with the same concentration of starch (80%), adding microcapsules with varying amounts has led to an increase in water solubility (F11 and F12). This might be due to the hydrophilic nature of starch and the wall material, triggering the ease of solubilization in water. As described in Section 3.2.1 results, the presence of polyphenols in the APE must have also improved the water absorption via hydrogen bonding. Similar effects have been observed in previous studies conducted using microencapsulated tea polyphenols in the starch matrix.^34,56

3.2.3. Water activity. The water activity of films varied from 0.393 to 0.414. If the value is below 0.60, then it can be speculated that the film is stable against microbial proliferation at room temperature. A similar range of values from 0.39 to 0.40 was recorded while synthesizing the film with microencapsulated blackberry pulp and arrowroot starch in a previous study.⁴⁶ However, there was a decreasing trend observed while increasing the starch percentage from 70% to 90%. On the other hand, increasing the microcapsule percentage enhanced the water activity in the films within the same starch concentration, as shown in Fig. S3.

3.2.4. Total phenolic content. The variation of total phenolic content among the film samples was mainly due to the addition of different percentages of microcapsules. Among the film samples, the total phenolic content ranged from 8.25 to 15.18 mg GAE per g for the dried films. This variation was mainly due to the addition of different percentages of microcapsules. Since F11 was synthesized with 10%, while others were with 20% microcapsules of APE, the increasing microcapsule addition improved the phenolic content in those films, almost doubling the amount of total phenolic content with double the amount of added microcapsule content, which is presented in Fig. S4. Piñeros-Hernandez et al. observed a similar increase in the total phenolic content of edible films with the increasing addition of rosemary extract into the film sample.³⁷

3.3. FT-IR and TGA

The selected films (F8, F11, F12, and F16) were subjected to chemical structure identification using FT-IR spectroscopy. The spectrum of the four films is shown in Fig. 1(a). The variations in the spectra of different films are mainly due to orientation and conformational changes from the different percentages of starch and microcapsules. Almost the same trend of vibrational pattern was observed for F11 and F12, which could be due to the addition of the same concentration of starch during the synthesis. In the spectra of the four films, the wide band appearing at 3000–3600 cm⁻¹ indicates the presence of O–H stretching from the intermolecular interaction of –OH groups.⁵⁷ The peak at 2929 cm⁻¹ is associated with the C–H bond from alkyl groups in the polymer chain.⁵⁸ The band designated at 1636 cm⁻¹ could have resulted from the O–H bending of the absorbed water present in the cassava peel starch's amorphous regions.^59,60 The peaks from all the films around 1543 cm⁻¹ denote the N–H stretching of amide II.³⁴ The peak at 1017 cm⁻¹ is associated with C–H deformation of the aromatic ring. Also, the peaks were located in the range of 920–758 cm⁻¹, which might have resulted from the C–O–C ring vibration of carbohydrates.⁶¹

Fig. 1 FT-IR spectra and TGA profiles of the selected films F8, F11, F12, and F16: (a) FT-IR spectra and (b) TGA profiles.

Thermogravimetric analysis (TGA) is an important technique to evaluate thermal decomposition, thermal tolerance, and mass loss concerning the temperature change.⁶² Based on the results shown in Fig. 1(b), with the increase in heating temperature, the variations in the thermal decomposition of all the film samples are clearly visible. The weight loss of the film samples were in the range of 14.28–80.49%, 9.48–80.43%, 10.75–82.50%, and 9.04–80.40% for F8, F11, F12, and F16 films, respectively, at different phases. The first stage of decomposition at 30–135 °C in all films is attributed to the loss of water molecules and glycerol present,⁵² which ranged from 9.04 to 14.28%. Next, the larger phase of decomposition at 190–400 °C with more than 70% of its initial weight is associated with the decomposition of starch, followed by the carbon ring's decarboxylation and generation of many gaseous products leading to solid char formation.⁶² The most apparent weight loss was observed in the temperature range of 200–320 °C, which could be due to the decomposition of the polymer backbone.

The weight loss beyond 400 °C could be ascribed to the generation of demolished solid char stacks with polyaromatic structures.⁶² Among the four films, F16 showed the least weight loss at high temperatures, confirming its high stability, which might be associated with the synergistic effect of the high content of APE microcapsules and starch percentage.⁵² One of the previous studies also reported increasing thermal stability in a clove essential oil-loaded pullulan-gelatin based edible film,⁴⁰ where it has also been reported that the slow release of bioactive compounds from microcapsules had a positive impact on the mass retention rate of the film samples.

3.4. Optimization of the films by sensory evaluation

Ten trained and ten untrained panellists, including 85% women and 15% men, were employed for the evaluation of the sensory properties of the PFL sample packed with F8, F9, F11, and F12 films and the control.

A statistically significant difference was observed in all the sensory properties, including appearance, aroma, color, texture, taste, and overall acceptance with the different film samples and control (p < 0.05). Among these sensory attributes, texture got the lowest score as compared to others, whereas aroma got the highest score (Fig. 2). Based on the results from different sensory properties, film F8 turned out to be the best among others in terms of aroma. This might be due to the provision of proper masking of unfavourable aroma from PFL. Therefore, based on the sensory evaluation results, the F8 film was selected as the most suitable film for PFL wrapping. It was then considered as the test film, and the film without the microcapsules having the same concentration of starch (F5) was considered as the control film, and these two films were subjected to further analyses.

Fig. 2 Radar chart of sensory attributes.

3.5. SEM analysis

The 3D appearance of the edible film after the incorporation of encapsulated APE, the surface morphology, and the size of the microcapsules were observed under SEM, as depicted in Fig. 3(b). The control film (Fig. 3(a)) exhibited a smooth and flat surface, whereas the test film (Fig. 3(b)) exhibited a rough surface with protuberances, with the dispersion of tiny globular-shaped microparticles. Fig. 3(b) clearly shows the wide dispersion of microcapsules on the surface of the film, resulting from their entrapment in the gelatinized starch suspension. The partial dissolution of these microcapsules is mainly due to the presence of the APE inside the wall material, which prevents homogenous dissolution, leading to the formation of a rough surface. The same phenomenon has been observed in a previous study.³⁴

Fig. 3 Scanning electron microscopy images of control (F5) and test films (F8): (a) control film and (b) test film.

The size of the spray-dried microcapsules infused inside the films was found to be in the size range of 3.5–6.5 μm. The microcapsules obtained from spray drying showed spherical-shaped particles, a wall with fewer cracks or fissures and a rough surface. During spray drying, the high inlet temperature must have caused faster solidification of the wall, which could create a strong layer on the surface retarding complete bubble inflation, thus resulting in a non-homogeneous spherical surface.⁶³ The same results were reported by other authors while using maltodextrin as a wall material to obtain spray-dried microcapsules.²⁷

3.6. Application of the films

PFL samples were packed with F5 (control film) and F8 (test film) for 50 days of storage at 25 °C to check the weight loss, moisture content, water activity, color index, microbial count, and texture. The visual image of the PFL samples packed with both films is shown in Table S2.

In the case of weight loss, there was a statistically significant difference in the percentage of weight loss in the PFL samples packed with different packaging materials at different storage periods (p < 0.05). The percentage of weight loss ranged from 0.00 to 6.12% in the PFL samples packed with film F5 and 0.00–3.97% in the PFL samples packed with film F8 throughout the storage period. Table 2 shows the increasing trend of the weight loss percentage in PFL samples packed with films F5 and F8. After 50 days of storage, the PFL sample packed with F5 and F8 showed a weight loss of 6.12% and 3.97%, respectively. From this, it can be observed that the PFL sample packed with F8 (microcapsule loaded film) had a lower weight loss than that packed with F5 (without microcapsule loaded film). This might be due to the lower migration of existing water from the PFL³⁶ and also due to the stability of other compounds present in the PFL sample. Based on the values given in Table 1, F8 has a lower water vapor transmission rate than F5, but it has a higher thickness and a lower light transmission rate than F5. So, as a combination of these properties, F8 could act as a more stable film on the fruit leather sample than F5 to facilitate higher protection against weight loss.

Table 2 The physicochemical, microbial and textural properties of the PFL samples packed with F5 and F8 during the 50-day storage period at 25 °C

Sample	Storage time (days)
	0	10	20	30	40	50
Dissimilar lowercase letters within the similar column denote significant variations (p < 0.05) among the mean values. All values are presented in the form of mean ± SD (n = 3).
Weight loss (%)
F5	0.00 ± 0.00^a	0.58 ± 0.16^ab	1.60 ± 0.23^c	2.86 ± 0.33^d	4.41 ± 0.46^e	6.12 ± 0.58^f
F8	0.00 ± 0.00^a	0.46 ± 0.08^ab	1.16 ± 0.11^bc	1.88 ± 0.08^c	2.85 ± 0.15^d	3.97 ± 0.27^e
pH
F5	4.18 ± 0.02^ab	4.20 ± 0.05^ab	4.20 ± 0.03^ab	4.20 ± 0.04^ab	4.20 ± 0.03^ab	4.22 ± 0.02^b
F8	4.17 ± 0.03^ab	4.14 ± 0.04^ab	4.14 ± 0.03^ab	4.12 ± 0.02^a	4.15 ± 0.02^ab	4.14 ± 0.03^ab
Moisture (%)
F5	13.96 ± 1.16^b	10.93 ± 1.58^ab	10.49 ± 1.74^ab	9.48 ± 2.04^ab	8.64 ± 1.60^ab	8.13 ± 1.53^a
F8	11.86 ± 2.44^b	11.17 ± 1.99^ab	10.79 ± 2.03^ab	10.45 ± 1.97^ab	10.19 ± 1.98^ab	8.64 ± 1.60^ab
Water activity
F5	0.423 ± 0.01^b	0.408 ± 0.010^ab	0.400 ± 0.02^ab	0.398 ± 0.02^ab	0.394 ± 0.02^ab	0.390 ± 0.02^a
F8	0.423 ± 0.02^b	0.419 ± 0.003^ab	0.415 ± 0.00^ab	0.412 ± 0.00^ab	0.407 ± 0.01^ab	0.405 ± 0.01^ab
Color index
L*
F5	22.53 ± 1.70^a	22.51 ± 1.68^a	22.48 ± 1.67^a	22.47 ± 1.65^a	22.50A ± 1.67^a	21.50 ± 0.32^a
F8	23.79 ± 3.46^a	23.79 ± 3.45^a	23.78 ± 3.46^a	23.77 ± 3.43^a	23.75 ± 3.47^a	23.72 ± 3.47^a
a*
F5	10.36 ± 0.37^a	15.73 ± 1.99^bc	18.76 ± 0.71^cd	22.02 ± 3.28^d	15.94 ± 1.05^bc	12.77 ± 2.08^ab
F8	11.70 ± 1.66^ab	12.10 ± 0.98^ab	15.34 ± 1.05^bc	11.71 ± 1.52^ab	12.24 ± 0.60^ab	11.87 ± 0.14^ab
b*
F5	6.33 ± 4.71^a	6.30 ± 4.71^a	6.26 ± 4.68^a	6.20 ± 4.65^a	6.15 ± 4.60^a	6.11 ± 4.58^a
F8	10.32 ± 1.32^a	10.36 ± 1.29^a	10.53 ± 1.35^a	10.60 ± 1.40^a	10.67 ± 1.36^a	10.69 ± 1.27^a
Total plate count (log CFU g^−1)
F5	0.85 ± 0.03^abc	0.98 ± 0.03^def	1.07 ± 0.02^fgh	1.17 ± 0.03^h	1.14 ± 0.03^gh	1.04 ± 0.05^efg
F8	0.83 ± 0.06^ab	0.91 ± 0.04^bcd	0.93 ± 0.06^bcde	1.02 ± 0.03^defg	0.96 ± 0.04^cde	0.75 ± 0.02^a
Total yeast and mold count (log CFU g^−1)
F5	0.40 ± 0.05^cd	0.42 ± 0.14^cd	0.13 ± 0.09^ab	NF	NF	NF
F8	0.46 ± 0.05^d	0.28 ± 0.06^bc	NF	NF	NF	NF
Hardness (×10^4) N m^−2
F5	1.32 ± 0.16^a	1.76 ± 0.23^abcd	2.10 ± 0.21^cde	2.32 ± 0.17^de	2.41 ± 0.32^ef	2.97 ± 0.20^f
F8	1.44 ± 0.18^ab	1.61 ± 0.14^abc	1.77 ± 0.15^abcd	1.97 ± 0.18^bcde	2.22 ± 1.76^de	2.31 ± 0.14^de

---

The pH value of the PFL sample packed with film F5 ranged from 4.18 to 4.22, whereas in the PFL samples packed with F8 ranged from 4.14 to 4.17. Safaei, Sadeghi and Jahed Khaniki reported that the pH of the fruit leather sample must range from 2.5 to 4.5, and the pH values obtained from the present study are in accordance with the reported values.⁶⁴ From Table 2, there is a slightly increasing trend observed in the PFL sample packed with F5, whereas a decreasing trend is observed in the PFL sample packed with F8, which might have resulted from the migration of the bioactive phytochemicals, especially phenolic compounds, present in the microcapsules to the fruit leather sample. Ellagic acid, chebulinic acid, gallic acid, chebulagic acid, etc. are some of the major polyphenols present in amla,⁶⁵ which are acidic in nature, resulting in the reduction of pH during their migration to the sample. The increasing pH content over the storage time might be due to the liberation of protein metabolites (amines), during microorganisms' breakdown of protein in the fruit leather sample.⁴¹ Since the F5 film lacks amla pomace microcapsules, the microbial activity might promote a slightly high pH value for the PFL sample packed with film F5.

The ANOVA test was carried out to determine the differences in the moisture content of PFL samples packed with films F5 and F8 at different storage times. It revealed that there were statistically significant differences between the moisture content among the samples packed with films F5 and F8 at different storage times. The PFL samples packed with films F5 and F8 showed different moisture content values of 8.14–13.96% and 8.64–11.86%, respectively. They are in the acceptable standard range of moisture content for fruit leather, which must be below 15%.⁶⁴ From Table 2, a decreasing trend of the moisture content was observed in the PFL sample packed with both films with storage time, but the sample packed with F5 showed a stronger decline than the sample packed with F8. The decline in both samples might have been caused by the moisture evaporation into the headspace or moisture condensing on the packaging itself. Furthermore, film F8 had a high thickness and better barrier properties as compared to film F5, which might have led to lower moisture content lost from the PFL sample packed with film F8 compared to the sample packed with F5 within the storage period being considered. Sharma et al. have reported a similar decreasing trend in the moisture content over the storage of meat nuggets packed with Commiphora wightii based bioactive edible films.⁴¹

From Table 2, the PFL samples packed with films F5 and F8 exhibited water activity values in the range of 0.390–0.423 and 0.405–0.423, respectively. There was a statistically significant difference (p < 0.05) in the water activity of the PFL samples packed with both films throughout the storage period. The PFL samples packed with both films showed a decrease in water activity with the increase of the storage time, but there was a higher water activity reduction in the sample packed with film F5 than the other samples. The decreasing water activity values might be attributed to the decreasing moisture content in both samples with the increasing storage time. The same decreasing trend of water activity was observed in bread crumbs packed with an edible film made with vegetable oil and egg protein.

The color indices of the edible film may influence the appearance of the food that it is used to wrap. Since the color of the edible film can change with the materials used to produce it, the color of the food item is one of the crucial properties that severely influence the customer's purchasing decision. The color and the overall appearance of the PFL sample packed with F5 and F8 films seemed to be similar in the visual observation shown in Table S2. There was no significant difference observed in both L* and b* (p > 0.05), but there was a statistically significant difference (p < 0.05) in a* among the PFL samples packed with F5 and F8 over 50 days of storage. Table 2 shows the variation of the L* index, where the sample packed with F8 remains almost stable, while the sample packed with F5 shows a slight decline. This means that PFL wrapped with F5 could become darker over time. The L* values for the PFL samples packed with films F5 and F8 were 21.50–22.53 and 23.72–23.79, respectively. Next, there was an increasing trend observed in a* in the PFL sample packed with F8 up to day 30, after which the values were reduced, whereas the PFL packed with film F5 showed a fluctuating trend. The a* values were in the range of 10.36–22.02 and 11.70–15.34 in PFL samples packed with films F5 and F8, respectively.

In the case of b*, there were no considerable changes in the PFL samples packed with both films (Table 2). The b* index in the PFL sample packed with F5 and F8 varied in the range of 6.11–6.33 and 10.32–10.69, respectively. Based on the results, the b* value in the PFL sample packed with F8 was higher than that in the sample packed with F5. The higher the b* value, the more intense the yellow color. Thus, the results might be attributed to the color of the film, which was slightly yellow (F8), caused by the migration of the microcapsules from the film.

The microbial count in the PFL sample is relatively low compared to other food items such as meat and milk products (Table 2). This is mainly due to the relatively low pH, water activity, and moisture content in the fruit leather sample.⁶⁴ It has been reported that the number of aerobic bacteria that contribute to the total plate count and the total yeast and mold count must be less than 200 colonies per g of the sample.⁶⁴ The results of the present study comply with this standard value, indicating that fruit leather samples packed with both F5 and F8 maintained good quality as edible products even at the end of the 50th day.

The total plate counts in the PFL samples packed with both F5 and F8 throughout the 50 days of storage were significantly different (p < 0.05) from each other. These values were in the range of 0.85–1.17 log CFU g⁻¹ and 0.75–1.02 log CFU g⁻¹, respectively. As shown in Fig. 4(a), palmyrah samples packed with both F5 and F8 show an increasing trend of the total plate count up to 30 days of storage. Similar observations have been reported by Sharma et al. in the case of meat nuggets packed with Commiphora wightii based active films.⁴¹ However, after the 30th day of storage, there was a decline in the total plate count in both samples. This decline might have resulted from the low moisture content and water activity in both samples, which are usually favourable conditions to retard the growth and proliferation of microorganisms. It can be observed from Table 2 that both samples showed decreasing moisture and water activity after the 30th day than during the initial days of the storage. The PFL sample packed with F5 showed a higher microbial count than the sample packed with F8 throughout the storage period. This might be attributed to the release of APE from the microcapsules that are embedded within the F8 film, leading to excellent antimicrobial activity.

Fig. 4 Microbial count variations in PFL samples with the films over the storage time: (a) total plate count variation among the films and (b) total yeast and mold count variation among the films. Different lowercase letters (a–h) above the bars denote statistically significant differences (p < 0.05).

This could be mainly due to the myriad of bioactive compounds with strong anti-microbial properties present in APE. Khurana et al. have reported that amla extract showed a higher inhibition zone against various Gram-positive and Gram-negative bacteria like Escherichia coli, Staphylococcus aureus, Bacillus cereus, Klebsiella pneumoniae, etc. due to the presence of phenols, flavonoids, alkaloids, saponins, and tannin terpenoids.⁶⁵ Therefore, the F8 film-packed PFL sample produced more satisfactory results than those packed with film F5 in countering food-borne bacteria.

There was a significant difference observed in the total yeast and mold count between the PFL samples packed with both films (p < 0.05). The total yeast and mold count was in the range of >0.00–0.40 log CFU g⁻¹ and >0.00–0.46 log CFU g⁻¹ for the F5 film-packed PFL sample and the F8 film-packed PFL sample, respectively. The total yeast and mold count decreased significantly from 0.40 log CFU g⁻¹ to >0.00 log CFU g⁻¹ from the initial day to the 20^th day of storage, after which there were no fungal colonies observed in the PFL sample packed with film F5, whereas, in the F8 film-packed PFL sample, the total yeast and mold count decreased significantly from 0.46 log CFU g⁻¹ to >0.00 log CFU g⁻¹ from the initial day to the 20^th day of storage, as shown in Fig. 4(b). The excellent antifungal properties of the F8 film-packed PFL sample could be attributed to the APE release from the microcapsules in the F8 film. Amla has been reported to have appreciable antifungal activity against food-borne fungal species communities, especially eight species of Aspergillus, which is one of the most food-degrading fungi, and some others, including Penicillium chrysogenum, Candida albicans, etc.⁶⁵ This aligns with the findings of Jain et al.,¹⁹ who demonstrated that incorporating amla pomace significantly improved the antimicrobial performance of sugarcane bagasse-based packaging materials.

The texture of fruit leather is one of the major sensorial properties and is required to be slightly elastic to ensure pleasant consumption. The hardness of the PFL sample was analyzed as a textural property to check the variations. There was a significant difference in the hardness of the fruit leather samples packed with both F5 and F8 (p < 0.05). The hardness of the PFL sample packed with film F5 ranged from 1.32 × 10⁴ to 2.97 × 10⁴ N m⁻², whereas that of the PFL sample packed with F8 ranged from 1.44 × 10⁴ to 2.31 × 10⁴ N m⁻². The variation of hardness in both samples is presented in Table 2, where the PFL sample packed with F5 shows a much higher hardness value than the sample packed with F8. Loss of moisture leads to increased rigidity of the food, whereas absorption of moisture results in a soggy-like texture.⁶⁶ In the present study, the highest hardness of the PFL sample packed with F5 might be attributed to the lowest moisture content. Also, it has been reported that water activity also influences the textural properties of food items, where decreasing water activity could lead to unfavorable attributes like hardness and staleness.⁶⁷ The findings of the present study align with this report, as the hardness of the PFL samples packed with both films increased with the reduction in water activity (Table 2).

4. Conclusions

This study demonstrated the successful synthesis of edible films from cassava peel starch incorporated with amla pomace extract (APE) microcapsules. The microcapsules were spherical in shape with a size of 3.5–6.5 μm. Their incorporation, along with varying cassava peel starch percentages, significantly influenced the films’ thickness, tensile strength, elongation, and barrier properties (p < 0.05). More specifically, the increased addition of cassava peel starch (from 60% to 90%) has led to the improvement of film thickness, water vapor transmission, and mechanical properties while decreasing the light transmission. On the other hand, the increased addition of APE microcapsules has increased the film thickness but decreased the tensile strength and barrier properties. Moreover, it has also led to irregularities in elongation at break. Among 16 formulations, film F8 was identified as optimal due to its balanced mechanical and barrier performance. When applied to palmyrah fruit leather (PFL), F8 effectively reduced weight loss, controlled moisture and water activity, reduced color changes, minimized microbial growth, and maintained desirable texture over 50 days of storage compared with the control film (F5).

This study presents significant outcomes by advancing the development of functional and sustainable bioactive food packaging materials from cassava peel starch and APE microcapsules, which adds value to agricultural waste and provides an alternative to traditional petrochemical-based packaging. These findings support trends in food safety, sustainability, and functional foods and are especially relevant to the development of edible, biodegradable, and bioactive packaging solutions for future food applications. Additionally, this study lays the groundwork for future investigations on the effective and sustained incorporation of bioactive compounds derived from plants into starch-based films for improved food preservation.

Author contributions

Tharmika Sothilingam: investigation, methodology, formal analysis, writing – original draft. Danushika C. Manatunga: supervision, writing – review and editing. Anuluxshy Balasubramanium: supervision, writing – review and editing. Rohan S. Dassanayake: writing – review and editing. S. Srivijeindran: writing – review and editing. The manuscript was read and approved by all authors.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this study are included in the form of text, figures, and tables in this article and in the SI. The SI section contains the information related to variations in the moisture content in different films, variations of the water solubility in different films, variations of water activity among different film samples, variations in the Total Phenolic Content (TPC) among different film samples, the experimental design used for the development of the edible films and the visual images of PFL samples packed with F5 and F8 during the 50 days of storage. See DOI: https://doi.org/10.1039/d5nj02911b.

Acknowledgements

This work was supported by the Palmyrah Research Institute, Kaithady, Jaffna, Sri Lanka. The authors would like to thank Dr Suranga M. Rajapaksha at the Department of Materials and Mechanical Technology, Faculty of Technology, University of Sri Jayewardenepura, for supporting sample analysis using FTIR and the tensile testing machine and for data collection. The authors would also like to express their gratitude to the analytical services division of the University of Moratuwa for providing support in accessing the SEM and obtaining microscopic images.

References

1. S. S. Nair, J. Trafiałek and W. Kolanowski, Appl. Sci., 2023, 13, 8234.
2. A. Trajkovska Petkoska, D. Daniloski, N. M. D’Cunha, N. Naumovski and A. T. Broach, Food Res. Int., 2021, 140, 109981.
3. A. Manzoor, S. Khan, A. H. Dar, V. K. Pandey, R. Shams, S. Ahmad, G. Jeevarathinam, M. Kumar, P. Singh and R. Pandiselvam, J. Food Saf., 2023, 43, e13046.
4. J. Kadzińska, M. Janowicz, S. Kalisz, J. Bryś and A. Lenart, Packag. Technol. Sci., 2019, 32, 483–495.
5. A. W. Wasistha, M. R. S. Dika, A. S. Aulia, N. M. Samudra and D. N. Putri, Agrisaintifika J. Ilmu-Ilmu Pertan., 2021, 5, 149–158.
6. E. G. Fadhallah, A. S. Zuidar, A. H. Dameswary, I. N. Assa’diyah, N. Juwita, S. Tullaila and M. K. K. Yudistiro, Molekul, 2024, 19, 36–45.
7. G. L. Utama, I. Dinika, S. Nurmilah, N. Masruchin, B. Nurhadi and R. L. Balia, Membranes, 2022, 12, 636.
8. D. F. Nury, M. Z. Luthfi and M. P. Ramadhan, BIO Web Conf., 2023, 77, 01005.
9. F. M. Fakhoury, S. Maria Martelli, L. Canhadas Bertan, F. Yamashita, L. H. Innocentini Mei and F. P. Collares Queiroz, LWT, 2012, 49, 149–154.
10. V. Siracusa, P. Rocculi, S. Romani and M. D. Rosa, Trends Food Sci. Technol., 2008, 19, 634–643.
11. M. Tshamisane, J. O. Adeyemi and O. A. Fawole, Mater. Adv., 2025, 6, 1051–1066.
12. S. Siju and K. K. Sabu, Plant Genet. Resour. Charact. Util., 2020, 18, 445–453.
13. P. C. Vengaiah, S. Kaleemullah, M. Madhava, A. Mani and B. Sreekanth, Pharma Innov. J., 2021, 10, 1920–1925.
14. S. Mahilrajan, S. Thuraisingam and J. Prabagar, Food Chem. Adv., 2025, 6, 100880.
15. S. Srivijeindran, S. Mohanadas and C. D. Jayaweera, J. Agric. Sci., 2025, 20, 226–238.
16. A. Yashwanth, R. Huang, M. Iepure, M. Mu, W. Zhou, A. Kunadu, C. Carignan, Y. Yegin, D. Cho, J. K. Oh, M. T. Taylor, M. E. S. Akbulut and Y. Min, Compr. Rev. Food Sci. Food Saf., 2025, 24, 1–40.
17. P. Prasad, Nutraceuticals from Fruit and Vegetable Waste, 2024, pp. 81–96.
18. K. Muzaffar, S. Rafiq, S. A. Sofi, J. A. Rather, F. Allai, H. A. Makroo, D. Majid and B. N. Dar, Int. J. Food Sci. Technol., 2023, 58, 5357–5365.
19. P. Jain, S. Tripathi, R. K. Deshmukh, K. K. Gaikwad and S. Singh, Biomass Convers. Biorefinery, 2024, 14, 20959–20972.
20. G. R. de Carvalho, G. S. Marques, L. M. de Matos Jorge and R. M. M. Jorge, Brazilian J. Chem. Eng., 2021, 38, 341–349.
21. A. Edhirej, S. M. Sapuan, M. Jawaid and N. I. Zahari, Fibers Polym., 2017, 18, 162–171.
22. W. Abotbina, S. M. Sapuan, R. A. Ilyas, M. T. H. Sultan, M. F. M. Alkbir, S. Sulaiman, M. M. Harussani and E. Bayraktar, Materials, 2022, 15, 6992.
23. T. Rahmadi Putri, A. Adhitasari, V. Paramita, M. Endy Yulianto and H. Dwi Ariyanto, Mater. Today Proc., 2023, 87, 192–199.
24. Y. Zhang, J. Xie, W. O. Ellis, J. Li, W. O. Appaw and B. K. Simpson, Ind. Crops Prod., 2024, 213, 118427.
25. S. Mohd Asharuddin, N. Othman, N. S. Mohd Zin, H. A. Tajarudin, M. F. Md Din and V. Kumar, Int. J. Integr. Eng., 2018, 10, 185–192.
26. D. Krishnaiah, A. Bono, R. Sarbatly, R. Nithyanandam and S. M. Anisuzzaman, J. King Saud Univ., Eng. Sci., 2015, 27, 26–36.
27. J. Zhu, X. Li, L. Liu, Y. Li, B. Qi and L. Jiang, LWT, 2022, 154, 112874.
28. W. Tongdeesoontorn, L. J. Mauer, S. Wongruong, P. Sriburi, A. Reungsang and P. Rachtanapun, Polymers, 2021, 13, 1117.
29. J. I. Castro, D. P. Navia-Porras, J. A. A. Cortés, J. H. M. Hernández and C. D. Grande-Tovar, Molecules, 2022, 27, 2264.
30. I. K. Sani, S. P. Geshlaghi, S. Pirsa and A. Asdagh, Food Hydrocolloids, 2021, 117, 106719.
31. R. Zhao, W. Guan, P. Zheng, F. Tian, Z. Zhang, Z. Sun and L. Cai, Food Control, 2022, 132, 108545.
32. E. Karakuş, Z. Ayhan and G. Haskaraca, Food Packag. Shelf Life, 2023, 37, 101092.
33. I. R. A. P. Jati, J. Elaine, E. Setijawaty and A. R. Utomo, BIO Web Conf., 2024, 98, 1–12.
34. S. W. Rajapaksha and N. Shimizu, Molecules, 2021, 26, 3898.
35. L. D. Pérez-Vergara, M. T. Cifuentes, A. P. Franco, C. E. Pérez-Cervera and R. D. Andrade-Pizarro, NFS J., 2020, 21, 39–49.
36. G. F. Nogueira, B. O. Leme, G. R. S. dos Santos, J. V. da Silva, P. B. Nascimento, C. T. Soares, F. M. Fakhouri and R. A. de Oliveira, Polysaccharides, 2021, 2, 373–386.
37. D. Piñeros-Hernandez, C. Medina-Jaramillo, A. López-Córdoba and S. Goyanes, Food Hydrocolloids, 2017, 63, 488–495.
38. J. Ashfaq, I. A. Channa, A. A. Shaikh, A. D. Chandio, A. A. Shah, B. Bughio, A. Birmahani, S. Alshehri and M. M. Ghoneim, Materials, 2022, 15, 1046.
39. N. Singh, R. Chawla and S. Sivakumar, J. Packag. Technol. Res., 2021, 5, 29–40.
40. Y. Shen, Z. J. Ni, K. Thakur, J. G. Zhang, F. Hu and Z. J. Wei, Int. J. Biol. Macromol., 2021, 181, 528–539.
41. R. Sharma, Z. F. Bhat, A. Kumar, S. Kumar, A. E.-D. A. Bekhit and Z. Naqvi, J. Food Saf., 2021, 41, e12909.
42. C. Cai, R. Ma, M. Duan, Y. Deng, T. Liu and D. Lu, LWT, 2020, 131, 109700.
43. D. Fransiska, S. Nurbaity Kartika Apriani, M. Murdinah and S. Melanie, KnE Life Sci., 2015, 2, 63.
44. T. Sarkar and R. Chakraborty, Asian J. Water, Environ. Pollut., 2018, 15, 79–92.
45. K. Zahir, G. Nejib, S. Vandita, M. S. Rahman and A. Lyutha, Food Sci. Technol. Res., 2013, 19, 531–538.
46. G. F. Nogueira, F. M. Fakhouri and I. Velasco, Polymers, 2021, 373–386.
47. H. Ismail and N. Fasihah Zaaba, J. Elastomers Plast., 2012, 46, 96–109.
48. C. P. Y. Loo and N. M. Sarbon, Food Biosci., 2020, 35, 100589.
49. M. Wahidin, A. Srimarlita, I. Sulaiman and E. Indarti, IOP Conf. Ser. Earth Environ. Sci., 2021, 667, 012030.
50. A. Edhirej, S. M. Sapuan, M. Jawaid and N. I. Zahari, Int. J. Biol. Macromol., 2017, 101, 75–83.
51. K. M. Dang and R. Yoksan, Carbohydr. Polym., 2015, 115, 575–581.
52. K. A. Sharifi and S. Pirsa, Mater. Chem. Phys., 2021, 267, 124580.
53. R. G. M. van der Sman, Food Hydrocolloids, 2023, 139, 108481.
54. P. Jain, S. Tripathi, R. K. Deshmukh, K. K. Gaikwad and S. Singh, Biomass Convers. Biorefinery, 2024, 14, 20959–20972.
55. M. Mujtaba, L. Akyuz, B. Koc, M. Kaya, S. Ilk, D. Cansaran-Duman, A. S. Martinez, Y. S. Cakmak, J. Labidi and S. Boufi, Food Hydrocolloids, 2019, 89, 20–28.
56. L. Gao, T. Zhu, F. He, Z. Ou, J. Xu and L. Ren, Coatings, 2021, 11, 1–13.
57. P. Jamwal, G. S. Chauhan, P. Kumar, B. Kumari, K. Kumar and S. Chauhan, Sustain. Chem. Pharm., 2023, 32, 100950.
58. B. Kumari, S. Chauhan, G. S. Chauhan, K. Kumar, P. Jamwal and S. Ranote, Environ. Sci. Pollut. Res., 2024, 31, 21591–21609.
59. S. Chauhan, P. Jamwal, G. S. Chauhan, K. Kumar, B. Kumari and S. Ranote, Int. J. Biol. Macromol., 2024, 279, 135313.
60. B. Kumari, S. Chauhan, K. Kumar, S. Singh, S. Ranote, R. Kumar and G. S. Chauhan, Chemosphere, 2024, 364, 143103.
61. A. H. D. Abdullah, S. Chalimah, I. Primadona and M. H. G. Hanantyo, IOP Conf. Ser. Earth Environ. Sci., 2018, 160, 012003.
62. S. Shivangi, D. Dorairaj, P. S. Negi and N. P. Shetty, Food Hydrocolloids, 2021, 121, 107046.
63. J. O. Airouyuwa and T. Kaewmanee, Food Sci. Technol. Int., 2019, 25, 533–543.
64. P. Safaei, Z. Sadeghi and G. J. Khaniki, Jundishapur J. Health Sci., 2019, 11, 11–15.
65. S. K. Khurana, R. Tiwari, K. Sharun and M. Iqbal, J. Pure Appl. Microbio.l, 2019, 13, 1995–2012.
66. T. De Pilli, Food Control, 2020, 114, 107273.
67. D. Shende, M. Kour and A. K. Datta, J. Food Meas. Charact., 2020, 14, 3227–3237.

---