Development of an Encapsulation Method for Trapping the Active Materials from Sour Cherry Biowaste in Alginate Microcapsules

This study aims to contribute to those valorization approaches for the recovery process of high-value-added substances in environmentally friendly ways. In this study, one of the most consumed juice products was selected for providing waste byproducts (peel). Sour cherry peels were subjected to automatic solvent extraction using a GRAS solvent (aqueous 80% ethanol, v/v). Then, encapsulation for the preservation of the related extract was performed by ionic gelation in alginate beads. The process conditions (gelling medium concentration, wall material concentration, and hardening time) were optimized by a Box–Behnken design (statistical experimental design approach). An almost 80% encapsulation efficiency was achieved under the proposed method (7.8% CaCI2, 1.3% alginate, and 26 min). The inhibition effect of the produced capsules against DPPH (2,2-diphenyl-1-picrylhydrazil) radicals also shows that the current products might represent potential alternative natural antioxidants for food formulations. The morphological properties were also measured.


Introduction
The valorization of biowaste derived from agricultural and food processing is a very important study in terms of the circular economy since the related residues possess various beneficial phytochemicals that minimize environmental damage [1]. Sour cherry pomace is a rich source of natural antioxidants [2]. A total of 25% of the total weight remains as residual waste from wine and juice processing [3]. Therefore, an investigation for the recovery of downstream processing is of great value to the production of high-value-added substances. On the other hand, these fine chemicals have been reported to represent alternatives as preservatives in food and cosmetic formulations [4][5][6]. Sour cherry pomace has been known to have a significant number of phenolic substances [7]. The dried extract powder contained 91.29 mg of total phenolic substance per gram of powder as gallic acid based on the report of Çilek et al. [8]. The main phenolic compounds in the extract are anthocyanins, such as cyanidin glycosides [9]. Its antioxidant activity was also comparable to that of butylated hydroxyanisole (one of the most consumed synthetic additives) and ascorbic acid [10]. However, the stability of these compounds should be improved due to the degradation problems when exposed to external influences, such as heat, light, humidity, and the acidity of the medium [11].
Ionic gelation can be seen as an option for an encapsulation method due to its simplicity, efficiency, and low cost [12]. There is no need to use complex systems like freeze-drying methods. Unlike spray drying, there is also no high-temperature application [13]. This is very valuable in terms of preventing the decomposition of the phenolic components, which are especially heat-sensitive components, during processing. In the presence of A commercial juice company (DİMES) located inİzmir (Turkey) supplied the sour cherry samples. The fruit samples were subjected to some physicochemical analyses, such as pH, total soluble solids content ( • Brix), and total titratable acidity (g-malic acid/100 g). The total soluble solids content was measured at 20 • C using a digital refractometer (Atago PAL-1 BLT/i, Tokyo, Japan), while pH analyses were performed with a pH meter (Orion 4-Star pH meter, Thermo Scientific, Waltham, MA, USA). Titratable acidity was determined by potentiometric titration with 0.1 N NaOH up to pH 8.1 [23]. The total soluble solids content of the fruits was 13 • Brix. The pH of the samples was 3.2. Total titratable acidity was measured as 1.5 g-malic acid/100 g. The peels of the samples were separated by hand before the fresh peels were used in the extraction.

Automatic Solvent Extraction
The extraction unit (Velp Scientifica, Usmate, Italy) was provided by the VELP ® society. The extraction process was performed with 80 mL of ethanol: water: 0.1 N HCI mixture (68:20:12, v/v/v). HCI was particularly added into the solvent since acidic solutions prevent the degradation of anthocyanin substances by balancing the anthocyanins in flavylium ion form [24]. The pH of the solution was~3. The amount of raw material to be extracted was 1 g. Total extraction time was 92 min, including immersion time (20 min), washing time (40 min), recovery time (30 min), and cooling time (2 min) [25]. Briefly, the fresh samples were put in cellulose soxhlet extraction thimbles (33 mm × 80 mm). Then, the thimbles were placed in glass cups containing the solvent. Expectedly, the level of solvent decreased since the remaining was recovered in a collection tank. The extract filtered from the cellulose thimble was not subjected to filtration.

Encapsulation Method
Alginate-based ionic gelation was used as the encapsulation method. Wall material (alginate) concentration was adjusted to 10 g/L, 15 g/L, and 20 g/L, respectively. On the other hand, the gelling medium (CaCl 2 ) concentration was adjusted to 20 g/L, 85 g/L, and 150 g/L, respectively. Then, the mixture of the core to the wall material (1/2, v/v) was mixed by vortex for 3 min. In total, almost 170 mL of extract and 340 mL of alginate solution were consumed. The mixture was added to the medium (250 mL CaCl 2 ) on a magnetic stirrer. Then, the capsules were left to be mixed in the CaCl 2 solution for several time periods (15,30, and 45 min) at ambient conditions. The agitation of the magnetic stirrer was 275 rpm.

Shape and Morphology of the Alginate Microcapsules
The morphology of the capsules was performed by using a stereo microscope (Stemi 2000-C, Zeiss, Germany) connected to an AxioCam digital camera. The capsule size was determined by using the Zeiss (Carl Zeiss Microscopy GmbH, Jena, Germany) program. Sphericity factor (SF) and the roundness (Rn) values of the capsules were calculated according to Equations (1) and (2) [26]: where d max is the maximum diameter, d min is the minimum diameter, P is the perimeter, and A is protection area of the capsules. Scanning electron microscopy (SEM) was used to determine the morphology of the capsules. The capsules were fixed onto the stubs of Zeiss LEO 1430 model SEM device. Coating treatment was applied with a thin layer of gold.

Physicochemical Analysis of the Alginate Microcapsules
Moisture content of the capsules was determined by drying the samples in an oven maintained at 70 • C for 24 h until a constant weight. Measurement of water activity was carried out by using a water activity meter (Testo 650 Water Activity System, Cole-Parmer, Vernon Hills, IL, USA). Additionally, the bulk density of the microcapsules was calculated by application of weight/volume ratio after approximately 2 g of powder was transferred to the 10 mL graduated cylinder [27].

Bioactivity Measurements
A total of 100 mg of microcapsules were dissolved in 3 mL of ethanol: acetic acid: water mixture (50:8:42, v/v/v). After mixing the mixture with a vortex for 1 min, the mixture was subjected to an ultrasonic bath at ambient conditions. Using the Folin-Ciocalteau method, 100 µL of the microcapsule mixture was taken into 2000 µL of Folin-Ciocalteau reagent (10%, v/v). After this mixture was kept in the dark for 5 min, 1800 µL of the sodium carbonate solution (7.5%, w/v) was added to the mixture and mixed. After the resulting mixture was kept in the dark for 1 h, the absorbance values were recorded at 765 nm in a UV spectrophotometer (PG Instruments, T60 / Leicestershire and England). Total phenolic content (TPC) was given in mg gallic acid equivalent (mg-GAE) per liter (ppm) [21].
In order to evaluate the efficiency of the microcapsules, the surface phenolic content (SPC) of the microcapsules was also calculated. A total of 100 mg of beads were dissolved in 3 mL ethanol-methanol solution (1:1, v/v) by keeping them in an ultrasonic bath (Protech, Istanbul, Turkey) for 5 min at ambient temperature [28]. The SPC results are also given in ppm units.
In the case of antioxidant activity, in vitro DPPH assay was adopted [28]. A total of 15 mg of microcapsules were dissolved in 3 mL ethanol:acetic acid:water (50:8 After mixing the mixture with a vortex for 1 min, the mixture was subjected to an ultrasonic bath at 40 • C. DPPH solution was added into the mixture. Then, the samples were kept in the dark for 1 h. The measurements were read at 517 nm. The results were given in mg trolox equivalent per gram dried microcapsule (mg-TEAC/g-DM).

Encapsulation Efficiency
Encapsulation efficiency (EE) shows the success of the microencapsulation method with quantitative findings. It is expressed by the following equation:

Experimental Design
Response surface approach was utilized to determine the impacts of the ionic gelation factors (A, B, and C) with their interactions (AB, AC, and BC) and quadratic (A 2 , B 2 , and C 2 ) terms, as well as performing the modeling and optimization studies. Box-Behnken design was selected as the subtype of the RSM. While encapsulation efficiency (Y 1 ) and antioxidant activity (Y 2 ) were the responses, the CaCI 2 concentration (A), alginate concentration (B), and hardening time (C) were ionic-gelation-process parameters. The relationship between A, B, and C was formed to estimate Y 1 and Y 2 . Design-Expert software (12.0.1.0) was used for the application of the RSM. Table 1 demonstrates the process factors, their levels, and the codes for the experimental design of the microencapsulation of active substances in sour cherry peels.

Microencapsulation of the Active Substances from the Sour Cherry Peels in the Alginate Microcapsules
Design-Expert software produced 17 experimental values with several combinations of process parameters, depending on the design type (Box-Behnken) of the RSM (Table 2). An encapsulation efficiency of ≈80% was accomplished by the proposed method. On the other hand, the free radical scavenging activity against the DPPH radical varied from 2.50 to 3.26 mg-TEAC/g-DM. Luca et al. reported 86.07-98.29 % EE for the encapsulation of the same extract with a freeze-drying method using different wall materials (gum Arabic and maltodextrin mixture) [29]. When soy protein and whey were used in freezedrying encapsulation method, the yields of each wall material were 94.9% and 90.1%, respectively [30]. It was also reported that EE with a 1/10 core to wall material changed between 69.38% and 77.83% [8]. Then, EE varied from 78.80% to 92.26% if the ratio was adjusted to 1/20. Oancea et al. encapsulated the sour cherry peel extract in whey proteins isolate with lower yield (almost 70% EE) [31]. Actually, it seems that the yield of the encapsulation depends on many factors such as the encapsulation method, wall material type, and the ratio of active material to wall material. As a simple, efficient and inexpensive method, ionic gelation might be a good option with the proposed yields [12].

Effects of the Ionic Gelation Parameters
Three-dimensional (3D) visuals can also be used to visually present the effect of the process parameters. These 3D images (Figures 1-3) were produced with Design-Expert software. As seen in Figure 1a,b, the EE and antioxidant activity of the microcapsules increased sharply with CaCI 2 concentration at first due to the fact that the Ca ++ ions are necessary to form the capsules. The strength of the gel is also enhanced by increasing the concentration [32]. However, both of the responses reduced visibly after around a 10% concentration for CaCI 2 (Figure 2a,b). A similar tendency of the gelling medium effect was also reported by Yousefi et al. during the encapsulation of Viola odorata Linn. extract by ionic gelation [33]. According to Halder et al. [34], the bivalent Ca ++ ions of the gelling-medium transferred to the beads, leading to the substitution of the active substance by the Ca ++ ions. Therefore, the yields of the encapsulation dropped by a higher degree for the CaCI 2 concentration. Alginate concentration also favored encapsulation at first. Then, it started to decrease, as seen in Figure 1. This outcome is in agreement with the earlier reports on the alginate-based ionic gelation of several active materials [33,[35][36][37]. Zam et al. explained this finding, with the increase in viscosity as the major reason for encapsulation efficiency [35]. Niizawa et al. also observed the same alginate effect on the encapsulation of astaxanthin in alginate microcapsules [38]. This was attributed to the occupation of the free spaces in the capsule, with excess alginate instead of the active material of the extract.
reports on the alginate-based ionic gelation of several active materials [33,[35][36][37]. Zam et al. explained this finding, with the increase in viscosity as the major reason for encapsulation efficiency [35]. Niizawa et al. also observed the same alginate effect on the encapsulation of astaxanthin in alginate microcapsules [38]. This was attributed to the occupation of the free spaces in the capsule, with excess alginate instead of the active material of the extract.      When the time for hardening is considered, EE was not significantly influenced, as seen in Figures 2a and 3a. Additionally, the antioxidant activity of the capsules increased with time until ~30 min, as seen in Figures 2b and 3b. Then, the activity against the free radicals began to drop with time. This is most probably because of the fact that the diffusion of the polar phenolic antioxidants into the polar gelling medium takes place due to the overexposure to the medium [39]. Similar time impacts were also observed during When the time for hardening is considered, EE was not significantly influenced, as seen in Figures 2a and 3a. Additionally, the antioxidant activity of the capsules increased with time until~30 min, as seen in Figures 2b and 3b. Then, the activity against the free Foods 2023, 12, 130 9 of 16 radicals began to drop with time. This is most probably because of the fact that the diffusion of the polar phenolic antioxidants into the polar gelling medium takes place due to the overexposure to the medium [39]. Similar time impacts were also observed during the encapsulation of cumin seed essential oil in alginate beads [27] and the encapsulation of yerba mate extract in alginate/chitosan microcapsules [40], respectively.

Modeling Study
The Box-Behnken design (as an independent quadratic design approach) was adopted for the present ionic gelation system. Since there are three factors with three levels, the Box-Behnken design was preferred. A total of 17 experimental runs with five replications at the center points (8.5% calcium chloride, 1.5% sodium alginate, and 30 min) were generated by the Design-Expert software. Figure 4 illustrates the classical quadratic designs for the current three factors of EE (Figure 4a) and antioxidant activity (Figure 4b), respectively. The experimental data were subjected to the given quadratic model (Equation (4)): Y represents the response variable (EE and antioxidant activity), while k is the number of independent factors (three in this process). β o is the constant, whereas β i , β ii , and β ij are the linear, quadratic, and interaction effects of the factors, respectively.
The derived second order models (depending on Equation (4)) are given by Equations (5) and (6): ANOVA (analysis of variance) was used to analyze the models statistically. Tables 3  and 4 are the ANOVA findings with a significance level of 5% for Y 1 and Y 2 , respectively. A model F-value of 24.59 means that the model is significant (Table 3), whereas a second model F-value of 14.67 implies the related model is significant, as seen in Table 4. On the other hand, a p-value < 0.05 denotes the significance of the model terms. In this instance, all the selected parameters and their second-order terms (A, B, C, A 2 , B 2 , and C 2 ) are found to be statistically significant, as seen in Table 3. Table 4 indicates that C, BC, A 2 , and C 2 are significant at p < 0.05 in the case of Y 2 . When the other model-fit statistics, such as the F-value, p-value, coefficient of variance (CV), coefficient of determination (R 2 ), adjusted R 2 , predicted R 2 , and lack of fit, are considered, both of the models seem satisfactory for the representation of the experimental finding. Lack-of-fit F-values of 0.65 and 0.04 imply the lack of fit values for both responses (Y 1 and Y 2 ) are not significant. CV values (4.36% and 2.78%) of less than 10% are desirable since CV represents the dispersion of the data around the mean [41]. So, the smaller the CV, the less variation around the mean. Furthermore, the predicted R 2 of 0.8065 is in compliance with the adjusted R 2 of 0.9299 (Table 3), whilst the predicted R 2 of 0.9020 is in conformity with the adjusted R 2 of 0.8849 (Table 4). It is desirable that the difference between the predicted R 2 and adjusted R 2 be less than 0.2. adopted for the present ionic gelation system. Since there are three factors with th levels, the Box-Behnken design was preferred. A total of 17 experimental runs with replications at the center points (8.5% calcium chloride, 1.5% sodium alginate, and 30 m were generated by the Design-Expert software. Figure 4 illustrates the classical quadr designs for the current three factors of EE (Figure 4a

Optimization and Validation Studies
An optimization study was performed under the ranges of each process parameter to achieve the maximum EE and antioxidant activity. Table 5 gives the optimum conditions with the maximum yields, with a 0.89 desirability. In order to verify the predicted conditions, experimental studies under the optimum conditions (7.8% CaCI 2 , 1.3% alginate, and 26 min) were performed. The difference between the actual and predicted EE and antioxidant values (less than 2%) shows the reliability of the outcome of this study. Figure 5a (EE) and Figure 5b (antioxidant activity) also demonstrate the reliability of the produced models when less dispersion around the regression line is considered.  Figure 5. Relationship between the actual and predicted data under optimal conditions (7.8% CaCI2, 1.3% alginate, and 26 min).

Morphology and Size of the Alginate Microcapsules
As spherical beads with a uniform size distribution are preferred for the delivery of active substances, the size and shape of the beads represent important parameters for food and pharmaceutical products [42]. The SF value was calculated via the ratio of the minimum capsule diameter to the maximum capsule diameter. This is mostly used for the teardrop and pear-shaped particles. While the SF value calculated by the ratio of the minimum and maximum diameters (to each other) is expected to be close to zero in a perfect sphere, it is also acceptable to be less than 0.05. Frequently, the Rn value indicates the spherical and elliptical characteristics of the particles. Although the Rn and SF values show how perfect the sphericity is, the Rn value is expected to be close to 1, contrary to the previously mentioned SF value [26]. The images of the alginate beads, including those from the sour cherry peel extract, are shown in Figure 6. The SF and Rn values were determined as 0.047 ± 0.017 and 1.126 ± 0.023, respectively. It is possible to say that the capsules are in an acceptable shape depending on the SF and Rn findings.

Morphology and Size of the Alginate Microcapsules
As spherical beads with a uniform size distribution are preferred for the delivery of active substances, the size and shape of the beads represent important parameters for food and pharmaceutical products [42]. The SF value was calculated via the ratio of the minimum capsule diameter to the maximum capsule diameter. This is mostly used for the teardrop and pear-shaped particles. While the SF value calculated by the ratio of the minimum and maximum diameters (to each other) is expected to be close to zero in a perfect sphere, it is also acceptable to be less than 0.05. Frequently, the Rn value indicates the spherical and elliptical characteristics of the particles. Although the Rn and SF values show how perfect the sphericity is, the Rn value is expected to be close to 1, contrary to the previously mentioned SF value [26]. The images of the alginate beads, including those from the sour cherry peel extract, are shown in Figure 6. The SF and Rn values were determined as 0.047 ± 0.017 and 1.126 ± 0.023, respectively. It is possible to say that the capsules are in an acceptable shape depending on the SF and Rn findings. As seen in Figure 7, the SEM images support the previous size findings. The capsules are smooth and homogeneous because of the bond between the alginate and phenolic substances [43].

Moisture Content, Water Activity, and Bulk Density of the Alginate Microcapsules
The microcapsules produced under optimum conditions (7.8% CaCI2, 1.3% alginate, and 26 min) were subjected to physicochemical analyses, such as moisture content, water activity, and bulk density. The moisture content and bulk density of the beads were 52.30 ± 3.01% and 0.54 ± 0.07 g/mL, respectively. The water activity of the samples was determined to be 0.684 ± 0.02. Gholamian et al. mentioned that the high water-absorbing capacity of alginate gave rise to the formation of a 3D network with high moisture content [27]. Additionally, the macropores of the alginate matrix caused a low bulk density for the hydrogels. On the other hand, the hydrophilic or lipophilic character of the core material influenced the moisture content and density of the hydrogels. As seen in Figure 7, the SEM images support the previous size findings. The capsules are smooth and homogeneous because of the bond between the alginate and phenolic substances [43].

Morphology and Size of the Alginate Microcapsules
As spherical beads with a uniform size distribution are preferred for the delivery of active substances, the size and shape of the beads represent important parameters for food and pharmaceutical products [42]. The SF value was calculated via the ratio of the minimum capsule diameter to the maximum capsule diameter. This is mostly used for the teardrop and pear-shaped particles. While the SF value calculated by the ratio of the minimum and maximum diameters (to each other) is expected to be close to zero in a perfect sphere, it is also acceptable to be less than 0.05. Frequently, the Rn value indicates the spherical and elliptical characteristics of the particles. Although the Rn and SF values show how perfect the sphericity is, the Rn value is expected to be close to 1, contrary to the previously mentioned SF value [26]. The images of the alginate beads, including those from the sour cherry peel extract, are shown in Figure 6. The SF and Rn values were determined as 0.047 ± 0.017 and 1.126 ± 0.023, respectively. It is possible to say that the capsules are in an acceptable shape depending on the SF and Rn findings. As seen in Figure 7, the SEM images support the previous size findings. The capsules are smooth and homogeneous because of the bond between the alginate and phenolic substances [43].

Moisture Content, Water Activity, and Bulk Density of the Alginate Microcapsules
The microcapsules produced under optimum conditions (7.8% CaCI2, 1.3% alginate, and 26 min) were subjected to physicochemical analyses, such as moisture content, water activity, and bulk density. The moisture content and bulk density of the beads were 52.30 ± 3.01% and 0.54 ± 0.07 g/mL, respectively. The water activity of the samples was determined to be 0.684 ± 0.02. Gholamian et al. mentioned that the high water-absorbing capacity of alginate gave rise to the formation of a 3D network with high moisture content [27]. Additionally, the macropores of the alginate matrix caused a low bulk density for the hydrogels. On the other hand, the hydrophilic or lipophilic character of the core material influenced the moisture content and density of the hydrogels.

Moisture Content, Water Activity, and Bulk Density of the Alginate Microcapsules
The microcapsules produced under optimum conditions (7.8% CaCI 2 , 1.3% alginate, and 26 min) were subjected to physicochemical analyses, such as moisture content, water activity, and bulk density. The moisture content and bulk density of the beads were 52.30 ± 3.01% and 0.54 ± 0.07 g/mL, respectively. The water activity of the samples was determined to be 0.684 ± 0.02. Gholamian et al. mentioned that the high waterabsorbing capacity of alginate gave rise to the formation of a 3D network with high moisture content [27]. Additionally, the macropores of the alginate matrix caused a low bulk density for the hydrogels. On the other hand, the hydrophilic or lipophilic character of the core material influenced the moisture content and density of the hydrogels.

Conclusions
The residuals from one of the most consumed juices were investigated in terms of the valorization of a rich source of natural antioxidants. In order to preserve the active material from many factors, an ionic gelation method was proposed using alginate microcapsules. A Box-Behnken design for the response surface method was adopted to develop a microencapsulation method for trapping the bioactive materials in the extract after the sour cherry peels were subjected to a green, environmentally friendly extraction method (automatic solvent extraction with aqueous ethanol). The optimum conditions (7.8% CaCI 2 , 1.3% alginate, and 26 min) were verified with a less than 2% error (the difference between the actual and predicted response values). All the fitting statistics (F-value (24.59 and 14.67), p-value (<0.05), lack of fit (>0.05), CV (<10%), R 2 (>0.94) adjusted R 2 (>0.88), and predicted R 2 (>0.80)) denoted the satisfactory model equations for the microencapsulation of sour cherry peel extract in alginate beads. As a result of the increasing importance of functional foods recently, the microencapsulation process has gained more meaning for the food industry. Considering that the current samples have antioxidant activity, these products might find application as a potential food additive. In this way, the physical properties of the related food product might be improved by preventing the possible bitter taste of the extracts from being used directly and the release of undesirable colors into the food. Moreover, the capsules are in an acceptable shape (smooth and homogeneous), considering that regular size dispersion is preferred for the transportation of active substances in food applications.