Response Surface Methodology-Aided Optimization of Bioactive Compound Extraction from Apple Peels Through Pulsed Electric Field Pretreatment and Ultrasonication

Abstract

Apple by-products (i.e., peels) are often thrown away, yet they are highly nutritious and provide numerous advantages as they contain a variety of nutrients such as vitamins, minerals, and antioxidants. Apple peels also comprise a high level of antioxidants, particularly polyphenols and flavonoids. This research aimed to determine the most efficacious extraction techniques and parameters to accomplish maximum bioactive compounds recovery from apple peels. Several extractions were conducted, including stirring, ultrasonication, and pulsed electric field-assisted extractions. Response surface methodology and several factors such as temperature, extraction duration, and solvent composition were considered to have a major impact on the isolation of bioactive compounds. The findings indicated that the most practical and efficient approach was to combine the pulsed electric field process with ultrasonication and stirring at 80 °C for 30 min, while 75% aqueous ethanol comprised the optimal solvent concentration, demonstrating the critical role of the solvent in optimizing extraction efficiency. The optimal conditions were obtained through response surface methodology with a statistical significance of p < 0.05. The extract exhibited a total polyphenolic content (TPC) of 17.23 mg gallic acid equivalents (GAE) per g of dry weight (dw), an ascorbic acid content (AAC) of 3.99 mg/g dw, and antioxidant activity of 130.87 μmol ascorbic acid equivalents (AAE)/g dw, as determined by FRAP and 95.38 μmol AAE/g dw from the DPPH assay. The measured antioxidant activity highlighted the significant potential of apple peels as a cost-effective source of exceptionally potent extracts.

1. Introduction

The apple (Malus domestica Borkh) is a globally renowned and extensively cultivated fruit [1]. It is a member of the Rosaceae family and produces a vast array of goods, including juices, jams, compotes, tea, wine, and dried apples [2]. Currently, the cultivated apple ranks as the third most widely cultivated fruit crop globally [3]. In 2018, the Food and Agriculture Organization (FAO) reported a total apple production exceeding 86 million tons [4], of which 11 million were generated in the Mediterranean area [5]. Apple peels are commonly regarded as waste materials in the canned apple and apple sauce manufacturing sectors, despite their untapped potential and possible health advantages. Therefore, there is a need to improve the exploitation of these peels [6]. Apple peels include five primary categories of polyphenolic chemicals (Figure S1), including phenolic acids (mostly chlorogenic acid), flavan-3-ols (specifically (+)-catechin, and (−)-epicatechin), flavonols (primarily various quercetin glycosides), flavanones (specifically hesperidin), and anthocyanins (mostly pelargonin chloride) [7]. Monomeric and polymeric flavan-3-ols compounds provide a significant proportion (about 60%) of the overall polyphenol content found in apple peels. In contrast, flavonols, hydroxycinnamic acids, dihydrochalcones, and anthocyanins contribute to 18%, 9%, 8%, and 5% of the total polyphenol content, respectively [7].

Utilizing by-products as additives for the development of new goods could reduce waste within the food processing sector [8]. Moreover, wastes usually exhibit notably elevated levels of polyphenols [9], which could effectively be used as natural antioxidants for humans. Thus, antioxidant and antibacterial compounds found in apple peels, such as ferulic acid, dopamine, and caffeic acid, could make them suitable for food preservatives. The appeal of apple peel lies in its antioxidant characteristics, which have the potential to facilitate the creation of innovative goods. These goods may consist of either human-consumable foods that can improve the health of the human body or animal feed with comparable characteristics [10]. Furthermore, apple peels could be employed to enrich apple juices with bioactive and antioxidant compounds.

Over the past few years, efforts have been undertaken by industries and researchers to apply green extraction techniques on waste, due to ecological sustainability. Green technologies such as pulsed electric field (PEF) and ultrasonication (US), generally, necessitate lower energy consumption in comparison to conventional processes, such as stirring [11]. PEF is a highly efficient method that has multiple benefits. It helps inactivate microorganisms, improving mass transfer in food products and recovering valuable bioactive compounds from food waste [12]. PEF has gained significant popularity among different food industries in recent years, offering extra motivation for businesses to minimize waste and the resulting harm to the environment [13]. Ultrasonication, on the other hand, is also a green technique that has several advantages, including reduced extraction duration, less energy and power consumption, minimized bioactive thermal degradation, and the production of extracts characterized by their high quality [14]. Tian et al. [15] reported a total polyphenol content of ~42 mg/g extracted from apple peels utilizing 75% methanol as a solvent and 20 min ultrasonication in triplicate. Furthermore, Sethi et al. [16] determined polyphenol content at a range of 708–1232 mg GAE/100 g dw and ascorbic acid content ranging from 6.81 to 27.68 mg/100 g dw from the peels of twelve different apple cultivars. Fotirić Akšić et al. [17] also reported a total polyphenol content ranging from ~3 to 21 mg GAE/kg dw in apple peels. Therefore, studying the optimal extraction method is important because apple peels contain sufficient bioactive compounds.

Green extraction techniques may be employed either as a pretreatment or as an exclusive extraction technique in order to recover various bioactive compounds from plant sources in a sustainable manner [18]. Moreover, green techniques might enhance the efficiency of the extraction, when compared to conventional ones. Recently, an increasing number of environmentally conscious extraction techniques have been applied to eliminate toxic organic solvents and reduce extraction duration and energy use [19]. PEF has been utilized in various food waste and by-products, such as orange peels, lemon peels, olive leaves and kernels, and many others [12], resulting in increased polyphenolic and antioxidant compound yields. Although apple peels have numerous advantages for humans, there is insufficient research concerning the environmentally sustainable production of extracts rich in antioxidant polyphenols.

The main objective of this investigation was to determine the most efficient combination of green pretreatment techniques for the production of extracts that are abundant in antioxidant compounds, including ascorbic acid and polyphenols. These extracts are a highly promising alternative for the food and pharmaceutical sectors. Although the extraction of bioactive and antioxidant compounds from apple peels has been extensively studied, to the authors’ best knowledge, not enough attention has been devoted to combining green pretreatment techniques to maximize yield in specific byproducts. Even though PEF [20] and US [21] have been employed in apple peels, the combination of the two has not been studied yet. Moreover, it is important to emphasize that environmentally friendly solvents are used to maintain an environmentally conscious profile, free of toxic organic solvents. The optimization of the extraction was accomplished through response surface methodology (RSM). The study focused on investigating the effects of green solvent combinations comprising water and ethanol, as well as the temperature impact and extraction duration on the extraction process. In addition to conventional extraction through stirring (ST), green sample pretreatment techniques, including PEF and US, were employed to further optimize this procedure. Moreover, a partial least squares (PLS) model was utilized to determine the optimal conditions.

2. Materials and Methods

2.1. Chemicals and Reagents

Iron (III) chloride was obtained from Merck (Darmstadt, Germany). Trichloroacetic acid, hydrochloric acid, methanol, aluminum chloride, 2,2-diphenyl-1-picrylhydrazyl (DPPH), L-ascorbic acid, 2,4,6-tris(2-pyridyl)-s-triazine (TPTZ), and all polyphenolic standards for the HPLC determination were bought from Sigma-Aldrich (Darmstadt, Germany). Gallic acid, Folin–Ciocalteu reagent, and ethanol were procured from Panreac Co. (Barcelona, Spain). Finally, anhydrous sodium carbonate was bought from Penta (Prague, Czech Republic). A deionizing column was used to generate deionized water, which was used for the conducted experiments.

2.2. Instrumentation

The lyophilization of apple peels was performed using a Biobase BK-FD10P freeze-dryer (Jinan, China). After extraction, the centrifugation was carried out using NEYA 16R (Remi Elektrotechnik Ltd., Palghar, India) to separate the supernatant from the solid residue. For PEF processing of the samples, we used a UPG100 mode/arbitrary waveform generator (ELV Elektronik AG, Leer, Germany), a Leybold high-voltage power generator (LD Didactic GmbH, Huerth, Germany), a Rigol DS1052E digital oscilloscope (Beaverton, OR, USA), and two custom stainless-steel chambers (Val-Electronic, Athens, Greece). To conduct US pretreatment, an Elmasonic P70H ultrasonication bath (Elma Schmidbauer GmbH, Singen, Germany) was utilized. Spectrophotometric analyses were conducted with a Shimadzu UV-1900i PharmaSpec spectrophotometer (Kyoto, Japan). Finally, to perform chromatographic analyses and quantify the identified individual polyphenols, we used a Shimadzu CBM-20A liquid chromatograph with a Shimadzu SPD-M20A diode array detector (DAD) (Shimadzu Europa GmbH, Duisburg, Germany). The separation of targeted compounds occurred in a Phenomenex Luna C18(2) column (Phenomenex Inc. Torrance, CA, USA) maintained at 40 °C (100 Å, 5 μm, 4.6 mm × 250 mm).

2.3. Sample and Extract Preparation

Red apple (Malus domestica) fruits from the ‘Starking Delicious’ cultivar were purchased in Karditsa, Greece, at a neighborhood store. The apples were thoroughly rinsed with running tap water and then dried using paper towels. At first, the apple peels were cut off with a blade, and then manually chopped into small pieces and lyophilized. The peels were finely ground to a diameter of less than 400 μm and then stored in the freezer at −40 °C, pending further analysis.

2.4. Plant Extraction Procedure

Various extraction combinations of different pre-treatment techniques were employed to determine the best conditions for recovering antioxidant compounds from apple peels. For each experiment, an amount of 1 g of dried apple peel powder was mixed with solvent at a solid-to-liquid ratio of 1:20 g/mL. The solvents used ranged from 0% to 100% v/v aqueous ethanol. PEF and US techniques were employed to enhance the conventional stirring extraction process. Once these techniques were used, the peel powder was hydrated by including the suitable solvent and allowed to sit for 10 min. The sample underwent a 20-min treatment with PEF or US individually. When both techniques were used together, the sample received a 20-min PEF treatment, then a 20-min US treatment. Ultimately, all samples underwent a conventional stirring extraction process with the use of a stirring hotplate. For the samples that demanded PEF treatment, an amount of 1.0 kV/cm of electric field strength was applied, with a pulse length of 10 μs within a pulse period of 1 ms (frequency: 1 kHz). For the US treatment, the equipment operated at a nominal frequency of 37 kHz with the temperature being stable at 30 °C. PEF and US conditions were chosen according to our previous methodology [19].

The mixtures comprising the appropriate solvent and apple peels were placed in screw-capped glass bottles and subjected to temperatures that ranged from 20 to 80 °C for a duration varying between 30 and 150 min while being stirred continuously at 500 rpm. The sample extracts were centrifuged for 10 min at 10,000× g after the extraction process was finished. The supernatants were gathered and kept refrigerated at −40 °C until further analysis. The extraction was performed using various combinations of the analyzed parameters, which are represented by the coded levels in

Table 1. The independent variables with the respective actual and coded levels were utilized for the optimization process.

Independent Variables	Code Units	Coded Variable Level
		1	2	3	4	5
Technique	X1	ST	PEF + ST	US + ST	PEF + US + ST	–
C (%, v/v)	X2	0	25	50	75	100
t (min)	X3	30	60	90	120	150
T (°C)	X4	20	35	50	65	80

.

2.5. Experiment Design and Optimization Using Response Surface Methodology (RSM)

The RSM method was utilized to assess the antioxidant capacity and quantity of bioactive compounds in apple peel extracts. The extraction of targeted bioactive compounds (i.e., individual polyphenols and ascorbic acid) could be enhanced through an optimization process made by RSM. This optimization could probably increase the antioxidant activity of the extracts. The optimization involved refining key extraction parameters including the extraction method (i.e., conventional stirring with or without pretreatment green techniques such as PEF and US), adjusting the ethanol-to-water ratio (C, % v/v), and adjusting the extraction duration (t, min) in several temperature ranges (T, °C). The Main Effect Screening design of the experiment was employed for that reason, with 20 design points being examined. The process variables were set at five levels according to the experimental design. A minimum confidence level of 95% using analysis of variance (ANOVA) and summary-of-fit tests was used to assess the overall model significance (R², p-value) and the significance of the model coefficients. The following second-order polynomial model was employed to predict the response variable as a function of the independent variables under investigation:

$${\mathrm{Y}}_{\mathit{k}}={\beta }_0+\sum _{\mathit{i}=1}^2{\beta }_{\mathit{i}}{\mathit{X}}_{\mathit{i}}+\sum _{\mathit{i}=1}^2{\beta }_{\mathit{ii}}{\mathit{X}}_{\mathit{i}}^2+\sum _{\mathit{i}=1}^2\mathrm{\sum }_{\mathit{j}\mathit{=}\mathit{i}+1}^3{\beta }_{\mathit{ij}}{\mathit{X}}_{\mathit{i}}{\mathit{X}}_{\mathit{j}}$$

(1)

where the intercept and regression coefficients for the initial, linear, quadratic, and interaction terms are denoted by β₀, β_i, β_ii, and β_ij, respectively. Terms X_i and X_j represent the independent variables, while Y_k denotes the expected response variable.

The largest peak area was determined through RSM wherein the effect of a significant independent variable on the response was also assessed. The results from the above model equation were visually represented through 3D plots, which illustrate surface response graphs.

2.6. Bioactive Compounds Quantification

2.6.1. Determination of Total Polyphenol Content (TPC)

The TPC determination procedure was based on a previously established methodology [22]. In summary, 0.2 mL of sample volume was combined with 0.2 mL of Folin–Ciocalteu reagent. Right after 2 min, 1.6 mL of a sodium carbonate aqueous solution (5% w/v) was inserted into the mixture in a 2 mL Eppendorf tube. The incubation process at 40 °C for 20 min was followed for the mixture with its absorbance being measured at 740 nm. A calibration curve of gallic acid was used to determine the concentration of total polyphenols (C_TP), in which TPC was expressed as mg of gallic acid equivalents (GAE)/g dw, calculated using the following equation:

$$\mathrm{TPC} (\mathrm{mg} \mathrm{GAE}/\mathrm{g} \mathrm{dw})={\displaystyle \frac{C_{\mathrm{TP}} \text{×}V}{w}}$$

(2)

where V denotes the total volume (in L) of the extraction medium, and w represents the dry weight of the sample (in g).

2.6.2. Chromatographic Polyphenol Quantification

An established methodology from our previous research [23] involving high-performance liquid chromatography (HPLC) was used to identify and quantify individual polyphenols from the apple peel extracts. Mixtures of formic acid constituted the mobile phase, since 0.5% aqueous formic acid (A) and 0.5% formic acid in acetonitrile (B) were used. The gradient program was set to start from 0 to 40% B, increase to 50% B over 10 min, then to 70% B in another 10 min, and hold steady for 10 min. We kept the flow rate stable at 1 mL/min. The identification of targeted compounds was made through the comparison of their absorbance spectra and their respective retention times with pure analytical standards, using calibration curves ranging from 0 to 50 mg/L of excellent linearity (>0.99).

2.6.3. Determination of Ascorbic Acid Content (AAC)

AAC analysis was carried out following a protocol previously described in reference [24]. Briefly, a mixture of 0.1 mL of extract volume and 0.5 mL of aqueous Folin–Ciocalteu reagent (10% v/v) was combined with 0.9 mL of aqueous trichloroacetic acid (10% w/v) within an Eppendorf tube. After 10 min intervals, the absorbance at 760 nm was recorded. Ascorbic acid was used as a calibration standard.

2.7. Antioxidant Assays

2.7.1. Ferric-Reducing Antioxidant Power (FRAP) Assay

The protocol published by Shehata et al. [25] was utilized for the evaluation of FRAP. A 0.1 mL aliquot of an appropriately diluted sample was combined with 0.1 mL of 4 mM FeCl₃ solution, which was prepared in 0.05 M HCl. This mixture underwent incubation for 30 min at 37 °C. Subsequently, 1.8 mL of TPTZ solution (1 mM in 0.05 M HCl) was added, with the mixture being vortexed. The absorbance was recorded at 620 nm after 5 min. The antioxidant activity through ferric-reducing power (P_R) was determined using an ascorbic acid calibration curve in 0.05 M HCl, with its concentrations (i.e., C_AA) ranging from 50 to 500 μM. The P_R was expressed as μmol of ascorbic acid equivalents (AAE) per g of dw, calculated using Equation (3) as follows:

$$P_{\mathrm{R}} (\mathsf{\mu }\mathrm{mol} \mathrm{AAE}/\mathrm{g} \mathrm{dw})={\displaystyle \frac{C_{\mathrm{A}\mathrm{A}} \text{×} V}{w}}$$

(3)

where V denotes the total volume of the extraction medium (in L), and w represents the dry weight of the sample (in g).

2.7.2. DPPH^• Antiradical Activity Assay

The antioxidant activity of apple peel extracts was also evaluated through radical inhibition activity, employing a modified DPPH^• methodology as described by Shehata et al. [25]. In this method, 0.1 mL of the extract was combined with 3.9 mL of a 100 μM DPPH^• solution in methanol. The final mixture was then incubated for 30 min in the dark at ambient temperature. Subsequently, the absorbance at 515 nm was recorded. Additionally, a blank sample containing the DPPH^• solution and methanol was used, with its absorbance being immediately recorded in the same wavelength. The percentage of inhibition was computed using Equation (4):

$$\mathrm{Inhibition} \left(\%\right)={\displaystyle \frac{A_{515\left(\mathrm{i}\right) }-A_{515\left(\mathrm{f}\right)}}{A_{515\left(\mathrm{i}\right)}}} \text{×} 100$$

(4)

Different ascorbic acid concentrations (C_AA) were used in the calibration curve, represented by Equation (5), which was utilized to assess the antiradical activity (A_AR), expressed as μmol AAE per g of dw as follows:

$$A_{\mathrm{AR}}\left(\mathsf{\mu }\mathrm{mol} \mathrm{AAE}/\mathrm{g} \mathrm{dw}\right)={\displaystyle \frac{C_{\mathrm{AA}} \text{×} V}{w}}$$

(5)

where V denotes the total volume of the extraction medium (in L), and w represents the dry weight of the sample (in g).

2.8. Statistical Analysis

A statistical study focusing on distribution analysis and response surface methodology was carried out using JMP^® Pro 16 software (SAS, Cary, NC, USA). Quantitative analyses were conducted in triplicate, with each batch of apple peel extracts undergoing the extraction process at least twice. To ensure data normality, the Kolmogorov–Smirnov test was used. One-way analysis of variance (ANOVA) was utilized to identify statistically significant differences, which was followed by post hoc Tukey HSD (honestly significant difference) test calculations applying the Tukey–Kramer method. Results are presented as means ± standard deviations. Additionally, Pareto plot and partial least squares (PLS) analyses were conducted for extraction optimization, whereas correlation analyses (i.e., principal component analysis (PCA), and multivariate correlation analysis (MCA)) were used to interpret any correlations between the variables under investigation. All mentioned analyses were performed using JMP^® Pro 16 software.

3. Results and Discussion

The apple fruits weighed 171.35 ± 12.85 g. The total soluble solid value was 14.17 ± 0.91 °Brix, the pH value was 4.08 ± 0.06, and the titratable acidity (given as % malic acid) was measured at 0.39 ± 0.01. These findings fall within the typical range for fruits, indicating a balance between acidity and sweetness. Also, these values are vital in determining the ripeness, flavor profile, and overall quality of the fruit, which are essential for consumer satisfaction and processing requirements.

Despite similarities in the methodology with our previously mentioned study [19], it is worth mentioning that apples create a vast amount of waste annually. Apples have a significantly higher global annual production (~86 million tons) [26] than mandarins (~38 million tons) [27]. To that end, this research was concerned with the extraction optimization of polyphenols from apple peels. The conditions and combinations of extraction techniques were examined through RSM in order to identify the optimal model for generating extracts that are abundant in bioactive compounds (i.e., polyphenolic compounds and ascorbic acid) and possess significant antioxidant activity. Additionally, previous studies have demonstrated that incorporating PEF and US into the extraction process can enhance its effectiveness [28]. Ethanol has the potential to be mixed with water to provide an extraction solvent that is well suited for application within the food industry [29]. However, it is crucial to take into account that polyphenols are thermolabile chemicals [30] for which the ideal temperature range for conventional extraction procedures to attain the maximum polyphenol recovery is typically between 50 and 80 °C [31,32]. An extensive examination is required to evaluate the influence of duration on extraction, considering the established effectiveness of both brief and extended extraction times in prior investigations [33]. PEF and the US are environmentally sustainable as they require reduced extraction time and energy consumption [34].

3.1. Extraction Parameters Optimization

The recovery of bioactive compounds may encounter challenges that stem from changes in the solubility and polarity of their chemical structures [35]. Hence, it is imperative to enhance the efficiency of this procedure [36]. The solvent’s composition is critical, as its features significantly influence chemical extraction [37]. Moderately polar compounds, such as polyphenols, are challenging to extract using highly polar solvents like water. Therefore, solvents with moderate polarity such as organic solvents are frequently utilized to further enhance the extraction process. Ethanol, for example, is a unique organic solvent (i.e., an alcohol) that can be mixed with water to create an extraction solvent appropriate for use in the food sector [29]. The extracts underwent a screening process that included performing spectrophotometric analysis for the specified assays. The assessment of polyphenols was carried out using the widely known Folin–Ciocalteu spectrophotometric method, which is a rapid, cost-effective, and highly sensitive methodology for measuring the total amount of polyphenols [36]. This method is widely recognized for its significant correlation with the liquid chromatographic method of measurement [38,39]. The obtained model from partial least squares analysis determined the optimal sample with high levels of polyphenols and strong antioxidant properties. Subsequently, liquid chromatography was employed to identify the specific polyphenols found in apple peel. In

Table 2. Findings from the experiment on the four independent variables and their corresponding responses of the dependent variables.

Design Point	Independent Variables				Responses
	X1	X2	X3	X4	TPC 1	FRAP 2	DPPH 3	AAC 4
1	3	1	3	4	3.04 ± 0.17 j	16.98 ± 0.51 j	27.63 ± 1.66 k,l	0.56 ± 0.04 n
2	3	2	1	3	10.42 ± 0.58 e,f	76.39 ± 4.66 f,g	59.71 ± 3.58 e,f,g	2.77 ± 0.06 c,d,e,f
3	2	3	4	3	11.17 ± 0.78 d,e	101.98 ± 2.45 c,d	67.89 ± 3.06 c,d,e	2.13 ± 0.04 i,j,k
4	2	4	5	4	10.78 ± 0.57 d,e,f	102.24 ± 3.78 c,d	73.29 ± 4.10 c,d	2.45 ± 0.18 f,g,h,i
5	3	5	4	2	6.53 ± 0.23 i	55.76 ± 2.23 i	38.38 ± 1.65 i,j	2.53 ± 0.16 e,f,g,h
6	4	1	4	5	3.99 ± 0.12 j	28.42 ± 1.02 j	22.95 ± 0.80 l,m	1.49 ± 0.06 l,m
7	4	2	3	1	10.36 ± 0.71 e,f	71.18 ± 1.85 f,g,h	51.63 ± 1.14 g,h	2.36 ± 0.13 g,h,i,j
8	1	3	3	2	14.40 ± 0.71 b	128.06 ± 4.74 a	105.66 ± 2.54 a	2.50 ± 0.11 e,f,g,h
9	1	4	4	1	13.96 ± 0.82 b	127.88 ± 6.52 a	90.74 ± 2.63 b	2.85 ± 0.08 c,d,e
10	1	5	1	4	8.89 ± 0.30 g,h	82.22 ± 5.51 e,f	64.57 ± 1.68 d,e,f	2.18 ± 0.14 h,i,j
11	1	1	2	3	2.70 ± 0.18 j,k	19.17 ± 0.40 j	32.07 ± 2.34 j,k	1.14 ± 0.09 m
12	1	2	5	5	13.65 ± 0.30 b	126.73 ± 9.25 a,b	97.62 ± 5.37 a,b	2.96 ± 0.12 c,d
13	4	3	2	4	15.93 ± 0.56 a	114.91 ± 4.14 a,b,c	71.30 ± 4.06 c,d	2.94 ± 0.09 c,d
14	3	4	2	5	13.37 ± 0.27 b	113.82 ± 3.41 b,c	73.18 ± 5.34 c,d	3.53 ± 0.23 b
15	2	5	3	5	7.48 ± 0.18 h,i	68.24 ± 2.80 g,h,i	42.96 ± 2.23 h,i	2.01 ± 0.10 j,k
16	2	1	1	1	1.57 ± 0.06 k	21.43 ± 0.96 j	16.71 ± 1.17 m	0.62 ± 0.04 n
17	2	2	2	2	9.41 ± 0.37 f,g	79.61 ± 4.06 f,g	57.17 ± 3.49 f,g	1.80 ± 0.12 k,l
18	3	3	5	1	11.86 ± 0.34 c,d	94.23 ± 2.17 d,e	73.87 ± 1.63 c	3.05 ± 0.09 c
19	4	4	1	2	13.15 ± 0.55 b,c	105.28 ± 3.26 c,d	58.90 ± 2.47 e,f,g	2.68 ± 0.10 d,e,f,g
20	4	5	5	3	9.49 ± 0.21 f,g	57.57 ± 2.88 h,i	29.71 ± 1.57 j,k,l	4.16 ± 0.11 a

Values are calculated as the mean values of triplicates (±standard deviation). Lowercase letters (e.g., a–n) within each column denote the statistically significant differences (p < 0.05); ¹ values in mg GAE/g dw; ² values in μmol AAE/g dw; ³ values in μmol AAE/g dw; ⁴ values in mg AA/g dw.

, the experimental findings out of the four independent variables are displayed. The TPC values of the extract ranged from 1.57 to 15.93 mg GAE/g dw, with design points 16 and 13 representing the lower and the higher TPC values, respectively. These values are consistent with the literature, considering that Villamil-Galindo and Piagentini [40] reported a similar range, 6.33 to 11.90 mg GAE/g dw, on ‘Granny Smith’ apple peels. As for the antioxidant activity of the extracts, the higher values are reported on design point 8 on both FRAP and DPPH assays, and the reported values are 128.06 and 105.66 μmol AAE/g dw, respectively, while the lowest antioxidant activity is on design point 1 for FRAP with a value of 16.98 μmol AAE/g dw, and 16 for DPPH, with a value of 16.71 μmol AAE/g dw. The highest ascorbic acid content resulted on design point 20, where the AAC is 4.16 mg/g dw, and the lower value is observed in design point 1, with a value of 0.56 mg/g dw.

The statistical parameters, including second-order polynomial equations (models) and their respective coefficients (R² > 0.97) derived from each model, are displayed in

Table 3. Mathematical models employing RSM have been utilized to optimize the extraction process of apple peels.

Responses	Second-Order Polynomial Equations (Models)	R2 Predicted	R2 Adjusted	p-Value	Equation
TPC	Y = −4.03 − 5.13X1 + 15.79X2 − 2.39X3 + 0.14X4 + 1.05X12 − 2.39X22 − 0.003X32 + 0.47X42 − 0.0004X1X2 + 0.18X1X3 − 0.27X1X4 + 0.48X2X3 − 0.47X2X4 − 0.04X3X4	0.9779	0.9159	0.0033	(6)
FRAP	Y = −24.83 − 33.24X1 + 140.01X2 − 18.61X3 − 16.43X4 + 4.79X12 − 19.82X22 + 0.14X32 + 4.68X42 + 0.04X1X2 + 1.06X1X3 − 0.61X1X4 + 1.6X2X3 − 3.84X2X4 + 1.99X3X4	0.9896	0.9606	0.0005	(7)
DPPH	Y = −25.99 − 22.48X1 + 97.17X2 − 13.59X3 + 4.52X4 + 3.5X12 − 13.76X22 − 0.14X32 + 3.16X42 − 1.14X1X2 + 1.82X1X3 − 2.55X1X4 + 2.33X2X3 − 4.24X2X4 − 0.15X3X4	0.9662	0.8716	0.0090	(8)
AAC	Y = 1.17 − 1.5X1 + 3.96X2 − 2.62X3 + 0.21X4 + 0.2X12 − 0.6X22 + 0.03X32 + 0.33X42 + 0.06X1X2 + 0.38X1X3 − 0.26X1X4 + 0.39X2X3 − 0.37X2X4 − 0.006X3X4	0.9766	0.9109	0.0038	(9)

. Moreover, the adjustment to the R² model is provided for each equation, and all values are higher than ~0.87, indicating a good fit of the equations to the model. An excellent fit between the produced models and the observed data was observed based on the data. Additionally, all responses exhibit low p-values (ranging from 0.0005 to 0.0090), enhancing the above statement. Figure S2–S5 present plots illustrating the relationship between the actual and the predicted response for each parameter under examination and the corresponding desirability functions. Figure 1 displays 3D response plots for TPC, whereas Figure S6–S8 display 3D response plots for the other responses under investigation. In Figure S2, it is evident that only factor X₂ (i.e., solvent composition) significantly affects TPC, as it is the only factor that has a different influence on polyphenol recovery, while the other factors seem to lead to maximum recovery at each value, with little variation between them, with a desirability function of ~0.87. In Figure 1, it is profound that the TPC is maximized when only ST is employed at high temperatures and for a short duration of time. Regarding the antioxidant capacity through the FRAP assay, in Figure S3 it is noted that the antioxidant capacity of the extracts is maximized when ST and low temperatures are employed for relatively long times, with a desirability function of ~0.89. The suitable solvent composition for this assay is relatively non-polar. These conclusions are also enhanced by the 3D plots in Figure S6. Similar results are drawn regarding Figures S4 and S7 for the DPPH assay, with a desirability function of ~0.91. Concerning the AAC of the extracts, Figures S5 and S8 imply that the most favorable extraction technique is the combination of PEF, US, and ST for an intermediate temperature and at a short time, and the desirability function is ~0.84.

3.2. Impact of Extraction Parameters to Assays Through Pareto Plot Analysis

On the notion of statistical significance (p < 0.05), the main effects and their interactions were assessed through a standardized Pareto plot. The independent variables (extraction technique, X₁; solvent composition, X₂; extraction duration, X₃; temperature, X₄) and their interactions that impacted TPC, FRAP, DPPH, and AAC are illustrated in Figure 2. The depicted values also include the orthogonal coded estimates, which are obtained through the orthogonalization of the transformation estimates. Concerning all assays, it is obvious that factor X₂*X₂ has a negative impact on all TPC, FRAP, DPPH, and AAC. On the contrary, factor X₂ positively affects these values. As the solvent has a high level of efficacy in the recovery of polar compounds, it is now ascertained that a binary solvent is necessary to enhance the efficacy of the extraction [41]. It is also important to mention that factor X₁ has a negative impact on both FRAP and DPPH, but it positively impacts AAC. This could lead to the conclusion that the pre-treatment of the extracts with PEF or US led to the degradation of some polyphenols [12], and so FRAP and DPPH were negatively affected, whereas AA was not affected at all by these techniques.

3.3. Optimized Extraction Conditions

To achieve the maximum yields of TPC, FRAP, DPPH, and AAC, the desirability function was employed, and the results are displayed in

Table 4. Optimal extraction conditions and maximum predicted responses for dependent variables.

Responses	Optimal Conditions
	Maximum Predicted Response	Technique (X1)	C (%, v/v) (X2)	t (min) (X3)	T (°C) (X4)
TPC (mg GAE/g dw)	15.45 ± 2.33	PEF + US + ST (4)	50 (3)	60 (2)	65 (4)
FRAP (μmol AAE/g dw)	127.68 ± 17.57	ST (1)	75 (4)	120 (4)	20 (1)
DPPH (μmol AAE/g dw)	103.06 ± 18.63	ST (1)	50 (3)	120 (4)	65 (4)
AAC (mg/g dw)	3.92 ± 0.63	PEF + US + ST (4)	100 (5)	150 (5)	50 (3)

. It can be noted that TPC and AAC are favored when PEF, US, and ST are combined, with their predicted responses at 15.45 mg GAE/g dw and 3.92 mg/g, respectively, while the antioxidant activity of the extracts is maximized only through ST, with predicted responses 127.68 μmol AAE/g dw for FRAP and 103.06 μmol AAE/g dw for DPPH. The solvent composition ranges from intermediate polarity to low polarity, while a long extraction duration seems to favor all responses. The extraction temperature varies from 20 to 65 °C. Since these conditions are fundamentally distinct, RSM, ANOVA, and PLS statistical analyses are necessary to determine the general best settings for all four responses to simultaneously extract the greatest possible quantities.

3.4. Correlation Analyses

PCA (Figure 3) and MCA (

Table 5. Multivariate correlation analysis of the measured variables.

Responses	TPC	FRAP	DPPH	AAC
TPC	–	0.9616	0.8757	0.7759
FRAP		–	0.9474	0.6755
DPPH			–	0.5269
AAC				–

) were utilized to further examine the responses more thoroughly and provide information about the correlations among the examined responses. The objective of MCA is to identify patterns and relationships among multiple variables concurrently, while PCA is employed to represent a multivariate data table as a reduced set of variables (summary indices) to facilitate the observation of trends, anomalies, clusters, and outliers. PCA is particularly advantageous for examining complex datasets, enabling a more profound comprehension of the data and their correlation. AAC, TPC, FRAP, and DPPH have a positive correlation with both factors X₂ and X₃ but a negative correlation with factors X₁ and X₄. As for MCA analysis, TPC seems to have a high correlation with all three responses and, more specifically, it correlates with FRAP at ~0.96, at ~0.88 with DPPH, and ~0.78 with AAC. FRAP and DPPH are also in line with one another. However, a relatively low correlation is observed between AAC and FRAP (~0.68) and between AAC and DPPH (~0.53).

3.5. Partial Least Squares (PLS) Analysis

The impact of the four extraction parameters was assessed by applying a PLS model. Figure 4 illustrates the PLS model used to create a correlation loading plot in which the impact of extraction conditions of apple peels is visually displayed. The projection factor with a value over 0.8 indicates that this variable exhibits a high contribution. It is obvious that the X₁ variable, just stirring, was far from enough to yield maximum polyphenol recovery, the antioxidant activity of the extracts, and ascorbic acid content simultaneously; however, this became feasible when PEF, US, or coupled PEF and US pretreated extractions were executed. This phenomenon can be attributed to the electroporation processes associated with PEF, which has proven efficacy, as the current disrupts the membrane of cells in apple peels, facilitating the recovery of targeted bioactive compounds [12]. The cavitation effect of ultrasound, which induces the creation of microcracks on solid surfaces due to the release of high energy upon bubble collapse, also results in increased yields [42]. Regarding the X₂ variable, it was observed that a moderate-polarity solvent comprising 75% v/v ethanol was ideal in all assays. The explanation for this may be in the moderate-polarity polyphenols recovered in comparison to water [43]. Extraction duration (i.e., variable X₃) had the most effective impact when short periods were applied in all assays. Finally, in all cases, it was observed that temperatures as high as 80 °C (variable X₄) favored the maximization of the extraction yields. The desirability of this model is ~0.92, which implies a good fit of the model in the data.

After comparing the PLS model values with those derived from experimental investigations, a correlation of 0.9845 between them is observed and they exhibit no significant deviations, possessing a p-value < 0.0001.

3.6. Analysis of the Optimal Extract

The optimal extraction conditions were determined as a combination of PEF, US, and ST, utilizing 75% v/v aqueous ethanol as a solvent, for 30 min at 80 °C. There is a plethora of research in the literature where the combination of PEF, US, and ST is shown to be effective for the recovery of bioactive compounds [12]. The combination of electroporation and cavitation effect, as pretreatment techniques, appear to considerably enhance the performance of conventional extraction through stirring, and thus it is possible to obtain higher yields in shorter times (about 30 min), and thus less energy is expended. The polyphenols recovery, along with the antioxidant capacity (FRAP, DPPH, and AAC values) are exhibited in

Table 6. Optimal extraction conditions (X₁:4, X₂:4, X₃:1, and X₄:5) and maximum desirability for all variables using the partial least squares (PLS) prediction profiler.

Variables	PLS Model Values	Experimental Values
TPC (mg GAE/g dw)	19.84	17.23 ± 0.65
FRAP (μmol AAE/g dw)	152.79	130.87 ± 6.15
DPPH (μmol AAE/g dw)	86.79	95.38 ± 3.05
AAC (mg/g dw)	4.44	3.99 ± 0.13

. Moreover, in

Table 7. Polyphenolic compounds identified under optimal extraction conditions (X₁:4, X₂:4, X₃:1, and X₄:5).

Polyphenolic Compound	Optimal Extract (mg/g dw)
Pelargonin chloride	2.56 ± 0.19
Catechin	1.58 ± 0.03
Chlorogenic acid	0.61 ± 0.04
Homovanillic acid	0.01 ± 0
Epicatechin	0.11 ± 0
Quercetin 3-D-galactoside	0.60 ± 0.02
Hesperidin	0.59 ± 0.04
Total identified	6.06 ± 0.34

, the individual polyphenolic compounds determined through HPLC-DAD are displayed, and in Figure S9 a representative chromatograph of these polyphenolic compounds is depicted. The TPC recovered under the optimal conditions, which is 17.23 mg GAE/g dw, is very close to the value predicted by the PLS program. This result is almost double the one reported by Tian et al. [15], who determined 9.22 mg GAE/g dw on apple peel flesh. Kunradi Vieira et al. [44] subjected apple fruit and peels in the US for 15 min and 80% v/v acetone and reported a TPC on apple peels that is ~199% lower than ours. It is evident that the selection of a suitable solvent is crucial for the recovery of bioactive compounds, and that an ethanol–water mixture, which produces a solvent of moderate polarity, results in much greater yields compared to acetone, a non-polar solvent. Yue et al. [21] reported a TPC of ~13 mg GAE/g dw from apple peels when 50% ethanol was utilized with 30 min of ultrasonication at 50 °C. It is important to highlight that in our study, the TPC obtained under the optimal conditions utilized lower temperature and ultrasonication duration. Wang et al. [20] studied how PEF affects the extraction yield and the degradation of bioactive compounds from apple peels, and the results indicated that all PEF parameters play a crucial role in the TPC yield. Sethi et al. [16] determined the AAC on several cultivars of apple peels, and the reported content is ~1187 lower than the one reported in this study. It is once again evident that the nature of the raw material, in this case, the different varieties of the same fruit, clearly affects the extraction yield. The most abundant polyphenolic compound in apple peels identified in this research is pelargonin chloride. Pelargonin is an anthocyanin to which the red color of the apple peels is attributed [45]. Several pelargonidin pigments have been found in red apples by other researchers [46]. The second most abundant polyphenolic compound is catechin. Kondo et al. [47] also determined catechin in three different apple cultivars, and the amounts are from ~147% to ~394% lower than ours. Moreover, Karaman et al. [48] explored the polyphenolic compounds in apple flesh and peels from different cultivars, and they also determined catechin and chlorogenic acid. The highest amount of catechin they found was in the peels of the ‘Amasya’ cultivar and of chlorogenic acid in the ‘King Luscious’ cultivar, but these results are ~888% and ~663% lower than ours, respectively. The same research team also determined epicatechin in their apple samples, and the highest amount in apple peels was in ‘Ervin Spur’ for epicatechin, which is in line with our value.

4. Conclusions

By conducting thorough research and carefully analyzing various conditions, this study sought to identify the most effective method for extracting bioactive compounds from apple peels. Utilizing the RSM approach, we were able to identify the key parameters for extraction. However, the PLS model provided us with the opportunity to further refine and optimize these parameters. The significance of incorporating PEF and US into traditional extraction methods was discovered. It is worth mentioning that additional research could explore alternative green techniques or extraction conditions to enhance the recovery of bioactive compounds. The findings of this study highlight the effectiveness of green extraction techniques for specific byproducts that can be utilized to develop food additives, animal feed, dietary supplements, and cosmetics. Additionally, they offer an effective environmentally friendly alternative for minimizing waste in the food sector.