Performance of Combined Olive Mills Wastewater Treatment System: Electrocoagulation-Assisted Adsorption as a Post Polishing Sustainable Process

Abstract

This study investigates the effectiveness of electrocoagulation (EC) with locally sourced iron electrodes for treating olive mill wastewater (OMW) prior to adsorption with olive stone (OS). Using Response Surface Methodology (RSM), 60 experiments were conducted to evaluate various operational parameters, including current density (CD), reaction time (T), distance between electrodes (D), and the number of electrodes (N). The optimal conditions identified were a reaction time of 53.49 min, a current density of 15.1104 mA/cm², 1 cm electrode spacing, and six electrodes. Under these conditions, the removal efficiencies achieved were 54.46% for total phenols (TPh), 73.25% for total Kjeldahl nitrogen (TKN), 92% for turbidity, 58.91% for soluble chemical oxygen demand (COD_soluble), and 58.55% for total COD (COD_total), with an energy consumption of 14.3146 kWh/m³ and a projected cost of USD 3.92/m³. Following the EC process, the treated OMW underwent further adsorption using OS, enhancing pollutant removal. The combined EC and adsorption (ECA) method demonstrated superior performance, achieving TPh removal at 62.63%, TKN removal at 77.52%, and turbidity reduction at 83.73%. Additionally, COD_total removal increased to 72.88% with COD_soluble removal at 70.04%. This integrated approach significantly improves pollutant removal, presenting a promising solution for effective OMW treatment.

1. Introduction

Water is a vital natural resource necessary for life. Nevertheless, freshwater resources face significant threats worldwide. These threats stem from rising consumption rates, pollution from human activities, and the impacts of global climate change, which disrupts normal precipitation patterns [1]. According to reports from the World Health Organization and the United Nations International Children’s Emergency Fund, approximately 2.1 billion individuals lack access to safe drinking water at home, and 4.5 billion people do not have access to adequate sanitation [2]. Activities such as industrialization, agriculture, and urbanization significantly threaten freshwater supplies by discharging large volumes of inadequately treated or untreated wastewater [3]. The emergence of pollutants of growing concern from these anthropogenic sources has become a pressing issue due to their detrimental effects on ecosystems [1,4].

Over the past two decades, the global demand for olive oil has surged, with Mediterranean countries being the largest producers, accounting for over 97% of worldwide olive oil production [5,6]. This agro-industry, however, is associated with significant water consumption and the generation of liquid effluents and solid waste [7]. The global production of olive mill wastewater (OMW) is estimated to reach 40 million m³ annually, with a significant number of oil extraction facilities primarily located in Mediterranean countries [8]. These nations collectively account for over 75% of the world’s OMW production, approximately 30 million m³ per year [7]. On average, the processing of each ton of olives generates around 1.3 m³ of this black liquid effluent [9]. OMW is toxic and varies in its concentrations of phenolic compounds, proteins, long-chain fatty acids, and suspended solids [8], influenced by factors such as cultivation practices, climatic conditions, olive variety, and the extraction methods used (traditional press, two-phase, and three-phase extraction) [8,10,11].

Olive mill wastewater (OMW), a byproduct of the olive oil production process, represents a considerable environmental pollution threat. Although OMW is produced seasonally and in smaller quantities compared to other industries, its harmful effects on the environment are significant [12], necessitating immediate and comprehensive management to mitigate its environmental impact. The high toxicity and resistant nature of OMW complicate its decontamination and reduce its biodegradability [10,13]. OMW is characterized by its acidic nature (pH levels ranging from 2.2 to 6), high suspended solids (up to 190 g/L), chemical oxygen demand (COD) levels (up to 200 g/L), phenolic compound levels of 4 to 6 g/L, residual fats, organic matter, and high phytotoxicity, which inhibits seed germination and negatively impacts plant growth [14,15]. As one of the most hazardous types of food processing waste, OMW poses significant risks to both the environment and human health [14]. Its infiltration into surrounding ecosystems can lead to soil compaction and clogging, deterioration of surface and groundwater quality, and inhibition of plant growth [16]. In regions such as Jordan, where water resources are already constrained, the inadequate disposal of OMW significantly exacerbates the challenges associated with water scarcity and environmental pollution [10].

Therefore, developing effective, sustainable, and cost-efficient treatment processes for OMW is crucial to mitigate its environmental impact. The absence of adequate treatment systems often leads to the disposal of OMW in valleys, where it mixes with rainwater and increases the burden on existing wastewater treatment facilities [17]. Current treatment processes include membrane filtration [18], advanced oxidation processes (AOPs) [19,20], coagulation/flocculation [21], sequencing batch reactors [22], and both aerobic and anaerobic processes [23,24]. However, these methods often fall short due to their complexity, inefficiency, high energy requirements, chemical usage, limited biodegradability, economic impracticality, and the generation of secondary pollutants during treatment [25,26]. Consequently, there is an urgent need for simple and cost-effective OMW treatment methods [10] that do not compromise the recovery of phenolic compounds [27]. Therefore, innovative strategies such as electrocoagulation (EC) and adsorption methods are being explored. Specifically, the combination of iron electrodes in the EC process with adsorption using olive stone (OS) may effectively reduce organic loads and phenolic content while facilitating the recovery of valuable compounds from OMW.

Researchers regard the EC technique as an effective method for treating both organic and inorganic pollutants in water and wastewater [28]. The fundamental principle of this method is to generate iron (Fe) ions by dissolving them on Fe electrodes and utilizing them as coagulants in ion formation. Metal ions are produced at the anode, while hydrogen gas is formed at the cathode [29]. The flocculated particles are carried to the water’s surface by the hydrogen gas bubbles that develop on the cathode. The advantages of EC include its flexibility, simplicity, lack of need for chemical additives, rapid startup time, ability to treat various pollutants, and reduced sludge generation [30]. Moreover, integrating EC with pre- or post-treatment methods can significantly improve its removal efficiency [10,31]. Studies show that combined techniques, such as electrocoagulation with adsorption (ECA), can achieve removal efficiencies up to 20% higher than EC alone, resulting in highly purified treated wastewater and facilitating efficient water reuse [32,33].

Adsorption is a highly effective and low-energy treatment method capable of achieving up to 99.9% removal efficiency for both soluble and suspended contaminants, addressing organic and inorganic pollutants [34]. This technique is favored for its simplicity, low initial costs, and minimal land requirements, making it a space-efficient option for water treatment [34]. There is increasing interest in using inexpensive adsorbents derived from local resources, particularly natural materials and agricultural byproducts [34]. In Jordan, OS, a byproduct of olive oil production, serves as a cost-effective source for activated carbon production [25]. Locally known as Jift, olive pomace has demonstrated effective adsorption of toxic compounds, making it suitable for applications such as groundwater treatment and drinking water purification [10]. Utilizing local agricultural waste for adsorption reduces disposal costs and provides a viable alternative to conventional activated carbon. Studies have confirmed the effectiveness of OS in wastewater purification [35]. In recent decades, various experimental design models have been utilized to optimize treatment methods, reduce time and energy consumption, and evaluate the effects of individual parameters and their interactions [8,36]. One notable statistical approach for this purpose is Response Surface Methodology (RSM). Among RSM models, Central Composite Design (CCD) stands out for its efficiency, as it requires fewer design points while still providing adequate data for assessing the experimental model’s validity [37,38].

This study specifically focuses on optimizing the treatment of OMW using EC with locally sourced iron electrodes. The efficiency of the EC process was evaluated based on the removal of COD, total phenols (TPh), total nitrogen (TN), turbidity (TR), and energy consumption (ENC). Following the EC treatment, an adsorption method employing OS was implemented as a subsequent polishing step under optimal conditions. This study utilizes OS sourced from the same OMW as an adsorbent. This approach enhances OMW treatment while repurposing waste from the same source. The ECA system was analyzed through RSM to identify the critical factors influencing pollutant removal. A mathematical model was developed using CCD to examine the relationships between removal efficiencies and parameters such as the reaction time (T), number of electrodes (N), distance between electrodes (D), and current density (CD). The treatment system consists of an initial EC process using iron electrodes, followed by adsorption to eliminate any remaining pollutants.

2. Materials and Methods

2.1. Wastewater Collection

The wastewater samples for our study were collected from an olive oil mill in Ajloun, Jordan, which employs a three-phase extraction process, between October 2023 and December 2023, coinciding with the olive harvesting season. Importantly, this facility’s production process does not utilize any chemical additives, ensuring that the samples reflect the natural characteristics of the effluent. The methods used for sampling, conditioning, and storage significantly impact the analytical results and their interpretation. To maintain sample integrity, the wastewater was promptly transported to the laboratory in a clean 20 L plastic container and stored at a low temperature of 4 °C until further analysis. The unpreserved samples were stored for no longer than 24 h to ensure that their characteristics remained unchanged. The experiments were conducted at the Environmental Engineering Lab at the University of Jordan.

Before treatment, a preparatory process was implemented to condition the samples. This involved decanting the samples to separate solids and filtering the supernatant through a cloth filter. This pre-treatment step was crucial for removing particulate matter, ensuring that the EC and adsorption processes were performed on a liquid phase that closely resembles the conditions of actual olive mill wastewater. This approach enhances the relevance and applicability of this study’s findings to real-world wastewater treatment scenarios within the olive oil production industry.

Table 1. Initial characteristics of OMW.

Parameter	Unit	Value
Color	-	Grey (G)
pH	-	5
Temperature	°C	13.8
Initial Total COD	g O2/L	123.9
BOD5	g O2/L	5
BOD5/COD	-	0.04
TPh	mg Gallic Acid/L	8562.8
TN	g O2/L	403.4
Turbidity	NTU	1770
Total solid (TS)	g/L	65.132
TSS	g/L	6.355
Total dissolved solid (TDS)	g/L	58.777

presents the initial characteristics of the OMW samples used in this study.

2.2. Chemicals

All chemicals used in this study were of analytical grade. The required chemicals and equipment were either purchased or already available in the laboratory. The following chemicals were sourced from Alnoorien Est. for Laboratories & Scientific Supplies in Amman, Jordan: potassium sulfate (K₂SO₄), copper sulfate (CuSO₄), sodium hydroxide (NaOH), and boric acid (4%). Additional chemicals, including potassium hydroxide (KOH), 2-Chloro-6-(trichloromethyl) pyridine (C₆H₃Cl₄N), COD kit HR (0–15,000 mg/L), potassium hydrogen phthalate (KHP), sodium chloride (NaCl), potassium nitrate (KNO₃), hydrochloric acid (HCl), ethanol (C₂H₆O), and sulfuric acid (H₂SO₄), were obtained from Hilmi Abu Sham and Partners Company in Amman, Jordan. The Folin–Ciocalteu reagent, anhydrous sodium carbonate (Na₂CO₃), and dry gallic acid (GA) were procured from Sidor Company in Irbid, Jordan.

2.3. Analytical Measurement

The test results obtained from the environmental engineering laboratory were verified at external laboratories to enhance the accuracy and reliability of this study. Specifically, analyses for COD, BOD, TN, and TPh were conducted in these external laboratories. The average values from these repeated measurements were used for our analyses, ensuring that the findings met the highest quality standards. This verification process was essential for informing key decisions and recommendations throughout this study.

2.3.1. Quality Parameter Analysis

To assess the effectiveness of the EC and adsorption methods, several quality parameters were analyzed both prior to and following treatment. These parameters included total COD, soluble COD, TPh, TN, and turbidity, along with any chemical alterations in the treated OMW samples. Temperature and pH were measured using a HAANA pH Meter (Hanna HI 2002-02, Hanna Instruments Ltd, Bedfordshire, UK), while electrical conductivity was evaluated with a Delta OHM meter (HD 3406.2, Senseca Germany Group, Senseca, Italy), with results presented in mS/cm. Turbidity was quantified using a Velp Scientifica TB1 turbidity meter (VELP Scientifica Srl, Usmate, Italy), with results given in NTU. An analytical balance from BEL Engineering was utilized for accurate measurements, including the assessment of changes in electrode mass.

2.3.2. COD and BOD Testing

COD was determined using a spectrophotometer (HACH Lange DR 3900, Hach Company, Loveland, CO, USA), which measures the oxygen consumed during oxidation reactions, with results reported in mg/L. For the BOD test, a specific sample preparation procedure was followed. The appropriate sample size was determined, and a suitable volume was placed in a BOD container along with a magnetic stirrer (Velp F20500051 Heating Magnetic Stirrer AREC.T, Velp Scientifica, Jakarta, Indonesia). To prevent nitrification, several drops of a nitrification inhibitor (ATH) were added. Additionally, 3–4 drops of KOH solution were applied to the sealing gasket to absorb carbon dioxide. After sealing the container, BOD sensors were installed, and the setup was incubated for five days at 20 °C, in accordance with the standard protocols outlined by Baird et al. [39].

2.3.3. Total Nitrogen Measurement

The TN content was measured using the macro-Kjeldahl method, as outlined by Halahsheh et al. [40]. A digestion solution was combined with 25 mL of raw wastewater and heated in a Velp Scientifica (DKL 8 Automatic Digestion Unit, VELP Scientifica Srl, Usmate, Italy). This process converted organically bound nitrogen to ammonia. The resulting solution was distilled in a Velp Scientifica (UDK 140 Distillation Units, VELP Scientifica Srl, Usmate, Italy), where ammonium was converted to ammonia and captured in a boric acid solution. The unreacted boric acid was then titrated with 0.02 N H₂SO₄ to a pH of 4.9.

2.3.4. Total Phenols Analysis

The TPh content was determined following the method outlined by the American Public Health Association [41] and Singleton and Rossi [42]. A stock solution was prepared by dissolving 0.5 g of GA in 10 mL of ethanol and diluting to 100 mL. Various concentrations of GA were created for calibration with a HACH Lange DR 3900 spectrophotometer at 765 nm. In labeled test tubes, 1.58 mL of distilled water and 0.02 mL of the sample were mixed with 0.1 mL of Folin–Ciocalteu reagent (a commercially available reagent for phenolic analysis). After a 30 s to 8 min reaction period, 0.3 mL of sodium carbonate (Na₂CO₃) solution was added, and the tubes were heated in a Mastest A005-01 KIT oven (Via Delle Industrie, Treviolo, Italy) at 40 °C for 30 min. The absorbance was then measured at 765 nm, and the TPh content was calculated using the calibration curve.

The removal efficiency (R%) is calculated using the following equation [43]:

$$R(\%)={\displaystyle \frac{{(\mathrm{C}}_0-{\mathrm{C}}_{\mathrm{e}})}{{\mathrm{C}}_{\mathrm{e}}}}\ast 100$$

(1)

where C₀ and C_e represent the initial and final concentrations (mg/L), respectively.

2.4. The Integrated ECA System

This research utilized a two-step combined system, starting with EC using Fe electrodes followed by adsorption on OS. The adsorption method acts as a polishing stage to achieve highly purified and treated wastewater.

2.4.1. Electrocoagulation Process

The laboratory-scale EC setup consists of a plastic rectangular reactor with a 3 L capacity, measuring 29 cm × 8 cm × 13 cm, designed for easy addition and removal of substances. The electrodes, made of locally sourced iron plates measuring 10 cm by 6 cm with a thickness of 2 mm, were submerged vertically to a depth of 4 cm in the reactor’s center, allowing the effluent to flow between them. The electrodes were sourced from a local blacksmith, who cut them into rectangular shapes. They were arranged in a Monopolar series configuration, with wood spacers attached to maintain their vertical parallelism. The EC system included eight electrodes: four anodes and four cathodes. An agitator, rotating at 350 rpm, ensures homogeneous mixing of the solution. The electrochemical cell is powered by a direct current generator, with the DC power supply (Sunshine-P-3005A, Matrix Jordan, Amman, Jordan) and wires obtained from an electronics store. To enhance electrical conductivity and reduce energy consumption, a small amount of NaCl is dissolved in distilled water and added to the effluent as a supportive electrolyte, improving the process’s efficiency. After treatment, the solution is allowed to settle, completing the process. The total effective surface area of the anodes submerged in the OMW solution was calculated as the combined area of each anode on both sides, resulting in an effective area of 240 cm² (10 cm × 4 × 2 × 3). Figure 1 shows the reactor used in the EC process during the research.

As mentioned above, this study used Fe as an electrode material. According to the following chemical reactions, ferric hydroxide is formed according to chemical Equations (2)–(4), and then, it acts as a coagulant for the pollutants found in the wastewater [44]:

$$Anode:F{e}_{(s)}\leftrightarrow {{Fe}^{2+}}_{(aq)}+2e^{-}$$

(2)

$$Cathode:2H_2{O}_{(I)}+2e^{-}\leftrightarrow {H_2}_{(g)}+2{{OH}^{-}}_{(aq)}$$

(3)

$$Overall:F{e}_{(s)}+2H_2{O}_{(I)}\leftrightarrow {Fe(OH)}_{2_{(s)}}+{H_2}_{(g)}$$

(4)

The OMW sample was introduced into the EC reactor for each experiment, which was conducted under varying conditions. These conditions included the number of electrodes (ranging from 1 to 3 pairs), inter-electrode distance (1 to 4 cm), applied current (1 to 5 A), and contact time (5 to 65 min). A concentration of 1000 mg/L NaCl was added to enhance conductivity during the EC process.

During the EC process, an oxide layer is formed at the anode. After each experiment, the electrodes were rinsed with a 0.1 N HCl solution to remove any solid residues from the surfaces of the reactors and electrodes. They were then rinsed again with distilled water to prevent electrode passivation. Additionally, the used electrodes were restored by polishing the oxide layer with abrasive paper, followed by washing in a 0.1 N HCl solution, rinsing with distilled water, and drying with absorbent paper. The electrodes were then weighed, and these values were used in the calculations of the total operating cost. Samples were collected from the bottom of the reactor using a pipette and filtered for analysis. These procedures were repeated for the other operational parameters.

2.4.2. Adsorption Process

The adsorption process was implemented as a polishing step following the EC process, utilizing biochar derived from OS. This integrated approach provides an environmentally and economically sustainable solution, leveraging the high adsorption capacity of biochar to effectively remove contaminants from wastewater. While EC efficiently eliminates larger pollutants, the subsequent adsorption step enhances the overall treatment by addressing residual pollutants. The experimental tests for the adsorption process included adsorption kinetics, conducted using a batch method. Key parameters investigated included particle size, pH, and initial concentration. A solution pH of 4 was chosen because it corresponds to significantly negative zeta potential values for the biochar. As pH increases beyond 5, the zeta potential becomes less negative, which can reduce the stability and adsorption capacity of the biochar. Therefore, pH 4 was deemed optimal for maximizing adsorption efficiency while preserving the desired surface properties of the adsorbent. Each experiment was conducted under controlled conditions, maintaining this pH, except during specific tests that evaluated varying pH levels. Shaking was performed for 3 h at 200 RPM, with adjustments made for the experiment assessing the impact of contact time. All experiments were carried out at room temperature (20 °C) after agitation. Filtration of the samples was conducted using a syringe filter with a pore size of 0.45 μm.

Preparation of Biochar

The preparation of biochar from OS involved several systematic steps. Initially, the OS was cleaned with distilled water to remove impurities and then dried in an oven at 50 °C for 24 h. After drying, the OS was sifted through a 4.75 mm sieve to eliminate any remaining leaves and stones. Next, the cleaned OS underwent heat treatment. The samples were divided into known weights (100 g), wrapped in tin foil, and heated in a muffle furnace (Nabertherm, Lilienthal, Germany, up to 3000 °C) at temperatures of 400, 500, and 600 °C, with heating durations of 1, 3, and 5 h tested for each temperature. After heat treatment, the samples were ground using a grinder (Silver Crest, Hong Kong, SC-1880) and then sieved with a shaker sieve (Electromagnet A059-02 KIT, Arcore, Italy) to obtain three particle sizes: 0.3 mm, 0.15 mm, and 0.075 mm. This systematic approach ensured the production of an effective adsorbent for the treatment process. Afterwards, the biochar was retained and stored in glass vials for subsequent works.

Biochar Properties

The yield of biochar was calculated as a percentage of the dry weight relative to the initial biomass. The pH of the biochar was assessed by soaking it in distilled water at a ratio of 3:50 (biochar to water) for 2 h with continuous agitation, followed by measurement using a pH meter (Hanna HI 2002-02, Hanna Instruments Ltd., Bedfordshire, UK) [45]. Electrical conductivity was determined using a conductivity meter (CON 6-LaMotte, Lamotte Co., Washington, DC, USA).

Ash content was evaluated according to a modified ASTM standard, which involved measuring weight loss [45,46]. Approximately 5 g of oven-dried biochar (dried for 24 h at 105 °C) was weighed and subsequently combusted at 750 °C for 6 h [47]. After combustion, the samples were allowed to cool to room temperature in desiccators before weighing [47]. Zeta potential measurements were also conducted using a DTS1070 cell. The microscopic morphology of the adsorbent was analyzed using scanning electron microscopy (SEM) (Tescan Vega3, TESCAN GROUP, Brno-Kohoutovice, Czech Republic). Fourier Transform Infrared Spectroscopy (FTIR) was employed to identify the functional groups present on the surface of the biochar using a Spectrum Two PerkinELmer FTIR spectrometer (L160000A, Shelton, CT, USA).

Adsorption Kinetic

For the adsorption kinetics experiments, four tubes containing OMW were prepared, each filled with 25 mL of adsorbate solution at varying initial concentrations of 0, 250, 500, and 750 mg/L. An equal amount of 0.1 g of the biochar adsorbent was added to each tube, along with glass beads, and the flasks were sealed. The samples were shaken at 200 rpm for specified durations: 5, 10, 15, 20, 30, 60, 90, 120, 150, and 180 min. TPh and COD tests were performed to monitor changes in adsorbate concentration over time. This setup allowed for the assessment of the adsorbent’s efficiency in removing contaminants throughout the adsorption process.

2.5. Material Design and Procedure

To determine the optimal conditions for total COD, soluble COD, TPh, TKN, turbidity, and electrode consumption, Minitab 22.2 software was employed to conduct a full-factorial CCD using RSM. The experiments included three continuous variables: reaction time, current density, and electrode distance, along with a categorical variable: the number of electrodes. These variables were selected based on prior research and published studies [26,48,49].

Table 2. Ranges of the input variables.

	Coded Values
Continuous Variable	−1.68179	−1	0	1	1.68179
	Uncoded Values
Time (min)	5	10	27.5	50	65
Current Density (mA/cm2)	4.17	8.33	12.5	16.67	20.833
Distance between electrodes (cm)	1	2	3	3.5	4
	Coded Values
Categorical Variables	1	2			3
	Uncoded Values
Number of Electrodes	2	4			6

presents the input variables and their levels in both coded and uncoded formats, with ranges determined from preliminary experiments.

The preliminary trials established the varying ranges for the parameters. Using Minitab 22.2, a CCD was created based on these parameters, resulting in a design matrix comprising 60 experiments. An analysis of variance (ANOVA) was conducted to assess the statistical significance among the samples and their means. Initially, the data were fitted to linear regression models to establish a mathematical relationship between the response variables and the influencing factors. Subsequently, second-order regression was applied to refine the model by incorporating quadratic and interaction terms. The model’s significance was evaluated using ANOVA.

Table 3. The design matrix.

	Coded Value				Uncoded Value
Run	CD, mA/cm2	Reaction Time, min	Distance Between Electrodes, cm	Number of Electrodes	CD, mA/cm2	Reaction Time, min	Distance Between Electrodes, cm	Number of Electrodes
1	0	0	−1	3	12.5	27.5	2	6
2	0	0	1.68179	2	12.5	27.5	4	4
3	0	0	0	3	12.5	27.5	3	6
4	1.68179	1	0	3	20.833	50	3	6
5	0	0	−1	1	12.5	27.5	2	2
6	0	1.68179	−1	2	12.5	65	2	4
7	0	0	−1.68179	1	12.5	27.5	1	2
8	−1	0	−1	3	8.333	27.5	2	6
9	0	0	−1	1	12.5	27.5	2	2
10	0	0	−1	1	12.5	27.5	2	2
11	−1.68179	1	0	3	4.17	50	3	6
12	0	0	−1	2	12.5	27.5	2	4
13	1	1	0	1	16.67	50	3	2
14	1.68179	0	−1	3	20.833	27.5	2	6
15	1	1	−1.68179	2	16.67	50	1	4
16	0	−1	−1	2	12.5	10	2	4
17	1.68179	0	−1	2	20.833	27.5	2	4
18	−1.68179	−1.68179	−1.68179	1	4.17	5	1	2
19	0	−1	−1	3	12.5	10	2	6
20	0	0	−1.68179	2	12.5	27.5	1	4
21	0	−1.68179	−1	1	12.5	5	2	2
22	1	1	−1.68179	2	16.67	50	1	4
23	0	−1.68179	−1	2	12.5	5	2	4
24	0	−1.68179	−1	3	12.5	5	2	6
25	0	1.68179	−1	3	12.5	65	2	6
26	−1.68179	1	0	1	4.17	50	3	2
27	−1.68179	−1.68179	0	2	4.17	5	3	4
28	−1.68179	1	0	2	4.17	50	3	4
29	1	−1.68179	0	1	16.67	5	3	2
30	1.68179	0	−1	1	20.833	27.5	2	2
31	0	1	−1	1	12.5	50	2	2
32	1.68179	−1.68179	−1.68179	2	20.833	5	1	4
33	−1.68179	1	−1.68179	1	4.17	50	1	2
34	0	0	−1	1	12.5	27.5	2	2
35	0	0	−1	1	12.5	27.5	2	2
36	0	0	−1	3	12.5	27.5	2	6
37	0	1	−1	2	12.5	50	2	4
38	0	0	−1	3	12.5	27.5	2	6
39	0	1.68179	−1	1	12.5	65	2	2
40	1.68179	1	0	2	20.833	50	3	4
41	0	−1.68179	−1	2	12.5	5	2	4
42	1.68179	−1.68179	0	2	20.833	5	3	4
43	−1.68179	−1.68179	−1.68179	2	4.17	5	1	4
44	1.68179	−1.68179	0	3	20.833	5	3	6
45	0	0	−1	3	12.5	27.5	2	6
46	0	0	−1	2	12.5	27.5	2	4
47	1.68179	−1.68179	−1.68179	1	20.833	5	1	2
48	0	0	−1	3	12.5	27.5	2	6
49	0	0	−1	2	12.5	27.5	2	4
50	1.68179	−1.68179	−1.68179	3	20.833	5	1	6
51	−1.68179	1	−1.68179	3	4.17	50	1	6
52	−1.68179	−1.68179	−1.68179	3	4.17	5	1	6
53	−1.68179	−1.68179	0	3	4.17	5	3	6
54	−1.68179	−1.68179	0	1	4.17	5	3	2
55	0	0	−1.68179	3	12.5	27.5	1	6
56	1.68179	1	−1.68179	1	20.833	50	1	2
57	0	−1	−1	1	12.5	10	2	2
58	1.68179	1	−1.68179	3	20.833	50	1	6
59	−1	0	−1	2	8.333	27.5	2	4
60	−1	0	−1	1	8.333	27.5	2	2

presents the design matrices used in the experiments. These matrices outline the various combinations of the input variables and their corresponding levels, facilitating the analysis of the effects of different parameters on the response variables. Each row in the table represents a specific experimental run, detailing the values assigned to the continuous and categorical variables as per the central composite design.

Coded values are typically used instead of original units for data analysis and interpretation, as they mitigate issues related to varying numerical scales and enhance the clarity of interactions between variables. Coded values also facilitate the assessment of the relative impact of different factors. The relationship between coded and original values is expressed in Equation (5) [50]:

$$Coded value={\displaystyle \frac{Actual value-Mean of the low and high values}{The difference between the low and high values}}$$

(5)

To establish a functional relationship between the responses and the variables, a second-order polynomial is employed. This model describes how the variables affect the outcomes. The relationship is represented in Equation (6) [50]:

$$Response=a_0+{\sum }_{i=1}^n{a}_i{X}_i+{\sum }_{i=1}^n{a}_{ii}{X}_i^2+{\sum }_{i<j}^n{\sum }_j^n{a}_{ij}{X}_i{X}_j+\beta$$

(6)

where a₀ is the constant coefficient, a_i is the linear coefficient, a_ii is the quadratic coefficient, a_ij is the interactive coefficient, X is the uncoded value of the factors, and β is the error value.

During each run, the current intensity and voltage readings from the DC power supply are recorded. One response variable, the ENC, can be calculated using these values. ENC is determined using Equation (7) [51]:

$$ENC \left({\displaystyle \frac{kWhr}{m^3}}\right)={\displaystyle \frac{VIt}{Volume of sample}}$$

(7)

where I is the applied current in A, t is the time in h, V is the voltage in volts, and the volume of sample is measured in m^3.

The theoretical and actual electrode consumption are calculated using Equations (8) and (9), respectively [52]:

$$EL{C}_{(theoritical)}={\displaystyle \frac{{\mathrm{It}}_{\mathrm{EC}}{M}_w}{\mathrm{ZFV}}}$$

(8)

where ELC_{(theoretical)} represents the theoretical electrode consumption in kg/m³, I represents the applied current in A, t_EC denotes the EC duration in seconds, Z refers to the chemical equivalence of the electrode with a value of 2 for Fe and Al (Z = 2), M_W represents the molecular weight of the electrode metal, which is 56 g/mol for Fe, and F represents the Faraday’s constant, which is equal to 96,500 C/mol for Fe. Lastly, V is used to indicate the volume of the treated wastewater in m³.

The actual ELC is then calculated based on the measured parameters from the experiment.

$$EL{C}_{(actual)}={\displaystyle \frac{{(\mathrm{m}}_{\mathrm{i}}-m_f)}{\mathrm{v}}}$$

(9)

where ELC_(actual) is the consumed number of electrodes per unit volume of pharmaceutical wastewater sample (kg/m³), m_i is the initial weight of electrodes (kg), m_f is the final weight of electrodes after treatment (kg), and V is the volume of the OMW sample (m³).

3. Result and Discussions

3.1. OMW Characterization

The wastewater samples for this study were collected from an olive oil mill in Ajloun, Jordan. All experiments were conducted in the Environmental Engineering Lab at the University of Jordan.

Table 4. Initial characteristics of OMW from current study compared with other studies.

Parameter	Unit	Current Study	[53]	[54]		[55]	[56]	[57]	[58]	[59]
				1st Sample	2nd Sample					1st Sample	2nd Sample
pH	-	5	4.91	5.2	5.6	6.5	4.6–5.1	4.2–5.5	4.62 ± 0.01	4.53 ± 0.02	4.47 ± 0.03
COD	mg O2/L	123,900	58,614	180,000	121,000	2500	25,800–146,000	45,230–106,800	52,100	100,800	118,500
BOD5	mg O2/L	5000	36,329	8900	22,000	1050	5260	17,640–41,720	-	22,500	28,400
BOD5/COD	-	0.04	0.62	0.049	0.182	0.42	0.204–0.036	0.39–0.391	-	0.22 ± 0.03	0.24 ± 0.024
TPh	mg/L	8563	2269	5500	4500	-	1540	-	6920	8710	7420
TKN	mg O2/L	403,400	544	-	-	-	-	-	-	-	-
Turbidity	NTU	1770	-	-		323	1264	-	-	-	-
TS	mg/L	65,132	-	91,200	72,000	3400	-	-	-	-	-
TSS	mg/L	6355	-	-	-	-	12,760	-	-	25,500	22,660

compares the initial characteristics of the OMW from the current study with those from other studies. This table provides insights into the composition and quality of OMW, highlighting variations in parameters such as COD, TPh, turbidity, and other relevant metrics across different research efforts.

As shown in Table 4, the results from our study indicated that the OMW had an acidic pH of 5. The COD was measured at 123,900 g/L, while the BOD₅ stood at 5000 mg/L, leading to a low BOD₅/COD ratio of 0.04. This low ratio suggested a high level of organic matter that was not readily biodegradable. Additionally, the TPh was significant at 8563 mg/L, indicating a rich presence of phenolic compounds. The TKN was recorded at 403.4 mg/L, reflecting the nitrogen content in the wastewater. Furthermore, the TSs were at 65.132 mg/L, and the TSSs were measured at 6.355 mg/L. These characteristics highlighted the complexity and richness of the OMW from this study, particularly in terms of organic and phenolic content. In contrast, Ayoub et al. [53] reported a BOD₅ of 36,329 mg/L and a COD of 58,614 mg/L, resulting in a BOD₅/COD ratio of about 0.62, indicating a more significant proportion of biodegradable organic material. Al-Qodah et al. [54] presented varying results; one sample had a BOD₅ of 8900 mg/L and a COD of 180,000 mg/L, yielding a ratio of 0.05, while another sample showed a BOD₅ of 22,000 mg/L and a COD of 121,000 mg/L, resulting in a ratio of 0.18. Khani et al. [55] reported a BOD₅ of 1050 mg/L and a COD of 2500 mg/L, leading to a BOD₅/COD ratio of 0.42, indicating a better balance of biodegradable content relative to the overall organic load. Shahawy et al. [56] showed a BOD₅ of 5260 mg/L with a COD that varied from 25,800 to 146,000 mg/L, demonstrating a range of ratios depending on the specific sample. Overall, the current study’s BOD5/COD ratio reflects a significant difference in the organic matter composition compared to previous studies, emphasizing the importance of processing methods and sampling timing in determining the environmental impact of OMW.

3.2. Biochar Characterization

The raw OMW exhibited an acidic pH of 5.

Table 5. Physicochemical properties and surface characteristics of biochar produced at a pyrolysis temperature of 400, 500, and 600 °C.

Parameters	Unit	Biochar
		400 °C	500 °C	600 °C
pH	-	10.53	10.87	10.92
Ash Contents	%	3.562	7.79	6.067
Yield	%	32.08	28.11	24.62
Moisture Contents	%	−0.121	1.20	1.79
Zeta Potential	mV	−22.72	−26.39	−25.2
Electrical Conductivity	mS/cm	0.0113	0.078	0.018
Surface Structure
Langmuir Surface area	m2/g	2.73	70.50	14.84
BET Surface area	m2/g	1.90	44.19	9.96
Single point Surface area at P/Po = 0.297	m2/g	1.98	38.92	9.91
Total pore volume	cm3/g	0.0012	0.026	0.0055
Average pore width	Å	24.53	23.99	21.95

presents the main characteristics and surface area of the biochar produced at pyrolysis temperatures of 400, 500, and 600 °C.

The pH values of biochar produced at 400 °C were lower than the biochar produced at 500 and 600 °C. Biochar corresponding to pyrolysis temperature of 400, 500, and 600 °C had yields of 32.08%, 28.11%, and 24.62%, respectively. This yield trend is consistent with the general understanding that higher pyrolysis temperatures result in the increased loss of volatile compounds, thereby reducing the overall yield of biochar. These results align with previous research by Abdelhadi et al. [60] and Abid et al. [61], which reported similar yield ranges for biochar derived from olive mill waste. The ash content of the biochar increased from 3.562% at 400 °C to 7.79% at 500 °C before decreasing to 6.067% at 600 °C. The increase at 500 °C suggests enhanced mineral retention and nutrient availability, which may benefit soil applications. The subsequent decrease at 600 °C may indicate the volatilization or transformation of inorganic compounds, highlighting the complex interactions at elevated temperatures.

The moisture content exhibited a noteworthy increase from negative values at 400 °C to 1.79% at 600 °C, suggesting improved retention of moisture at higher temperatures. Electrical conductivity measurements increased significantly at 500 °C, supporting the notion of enhanced ionic mobility due to the higher mineral content. The surface area measurements revealed significant variations. The Langmuir surface area was highest at 500 °C (70.50 m²/g) and decreased to 14.84 m²/g at 600 °C. The BET surface area followed a similar trend, peaking at 44.19 m²/g at 500 °C, while the single-point surface area at P/P_o = 0.297 was also highest at this temperature (38.92 m²/g). The increase in surface area at 500 °C can facilitate better adsorption of contaminants. However, a decline in these values at 600 °C suggests structural changes that may affect biochar’s effectiveness. These results indicate that 500 °C is optimal for developing a porous structure. The zeta potential values ranged from −22.72 mV at 400 °C to −25.2 mV at 600 °C. The increasing negativity of zeta potential indicates a rise in surface charge density and enhanced stability of biochar in aqueous solutions, which enhances the biochar’s adsorption capacity for pollutants.

The SEM micrograph of OS shown in Figure 2a displays a rough surface texture with no visible pores, indicating a compact structure typical for natural biomass. This lack of porosity may limit its effectiveness in applications such as adsorption, as it provides limited surface area. In contrast, the SEM image of biochar pyrolyzed at 400 °C, presented in Figure 2b, reveals a significant transformation in surface morphology. The presence of visible pores suggests an increase in surface area, enhancing its adsorption properties. The particles appear grainy with uneven shapes, indicative of a more complex structure compared to the OS.

As the temperature increases to 500 °C, the SEM micrograph of biochar in Figure 2c shows a further enhancement in porosity, with more visible pores distributed across the surface. This increase in temperature facilitates the development of a more porous structure, potentially improving the biochar’s capacity for adsorbing contaminants due to higher surface area and reactivity. Finally, the SEM image of biochar pyrolyzed at 600 °C, illustrated in Figure 2d, reveals an even more pronounced porous structure, with numerous pores that may significantly enhance its functional capacity as a sorbent.

The FTIR analysis of raw OMW, as illustrated in Figure 3a, reveals a complex mixture of materials. The prominent band at 421 cm⁻¹ is indicative of inorganic compounds or metal–ligand interactions, suggesting a complex mixture of materials present in the sample. At 551 cm⁻¹, the peak may correspond to double bonds or specific functional groups, further reflecting the intricate organic nature of OMW. The band at 1040 cm⁻¹ is associated with C-O stretching vibrations, confirming the presence of alcohols and carboxylic acids. Additionally, the peak at 1255 cm⁻¹ signifies C-C or C-H stretching, pointing to aromatic compounds or alkanes in OMW. The band at 1544 cm⁻¹ is linked to C=C stretching vibrations, indicating the presence of aromatic rings, which aligns with findings from Tran et al. [62], who reported similar observations in olive oil byproducts. Furthermore, the peak at 1656 cm⁻¹ corresponds to C=O stretching, indicating ketones or carboxylic acids, which are important for biological activity. The peaks at 2857 cm⁻¹ and 2927 cm⁻¹ reflect C-H stretching, indicating aliphatic hydrocarbons, while the strong band at 3331 cm⁻¹ demonstrates O-H stretching, characteristic of alcohols and phenolic compounds, aligning with the observations of Abid et al. [61], Hafidi et al. [63], and El Hajjouji et al. [64], who noted the significance of such compounds in the biological activity of OMW.

The FTIR spectrum of biochar produced at 400 °C (Figure 3b) shows that the peak at 409 cm⁻¹ (transmittance: 88.46%) indicates inorganic compounds or metal–ligand interactions. The band at 817 cm⁻¹ (transmittance: 91.15%) is associated with out-of-plane bending vibrations of aromatic C-H bonds [62]. The peak at 1204 cm⁻¹ (transmittance: 89.37%) corresponds to C-O stretching groups [65], while the band at 1420 cm⁻¹ (transmittance: 90.15%) relates to C-H bending vibrations [65]. The peak at 1592 cm⁻¹ (transmittance: 90.45%) indicates C=C stretching, confirming the presence of aromatic rings [61].

For biochar produced at 500 °C (Figure 3c), the peak at 408 cm⁻¹ (transmittance: 76.5%) also indicates inorganic compounds or metal–ligand interactions. The band at 1644 cm⁻¹ (transmittance: 88.86%) highlights the aromatic nature of the biochar through C=C stretching. The peak at 3286 cm⁻¹ (transmittance: 83.9%) is associated with O-H stretching, suggesting hydroxyl or phenolic groups. Finally, the FTIR spectrum of biochar produced at 600 °C (Figure 3d) shows a peak at 443 cm⁻¹ (transmittance: 86.85%) suggesting inorganic compounds or metal–ligand interactions. The band at 454 cm⁻¹ (transmittance: 87.05%) reinforces the presence of residual minerals in the biochar.

Figure 3a–d of biochar produced at different temperatures (400 °C, 500 °C, and 600 °C) exhibit chemical properties that significantly influence their efficiency in pollutant removal. The biochar produced at 400 °C shows peaks indicating the presence of inorganic compounds and aromatic structures, which enhance its ability to adsorb pollutants. At 500 °C, the efficiency of biochar increases due to a higher proportion of aromatic content and the presence of O-H groups, facilitating better interactions with organic pollutants. Conversely, the biochar produced at 600 °C retains some beneficial properties but may lose the desirable aromatic structure. Compared to raw OMW, which contains complex organic components, biochar demonstrates superior pollutant removal efficiency due to its aromatic characteristics and available functional groups. Therefore, the optimal temperature for biochar production for pollutant removal is 500 °C.

3.3. Experimental Results and Statistical Analysis

The experiments were successfully conducted using the design matrix created through the CCD. After each experiment, the samples were allowed to settle for one hour before testing the resulting supernatant. The COD_total, COD_soluble, TPh, TKN, and turbidity of the samples were measured according to the procedures outlined in the methodology. The removal efficiencies were then calculated to serve as response variables for subsequent analyses in Minitab 22.2. Using the recorded voltage and applied current for each run, the ENC was computed and used as an additional response variable. Statistical analyses were performed, and a response surface model was applied to the obtained results to determine the effects and significance of the variables, along with the regression equations. Response surface graphs were also generated to further explore the impact of the variables.

Analysis of Variances

The results from the CCD indicated that the quadratic model provided a good fit for the eight responses, as determined by the response surface regression analysis. Based on the experimental data, the following regression equations were derived:

$$\begin{array}{cc}\hfill \mathrm{C}\mathrm{O}\mathrm{D}\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\left(\mathrm{\%}\right)=& 2.66+1.026 \mathrm{T}+1.661 \mathrm{C}\mathrm{D}+3 \mathrm{D}-0.01565 \mathrm{T}\ast \mathrm{T}-0.0782 \mathrm{C}\mathrm{D}\ast \mathrm{C}\mathrm{D}-1.184 \mathrm{D}\ast \mathrm{D}\hfill \\ & +0.02204 \mathrm{T}\ast \mathrm{C}\mathrm{D}-0.0238 \mathrm{T}\ast \mathrm{D}+0.0836 \mathrm{C}\mathrm{D}\ast \mathrm{D}\hfill \end{array}$$

(10)

$$\begin{array}{cc}\hfill \mathrm{C}\mathrm{O}\mathrm{D}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{a}\mathrm{b}\mathrm{l}\mathrm{e}\left(\mathrm{\%}\right)& =13.67+1.335 \mathrm{T}+1.9 \mathrm{C}\mathrm{D}-6.76 \mathrm{D}-0.01685 \mathrm{T}\ast \mathrm{T}-0.062 \mathrm{C}\mathrm{D}\ast \mathrm{C}\mathrm{D}+1.15 \mathrm{D}\ast \mathrm{D}+0.01707 \mathrm{T}\ast \mathrm{C}\mathrm{D}\hfill \\ & +0.0173 \mathrm{T}\ast \mathrm{D}-0.108 \mathrm{C}\mathrm{D}\ast \mathrm{D}\hfill \end{array}$$

(11)

$$\begin{array}{cc}\mathrm{T}\mathrm{P}\mathrm{h}\left(\mathrm{\%}\right)=18.24+1.293 \mathrm{T}+0.944 \mathrm{C}\mathrm{D}-9.21 \mathrm{D}-0.01169 \mathrm{T}\ast \mathrm{T}\hfill & \hfill -0.0186 \mathrm{C}\mathrm{D}\ast \mathrm{C}\mathrm{D}+2.458 \mathrm{D}\ast \mathrm{D}\\ \hfill -0.00144 \mathrm{T}\ast \mathrm{C}\mathrm{D}-0.0179 \mathrm{T}\ast \mathrm{D}-0.042 \mathrm{C}\mathrm{D}\ast \mathrm{D}& \end{array}$$

(12)

$$\begin{array}{cc}\hfill \mathrm{T}\mathrm{u}\mathrm{r}\mathrm{b}\mathrm{i}\mathrm{d}\mathrm{i}\mathrm{t}\mathrm{y}\left(\mathrm{\%}\right)=& 74.77+0.025\ast \mathrm{T}+1.396\ast \mathrm{C}\mathrm{D}-3.71\ast \mathrm{D}+0.00198\ast \mathrm{T}\ast \mathrm{T}-0.0435\ast \mathrm{C}\mathrm{D}\ast \mathrm{C}\mathrm{D}+0.41\ast \mathrm{D}\hfill \\ & \ast \mathrm{D}+0.00242\ast \mathrm{T}\ast \mathrm{C}\mathrm{D}+0.0042\ast \mathrm{T}\ast \mathrm{D}+0.012\ast \mathrm{C}\mathrm{D}\ast \mathrm{D}\hfill \end{array}$$

(13)

$$\begin{array}{cc}\hfill \mathrm{T}\mathrm{K}\mathrm{N}\left(\mathrm{\%}\right)=23.64+1.343 \mathrm{T}+3.751 \mathrm{C}\mathrm{D}-8.47 \mathrm{D}-0.01472 \mathrm{T}\ast \mathrm{T}& -0.0938 \mathrm{C}\mathrm{D}\ast \mathrm{C}\mathrm{D}+1.034 \mathrm{D}\ast \mathrm{D}\hfill \\ \hfill -0.0104 \mathrm{T}\ast \mathrm{C}\mathrm{D}+0.0609 \mathrm{T}\ast \mathrm{D}-0.1848 \mathrm{C}\mathrm{D}\ast \mathrm{D}& \end{array}$$

(14)

$$\begin{array}{cc}\hfill ENC\left({\displaystyle \frac{kWhr}{m^3}}\right)=& 19.8+0.016 \mathrm{T}-4.55 \mathrm{C}\mathrm{D}+0.7 \mathrm{D}-0.01196 \mathrm{T}\ast \mathrm{T}+0.1257 \mathrm{C}\mathrm{D}\ast \mathrm{C}\mathrm{D}-2.14 \mathrm{D}\ast \mathrm{D}+0.0606 \mathrm{T}\hfill \\ & \ast \mathrm{C}\mathrm{D}+0.231 \mathrm{T}\ast \mathrm{D}+0.537 \mathrm{C}\mathrm{D}\ast \mathrm{D}\hfill \end{array}$$

(15)

The Equations (10)–(15) elucidate the effects of individual and quadratic coefficients on various targeted measurements when using six Fe electrodes in OMW treatment. Each equation focuses on a specific parameter related to pollutant removal efficiency. In Equation (10), which represents COD_total, all individual coefficients—current density, reaction time, and distance between electrodes—are positive. This indicates that increasing these parameters enhances the removal efficiency of total COD. However, the presence of negative quadratic coefficients suggests that as these parameters increase beyond a certain point, the benefits may diminish, highlighting a nonlinear relationship. Equation (11) for COD_soluble shows that while the coefficients for current density and reaction time are positive, the coefficient for distance is negative. This means that increasing current density and reaction time promotes the removal of soluble COD, whereas increasing the distance between electrodes may hinder this efficiency. The negative quadratic terms further suggest that while enhancing current density and reaction time is beneficial, excessive distance can lead to diminishing returns in removal efficiency. Equation (12) for TPh indicates that current density and reaction time positively contribute to phosphorus removal, but increased distance negatively affects efficiency.

In Equation (13) for turbidity, the positive coefficient for current density supports reduction, whereas negative coefficients for reaction time and distance indicate potential obstructions to performance. Equation (14) for TKN highlights the importance of current density and reaction time for nitrogen removal, with a negative distance coefficient suggesting that greater spacing can be detrimental. Finally, Equation (15) for ENC presents a constant value of 39.3, accompanied by a negative coefficient for current density. This suggests that higher current densities can lead to increased energy consumption. The negative quadratic terms for current density², reaction time², and distance² indicate that as these parameters rise, energy usage tends to increase disproportionately, underlining the need to balance operational efficiency with energy costs. The coefficients provide insights into how to optimize conditions for effective wastewater treatment while also managing ENC.

Table 6. The coefficient of determination for six models.

Model	R2 (%)	R2adj (%)	R2adj (Pred) (%)
%CODtotal	93.69	91.13	82.43
%CODsoluble	88	83.15	72.36
%TPh	94.81	92.72	90.17
%TKN	70.66	58.79	39.27
%Turbidity	94.41	92.15	84.02
ENC	51.5	31.88	22.2

shows the coefficient of determination (R²) and its adjusted values for six models assessing pollutant removal efficiencies and energy consumption metrics. The table includes three key metrics: R², R²_adj, and R²_adj (Pred). An R² value above 80% is generally considered adequate for a model to effectively explain the effects of the variables [66]. In this context, all models exhibit good R² values, indicating that they effectively explain the variability in the response variables.

As shown in Table 6, the COD_total model has an R² of 93.69%, meaning that only about 6.31% of the variation remains unexplained, which demonstrates its strong performance. Similarly, the TPh model has an even higher R² of 94.81%, suggesting robust explanatory capability. The R²_adj values provide insight into the model’s goodness of fit by accounting for the number of predictors included. All models show satisfactory R²_adj values, confirming their reliability. The R²_adj (Pred) values further indicate how well the models can predict new data. For example, the COD_total model has an R²_adj (Pred) of 82.43%, demonstrating good predictive power, while the TPh model achieves a value of 90.17%. In contrast, the TKN model shows lower performance with an R² of 70.66% and an R²_adj of 58.79%, and a notably low R²_adj (pred) of 39.27%, suggesting limitations in its predictive ability. The Turbidity model, on the other hand, achieves an R² of 94.41% and an R²_adj of 92.15%, indicating strong explanatory power for turbidity removal. Lastly, the ENC model exhibits the weakest performance, with an R² of 51.5% and an R²_adj of 31.88%, suggesting limitations in explaining energy consumption metrics.

The t values of the coefficients of the COD_total, COD_soluble, TPh, TKN, turbidity, and ENC model are shown in

Table 7. ANOVA for the COD_total, COD_soluble, TPh, TKN, turbidity, and ENC models (* is a multiplication sign).

Term	F Value	CODtotal		F Value	CODsoluble		F Value	TPh		F Value	Turbidity		F Value	TKN		F Value	ENC
		t Value	p Value		t Value	p Value		t Value	p Value		t Value	p Value		t Value	p Value		t Value	p Value
Model	36.65		0.000	18.12		0.000	45.18		0.000	5.59		0.000	41.74		0.000	2.62		0.006
Linear	39.99		0.000	15.03		0.000	47.41		0.000	12.67		0.000	30.16		0.000	4.87		0.001
T	71.47	8.45	0.000	53.85	7.34	0.000	194.66	13.95	0.000	42.76	6.54	0.000	113.49	10.65	0.000	8.45	2.91	0.006
CD	67.18	8.20	0.000	17.12	4.14	0.000	17.83	4.22	0.000	10.24	3.20	0.003	0.27	0.52	0.608	13.21	3.63	0.001
D	5.34	−2.31	0.000	0.16	0.40	0.692	3.17	1.79	0.082	1.22	−1.11	0.275	3.50	−1.87	0.068	1.03	1.01	0.317
N	30.03	−7.73	0.000	3.13	−2.46	0.054	16.82	−4.28	0.000	4.79	−3.09	0.013	15.62	−4.36	0.000	1.50	1.69	0.235
Square	55.05		0.000	25.61		0.000	36.97		0.000	1.93		0.139	67.85		0.000	0.96		0.419
T*T	87.88	−9.37	0.000	53.60	−7.32	0.000	90.05	−9.49	0.000	1.27	1.13	0.266	107.63	−10.37	0.000	1.94	−1.39	0.171
CD*CD	17.47	−4.18	0.000	5.78	−2.40	0.021	1.82	−1.35	0.184	4.88	−2.21	0.033	34.82	−5.90	0.000	1.71	1.31	0.198
D*D	1.48	−1.22	0.231	0.74	0.86	0.396	11.73	3.42	0.001	0.16	0.40	0.690	1.56	1.25	0.218	0.18	−0.43	0.671
Interaction	5.38		0.000	1.53		0.170	0.32		0.963	0.72		0.692	5.29		0.000	1.33		0.249
T*CD	25.54	5.05	0.000	8.07	2.84	0.007	0.20	−0.45	0.657	0.28	0.53	0.601	7.88	−2.81	0.008	7.31	2.70	0.010
T*D	0.43	−0.65	0.516	0.12	0.35	0.731	0.45	−0.67	0.508	0.01	0.11	0.912	3.89	1.97	0.055	1.53	1.24	0.222
T*N	8.84	−4.17	0.001	0.63	−1.11	0.538	0.69	−0.99	0.505	0.92	1.22	0.405	0.13	−0.44	0.878	0.02	0.13	0.983
CD*D	0.73	0.85	0.399	0.64	−0.80	0.429	0.34	−0.58	0.565	0.01	0.11	0.910	4.91	−2.22	0.032	1.13	1.06	0.293
CD*N	1.53	−1.64	0.227	0.35	−0.82	0.708	0.04	−0.09	0.960	0.10	0.01	0.905	13.33	−3.95	0.000	0.83	−1.12	0.444
D*N	0.48	−0.08	0.621	1.49	1.26	0.236	0.22	−0.57	0.806	2.05	−1.71	0.142	2.03	1.57	0.144	0.17	0.58	0.843
Lack of Fit	2083.95		0.000	2.08		0.099	-		-	-		-	-			975,409.98		0.000

.

The significance of each term in the model was assessed using the F-value, where higher F-values indicate greater significance [66]. The p-value, particularly when below 0.05, serves as a threshold for determining the adequacy of the F-value, indicating whether the model’s significance is sufficient at a 95% confidence level. According to Table 7, the results reveal that the variable time (T) has a significant impact on all parameters, with p-values consistently below 0.05. This highlights the crucial role of time in the model, suggesting it is a key factor influencing the response variables such as COD_total, COD_soluble, TPh, TKN, and ENC. The CD also shows a significant effect on all parameters except for TKN. In contrast, the number of electrodes (N) and distance between electrodes (D) show limited significance across the models.

Notably, the quadratic terms for time and CD also demonstrate statistical significance, reinforcing their roles in the model. Meanwhile, the quadratic term for D has a notable impact only on TPh, suggesting its limited relevance. Furthermore, the interaction between time and current density significantly impacts COD_total, COD_soluble, TKN, and ENC, emphasizing the interdependence of these variables in pollutant removal. The interaction between time and the number of electrodes affects COD_total alone, indicating a specific context where this relationship is significant. Lastly, the interaction between current density and the number of electrodes shows an effect on TKN. The statistical results indicate that by focusing on significant variables, the models can be refined and optimized, potentially through methods such as stepwise regression, which would enhance their predictive capabilities by excluding insignificant terms.

Figure 4a, Figure 4b, and Figure 4c illustrate the comparison between experimental and predicted values for the removal efficiencies of COD_total, TKN, and TPh models, respectively.

The data points presented in Figure 4a–c closely align along the diagonal line, suggesting a strong correlation between the experimental and predicted values. This proximity indicates that the model effectively captures the dynamics of both datasets. Furthermore, the R² values for each model reinforce this observation, with COD_total achieving an R² of 94.57% and with the values for TKN at 92.64% and TPh at 93.96%. Such high R² values, exceeding 90%, signify a robust relationship between the experimental and predicted outcomes, as noted by Hamid et al. [50]. This level of correlation underscores the model’s reliability in predicting removal efficiencies.

3.4. The Effect of the Variables

3.4.1. Variables Effect on COD_total Removal Efficiency

To enhance the understanding of the interaction between independent variables and to estimate the efficiency of COD_total removal, three-dimensional (3D) contour-overlaid response surface plots were generated based on the regression equation. These plots depict the relative effects of any two variables on COD_total removal efficiency while keeping the other variables constant, as analyzed using Minitab 22.2. In this case, the constant values were set at a reaction time of 35 min and with six Fe electrodes. The response variable is represented on the vertical axis, while the two parameters are plotted on the horizontal axes, facilitating a comprehensive visualization of the interactions among the various factors influencing the response variables in this study. As illustrated in Figure 5, the surface plots are created for the parameters of current density, reaction time, and the distance between electrodes.

Figure 5a presents a 3D surface plot that illustrates the relationship between current density, reaction time, and COD_total removal efficiency. The plot indicates a nonlinear relationship, demonstrating that the interactions among these parameters significantly influence COD_total removal effectiveness.

As reaction time increases, COD_total removal efficiency rises, peaking at around 60 min, where it can reach 60% at a CD between 17 and 20 mA/cm². At a lower CD (approximately 5 mA/cm²), removal efficiency remains low, but it improves notably as the CD increases to moderate levels (10–15 mA/cm²) due to enhanced generation of reactive oxidizing agents. However, beyond a threshold of about 20 mA/cm², efficiency may decline slightly due to side reactions. This finding aligns with previous studies [67,68], which suggest that at higher current densities, Fe⁺ ions undergo hydrolysis, resulting in the production of iron hydroxides. This process leads to increased sludge formation, significantly enhancing COD_total removal efficiency [68].

It is well established that current not only determines the coagulant dosage rate but also influences bubble production rates, bubble sizes, and floc growth, all of which affect the treatment efficiency of EC [69,70]. At higher currents, the amount of oxidized Fe increases, resulting in a larger quantity of precipitate for pollutant removal [70,71]. Furthermore, studies have shown that bubble density increases while bubble size decreases with rising current, leading to greater upward flux and faster pollutant removal and sludge flotation [72]. As current decreases, the time required to achieve similar efficiencies increases, as treatment efficiency is primarily affected by charge loading (Q = It), as reported by Chen et al. [73] and Ahmad et al. [74]. A similar pattern was observed by Holt et al. [69], who explained that at higher currents, the rapid generation of Fe⁺ ions outpaces the coagulation process, resulting in a decrease in removal efficiency when calculated on an equivalent Fe basis. Additionally, the quicker removal of Fe(OH)₂ from the solution through flotation reduces the probability of collision between the pollutants and the coagulant.

In Figure 5b, the surface plot depicts the relationship among the CD, distance between electrodes, and COD_total removal efficiency. The analysis reveals that optimal COD_total removal efficiency is achieved at an electrode distance closer to 2 cm. At this distance, the plot indicates that the removal efficiency is maximized across a range of CDs, similar to those noted in Figure 5a. This suggests that the spacing between electrodes is a critical factor influencing the effectiveness of the electrochemical process. When the distance between electrodes is maintained at around 2 cm, the efficiency of electrochemical reactions improves. This distance allows for an adequate electric field strength, facilitating the movement of ions and enhancing the interaction between reactants. If the distance is too large, the electric field may weaken, leading to reduced reaction rates and lower COD removal efficiency. Conversely, if the distance is too small, it may lead to increased resistance and potential issues such as short-circuiting [75]. These findings align with previous research, which indicates that optimal electrode spacing is essential for maximizing treatment efficiency [75,76]. Additionally, Figure 5c shows that optimal conditions are achieved at minimal distances between the electrodes, while keeping the reaction times consistent with those seen in the previous plots. The analysis reveals that as reaction time increases, COD_total removal efficiency generally improves, indicating that longer treatment durations enhance the degradation of organic contaminants [76]. As the distance increases to 4 cm, the efficiency tends to decline, likely due to reduced electric field strength and limited movement of reactive species.

3.4.2. Variables Effect on COD_soluble Removal Efficiency

Figure 6 presents the surface plots related to the removal of COD_soluble, illustrating the effects of CD, reaction time, and distance between electrodes. The analysis reveals a clear relationship between these parameters and COD_soluble removal efficiency.

Based on Figure 6a, when the reaction time increases, the COD_soluble removal efficiency tends to rise, suggesting that prolonged treatment allows for more effective breakdown of soluble organic contaminants [77]. Additionally, a higher CD also enhances removal efficiency, with the plot demonstrating that optimal COD_soluble removal occurs at an increased CD of around 10–15 mA/cm². However, the efficiency does not significantly improve beyond this point, indicating a potential threshold effect. The CD is essential in influencing coagulant dosage and bubble generation rates, which in turn affect solution mixing and mass transfer at the electrodes. A higher CD results in an increased production of hydrogen bubbles at the cathode, promoting greater upward flux and enhancing COD_soluble removal [67,78]. Figure 6b shows that optimal COD_soluble removal occurs when the distance between electrodes is approximately 2 cm, akin to the results in Figure 5b.

Furthermore, Figure 6c illustrates the relationship between COD_soluble removal, reaction time, and the distance between the electrodes. It shows that the removal percentage of COD_soluble increases with the reaction time until it reaches an optimal removal rate of approximately 58% at around 20 min. Beyond this point, the removal percentage decreases to about 40% at 60 min. Additionally, the data indicate that the removal efficiency remains relatively stable across different distances between the electrodes. This suggests that while reaction time plays a critical role in enhancing COD_soluble removal, the distance between the electrodes does not significantly impact the COD_soluble removal efficiency.

3.4.3. Variables Effect on TPh Removal Efficiency

Figure 7 displays the surface plots related to TPh removal, showcasing the impacts of CD, reaction time, and electrode distance. The analysis highlights a distinct relationship among these parameters and their influence on TPh removal efficiency.

Figure 7a–c collectively illustrate the intricate relationship between TPh removal efficiency, CD, reaction time, and electrode distance. In Figure 7a, it is evident that TPh removal efficiency significantly increases with longer reaction times, peaking between 50 and 60 min. This finding highlights the importance of allowing sufficient reaction time for effective removal. While higher CD also contributes positively to TPh removal, its impact is less pronounced compared to the effect of reaction time, suggesting that the duration of the reaction is more critical for achieving optimal efficiency.

The lower efficiency of TPh removal compared to other variables can be attributed to the autooxidation and subsequent polymerization of phenolic compounds during the storage of OMW. As noted by Assas et al. [79] and Safaa et al. [80], the storage process leads to the formation of dark-colored, recalcitrant compounds that are resistant to biological treatment. These changes complicate the removal of TPh, as the phenolic compounds undergo transformations that increase their stability and resistance to degradation.

Figure 7b illustrates the relationship between the removal efficiency of TPh, CD, and distance. The efficiency of TPh removal varies with distance, displaying notable peaks and troughs within Figure 7b. This indicates that the removal efficiency does not increase uniformly with distance, suggesting potential influences on the distribution of CD and its interaction with phenolic compounds. Furthermore, an increase in CD generally correlates with improved TPh removal efficiency.

Figure 7c presents a 3D representation of the relationship between TPh removal efficiency, distance between electrodes, and time. It reveals that as the distance between electrodes increases from 1 cm to 4 cm, TPh removal efficiency exhibits notable variations. Specifically, maximum efficiency is observed at a distance of 4 cm, reaching approximately 50%. Conversely, at a distance of 1 cm, the efficiency is around 20%, indicating that closer electrode placements may not optimize the removal process effectively. Additionally, the efficiency increases steadily over time, particularly within the first 40 min. After this time, the rate of increase begins to level off. The highest recorded efficiency of approximately 50% occurs at around 60 min, indicating that extended reaction times can contribute to improved removal rates.

3.4.4. Variables Effect on TKN Removal Efficiency

Figure 8 presents the surface plots related to TKN removal, highlighting the effects of reaction time, CD, and distance between electrodes on removal efficiency.

Figure 8a illustrates that as CD increases from 0 to 20 mA/cm² and reaction time extends from 0 to 60 min, TKN removal efficiency significantly increases, peaking at approximately 70% at 15 mA/cm² and 50 min. However, at 20 mA/cm² with a reaction time of 60 min, the efficiency drops to about 65%. This indicates that while a higher CD initially enhance TKN removal, there is a threshold beyond which further increases may result in diminishing returns. Figure 8b illustrates the relationship between TKN removal efficiency, CD, and electrode distance. It reveals that at a distance of 1 cm, TKN removal efficiency starts at a relatively low level. As the distance increases to 2 cm, efficiency rises significantly, reaching approximately 60%. This increase can be attributed to improved mass transfer and enhanced interaction between the reactants and the electrodes [81]. However, at distances greater than 2 cm, the efficiency begins to decline. This reduction may be due to increased resistance to mass transfer, which can hinder the effectiveness of the electrochemical reactions. Additionally, it reveals that TKN removal efficiency increases with rising CD, peaking at approximately 70% when the CD reaches around 10 mA/cm². This enhancement can be attributed to improved electrochemical reactions facilitated by higher CDs, which promote more effective oxidation and reduction processes [81].

Figure 8c further integrates the analysis by examining the relationship between TKN removal efficiency, reaction time, and electrode distance. As the reaction time increases, TKN removal efficiency generally improves. Initially, within the first 30 min, significant increases in efficiency are observed. In terms of electrode distance, it shows that efficiency varies with distance. At shorter distances, such as 1 cm, TKN removal efficiency is relatively low. However, as the distance increases to about 2 to 3 cm, the efficiency peaks, reaching values above 60%. Beyond this distance, the efficiency may plateau or decline slightly, indicating that excessive distance could hinder the effectiveness of the electrochemical processes.

3.4.5. Impact of Variables on Turbidity Removal Efficiency

Figure 9 presents the surface plots related to turbidity removal efficiency.

The analysis of turbidity removal efficiency reveals significant interactions among CD, reaction time, and distance. Figure 9a indicates that as the CD increases, turbidity removal efficiency also increases, particularly at higher CDs, reaching efficiencies of around 90% within the optimal range of 10 to 15 mA/cm². This enhancement can be attributed to the intensified electrochemical reactions that facilitate the aggregation and removal of suspended particles [82]. Moreover, the effect of time is also significant. Turbidity removal efficiency rises sharply within the first 30 min of treatment, suggesting that most turbidity reduction occurs relatively early in the process. Figure 9b illustrates the relationship between turbidity removal efficiency, CD, and electrode distance. As CD increases, turbidity removal efficiency generally improves, reaching efficiencies above 85% at optimal current densities of around 10 to 15 mA/cm². Figure 9c further explores the impact of reaction time and distance, revealing that at 5 min and 1 cm, efficiency is 80%, increasing to 88% at 60 min. Conversely, at 4 cm and 5 min, efficiency decreases from 85% to 74%, indicating that greater distances hinder performance. However, increasing reaction time to 60 min at 4 cm did not affect removal efficiency.

3.4.6. Impact of Variables on ENC Removal Efficiency

Finally, Figure 10 displays the surface plots for the ENC, emphasizing the substantial effects of reaction time, CD, and the distance between electrodes on this response variable. Gaining insights into ENC is essential, as it can facilitate the optimization of other response variables by reducing energy consumption.

In Figure 10a, the analysis of ENC reveals significant insights into how reaction time and CD impact energy usage in EC processes. As both parameters increase, the ENC rises markedly, peaking at 45 kWh/m³ when the current density is set to 20 mA/cm² and the reaction time is extended to 60 min. This peak can be attributed to the greater energy required for ion movement at higher CDs and the additional power needed to facilitate prolonged electrochemical reactions [83]. At lower settings, such as 5 mA/cm² with a reaction time of 5 min, the ENC is recorded at 10 kWh/m³, indicating a moderate energy demand that reflects efficient system operation. The analysis of Figure 10b reveals a clear relationship between ENC, CD, and the distance between the electrodes. It reveals that ENC tends to increase with both CD and distance. As the distance increases, there is a marked escalation in ENC. This increase can be attributed to the greater energy required to facilitate ion movement across a larger distance, necessitating additional power to maintain reaction efficiency [84]. The ENC escalates with increased electrode distance due to the need for more energy to sustain the same level of electrical conductivity and ion transfer. When the distance between electrodes increases, the resistance in the electrolyte solution rises, leading to higher voltage requirements to drive the same current. The increased resistance not only results in greater ENC but also contrasts with trends observed for other response variables, such as TPh, TKN, COD_total, and COD_soluble, where removal efficiencies decrease with increasing distance. Similarly, Figure 10c confirms that ENC escalates with longer reaction times and increased distances, highlighting the need for optimization in EC processes.

3.4.7. Optimization of Operating Parameters and Energy Cost Assessment

Operating Parameters for Fe Electrodes

Response optimization can also be conducted using Minitab 22.2. The optimization process involves configuring the responses to be maximized, minimized, or set to specific target values. In this case, the ENC response is targeted for minimization, while the removal efficiencies of TPh, TKN, COD_total, and COD_soluble are aimed to be maximized.

Table 8. Optimized operational parameters using Fe electrodes.

	Time, min	CD, mA/cm2	Number of Electrode	Distance, cm
Optimized Values	53.4848	15.1104	6	1

below presents the optimized conditions.

The optimal conditions identified for the EC treatment process include a current density of 15.1104 mA/cm², a reaction time of 53.4848 min, an electrode spacing of 1 cm, the use of Fe as the electrode material, and a maximum of six electrodes. Under these conditions, the ENC is calculated to be 14.3146 kWh/m³, with removal efficiencies of 73.2516% for TKN, 92.2516% for turbidity, 54.4583% for TPh, 58.9098% for COD_soluble, and 58.5484% for COD_total.

Figure 11 presents an optimization plot that illustrates the relationships among COD_soluble, COD_total, TKN, turbidity, and TPh removal efficiencies, along with ENC values for the Fe electrode. These Figures are derived from Equations (7)–(12). It is noteworthy that prolonging the reaction time may result in increased ENC due to greater ion weight loss and modifications in the chemical composition of the wastewater. In this study, the costs associated with electricity consumption and electrode usage are critical factors. These costs can fluctuate based on various elements, including wastewater conductivity, composition, global market trends, treatment efficiency, and the type of sludge generated [56]. The total operating cost (OPC) for wastewater treatment is calculated using the following equation [85,86]:

$$OPC=aENC+bELC+cCHC+BiocharCos t+SludgeCost$$

(16)

where OPC represents the total operational cost of EC (in JD/m³ or USD/m³), CHC denotes the consumption of the neutralizing chemical (in kg/m³), a is the cost of electrical energy (in JD/kWh or USD/kWh), and b and c refer to the electrodes and biochar prices (JD/kg or USD/kWh), respectively.

For instance, the estimated OPC for treating 1 m³ of OMW using EC under optimal conditions is based on an electricity cost of 0.13 USD/kWh and an Fe cost of 0.89 USD/kg, with a current density of 15.1104 mA/cm², an effective electrode area of 240 cm², an electric potential of 14.94 V, and an electric current of 3.626 A. The ENC is measured at 14.3146 kWh/m³. The actual energy loss or ELC_actual in the EC cell is measured at 2.25 kg/m³. Additionally, the estimated cost for sludge treatment is to be 0.05 USD/m³, and the cost for adsorbent is approximately 0.00467 USD/m³. Consequently, the total estimated OPC for treating 1 m³ of OMW is approximately 3.92 USD/m³, including all relevant costs.

The Fe electrodes operate at a current density of 15.1104 mA/cm² and a longer reaction time of 53.4848 min, resulting in an energy consumption of 14.3146 kWh/m³. Although the removal efficiencies are comparable, the increased current density and extended reaction duration contribute to the higher energy costs associated with Fe electrodes. The higher over potential and the generation of complex byproducts, such as iron hydroxides, further complicate the energy dynamics for Fe. These results are consistent with the studies conducted by [87,88].

3.5. Performance of the Adsorption Method

In this study, the adsorption process was utilized as a final polishing step following the EC procedure. This approach is beneficial as the EC method effectively reduces pollutant concentrations to levels conducive for subsequent adsorption, thus improving operational efficiency and producing treated OMW that is sufficiently clean for safe reuse. Adsorption is characterized by the tendency of molecules in the liquid phase to adhere to a solid surface, a phenomenon driven by the formation of a low-potential energy region near the solid that increases the molecular density at the surface [84]. In this research, the supernatant obtained after separating the sludge from the EC process was subjected to the adsorption method, specifically targeting OMW treated under the optimal parameters established for the EC process.

Effect of Particle Size

The phenomenon of sorption is significantly influenced by the contact surface area between the sorbent and the liquid phase. To investigate the impact of OS particle size on pollutant removal efficiency, particle sizes of 0.3 mm, 0.15 mm, and 0.075 mm were utilized. Additionally, these particle sizes were applied in the EC process under optimal parameters using iron electrodes, which further enhances the understanding of how OS particle size influences the effectiveness of the treatment method. The relationship between pollutant removal efficiency and OS particle size is illustrated in Figure 12.

The results presented in Figure 12 emphasize the significant impact of OS particle size on pollutant removal efficiency during the adsorption process. At a particle size of 0.075 mm, the removal efficiencies, while the highest among the tested sizes, remain relatively low overall. Specifically, TPh were removed at 9.21%, TKN at 20.31%, COD_total at 18.47%, COD_soluble at 29.71%, and turbidity at 3.82%. The higher removal efficiencies observed with smaller particle sizes can be attributed to several factors. Firstly, smaller particles possess a greater surface area-to-volume ratio, which enhances their availability for adsorption interactions. This increased surface area allows for more active sites where pollutants can bind, facilitating a more effective removal process. Additionally, the increased reactivity of smaller particles may lead to enhanced interaction with the contaminants, improving the adsorption process [31,43].

Conversely, at a larger particle size of 0.15 mm, the removal efficiencies decline, with TPh at 10.83%, TKN at 17.24%, COD_total at 12%, COD_soluble at 23.81%, and turbidity at 8.03%. This reduction can be explained by the decreased surface area available for adsorption, which limits the interaction between the particles and the pollutants. The lower removal rates for TPh and TKN indicate that larger particles may not engage as effectively with the pollutants in the liquid phase. At a particle size of 0.3 mm, the removal efficiencies decrease even further, with TPh at 12.5%, TKN at 8.47%, COD_total at 7.52%, COD_soluble at 14.29%, and turbidity at 9.52%. This trend is consistent with expectations that larger particles provide less surface area for adsorption, resulting in the lowest removal efficiencies across all measured parameters.

Overall, the data illustrate that smaller OS particle sizes (0.075 mm) enhance pollutant removal due to their larger surface area and higher reactivity, while larger particles (0.3 mm) result in reduced efficiencies.

4. Combined Treatment Process

The performance of EC, adsorption, and the combined electrocoagulation and adsorption (ECA) methods in reducing pollutants from OMW was evaluated under optimal conditions. These conditions included six iron electrodes, an electrode distance of 1 cm, a current density of 17.64 mA/cm², an operating time of 52.27 min, and an OS particle size of 0.075 mm. The comparative results are illustrated in Figure 13.

The results depicted in Figure 13 reveal that the ECA method achieved the highest pollutant reduction efficiencies across all measured parameters. Specifically, the removal efficiencies were as follows: TPh at 59.3%, TKN at 79.18%, COD_total at 72.88%, COD_soluble at 70.04%, and turbidity at 96.7%. These findings highlight the enhanced effectiveness of integrating EC with adsorption processes.

In contrast, the EC method alone demonstrated substantial removal efficiencies, with TPh at 54.78%, TKN at 74.1%, COD_total at 59.81%, COD_soluble at 58.5%, and turbidity at 93.36%. While these results are noteworthy, they do not match the efficiencies achieved through the combined ECA method, indicating that the integration of both processes significantly enhances pollutant removal.

On the other hand, when considering the adsorption method alone, which utilized an OS particle size of 0.075 mm, the removal efficiencies were considerably lower. The results showed TPh removal at 9.21%, TKN at 20.31%, CODtotal at 18.47%, CODsoluble at 29.71%, and turbidity at 4.26%. This contrast underscores the limitations of using adsorption as a standalone treatment for OMW.

These findings are consistent with previous research that indicated the superiority of combining EC and adsorption processes for improved pollutant removal efficiency [43,89,90]. The significant enhancements observed in the ECA method reinforce the potential of integrated treatment approaches to optimize wastewater treatment systems, suggesting that such combinations could lead to more effective and sustainable solutions for managing OMW.

5. Conclusions

This study utilized RSM to conduct a comprehensive evaluation of the effectiveness of EC in treating OMW, followed by adsorption using OS under optimal conditions. A total of 60 experiments were performed based on a CCD matrix, which allowed for an in-depth examination of various operational parameters, including reaction time, current density, electrode spacing, and the number of electrodes. Under optimal conditions for iron electrodes, which included a reaction time of 53.48 min and a current density of 15.1104 mA/cm², the removal efficiencies achieved were significant: 54.46% for TPh, 73.25% for TKN, and 92% for turbidity. Additionally, the COD_total removal was recorded at 58.55%, while the COD_soluble removal reached 58.91%. These results underscore the effectiveness of the EC process in reducing both organic and inorganic pollutants. Additionally, the ENC for the EC process was measured at 14.3146 kWh/m³, with OPC amounting to USD 3.92/m³.

The ECA yielded the highest pollutant reduction efficiencies across all measured parameters. Specifically, TPh removal improved to 62.63%, TKN removal reached 77.52%, and turbidity reduction was at 83.73%. Furthermore, COD_total removal increased to 72.88%, and COD_soluble removal was recorded at 70.04%. These results highlight the enhanced effectiveness of combining these two treatment processes. In contrast, the adsorption method alone, utilizing an OS particle size of 0.075 mm, demonstrated significantly lower removal efficiencies, with TPh at 9.21%, TKN at 20.31%, COD_total at 18.47%, COD_soluble at 29.71%, and turbidity at 4.26%. This stark contrast underscores the limitations of using adsorption as a standalone treatment method for OMW.

In summary, the findings of this research confirm that the ECA process significantly enhances pollutant removal efficiency, including both COD_total and COD_soluble. This integrated approach not only optimizes the removal of contaminants but also presents a promising solution for improving the sustainability of wastewater treatment practices. By leveraging the strengths of both methods, this study offers a viable pathway for advancing treatment technologies in the management of OMW, ultimately contributing to better environmental outcomes.

6. Recommendations and Future Research

This research studied and optimized the performance of a combined batch system for the treatment of olive mills wastewater. The combined system compromises two steps: electrocoagulation as a pre-treatment step and adsorption as a post-polishing step. Based on the promising findings including the overall pollutants’ removal efficiencies, the authors present several recommendations, as insights for future research directions and a guide for new studies in this field. These recommendations highlight many gaps that require further investigation.

• The literature presents many studies dealing with EC combined treatment processes. However, most of these studies are still at the lab-scale. Intensive research should consider pilot or industrial scale processes to optimize them and apply at larger scales in real wastewater industrial treatment systems;
• The choice of electrode material has a crucial role in the overall performance of EC process. Al and Fe are the most commonly used electrodes material. Al electrodes are generally more efficient in pollutant removal; however, their higher cost and their generated sludge requires specialized management. For this reason, the use of other cheap and efficient materials needs further extensive research;
• The main drawback of EC treatment system is the use of electricity in the redox reaction causing a substantial increase in the cost and cause an increased environmental pollution from fossil fuels emissions. Accordingly, the employment of cheap renewable energy sources such as solar, wind, or tidal energy to power EC combined systems is of first priority;
• Most of the published research on EC treatment systems is still at the lab-scale. More research is needed to assess the performance of continuous EC combined systems in treating different industrial wastewater, such as heavy metals and organic contaminants;
• The literature survey indicates that mathematical modeling of batch and continuous EC combined processes is still limited. Accordingly, more research is needed to develop models that describe the experimental results of EC combined systems to facilitate their scale-up;
• Electrode passivation is a significant problem that reduces the electrodes’ performance, due to the accumulated sludge on the electrode surface. Optimization of the parameters affecting electrode passivation such as electrode shape, arrangement, applied voltage, current density, and others needs more studies;
• The use of other combined EC treatment systems such as the integration of advanced oxidation or membrane technologies is necessary to enhance the EC treatment process further and expand its scope.