Black Tea Extracts/Polyvinyl Alcohol Active Nanofibers Electrospun Mats with Sustained Release of Polyphenols for Food Packaging Applications

The efficiency in the capabilities to store and release antioxidants depends on the film morphology and its manufacturing process, as well as on the type and methodology used to obtain the polyphenol extracts. Here, hydroalcoholic extracts of black tea polyphenols (BT) were obtained and dropped onto different polyvinyl alcohol (PVA) aqueous solutions (water or BT aqueous extract with and without citric acid, CA) to obtain three unusual PVA electrospun mats containing polyphenol nanoparticles within their nanofibers. It was shown that the mat obtained through the nanoparticles precipitated in BT aqueous extract PVA solution presented the highest total polyphenol content and antioxidant activity, and that the addition of CA as an esterifier or PVA crosslinker interfered with the polyphenols. The release kinetics in different food simulants (hydrophilic, lipophilic and acidic) were fitted using Fick’s diffusion law and Peppas’ and Weibull’s models, showing that polymer chain relaxation is the main mechanism in all food simulants except for the acidic, which presented an abrupt release by Fick’s diffusion mechanism of about 60% before being controlled. This research provides a strategy for the development of promising controlled-release materials for active food packaging, mainly for hydrophilic and acidic food products.


Introduction
The circular economy is the upcoming trend in the packaging industry as a result of the accumulation of plastic waste and pollutants that produce catastrophic results for the environment [1,2]. International organizations recognize plastic marine pollution as an obstacle to sustainable development. Companies seeking to be part of the sustainable development goals (SDGs) have proposed the use of biodegradable components to replace or reduce the use of traditional single-use plastics [3]. In addition, market surveys have shown consumer desire for sustainable materials that also have functional properties [4]. In this context, the electrospinning technique emerges as a very attractive alternative to obtain active ultralight packaging. Electrospun mats consist of entanglements of nanofibers with void spaces between them (forming a non-woven fabric) that can encapsulate different active components inside or on the surface of the fibers, and/or in the space between them [5]. It was demonstrated that the encapsulation of polyphenols led to higher microbiological stability and better control of their release from the mat [6][7][8][9]. The use of nanostructured mats with encapsulated polyphenols as biodegradable packaging is highly Polymers 2023, 15, 1311 3 of 16 same source of polyphenols as the nanoparticles. Unbelievably, to date black tea has not yet been considered in such a matter.
Here, we hypothesize that using the black tea extract's polyphenol nanoparticles formed in situ in an aqueous PVA solution, where the solvent is black tea extract, can maximize the total polyphenol content and antioxidant activity of an electrospun mat. In addition, we exhibited that heat treatment can be performed to achieve water insoluble PVA mats, avoiding the use of citric acid as esterifier or crosslinker. The presented results are significant because they demonstrate that electrospun mats obtained with nanoparticles precipitated on a black tea aqueous extract PVA doubling the total extracted polyphenols content with about 24% higher antioxidant activity than those containing nanoparticles formed in a water-PVA solution. Moreover, the release in both mats exceeded 64% in the hydrophilic simulant and 98% in the lipophilic one. In the case of the mat with citric acid, there could be an interaction with the polyphenols, inhibiting their complete release. The polyphenols' kinetic release of the mats in food simulants (FS, hydrophilic, lipophilic and acidic) was fitted by Fick's diffusion law and Peppas' and Weibull's models, showing that polymer chain relaxation is the leading mechanism in all FS, except for the mat with CA that presented an abrupt release by Fick's diffusion mechanism in the acid medium.
This study is original because it proposes a new strategy for the development of active nanofibrous PVA mats, which could be very promising as a type of packaging to extend the shelf life of food products.

Black Tea Aqueous Extract
The black tea aqueous extract (BT Aq ) was obtained following the procedure mentioned by Bruni et al. (2020) with some modifications; 30 g of black tea was dispersed in 100 mL of distilled water at 100 • C for 40 min [35]. The aqueous extract was cooled to room temperature and filtered (mat pore size 0.8 µm). The remaining solution was newly filtered through a 0.45 µm pore nylon membrane (MSI, WESTBORO, Massachusetts, USA). The solution was kept in amber sealed containers and refrigerated until further use.

Black Tea Ethanolic Extract
The polyphenols' ethanolic extract from black tea (BT) was obtained following the method described by Estevez-Areco et al. (2020) with some modifications, 10 g of black tea were dispersed in 50 mL of a mixture of ethanol:water (70:30) at 50 • C for 55 min. The ethanolic extract was then cooled and filtered following Section 2.2.1 [36].

Solid's Extraction Yields
The solid's extraction yield was calculated gravimetrically following the method used by Estevez-Areco et al. (2018): 5 mL of each extract was dried at 50 • C until constant weight [15].

Total Polyphenol Content (TPC) and Antioxidant Activity
The total polyphenol content (TPC) of all extracts was determined by the Folin-Ciocalteu method employing the UV-Vis technique at 760 nm (spectrophotometer UV-1800 SHIMADZU, Tokyo, Japan). Absorbance results were compared to a calibration curve built using gallic acid as a standard (10-150 mg/L) and the TPC was reported as gallic acid equivalents (GAE/mL of each extract).
Antioxidant activity of every sample was obtained following the method proposed by Brand-Williams et al. (1995) and using Equation (1) [37].
where A b is the absorbance of DPPH at the initial time and A s is the absorbance of DPPH·at 30 min. The absorbance results were compared to a calibration curve built with TROLOX (in ethanol) as a free radical reagent (50-350 mg/L). Furthermore, the concentration of polyphenols needed to decrease DPPH·concentration by 50%, also known as efficient concentration (EC50), was obtained, and both results were expressed in µmol TROLOX equivalent (µg TE/mL of extract).
The solid's extraction yield, total polyphenols concentration (TPC), antioxidant activity and EC50 of both extracts are shown in Table 1. As can be seen in Table 1, both the solid's yield and the antioxidant activity are much higher for BT than for BT Aq . BT doubled the solid's extraction yield and increased the antioxidant activity by one order. This agrees with the literature, which has shown that with ethanol:water mixtures, the extraction of polyphenols is enhanced [38,39]. Possibly this is a consequence that the alcohol-soluble polyphenols, as well as the water-soluble polyphenols, are extracted.

Preparation of Nanoparticles and Electrospun Mats
Four electrospinning PVA systems were prepared. One of them consisted of PVA in water and another three containing black tea polyphenols' nanoparticles in different aqueous PVA solutions (water or BT Aq , with and without citric acid). PVA solutions were first heated at 80 • C for 1 h under constant stirring and then allowed to cool to room temperature. For the case of the nanoparticle systems, the nanoparticles were formed in situ in the different PVA solutions using the solvent displacement technique. For this, a black tea ethanolic extract (BT) was incorporated dropwise to the aqueous PVA solution (named: PVA + BT NP ) and the aqueous black tea extract PVA solution, with or without 5 wt% (5 g/100 g of PVA) of citric acid (CA) (PVA + CA + BT Aq + BT NP and PVA + BT Aq + BT NP , respectively). As expected, these solutions become turbid as a consequence of the nanoparticles' precipitation due to the solvent displacement phenomenon.
The composition of all the PVA solutions to be electrospun, as well as their electrical conductivity and viscosity, measured at room temperature using an Orion™ Versa Star Pro™ (Thermo Fisher Scientific, Waltham, MA, USA) and a Brookfield viscometer model LV DV-E™ (AMETEK Brookfield, Marlborough, MA, USA), respectively, are shown in Table 2. PVA mats were obtained using an in-house electrospinning equipment [40]. The solutions were placed in a 10 mL plastic syringe connected to a six-needle injector with a 0.8 mm inner diameter. They were then ejected at a rotating drum collector (6 cm in diameter) and placed at 12.5 cm from the needles by applying a 28 kV between the collector and the needles. The syringe was placed at a syringe pump PC11U (APEMA, Buenos Aires, Argentina) with a controlled feed of 2.5 mL/h. Once the mats were obtained, a thermal treatment (190 • C for 10 min) was performed to improve their solubility in water [22,40,41].
A schematic of the preparation of the electrospinning solutions and active PVA electrospun mats is shown in Figure 1.  Table 2. PVA mats were obtained using an in-house electrospinning equipment [40]. The solutions were placed in a 10 mL plastic syringe connected to a six-needle injector with a 0.8 mm inner diameter. They were then ejected at a rotating drum collector (6 cm in diameter) and placed at 12.5 cm from the needles by applying a 28 kV between the collector and the needles. The syringe was placed at a syringe pump PC11U (APEMA, Buenos Aires, Argentina) with a controlled feed of 2.5 mL/h. Once the mats were obtained, a thermal treatment (190 °C for 10 min) was performed to improve their solubility in water [22,40,41].
A schematic of the preparation of the electrospinning solutions and active PVA electrospun mats is shown in Figure 1.

Nanoparticles Size and Mats Morphology
The size of the BT nanoparticles and fiber's diameter were analyzed by field emission scanning electron microscopy (FEG-SEM, FEI-QUANTA TM 250, Oregon, USA). Nanoparticle samples were obtained by dripping ethanolic extract on water and on BTAq. Drops of these dispersions were placed on single crystal silicon pieces which were dried at 50 °C for 24 h and sputter-coated with platinum using an Edward S150B sputter coater, Heathfield, UK (70 s, 0.06 mbar of Ar); mats samples were placed in carbon tapes and sputtercoated with platinum (40 s, 0.06 mbar of Ar) before SEM observations. Finally, the nanoparticles' size and fiber's diameter, which was considered cylindrical [15], were determined by measuring the diameter of at least 100 nanoparticles and 200 fibers from the SEM images, using ImageJ (free software by Wayne Rasband, National Institutes of Health, NIH in USA).

FTIR
Absorbance spectra were obtained using an Attenuated total Fourier Transform Infrared (ATR-FTIR 4100 spectrophotometer, Jasco Inc., Hachioji, Japan), in the range of 4000 to 650 cm −1 at a resolution of 4 cm −1 and the average of 64 scans. All mats' spectra  The size of the BT nanoparticles and fiber's diameter were analyzed by field emission scanning electron microscopy (FEG-SEM, FEI-Quanta TM 250 FEG, Hillsboro, OR, USA). Nanoparticle samples were obtained by dripping ethanolic extract on water and on BT Aq . Drops of these dispersions were placed on single crystal silicon pieces which were dried at 50 • C for 24 h and sputter-coated with platinum using an Edward S150B sputter coater, Heathfield, UK (70 s, 0.06 mbar of Ar); mats samples were placed in carbon tapes and sputter-coated with platinum (40 s, 0.06 mbar of Ar) before SEM observations. Finally, the nanoparticles' size and fiber's diameter, which was considered cylindrical [15], were determined by measuring the diameter of at least 100 nanoparticles and 200 fibers from the SEM images, using ImageJ (free software by Wayne Rasband, National Institutes of Health, NIH in USA).

FTIR
Absorbance spectra were obtained using an Attenuated total Fourier Transform Infrared (ATR-FTIR 4100 spectrophotometer, Jasco Inc., Hachioji, Japan), in the range of 4000 to 650 cm −1 at a resolution of 4 cm −1 and the average of 64 scans. All mats' spectra were normalized using the C-H band (1372 cm −1 ), which corresponds to a functional group that is not expected to interact with any of the components [15]. For comparative purposes, powdered extracts samples were tested and their spectra were normalized using the 844 cm −1 band as it does not change. Tests were carried out in triplicate with no significant differences.

Contact Angle (θ)
Essays were carried out to study the hydrophobicity of the mats [42]. For the essays, a "one attention theta lite optical tensiometer" (Biolin Scientific, Västra Frölunda, Sweden) was used at room temperature, dripping a drop of water (~13 µL) on the surface of each material. The contact angle (θ) was determined as the angle formed by the intersection of the drop of water-surface of the mat and the liquid-steam (tangent on the boundary of the drop). The mean of the test was reported with five measurements.

Water Solubility (S)
The water solubility test was performed by taking~25 mg (m si ) of each mat and submerging it in 50 mL of distilled water at room temperature for 24 h under constant stirring; then was dried at 100 • C for 24 h and weighed (m sf ). Results were calculated following Equation (2) and expressed as the average of three samples.

Total Extracted Polyphenol Content (TEPC) and Antioxidant Activity of Mats
To measure the TEPC in the mat and antioxidant activity, a polyphenol extraction from all mats was performed. For TEPC, a square of 25 mg of each mat was placed in 100 mL of ethanol:water solution (70:30 v/v) at 50 • C with constant stirring and reflux for 32-h. The polyphenol content and antioxidant activity were measured by the Folin-Ciocalteu and DPPH·methods, respectively. Briefly, 0.4 mL of each sample was mixed with 2 mL of Folin-Ciocalteu reagent (previously diluted 1:10), and 1.6 mL of sodium carbonate (7 g/100 mL) was incorporated into the mix afterwards. Every mixture was allowed to proceed for 30 min in the darkness. Then, TEPC and the antioxidant activity were determined following the procedure described in Section 2.2.4.
Theoretical TPC, also known as the total polyphenol content loaded (TPL), for each system was calculated for comparative purposes, obtaining~26.9 mg GAE/PVA + BT NP mat and~47.6 mg GAE/PVA + BT Aq + BT NP mat (with and without citric acid, respectively) in 100 mL of the precursor solution.
To evaluate the stability of the extracts released from the PVA mats, TEPC and antioxidant activity were determined after 4 months of storage at room temperature.

Release of Polyphenols from the Mats
Polyphenols release from the mats was studied in three different food simulants: [43]. Samples of~25 mg of each mat were immersed in 20 mL of each food simulant and shaken at 120 rpm. Aliquots were taken at different times (20 s, 20 min, 40 min, 1 h, 2 h, 4 h, 6 h, 24 h, 48 h and 72 h) and polyphenols released were measured using the Folin-Ciocalteu method described in Section 2.4.5. Furthermore, the released polyphenols (%) were calculated using Equation (3), where M pt is the amount of the polyphenol released at time t and M o is the total extracted polyphenol content (TEPC) normalized to the mass of the mat. The results are reported as the mean of three essays and then fitted with Fick's diffusion law and the Weibull function using SciPy [15].
If the system is an electrospun mat, nanofibers must be considered as cylindrical geometry and the Fick's equation to the second law of diffusion solution (Equation (4)) is [44]: Here M t is the relation, M pt /M o , M ∞ is the polyphenol amount at equilibrium (infinite time), D is the diffusion coefficient, r is the fiber average radius and r·α n are the Bessel's positive roots to this function of first kind of zero order [15]. In this work the first ten roots of the Bessel function were significant for the fitting. For systems that do not follow a pure diffusional process, the Power Law [45] by Korsmeyer-Peppas is an alternative. This law only considers the first portion of the release where M t /M ∞ < 0.6 and can describe the Fickian and non-Fickian release based on the parameters of Equation (5): Here k is the kinetic constant and n a diffusional exponent, which characterizes the release system mechanism. If n is lower than 0.5 it represents a pseudo-Fickian diffusion mechanism, close to 0.5 the principal mechanism is Fick diffusion, close to 1 it indicates Case II transport (where polymer chain relaxation domains polyphenols release), and when n is between 0.5 and 1 it denotes an anomalous release. Another alternative to describe the release behavior is the empirical Weibull function (Equation (6)): Here a and b are constants. A b value smaller than 0.75 indicates that the release mechanism occurs according to Fick's diffusion Law combined with Case II transport, intermediate values (0.75-1) indicate Case II transport and b > 1 indicate complex release mechanisms [46,47].

Data Treatment and Statistical Analysis
All obtained results were analyzed through two-way ANOVA with a 95% confidence level (p < 0.05) and Tukey's Post-Hoc test. Data were reported as the main and standard deviation.

Characterizations of Nanoparticles and Mats
Nanoparticles Size and Mats Morphology  [34]. The use of CA did not present significant differences in the nanoparticle size distribution with respect to BT Aq + BT NP ( Figure 2B). All mats exhibited a homogeneous fibers morphology, as shown in Figure 2C-F. However, it can be noted that the main diameter of the nanofibers and their size distribution depend on the solution studied. It is known that the evaporation of the solvent in the polymeric solution plays a fundamental role in the final morphology of the electrospun mats. According to the literature, a faster solvent evaporation rate leads to wider nanofiber diameters, regardless of the polymer used in the electrospinning process [48][49][50]. This effect is associated with the earlier solidification of the nanofiber that is expelled from the needle: the higher the evaporation rates, the lower the elongation of the nanofiber, solidifying before others that have had more time to stretch (resulting in a narrower nanofiber). PVA + BTNP, which has water, polyphenols and ethanol, contrary to the PVA mat that only contains water, showed an increase in the average fiber size ( Figure 2C). Estevez-Areco et al. (2018) reported that the addition of ethanol as a solvent in the electrospinning solution decreases the surface tension of the PVA solutions and consequently the evaporation rate during the electrospinning process increases, leading to wider nanofiber diameter and size distributions [15].
The mean diameter of the PVA + BTAq + BTNP nanofiber ( Figure 2E) is greater than that of PVA but slightly lower than that of PVA + BTNP. This material contains the greatest amount of polyphenols but has the highest electrical conductivity (Table 2), which, according to the literature, leads to thinner fibers [5]. Moreover, in PVA + BTAq + BTNP, several fibers were stuck together. However, when citric acid was present in the same solution ( Figure 2F), the diameter of the fiber increased to 195 ± 73 nm, leading to the highest value among the active PVA mats. As was reported, when CA concentration increases in an electrospinning precursor solution, the conductivity decreases, leading to a notably lower nanofiber diameter and a higher dispersion [51,52]. This effect has been associated with the polyphenols avoiding the PVA chains from interacting via inter/intramolecular hydrogen bonds and creating void spaces with the polyphenols entrapped within, leading to wider fibers [27,41,49,50]. The wider dispersion of the fibers' diameter obtained for the mats prepared with both extracts (PVA + BTAq + BTNP and PVA + BTAq + CA + BTNP) was All mats exhibited a homogeneous fibers morphology, as shown in Figure 2C-F. However, it can be noted that the main diameter of the nanofibers and their size distribution depend on the solution studied. It is known that the evaporation of the solvent in the polymeric solution plays a fundamental role in the final morphology of the electrospun mats. According to the literature, a faster solvent evaporation rate leads to wider nanofiber diameters, regardless of the polymer used in the electrospinning process [48][49][50]. This effect is associated with the earlier solidification of the nanofiber that is expelled from the needle: the higher the evaporation rates, the lower the elongation of the nanofiber, solidifying before others that have had more time to stretch (resulting in a narrower nanofiber). PVA + BT NP , which has water, polyphenols and ethanol, contrary to the PVA mat that only contains water, showed an increase in the average fiber size ( Figure 2C). Estevez-Areco et al. (2018) reported that the addition of ethanol as a solvent in the electrospinning solution decreases the surface tension of the PVA solutions and consequently the evaporation rate during the electrospinning process increases, leading to wider nanofiber diameter and size distributions [15].
The mean diameter of the PVA + BT Aq + BT NP nanofiber ( Figure 2E) is greater than that of PVA but slightly lower than that of PVA + BT NP . This material contains the greatest amount of polyphenols but has the highest electrical conductivity (Table 2), which, according to the literature, leads to thinner fibers [5]. Moreover, in PVA + BT Aq + BT NP, several fibers were stuck together. However, when citric acid was present in the same solution ( Figure 2F), the diameter of the fiber increased to 195 ± 73 nm, leading to the highest value among the active PVA mats. As was reported, when CA concentration increases in an electrospinning precursor solution, the conductivity decreases, leading to a notably lower nanofiber diameter and a higher dispersion [51,52]. This effect has been associated with the polyphenols avoiding the PVA chains from interacting via inter/intramolecular hydrogen bonds and creating void spaces with the polyphenols entrapped within, leading to wider fibers [27,41,49,50]. The wider dispersion of the fibers' diameter obtained for the mats prepared with both extracts (PVA + BT Aq + BT NP and PVA + BT Aq + CA + BT NP ) was possibly due to the presence of water and ethanol soluble polyphenols, contrary to PVA + BT NP in which only water soluble polyphenols are present. Figure 3a shows FTIR spectra of BT NP and BT Aq + BT NP powder and the characteristic polyphenol bands. The wide bands around 3282 cm −1 , 1142 cm −1 and 1032 cm −1 are attributed to the stretching of -OH and C-O groups and the bending of the C-OH bonds, respectively, from phenol groups present in the extracts [18,28,53]. The bands at 2922 and 2848 cm −1 are ascribed to symmetric and asymmetric C-H stretching, and the band at 823 cm −1 corresponds to the C-H deformation out of the plane. The bands around 1695 cm −1 and 1607 cm −1 are associated to C=O from carbonyl groups and C=C stretching vibration of the aromatic rings, respectively. Finally, the band around 1448 cm −1 can be assigned to the vibrations of aliphatic hydroxyl groups [54]. As expected, all the bands associated with the polyphenols show a greater increase in their intensity in BT Aq + BT NP due to their higher TPC ( Table 1). The intensity of some bands differs depending on the extract, suggesting that they have different types of polyphenols, which is expected due to the different extraction methods used. In this sense, the presence of new bands at 1327 and 1367 cm −1 is observed in the BT Aq + BT NP that have been associated with the C-O interactions and the O-H stretching of the polyphenols, respectively [55].  Figure 3a shows FTIR spectra of BTNP and BTAq + BTNP powder and the characteristic polyphenol bands. The wide bands around 3282 cm −1 , 1142 cm −1 and 1032 cm −1 are attributed to the stretching of -OH and C-O groups and the bending of the C-OH bonds, respectively, from phenol groups present in the extracts [18,28,53]. The bands at 2922 and 2848 cm −1 are ascribed to symmetric and asymmetric C-H stretching, and the band at 823 cm −1 corresponds to the C-H deformation out of the plane. The bands around 1695 cm −1 and 1607 cm −1 are associated to C=O from carbonyl groups and C=C stretching vibration of the aromatic rings, respectively. Finally, the band around 1448 cm −1 can be assigned to the vibrations of aliphatic hydroxyl groups [54]. As expected, all the bands associated with the polyphenols show a greater increase in their intensity in BTAq + BTNP due to their higher TPC ( Table 1). The intensity of some bands differs depending on the extract, suggesting that they have different types of polyphenols, which is expected due to the different extraction methods used. In this sense, the presence of new bands at 1327 and 1367 cm −1 is observed in the BTAq + BTNP that have been associated with the C-O interactions and the O-H stretching of the polyphenols, respectively [55]. Several changes have been produced by the incorporation of the polyphenols into the PVA mats (Figure 3b). Neat PVA mat spectrum showed a band around 3304 cm −1 corresponding to the stretching of the hydroxyl groups. Furthermore, there is the double peak in 2938 cm −1 and 2908 cm −1 that can be assigned to the symmetric and asymmetric C-H stretching. The band around 1707 cm −1 corresponds to C=O, 1088 cm −1 to the stretching of the O-H groups and that at 1328 and 1431 cm −1 to the C-H bending and that around 1142 cm −1 to the C-O bending [41,56]. The black tea extract caused increases in the O-H absorption band of PVA mats and the appearance of a new band around 1606 cm −1 associated with C=C vibrations of the aromatic ring of the extract, confirming the presence of polyphenol compounds [28,56]. On the other hand, a broadening in the band at 1088 cm −1 for mats containing the black tea extract was observed (Figure 3b), possibly due to an overlap with the C-OH band from phenolic compounds [15].

FTIR
Different effects can be observed in the range of the OH vibration (Figure 3c): almost no displacement in the case of the PVA + BTNP mat, which could be due to the lower polyphenol content, and a shift to lower wavenumber in the PVA + BTAq + BTNP mat that could be due to hydrogen bonding interactions between PVA and black tea extract components [28]. However, in the PVA + BTAq + CA + BTNP mat, this shift is reverted possibly due to the esterification/crosslink reaction between CA and -OH groups of the PVA chain Several changes have been produced by the incorporation of the polyphenols into the PVA mats (Figure 3b). Neat PVA mat spectrum showed a band around 3304 cm −1 corresponding to the stretching of the hydroxyl groups. Furthermore, there is the double peak in 2938 cm −1 and 2908 cm −1 that can be assigned to the symmetric and asymmetric C-H stretching. The band around 1707 cm −1 corresponds to C=O, 1088 cm −1 to the stretching of the O-H groups and that at 1328 and 1431 cm −1 to the C-H bending and that around 1142 cm −1 to the C-O bending [41,56]. The black tea extract caused increases in the O-H absorption band of PVA mats and the appearance of a new band around 1606 cm −1 associated with C=C vibrations of the aromatic ring of the extract, confirming the presence of polyphenol compounds [28,56]. On the other hand, a broadening in the band at 1088 cm −1 for mats containing the black tea extract was observed (Figure 3b), possibly due to an overlap with the C-OH band from phenolic compounds [15].
Different effects can be observed in the range of the OH vibration (Figure 3c): almost no displacement in the case of the PVA + BT NP mat, which could be due to the lower polyphenol content, and a shift to lower wavenumber in the PVA + BT Aq + BT NP mat that could be due to hydrogen bonding interactions between PVA and black tea extract components [28]. However, in the PVA + BT Aq + CA + BT NP mat, this shift is reverted possibly due to the esterification/crosslink reaction between CA and -OH groups of the PVA chain [15,41,57]. This effect is also observed by the increase in the intensity of the C=O stretching band. The intensity ratio of the C=O band to the C-H one (I 1707 /I 1374 ) for PVA + BT Aq + CA + BT NP (6.96) is higher than that of the PVA + BT Aq + BT NP (5.03), PVA-BT NP (4.42) and PVA mat (2.06) samples. On the other hand, polyphenols can also interact with the -OH groups in the PVA mat, as shown above. These interactions may help to retain the polyphenols within the fibers, as was also observed for PVA with the CA mat [15]. These results, as will be shown in Section 3.5, correlated with the lower content of polyphenols extracted from this mat.

Contact Angle (θ)
Contact angle (θ) measurements are reported in Figure 4. The PVA + BT NP mat shows lower θ than the PVA mat, indicating an increase in its hydrophilicity, possibly due to the nanoprecipitation that leads to a higher surface-volume ratio of the BT nanoparticles [34]. [15,41,57]. This effect is also observed by the increase in the intensity of the C=O stretching band. The intensity ratio of the C=O band to the C-H one (I1707/I1374) for PVA + BTAq + CA + BTNP (6.96) is higher than that of the PVA + BTAq + BTNP (5.03), PVA-BTNP (4.42) and PVA mat (2.06) samples. On the other hand, polyphenols can also interact with the -OH groups in the PVA mat, as shown above. These interactions may help to retain the polyphenols within the fibers, as was also observed for PVA with the CA mat [15]. These results, as will be shown in Section 3.5, correlated with the lower content of polyphenols extracted from this mat.

Contact Angle (θ)
Contact angle (θ) measurements are reported in Figure 4. The PVA + BTNP mat shows lower θ than the PVA mat, indicating an increase in its hydrophilicity, possibly due to the nanoprecipitation that leads to a higher surface-volume ratio of the BT nanoparticles [34].
On the other hand, PVA + BTAq + BTNP had the lowest θ value; this can be associated with the incorporation of the aqueous polyphenol BT, suggesting that more available OH are on the surface. Moreover, the addition of CA could lead to an esterification/crosslink reaction with PVA, increasing the PVA + BTAq + CA + BTNP mat contact angle with respect to PVA + BTAq + BTNP, and thus, its hydrophilicity [15,23,41].

Water Solubility (S)
Water solubility (S) increases with the incorporation of the polyphenols (Figure 4). Results show a slight difference between PVA + BTNP and PVA + BTAq + BTNP but do not show significant differences between the mats with and without citric acid. This agrees with Liu et al. (2019), who attributed this effect to polyphenols content in the electrospun mat [58].

Total Polyphenol Extracted Content (TEPC) and Antioxidant Activity of Mats
Comparing the samples of the PVA + BTNP and PVA + BTAq + BTNP mats, the first has the lowest TEPC (Table 3). This was expected since the only source of polyphenols is BT. However, considering the theoretical TPC (Section 2.4.5) used in the precursor solutions, 88.9 ± 0.4% and 88.2 ± 0.3% of polyphenols' retention, respectively, were achieved. The difference between the theoretical data for these mats could be due to the loss of polyphenols during the electrospinning process and/or the heat treatment. Citric acid (PVA + BTAq + CA + BTNP) hindered the extraction of the polyphenols from the mat, releasing only 31.0 ± 0.4%. This effect may have been due to possible interactions between citric acid and polyphenols. Since CA interacts with PVA [15,17], the polyphenols could have been retained in the mat. On the other hand, PVA + BT Aq + BT NP had the lowest θ value; this can be associated with the incorporation of the aqueous polyphenol BT, suggesting that more available OH are on the surface. Moreover, the addition of CA could lead to an esterification/crosslink reaction with PVA, increasing the PVA + BT Aq + CA + BT NP mat contact angle with respect to PVA + BT Aq + BT NP , and thus, its hydrophilicity [15,23,41].

Water Solubility (S)
Water solubility (S) increases with the incorporation of the polyphenols (Figure 4). Results show a slight difference between PVA + BT NP and PVA + BT Aq + BT NP but do not show significant differences between the mats with and without citric acid. This agrees with Liu et al. (2019), who attributed this effect to polyphenols content in the electrospun mat [58].

Total Polyphenol Extracted Content (TEPC) and Antioxidant Activity of Mats
Comparing the samples of the PVA + BT NP and PVA + BT Aq + BT NP mats, the first has the lowest TEPC (Table 3). This was expected since the only source of polyphenols is BT. However, considering the theoretical TPC (Section 2.4.5) used in the precursor solutions, 88.9 ± 0.4% and 88.2 ± 0.3% of polyphenols' retention, respectively, were achieved. The difference between the theoretical data for these mats could be due to the loss of polyphenols during the electrospinning process and/or the heat treatment. Citric acid (PVA + BT Aq + CA + BT NP ) hindered the extraction of the polyphenols from the mat, releasing only 31.0 ± 0.4%. This effect may have been due to possible interactions between citric acid and polyphenols. Since CA interacts with PVA [15,17], the polyphenols could have been retained in the mat.
Moreover, Table 3 shows higher antioxidant activity for PVA + BT Aq + BT NP followed by PVA + BT NP , which presented similar values to that reported by Estevez-Areco et al.
(2018) for PVA mats with polyphenols from another natural source (120 ± 8 µmol/g of mat) [15]. In both cases, the polyphenol nanoparticles were precipitated over an aqueous PVA electrospinning solution. However, the results obtained for PVA + BT Aq + BT NP mat showed that when the nanoparticles were precipitated on an aqueous black tea extract PVA solution, the TEPC and the antioxidant activity were increased.
The stability of the mats with higher total extracted polyphenols content (PVA + BT NP and PVA + BT Aq + BT NP ) after 4 months of storage was evaluated. As can be seen in Table 3, the TEPC and the antioxidant activity of these mats slightly decreased (less than 1%). These results suggest that the polyphenols remained stable for at least during the studied storage time. Antioxidant activity (µmol/g of mat), t = 0 125 ± 1 a 155 ± 3 b 51 ± 2 c Antioxidant activity (µmol/g of mat), t = 4 months 120 ± 3 a 147 ± 4 b a,b,c Different letters in the same row indicate significant differences (p < 0.05).

Release of Polyphenols from the Mats
The release test provides information on the affinity between active materials and food products; thus, the most suitable active material can be selected for each type of food [32,59]. The compositions of polymeric composites and the food simulant determine the migration mechanisms between both phases. The chemical potential of each phase controls the diffusion coefficients and the equilibrium concentrations.
The release of polyphenols, normalized with respect to the TEPC, in the hydrophilic simulant was from 75% to 84%, in the lipophilic from 94% to 98% and in the acid from 52% to 66% ( Figure 5). The lipophilic medium showed the smallest difference in the final release. Moreover, Table 3 shows higher antioxidant activity for PVA + BTAq + BTNP followed by PVA + BTNP, which presented similar values to that reported by Estevez-Areco et al. (2018) for PVA mats with polyphenols from another natural source (120 ± 8 μmol/g of mat) [15]. In both cases, the polyphenol nanoparticles were precipitated over an aqueous PVA electrospinning solution. However, the results obtained for PVA + BTAq + BTNP mat showed that when the nanoparticles were precipitated on an aqueous black tea extract PVA solution, the TEPC and the antioxidant activity were increased.
The stability of the mats with higher total extracted polyphenols content (PVA + BTNP and PVA + BTAq + BTNP) after 4 months of storage was evaluated. As can be seen in Table  3, the TEPC and the antioxidant activity of these mats slightly decreased (less than 1%). These results suggest that the polyphenols remained stable for at least during the studied storage time.

Release of Polyphenols from the Mats
The release test provides information on the affinity between active materials and food products; thus, the most suitable active material can be selected for each type of food [32,59]. The compositions of polymeric composites and the food simulant determine the migration mechanisms between both phases. The chemical potential of each phase controls the diffusion coefficients and the equilibrium concentrations.
The release of polyphenols, normalized with respect to the TEPC, in the hydrophilic simulant was from 75% to 84%, in the lipophilic from 94% to 98% and in the acid from 52% to 66% ( Figure 5). The lipophilic medium showed the smallest difference in the final release.  The results of the polyphenols released in percent based on the TPL and the TEPC are reported in Table 4. As was expected, the polyphenols release based on TPL from PVA + BT NP and PVA + BT Aq + BT NP was lower than that based on TEPC for all food simulants. When CA was present in the mat (PVA + BT Aq + CA + BT NP ), the polyphenols release based on TPL tripled. This effect is due to the esterification/crosslinking reaction between CA and the -OH groups of the PVA chain as previously discussed. For all mats, the highest release occurred in the lipophilic simulant. Black tea nanoparticles were prepared in ethanol:water (70:30); therefore, they contain a large amount of ethanol-soluble polyphenols that can be released in a lipophilic simulant [17].
On the contrary, all mats showed a lower and more controlled polyphenol release in hydrophilic and acidic simulants (reaching its maximum at 72 h), indicating that they could be used as packaging for hydrophilic and acidic foods, providing them with a controlled release of polyphenols for a long time. The lowest release was observed in the acid simulant. This has been associated with the insolubility of nanoparticles in these simulants [15]. As it has been shown in Table 1, the TEPC and the antioxidant activity from PVA + BT Aq + BT NP are lower than PVA + BT NP and, comparing Figure 5a,b, higher release values between them are observed in the hydrophilic simulant. It might be related to the polyphenols' presence in the aqueous phase since they were more hydrophilic than those made from the ethanolic extract.
When citric acid is used (PVA + BT Aq + CA + BT NP ), the amount of released polyphenols was highest at 72 h in the lipophilic simulant, resulting in around 98% of the TEPC of this mat (being TEPC 31%, as discussed in Section 3.5, Figure 5c). This suggests that polyphenols in this mat are more soluble in a nonpolar simulant. In the hydrophilic simulant, the release is lower than the PVA + BT Aq + BT NP one. This could be related to a higher chemical affinity between the simulant and the polyphenols with the citric acid presence.
Furthermore, in the acid simulant, a faster release of polyphenols was observed ( Figure 5c). As previously discussed, CA can interact with both polyphenols and PVA, interrupting some interactions between polyphenols and PVA, thus, leaving free polyphenols to be more easily released.
Fick's diffusion law and Weibull's model were applied on all the release kinetics curves, and the Power Law model was only applied to the polyphenols release curves that met the method's condition (acid simulant). This method did not correctly fit the curves for the hydrophilic and lipophilic simulants as only five or fewer spots achieved release values less than 60%. The results (Table 5) suggested that Weibull's model is the appropriate one for all mats and Fick's diffusion model was only proper for the PVA + BT Aq + CA + BT NP mat in an acidic simulant. This could be explained by the close affinity between the simulant and the acidic mat. The value of the diffusion coefficient D = 3.3 ± 0.7 × 10 −15 cm 2 /s was higher than in the previous works with rosemary polyphenols extract [15], indicating a faster releasing process. Analyzing the b values of Weibull's model, all the mats' releases present a combined mechanism of a Fickian diffusion with a Case II transport. These results could indicate that polymer chain relaxation by the Case II transport is the principal mechanism in the polyphenol release since Fick's model was not adequate except for PVA + BT Aq + CA + BT NP in the acid medium. In this case, the Fickian diffusion mechanism in the combination could be the main one.

Conclusions
The investigation of ultralight and environmentally friendly active materials obtained through an emerging technique, such as electrospinning, is a very attractive alternative to replace petrochemical plastics in the packaging industry. In the present study, antioxidant electrospun mats were obtained by the solvent displacement technique, dripping ethanolic black tea extract onto three PVA solutions (water or aqueous extract of black tea, with and without CA). Our work demonstrated that the best strategy to maximize the total polyphenol content and the antioxidant activity of electrospun mats was to form black tea polyphenolic nanoparticles in a PVA-black tea aqueous extract solution (PVA + BT Aq + BT NP ), due to the synergy between the polyphenols of the nanoparticles and those from the PVA solution.
Regardless of the techniques developed in this work for the incorporation of black tea antioxidants in electrospun mats, all materials achieved the most controlled polyphenols release in hydrophilic and acid simulants. By applying Fick's diffusion law, and Peppas' and Weibull's models, on the release kinetics curves, it was shown that polymer chain relaxation was the principal mechanism in black tea polyphenol release.
This investigation provides an original and attractive strategy to achieve highly promising ultralight materials with controlled-released polyphenols for use as active and environmentally friendly food packaging.