Dynamic pH-Responsive Labeling System Based on Polyvinyl Alcohol/Arabinoxylan Nanofibers Incorporating Purple Cabbage Anthocyanins for Real-Time Food Freshness Monitoring

TL;DR

A smart label made from nanofibers and natural pigments changes color to show food freshness in real time, helping reduce waste and improve safety.

Contribution

A biodegradable, pH-responsive nanofiber label using purple cabbage anthocyanins for real-time food freshness monitoring is developed.

Findings

The nanofiber label shows distinct color changes across a wide pH range (2 to 12).

The label demonstrated significant color changes when applied to sausages and shrimp, indicating freshness.

The nanofiber label is biodegradable and environmentally friendly.

Abstract

The fabrication of a real-time intelligent indication label for food freshness has emerged as an effective strategy to reduce food waste and improve food safety. In this study, utilizing polyvinyl alcohol (PVA) and arabinoxylan (AX) as the polymer matrices, and incorporating purple cabbage anthocyanins (PCAs) as natural pH-responsive agents, we fabricated a PVA/AX/PCA nanofiber-based intelligent indication label via electrospinning. The results confirmed that the nanofibers exhibited uniform morphology and good structural stability, with the PCA successfully embedded within the nanofibers. The nanofiber membrane exhibits a low water contact angle (54°) and demonstrates a tensile strength of 5.34 ± 0.09 MPa with an elongation at break of 32.43 ± 1.02%, while maintaining a certain degree of flexibility. The nanofiber labels exhibited distinct color changes within a wide pH range (2 to 12), which confirms their pH-responsive characteristics. After being stored at 4 °C and 25 °C for 14 days, the maximum color difference related to storage stability was 1.53 ± 0.02. In practical applications at 25 °C, this intelligent label demonstrated significant color changes when monitoring low-temperature-cooked sausages and fresh shrimp, with total color differences of 41.57 and 53.06, respectively. Degradation experiments showed that the nanofiber labels gradually decomposed, reflecting good biodegradability and environmental-protection characteristics. In conclusion, the green intelligent indication label developed in this study offers a feasible solution for real-time monitoring of food quality.

Figures

(a) pH vs. TVB-N changes and (b) ΔE vs. TVB-N changes in low-temperature-cooked sausages; (c) correlation analysis of pH and ΔE with TVB-N in low-temperature-cooked sausages; (d) pH vs. TVB-N changes and (e) ΔE vs. TVB-N changes in fresh shrimp; (f) correlation analysis of pH and ΔE with TVB-N in fresh shrimp.

Visual monitoring photographs of freshness changes in low-temperature-cooked sausages and fresh shrimp during storage at 25 °C using APNFs and APPNFs-1.2%.

Sensory radar charts illustrating freshness changes of (a) low-temperature-cooked sausages and (b) fresh shrimp during storage.

Full text

1. Introduction

Food is prone to deterioration during storage and transportation, causing economic losses and potential health risks, thereby making dynamic food freshness monitoring a key means to ensure consumer safety and reduce waste [1,2]. However, for perishable animal products, the ‘shelf life’ printed on traditional labels only provides a fixed estimation based on preset storage conditions and cannot reflect the actual quality changes in the food during transportation and sales processes. This may lead to consumers mistakenly consuming spoiled food, causing food safety incidents, while food may be discarded at a stage when it is still safe to eat, resulting in a serious waste of resources [3]. Thus, real-time monitoring and feedback in intelligent packaging have become an important research direction in food science and technology [4].

An intelligent indication label (IIL) is an important type of smart food packaging. It can detect changes during the real-time assessment of food, conveying food freshness to consumers and helping them make more informed choices [5]. The pH-responsive color indicator labels demonstrate significant advantages in freshness detection due to their simplicity, intuitive color changes, and rapid response [6]. When food spoils, microorganisms and enzymes release alkaline gases like ammonia that raise the pH, while lactic acid from bacterial metabolism can lower it [7,8,9]. Incorporating pH-sensitive dyes into intelligent labels enables visual indication of spoilage levels through visible color shifts, providing reliable guidance for consumers. Traditionally, chemically synthesized pH-sensitive dyes (e.g., bromophenol blue, methyl red, and phenolphthalein) have been used in polymer matrices, but they pose a toxicity risk, potentially endangering human health and polluting the environment [10,11,12]. Natural pigments are safe and non-toxic and have good environmental protection and sustainability [13]. Among various plant-derived anthocyanins (e.g., purple sweet potato, black rice, and blueberry), purple cabbage anthocyanins (PCAs) exhibit a higher degree of acylation and improved color stability, which are advantageous for constructing reliable pH-responsive freshness indicators [14]. PCAs are plant-derived natural pH indicators with pronounced color variation over a wide pH range [15]. Due to the highly acylated structural characteristics of PCA, it is an excellent choice for preparing freshness-monitoring labels for protein-based foods [16]. The anthocyanin molecules contained in it will show obvious color changes under different pH environments, which can directly reflect food freshness through distinct color changes [17]. Compared with chemical synthetic dyes, PCA has the advantages of being natural, safe and non-toxic, which is very suitable for the current consumer demand for green and healthy food packaging [18]. However, anthocyanins remain susceptible to environmental factors, which limits their wide application [19]. Under the influence of light, temperature and pH changes, such compounds degrade, resulting in color fading and weakened functionality, which affects the stability and reliability of the label. Therefore, how to develop effective anthocyanin encapsulation technology remains a major obstacle that calls for an urgent solution in the field of IIL.

There are multiple methods for producing fiber materials, including extrusion, blow molding, casting, phase separation, self-assembly, and electrospinning [20]. Among various fiber material preparation techniques, electrospinning stands as a versatile and relatively simple nanofiber manufacturing technology [21]. It enables the fabrication of uniform nanofiber membranes with outstanding mechanical properties, high specific surface areas, and high porosities [22,23]. These membranes can effectively encapsulate sensitive compounds, forming fast-response materials [24]. Furthermore, electrospun nanofibers exhibit flexibility, ease of surface modification, and unique physicochemical properties due to their nanoscale dimensions [25]. For instance, research by Luo et al. utilized electrospinning technology to combine polyvinyl alcohol (PVA), chitosan, and purple sweet potato anthocyanins to create a composite label; this label took advantage of the property that anthocyanins change color with pH value to determine in real time whether beef is fresh, providing consumers with an intuitive judgment standard [26]. PVA is safe and non-toxic, and has good compatibility with living organisms [27]. PVA films have good transparency and high gas barrier properties, making them an ideal material for intelligent labels [28]. Polysaccharide-based films have garnered significant attention due to their excellent gas barrier properties, mechanical stability, and optical transparency [29]. Unlike chitosan, which dissolves primarily in acidic aqueous solutions, arabinoxylan (AX) is a plant-derived polysaccharide that dissolves under relatively mild conditions and offers good solution processability [30,31]. The abundant hydroxyl and polysaccharide chains in AX can form extensive hydrogen bond interactions with the polymer matrix and phenolic compounds, including natural colorants such as anthocyanins, thereby enhancing the structural stability of the composite system and improving the retention ability of the colorants [32,33,34]. Furthermore, AX possesses excellent film-forming ability and food-grade safety, which are extremely desirable characteristics for the preparation of electrospinning processes and the application of smart labels. Although AX has excellent biological characteristics and film-forming potential, its application in pH-responsive smart tags is still limited. In particular, systematic studies on the construction of AX-based nanofiber labels via electrospinning technology have not been thoroughly reported. Compared with previous reports on anthocyanin-based indicator membranes (prepared using chitosan, gelatin or starch matrices), this study is the first to introduce AX as a functional polysaccharide component into the electrospun nanofiber system. The introduction of AX not only enhanced the structural integrity of the nanofiber membranes, but also improved the stability and dispersion of anthocyanins through hydrogen bond interactions. Moreover, compared with traditional casting membranes, the electrospinning strategy can provide a higher specific surface area and a faster response speed, which offers significant advantages for real-time freshness-monitoring applications.

In this study, a green and environmentally friendly pH-responsive electrospun nanofiber label was developed for the visual monitoring of food freshness. The nanofiber label utilized arabinoxylan (AX) as the film-forming matrix, with polyvinyl alcohol (PVA) added to enhance its spinnability, and purple cabbage anthocyanins (PCA) incorporated as natural pH indicators to endow the label with sensitive color-responsive properties. During food spoilage, changes in environmental pH trigger a distinct color change in PCA. Concurrently, the addition of AX not only improved the spinning efficiency of PVA and the stability of the resulting nanofiber labels, but also enhanced the dispersion stability and durability of anthocyanins to some extent. By adjusting the PCA addition amount, this study systematically investigated the morphological structure, ultraviolet-visible (UV-vis) spectroscopy, X-ray diffraction (XRD), mechanical properties, thermogravimetric analysis (TGA), and pH-responsive properties of the nanofiber label. In addition, the nanofiber label was used to conduct spoilage simulation experiments on low-temperature-cooked sausages and fresh shrimp. Through monitoring the chromatic shift of the label, the feasibility and application potential of it in preservation monitoring were verified. Compared with pH-indicating labels prepared via conventional casting methods, the electrospun nanofiber label exhibits a higher specific surface area, faster response speed, and superior interfacial properties. As far as we are aware, this study represents an early attempt at the synergistic incorporation of AX and PCA into an electrospinning system for constructing nanofiber labels for food freshness indication.

2. Materials and Methods

2.1. Materials

The arabinoxylan (AX) (purity ≥ 98%, M_W_ 200–300 kDa) was obtained from Xi’an Tiancan Biotechnology Co., Ltd. (Xi’an, China) and was derived from wheat bran. The polyvinyl alcohol (PVA) (M_W_ 195,000 g/mol, hydrolysis 98%) was purchased from Macklin Biochemical Co., Ltd. (Shanghai, China). Purple cabbage anthocyanins (PCA) with a total anthocyanin content of 25% were purchased from Xi’an Tianhe Pharmaceutical Co., Ltd. (Xi’an, China). The remaining components mainly consist of water-soluble plant-derived compounds such as sugars, organic acids, and amino acids, which are commonly present in plant anthocyanin extracts. The low-temperature-cooked sausages were provided by Linyi Xincheng Jinluo Meat Products Group Co., Ltd. (Linyi, China), while the fresh shrimp were purchased from a local market in Changqing District, Jinan City, Shandong Province, China. All other reagents were of analytical grade and used without further purification. Demineralized water was produced with a Milli-Q system from Millipore (Bedford, MA, USA).

2.2. Formulation of Electrospinning Solutions

To formulate a 14% (w/v) electrospinning solution, a mixture of 0.7 g PVA and 10 mL deionized water was prepared. Complete dissolution of the PVA was achieved by magnetic stirring at 90 °C for 2 h. After that, 0.7 g of AX powder was added to the system, the mixture was stirred at 70 °C for 1 h, and finally cooled to ambient temperature for subsequent use. A concentration gradient of PCA was established at 0.3%, 0.6%, 0.9%, and 1.2% (w/v) relative to the AX/PVA mixed solution. After adding PCA powder corresponding to each concentration, the resulting mixture was agitated at 40 °C for 3 h. The obtained solution was treated under ultrasound for 10 min to remove the bubbles present in the solution. Finally, the PCA-containing solutions (with PCA concentrations of 0.3%, 0.6%, 0.9%, and 1.2%) were subjected to electrospinning to prepare the nanofibers. The resulting samples were denoted as APNFs (control, without PCA), APPNFs-0.3%, APPNFs-0.6%, APPNFs-0.9%, and APPNFs-1.2%, respectively.

2.3. Electrospinning Process

Electrospinning was performed using a NANON-03 electrospinning unit (MECC Co., Ltd., Fukuoka, Japan) which is equipped with a high-voltage power supply and a programmable injection pump. The electrospinning experiment was conducted under optimized conditions, with an applied voltage of 26 kV and a temperature of 30 °C. Through preliminary tests, the electrospinning parameters were optimized, including the applied voltage, flow rate, and the distance between the needle tip and the collector, to achieve stable jet flow formation and continuous fiber production, while avoiding the formation of beads. The flow rate of the electrospinning solution syringe was set at 0.3 mL/h, and the fibers were collected on a rotating drum 15 cm away from the needle tip. The drum speed was maintained at 200 rpm and the humidity was controlled at 40%. After electrospinning, the nanofibers were taken off from the collector and stored in a desiccator for further characterization. For the control experiment, the APNFs samples were prepared under identical conditions.

2.4. Electrical Conductivity and Rheological Properties

The electrical conductivity was measured using a DDS-307A conductivity meter (Shanghai INESA Scientific Instrument Co., Ltd., Shanghai, China), while the rheological properties were determined using an MCR 302 rheometer (Anton Paar GmbH, Graz, Austria) at shear rates ranging from 0.1 to 100 s^−1^.

2.5. Ultraviolet-Visible (UV-Vis) Absorption Spectroscopy Analysis

The UV-Vis absorption spectra of the PCA solutions (pH 2.0–12.0) in the scan range of 400–800 nm were recorded using the TU-1900 spectrophotometer (PureGinji Corporation, Beijing, China). The PCA solutions (pH 2–12) were prepared by pH adjustment using 0.1 mol/L HCl and 0.1 mol/L NaOH, and measured with an S210B pH meter (Mettler Toledo, Schwetzingen, Switzerland). A digital camera (Canon Inc., Tokyo, Japan) was used to capture the color changes in the solutions at different pH values.

2.6. Morphological Test

A Gemini SEM 500 scanning electron microscope (Carl Zeiss Microscopy GmbH, Jena, Germany) was employed to analyze the microstructure of the nanofibers with different PCA concentrations. The samples were fixed with a conductive binder and gold-sputtered for 60 s at 15 mA under vacuum. The SEM images were obtained at an accelerating voltage of 5 kV with the micrographs acquired at 5000× and 10,000×. Using ImageJ 2x software (version 2.1.5.0) (National Institutes of Health, Bethesda, MD, USA), the average diameter (AD) of the nanofibers was measured, with 50 fibers randomly chosen per sample.

2.7. FT-IR Spectral Analysis

The FT-IR spectra were obtained on an FT-IR spectrometer (Nicolet iS10; Thermo Fisher Scientific, Waltham, MA, USA) to investigate the intermolecular interactions among AX, PVA, PCA, and the prepared nanofibers (APNFs, APPNFs-0.3%, APPNFs-0.6%, APPNFs-0.9%, APPNFs-1.2%). Measurements were performed over the range of 4000–600 cm^−1^, with a spectral resolution of 4 cm^−1^, and 64 scans were conducted.

2.8. X-Ray Diffraction (XRD) Analysis

The XRD patterns were scanned using an Ultima IV X-ray diffractometer (Kuraray Co., Ltd., Tokyo, Japan) equipped with Cu-Kα radiation (λ = 1.5406 Å). Samples were recorded at room temperature, operating at 40 kV and 40 mA, with a 2θ range of 4–40°, a scan rate of 30°/min, and a step size of 0.02°. The calculation method of the crystallinity index (CrI%) was as shown in (1):

$$[eqn]$$

where A_y_ represents the area of the crystalline region, and A_q_ represents the total area of the crystalline and amorphous regions.

2.9. Thermal Stability Test

Thermogravimetric analysis (TGA) was performed using a STA 6000 thermogravimetric analyzer (PerkinElmer Inc., Shelton, CT, USA) to analyze the thermal stability of AX, PVA, PCA, and the nanofibers (APNFs, APPNFs-0.3%, APPNFs-0.6%, APPNFs-0.9%, APPNFs-1.2%). Thermal analysis was performed under a nitrogen atmosphere with a flow rate of 50 mL/min. Approximately 5 mg of the sample was heated from 30 °C to 600 °C at a heating rate of 20 K/min.

2.10. Water Contact Angle (WCA) Measurement

The surface wettability of different nanofiber samples was characterized by the DSA25 droplet shape analyzer (KRÜSS GmbH, Hamburg, Germany) and their WCA was measured.

2.11. pH Sensitivity Test

The nanofibers were cut into 10 mm × 20 mm rectangular pieces and soaked in a buffer solution with pH values between 2 and 12. Since APPNFs-1.2% undergoes color change within 5 s, a digital camera was used to capture the color transition in real time.

2.12. Reversible Response of Nanofibers to Ammonia and Their Sensitivity

For reversibility testing, the APPNFs-1.2% nanofiber membranes were fixed at the opening of a conical flask filled with 100 mL of ammonia solution (5%, v/v), sealed with a transparent glass cover, and allowed to react until the color change reached equilibrium. The nanofiber membranes were then immediately transferred to a 10% (v/v) acetic acid solution to evaluate the reversibility of the color response. By deliberately using a higher concentration of ammonia and acid, the interaction between the gas and the indicator can be accelerated, thereby clearly demonstrating the pH response mechanism and reversibility of the nanofiber system based on anthocyanins. This alternating exposure cycle (ammonia/acetic acid) was repeated to verify response stability. Color analysis was performed using Adobe Photoshop CS5 software (Version 12.0), and the total color difference (ΔE) was calculated using Equation (2):

$$[eqn]$$

where L*, a*, and b* represent the chromatic parameters of the membranes under various pH conditions, and L, a, and b represent those of the unmodified membranes. A whiteboard was used as the color reference standard (L = 79.48, a = 6.67, b = −2.11) [35]. All images were captured under uniform illumination.

Ammonia gas sensitivity was assessed following the method of Parya Ezati and Jong-Whan Rhim [36]. Briefly, the APPNFs-1.2% nanofiber membranes were placed within the headspace of a conical bottle containing 80 mL of 0.8 mol/L ammonia solution, which was then encapsulated. Subsequently, the color variations of the nanofiber membranes were captured every 5 min with a digital camera, and the corresponding RGB parameter values were extracted. The sensitivity of the nanofibers toward ammonia vapor was evaluated according to the following Equation (3):

$$[eqn]$$

where R_a_, G_a_, B_a_ are the initial RGB values, while R_b_, G_b_, B_b_ are the RGB values after exposure.

2.13. Mechanical Properties Test

The prepared nanofiber labels (38 mm × 13 mm × 0.16 mm) were subjected to mechanical property testing using a tensile tester (XLW, Labthink Instruments Co., Ltd., Jinan, China) at a tensile rate of 25 mm/min. The tensile strength (TS) and elongation at break (EAB) were simultaneously recorded. Each sample was tested in triplicate (n = 3), and the results are expressed as mean ± standard deviation.

2.14. Color Stability Test

The prepared APPNFs-1.2% samples were stored at 4 °C and 25 °C for 14 days. During storage, the color parameters were measured every two days using a Minolta Chroma Meter Model CR-400 (Minolta Co. Ltd., Osaka, Japan). The color stability was evaluated according to Formula (2).

2.15. Degradability Test

To evaluate the biodegradability of APNFs and APPNFs as green materials, we conducted soil burial tests. The dried nanofiber membranes were cut into 10 mm × 20 mm rectangles and placed in a 500 mL beaker containing natural dry soil, using polyethylene (PE) film as a control. The beakers were controlled at 25 ± 2 °C and 45 ± 5% relative humidity, and deionized water was sprayed every 24 h to maintain soil moisture. The nanofiber membranes were regularly retrieved to observe the morphological changes and measure mass loss, and then reburied in the same soil for continued degradation.

2.16. Application of Nanofibers in Food Freshness Monitoring

Low-temperature-cooked sausages (60.0 g ± 1.0 g) and fresh shrimp (30.0 g ± 1.0 g) were individually positioned in sterile culture dishes with a diameter of 120 mm. APNFs and APPNFs-1.2% specimens, cut into 15 × 30 mm strips, were attached to the top of the inner lid of the Petri dishes, which were then sealed and incubated at 25 °C and 75% relative humidity. For sausage storage, APPNFs-1.2% color was recorded at 0, 12, 36 and 60 h. For shrimp storage, APPNFs-1.2% color was recorded at 0, 6, 12, 24 and 36 h. Concurrently, the total volatile basic nitrogen (TVB-N) content was analyzed with a K9860 Kjeldahl nitrogen analyzer (Jinan Hannong Instrument Co., Ltd., Jinan, China). Sample pH was measured using an S210-B pH meter. The corresponding ΔE values were recorded at each sampling time.

2.17. Sensory Evaluation

To evaluate freshness changes in fresh shrimp stored at 25 °C and low-temperature-cooked sausages, a sensory evaluation was conducted. This followed the methodology of Sun et al., with appropriate modifications, as outlined below [37]. The panel consisted of 10 trained evaluators. A 9-point scale was used for assessment, with sensory attributes including appearance, color, odor, texture, and overall acceptability. A score of 5 was defined as the sensory acceptability threshold, and the scoring criteria were based on Table 1 and Table 2. The sensory evaluations for fresh shrimp were conducted at 0, 6, 12, 24, and 36 h, while the evaluations for the low-temperature-cooked sausages were performed at 0, 12, 36, and 60 h. All samples were randomly coded and presented to evaluators in random order. Sensory evaluation results were visualized using radar charts.

2.18. Data Analysis

The data from three independent experiments are presented as the mean ± standard deviation, with a p < 0.05 considered statistically significant. SPSS Statistics version 17.0 (SPSS Inc., Chicago, IL, USA) was used to perform statistical comparisons by one-way analysis of variance (ANOVA).

3. Results and Discussion

3.1. Ultraviolet-Visible Spectra and Colorimetric Responses of PCA Under Variable pH Conditions

The colorimetric changes, chemical structure transitions, and UV-Vis spectral data of PCA solutions across pH 2.0–12.0 are shown in Figure 1a–c. As pH increased from 2.0 to 12.0, the PCA solutions displayed a sequential color evolution: red (pH 2.0–4.0), pink (pH 5.0–6.0), purple (pH 7.0–8.0), blue (pH 9.0–11.0), and greenish-yellow (pH 12.0) (Figure 1a), which corresponded to characteristic structural transformations of PCA: from protonated flavylium cations (acidic conditions) to carbinol pseudobases (neutral pH), and then to neutral quinoidal bases (alkaline conditions), eventually forming chalcone species under strongly alkaline conditions [17]. In the acidic range (pH 2.0–4.0), the UV-Vis spectra showed a maximum absorption peak at ~530 nm, with absorbance decreasing gradually as pH increased (Figure 1b). Under alkaline conditions (pH 8.0–11.0), the absorption peak shifted to around 603 nm, and the color change was characterized by a shift from purple to blue. The phenomenon we observed was the same as the spectral characteristics of anthocyanins varying with pH previously reported by Huang et al. [38]. The occurrence of color and spectral shift is caused by the change in the chemical structure of PCA in different pH environments [39]. The above results indicated that PCA could serve as a natural pH indicator in smart tags, which is attributed to its structure’s ability to reversibly change with pH and generate corresponding color responses.

3.2. Morphological Analysis

The surface morphology of nanofibers with varying PCA concentrations was characterized via scanning electron microscopy (SEM). Figure 2 displays the SEM micrographs along with the diameter distribution. To form continuous electrospun nanofibers, the polymer chains need to be fully entangled with each other [40]. However, the main difficulty in using a single AX solution is the insufficient viscoelasticity of the spray and the poor stability of the Taylor cone, which makes it impossible to form continuous nanofibers [41]. As shown in Figure 2, all five samples exhibit clear nanofiber structures at 5000× magnification. These smooth and uniform cylindrical fibers are interlaced and connected, forming a nonwoven network structure. The experimental results showed that adding PVA to AX can effectively enhance the entanglement of molecular chains and significantly improve its electrospinning performance, thus successfully preparing AX-based nanofibers. At the same time, PCA incorporation further improved nanofiber uniformity by modifying AX-PVA chain entanglement via anthocyanin–matrix interactions [42]. Solution viscosity and electrical conductivity of the precursor solution were measured to evaluate its processability, as these parameters directly affect the jet stability and fiber formation during the electrospinning process. The solution viscosity, electrical conductivity, and AD are presented in Table 3. As PCA concentration increased from 0 to 1.2%, viscosity rose from 0.24 ± 0.01 to 0.84 ± 0.02 Pa·s, while conductivity increased from 231 ± 2.7 to 318.3 ± 1.2 μS/cm. For APPNFs, AD initially decreased slightly from 510 ± 85 nm (APNFs, 0% PCA) to 394 ± 86 nm with low PCA incorporation, then gradually increased to a maximum of 568 ± 80 nm at higher concentrations. This indicated that fiber diameter during electrospinning was jointly affected by solution viscosity and conductivity [43]. At lower solution concentrations, conductivity dominates, resulting in finer fibers. Conversely, as concentration increases, viscosity becomes the primary factor, limiting draw-out and leading to coarser fibers [44]. These results confirm that the properties of the solution play a crucial role in controlling the morphology of the nanofibers. No distinct PCA particles were detected in the SEM micrographs, confirming the uniform incorporation and fine dispersion of PCA within the AX/PVA fiber matrix.

3.3. FT-IR Analysis

FT-IR spectroscopy was employed to identify specific functional groups and chemical bonds in the interaction system [45]. The FT-IR spectra of AX, PVA, PCA and the electrospun nanofiber mats with varying PCA concentrations (APNFs, APPNFs-0.3%, APPNFs-0.6%, APPNFs-0.9%, APPNFs-1.2%) are shown in Figure 3a. In the range of 3200–3400 cm^−1^, all samples showed a wide and strong absorption peak, corresponding to the O-H stretching vibrations [46]. Compared to pure PVA, the O-H stretching band of APNFs and APPNFs showed a slight shift, indicating intermolecular hydrogen bonding between components [47]. A strong absorption band at ~2900 cm^−1^ was observed, which corresponds to the stretching vibrations of C-H. For PCA, characteristic bands at 1603 cm^−1^ (aromatic ring C=C stretching) and 1146 cm^−1^ (glucopyranose ring O-C stretching) have been reported [48]. Characteristic peaks corresponding to PCA were absent in the nanofibers, likely due to the low concentration of PCA. It was worth noting that PCA incorporation altered both the position and intensity of the characteristic bands, particularly at ~3300 cm^−1^ (O-H stretching) and ~2915 cm^−1^ (C-H stretching). Compared to the control (APNFs), APPNFs showed a redshift (decreased wavenumber) of these bands with increasing PCA content. This redshift suggested the formation of new hydrogen bonds among PCA, AX and PVA [49]. Therefore, FT-IR spectroscopy confirmed the successful encapsulation of PCA and tight molecular binding between the PVA/AX/PCA system via hydrogen bonding.

3.4. X-Ray Diffraction Analysis

Figure 3b displays the XRD profiles of AX, PVA, PCA, and the electrospun nanofiber mats with different PCA concentrations (APNFs, APPNFs-0.3%, APPNFs-0.6%, APPNFs-0.9%, APPNFs-1.2%). In the XRD pattern of PVA, the characteristic peaks corresponding to its crystal structure were observed at approximately 11.6°, 19.4°, and 22.7°, which are consistent with the research results of Lv et al. [43]. Its crystallinity index (CrI) was 31.50%. By contrast, PCA and AX lacked sharp diffraction peaks, exhibiting only a wide peak at 2θ = 20°, indicative of their amorphous structure, their CrI values were 17.45% and 9.74%, respectively [50]. Upon doping AX into PVA and electrospinning into nanofibers, APNFs displayed a significantly weakened diffraction peak at 2θ = 19.0° (vs. PVA), and its CrI decreased to 19.89%. This indicates that the incorporation of AX effectively disrupts the ordered crystalline regions of PVA. The decrease in CrI can be attributed to strong intermolecular interactions between AX and PVA, which hinder the regular alignment and arrangement of PVA molecular chains, indirectly confirming the excellent compatibility between the AX/PVA matrix [51]. The incorporation of PCA into APNFs resulted in weakened crystalline peak intensities, with the CrI values of APPNFs gradually decreasing from 16.64% to 15.04%, 12.11%, and finally to 7.72%, indicating a further decrease in material crystallinity. This may be attributed to the abundant hydroxyl groups in PCA promoting extensive hydrogen bonding interactions with PVA and AX, leading to redistribution of the hydrogen bond network and the formation of a highly amorphous structure [48]. The XRD patterns of APPNFs closely resemble those of APNFs, indicating that PCA is effectively incorporated into the AX/PVA matrix via hydrogen bonding without forming a distinct crystalline phase. This further confirms the excellent compatibility and uniform dispersion of PCA within the nanofiber system.

3.5. Thermal Stability Analysis

The decomposition behavior of the sample during the heating process was observed through thermogravimetric analysis (TGA) and derivative thermogravimetric curves (DTG) (Figure 4a,b). In the range of 30 to 130 °C, APPNFs showed a decline in quality, which was mainly caused by the evaporation of moisture adsorbed in the material due to heat. This observation could be ascribed to the porous structure and low density characteristics of electrospun nanofibers [52]. The main thermal degradation of the sample occurred in the temperature range of 200–350 °C. This process involved the main chain breakage of the electrospun PVA/AX matrix, accompanied by the partial decomposition of PCA. The DTG curve showed that the maximum decomposition temperature (T_max_) of PCA was 304.85 °C. This value was markedly higher than that of the composite film system. The T_max_ of APNFs was 246.17 °C, while the T_max_ of APPNFs containing 0.3%, 0.6%, 0.9% and 1.2% PCA were 248.17 °C, 247.53 °C, 247.55 °C and 247.06 °C, respectively. A secondary DTG peak was observed at around 338 °C for APNFs, which was attributed to the secondary degradation of the PVA/AX matrix. Compared with APNFs, the differences in these Tmax values were all less than 2 °C, indicating that the addition of PCA had little effect on the main degradation temperature, but it slightly broadened the DTG peak—this might have been caused by the overlapping of multi-component decomposition processes. Overall, the addition of PCA had minimal influence on the thermal stability of the electrospun nanofibers and caused no adverse effects. The PVA-based nanofibers had excellent thermal stability below 100 °C, indicating that they were suitable for food packaging [50].

3.6. WCA Analysis

The relationship between the WCA of the nanofibers and the concentration of PCA is shown in Figure 5. When the content of PCA increased from 0% to 1.2%, the WCA of the nanofibers decreased from 80° to 54°, indicating that with the increase in the addition amount of PCA, the hydrophilicity of the material gradually increased. The enhancement of hydrophilicity was attributed to the fact that PCA brought numerous polar functional groups (e.g., phenolic hydroxyl and carboxyl groups) onto the surface of nanofibers, enabling the formation of more hydrogen bonds with water molecules [53]. Even when the concentration of PCA was as low as 0.3%, the WCA value was still significantly reduced, indicating that a trace amount of PCA had a certain impact on the surface properties of fibers. When the concentration of PCA increased to 0.9–1.2%, the decrease in WCA became more significant, as PCA covered the fiber surface more completely. In addition, PCA altered the structure and pore distribution of fibers, which helped water droplets spread and wet the membrane surface more quickly; the literature on similar phenolic acid-modified systems has also reported the relationship between this concentration and response, supporting our experimental results [43]. Although enhanced hydrophilicity may improve the moisture absorption capacity, the storage stability test results (with ΔE < 5 during storage at 4 °C and 25 °C) indicate that the nanofiber membranes can still maintain stable color performance under normal storage conditions. It is worth noting that moderate hydrophilicity has advantages for the application of intelligent indicators, as it can facilitate the interaction between anthocyanin molecules and the volatile alkaline compounds produced during food spoilage. Therefore, the hydrophilicity induced by PCA effectively improves the sensing response performance without significantly affecting the storage stability.

3.7. pH Responsiveness and Ammonia Sensitivity

As can be seen from Figure 6a, APPNFs-1.2% showed a similar color change to the PCA solution under different pH environments. When the pH value increased, the color shifted from red-pink to blue-green. The buffering system reflects the overall pH behavior. During the actual process of food spoilage, the volatile amines in the headspace form a local alkaline environment on the membrane surface, resulting in more significant color changes. To test the color reactivity and reversibility, we placed APPNFs-1.2% successively in the environments of 5% volatile ammonia and 10% acetic acid, and observed and recorded the color changes. Using these relatively high concentrations to accelerate the interaction between the gas and the indicator, and clearly demonstrating the responsiveness of the system under controlled laboratory conditions. According to the data in Table 4, when APPNFs-1.2% was exposed to volatile ammonia gas for 50 s, the color changed significantly: the a value changed from 6.4 to 12.0, and the b value changed from −1.3 to 0.4. After being treated in acetic acid for 135 s, the color almost recovered, and the color difference ∆E dropped from 19.9 to 6.1. This reflected the material’s rapid response and recovery ability to volatile ammonia, demonstrating a strong color responsiveness. In order to further verify its reversibility, the gas exposure sequence was adjusted as shown in Table 5, and the color change was still confirmed. As can be seen from Figure 6c, the nanofibers showed a light blue after 5 min, and the blue deepened after 10 min. The S_RGB_ was 4.1% and 21.6%, respectively. After 15 min, the blue-green S_RGB_ reached 28.9%, and then the color further deepened and finally increased to 65.1% at 25 min. This color change was due to the formation of phenolic oxygen anions [54], verifying its potential as a volatile-amine-responsive indicator membrane for monitoring food quality [55].

Importantly, although these laboratory tests used accelerated ammonia concentrations, validation in actual food systems (low-temperature-cooked sausages and fresh shrimp) demonstrated a clear correlation between the color reaction and the TVB-N value. This indicates that the developed indicator remains sensitive under actual spoilage conditions.

3.8. Mechanical Properties

The mechanical properties of the samples are shown in Figure 7a,b. The TS of all samples exceeded the minimum acceptable value for food packaging materials (2.5 MPa), indicating that the prepared nanofiber mats meet the basic application requirements for food packaging labels in terms of mechanical properties [56]. Compared to OSAS, the introduction of AX significantly increased the TS of the samples, which may be attributed to the reinforcing effect of AX on the fiber network structure [23,57]. With increasing PCA content, both the TS and EAB of APNFs exhibited a gradual upward trend, reaching maximum values of 5.34 ± 0.09 MPa and 32.43 ± 1.02%, respectively. The improvement in the mechanical properties of the nanofiber mats is primarily attributed to enhanced intermolecular interactions between PCA and the PVA/AX fiber matrix, particularly the strengthened hydrogen bonding. This result is consistent with the enhanced molecular interactions observed in the FTIR analysis.

3.9. Color Stability

Color stability directly impacts the accuracy of food freshness assessment by smart indicator labels, making it a crucial parameter for evaluating their practical applicability. Figure 6b shows the color parameter changes in APPNFs-1.2% during 14 days of storage at 4 °C and 25 °C. Previous studies indicate that when the ΔE value is below 5, color differences in nanofiber membranes are difficult to discern with the naked eye [58]. In this study, the ΔE values at both storage temperatures consistently remained below 5, indicating that the system maintains relatively stable color characteristics during storage. The color difference value at 4 °C (ΔE = 0.99 ± 0.02) was lower than that at 25 °C (ΔE = 1.53 ± 0.02), suggesting superior stability under low-temperature conditions. This may be attributed to elevated temperatures weakening the structural stability of anthocyanin molecules, thereby accelerating their degradation process [59]. Although the indicator labels demonstrated good storage stability under laboratory conditions, exposure to light in actual packaging environments may affect the long-term stability of the anthocyanin-based indicator system. Therefore, the light stability under different light conditions needs to be further studied for practical applications.

3.10. Degradability Analysis

The degradation behavior of nanofibers under natural soil conditions was presented in Figure 8a. Among them, the control group was PE film. During the entire degradation process, the hydrophilic polymer materials absorbed water and expanded. The microorganisms and enzymes in the soil then decomposed them, destroying the structure of the nanofiber network [60]. In addition, the addition of PCA significantly accelerated the degradation process: anthocyanins in PCA have antioxidant properties and can react with soil microorganisms and water, thereby accelerating the degradation rate. The experimental results showed that the quality of PE film remained almost unchanged within three days, indicating that it was difficult to degrade in the environment, almost impossible to biodegraded, and had poor compatibility with the ecological environment. In contrast, the nanofibers we prepared have much better degradation performance, with a degradation rate as high as 90% within 72 h (Figure 8b). This indicates that the material can effectively decompose in the natural environment, has good environmental-protection performance and meets the requirements of sustainable development.

3.11. Application in Freshness Testing

To evaluate the application potential of APPNFs-1.2% in freshness detection, we measured the pH, TVB-N and ∆E values of low-temperature-cooked sausages and fresh shrimp (Figure 9a,b,d,e). During storage, APNFs showed no obvious color change, while APPNFs-1.2% exhibited a significant color change corresponding to the corruption index (Figure 10). As the storage time of low-temperature-cooked sausages increased, their color changed successively from purple (0 h) to purplish red (12 h), then to deep red (36 h), and finally to grayish pink (60 h). When the color turned deep red, the content of TVB-N was 23.5 mg/100 g (pH = 6.41, ∆E = 46.16), and at this time, the sausage was still safe to eat. When the TVB-N content in sausages is ≥25 mg/100 g, it indicates that they have gone completely bad and are no longer fit for human consumption [61]. At the grayish-pink stage, TVB-N increased to 35.65 mg/100 g (pH = 6.19, ∆E = 41.57), confirming complete spoilage. During the storage of fresh shrimp, APPNFs-1.2% showed obvious color changes: from purple (0 h), to blue (6 h), to light blue (12 h), to dark grayish blue (24 h), and finally to dark green (36 h). At the dark gray-blue stage, TVB-N was 24.18 mg/100 g (pH = 7.13, ∆E = 47.68). According to the provisions of the Chinese national standard GB 2733–2015 [62], the content of TVB-N in shrimp should be ≤30 mg/100 g. Compliance with this standard indicated that shrimp meat was safe for consumption. When the color turned dark green, the content of TVB-N reached 39.41 mg/100 g (pH = 7.47, ∆E = 53.06), which confirmed that the shrimp had completely gone bad and is inedible. Additionally, as shown in Figure 11a,b, sensory evaluation results indicate that scores gradually decline with extended storage duration. Odor and overall acceptability deteriorate more rapidly than appearance and texture. When the overall acceptability score falls below 5 points, the sample is deemed sensory non-compliant. This sensory non-compliance threshold aligns closely with the rapid upward trend in TVB-N values. Overall, APPNFs-1.2% showed noticeable color changes during storage. These color shifts were easily distinguishable to the naked eye. To further quantify the relationship between color reactions and spoilage indicators, we conducted a linear regression analysis on ΔE and TVB-N (Figure 9c,f). During the storage of low-temperature-cooked sausages (R^2^ = 0.9337) and fresh shrimp (R^2^ = 0.9261), a strong linear correlation was observed between TVB-N and ΔE, indicating that the color differences in APPNFs-1.2% are closely related to the degree of product spoilage. This result suggests that the color reaction is mainly triggered by volatile amine substances released during the spoilage process, rather than the overall change in pH value in the food matrix. This proved its practical value as an intelligent preservation indicator in real-time monitoring of food quality.

(a) pH vs. TVB-N changes and (b) ΔE vs. TVB-N changes in low-temperature-cooked sausages; (c) correlation analysis of pH and ΔE with TVB-N in low-temperature-cooked sausages; (d) pH vs. TVB-N changes and (e) ΔE vs. TVB-N changes in fresh shrimp; (f) correlation analysis of pH and ΔE with TVB-N in fresh shrimp.

Visual monitoring photographs of freshness changes in low-temperature-cooked sausages and fresh shrimp during storage at 25 °C using APNFs and APPNFs-1.2%.

Sensory radar charts illustrating freshness changes of (a) low-temperature-cooked sausages and (b) fresh shrimp during storage.

4. Conclusions

This study successfully developed a PVA/AX electrospun nanofiber intelligent indicator label based on purple cabbage anthocyanins for real-time monitoring of food freshness. PVA and AX form a stable polymer matrix, while PCA acts as a natural pH and ammonia-responsive agent, giving the nanofibers rapid and reversible color-changing ability. The nanofiber structure is uniform. The indicator labels exhibited good color stability during storage in the absence of spoilage-related volatiles (ΔE < 5), while significant color variation was observed during monitoring of low-temperature-cooked sausages and fresh shrimp. In the experiments of low-temperature-cooked sausages and fresh shrimp, the color change was highly consistent with the content of TVB-N, confirming that it can be used as a reliable visual indicator of food quality. This material can decompose well in natural soil, indicating that it can be used to develop environmentally friendly smart packaging. This indicator label is specifically designed to be attached to the inner surface of packaging materials, to avoid direct contact with food, thereby reducing the risk of migration and enhancing the compliance of food packaging applications. Subsequent research will focus on improving the mechanical strength and long-term color stability of the nanofiber membranes, quantitatively determining the detection limit of volatile amines, and exploring the scale-up integration schemes for commercial packaging systems. Overall, this research provides an innovative path for intelligent food packaging, which not only has a preservation monitoring function, but also takes into account environmental sustainability.