Intelligent indicator films based on cold plasma-modified chitosan loaded with curcumin-resveratrol nanoemulsions: pH-responsive mechanism, colorimetric performance and application for tilapia freshness monitoring

Abstract

Food safety risks associated with spoilage and plastic packaging have driven the development of sustainable intelligent packaging systems. In this study, zein-curcumin-resveratrol‑sodium alginate nanoemulsions were incorporated into chitosan films to construct pH-responsive active indicator films. Cold plasma (CP) modification was applied to regulate surface properties and enhance sensing performance. CP modification increased surface roughness and active sites, improving the interaction between films and volatile amines. The 30 kV CP-modified film exhibited optimal comprehensive properties, with improved mechanical strength (tensile strength of 5.36 ± 0.18 MPa and elongation at break of 46.95 ± 3.14%), antioxidant activities (73.81 ± 1.50% for DPPH and 70.26 ± 1.41% for ABTS), pH responsiveness, barrier properties and UV protection. Moreover, colorimetric responses (ΔE and a* values) showed a significant positive correlation with tilapia TVB-N accumulation (R² > 0.95), verifying the quantitative relationship between film discoloration and fish freshness deterioration. This study provides a practical intelligent packaging strategy for real-time aquatic product quality monitoring.

Graphical abstract

Highlights

• •Zein nanoemulsions encapsulated curcumin and resveratrol for pH-sensitive smart films.
• •Cold plasma (CP) modification regulated bioactive release in composite films.
• •CP at 30 kV endowed films with excellent mechanical, barrier and antioxidant properties.
• •The films showed pH-responsive color change to volatile amines from tilapia spoilage.

1. Introduction

As a globally important dietary resource with high yield and nutritional value, aquatic products are prone to fast spoilage in storage and transit. Such deterioration is driven by microbial growth and endogenous biochemical metabolism, which not only degrade product quality and trigger food safety risks but also cause severe post-harvest losses and restrict aquaculture development (Bhalavey et al., 2026). Traditional preservation and monitoring approaches cannot fulfill the need for real-time and visual quality control, as they rely on destructive sampling and conventional analytical procedures that fail to provide timely spoilage information to consumers and retailers (Das & Mishra, 2023). Therefore, the fabrication of functional packaging with preservation and intelligent freshness-monitoring capabilities is vital for curbing aquatic product spoilage losses and safeguarding food safety.

To address the environmental concerns raised by non-degradable packaging and the rapid quality deterioration of aquatic products, green films integrating preservation and real-time indication functions have emerged as a prominent research focus. With abundant natural sources, full biodegradability, and superior film-forming properties, chitosan serves as a viable replacement for traditional plastics (Li et al., 2026). Prior investigations have proven that chitosan coatings can retard fish quality deterioration and prolong their shelf life by 2 days under refrigerated conditions (Karsli et al., 2021). Nevertheless, chitosan films lack the capacity for real-time freshness monitoring. Meanwhile, their inherent antibacterial and antioxidant properties are inadequate for the prolonged preservation of aquatic products. In order to compensate for these drawbacks, bioactive additives are integrated into the base matrix. The resultant pH-responsive intelligent films can achieve effective aquatic product preservation and visual pH indication.

Among natural bioactive compounds, curcumin from turmeric is a typical pH-sensitive polyphenol. Curcumin has a keto enol tautomeric structure in its molecule and can respond sensitively to environmental pH changes, serving as a widely used colorimetric indicator in intelligent packaging (Hu et al., 2026; Huang et al., 2025). Resveratrol is a polyphenolic bioactive compound rich in hydroxyl groups, and it exhibits dual antioxidant and antibacterial properties by efficiently scavenging free radicals and suppressing pathogenic bacterial growth (Guo et al., 2024). The combination of curcumin and resveratrol achieves synergistic effects of freshness indication and antibacterial preservation, which significantly enhances the functionality and practicality of intelligent indicator films. However, curcumin and resveratrol suffer from inherent limitations of poor water solubility, low stability and easy aggregation. Direct incorporation into film matrices results in poor dispersion and impaired color responsiveness (Bouchard et al., 2026; Shen et al., 2022). The development of nanoemulsion (NEs) technology offers an effective strategy to overcome the above drawbacks. In contrast to the previously reported zein–curcumin particle systems (Yang et al., 2026), the NE-based delivery platform not only improved colloidal stability and achieved sustained release of curcumin, but also enhanced the pH-responsive colorimetric performance. NEs encapsulation protects curcumin and resveratrol against UV irradiation and oxidative degradation while promoting uniform dispersion of bioactive substances within film matrices (Aswathanarayan & Vittal, 2019; Manzoor et al., 2023). They not only effectively extend the food preservation period but also significantly enhance the pH response sensitivity and antioxidant efficacy of intelligent indicator films, thus offering a feasible strategy for developing intelligent food packaging.

As an eco-friendly treatment, cold plasma (CP) modification improves the overall performance of composite films through surface etching and polar group incorporation to tailor film structure and surface properties (Kongboonkird et al., 2024a). The material bulk structure remained intact after treatment, whereas composite film performance and active substance release stability were obviously improved. At present, CP modification can effectively enhance the physicochemical and functional characteristics of protein-polysaccharide composite films (Guo et al., 2026). This technique offers a solid technical basis for the performance optimization and practical utilization of intelligent indicator films.

Hence, this study aims to develop pH-responsive intelligent active chitosan films embedded with curcumin-resveratrol NEs for tilapia preservation and freshness visualization. Sodium alginate-stabilized NEs co-encapsulating curcumin and resveratrol were incorporated into chitosan matrices, followed by CP surface modification. We hypothesized that CP acts as a surface regulation strategy. Through surface etching and polar group incorporation, CP optimizes film surface characteristics and facilitates uniform dispersion of NEs droplets. This treatment is expected to enhance curcumin stability and pH sensitivity, optimize film surface-related functional performances, and realize visual tilapia spoilage indication. The film pH-responsive property and practical monitoring capacity were evaluated using TVB-N, pH, color difference and texture. This work provides theoretical support to develop eco-friendly intelligent aquatic packaging substituting traditional plastics.

2. Materials and methods

2.1. Materials

Zein (purity ≥95%, CAS: 9010–66-6, Lot#: H2414283, Aladdin Biotech, Shanghai, PRC), resveratrol (purity ≥99%, molecular weight: 228.24 g/mol, CAS: 501–36-0, Lot#: F2312859, Aladdin Biotech, Shanghai, PRC), curcumin (purity ≥95%, CAS: 458–37-7, Lot#: 211006 J, DASF Biotechnology, Nanjing, PRC), chitosan (degree of deacetylation ≥95%, molecular weight: 50–60 kDa, CAS: 9012-76-4, Lot#: C17786834, Macklin Biotechnology Co., Ltd., Shanghai, China) and sodium alginate (viscosity: 15–25 cP at 1% w/v in deionized water, mannuronic acid/guluronic acid ratio = 1.56, CAS: 9005-38-3, Lot#: MKCX6600, Sigma-Aldrich, USA) were used without further purification. Soybean oil (purity ≥99%) was purchased from a local supermarket in Hainan, China. All other chemical reagents and solvents were of analytical grade.

2.2. Preparation of the ZCRS NEs

The fabrication of zein-curcumin-resveratrol‑sodium alginate (ZCRS) NEs followed the approach described by Zhang, Sun, et al. (2025), with the dosages of key materials determined with reference to Chen, Han, et al. (2020) and slightly revised. Zein (0.5 g) and curcumin (0.05 g) were dissolved in 80% (v/v) aqueous ethanol solution (50 mL). The mixture was continuously stirred at 500 rpm for 5 h with a magnetic stirrer (HJ-6B, Liangyou Instruments Co., Ltd., Jintan District, Changzhou). The above solution was added dropwise into deionized water at 1:3 (v/v) to fabricate zein-curcumin (ZC) nanoparticles (NPs). Subsequently, 10 mL of resveratrol solution (5 mg/mL) was injected into the aforementioned mixture to prepare zein-curcumin-resveratrol (ZCR) NPs. Ultimately, sodium alginate was dissolved in deionized water to prepare a 1% (w/v) solution. The ZCRS NPs were then fabricated by mixing this solution with the previously prepared ZCR NPs dispersion at a 1:1 (v/v) ratio. NPs were mixed with soybean oil and homogenized using a shear homogenizer (T18 digital, IKA, Germany) at 12000 rpm for 5 min to obtain an emulsion with 5% (v/v) oil. The resultant emulsion was subjected to three consecutive homogenization cycles in a high-pressure homogenizer (Nano Genizer 30 k, Suzhou Microfluidic Nano Technology, China) at 100 MPa to yield ZCRS NEs. The ZCRS NEs were aliquoted and kept at 4 °C for later characterization and evaluation.

2.3. Characterization of the ZCRS NEs

The ZCRS NEs were diluted with deionized water (1:100, v/v), and their average particle size and zeta potential were analyzed via the laser particle sizer (Zetasizer Nano ZS90, Malvern Instruments Ltd., UK). The morphology of lyophilized ZCRS NEs was observed by scanning electron microscopy (Verios G4 UC, Thermo Fisher, USA) at 2500× magnification. Additionally, the microstructure of 5 μL ZCRS NEs deposited onto a glass slide was characterized under an optical microscope.

Curcumin (Cur) and resveratrol (Res) encapsulation efficiencies (EE) were determined according to the procedure described in Wang, Yin, et al. (2025). The ZCRS NEs were dissolved in 80% (v/v) ethanol, followed by vortexing for 5 min. The supernatant collected after emulsion centrifugation was used to quantify curcumin and resveratrol at 425 nm and 306 nm via a UV spectrophotometer. Calculations of the encapsulation efficiencies of Cur (EE₁) and Res (EE₂) were performed with the aid of Eqs. (1) and (2), respectively.

$${\mathrm{EE}}_1=\frac{\text{totalCur}-\text{freeCur}}{\text{totalCur}}\times 100\%$$ (1)

$$\begin{array}{c}{\mathrm{EE}}_2=\frac{\text{totalRes}-\text{freeRes}}{\text{totalRes}}\times 100\%\end{array}$$ (2)

2.4. Preparation of CP-modified composite films

The composite films were fabricated according to the approach reported by Fang et al. (2025). Briefly, chitosan (2 g) was dissolved in dilute acetic acid (1%, v/v) to prepare a 2% (w/v) chitosan solution (100 mL). Glycerol (0.4 g, 20%, w/w, based on dry chitosan weight) was incorporated into the above system, followed by magnetic stirring at 55 °C and 500 rpm for 1 h. The resultant mixture was homogenized with the NEs at a volume ratio of 5:1 (v/v) via high-speed shearing (10,000 rpm, 3 min) to obtain the composite solution. Subsequently, 30 mL of the composite solution was cast into Petri dishes (90 × 15 mm), followed by drying at 40 °C for 48 h to yield C-NEs films.

The C-NEs films were subjected to cold plasma (CP) modification. Specifically,

the films were placed on a quartz sample stage in a plasma reactor and treated with dielectric barrier discharge cold plasma (BK130/36, Phoenix Electric Co. Ltd., USA). Ambient air served as the discharge gas, operated at a current of 0.024 mA and a fixed frequency of 120 kHz for 3 min, with the applied voltage adjusted to three levels (10 kV, 30 kV, and 50 kV). The resulting films were coded as C-NEs-0 kV, C-NEs-10 kV, C-NEs-30 kV, and C-NEs-50 kV.

2.5. Microscopic characterization of NEs composite films

2.5.1. Scanning electron microscopy (SEM)

The surface features of the indicator films were characterized by SEM (Verios G4 UC, Thermo Fisher, USA). The film samples were mounted on copper stubs using conductive adhesive, followed by gold sputter coating for 1 min to enhance electrical conductivity. Microscopic imaging observations were carried out at 5 kV with a magnification of 100 × .

2.5.2. X-ray diffraction (XRD)

A Rigaku Smart Lab SE X-ray diffractometer (Rigaku, Japan) was used to characterize the XRD patterns of the films. The measurements were performed with Cu Kα radiation over a 2θ range of 5° to 40°. The instrument was operated at 40 kV and 30 mA, with a scanning rate of 10°/min.

2.5.3. FTIR

The molecular structure of all samples was characterized by Fourier transform infrared spectroscopy (FTIR) using a Bruker Tensor 27 spectrometer (Bruker, Germany). The film samples (2 mg) were mixed with dried KBr (200 mg) and compressed into transparent pellets, followed by spectrum acquisition over the wavenumber range of 4000–500 cm⁻¹ at a resolution of 4 cm⁻¹.

2.6. Characterization of NEs composite films

2.6.1. Thickness and mechanical properties of films

A digital micrometer was used to measure six random positions of the films. The mean value was regarded as film thickness. Tensile strength (TS) and elongation at break (EAB) of the films were measured with a TA.XT Plus texture analyzer (Stable Micro Systems, UK).

2.6.2. Color parameters, opacity and transparency

The color parameters L* (lightness), a* (redness) and b* (yellowness) of the films were measured with a colorimeter (CR-410, Konica Minolta, Japan) that had been calibrated against a standard white plate. Spectrophotometric measurements of film transmittance and opacity were performed on a UV-2700 instrument (Shimadzu, Japan) within a 200–800 nm wavelength range. ΔE and film opacity were determined using Eqs. (3), (4).

$$\begin{array}{c}∆\mathrm{E}=\sqrt[]{{\left(∆L^{\ast }\right)}^2+{\left(∆a^{\ast }\right)}^2+{\left(∆b^{\ast }\right)}^2}\end{array}$$ (3)

$$\begin{array}{c}\text{Opacity}=\frac{{\mathrm{A}}_{600}}{\mathrm{T}}\end{array}$$ (4)

where A₆₀₀ and T are absorbance at 600 nm and film thickness (mm), respectively.

2.6.3. Water vapor permeability (WVP) and oxygen permeability (OP)

The WVP and OP of films were determined following the procedure reported by Ma et al. (2025) with slight modifications. Beaker openings were covered and sealed with composite films (50 mm × 50 mm). 3 g of anhydrous calcium chloride and 3 g of composite oxygen absorber (1:2:3 mass ratio of iron powder to sodium chloride to activated carbon) were placed in separate containers for WVP and OP tests. After weighing the beakers to obtain their initial mass, all samples were maintained at 25 °C with 75% relative humidity for 48 h. Subsequent reweighing of the beakers was performed to obtain mass changes. Calculations of WVP and OP were performed with the following formulas.

$$\begin{array}{c}\mathrm{WVP}=\frac{\mathrm{\Delta m}\times \mathrm{\delta }}{\mathrm{A}\times \mathrm{\Delta t}\times \mathrm{\Delta P}}\end{array}$$ (5)

$$\begin{array}{c}\mathrm{OP}=\frac{\mathrm{\Delta m}\times \mathrm{\delta }}{\mathrm{A}\times \mathrm{t}}\end{array}$$ (6)

where Δm is the weight change of the beaker (g), δ is the film thickness (m), A is the effective permeation region (m²), t is the equilibration time (d), and ΔP is the pressure discrepancy through the film (Pa).

2.6.4. Moisture content (MC), degree of swelling (DS) and water solubility (WS)

The MC, DS and WS of the films were measured according to the approach adopted by Li, Yang, et al. (2025). Film samples (2 cm × 2 cm) were cut and weighed (denoted as m₁), then dried in an oven at 105 °C for 24 h and reweighed (denoted as m₂). Subsequently, the films were submerged in 40 mL of deionized water at 25 °C for 24 h. Excess surface water was carefully removed with filter paper. The films were subsequently weighed (denoted as m₃) and dried again at 105 °C for another 24 h to obtain the final dry weight (denoted as m₄). The MC, DS, and WS were conducted using equations as follows.

$$\begin{array}{c}\mathrm{MC}=\frac{\left({\mathrm{m}}_1-{\mathrm{m}}_2\right)}{{\mathrm{m}}_1}\times 100\%\end{array}$$ (7)

$$\begin{array}{c}\mathrm{DS}=\frac{\left({\mathrm{m}}_3-{\mathrm{m}}_2\right)}{{\mathrm{m}}_2}\times 100\%\end{array}$$ (8)

$$\begin{array}{c}\mathrm{WS}=\frac{{\mathrm{m}}_2-{\mathrm{m}}_4}{{\mathrm{m}}_2}\times 100\%\end{array}$$ (9)

2.6.5. Water contact angle (WCA)

The WCA of composite films was measured using an interfacial tension analyzer (OSA OT100, Xinjie Scientific Instruments, China). A microsyringe was utilized to deliver deionized water (5 μL) on the film surface. WCA images were captured by the high-definition optical camera system at the initial contact moment (0 s) and after 60 s of equilibration.

2.6.6. Antioxidant activity

The antioxidant capacity of the prepared films was determined by DPPH and ABTS radical scavenging assays as reported by Li, Wei, et al. (2025). Briefly, the composite films (100 mg) were soaked in 25 mL of deionized water for 30 min. The resulting mixture was then centrifuged at 6000 rpm for 10 min to obtain a clear supernatant. Subsequently, the supernatant was individually blended with DPPH or ABTS working solution at 1:1 (v/v). The mixture underwent 30 min of light-protected incubation before centrifugation, and absorbance values were recorded at 517 nm (DPPH) and 734 nm (ABTS). The radical scavenging capacities were quantified by means of the equations below.

$$\begin{array}{c}\text{Radical scavenging activity}=\frac{{\mathrm{A}}_0-{\mathrm{A}}_1}{{\mathrm{A}}_0}\times 100\%\end{array}$$ (10)

where A₀ and A₁ are the absorbances of the control and film samples, respectively.

2.6.7. Release of curcumin from composite film

The release profiles of curcumin and resveratrol from the films in semi-fatty (50% ethanol, v/v) and fatty (95% ethanol, v/v) food simulants were analyzed following the approach proposed by Zhang, Chen, et al. (2022). The Film (20 mm × 20 mm) was immersed in 50 mL of each simulant in Erlenmeyer flasks, followed by gentle shaking at 100 r/min. In the experimental cycle of 72 h, 5 mL aliquots were taken from the food simulant at scheduled time points. The absorbance of each sample was measured at 426 nm for curcumin and 306 nm for resveratrol. The release rates of both compounds at different time intervals were calculated using the following equation.

$$\begin{array}{c}\text{Release rate}\left(\%\right)=\frac{{\mathrm{M}}_{\mathrm{t}}}{{\mathrm{M}}_0}\times 100\end{array}$$ (11)

where M_t is the content of active components in food simulants, and M₀ is the initial content of active components in composite films.

2.6.8. Ammonia, acetic acid and pH-sensitivity tests

Film samples were shaped into flower shapes with an embosser. The sensitivity to pH, ammonia and acetic acid was subsequently tested following the procedure of Zhang, Li, et al. (2025).

For pH sensitivity test, films were soaked in buffer solutions (pH 2–12) for 5 min. After immersion, residual surface moisture on the samples was wiped away, and the color variations of the samples were imaged with a camera.

For ammonia and acetic acid sensitivity tests, samples were placed 1 cm above Petri dishes containing 20% (v/v) aqueous ammonia and 40% (v/v) acetic acid solution, respectively, for 40 min. During exposure, color changes were captured by a camera at 10 min intervals.

2.7. Monitoring fish freshness fish fillets

2.7.1. Preparation of tilapia fillet with indicator films

Fresh tilapia (average weight: 1500 ± 150 g; average body length: 28 ± 3 cm) were purchased from a local supermarket in Hainan, China, and sourced from certified freshwater farms with valid quarantine certificates. All fish were humanely slaughtered by certified supermarket staff before purchase, in strict accordance with animal welfare principles via rapid cranial stunning followed by branchial artery exsanguination. Upon arrival at the laboratory, the fish were further pretreated by removing scales, skin and bones, and then cut into uniform fillets (∼2.5 × 2.5 × 2 cm³). Each tilapia fillet was placed at the bottom of a sterile Petri dish, and the indicator film was adhered to the inner surface of the Petri dish lid without direct contact with the fillet. The sealed Petri dishes were kept at 4 °C for 4 days. During this period, the color parameters of both the films and fillets were monitored, along with the pH, total volatile basic nitrogen (TVB-N), textural properties, total viable count (TVC) and thiobarbituric acid reactive substances (TBARS) of the tilapia fillets. These quality indicators were determined to investigate the performance of the indicator film for monitoring the freshness of tilapia fillets during cold storage. All procedures complied with national and institutional animal ethics regulations. As the experimental subjects were commercially available edible aquatic products limited to post-slaughter quality testing, this study was exempt from separate ethical approval.

2.7.2. Color

The color of the indicator films and fish fillets was measured with a colorimeter. Random measurements were conducted at five locations on each sample, with averages computed thereafter. Meanwhile, color variations of films and fish fillets were recorded photographically.

2.7.3. pH

The fillet (4 g) was subjected to homogenization with 80 mL of deionized water. Following homogenization, the mixture was kept static for 30 min. Subsequently, the pH of the resulting homogenate was determined using a digital pH meter.

2.7.4. TVB-n

The TVB-N of the tilapia fillets was quantified with a Kjeldahl nitrogen analyzer. The tilapia fillet samples (2 g) were cut into pieces and soaked in 15 mL of distilled water for 30 min. The mixture was transferred into a distillation tube containing 0.2 g of MgO, followed by distillation for 3 min. After the cessation of the distillation reaction, an indicator solution was introduced, and the ensuing distillate was subjected to titration with a 0.1009 mol/L HCl standard solution until the endpoint was attained.

2.7.5. Texture

A texture analyzer was used to analyze the hardness and springiness of fish fillets. With a P/36R probe and TPA compression mode, the test speed, trigger force and compression ratio were set to 3 mm/s, 5 g and 30%, respectively.

2.7.6. TVC

Fish meat (1 g) was homogenized with 9 mL of sterile physiological saline for 1 min and subjected to serial dilution. Then, 200 μL of the dilution was inoculated onto plate count agar and incubated at 37 °C for 48 h before colony enumeration.

2.7.7. TBARS

Fish meat(5 g) was homogenized with 50 mL of mixed solution containing 0.46 mol/L trichloroacetic acid (TCA) and 0.1% (w/v) ethylenediaminetetraacetic acid disodium salt. Subsequently, the homogenate was centrifuged for 10 min at 4 °C and 6000 r/min. An equal volume of the supernatant was mixed with 0.02 mol/L thiobarbituric acid solution, heated in a boiling water bath for 30 min, and then cooled to room temperature. Finally, the absorbance of the solution was measured at a wavelength of 532 nm.

2.8. Statistical analysis

Data were obtained in triplicate and presented as mean ± standard deviation. SPSS 19.0 was used for statistical analysis. To compare significant differences (p < 0.05), one-way ANOVA was used, followed by Duncan's multiple range test.

3. Results and discussion

3.1. Characterization of ZCRS NEs

The measured average particle size, polydispersity index (PDI) and ζ-potential values of ZCRS NEs were presented in Fig. 2C. The −44.1 mV ζ-potential was ascribed to anionic sodium alginate coating of zein nanoparticles and conferred a negative charge to the system. The electrostatic repulsion between particles effectively maintained the colloidal stability of the ZCRS NEs (Gao et al., 2022). The average diameter and PDI of the ZCRS NEs were 200.2 nm and 0.185, respectively, which were consistent with the findings of relevant studies on zein-based nanoparticles co-loaded with active substances (Yang et al., 2022). According to the microscopic data in Fig. 2A and B, the ZCRS NEs displayed a regular spherical morphology, and the measured particle size was in agreement with that observed using SEM and optical microscopy. The observed particle aggregation in some regions could be attributed to local bridging flocculation between zein nanoparticles and sodium alginate chains during the self-assembly process. The EE of Cur (78.43 ± 1.21%) and Res (71.2 ± 0.97%) (Fig. 2D) confirmed the effective loading of both polyphenols within the zein-SA matrix, driven by non-covalent interactions (Meng et al., 2021). The detailed preparation procedures for NEs and films are shown in Fig. 1.

Fig. 2.

Scanning electron micrographs (A), optical micrographs (B), particle size distribution (C), and encapsulation efficiency (D) of nanoemulsions.

Fig. 1.

The fabrication schematic of composite films. Abbreviations: ZC NPs, zein-curcumin nanoparticles; ZCR NPs, zein-curcumin-resveratrol nanoparticles; ZCRS NPs, zein-curcumin-resveratrol-sodium alginate nanoparticles; ZCRS NEs, zein-curcumin-resveratrol-sodium alginate nanoemulsions; CP modification: Cold plasma modification, NEs films: nanoemulsions films.

3.2. Microscopic characterization of NEs composite films

3.2.1. SEM

Distinct variations in surface morphology were found for different composite films (Fig. 3A). The chitosan film showed homogeneous and smooth surface morphology, which was in agreement with earlier reports (Xiao et al., 2023). The composite films without CP modification displayed subtle protrusions and microscale wrinkled rough structures. These phenomena were attributed to the precipitation and uneven distribution of emulsion particles on film surface during the curing process (Zhang et al., 2021). Such particle aggregation and irregular arrangement greatly disrupted the surface flatness and morphological uniformity of the composite films. Corresponding observations have also been documented in previous studies by Zhang, Cai, et al. (2025). Furthermore, CP modification further increased surface roughness due to plasma etching and charge-induced uneven deposition, which contributed to enhanced interfacial interactions (Kongboonkird et al., 2024b). Large quantities of free radicals and high-energy particles were generated under 50 kV plasma treatment, with surface protrusions preferentially bombarded by these energetic species and flattened via selective sputtering, which reduces surface roughness (Hölzel et al., 2023).

Fig. 3.

Scanning electron microscopy (A), X-ray diffraction (B), Fourier transform infrared spectroscopy (C), and formation mechanism and interaction diagram (D) of composite films.

3.2.2. XRD

The XRD patterns of films were displayed in Fig. 3B. The chitosan film displayed a typical broad diffraction peak around 2θ = 20°, corresponding to its semicrystalline structure. The XRD patterns of composite films prepared by blending NEs with chitosan matrix showed no new characteristic diffraction peaks, indicating that the two polyphenolic active substances achieved homogeneous amorphous dispersion throughout the polymer network. However, the characteristic peak of the unmodified composite film at 20°–25° became weaker. This phenomenon was caused by the combined effects of hydrogen bond competition and steric hindrance between chitosan and NEs (Li et al., 2024). Notably, after CP modification, the diffraction peaks of films were further broadened alongside a continuous decline in crystallinity. This was attributed to the generation of high energy reactive species on film surfaces during modification, which induced polymer chain rearrangement and expanded amorphous regions (Song et al., 2026). In summary, the XRD results indicate that NEs exhibit interfacial interaction with the chitosan matrix and CP modification effectively regulates the microstructure of the film.

3.2.3. FTIR

FTIR spectroscopy (Fig. 3C) characterized the functional group composition and intermolecular interactions of the films. After blending NEs with the chitosan matrix, the composite film exhibited no new characteristic peaks. The –OH/–NH₂ characteristic peak shifted to a higher wavenumber, and a further shift was observed following CP modification. This spectral variation confirmed intermolecular hydrogen bonding between NEs and the chitosan matrix (Zhang, He, et al., 2022). CP modification could strengthen such interactions, thereby enhancing the mechanical behaviors and stable structural features of the films (Ahmadian et al., 2025). Meanwhile, the significant enhancement in the –CH characteristic peak intensity indicated an increased content of aliphatic chains, which could be rationalized by the incorporation of NEs components. Variations in the peak position and strength of amide I and amide II bands within 1500–1700 cm⁻¹ were clearly observed. These spectral changes verified the occurrence of synergistic electrostatic attraction and hydrogen bonding interactions between chitosan matrix and NEs components (Wang, Xu, et al., 2025). Therefore, FTIR analysis demonstrated that both NEs incorporation and CP modification formed a stronger intermolecular network, which accounted for the enhancement in mechanical and barrier behaviors of the films.

3.3. Characterization of NEs composite films

3.3.1. Thickness and mechanical properties

The thickness of the films is presented in Table 1. The chitosan film had a thickness of 0.0876 ± 0.002 mm. All composite films were thicker, with values ranging from 0.1070 to 0.1096 mm. The increased thickness of composite films was primarily attributed to NEs incorporation. NEs not only elevated the solid content but also filled voids among molecular chains and enhanced network compactness (Zhao et al., 2023). Furthermore, the CP modification exhibited no remarkable influence on the film thickness.

Table 1.

Thickness, color and optical properties of all films.

Sample	Thickness(mm)	L*	a*	b*	ΔE	Opacity
Chitosan	0.0876 ± 0.002b	84.63 ± 0.27a	−0.62 ± 0.11d	2.72 ± 0.14c	12.77 ± 0.29d	1.31 ± 0.01d
C-NEs-0 kV	0.1096 ± 0.003a	80.86 ± 0.47b	4.35 ± 0.10c	39.92 ± 0.82b	42.76 ± 0.71c	5.90 ± 0.02c
C-NEs-10 kV	0.1070 ± 0.002a	79.54 ± 0.48c	4.41 ± 0.08bc	41.25 ± 1.11ab	44.90 ± 0.88b	6.01 ± 0.02b
C-NEs-30 kV	0.1082 ± 0.003a	78.19 ± 0.51d	4.51 ± 0.10b	41.70 ± 1.03a	45.86 ± 0.84ab	6.16 ± 0.06a
C-NEs-50 kV	0.1074 ± 0.002a	77.33 ± 0.72e	4.77 ± 0.08a	42.27 ± 0.79a	46.76 ± 0.90a	6.10 ± 0.03a

As shown in Fig. 4B, pure chitosan films exhibited inferior mechanical properties. This was because the molecular chains aggregated tightly via strong hydrogen bonding, forming a dense but brittle structure. The incorporation of NEs induced filling and crosslinking effects between emulsion and matrix, thus increasing the tensile strength (TS) and elongation at break (EAB) of the films (Yang et al., 2024). The above observations coincided with the findings documented by Cai et al. (2025), and similar trends in TS and EAB were observed when emulsions were incorporated into the chitosan matrix. The mechanical properties of the composite films were further enhanced through CP modification, which induced surface cross-linking and structural rearrangement (Moosavi et al., 2020). Nevertheless, excessive voltage resulted in molecular chain scission and compromised crosslinking capacity, with high-energy particle damage further exacerbating the overall property degradation of the material (Li et al., 2022).

Fig. 4.

Transmittance spectra (A), mechanical properties (B), water vapor permeability and oxygen permeability (C), and antioxidant properties (D) of composite films. Photographs of mechanical flexibility of films under rolling, twisting, folding and 500 g loading (E).

3.3.2. Color and optical properties

Color parameters of the films are presented in Table 1. After NEs were incorporated into the chitosan matrix, film lightness (L*) decreased. Conversely, a* and b* values increased markedly compared with pure chitosan films. This phenomenon was mainly ascribed to the inherent yellowness of nanoparticles within NEs, which rendered composite films darker and more yellowish in appearance. Additionally, the brightness of the films was further reduced after CP modification.

Fig. 4A presents the UV-light transmittance of different films. Chitosan film exhibits high transmittance across the full 200–800 nm range, covering both ultraviolet and visible light regions. Incorporation of NEs leads to a globally decreased UV–Vis transmittance of the film. This effect was primarily ascribed to the light scattering and reflection triggered by the nanomaterials and the ultraviolet attenuation from aromatic phenolic compounds within the NEs (Tan et al., 2025). Subsequent CP modification further reduces the transmittance of composite films. This outcome is jointly induced by energetic plasma particles eroding the film surface to increase roughness and promoting cross-linking reactions, forming a denser film structure (Chen, Chen, et al., 2020). The above results corresponded well to the characterization data from SEM and FTIR.

3.3.3. Barrier properties

The quality of packaged food was determined by the water vapor and oxygen barrier capacities of the films. The chitosan films showed a WVP of 7.21 ± 0.17 × 10⁻⁶ g·m⁻¹·d⁻¹·Pa⁻¹ and an OP of 2.50 ± 0.10 g·m⁻¹·d⁻¹·atm⁻¹ (Fig. 4C). After mixing with NEs, the barrier performance of the composite films improved significantly. These findings could be rationalized by strong interfacial bonding between NEs and the chitosan matrix as well as the increased film density, both of which prolonged the diffusion pathways of water and gas molecules (Lin et al., 2023). The composite films treated with CP modification showed lower WVP and OP values than both chitosan films and unmodified composite films. Under 30 kV CP modification, the composite films achieved optimal barrier properties (WVP of 4.87 ± 0.11 × 10⁻⁶ g·m⁻¹·d⁻¹·Pa⁻¹, OP of 1.37 ± 0.07 g·m⁻¹·d⁻¹·atm⁻¹). As reported by Wan et al. (2025), CP modification introduced polar functional groups onto the film surface and promoted cross-linking reactions. This strengthened interfacial bonding and raised the diffusion resistance of permeants, thereby improving film barrier performance. However, more high-energy particles were generated under modification at 50 kV, inducing excessive surface etching and micropore formation that facilitate the penetration of water vapor and gases and deteriorate barrier properties (Hsu et al., 2018).

3.3.4. Water resistance ability

The MC, WS, and DS values are presented in Fig. 5A–C. Incorporation of NEs significantly elevated the water resistance of chitosan films. This improvement was explained by interfacial hydrophobic effects and hydrogen bonding between NEs and chitosan molecules, leading to decreased binding of hydrophilic groups to water molecules (Tian et al., 2024). CP modification contributed to a further reduction in the MC, WS and DS of the NEs film. Specifically, the MC, WS and DS decreased progressively with increasing voltage up to 30 kV. However, a slight recovery of these indices was observed when the voltage was elevated to 50 kV. This phenomenon originated from the dual synergistic mechanism of CP modification (Yang et al., 2026). At low voltages, CP modification generates reactive oxygen and nitrogen species that deplete hydrophilic hydroxyl and amino groups on the film surface. Mild etching creates a rough surface morphology, which facilitates the accumulation of curcumin and resveratrol at the film surface and compacts the film matrix. Such synergistic effects hinder water permeation and enhance the water barrier property of composite films. Conversely, 50 kV plasma exposure releases abundant energetic particles that trigger over-etching and produce massive micropores and cracks. Extensive cleavage of polymer backbones loosens the superficial layer, which enables easier water ingress and deteriorates the waterproof performance.

Fig. 5.

Moisture content (A), degree of swelling (B), water solubility (C), and water contact angle (D) of composite films.

3.3.5. WCA

The WCA of pure chitosan films reached 81.3° (Fig. 5D). However, chitosan films incorporated with NEs exhibited a decreased WCA, which was ascribed to the abundant hydrophilic groups (—OH, —COONa) contained in NEs. CP modification enabled the recovery of water contact angle by strengthening intermolecular interactions and cohesion of the film, constructing a dense network, and encapsulating partial hydrophilic groups. Observations indicated that the WCA of NEs composite films decreased at a slower rate than that of chitosan films. The above observation was primarily due to the enhanced cohesion and compact network structure of composite films, which slowed the penetration and spreading of water molecules (Zhang, Guo, et al., 2025). After CP modification, the WCA of composite films exhibited an initial increase followed by a reduction, peaking at 30 kV (78.7° at 0 s; 75.5° at 60 s). Meanwhile, the hydrophobicity of the films was greatly improved. This was because SEM observations revealed that CP modification roughened the film surface, thus altering the water resistance of the films. The WCA presented an upward trend as surface roughness increased (Roy & Rhim, 2021). These results showed that the incorporation of NEs and different modification methods could significantly affect the hydrophilicity and hydrophobicity of films by regulating their structure and surface properties. This was consistent with the variation trend in the barrier properties of composite films described above, namely that structural densification and enhanced surface hydrophobicity jointly endowed the films with superior barrier capacity.

3.3.6. Antioxidant activity

As illustrated in Fig. 4D, the chitosan films displayed low inherent antioxidant capacity, with DPPH and ABTS scavenging rates reaching 35.73 ± 1.78% and 34.66 ± 2.44%, respectively. The incorporation of NEs effectively improved the antioxidant capacity of composite films relative to chitosan films. This enhancement was mainly attributed to bioactive components including curcumin within NEs, which efficiently provided hydrogen protons or electrons for radical quenching, thereby producing antioxidant activity (Liu et al., 2024). Plasma-induced etching arising from CP modification contributed to a remarkable increase in film surface roughness (Mohammadi et al., 2025). This provided more reactive sites for the contact between active substances in the films and free radicals. Higher plasma voltages produced abundant reactive species, which triggered oxidation and molecular chain scission within the chitosan matrix. This process accelerated the exposure and release of phenolic compounds derived from curcumin and resveratrol, consequently improving antioxidant capacity.

3.3.7. Food simulant release behavior

As shown in Figs. 6A–D, Cur and Res exhibited typical biphasic release kinetics in 50% and 95% ethanol food simulants. This release pattern was tightly linked to the physicochemical properties of active compounds, solvent characteristics and composite film matrix structure. The release process was divided into two stages. The initial rapid release stage occurred within 0–24 h, during which the cumulative release rate increased sharply. The subsequent sustained slow release stage lasted from 24 to 72 h, in which the release rate decelerated and the cumulative release rate gradually leveled off. This biphasic release behavior resulted from the synergistic effect of two key factors. First, ethanol enhanced the solubility of Cur and Res, two hydrophobic compounds, reduced their molecular aggregation tendency and facilitated the rapid dissolution of active substances (Wang et al., 2024). Secondly, ethanol molecules penetrated the film matrix, disrupting the intermolecular interactions within the chitosan‑sodium alginate composite film. Such behavior induced polymer network swelling and expanded the internal free volume, thereby providing channels for the diffusion of active substances (Liu et al., 2017). CP modification regulated the release behavior in a voltage-dependent manner. The 10 kV low-voltage modification accelerated the release rate by means of surface etching and pore formation, which shortened the diffusion path of active substances. In contrast, the 30–50 kV medium-high voltage treatment enhanced polymer cross-linking and increased the surface density of polar functional groups, thereby strengthening hydrogen bonding interactions between active compounds and the matrix (Chen et al., 2018). These effects restrained the dissolution and diffusion of active substances. Res presented a slightly higher cumulative release than Cur. Cur was more sensitive to the changes in ethanol polarity. These differences reflected the variations in molecular hydrophobicity and binding affinity to the matrix.

Fig. 6.

Release profiles of curcumin from composite films in 50% ethanol (A) and 95% ethanol (B) food simulants. Release profiles of resveratrol from composite films in 50% ethanol (C) and 95% ethanol (D) food simulants.

3.3.8. Color indicator response

As shown in Fig. 7A, composite films showed clear color response characteristics after being submerged in buffer solutions of varying pH and exposed to volatile ammonia and acetic acid gases. Specifically, pristine composite films without CP modification displayed bright golden yellow in acidic buffer solutions. As the pH value gradually increased during storage, the brightness of the yellow color displayed by the intelligent films underwent a slight downward trend. In contrast, the films rapidly turned deep orange-yellow in alkaline buffer solutions. The color darkened with increasing pH. At pH 10, the films exhibited a deep orange-brown color, with the color change clearly distinguishable by the naked eye. A similar color change trend was also observed over time after exposure to ammonia/acetic acid gases. Moreover, the color response of the films was more pronounced after CP modification. This observation further corroborated the enhancement effect of CP modification on the pH-responsive performance of the films. The underlying mechanism was attributed to the etching effect induced by CP modification, which increased film porosity and facilitated the permeation of buffer media and gaseous molecules (Zhang et al., 2023). Notably, the 30 kV plasma modification promoted moderate chain scission and surface oxidation of chitosan, improving ion diffusion without compromising the integrity of the polymer network. However, excessive plasma energy at 50 kV caused severe degradation of chitosan chains and disruption of the ordered network structure (Sun et al., 2020). Although the increased structural looseness facilitated curcumin release, it reduced the structural responsiveness required for effective pH-induced color variation, resulting in a slower response and lower colorimetric sensitivity. The fundamental mechanism underlying the aforementioned color response lies in the pH-sensitive isomerization of the β-diketone moiety in curcumin (Shi et al., 2024). Specifically, curcumin molecules predominantly adopt the ketone form of the β-diketone structure under acidic or neutral pH conditions, thus exhibiting a bright yellow color. In contrast, it converts to the enol form in alkaline conditions. The color change is reversible upon pH reversal, confirming that the indicator mechanism relies on reversible protonation/deprotonation of curcumin rather than irreversible chemical degradation. These results confirm that the films possess pH-sensitive indicator properties with practical applicability for real-time food freshness monitoring.

Fig. 7.

Color variations at different pH levels of composite films (A), acetic acid vapor and ammonia vapor sensitivity of composite films (B).

3.4. Application of films to monitor tilapia freshness

To evaluate the practical applicability of the developed indicator films for freshness monitoring, the correlation between film colorimetric response and tilapia quality parameters was systematically analyzed. Fig. 8A presented representative photographs of the indicator films attached to tilapia fillet packages during storage at 4 °C for 4 days. This study determined the TVB-N, pH, texture, color, TVC and TBARS parameters of tilapia fillets during different storage periods. It also monitored the color response of composite films to verify their feasibility and accuracy for freshness detection. Microbial proliferation and metabolic activities were the primary causes of spoilage in tilapia fillets during storage. As presented in Fig. 8F, the pH of tilapia fillets exhibited an initial decrease followed by an increase. This phenomenon was closely linked to the lactic acid and other acidic metabolites produced by muscle glycogen glycolysis in the early storage stage, whereas microbial decomposition of muscle proteins generated alkaline substances such as ammonia and amines in the late storage stage, triggering a pH increase (Wu et al., 2023). Fresh tilapia fillets showed an initial TVB-N reaching 5.61 ± 0.90 mg/100 g. The value met the requirement for first-grade freshness (≤15 mg/100 g) specified in the aquatic product freshness evaluation standards of China, indicating good freshness of the raw material. On the 4th day of storage, the TVB-N in fillets rose sharply to 23.34 ± 1.36 mg/100 g. This concentration was higher than the regulatory spoilage limit (20 mg/100 g), signifying full spoilage and inedibility of the fillets. As shown in Fig. 8G, the hardness and springiness of the fillets exhibited a continuous downward trend with prolonged storage time. Specifically, the hardness decreased from an initial value of 1503.96 g to 774.15 g, while the springiness decreased from 0.83 mm to 0.64 mm. This result mainly arose from the synergistic breakdown of muscle fibrillar proteins and connective tissues induced by microorganisms and endogenous proteases (Yu et al., 2022). The destruction of fillet structural integrity further resulted in texture softening and decreased springiness. Color parameter changes on the fillet surface were equally significant. The a*, b* and ΔE values showed a gradual increasing trend throughout storage, while the L* value decreased continuously. These changes indicated that the fillets gradually exhibited visual deterioration characteristics such as yellowing and darkening, consistent with the change trends of TVB-N and texture parameters. The dynamic changes in TVC and TBARS were the core intrinsic cause for the overall deterioration of pH, TVB-N, texture and color in tilapia fillets. As shown in Fig. 8H, both TVC and TBARS rose continuously during refrigerated storage. The increase in TVC reflected microbial proliferation. Enzymes secreted by microorganisms decomposed muscle proteins and amino acids and generated volatile basic nitrogen, which triggered changes in TVB-N content and pH value. In addition, the synergistic degradation of proteins impaired tissue integrity, leading to a decline in the hardness and springiness of fillets. Meanwhile, the increase in TBARS indicated sustained lipid oxidation throughout storage. Secondary oxidation products such as malondialdehyde not only gave rise to rancid off-flavors, but also formed covalent cross-links with muscle proteins, further exacerbating protein denaturation and textural deterioration. Moreover, the accumulation of oxidative products and the release of pigments from degraded proteins were highly correlated with the color deterioration of fish fillets. Overall alterations in microbial proliferation, lipid oxidation and protein degradation worked in concert to induce spoilage deterioration in refrigerated tilapia fillets. These spoilage-derived volatile substances were trapped by the indicator films, which allowed real-time tilapia freshness to be visualized through color variations of the films.

Fig. 8.

Freshness indication effect during tilapia preservation (A), colorimetric response mechanism of composite films (B), variations in a* value (C) and ΔE (D) of composite films, color changes (E), total volatile basic nitrogen and pH (F), hardness and springiness (G), thiobarbituric acid reactive substances and total viable count (H) of tilapia fillets.

Freshness monitoring of tilapia fillets using composite films revealed that the ΔE and a* values of all composite films showed almost no obvious changes during the first two days of storage, with this observation consistent with the quality characteristics of fresh fillets. Significant increases in ΔE and a* were observed for all film groups over storage. These changes were accompanied by the gradual accumulation of volatile nitrogenous compounds generated by fish fillet metabolism, which induced a slow color shift of the indicator films from yellow to reddish yellow. The pH color response of composite films is governed by curcumin, and synergistic effects are achieved via resveratrol. Deprotonation of the β-diketone conjugated skeleton and phenolic hydroxyl groups of curcumin takes place under alkaline conditions, and the resulting extended conjugated system causes color transformation. Hydrogen bonds formed between resveratrol and curcumin stabilize dissociated phenoxide anions and increase the stability of color development. Such obvious color evolution provided an intuitive visual indicator for monitoring the real-time spoilage process of fish fillets. Among all treatment groups, the composite films modified by plasma at 30 kV exhibited the most significant color change (Fig. 8C and D). This was because plasma modification at 30 kV optimized the surface pore structure and active group distribution of the films, thereby enhancing ammonia sensitivity (Chizoba Ekezie et al., 2017; Zhang et al., 2023). Conversely, plasma modification at 50 kV damaged the microstructural integrity of the films and reduced the availability of active sites, thus weakening the color response performance (Nasiru et al., 2025). These results collectively demonstrated that the CP-modified indicator films provided a reliable, quantitative, and visually accessible method for monitoring tilapia freshness during refrigerated storage.

4. Conclusions

This study incorporated ZCRS NEs into chitosan-based films to fabricate novel pH-responsive active composite films via CP modification (0 kV, 10 kV, 30 kV, 50 kV). Characterization results confirmed that CP modification effectively modulated the surface structure and chemical characteristics of films and introduced polar functional groups to increase surface active sites, consequently improving film functional properties. Notably, the film treated with 30 kV CP achieved the optimal comprehensive performance: it not only showed significantly optimized colorimetric response compared with the unmodified one, but also possessed greatly improved mechanical strength, barrier properties and ultraviolet protection capability, while retaining excellent antioxidant activities. A strong linear correlation was observed between the film color change and tilapia spoilage degree, supporting reliable visual freshness monitoring. The color response mechanism was systematically elucidated based on the reversible keto-enol tautomerism of curcumin, with the color transition occurring specifically at the pH threshold corresponding to fish spoilage. The color parameters changed progressively with the accumulation of TVB-N. The strong correlation between the colorimetric response and quality indicators confirmed the accuracy and reliability of the developed films for freshness monitoring. Overall, this work adopted zein NEs to co-encapsulate curcumin and resveratrol, providing an effective pathway to integrate dual bioactive components into pH-sensitive smart packaging films. As a green physical modification strategy, CP modification further modulated the release behavior of these embedded bioactive substances. The distinct pH-triggered color change in response to volatile amines from spoiled tilapia makes the developed films well-suited for visual freshness assessment of aquatic products. This study offers a green and feasible strategy for developing bio-based intelligent packaging materials and provides technical support for the practical application of freshness-indicating films in the aquatic food industry.

CRediT authorship contribution statement

Meng Qiu: Writing – original draft, Validation, Methodology, Formal analysis, Data curation. Tingting Yang: Validation, Methodology, Conceptualization. Guanghua Xia: Supervision, Conceptualization. Jiamei Wang: Supervision, Project administration, Funding acquisition. Zhongyuan Liu: Supervision, Formal analysis. E Liao: Resources, Data curation. Wei Xu: Resources, Conceptualization. Liming Zhang: Writing – review & editing, Supervision, Funding acquisition. Wenmin Wu: Writing – review & editing, Supervision.

Ethics declaration

This study was conducted in accordance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines. Ethics committee approval was not required. This article does not contain any studies involving human participants or animals performed by any of the authors.

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.

Data availability

Data will be made available on request.