Optimization of fumed silica and melatonin loadings in polyvinyl alcohol-based films for postharvest preservation of rambutan fruit

Abstract

To address high hydrophilicity of polyvinyl alcohol (PVA) and agglomeration tendency of fumed silica (FS), melatonin (MT)-functionalized PVA/FS composite films were prepared to enhance their performance and applicability. MT (1–3%) mitigated FS aggregation and promoted interfacial adhesion, forming a denser network. The P/S-M2 (2% MT) film exhibited the best performance among all tested films, showing a 17.4% increase in tensile strength, an 88.1% decrease in water vapor permeability, and reductions of 78.8%, 46.0%, and 32.6% in moisture content, water solubility, and swelling ratio, respectively, compared to the P/S film. The P/S-M2 film also demonstrated enhanced biodegradability (74.7% degradation after 120 days) and antioxidant activity (85.3% ABTS radical scavenging rate). Packaging application of composite films to rambutan fruit delayed the changes of weight loss, chromaticity L*, and soluble solid contents, with the P/S-M2 film demonstrating the optimal effect. Overall, MT-functionalized PVA/FS films offer a promising packaging solution for rambutan preservation.

Highlights

• •Melatonin (MT) acted as a novel dispersant in PVA/silica (P/S) composites.
• •Composite film containing 2% MT (P/S-M2) showed optimal mechanical/barrier traits.
• •P/S-M2 film demonstrated strong antioxidant activity and improved biodegradability.
• •P/S-M2 film as a packing agent exerted a good preservation effect on rambutan fruit.

1. Introduction

Plastic packaging is widely adopted in the food industry owing to its superior moldability and barrier properties. However, its persistent environmental impact and potential risks to biosafety are raising growing concerns. Consequently, the development and utilization of biodegradable materials have become increasingly important. Polyvinyl alcohol (PVA), a polyhydroxyl polymer, has excellent film-forming ability, cytocompatibility and controlled biodegradability (Qiao et al., 2022). However, the high hydrophilicity of PVA leads to poor water resistance, inadequate barrier properties, and limited antioxidant capacity, severely restricting its practical application in food packaging (Song et al., 2025).

To overcome the shortcomings of PVA, PVA-based composite materials have been developed and extensively applied in multiple fields (Zhang et al., 2024). Composite strategies integrate multiple functional components (e.g., nanoparticles and bioactive substances) into polymer matrices, leveraging synergistic effects to enhance the mechanical strength, barrier performance, and multifunctionality for targeted applications (Oun et al., 2022). For example, in PVA-based composites, amphiphilic chitosan, when blended with PVA at an optimal mass ratio of 5:5 (PVA to chitosan), enhances mechanical and barrier properties by balancing interfacial interactions (Pan et al., 2023). Du et al. (2024) designed a PVA film containing Zn-doped carbon quantum dots (Zn-CQDs) and demonstrated that the mechanical properties of Zn-CQDs/PVA films improved with increasing Zn-CQDs content. Fumed silica (FS), an inorganic material with nanoscale dimensions, high specific surface area, high surface adsorption capacity, and cytocompatibility, is frequently incorporated into films as both a nanofiller and a carrier for bioactive substances (Jain et al., 2024; Videira-Quintela et al., 2022). This functional versatility is attributed to its surface hydroxyl groups and three-dimensional network structure, which synergistically enhance critical film properties, including the mechanical strength, barrier performance, water resistance, and sustained release of active substances (Velásquez et al., 2024; Yong et al., 2024). Nevertheless, the high surface energy of FS nanoparticles easily causes their aggregation in hydrophilic polymer matrices (e.g., PVA), leading to structural inhomogeneity and functional deterioration of the films (Wu et al., 2022). To address this issue, polysaccharides, polyphenols, and enzymes, which serve as ideal stabilizers and modifiers, tend to be added to nanomaterials. These substances reduce surface energy through diverse mechanisms involving charge regulation, hydrogen bonding and hydrophobic interactions, thereby improving the dispersion and stability of nanomaterials (Feng, Zhang, et al., 2024; Jain et al., 2024). Similarly, biopolymer-based systems such as cellulose microfibrils/guar gum hydrogels have been explored for their structural and functional contributions, including post-printing antioxidant activity, which aligns with the development of active packaging materials (Adedeji et al., 2024).

Melatonin (MT), an indole heterocyclic compound, has attracted extensive attention because of its remarkable biological and chemical activities, and is widely used in the agriculture, food, and medical industries (Wang et al., 2020). The MT molecule contains both polar groups and a hydrophobic indole ring, giving it an amphiphilic structure that enables intermolecular interactions with other substances through multiple mechanisms, including hydrogen bonding, π-π stacking, electrostatic interactions, and hydrophobic interactions. A study has shown that in zein/pectin nanoparticle systems, MT participates in nanoparticle formation through hydrophobic and hydrogen-bonding interactions (Zhang et al., 2025). Furthermore, MT acts as a biosurfactant to reduce interfacial tension between tamarind gum and pomelo essential oil, thereby preventing the aggregation of oil droplets (Feng, Wang, et al., 2024). These findings highlight the potential of MT as an effective surface modifier in composite systems. Recently, the functional applications of MT have been expanded to active packaging based on its beneficial effects on the material properties and preservation performance of agricultural products. For example, in a chitosan/carboxymethyl chitosan bilayer film system, the introduction of MT significantly enhances the tensile strength and water barrier properties, while endowing the film with antioxidant and antibacterial activities. This effectively maintains the color and texture of fresh-cut fruits and vegetables, such as cucumbers, broccoli, and melons (Zhao et al., 2020). Feng et al. (2024b) demonstrated that the addition of MT to glutenin/tamarind gum enhances the flexibility and barrier properties of the films, and the application of this MT-composite film to white mushrooms effectively delays browning by maintaining cellular energy homeostasis. These findings indicate that MT can enhance packaging performance by modulating the physical and chemical properties, offering a promising strategy for developing efficient and environmentally friendly preservation technologies.

Rambutan (Nephelium lappaceum L.) is a tropical fruit that belongs to the Sapindaceae family and is prized for its unique appearance, with a red pericarp and soft spine, succulent texture, sweet-and-sour taste, and rich nutritional profile (including carbohydrates, vitamins, minerals, and polyphenols) (Liu et al., 2025). However, rapid deterioration of postharvest quality caused by water loss, mechanical injury and pathogen infection severely limits its commercial distribution, resulting in a shelf life of only 2–3 days at ambient temperature (Shu et al., 2025). Appropriate packaging has been regarded as an effective means to maintain postharvest quality and extend the shelf life of rambutan fruit. Recent studies have explored active packaging strategies for rambutan fruit, such as using filter paper soaked with β-cyclodextrin/cinnamaldehyde nanoparticles (Shu et al., 2025) and pectin-based edible coatings incorporating trans-cinnamaldehyde (Sun et al., 2023), both of which delay fruit pericarp browning and prolong shelf life. However, these methods may raise potential hygiene and environmental concerns. In contrast, the PVA/FS/MT film developed in this study offers a more convenient and environmentally friendly solution for rambutan preservation.

To achieve this, MT was incorporated as a functional additive into a PVA/FS composite system. The regulatory effects of various MT concentrations on the microstructure, physical properties, and functional characteristics of the films were systematically assessed. Furthermore, the role of composite films in preserving the postharvest quality of rambutan fruit was investigated. This study provides novel insights into the development of active packaging films to extend the shelf life of perishable tropical horticultural products.

2. Materials and methods

2.1. Main reagents and materials

Polyvinyl alcohol (PVA; degree of saponification: 98–99%; molecular weight: 105,000) was purchased from Xilong Scientific Co., Ltd. (Shantou, China). Fumed silica (FS; molecular weight: 60.08; specific surface area: 260 m² g⁻¹; particle size: 7–40 nm) was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Polyethylene glycol 400 (PEG 400) was purchased from Beijing Solarbio Science & Technology Co., Ltd. (Beijing, China). Melatonin (MT; analytical grade, 99% purity) was purchased from Shanghai Yuanye Biotechnology Co., Ltd. (Shanghai, China). HepG2 hepatoma cells were provided by Xiamen Immocell Biotechnology Co., Ltd. (Xiamen, China). Rambutan (cv. Baoyan No. 7) fruit was harvested from a local orchard in Baoting County, Hainan Province, China. Fruit at a commercial mature stage, characterized by a bright red pericarp with green spinterns, uniform size, and absence of physical damage, were selected for the preservation experiment.

2.2. Film preparation

PVA (3.00 g) was dissolved in 100 mL of distilled water with magnetic stirring at 95 °C for 2 h to obtain a 3% (w/v) PVA solution. After cooling to room temperature, 0.18 g of PEG 400 was added to the PVA solution, followed by continuous stirring at 300 rpm for 1 h. Based on the evaluation of barrier and mechanical properties from the preliminary experiment, 0.015 g of FS was dispersed in 10 mL of anhydrous ethanol via magnetic stirring at 300 rpm for 1 min, yielding a clear and homogeneous suspension. Subsequently, different masses of MT (corresponding to 1%, 2%, and 3% relative to the PVA mass) were added to the FS suspension and stirred at 300 rpm for 1 min to prepare the FS/MT mixtures. The FS/MT mixtures were then incorporated into the PVA solution, followed by continuous stirring at 800 rpm for 1 h. Composite film solutions were obtained and designated as P/S-M1, P/S-M2, and P/S-M3, representing 1%, 2%, and 3% of MT relative to the mass of PVA, respectively. Finally, all the prepared film-forming solutions were cast into a circular plastic mold (radius 10 cm) and dried in an oven at 60 °C for 12 h to form PVA, P/S, P/S-M1, P/S-M2, and P/S-M3 films.

2.3. Characterization of the films

2.3.1. X-ray diffraction (XRD) characterization

XRD analysis was performed on an X-ray diffractometer (SmartLab SE, Rigaku Co., Saitama Prefecture, Japan) to evaluate the crystallinity and crystal structure that can influence the mechanical and barrier properties of the films (Jayakumar et al., 2023). Diffraction data were acquired over a 2θ range of 10–80° at a scan rate of 2° min⁻¹ in the continuous mode.

2.3.2. Fourier transform infrared (FTIR) analysis

FTIR analysis was performed on a Nicolet iS50 spectrometer (Thermo Fisher Scientific, Wisconsin, USA) to identify the functional groups and intermolecular interactions within the composite films (Shahvalizadeh et al., 2021). The instrument was configured to scan from 4000 to 400 cm⁻¹ (resolution: 4 cm⁻¹; 32 scans).

2.3.3. Scanning electron microscopy (SEM) observations

Field-emission scanning electron microscopy (FE-SEM; TESCAN MIRA LMS, Brno, Czech Republic) was used to observe the microstructure of the films. Prior to imaging, the samples were coated with a 10-nm-thick gold layer using a sputter coater (Leica EM ACE600, Wetzlar, Germany) to enhance conductivity. Micrographs were captured under high vacuum conditions and an accelerating voltage of 3.0 kV (Wu et al., 2022).

2.3.4. Thermogravimetric analysis (TGA)

To evaluate the thermogravimetric behavior of the composite films, TGA was performed using a TGA 550 analyzer (TA Instruments, New Castle, USA). The samples were heated from 30 °C to 600 °C under a nitrogen atmosphere at a heating rate of 10 °C/min (Li et al., 2024).

2.3.5. Thickness and optical properties

The thickness of the film was measured using a portable micrometer with a precision of 0.001 mm (IP64, Greenwood Tools Co., Ltd., Yantai, China). Ultraviolet–visible (UV–Vis) absorption spectra were recorded from 200 to 800 nm using a TU-1901 spectrophotometer (Puxi General Instrument Co., Ltd., Beijing, China). The film opacity at 600 nm was calculated as follows:

$$\mathit{Opacity}=-\log \left(T\right)/d$$

where T represents the transmittance of the film at 600 nm and d is the thickness of the film (mm).

2.3.6. Mechanical properties

The tensile strength (TS) and elongation at break (EB) of the films were measured using a texture analyzer (TA; XT plusC, Stable Micro Systems Ltd., Surrey, UK). The film samples were cut into 50 mm × 10 mm strips, clamped between the tensile grips, with an initial gauge length of 30 mm and a crosshead speed of 1 mm s⁻¹ (Zhao et al., 2020).

2.3.7. Water vapor permeability (WVP)

The WVP was determined using the gravimetric method (Wei et al., 2024). Briefly, a conical flask (bottle mouth diameter: 28 mm; depth: 57 mm) containing 5.0 g of anhydrous CaCl₂ was hermetically sealed with the test film. The conical flask was placed under controlled conditions (25 °C, 60% RH), and its mass was measured every 12 h. The WVP was calculated using the following formula:

$$\mathit{WVP}=\left(\mathrm{\Delta }W\times d\right)/\left(\mathrm{\Delta }t\times A\times \mathit{\Delta P}\right)$$

where ΔW is mass change (g), d is film thickness (m), Δt is time (s), A is effective film area (m²), and ΔP is vapor pressure difference (Pa).

2.3.8. Moisture content (MC), swelling ratio (SR), and water solubility (WS)

The methods reported in Wei et al. (2024) were employed to determine the MC, SR, and WS of the films. Each film strip (20 mm × 20 mm) was weighed (W₀), dried at 105 °C to a constant weight (W₁), and then soaked in a distilled water bath for 12 h at room temperature (W₂). Next, the undissolved film was dried at 105 °C (W₃) for 6 h. The MC, SR, and WS were calculated as follows:

$$\mathit{MC}\left(\%\right)=\left[\left(W_{\mathit{0}}-W_{\mathit{1}}\right)/W_{\mathit{0}}\right]\times 100$$

$$\mathit{SR}\left(\%\right)=\left[\left(W_{\mathit{2}}-W_{\mathit{1}}\right)/W_{\mathit{1}}\right]\times 100$$

$$\mathit{WS}\left(\%\right)=\left[\left(W_{\mathit{1}}-W_{\mathit{3}}\right)/W_{\mathit{1}}\right]\times 100$$

2.3.9. Antioxidant activity

The 2,2-diphenyl-1-picrylhydrazyl (DPPH) and 2,2-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical scavenging assays were adapted from Yu et al. (2024). For the DPPH assay, an 80 mg film sample was incubated with 20 mL of DPPH ethanolic solution (60 μM) in the dark for 24 h, and then centrifuged at 8000 rpm for 5 min. The absorbance (A₁) of the supernatant was measured at 517 nm. The DPPH solution without sample served as a control (A₀). For the ABTS assay, the working solution was prepared by mixing 7 mM ABTS with 2.45 mM potassium persulfate (1:1, v/v). The mixture was incubated at 25 °C for 16 h in the dark and then diluted to an absorbance of 0.70 ± 0.01 at 734 nm. A film sample (200 mg) was mixed with 20 mL of ABTS solution at 25 °C in the dark for 24 h and then centrifuged at 8000 rpm for 5 min, after which A₁ was measured at 734 nm.

The radical scavenging activity (%) for both assays was calculated as follows:

$$\mathit{Scavenging\; activity}\left(\%\right)=\left[1-\left(A_{\mathit{1}}/A_0\right)\right]\times 100$$

2.3.10. Biodegradability and biosafety evaluation

The biodegradability of the composite films was evaluated using the soil burial method (Fan et al., 2016). Briefly, film samples (5.0 cm × 5.0 cm) were buried in soil at a depth of 6.0 ± 0.5 cm for 120 days. They were retrieved at 30-day intervals, rinsed with distilled water, and dried at 25 °C and 60% RH for 24 h. The degradation rate was calculated by the following formula:

$$\mathit{Degradation\; rate}\left(\%\right)=\left[\left(W_0-W_t\right)/W_0\right]\times 100$$

where W₀ is the initial weight and W_t is the weight after degradation.

The biocompatibility of the films was assessed by measuring cell viability via the CCK-8 assay (Cai et al., 2023), which evaluated their cytotoxicity to HepG2 hepatoma cells in vitro. To obtain the extract solution, 0.10 g of sterile film sample was immersed in 100 mL of Dulbecco's Modified Eagle Medium (DMEM), followed by an ultrasonic treatment for 72 h. The extract was then diluted with DMEM to final concentrations of 50, 100, 200, 400, 600, 800, and 1000 μg mL⁻¹. HepG2 cells were seeded into 96-well plates at a density of 5 × 10⁴ cells per well (100 μL well⁻¹) and incubated for 24 h to allow cell attachment. The culture medium was then removed, and 100 μL of the diluted extract was added to each well. After 24 h of co-culture, the supernatant was discarded, and 90 μL of DMEM and 10 μL of CCK-8 reagent were added to each well. Following incubation for 2 h, the absorbance at 450 nm was recorded. Cell viability was calculated as follows:

$$\mathit{Cell\; viability}\left(\%\right)=\left({\mathit{OD}}_s–{\mathit{OD}}_b\right)/\left({\mathit{OD}}_c-{\mathit{OD}}_b\right)\times 100\%$$

where OD_s represents the absorbance of cell samples treated with the composite membrane extract, OD_c represents the absorbance of cells without treatment with the composite membrane extract, and OD_b represents the blank absorbance containing only the culture medium and CCK-8 reagent but no cells.

2.4. Application of the composite films for packaging rambutan fruit

2.4.1. Packaging processing

The selected rambutan fruit were disinfected with sodium hypochlorite solution (0.05%, v/v) for 3 min, rinsed with tap water, and air-dried. All fruit were packed in frustum-shaped polypropylene containers (upper diameter: 15 cm; lower diameter: 10 cm; height: 8 cm), with six fruit per container. Container openings were sealed with different films, including PVA, P/S, P/S-M1, P/S-M2, and P/S-M3 films, which represented five treatment groups. The containers without sealing served as the control group. Each treatment group consisted of three containers (3 replicates) at each analytical time point, which were used to evaluate the browning index (BI), chroma L*, soluble solids content (SSC), and fruit weight loss.

2.4.2. Preservation quality evaluation

The weight of both the fruit and the film from each container was recorded every two days (three containers per treatment, six fruits per container). The weight loss of the fruit was calculated as the percentage of reduced weight at each time point relative to the initial weight. The BI was determined according to the method of Wei et al. (2022), in which the degree of pericarp browning was classified into four grades, and the BI was calculated as: Σ (browning degree × number of fruit of this degree)/(4 × total number of fruit). The pericarp color was measured at two symmetric points on the equator of each fruit using a CR-400 chromameter (Konica Minolta, Tokyo, Japan), and the L* value (lightness) was recorded. The SSC of the fruit aril (pulp) was measured using a SW-35 T portable digital refractometer (Swevy Electronics, Guangzhou, China) and expressed as a percentage.

2.5. Statistical analysis

Three independent replicates were carried out for each parameter analysis, with data presented as the means ± standard errors (SEs). Statistical analysis was performed with one-way analysis of variance (ANOVA) and Duncan's multiple range test (P < 0.05), using SPSS 27.0 software. The data were graphed with GraphPad Prism 9.5.0 software.

3. Results and discussion

3.1. SEM observation

The surface and cross-sectional microstructures of the films are shown in Fig. 1. The surface of the PVA film was smooth and without cracks (Fig. 1A). In contrast, the addition of FS led to obvious agglomeration on the film surface, forming an irregular surface structure (Fig. 1B). One side of the cross-section of the P/S film presented a relatively dense fracture surface (Fig. 1G). This occurred because nanoscale FS with high surface activity and surface energy may nonuniformly disperse in PVA to form micrometer-sized agglomerates (Videira-Quintela et al., 2022).

Fig. 1.

SEM images of PVA (A), P/S (B), P/S-M1 (C), P/S-M2 (D) and P/S-M3 (E) film surfaces, and SEM images of PVA (F), P/S (G), P/S-M1 (H), P/S-M2 (I) and P/S-M3 (J) film cross-sections.

Compared with the P/S film, the addition of MT substantially improved the surface morphology of the composite film, reducing surface roughness and preventing aggregate formation (Fig. 1C and H). This improvement likely stems from the amphiphilic nature of MT, which facilitates interfacial integration between PVA and FS via hydrogen bonding and hydrophobic interactions, thus preventing FS agglomeration. This interfacial mechanism is consistent with the findings of Feng et al. (2024a), who demonstrated that MT acts as a surfactant analog at oil–water interfaces, promoting essential oil dispersion stability in gum matrices. Notably, particle formation was observed on the film surface at a 3% MT concentration (Fig. 1E). At high concentrations, excessive MT self-aggregates via π-π stacking of its indole rings, disrupting the polymer-filler network uniformity and film integrity.

3.2. XRD, FTIR, and thermal stability

XRD was employed to analyze the crystalline structure of the composite films. Pure PVA exhibited a broad diffraction peak at 2θ = 23° (Fig. 2A), which is characteristic of a semicrystalline nature (Jayakumar et al., 2023). FS incorporation into the PVA film decreased the diffraction peak intensity and reduced the d-spacing from 0.388 nm to 0.37 nm (Fig. 2A). These observations demonstrate that FS simultaneously enhances the molecular packing density in PVA while partially disrupting its crystalline structure. This dual effect may arise from nonuniform FS dispersion and hydrogen bonding interactions between FS and PVA, which compromise the intrinsic hydrogen-bonded network of PVA. In contrast, adding MT to the P/S films significantly increased the crystallinity and shifted the diffraction peaks. The peaks for P/S-M1, P/S-M2, and P/S-M3 were observed at 19.06°, 19.76°, and 19.64°, corresponding to d-spacings of 0.464 nm, 0.449 nm, and 0.452 nm, respectively (Fig. 2A). The increased d-spacing and enhanced crystallinity resulted from the formation of multiple hydrogen bonds among MT (–NH–CO–), PVA (–OH), and FS (Si–OH), combined with hydrophobic interactions. Such interactions reconstruct an ordered molecular structure, expand intermolecular spacing, and facilitate the alignment of PVA chains, hereby reducing amorphous disorder (Jayakumar et al., 2023). Similar findings were also reported for bionanocomposite films containing gelatin, tragacanth gum, and zinc oxide nanoparticles (Shahvalizadeh et al., 2021). However, excess MT (e.g., P/S-M3) led to structural heterogeneity due to MT aggregation, disrupting the crystal lattice and consequently reducing crystallinity.

Fig. 2.

Characterization of the composite films. X-ray diffraction patterns (XRD) (A), fourier transform infrared spectroscopy (FT-IR) (B), thermogravimetric curves (TG) (C), thermogravimetric derivative curves (DTG) (D), UV–vis transmittance profile (E), and physical photographs (F).

The FTIR spectroscopy results shown in Fig. 2B revealed the molecular interactions in the composite films. The characteristic peaks of pure PVA appeared at 3305 cm⁻¹ (O—H stretching vibration), 2910 cm⁻¹ (C—H stretching), 1440 cm⁻¹ (CH₂ bending), and 1090 cm⁻¹ (C—H bending/C–O stretching). Upon FS incorporation, the O—H stretching peak at 3305 cm⁻¹ red-shifted to 3281 cm⁻¹, confirming the presence of hydrogen bonding between FS silanol groups (Si–OH) and PVA hydroxyl groups (Yao et al., 2011). Additionally, the peak at 1090 cm⁻¹ was broadened and intensified due to the overlap of FS Si–O–Si and PVA C–H/C–O vibrations, confirming the incorporation of FS. The addition of MT induced changes in the FTIR spectra, with new vibrational bands at 1640 cm⁻¹ (amide I, C O stretching), 1560 cm⁻¹ (amide II, N—H bending coupled to C=C/C=N stretching), and 1320 cm⁻¹ (amide III, C—N stretching), verifying molecular incorporation. The composite O–H/N–H peak at 3279 cm⁻¹ exhibited significant sharpening and intensification, which is consistent with the enhanced hydrogen bond formed by the acetylamino group (–NHCO–) of MT that could restructure the PVA network. Synchronous enhancements at 2910 cm⁻¹ (C—H stretching) and 1444 cm⁻¹ (CH₂ bending) revealed reorganization of the alkyl chains, whereas the intensified 1090 cm⁻¹ band indicated improved FS dispersion uniformity. This synchronized peak evolution correlates with the microstructural homogeneity revealed by SEM. Notably, the absence of new absorption bands after FS/MT incorporation indicates that FS–PVA–MT interactions are primarily physical (hydrogen bonding and van der Waals forces).

The thermal degradation behavior of the composite films was characterized by TGA/DTG, as shown in Fig. 2C and D. All films exhibited similar four-stage decomposition profiles: initial loss of moisture and volatiles (60–170 °C), followed by hydroxyl dehydration and crystalline disruption (200–350 °C), main chain scission (350–430 °C), and residual decomposition (430–650 °C) (Fig. 2C). TGA revealed that the residual mass of pure PVA film at 313 °C was 72.3% (Fig. 2C). In contrast, the residual mass at 313 °C for the composite films increased to varying degrees, with the following order: P/S-M2 (79.0%) > P/S (77.7%) > P/S-M3 (75.4%) ≈ P/S-M1 (75.1%) (Fig. 2C), indicating that thermal stability increased after the addition of FS and MT. Furthermore, DTG revealed that the temperature corresponding to the maximum thermal decomposition rate increased significantly from 313 °C (PVA) to 319 °C (P/S), 350 °C (P/S-M1), 379 °C (P/S-M2), and 317 °C (P/S-M3), while the maximum decomposition rate decreased, with P/S-M2 exhibiting the most pronounced improvement (Fig. 2D). These findings collectively indicate that the incorporation of both FS and MT, particularly 2% MT in P/S-M2, enhances the thermal stability of the PVA film. This enhancement can be attributed to two main factors: (1) FS provides effective thermal insulation, restricting the chain transfer process within the material; (2) the C O and –NH groups in MT form hydrogen bonds with the –OH groups in PVA/FS, while π-π stacking between MT's indole rings and synergistically hydrophobic interactions further promote cross-linking and interfacial reinforcement, thereby enhancing thermal stability (Li et al., 2024; Wu et al., 2022).

3.3. Optical characteristics of the films

UV radiation induces photochemical degradation in foods, disrupting the physical structure, chemical composition, and nutritional value of photosensitive compounds (Ezati et al., 2023). Compared with pure PVA, the addition of FS (P/S film) resulted in a minimal improvement in the UV-blocking capacity, as indicated by the nearly overlapping UV absorption spectra (Fig. 2E). In contrast, MT significantly enhanced UV shielding in the PVA/FS composites. The optimized P/S-M2 film achieved complete blockage of UV-B (280–320 nm) and UV-C (200–280 nm), along with partial blockage of UV-A (320–400 nm) (Fig. 2E). This dramatic enhancement is primarily attributed to the intrinsic light-absorbing properties of the aromatic ring structure of MT (Singh et al., 2014). The opacity data revealed that FS addition caused a negligible opacity change (PVA: 0.63 ± 0.02; P/S: 0.65 ± 0.05), whereas MT progressively increased opacity from 0.97 ± 0.06 (P/S-M1) to 2.31 ± 0.10 (P/S-M3) (Table 1), which was correlated with reduced visible light transmittance (Fig. 2E). These results suggest that the MT provides essential UV protection for photosensitive food packaging films, despite increasing film opacity.

Table. 1.

Thickness, tensile strength (TS), elongation at break (EB), and opacity of PVA, P/S, P/S-M1, P/S-M2, and P/S-M3 composite films.

Film	Thickness (μm)	TS (MPa)	EB (%)	Opacity (mm−1)
PVA	61.50 ± 1.77b	50.18 ± 0.96a	286.71 ± 3.42a	0.63 ± 0.02d
P/S	67.33 ± 3.09ab	44.66 ± 1.35b	176.37 ± 6.19c	0.65 ± 0.05d
P/S-M1	65.67 ± 2.43ab	45.08 ± 1.17b	256.94 ± 4.16b	0.97 ± 0.06c
P/S-M2	66.00 ± 1.93ab	52.43 ± 1.87a	279.34 ± 3.50a	1.69 ± 0.09b
P/S-M3	70.50 ± 2.73a	39.28 ± 0.98c	285.07 ± 7.38a	2.31 ± 0.10a

Different letters represent significant difference between any two groups within one column (P < 0.05).

As shown in Fig. 2F, increasing the MT concentration changed the visual appearance of the P/S films. Although the PVA base film is colorless and transparent, higher MT concentrations reduce the transparency and add a yellowish hue, indicating changes in the optical and structural properties. The observed increase in opacity and development of a yellowish tint could compromise consumers' capacity to accurately evaluate the freshness and quality of packaged fruit, thereby potentially influencing their purchasing choices.

3.4. Thickness and mechanical properties of the films

Film thickness is a critical parameter that affects the barrier performance and mechanical properties of the film. The thicknesses of the composite films are presented in Table 1. The thinnest film was observed for pure PVA (61.50 ± 1.77 μm). With the addition of FS and MT, the thickness of the film tends to increase. This increase was attributed to the higher dry matter content resulting from the incorporation of FS and MT, a common observation in composite materials. A similar finding was reported by Tian et al. (2021) for a composite film incorporated with hops β-acids and chitosan-SiO₂, in which increasing dosage of β-acids and SiO₂ leads to the enhancement of film thickness.

The TS and EB are important indicators for evaluating the reliability of packaging film structures (Li et al., 2020). The effects of FS and MT on the TS and EB of the composite films are shown in Table 1. Pure PVA has excellent mechanical properties, with a TS of 50.18 ± 0.96 MPa and an EB of 286.71 ± 3.42%. However, the addition of FS reduced these values to 44.7 ± 1.4 MPa and 176.37 ± 6.19%, respectively. This reduction is due to stress concentration caused by silica aggregation,  which compromises the mechanical properties. Compared with P/S films, the addition of MT significantly improved the mechanical properties, although this effect depended on the MT dosage. The TS of the P/S-M2 film group (52.43 ± 1.87 MPa) was significantly higher than that of the P/S group (44.66 ± 1.35 MPa). The amphiphilic structure of MT enables multiple molecular interactions with the PVA matrix and FS nanoparticles, including hydrogen bonding, π-π stacking, and hydrophobic interactions, with both the PVA matrix and FS nanoparticles. These interactions facilitate the uniform dispersion of FS, reduce interfacial discontinuities, increase film density, and thereby enhancing mechanical robustness. However, a further increase in the MT concentration did not result in higher TS values, which is likely attributed to the disruption of the PVA chain structure at elevated MT concentrations. Additionally, the addition of MT increased the EB of the composite films, which indicates that MT can enhance the flexibility of the film. This finding is consistent with previous research results (Feng et al., 2023; Feng et al., 2024b). It is worth noting that MT's high loading led to particle aggregation and potential phase separation, which can compromise interfacial cohesion and mechanical integrity. This phenomenon was also reported in other composite film systems with phase-separated morphologies (Wang et al., 2025).

3.5. WVP

A low WVP is critical for inhibiting water and nutrient loss in foods, preserving their appearance and integrity, and extending the shelf life. As shown in Fig. 3A, the WVP of the pure PVA film was 9.22 × 10⁻¹¹ g m⁻¹ s⁻¹ Pa⁻¹. Upon incorporation of FS, the WVP decreased by 73.1% (Fig. 3A), which can be attributed to FS acting as a nanofiller to hinder the diffusion of water vapor molecules by creating a more tortuous path inside the film (Eze et al., 2024). MT addition further reduced the WVP, with values for P/S-M1, P/S-M2, and P/S-M3 of 4.22, 2.97, and 6.42 × 10⁻¹² g m⁻¹ s⁻¹ Pa⁻¹,  respectively (Fig. 3A). The superior performance of P/S-M originates from the amphiphilic structure of MT, where hydrophilic groups (–NH₂ and –OH) and hydrophobic moieties (the indole ring and methyl groups) synergistically enhance the molecular interactions between FS and PVA. These interactions promote the formation of a dense and ordered network, which physically impedes water vapor diffusion through two mechanisms: (1) the increased network density reduces the channels for water molecules; (2) the well-dispersed FS nanoparticles create a more tortuous diffusion pathway in the film matrix, slowing down the diffusion of water vapor (Feng et al., 2024b). This approach aligns with recent advances in developing humidity-resistant biopolymer barriers, such as water-resistant cellulose nanofibril films designed for exceptional barrier performance under high-humidity conditions (Zheng et al., 2026).

Fig. 3.

Water barrier properties and antioxidant activities of the composite films. Water vapor permeability (WVP) (A), moisture content (MC) (B), swelling ratio (SR) (C), water solubility (WS) (D), DPPH radical scavenging activity (E), and ABTS radical scavenging activity (F). Data are presented as the means ± SEs (n = 3). Different letters above the bars represent significant differences among the groups (P < 0.05).

3.6. MC, SR, and WS of the films

The MC, SR, and WS are critical parameters for evaluating the physical, mechanical, and processing properties of films (Wu & Li, 2022). The pure PVA films demonstrated the highest values (MC: 10.6 ± 0.4%; SR: 387.0 ± 1.3%; WS: 29.1 ± 0.4%) in Fig. 3B–D, attributed to water absorption by hydrophilic –OH groups and consequent amorphous region expansion (Deng et al., 2024). The incorporation of FS into the film significantly reduced the values of these parameters, with a MC of 7.5 ± 0.1%, SR of 334.7 ± 0.8%, and WS of 24.8 ± 0.7%, corresponding to 29.0%, 13.5%, and 14.7% reductions, respectively, relative to those of the pure PVA film (Fig. 3B–D). This phenomenon is attributed to FS filling the polymer voids and restricting molecular chain mobility, thereby impeding water permeation. The P/S-M2 film exhibited the lowest parameter values (MC: 1.6 ± 0.1%; SR: 225.5 ± 1.4%; WS: 13.4 ± 1.0%) (Fig. 3B–D), resulting from MT-mediated cross-linking and surface modification that collectively decrease free –OH groups and water absorption (Wu & Li, 2022). Notably, excessive MT increased water absorption, as indicated by the higher MC (9.5 ± 0.3%) of the P/S-M3 film than those of the P/S, P/S-M1, and P/S-M2 films (Fig. 3B), which is likely due to disruption of the PVA molecular structure by high concentrations of MT.

3.7. Antioxidative ability of the PVA, P/S, and P/S-M films

The radical-scavenging ability of packaging films is critical for preventing food oxidation and extending their shelf life. As shown in Fig. 3E, the PVA, P/S, and P/S-M1 films exhibited negligible DPPH scavenging activity. The incorporation of 2 wt% and 3 wt% (P/S-M2 and P/S-M3) increased the DPPH radical scavenging activity by 20.2-fold and 21.2-fold, respectively, compared with that of the pure PVA film.

The ABTS radical scavenging activity of the films is shown in Fig. 3F. Among all the tested films, the PVA film had the lowest ABTS radical scavenging activity (6.3 ± 1.2%), which is attributable to the limited reducibility of its hydroxyl groups (Dheeravathu & Shetty, 2025). Compared with the pure PVA film, the P/S films demonstrated significantly higher ABTS radical scavenging activity (14.5 ± 1.0%),  primarily due to radical adsorption and redox reactions mediated by the Si–OH groups on the silica nanoparticles. The incorporation of MT dramatically enhanced ABTS radical scavenging activity to 85.3 ± 0.7%, indicating that MT is the primary contributor to the antioxidant capacity of the composite films. This enhancement originates from electron/hydrogen donation by the indole rings and amide groups of MT, which effectively terminate free radical chain reactions (Wang et al., 2024).

3.8. Biodegradability and cytocompatibility of the films

The biodegradation kinetics of composite films are influenced by their chemical composition, including their functional group distribution and network cross-linking density (Jayakumar et al., 2023). As shown in Fig. 4, all the films underwent substantial degradation after 120 days of soil burial. The residual mass percentages for PVA, P/S, P/S-M1, P/S-M2, and P/S-M3 were 58.0%, 42.7%, 29.3%, 25.3%, and 23.3%, respectively. The greater extent of biodegradation observed in P/S-M films suggests an interaction between FS and MT, which may contribute to the formation of hydrolytically labile sites on the film surface, consequently accelerating material degradation. A cytotoxicity assessment demonstrating the biocompatibility of the materials is shown in Fig. S1. None of the tested films significantly affected HepG2 cell viability, with all the film extract solutions (at concentrations of 50–1000 μg mL⁻¹) maintaining cell viability at approximately 95.0%. These results have significant practical implications: the rapid degradation of P/S-M films under soil burial conditions offers an advantage over conventional plastics and certain bioplastics that require industrial composting; meanwhile, their excellent cytocompatibility supports their safety for direct food contact.

Fig. 4.

Biodegradability of PVA, P/S, P/S-M1, P/S-M2, and P/S-M3 films after 120 days of soil burial. Data are presented as the means ± SEs (n = 3). Different letters represent significant difference between any two groups at same time point or same film concentrations (P < 0.05).

3.9. Effects of composite films on rambutan preservation

Rapid browning and senescence are bottlenecks that restrict the marketing of rambutan fruit (Liu et al., 2025). The development of active packaging materials that can modify the in-package O₂/CO₂ ratio and maintain high humidity is an effective strategy for preserving rambutan quality and extending its shelf life. In the present study, the control fruit showed progressive weight loss, pericarp browning, discoloration (chromaticity loss), and a decrease in SSC during storage, and the quality of the fruit was almost completely lost after 6 days of storage (Fig. 5A–E). The application of the films delayed the changes in weight loss, BI, and chromaticity L* to varying degrees in the rambutan fruit, with the P/S-M2 film demonstrating the most efficacy among all the tested films (Fig. 5A–E). Interestingly, the SSC in control group was higher than that in the other treatment groups with films, which could be attributed to cytosol concentration of pulp cell due to the extremely rapid desiccation in control fruit. The pure PVA film showed a relatively weaker preservation efficacy for rambutan fruit, compared with the other tested films (Fig. 5A–E). This finding may result from the insufficient moisture barrier capacity and strong hydrophilicity of the PVA film (as reflected by the higher WVP and MC, Fig. 3A and B), which enable the PVA films to spontaneously absorb moisture from rambutan fruit while releasing vapor to the external environment, thereby promoting fruit water loss and reducing in-package humidity, ultimately accelerating the deterioration of fruit quality. In addition, the inferior preservation effect of the P/S-M3 film compared with that of the P/S-M2 film may be attributed to its higher MC, which is related to the disruptive effect of the high MT concentration on the microstructure of the film (Fig. 1E and J, and Fig. 3B).

Fig. 5.

Fruit appearance (A), weight loss (B), browning index (BI) (C), chromaticity L* (D) and soluble solids content (SSC) (E) in ‘Baoyan 7’ rambutan fruit during storage at 25 °C after sealing with PVA, P/S, P/S-M1, P/S-M2 and P/S-M3 films. Fruit in the package without sealing with films were served as control. Data are presented as the means ± SEs (n = 3). Different letters represent significant difference between any two groups at same time point (P < 0.05).

4. Conclusion

In this study, a multifunctional PVA-based composite film was developed by incorporating FS and MT. Structural characterization (SEM, FTIR, and XRD) revealed that the amphiphilicity of MT facilitated hydrogen bonding between PVA hydroxyl groups and FS silanol groups, enabling homogeneous nanoparticle dispersion and the formation of a three-dimensional polymer–nanoparticle network. Variations in the MT concentration differentially affected the mechanical integrity, thermal resilience, antioxidant capacity, and water vapor barrier properties of the composite film. Notably, the P/S-M2 composite film exhibited superior mechanical robustness and thermal stability, coupled with a relatively low WVP and enhanced antioxidant activity. The application experimental results demonstrated that the preservation efficacy of the PVA, P/S, and P/S-M films on rambutan fruit was varied. Among them, the P/S-M2 film showed the best performance, which could be attributed to its excellent physical and structural properties that effectively reduced the water loss and delayed pericarp browning. This study developed a theoretical framework for the application of MT-functionalized composite packaging materials to enhance the postharvest preservation efficacy of tropical fruits. To further validate the general applicability and practical performance of MT-functionalized PVA/FS films, future work will involve comparing them with other commercial packaging films and investigating their preservation effects on other Sapindaceae fruits (e.g., litchis and longans) stored under ambient and cold conditions.

CRediT authorship contribution statement

Jingyi Guo: Writing – original draft, Software, Methodology, Investigation, Data curation, Conceptualization. Lipeng Wei: Software, Methodology, Investigation. Jialiang Liu: Writing – review & editing, Methodology. Xiangbin Xu: Writing – review & editing. Weimin Zhang: Writing – review & editing, Project administration, Conceptualization. Zhengke Zhang: Writing – review & editing, Supervision, Project administration, Investigation, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary material

Data availability

Data will be made available on request.