Valorization of spent black tea by recovery of antioxidant polyphenolic compounds: Subcritical solvent extraction and microencapsulation

Abstract Spent black tea (SBT), waste remaining after producing tea beverages, is potentially an underutilized source of antioxidant phenolic compounds. This study evaluated the integrated processes of subcritical solvent extraction of polyphenols from SBT followed by microencapsulation to improve the stability of obtained extract. Optimization of extraction conditions was carried out by response surface methodology for the best recovery of antioxidant phenolic compounds. Two variables [temperature (°C) and ethanol concentration (%)] were used to design the optimization model using central composite inscribed. Extraction temperature of 180°C and ethanol concentration of 71% were optimal for the highest yield of total polyphenols (126.89 mg gallic acid equiv./g SBT) and 2,2‐diphenyl‐1‐picrylhydrazyl scavenging activity (69.08 mg gallic acid equiv./g SBT). The extract was encapsulated using pectin, sodium caseinate, and a blend of these compounds (ratio 1:1) as wall materials by spray drying. The wall material significantly influenced (p < .05) encapsulation efficiency, particle size, morphology, thermal stability, crystallinity, and storage stability. The blend of wall materials produced an amorphous powder with the highest phenolic retention (94.28%) in the accelerated storage at 45°C for 40 days. The microcapsules prepared with sodium caseinate were smaller with lowest mean diameter and highest thermal stability than the other types of materials. Obtained microencapsulates have potential use in different food systems to enhance their antioxidant property.

Because of its convenience of consumption, ready-to-drink (RTD) or bottled tea is produced commercially in most parts of the world, in particular, Japan. As a consequence, a large amount of spent black tea (SBT), the residues after manufacturing this tea product, is generated annually (Kondo, Hirano, Kita, Jayanegara, & Yokota, 2018). These used tea leaves mostly become waste with only a small percentage used as feedstock or turned to compost (Sagar, Pareek, Sharma, Yahia, & Lobo, 2018). As the conditions used for tea brewing are mild, a significant amount of polyphenols with a high antioxidant power is retained in SBT (Abdeltaif, SirElkhatim, & Hassan, 2018). Hence, utilizing food manufacturing waste such as SBT is a sustainable and economically attractive way to recover antioxidant phenolic compounds but an efficient method for extracting polyphenols from SBT has not yet been developed and tested.
Several techniques have been investigated for extracting phytochemicals from food waste (Sagar et al., 2018;Yanagida, Shimizu, & Kimura, 2005). Of these techniques, subcritical solvent extraction (SSE) is a greener and faster method (Munir, Kheirkhah, Baroutian, Quek, & Young, 2018), which uses a pressurized liquid kept below its critical point (374°C for water) and above its boiling point (100°C for water). These conditions allow fluids to remain in a liquid state due to the applied pressure and it creates low polar water with equivalent to organic solvents at ambient temperature (Shimizu, Ushiyama, & Itoh, 2019;Zhang, Baroutian, Munir, & Young, 2017;Zhang & Wolf 2019). This technique facilitates rapid extraction without the loss or changing the chemical integrity of thermolabile compounds (Essien, Young, & Baroutian, 2020). Combining subcritical water with an organic solvent such as ethanol and methanol has also been used to improve the yield, extraction time and solubility of compounds (Kwon & Chung, 2015;Pronyk & Mazza, 2009). Response surface methodology (RSM) is commonly used for optimizing the process parameters for the extraction of phytochemicals. This is a useful mathematical and statistical tool for defining the effect of independent variables and their interactions on a particular response, such as the yield.
However, effectiveness of polyphenols mainly depends on their stability, bioactivity, and bioavailability. The unsaturated bonds in the molecular structure of polyphenols make them vulnerable to oxidants, light, and heat, thus reducing their activity (Kailaspathy, 2015).
Therefore, protecting phenolic compounds by encapsulation following their extraction would be a better way to maintain the structural integrity of polyphenols until their industrial application. The microencapsulation of phenolic extracts not only preserves them but also produces a powdered product that is convenient for food application. At present, spray drying is the most widely used technique for the microencapsulation of polyphenols and other heat labile compounds because of its short thermal contact time, cost-effectiveness, and suitability for industrial application.
As well as the technique used for microencapsulation, selecting a coating or wall material is also crucial for efficient spray drying (Ushiyama & Shimizu, 2018). Of the different types of wall material, polysaccharides and protein agents are commonly used either alone or in combination because of their distinct properties. Pectin is a polysaccharide with strong film-forming, gelling, and binding abilities. Its ability to form stable dispersions at low concentrations facilitates microencapsulation by spray drying (Rehman et al., 2019).
Sodium caseinate is the salt of casein, a major milk protein fraction.
Generally, milk proteins act as effective film-formers and emulsifiers while polysaccharides act as filler materials (Augustin & Oliver, 2014).
Ultimately, encapsulated SBT would exhibit important properties that would facilitate the shelf life of polyphenols because the coating materials can act as a barrier against adverse environmental conditions. Therefore, the objective of the present study is to develop a method for recovering phenolic compounds from SBT by integrating the processes of SSE optimization and the subsequent encapsulation of SBT using different wall materials by spray drying. The encapsulated powder using pectin, sodium caseinate, and a mixture of these compounds as wall materials will be characterized to evaluate their encapsulation efficiency, morphology and size, thermal stability, crystallinity, and storage stability.

| MATERIAL S AND CHEMI C AL S
Low grown unblended black tea was supplied by Nawa withana Kanda tea factory in Sri Lanka, gallic acid by Sigma-Aldrich Distilled water was used in all the experiments. All other chemicals and solvents used were analytical grade.

| Preparation of spent black tea
Black tea leaves (20 g) were brewed in 1,000 ml of boiling water (100°C) for 6 min. The infusion was then filtered using a tea strainer, and the residue was dried in an air-drying oven at 45°C overnight.

| Subcritical solvent extraction
Subcritical solvent extraction was performed using an organic synthesizer (Chemi-station PPV 3000, Tokyo Rikakikai Co. Ltd, Tokyo, Japan) with an agitator and an 11-ml reactor with a maximum temperature and pressure of 200°C and 5 MPa, respectively. For each experimental run, 0.5 g of SBT was mixed with 10 ml of solvent at a solid: solvent ratio of 1 g: 20 ml. The extraction reactor was filled and then purged three times with nitrogen gas to remove the atmospheric oxygen present in the reactor vessel, and then, an initial pressure of 2.0 MPa was applied. The heating control was adjusted to obtain the desired temperature, which was then maintained for 10 min. During extraction, the agitation speed was kept at 17 g to prevent any local overheating and to increase the mass transfer. The extraction process was conducted in at various ethanol concentration (0%-100%) and temperature (100°C-180°C) ranges, based on the RSM design given in Table 1. After the extraction, the reactor was immediately cooled by placing it in a container of cold water. The extracts were filtered through filter paper (6 μm) under vacuum after the vessel pressure reached the initial pressure, and then, the filtrate was lyophilized. The lyophilized powder of SBT extract was stored at 4°C until further analysis.

| Determination of total phenolic content
The TPC of the dried extract was measured colorimetrically using the Folin-Ciocalteu (FC) method described by Dranca and Oroian (2016) with little modifications. Briefly, the dried extract obtained was diluted with a dilution factor of 100, and then, a 1.0-ml aliquot of the extract in triplicate was transferred into a test tube and mixed thoroughly with 5.0 ml of FC reagent diluted 1:10 with distilled water.
After keeping for 3 min, 5.0 ml of sodium carbonate (7.5%, w/v) was added and mixed. The mixtures were then allowed to stand for 1 hr in the dark before measuring the absorbance using a UV-Vis spectrophotometer (JASCO V-560, JASCO corporation, Tokyo, Japan) at 756 nm against the blank. Gallic acid was used as the standard for preparation of the standard curve (7.812-250 μg/ml, R 2 = .998). The TPC values were expressed as milligrams of gallic acid equivalent/g (dry weight) material (mg GAE)/g SBT).
DPPH scavenging capacity was expressed as milligrams of gallic acid equivalent/g (dry weight) material (mg GAE)/g SBT).

| Response surface methodology design
The study was conducted as a two-factor full factorial experiment with the influence of two independent variables (temperature and ethanol concentration) on the responses (total phenolic content and DPPH scavenging capacity) being evaluated ( Table 2). The CCI design consisted of 13 experiments using 5 centers, 4 axial, and 4 factorial points. The experimental data obtained were fitted to a second order polynomial model of the form: where y is the predicted values of TPC or DPPH scavenging capacity; x i , the coded levels of the design variables (Temperature and ethanol concentration); b o , a constant; b i , the linear effect; b ii , the quadratic effect; and b ij , interaction effects.
The statistical significance of differences between the mean values of variables was determined at the 5% probability level (p < .05), and the data were analyzed by ANOVA. Minitab 19.1.1 software (Minitab Inc., State College, PA, USA) was used to generate the surface plots and the optimized conditions. All assays for characterizing the SBT extract were performed in triplicate.

| Preparation of feed solutions
The coating materials (3 g) were dissolved in 100 ml of distilled water at 90°C and then stirred until a clear dispersion was achieved. Three coating materials were evaluated as follows: 100% pectin (PE), 100% sodium caseinate (SCN), and a 50%:50% combination of pectin and sodium caseinate (PE + SCN). The prepared PE solution was kept at room temperature while the SCN and PE + SCN solutions were kept in a refrigerator overnight to allow complete hydration to occur. The next day, SBT extract concentrated by a rotary evaporator (N-1210 and SB-1300 water bath, EYELA Tokyo Rikakikai Co., Ltd, Tokyo, Japan) was added dropwise to the prepared biopolymer solutions heated to 40°C with magnetic stirring at 21.5 g for 20 min. The prepared feed solutions were sonicated for 20 min then homogenized (HERACLES-16g, Koike Precision Instruments, Tokyo, Japan) with stirring for 30 min before further processing. The feed solution contained 20 g of the carrier solution and 1 g of the concentrated SBT extract. All the prepared feed solutions were then spray dried.

| Measurement of viscosity
Before spray drying, the viscosity of all feed solutions was measured using a Sine-wave Vibro Viscometer SV-10 (A&D Co. Ltd., Tokyo, Japan). All measurements were carried out at room temperature.
Each experiment was performed three times, and the average value was taken as the final value.

| Spray drying conditions
The liquid feeds were spray dried using a laboratory scale spray dryer OSK 55MO102 (Osaka Seimitsu Kikai Co. Ltd, Osaka, Japan).
The values of the operational parameters established for the drying process were as follows: solid concentration, 3% (g/g); inlet air tem-

| Encapsulation efficiency
The encapsulation efficiency (EE%) of the powders was calculated as: where TPC is the total phenolic content, and SPC is the surface phenolic content (Kaderides & Goula, 2019).
TPC was determined by dissolving 10 mg of the sample in 4 ml of ethanol and methanol followed by thorough agitation and sonication for 40 min to completely break down the microencapsulates. Then, the solution was filtered through a 0.45-μm filter. SPC was measured by washing a powder sample (10 mg) into a filter paper (0.45 μm) using 4 ml of ethanol and methanol.

| Morphology and particle size analyses
The morphology of the particles was examined using scanning electron microscopy (SEM, JSM-6301F, JEOL Ltd., Tokyo, Japan) at a beam voltage of 10 kV and a working distance of 39 mm. From the micrographs, the particle diameter was calculated using ImageJ open-source software (imagej.net). For measuring the size distribution, 100 particles were counted.

| Thermogravimetric analyses
Thermogravimetric analyses (TGA) were carried out using a

| Crystallinity of powders
The crystallinity of the encapsulated phenolic extract was evaluated using a Rigaku Rint-Ultima III X-ray diffractometer (Rigaku Corp.) with Cu-Kα radiation generated at 40 kV/40 mA at a wavelength of 0.154187 nm. The scanning range was 5-40° 2θ at a speed of 2° 2θ/min. The degree of crystallinity was calculated as described by Ahmadian, Niazmand, and Pourfarzad (2019): where CD is the degree of crystallinity (%); I net , the crystalline intensity of peaks; and I total , the overall intensity.

| Accelerated storage stability study
Glass vials containing three types of microencapsulated powders loaded with SBT extract (20 mg) were stored in an incubator at 45°C for 40 days. After given periods (0, 20, and 40 days), samples were collected from each batch, and then, their retained phenolic content was determined.

| Statistical analysis
Statistical analyses were carried out using Minitab 19. The results were reported as the mean value of three repeated experimental data.  (Table 3).
using water extraction to be low, or possibly water alone cannot cleave hydrogen bonds (Miralai, Khan, & Islam, 2008). However, ethanol can precipitate polysaccharides and expel them from the solution (Xu et al., 2014). The interactive effect of two variables also showed a positive effect on the TPC. The polynomial equation obtained for the apparent phenolic content was as follows:

| Effect variables on antioxidant activity
The antioxidant activity was determined by the DPPH assay. Like

| Optimization of extraction process and experimental validation
The SSE was optimized to yield an extract with a high content of phenolic compounds and a high antioxidant activity. A graphical optimization based on the effect of the two factors on the responses was conducted using the highest desirability level ( Figure 3). This shows that under the optimal conditions (180°C temperature and 71% ethanol concentration), a TPC of 126.89 mg GAE/g SBT and DPPH activity of 69.08 mg GAE/g SBT can be obtained. These optimal conditions were used later for validation, and the results obtained for TPC (127.15 ± 1.67 mg GAE/g SBT) and DPPH activity (71.31 ± 3.40 mg GAE/g SBT) were very close to those predicted. Under the optimal conditions, the phenolic content and antioxidant activity of the raw and spent black tea were compared (Figure 4). The results showed that only 33% of the antioxidant phenolic compounds had been lost during tea brewing while the remaining 67% could be recovered using the optimized conditions for SSE. The optimal values of TPC from SSE in the present study were considerably higher than that (91.06 mg GAE/g SBT) reported in a previous study that used the maceration-mediated liquid-liquid extraction of SBT (Mukhtar, Mushtaq, Akram, & Adnan, 2018).

| Particle size analyses
The particle size and size distribution are important characteristics of powder that can affect their storage and handling. To evaluate the effect of the type of wall material, the average particle diameter and size distribution were measured (Table 4). This revealed that the type of wall material significantly influenced the diameter of the particles (p < .05). The standard deviation values of the powder diameter  (Table 4). The solution with pectin possessed the highest viscosity and sodium caseinate the lowest. The particles with the mixture of wall materials exhibited the narrowest size distribution (1.45 μm) so were of a more uniform size than particles coated with the other two wall materials. This could have been caused by the heating of the pectin-sodium caseinate blend during its preparation, thus decreasing the polydispersity index by forming a more compact and dense particulate structure F I G U R E 3 Optimization plot for the responses of total phenolic content (TPC, mg GAE/ g SBT) and antioxidant activity (DPPH, mg GAE/ g SBT)

F I G U R E 4
Comparison of the extract from raw black tea (RBT) and spent black tea (SBT) obtained under optimal conditions (180°C temperature and 71% ethanol concentration)

TA B L E 4
Effect of different wall materials on the viscosity of the feed solutions before spray drying, average particle diameter, and encapsulation efficiency of spray dried spent black tea powders through re-arranging the pectin molecules on the sodium caseinate surface (Liang & Luo, 2020).

| Encapsulation efficiency (EE%)
The encapsulation efficiency of SBT powder with coatings varied between 60.06% and 81.55%, thus confirming the successful entrapment of phenolic compounds within the wall materials (Table 4). The results also revealed that the type of coating agent used for encapsulation had an important role in retaining phenolic compounds in the carrier matrix (p < .05). The best result was achieved by the sample with SCN + PE (1:1 mass/mass). Pectin offers the advantageous as a protective carrier and its capability of interacting with hydrophobic molecules (Rehman et al., 2019). However, carbohydrates such as pectin mostly lack the interfacial functionality (Livney, 2010). Hence, it is better to combine polysaccharides with surface active biopolymer-like caseinate to achieve successful encapsulation (Hogan, McNamee, O'Riordan, & O'Sullivan, 2001). Nevertheless, milk protein such as sodium caseinate can bind with polyphenols, especially with catechin, with this mechanism being promoted by a preheating treatment (Haratifar & Corredig, 2014;Shpigelman, Israeli, & Livney, 2010).

| Morphology
SEM was used to observe the microstructure of the encapsulated SBT extracts ( Figure 5). Figure

| Thermal stability
The thermogravimetric (TG) and derivative thermogravimetric (DTG) curves obtained from microparticles allow the visualization of their thermal degradation behavior (Figure 6a). All the tested samples ex- This was possibly because, with its flexible and disordered structure, it is less sensitive to changes in temperature (McClements, 2018). Figure 6a shows that the rate of weight loss during decomposition was lowest in the particles coated with a combination of wall materials possibly because of the enhanced electrostatic interaction between pectin and sodium caseinate (Chang et al., 2017). Pectin is an anionic polysaccharide that can interact with casein mainly through electrostatic, steric, or covalent interactions.

| Crystallinity
The XRD patterns of the encapsulated SBT extracts coated with three polymer mixtures are shown in Figure 7. The results of XRD were evaluated after smoothing by using Match! Software (Crystal Impact, Bonn, Germany). Normally, a crystalline fraction diffracts X-ray coherently according to Bragg′s law to give a sharp peak while an amorphous fraction diffracts incoherently to give a diffuse halo (Chung, 1999). Thus, in Figure 7, the XRD patterns with their wide in the particles coated with a mixture of wall materials. These results agreed with XRD results in previous studies on pectin (Hosseinnia, Khaledabad, & Almasi, 2017) and sodium caseinate (Pan, Zhong, & Baek, 2013). Forming of an amorphous structure enhances the rate of dissolution and solubility compared with a crystalline material thereby increasing the bioavailability of the active material (Kanaujia, Poovizhi, Ng, & Tan, 2015).

| Accelerated storage stability
The stability of the encapsulated SBT extract was evaluated Values represent the mean (n = 3). Different letters in each coating material indicated a statistically significant difference (p < .05) between time points composite inscribed design could be used for the subcritical solvent extraction of phenolic compounds in spent black tea, and the subsequent microencapsulation could enhance the stability of the polyphenols. Therefore, antioxidant phenolic compounds can be effectively recovered from spent black tea a food manufacturing waste product with subsequent microencapsulation turning it into valuable food ingredient.

ACK N OWLED G M ENTS
Author gratefully acknowledges the help of Prof. Kiyonori Takahashi, Research Institute for Electronic Science, Hokkaido University for granting the use of the X-ray diffraction facility. The TGA analysis was performed using TG-DTA analyzer at the Global Facility Centre of Hokkaido University, and the scanning microscopy was provided by the Electron Microscope Laboratory in the Research Faculty of Agriculture, Hokkaido University. We thank Philip Creed, PhD, from Edanz Group (www.edanz editi ng.com/ac) for editing a draft of this manuscript.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

E TH I C A L A PPROVA L
Ethical approval is not required for this research.

DATA AVA I L A B I L I T Y S TAT E M E N T
The research data are not shared.