Cost-Effective Processes for Denim Production Wastewater: Dual Criterial Optimization of Techno-Economical Parameters by RSM and Minimization of Energy Consumption of Photo Assisted Fenton Processes via Direct Photovoltaic Solar Panel Integration

Abstract

Denim production wastewater is an industrial wastewater with a high organic pollutant content. The aim of this study was to improve a cost-effective method via solar panel integration to the photo Fenton process (PFP) and photo electrochemical Fenton process (PEFP) for removing high chemical oxygen demand (COD) from denim production wastewater. To determine process parameter values, the double criterial optimization option was used. The results that maximized the COD removal efficiency and minimized the operating cost of two oxidation processes were determined by response surface methodology (RSM). Optimum operation conditions for the PFP process were 3.18 initial pH, 2.3 g/L Fe²⁺ concentration, and 27 g/L H₂O₂ concentration while they were 3.00 initial pH, 27.06 A/m² current density, and 28.16 g/L H₂O₂ concentration for PEFP. At the optimum conditions, COD and the total organic carbon (TOC) removal efficiency of PFP were 85% and 61%, respectively. They were determined as 90% and 73% in PEFP. Carbon oxidation state (COS) and average oxidation state (AOS) parameters were used to obtain the biodegradation capability of organic materials. The biodegradability capability of wastewater was observed as high after the Fenton processes. As a result of the optimization of technical parameters, total operating cost was obtained as USD 14.62/m³ (USD 4.25/kgCOD_removed) in PFP and USD 13.79/m³ (USD 3.73/kgCOD_removed) in PEFP. After the integration of the photovoltaic solar panel to the processes, the total operating cost of PFP and PEFP decreased in a ratio of 61% and 64%, respectively.

1. Introduction

Denim is a type of fabric used in jean production. During the production of denim jeans, chemicals that prevent the shrinkage of the jeans, stabilizers, and abrasives are used. Denim jeans, which are washed with chemicals to gain these properties, are then rinsed at least two times. Each denim product goes through at least three washing processes, one of which is chemical washing and two of which are rinsing. Depending on the size of the washing machines, 300 to 500 L of water is used for each wash. This shows how large the amount of water and occurring wastewater is in denim jean production. In short, there is a very high-water requirement in the production of denim jeans, and accordingly, a high amount of wastewater is produced after use. Additionally, this wastewater occurring in the production processes has high concentrations of organic components such as COD, TOC, BOD, and color. COD is an especially important indicator of the degree of pollution in the effluent. COD is determined using a strong chemical oxidant under standard conditions. Biodegradable organic compounds, non-biodegradable compounds, and inorganic oxidizable compounds contribute to COD [1]. Therefore, the removal of COD from wastewaters has a great importance compared to other parameters. The discharge of textile wastes into water bodies causes toxicity, poor degradability, and the deterioration of environmental quality, such as the reduced self-treatment capacity of receiving waters [2]. Because of these problems, wastewaters occurring in the textile sector have to be treated before discharging them into water bodies. In the treatment of these kind of wastewaters, some methods such as coagulation [3], electrocoagulation [4], and biological [5] and oxidation processes [6] are used. The Fenton process is one of the advanced treatment techniques in which Fe²⁺ and H₂O₂ are used together to form homogeneous OH^·. In recent years, the usage of Fenton/Fenton-adopted processes has increased in a wide scope [7,8]. While Fenton processes are effective in removing high amounts of organic pollutants in wastewater, they are also effectively used in the treatment of different types of wastewater such as pharmaceutical industry wastewater [9], tannery wastewater [10], dye manufacturing wastewater [11], landfill leachate wastewater [12], coking wastewater [13], etc. Fenton, which is one of the processes used in the treatment of such wastewater, differs from other processes with its various advantages such as treating wastewater with a high organic content with high efficiency, the ease of finding reagents, being effective in a short time [14], relatively low cost and simple operation [15], chemicals of Fenton reagents have low toxicity, and the process works in basic conditions such as at room temperature and atmospheric pressure [16]. To develop the capability of organic material degradation by the Fenton process and to obtain a greater hydroxyl radical formation that offers the usage of a lower concentration of ferrous ion in the Fenton process, UV irradiation was added to the Fenton process, which is called photo-Fenton [17,18]. As a result, it is stated that the organic pollutant removal efficiency of the photo-fenton process is more effective than the classical fenton process [19]. In the photo-Fenton process, the formation of OH occurs by the following reactions (Equations (1) and (2)) [20]:

$${\mathrm{H}}_2{\mathrm{O}}_2+\mathrm{h}𝓋\to {\mathrm{OH}}^{·}+{\mathrm{OH}}^{·} (\mathrm{photolysis} \mathrm{of} \mathrm{hydrogen} \mathrm{peroxide})$$

(1)

$${\mathrm{Fe}}^{3+}+{\mathrm{H}}_2\mathrm{O}+\mathrm{h}𝓋\to {\mathrm{Fe}}^{2+}+{\mathrm{OH}}^{·}+{\mathrm{H}}^{+} (\mathrm{photo-Fenton} \mathrm{reaction})$$

(2)

The main difference between photo-Fenton and photo-electrochemical Fenton is the additional method of adding Fe²⁺ ions into the wastewater. In the photo-Fenton process, Fe²⁺ ions are added externally, while in the photo-electrochemical Fenton process, Fe²⁺ is added by dissolving it from the Fe electrodes by electrolysis. The electro Fenton process mineralizes almost any organic pollutant in wastewater into carbon dioxide, water, and inorganic species at room temperature and atmospheric pressure. In this respect, the electro-Fenton process can be expressed as cold incineration and allows the effective treatment of wastewater containing toxic, non-biodegradable or persistent substances that cannot be oxidized by conventional processes [16]. To enhance the removal efficiency of Fenton processes, UV is added to the electrochemical Fenton, which is called photo-electrochemical Fenton [21,22]. This process takes place in two ways: (i) the production of hydroxyl radicals by the photo reduction of Fe(OH)²⁺ (Equation (3)) and (ii) the production of Fe²⁺ by the photolysis of Fe³⁺ (Equation (4)):

$$\mathrm{Fe}\left(\mathrm{OH}\right)\stackrel{2+\mathrm{h}𝓋\left({\mathrm{k}}_4\right)}{\to }{\mathrm{Fe}}^{2+}+{}_{}{}^{·}\mathrm{O}\mathrm{H}$$

(3)

$$\mathrm{R}\left({\mathrm{CO}}_2\right)-\mathrm{Fe}\left(\mathrm{III}\right)+\mathrm{h}𝓋\to \mathrm{R}\left({}_{}{}^{·}\mathrm{C}{\mathrm{O}}_2\right)+\mathrm{Fe}\left(\mathrm{II}\right)\to {}_{}{}^{·}\mathrm{R}+{\mathrm{CO}}_2$$

(4)

The operating cost of Fenton/electrochemical-Fenton processes varies according to the current density given to the system, the amount of Fenton reagent supplied to the system, and the energy consumption depending on the electrolysis time and the amount of sludge formed. The most important parameter that creates the operating cost in the system is energy consumption. For this reason, renewable energy studies with integrated electrochemical processes have increased in recent years. Solar panel-integrated electrochemical processes give especially good results for minimizing energy consumption and operating cost [23]. Minimizing energy consumption, which constitutes a very high part of operating costs, contributes to the spread of these processes.

Although there are many articles about Fenton/Fenton-like processes in the literature, studies on the optimization of the techno-economical parameters of the processes are limited. In addition, in the operating cost analysis of Fenton processes, it was determined that the economic parameters of the process such as sludge disposal and chemical costs were not examined in detail. It is thought that this study will make important contributions to the literature in this respect. Despite Fenton/Fenton-adapted processes being effective for organic pollutant removal, the operating costs need to be evaluated in terms of applicability. For this reason, examining the treatment efficiency as well as the operating costs of the treatment techniques provides convenience to the treatment technology selectors in practice. In this study, COD removal, which is the wastewater pollution parameter, and the operating costs of photo-Fenton and electrochemical-Fenton processes were determined. Results that maximize COD removal efficiency and minimize operating costs were developed by the RSM, and after the optimization studies, the effect of photovoltaic solar panel integration into the photo-assisted processes was analyzed.

2. Materials and Methods

2.1. Experimental Flow-Chart and Characterization of Denim Jean Production Wastewater

The experimental chart of the study is given in Figure 1. Denim jean production wastewater was collected from washing machines of the fabric. The wastewater sample used in the experiments was brought to the laboratory under suitable conditions and stored in the refrigerator at about 4 degrees. Afterwards, preliminary experiments were carried out with the samples reaching room temperature, and the ranges of parameters affecting the Fenton processes and COD removal efficiency were determined. The experimental design was created by entering these intervals into the statistical software. The experiments were carried out according to the experimental design. The obtained results were entered into the software, ANOVA analysis was performed, and techno-economic parameters were optimized.

The characterization of denim jean processing wastewater is given in

Table 1. Raw wastewater characterization.

Parameter	Value/Concentration	Parameter	Value/Concentration
pH	6.82 ± 0.32	TDS	4.89 ± 0.40 g/L
Conductivity	5.16 ± 0.21 mS/cm	Color 460 nm	16 m−1
COD	4100 ± 2000 mg/L	Color 525 nm	35 m−1
TOC	2780 ± 200 mg/L	Color 620 nm	47 m−1
SS	3200 ± 190 mg/L

. Raw wastewater was supplied from a denim processing factory in Düzce/Türkiye.

2.2. PFP and PEFP Experimental Procedure

In the PFP experiments, 750 mL volume of denim jean production wastewater was added to the reactors (Figure 2a). The power of the UV lamp was 40 W. FeSO₄·7H₂O (99.5%) and H₂O₂ (hydrogen peroxide, 50% W/W) were used as a Fenton reagent in the Fenton experiments. Firstly, the pH of the wastewater was adjusted by adding 1 N H₂SO₄. FeSO₄·7H₂O was initially added to the pH-adjusted sample, and H₂O₂ was added in the second step. At the same time, after the addition of FeSO₄·7H₂O and H₂O₂, the samples were mixed with recirculation by a peristaltic pump (reaction time was optimized as 30 min). When the determined reaction times were completed, the pH of the wastewater sample was increased to 6.5–7.5 by 1 N NaOH solution. Then, the pollutants in the wastewater were allowed to settle for 30 min. The effluent was filtered using 0.45 µm filter papers, and COD, TOC, and other analyses were performed. All experiments were conducted at room temperature, and all the chemicals used in the experiments were of analytical grade.

In the PEFP experiments, 750 mL volume of denim jean production wastewater was added to the photo-electrochemical Fenton reactor as seen in Figure 2b. Reactor dimensions were 18 cm × 10 cm × 10 cm (H × L × B). The power of the UV lamp was 40 W. The outer surface of the reactor was covered with impermeable material so as not to reduce the effect of the UV light. Ten iron electrodes with dimensions of 1 cm × 20 cm and a thickness of 2 mm were used for the electrochemical Fenton process in the reactor. The samples were mixed with recirculation by a peristaltic pump (reaction time was optimized as 30 min). The pH of the wastewater was adjusted by adding 1 N H₂SO₄. A direct current was applied to the solution, and H₂O₂ was added. The samples were mixed with recirculation by a peristaltic pump (reaction time was optimized as 30 min). When the determined reaction times were completed, the pH of the wastewater sample was increased to 6.5–7.5 by 1 N NaOH solution. Then, the pollutants in the wastewater were allowed to settle for 30 min. The effluent was filtered using 0.45 µm filter papers, and COD, TOC, and other analysis were performed. All experiments were conducted at room temperature.

2.3. Photovoltaic Solar Panel Integration to the Fenton Processes

A photovoltaic solar panel was added to the UV-assisted Fenton processes after the optimization studies. The electrochemical reactor and solar panel integrated into the system are given in Figure 3.

Specifications of solar panel are given in

Table 2. Technical specifications of the solar panel.

Parameter	Technical Value	Parameter	Technical Value
Solar module type	SPE250	Max system voltage	1000 VDC
Max power (Pmax)	250 W	Dimensions	1001 × 1665 × 42 mm
Max power voltage (Vmp)	30.50 V	Application class	Class A
Max power current (Ipmax)	8.2 A	Weight	18.5 kg
Open circuit voltage (Voc)	37.8 V	Power tolerance up to	+4.9 W
Short circuit voltage (Isc)	8.7 A	Measurement tolerance	±3%

. Voltage and ampere produced by the solar panel were detected by the multimeter.

2.4. Analytic Methods

All analytical parameters were measured by the procedures described by the Standard Methods for the Examination of Water and Wastewater (SM) [24] and are given in

Table 3. Analytical methods.

Parameter	Method	Method No.	Instrument
COD	Photometric	SM 5220	Hach DR5000 UV-VIS
TOC	High temperature combustion method	SM 5310-B	TOC analyzer with NDIR detector
Suspended Solid	Gravimetric	SM 2540-D	Vacuum filtration unit
Turbidity	Photometric	SM 2130-B	Hach DR5000 UV-VIS
pH	Electrometric	SM 4500-B	Hanna Ins.
Conductivity	Electrometric	SM 2510-B	Hach 7100e
Color	Photometric	EN ISO 7887	Hach DR5000 UV-VIS

.

2.5. Response Surface Methodology (RSM)

The Box–Behnken Design was used for the optimization of PFP and PEFP for the degradation of COD from the denim jean production wastewater. In PFP, the parameters of pH (X₁), Fe²⁺ concentration (X₂), and H₂O₂ (hydrogen peroxide) concentrations (X₃) were handled as the variable, while pH (X₁), current density (X₂), and hydrogen peroxide concentration (X₃) were used in the PEFP. This experimental design involved 15 runs, based on a three-level Box–Behnken factorial design. The Design Expert Trial version was used for the statistical analysis of RSM. The parameter ranges and levels are given in

Table 4. The ranges and levels for PFP and PEFP.

			PFP			PEFP
				Levels			Levels
Coded Variables (Xi)	Factors	Unit	Low (−1)	Center (0)	High (+1)	Low (−1)	Center (0)	High (+1)
(X1)	pH	-	3	4.5	6	3	4.5	6
(X2)	Fe2+/C.D.	g/L-A/m2	2.3	3.06	3.83	24	48	72
(X3)	H2O2	g/L	23	38.5	54	23	38.5	54

.

The experimental data were analyzed by the RSM procedure, as seen in Equation (5). In Equation (5), Y is the response (COD removal efficiency (%)) or total operating cost (USD/m³); ${\mathsf{\beta }}_0{\mathsf{\beta }}_{\mathrm{i}} \left(\mathrm{i}=1,2,3\right),{ \mathsf{\beta }}_{\mathrm{ij}}\left(\mathrm{i}=1,2,3;\mathrm{j}=1,2,3\right)$ are the model coefficients; ${\mathrm{X}}_{\mathrm{i}}$ and ${\mathrm{X}}_{\mathrm{j}}$ are the coded independent variables.

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

(5)

2.6. Equations

Current density is an important parameter in electrochemical wastewater treatment techniques. Current density was calculated by Equation (6):

$$\mathrm{J}=\mathrm{I}/\mathrm{A}$$

(6)

In the equation, J is current density (A/m²), I is current (A), and A is active anode surface area (m² or cm²).

To calculate the COD removal efficiency of Fenton processes from denim jean production wastewater, Equation (7) was used:

$${\mathrm{E}}_{\mathrm{COD}}(\%)=\left(1-\frac{{\mathrm{C}}_{\mathrm{t}}}{{\mathrm{C}}_0}\right)\times 100$$

(7)

E is the removal efficiency of COD, C₀ is the initial concentration of COD, and C_t is the final concentration of COD.

To determine the energy consumption of the electrochemical systems, Equation (8) was used [25]:

$$\mathrm{E}\left(\frac{\mathrm{kWh}}{{\mathrm{m}}^3}\right)=-\frac{1}{\forall }\underset{0}{\overset{\mathrm{I}}{{\displaystyle \int }}}\mathrm{VA} \mathrm{dt}$$

(8)

In the equation, E is energy consumption, (Wh/m³), V is voltage (V), A is current (A), t is electrolysis time (hour), $\forall$ is volume (lt or m³).

The unit prices used in operating cost calculations are shown in

Table 5. Unit prices of chemicals, electrodes, and electricity used in the calculations.

Parameter	Unit Price	Parameter	Unit Price
a-Energy	USD 0.22/m3	e-NaOH	USD 1.74/m3
b-Fe Electrode	USD 1.00/kg	f-H2SO4	USD 1.53/m3
c-FeSO4·7H2O	USD 1.18/kg	g-Sludge Disposal	USD 0.19/kg
d-H2O2	USD 1.76/m3

.

The total operating cost of PFP consists of chemicals used before and after oxidation reactions and energy and sludge disposal costs (Equation (9)).

The abbreviations used in the equations are as follows.

OCPFP	Operating cost of PFP (USD/m3);
OCPEFP	Operating cost of PEFP (USD/m3);
EUV	Energy consumption of UV lamp (kWh/m3);
EDC	Energy consumption of DC power supply (kWh/m3);
EPP	Energy consumption of peristaltic pump (kWh/m3);
CFeSO4	Consumption of FeSO4 (kg/m3);
CH2O2	Consumption of H2O2 (m3/m3);
CNaOH	Consumption of NaOH (kg/m3);
CH2SO4	Consumption of H2SO4 (kg/m3);
CFe el.	Consumption of Fe electrode (kg/m3);
Vsludge	Sludge volume (kg/m3).

$${\mathrm{OC}}_{\mathrm{PFP}}=\mathrm{a}\times {\mathrm{E}}_{\mathrm{UV}}+\mathrm{a} \times {\mathrm{E}}_{\mathrm{PP}}+\mathrm{c} \times { \mathrm{C}}_{{\mathrm{FeSO}}_4}+\mathrm{d} \times { \mathrm{C}}_{{\mathrm{H}}_2{\mathrm{O}}_2}+\mathrm{e} \times { \mathrm{C}}_{\mathrm{NaOH}}+\mathrm{f} \times { \mathrm{C}}_{{\mathrm{H}}_2{\mathrm{SO}}_4}+\mathrm{g} \times { \mathrm{D}}_{\mathrm{sludge}}$$

(9)

For PEFP, the operating cost consisted of chemical, electrode, energy consumption, and sludge disposal cost (Equation (10)).

$${\mathrm{OC}}_{\mathrm{PEFP}}=\mathrm{a} \times { \mathrm{E}}_{\mathrm{DC}}+ \mathrm{a}\times { \mathrm{E}}_{\mathrm{UV}}+\mathrm{a} \times { \mathrm{E}}_{\mathrm{PP}}+\mathrm{b} \times { \mathrm{C}}_{\mathrm{Fe} \mathrm{Electrode}}+\mathrm{d} \times { \mathrm{C}}_{{\mathrm{H}}_2{\mathrm{O}}_2}+\mathrm{e} \times { \mathrm{C}}_{\mathrm{NaOH}}+\mathrm{f} \times { \mathrm{C}}_{{\mathrm{H}}_2{\mathrm{SO}}_4}+\mathrm{g} \times { \mathrm{V}}_{\mathrm{sludge}}$$

(10)

3. Results and Discussion

3.1. Statistical Analysis of COD Removal Efficiency of PFP and PEFP

According to the Box–Behnken Design developed in the RSM tool of Design Expert Software, PFP and PEFP processes were optimized for maximizing COD removal efficiency and minimizing total operating cost.

To determine the significance of the model and the effects of the parameters, DF (degrees of freedom of variance source), SS (sum of squares), MS (mean of squares), F values (F-value of variance source), and p values (probability of error to be significant) were obtained from the ANOVA analysis. Other statistical parameters such as R² (coefficient of regression), adjusted R², predicted R², PRESS (predicted residual error sum of squares), and C.V. (coefficient of variation) values are given in

Table 6. ANOVA results of PFP and PEFP for the COD removal (%) response (quadratic model).

	PFP					PEFP
Source	SS	DF	MS	F Value	p Value	SS	DF	MS	F Value	p Value
Model	1.15	9	0.13	116.59	<0.0001	0.94	9	0.10	225.77	<0.0001
X1-pH	0.99	1	0.99	903.68	<0.0001	0.83	1	0.83	1775.51	<0.0001
X2-Fe2+/i	0.031	1	0.031	28.41	0.0031	0.063	1	0.063	135.51	<0.0001
X3-H2O2	0.029	1	0.029	26.18	0.0037	4.5 × 10−4	1	4.5 × 10−4	0.97	0.3704
X1X2	1.6 × 10−3	1	1.6 × 10−3	1.45	0.2818	3.6 × 10−3	1	3.6 × 10−3	7.74	0.0388
X1X3	0.014	1	0.014	13.09	0.0152	2.25 × 10−4	1	2.25 × 10−4	0.48	0.5177
X2X3	4.9 × 10−3	1	4.9 × 10−3	4.45	0.0886	0.011	1	0.011	23.71	0.0046
X12	0.075	1	0.075	68.16	0.0004	2.792 × 10−3	1	2.792 × 10−3	6.00	0.0579
X22	2.3 × 10−5	1	2.308 × 10−5	0.021	0.8905	0.022	1	0.022	47.69	0.0010
X32	1.869 × 10−3	1	1.869 × 10−3	1.70	0.2492	0.021	1	0.021	44.67	0.0011
Residual	5.5 × 10−3	5	1.1 × 10−3			2.325 × 10−3	5	4.650 × 10−4
Lack of fit	5.3 × 10−3	3	1.767 × 10−3	17.67	0.0540	9.25 × 10−4	3	3.083 × 10−4	0.44	0.7491
Pure error	2 × 10−4	2	1 × 10−4			1.4 × 10−3	2	7 × 10−4
Cor total	1.16	14				0.95	14
	PFP					PEFP
	R2	0.99	Std. Dev.	0.033		R2	0.99	Std. Dev.	0.022
	Adj R2	0.97	Mean	0.63		Adj R2	0.99	Mean	0.62
	Pred R2	0.93	C.V. (%)	5.23		Pred R2	0.98	C.V. (%)	3.46
	A.P.	31.11	PRESS	0.085		A.P.	46.57	PRESS	0.018

.

R², adjusted R², and predicted R² were checked to determine the adequacy of the models. The quadratic model had a high signal, which is thought to explain the Fenton processes for COD removal response. In the PFP process, R², adjusted R², and predicted R² were determined as 0.99, 0.97, and 0.93, respectively. In the PEFP process, R², adjusted R², and predicted R² were determined as 0.99, 0.99, and 0.98, respectively. The larger the value of the multivariate correlation coefficient R², the better the correlation [13]. R² values for PFP and PEFP were close to 1, which indicates that the model has a good adequacy, and the responses obtained using the RSM model agreed well with the experimental data.

Statistically, the difference of adjusted R² and predicted R² is suggested to be <0.2, while C.V. is suggested to be <10%. As seen from Table 6, the difference of adjusted R² and predicted R² was smaller than 0.2, and C.V. was smaller than 10% (5.23% for PFP and 3.46% for PEFP). A high adjusted R² value indicates that the model terms are quite important.

The F-value of model was 117 and 226 for PFP and PEFP, respectively, with a very low probability value (<0.0001). This indicates that the model is statistically well-fitted and shows that the model is significant.

Lack of fit is an important factor for evaluating the reliability of the model [13]. The “lack fit p-value” of the model for PFP and PEFP was identified to be 0.05 and 0.75, respectively, which implied that there was no significant error in the data.

Adequacy of precision (A.P.) measures the signal to noise ratio. A ratio greater than 4 is desirable. In the study, the A.P. of PFP and PEFP was determined to be 31.11 and 46.57, respectively. Additionally, A.P. was greater than 4 in all processes.

To determine the fitting quality of the model at each point in the design, the PRESS value was used. This value is the sum of the squared differences between the estimated and actual values over all the points [26]. PRESS was obtained as 0.085 and 0.018 in PFP and PEFP, respectively.

A “Prob > F” less than 0.05 shows that the model terms are significant. Values greater than 0.10 indicate that the model terms are not significant. In this case, ${\mathrm{X}}_{1,}{ \mathrm{X}}_2, {\mathrm{X}}_3,{\mathrm{X}}_1{\mathrm{X}}_3,{\mathrm{X}}_1^2$ are significant model terms for PFP, and ${\mathrm{X}}_1,{ \mathrm{X}}_2{ \mathrm{X}}_1{\mathrm{X}}_2,{ \mathrm{X}}_2{\mathrm{X}}_3,{ \mathrm{X}}_2^2,{\mathrm{X}}_3^2$ are significant model terms for PEFP.

The significance of the main factors on the COD removal efficiency was: pH > Fe²⁺ concentration > H₂O₂ concentration in the PFP process, and pH > current density > H₂O₂ concentration in the PEFP process.

In conclusion, this model is reliable for optimizing PFP and PEFP parameters to achieve the highest COD removal.

The predicted values of the responses were obtained from the quadratic model. The response equations for the removal efficiency of COD by the PFP are given in Equation (11).

Here, Y₁ is the predicted COD removal efficiency for PFP (0 < Y₁ ≤ 100%), X₁, X₂, and X₃ are the pH (3 ≤ X₁ ≤ 6), Fe²⁺ concentration (2.3 g/L ≤ X₂ ≤ 3.83 g/L), and hydrogen peroxide concentration (23 g/L ≤ X₃ ≤ 54 g/L), respectively.

$${\mathrm{Y}}_1=\mathrm{COD} \mathrm{Removal} \mathrm{Efficiency} (\%)-\mathrm{PFP}=0.7-0.3525{\mathrm{X}}_1+0.0625{\mathrm{X}}_2+0.06{\mathrm{X}}_3+0.02{ \mathrm{X}}_1{\mathrm{X}}_2+0.06{ \mathrm{X}}_1{\mathrm{X}}_3-0.035{\mathrm{X}}_2{\mathrm{X}}_3-0.1425{\mathrm{X}}_1^2-0.0025{\mathrm{X}}_2^2+0.0225{\mathrm{X}}_3^{2 }$$

(11)

In PFP, although pH (p < 0.05) is the most important parameter affecting the COD removal efficiency in the ANOVA analysis, the coefficient of the pH parameter had a negative sign (b₁ = −0.35). This determines the conclusion that an increase in pH reduces the efficiency of COD removal. Likewise, in Equation (11), it is seen that Fe²⁺ concentration and hydrogen peroxide concentration, which are the main parameters, have a positive coefficient (b₂ = +0.063), (b₃ = +0.06). Increasing the Fe²⁺ concentration and hydrogen peroxide concentration increased the COD removal efficiency.

The predicted values of the responses were obtained from the quadratic model. The response equations for the removal efficiency of COD by PEFP are given in Equation (12).

$${\mathrm{Y}}_2=\mathrm{COD} \mathrm{Removal} \mathrm{Efficiency} (\%)-\mathrm{PEFP}=0.72-0.32{\mathrm{X}}_1+0.088{\mathrm{X}}_2+0.0075{\mathrm{X}}_3+0.03{ \mathrm{X}}_1{\mathrm{X}}_2+0.0075{\mathrm{X}}_1{\mathrm{X}}_3+0.053{\mathrm{X}}_2{\mathrm{X}}_3-0.028{\mathrm{X}}_1^2-0.078{\mathrm{X}}_2^2-0.075{\mathrm{X}}_3^2$$

(12)

Here, Y₂ is the predicted COD removal efficiency for PEFP (0 < Y₂ ≤ 100%), X₁, X₂, and X₃ are the pH (3 ≤ X₁ ≤ 6), current density (24 A/m² ≤ X₂ ≤ 72 A/m²), and hydrogen peroxide (23 g/L ≤ X₃ ≤ 54 g/L), respectively.

In PEFP, although pH (p < 0.05) is the most important parameter affecting the COD removal efficiency in the ANOVA analysis, the coefficient of the pH parameter had a negative sign (b₁ = −0.32). Increasing the pH decreased the efficiency of COD removal. Likewise, in Equation (12), it is seen that Fe²⁺ concentration and hydrogen peroxide concentration, which are the main parameters, have a positive coefficient (b₂ = +0.088), (b₃ = +0.0075).

As was mentioned by the researchers [27], if the normality plot of the residuals is close to a straight line, there is no need for a conversion of the response. In this study, the normality plot of residuals was close to a straight line, showing that the response did not need converted (Figure 4a,d).

As is seen in Figure 4b,e, the actual values obtained from the experiment were compatible with the predicted values of the model response for both the PFP and PEFP processes.

Main effects of X₁ (A), X₂ (B), and X₃ (C) parameters for PFP and for PEFP are given in Figure 4c,f. It is seen in Figure 4c,f, that, in PFP and PEFP processes, the main effects of X₂ (B) and X₃ (C) had a limited effect on the obtained quadratic model for the COD response. X₁ (A)-pH was the most effective parameter in all processes.

3.2. Interactive Effects of pH, Fe²⁺ Concentration/Current Density, H₂O₂ Concentration on PFP and PEFP for COD Degradation

After the ANOVA evaluation, 2D graphs were created for the interaction parameters. The interactive effects of pH/Fe²⁺ concentration on PFP and pH/current density on PEFP are given in Figure 5a and Figure 6a, respectively. COD removal efficiency positively affected pH for all Fenton processes because of the effect of occurring hydroxyl radicals on low pH values. Low pH values and a high Fe²⁺ concentration or current density had a positive effect on the high COD removal efficiency.

COD removal efficiency was >80% at the pH of 3 in PFP and PEFP. In PEFP, the pH of the solution was an important factor for both the electrolytic generation of H₂O₂ and in the production of free radicals. In the study, the removal efficiency of COD from textile wastewater by the electrochemical-Fenton process at optimum conditions was 82.1% (COD_i: 1310 mg/L, pH: 3, retention time (R.T.): 40 min, current density (C.D.): 4.76 mA/cm²) [28]. In another study, the COD degradation efficiency of photo-electro-Fenton at optimum conditions (pH: 3, C.D.: 0.30 A/dm², C_H2O2: 300 mg/L, electrode distance: 0.75 cm, UV source: 32 W, wave length: 254 nm, R.T.: 4 h) was 97% [12]. Acidic conditions are most effective in Fenton processes for the degradation of COD, as mentioned by other researchers [11,13,29,30].

The effect of pH/H₂O₂ concentration on the COD removal efficiency of PFP and PEFP is given in Figure 5b and Figure 6b, respectively. The pH parameter is more effective than the H₂O₂ concentration in all Fenton processes. Lower pH values provided an effective COD removal efficiency. At a pH > 5, the efficiency of the electro-Fenton process decreased rapidly. This is because the H₂O₂ is unstable in the solution at this pH. When pH > 7, H₂O₂ rapidly decomposes into oxygen and water [31,32,33]. The 2D plots show the same trend for both Fenton processes. The effect of the H₂O₂ concentration is thought to be limited, as the overdosing of H₂O₂ caused H₂O₂ to self-destruct into water and oxygen [34]. When the excessive H₂O₂ is given to the solution and there is not enough Fe ions in the wastewater, this causes an increase of the cost, and hydroxyl radical production is not provided; it also increases the COD concentration by interfering in the COD analysis [35,36].

The interaction effect of the Fe²⁺ concentration/H₂O₂ concentration on PFP and current density/H₂O₂ concentration on PEFP is given in Figure 5c and Figure 6c, respectively. Statistical results showed that pH, Fe²⁺ concentration, and H₂O₂ concentration parameters are effective in PFP, while pH and current density are effective in PEFP for COD degradation. Ferrous ion is one of the main parameters that influences the photo-Fenton processes because, while the concentration of Fe²⁺ increases, more hydroxyl radicals produce in the solution [20]. With the increase of Fe²⁺ concentration and H₂O₂ concentration, the COD removal efficiency of the PFP increased in the study. An increase of the H₂O₂ concentration occurred for the hydroxyl radical, which increased the removal efficiency of pollutants [37].

In PEFP, increasing the current density increased the COD removal efficiency (Figure 6b). Applying a higher current density increased the efficiency of Fenton chain reactions in a result of the higher electro-regeneration of ferrous ion from ferric ion (Equation (13)). However, after a while, the increase of the current density decreased the COD removal efficiency. When the current density was more, the COD removal efficiency was lower because a high current density causes competitive electrode reactions such as the discharge of oxygen at the anode by reaction (Equation (14)) and the evolution of hydrogen at the cathode by reaction (Equation (15)) [38].

$${\mathrm{Fe}}^{3+}+{\mathrm{e}}^{-}\to {\mathrm{Fe}}^{2+}$$

(13)

$$2{\mathrm{H}}_2\mathrm{O}\to 4{\mathrm{H}}^{+}+{\mathrm{O}}_2+4{\mathrm{e}}^{-}$$

(14)

$$2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-}\to {\mathrm{H}}_2$$

(15)

At the same time, when large amounts of H₂O₂ are added to the wastewater, the removal efficiency of Fenton processes decreases due to the hydroxyl radical scavenging effect of H₂O₂ (Equations (16) and (17)) and the recombination of hydroxyl radicals (Equation (18)) [39]. In the study, in PEFP, increasing the H₂O₂ concentration had a negative effect on the COD removal efficiency.

$${\mathrm{H}}_2{\mathrm{O}}_2+{}_{}{}^{.}\mathrm{O}\mathrm{H}\to {\mathrm{H}}_2\mathrm{O}+{\mathrm{HO}}_2^{.}$$

(16)

$${\mathrm{HO}}_2^{.}+{}_{}{}^{.}\mathrm{O}\mathrm{H}\to {\mathrm{H}}_2\mathrm{O}+{\mathrm{O}}_2$$

(17)

$${}_{}{}^{.}\mathrm{O}\mathrm{H}+{}_{}{}^{.}\mathrm{O}\mathrm{H}\to {\mathrm{H}}_2{\mathrm{O}}_2$$

(18)

3.3. Statistical Analysis of Total Operating Cost of PFP and PEFP

The operating cost of PFP consisted of the energy cost of the UV lamp and peristaltic pump, the chemical cost of FeSO₄, H₂O₂, NaOH, and H₂SO₄, and the sludge disposal cost.

The operating cost of the PEFP consisted of the energy cost of the UV lamp, peristaltic pump, and DC power supply, the chemical cost of H₂O₂, H₂SO₄, and NaOH, the cost of sludge disposal, and the cost of the Fe electrode.

All the parameters were calculated with the data obtained during the experiment. The calculated values were entered into the software, and as a result of ANOVA, it was seen that the total operating cost was compatible with the quadratic model.

Statistical parameters of the quadratic model for the total operating cost response are given in

Table 7. ANOVA results of PFP and PEFP for the operating cost (USD/m³) response (quadratic model).

	PFP					PEFP
Source	SS	DF	MS	F Value	p Value	SS	DF	MS	F Value	p Value
Model	20.41	9	2.27	46.18	0.0003	1090.12	9	121.12	47.94	0.0003
X1-pH	2.946 × 10−3	1	2.946 × 10−3	0.060	0.8163	14.32	1	14.32	5.67	0.0631
X2-Fe2+/i	19.43	1	19.43	395.50	<0.0001	969.03	1	969.03	383.58	<0.0001
X3-H2O2	0.048	1	0.048	0.99	0.3662	0.50	1	0.50	0.20	0.6764
X1X2	0.030	1	0.030	0.62	0.4670	0.50	1	0.50	0.20	0.6749
X1X3	0.061	1	0.061	1.23	0.3170	0.043	1	0.043	0.017	0.9018
X2X3	5.565 × 10−4	1	5.565 × 10−4	0.011	0.9194	5.05	1	5.05	2.00	0.2166
X12	0.78	1	0.78	15.92	0.0104	63.38	1	63.38	25.09	0.0041
X22	3.5 × 10−5	1	3.5 × 10−5	7.127 × 10−4	0.9797	6.30	1	6.30	2.49	0.1753
X32	0.092	1	0.092	1.88	0.2286	31.41	1	31.41	12.43	0.0168
Residual	0.25	5	0.049			12.63	5	2.53
Lack of Fit	0.25	3	0.082			9.08	3	3.03	1.70	0.3909
Pure Error	0.000	2	0.000			3.56	2	1.78
Cor Total	20.66	14				1102.75	14
			PFP					PEFP
	R2	0.98	Std. Dev.	0.22		R2	0.98	Std. Dev.	1.59
	Adj R2	0.97	Mean	16.17		Adj R2	0.97	Mean	25.03
	Pred R2	0.80	C.V. (%)	1.37		Pred R2	0.87	C.V. (%)	6.35
	A.P.	19.84	PRESS	3.93		A.P.	19.34	PRESS	153.22

. According to the quadratic model, in the PFP process, R², adjusted R², and predicted R² were determined to be 0.99, 0.97, and 0.80, respectively. In the PEFP process, R², adjusted R², and predicted R² were determined to be 0.98, 0.97, and 0.87, respectively. The results of the statistical analysis showed that the R² values for PFP and PEFP were close to 1, indicating that the experimental results and statistical values were compatible.

In the model, the adjusted R² and predicted R² difference was suggested to be <0.2, and CV was <10%. As is seen from Table 7, the adjusted R² and predicted R² difference was smaller than 0.2, and C.V. was <10% (1.37% for PFP and 6.35% for PEFP).

The F-value of the model for PFP and PEFP was 46 and 48, respectively, with a very low probability value (0.0003 and 0.0003), indicating that the model was statistically well-fitted. It was determined that the developed model is important in estimating the operating cost of the processes.

The “lack fit p-value” of the model for PFP and PEFP was identified to be 0.082 and 3.03, respectively, which implied that there was no significant error in the data.

In the study, in PFP and PEFP, A.P. was determined to be 19.84 and 19.34, respectively. Additionally, A.P. was greater than 4 in all processes.

Values of “Prob > F” less than 0.05 show that model terms are significant. Values greater than 0.10 indicate that the model terms are not significant. In this case, ${\mathrm{X}}_2$ and ${\mathrm{X}}_1^2$ were significant model terms for PFP, and ${\mathrm{X}}_2,{ \mathrm{X}}_1^2,{ \mathrm{X}}_3^2$ were significant model terms for PEFP.

The significance of the main factors for the operating cost was Fe²⁺ concentration > H₂O₂ concentration > pH in the PFP process, and current density > pH > H₂O₂ concentration in the PEFP process.

The predicted values of the responses were obtained from the quadratic model. The response equations for the total operating cost by the PFP and PEFP are given in Equations (13) and (14). Here, Y₃ is the predicted total operating cost for PFP (0 < Y₃ ≤ 100%), X₁, X₂, and X₃ are the pH (3 ≤ X₁ ≤ 6), Fe²⁺ concentration (2.3 g/L ≤ X₂ ≤ 3.83 g/L), and hydrogen peroxide concentration (23 g/L ≤ X₃ ≤ 54 g/L), respectively.

In PFP, it is seen in Equation (19) that the coefficient of the pH is negative (b₁ = −0.02). Increasing the pH decreased the total operating cost. The Fe²⁺ concentration (b₂ = +1.56) and H₂O₂ concentration (b₃ = +0.077) had positive signs. Increasing the Fe²⁺ concentration (b₂ = +1.56) and H₂O₂ concentration (b₃ = +0.077) increased the total operating cost.

$${\mathrm{Y}}_3=\mathrm{Total} \mathrm{operating} \mathrm{cost} (\%)-\mathrm{PFP}=15.84-0.02{\mathrm{X}}_1+1.56{\mathrm{X}}_2+0.077{\mathrm{X}}_3-0.087{ \mathrm{X}}_1{\mathrm{X}}_2-0.123{ \mathrm{X}}_1{\mathrm{X}}_3+0.012{\mathrm{X}}_2{\mathrm{X}}_3+0.46{\mathrm{X}}_1^2-0.003{\mathrm{X}}_2^2+0.16{\mathrm{X}}_3^2$$

(19)

Here, Y₄ is the predicted total operating cost for PEFP (0 < Y₂ ≤ 100%); X₁, X₂, and X₃ are the pH (3 ≤ X₁ ≤ 6), current density (24 A/m² ≤ X₂ ≤ 72 A/m²), and hydrogen peroxide (23 g/L ≤ X₃ ≤ 54 g/L), respectively.

In PEFP, it is seen in Equation (20) that the coefficient of the pH and Fe²⁺ concentration was positive (b₁ = +1.34, b₂ = +11.01), while the H₂O₂ concentration (b₃ = −0.25) was negative. Increasing the pH and Fe²⁺ concentration increased the total operating cost. Increasing hydrogen peroxide decreased the total operating cost. The coefficient efficiency of the Fe²⁺ concentration in the equation was quite high (b₂ = +11.01).

$${\mathrm{Y}}_4=\mathrm{Total} \mathrm{operating} \mathrm{cost} (\%)-\mathrm{PEFP}=28.10+1.34{\mathrm{X}}_1+11.01{\mathrm{X}}_2-0.25{\mathrm{X}}_3+0.35{ \mathrm{X}}_1{\mathrm{X}}_2-0.1{\mathrm{X}}_1{\mathrm{X}}_3+1.12{\mathrm{X}}_2{\mathrm{X}}_3-4.14{\mathrm{X}}_1^2+1.31{\mathrm{X}}_2^2-2.92{\mathrm{X}}_3^2$$

(20)

In the study, the normality plot of residuals of the Fenton processes for the response of total operating cost is given in Figure 7a,d.

As is seen from Figure 7b,e, the actual values obtained from the experiments were compatible with the predicted values of the model response for both PFP and PEFP processes for the total operating cost.

The main effects of the X₁ (A), X₂ (B), and X₃ (C) parameters for PFP and for PEFP are given in Figure 7c,f. In the PFP and PEFP processes, X₁ (A) and X₃ (C) had limited effects on the obtained quadratic model.

3.4. Optimization of Other Economical Parameters

Optimized techno-economical parameters, equation types, and equations obtained from ANOVA are given in

Table 8. Optimized techno-economical parameters of the processes.

PFP					PEFP
Parameter	Unit	Value	Eq.	ANOVA Results/Equations	Parameter	Unit	Value	Eq.	ANOVA Results/Equations
pH	-	3.00	-	-	pH	-	3.00	-	-
CFe2+	g/L	2.30	-	-	C.D.	A/m2	27.06	-	-
CH2O2	g/L	27.00	-	-	CH2O2	g/L	28.16	-	-
ECOD	%	84	Q	It is given in Equation (11)	ECOD	%	90	Q	It is given in Equation (12)
Vsludge	kg/m3	5.97	Q	R2 = 0.99, R2Ad j= 0.96, R2Pred = 0.80 Ysludge = 6.2 − 0.25 × X1 + 2.82 × X2 − 0.035 × X3 − 0.26 × X1X2 − 0.23 × X1X3 + 0.076 × X2X3+ 2.26 × X12 0.12 × X22 + 0.56 × X32	Vsludge	kg/m3	3.21	Q	R2 = 0.93, R2Adj = 0.81 Ysludge = 7.2 − 0.06 × X1 + 0.74 × X2 + 0.24 × X3 + 0.22 × X1X2 +0.06 × X1X3 + 0.053 × X2X3 − 2.3 × X12 − 1.08 × X22 − 0.84 × X32
CFeSO4	kg/m3	3.61	L	R2 = 1 R2Adj = 1, R2Pred = 1 YFeso4 = 4.82 − 7.85046e − 016 × X1 + 1.204 × X2 + 0 × X3	CFe electrode	kg/m3	4.12	Q	R2 = 0.98 R2Adj = 0.94, R2Pred = 0.75 YFe = 16.39 + 1.3 × X1 + 8.20 × X2 − 0.27 × X3 − 0.057 × X1X2 − 0.36 × X1X3 + 0.89 × X2X3 − 3.9 × X12 + 1.4 × X22 − 3.15 × X32
-					ECons.	kWh/m3 kWh/kgCOD	26.90 7.29	L	R2 = 0.96, R2Adj = 0.94, R2Pred = 0.91 Yenergy = 39.73 + 0.27 × X1+ 14.85 × X2 − 0.24 × X3
-					Celectrode	USD/m3 USD/kgCOD	4.08 1.106	Q	R2 = 0.98 R2Adj = 0.94, R2Pred = 0.75 YFe = 16.39 + 1.3 × X1 + 8.20 × X2 − 0.27 × X3 − 0.057 × X1X2 − 0.36 × X1X3 + 0.89 × X2X3 − 3.9 × X12 + 1.4 × X22 − 3.15 × X32
CChemical	USD/m3 USD/kgCOD	3.35 0.97	L	R2 = 0.97, R2Ad j= 0.97, R2Pred = 0.95 Ychemical = 4.3 + 0.045 × X1+ 0.85 × X2 + 0.087 × X3	CChemical	USD/m3 USD/kgCOD	0.105 0.028	L	R2 = 0.99 R2Adj = 0.99, R2Pred = 0.99 Ychemical = 0.13 – 0.005 × X1 – 0.0014 × X2 + 0.048 × X3
CEnergy	USD/m3 USD/kgCOD	9.52 2.76	L	-	CEnergy	USD/m3 USD/kgCOD	10.91 2.96	L	R2 = 0.94 R2Adj = 0.92, R2Pred = 0.87 Ychemical = 13.23 + 0.061 × X1+ 2.68 × X2 − 0.053 × X3
Csludge	USD/m3 USD/kgCOD	1.51 0.43	Q	R2 = 0.99 R2Adj = 0.96, R2Pred = 0.80 Yslduge = 1.57 − 0.06 × X1 + 0.71 × X2 − 0.009 × X3 − 0.07 × X1X2 − 0.058 × X1X3 + 0.019 × X2X3+ 0.57 × X12 + 0.030 × X22 + 0.14 × X32	Csludge	USD/m3 USD/kgCOD	0.61 0.16	Q	R2 = 0.93, Adj R2 = 0.81 Ysludge = 7.2 − 0.06 × X1 + 0.74 × X2 + 0.24 × X3 + 0.22 × X1X2 + 0.06 × X1X3 + 0.053 × X2X3 − 2.3 × X12 − 1.08 × X22 − 0.84 × X32
Total O.C.	USD/m3 USD/kgCOD	14.62 4.25	Q	It is given in Equation (19)	Total O.C.	USD/m3 USD/kgCOD	13.79 3.73	Q	It is given in Equation (20)

Q: Quadratic, L: Linear.

.

In PFP, at the optimum conditions (pH 3.00, C_Fe2+ 2.3 g/L, C_H2O2 27 g/L), 84% of COD was removed. In the study, 200 mg/L Fe²⁺ concentration, 300 mg/L H₂O₂ concentration and pH 3 were effective for a 93.2% removal of COD from textile wastewaters [8].

The sludge volume (S.V.) of PFP was determined to be 5.97 kg/m³, and the FeSO₄ consumption of the process was 3.61 kg/m³. Chemical, energy, sludge disposal, and total operating costs were obtained as USD 3.35/m³, USD 9.52/m³, USD 1.51/m³, and USD 14.62/m³ (USD 4.25/kgCOD), respectively.

At the optimum conditions of PEFP, (pH 3.00, C.D. 27.06 A/m², C_H2O2 g/L 28.16), 90% of COD was removed. Sludge volume was determined to be 3.21 kg/m³, and electrode and energy consumption of the process were 4.12 kg/m³ (1.11 kgFe/kgCOD) and 26.90 kWh/m³ (7.31 kWh/kg COD), respectively. In various electro-Fenton process studies in the literature, it was observed that the energy consumed per kg of COD varied in the range of 1.3–350 kWh without adding UV [29,40,41,42]. When UV was added to the process, it offered lower energy consumption per kg of COD when compared to the electrochemical Fenton processes without UV. The addition of UV increased the COD removal efficiency.

Electrode, chemical, energy, sludge disposal, and total operating costs of PEFP were obtained as USD 4.08/m³, USD 0.105/m³, USD 10.91/m³, USD 0.61/m³, and USD 13.79/m³ (USD 3.73/kgCOD), respectively.

The two main parameters that affected the total operating cost of the system are energy consumption and chemical (especially Fenton reagents) costs. The most important parameter among these two parameters is energy cost [43]. In the present study, it was determined that the energy cost of PEFP was the most important parameter that affected the total operating cost of the electro-Fenton processes, as mentioned by other researchers [44,45]. More than half of the total operating cost was due to energy costs [46]. The total operating cost of PEFP was USD 3.73 per removed COD. In a study with an initial COD concentration of 1827 mg/L and at the optimum conditions (pH = 3, current 1 A, [Fe²⁺] = 0.2 Mm, R.T. = 8 h), 96% of COD was removed from landfill leachate wastewater with an operating cost of USD 8.61–26.72/kgCOD [47]. In the literature, other economic parameters such as sludge disposal, acid, and base chemical cost were not considered. Despite this, the total operating cost of Fenton processes in this study per kg of removed COD is more feasible than other studies in the literature [40,48].

The energy cost (as a unit of USD/m³ of cost) of the PFP was 15% lower than the PEFP process because of the extra energy requirement of the DC power supply in the PEFP process. The sludge volume of PFP was 85% higher, and the sludge disposal cost of PFP was 2.5 times higher than PEFP. The total operating cost of PFP was 6% higher than PEFP.

It was determined that the chemical cost of the PEFP per kg of removed COD was 34.6 times lower, the sludge disposal cost was 2.7 times lower, and the energy cost was 7% higher than the PFP process. The total operating cost of the PEFP, per kg of removed COD, was 14% lower than PFP. A comparison of the present study and other Fenton studies in the literature is given in

Table 9. A comparison of the optimized techno-economical parameters for COD removal efficiency from different wastewaters with various Fenton/electrochemical processes.

					Total Operating Cost
Method	Type of Wastewater	Removal Efficiency (%)	Optimum Conditions	Sludge Volume Energy/Chemical Consumption	Before a Photovoltaic Solar Panel Integration	After a Photovoltaic Solar Panel Integration	Literature
Photo/Fenton	Denim jean production wastewater	COD: 84% (CODi: 4100 mg/L) TOC: 61%	pH 3, CFe2+: 2.3 g/L, H2O2: 27 gr/L, R.T.: 30 min	S.V.: 5.97 kg/m3 C FeSO4: 3.61 kg/m3	USD 4.25/kgCOD USD 14.62/m3	USD 1.61/kgCOD USD 5.7/m3	Present Study
Photo/Electrochemical Fenton		COD: 90% (CODi: 4100 mg/L) TOC: 73%	pH 3, CFe2+: 27 A/m2, H2O2: 28 g/L, R.T.: 30 min	S.V.: 3.21 kg/m3 E.C.:7.29 kWh/kgCOD (26.90 kWh/m3)	USD 3.73/kgCOD USD 13.79/m3	USD 1.34/kg COD USD 4.96/m3
Solar-Photo-electro-Fenton	Textile wastewater	COD: 83% (CODi: 545 mg/L)	pH: 4, C.D.: 40 mA/cm2, C[FeSO4]: 0.3 mM	-	USD 3.45/kgCOD USD 1.56/m3		[23]
Photo/Fenton	Dairy industry wastewater	COD: 60% (CODi: 2136 mg/L)	pH 3.5, C[FeSO4]: 198 mg/L, H2O2 14,000 mg/L, R.T.: 180 min	-	USD 40.24/kg COD		[29]
Electrochemical Fenton	Nanofiltration concentrate wastewater	COD: 71% (CODi: 3100 mg/L)	pH: 3, C.D.: 15 mA/cm2, C[FeSO4]: 560 mg/L, R.T.: 360 min	207 kWh/kgCOD	USD 15.93/kgCOD		[40]
Electro-Fenton	Landfill leachate wastewater	COD: 96% (CODi: 1827 mg/L)	pH: 3, C.D.: 1A, C[FeSO4]: 0.2 mM, R.T.: 480 min	110–350 kWh/kgCOD	USD 8.61–26.72/kgCOD		[41]
Electro-Fenton	Textile wastewater	COD: 96% (CODi: 544 mg/L)	pH: 3, C.D.: 0.32 A, C[FeSO4]: 0.53 mM, R.T.: 90 min	1.31 kWh/kgCOD	USD 5.76/kgCOD		[42]
Electro-Fenton	Landfill leachate wastewater	COD: 92.82% (CODi: 825 mg/L)	pH: 4, U: 5.5 V, H2O2/Fe2+: 2.5, R.T.: 50 min, (E:L = 1:2)	3.32 kWh/kgCOD	USD 1.719/kgCOD USD 1.41/m3		[45]
		COD: 93.35% (CODi: 792 mg/L)	pH: 4, U: 5.5 V, H2O2/Fe2+: 2.5, R.T.: 50 min, (E:L = 1:1)	3.44 kWh/kgCOD	USD 1.722/kgCOD USD 1.36/m3
		COD: 91.90% (CODi: 444 mg/L)	pH: 4, U: 5.5 V, H2O2/Fe2+: 2.5, R.T.: 50 min, (E:L = 2:1)	6.24 kWh/kgCOD	USD 1.92/kgCOD USD 0.85/m3

.

3.5. Effect of Solar Panel Integration on Processes

It was reported that electrochemical Fenton processes need to feed the solution with a continuous oxygen source in order to reduce both energy consumption and operating costs by producing H₂O₂ in situ [49]. Another method that minimizes energy consumption and operating costs in electrochemical Fenton processes is the use of solar energy, which is a renewable energy source. For this reason, a photovoltaic solar panel was integrated directly into the Fenton processes in the study.

The voltage/current from the photovoltaic solar panel was obtained as 35.87 V/0.916 A (Figure 8). Energy consumption was calculated as 32.87 Wh. In the study, the reaction time was obtained as 30 min. Considering that the reactor operated for 30 min, the energy obtained from the photovoltaic solar panel was calculated as 32.87 Wh/0.5 = 16.44 Wh. The energy obtained for liter volume was determined to be 21.91 Wh/lt (16.44 Wh/0.75 mL); 21.91 Wh/lt also equals 21.91 kWh/m³. Thus, 93% of the energy need of PFP and 81% of PEFP can be supplied from the solar panel.

Generally, 66% and 79% of the total operating costs of the PFP and PEFP processes, respectively, consist of energy costs. With the addition of the photovoltaic solar panel, it caused a 61% and 64% reduction in total operating costs for PFP and PEFP.

The total operating cost was calculated as USD 14.62/m³ (USD 4.25/kgCOD_removed) without the solar panel and USD 5.7/m³ (USD 1.61/kgCOD_removed) with the solar panel in PFP. In PEFP, it was calculated as USD 13.79/m³ (USD 3.73/kgCOD_removed) without the solar panel and USD 4.96/m³ (USD 1.34/kgCOD_removed) with the solar panel. With the integration of solar energy into electrochemical processes, energy consumption that affects operating cost can be minimized, thus making significant contributions to the environmental sustainability of these processes.

3.6. Biodegradability of Wastewater

The biodegradability of denim processing wastewater after Fenton processes was determined by the following equations (Equations (21) and (22)) [50]:

Carbon oxidation state (COS)

$$\mathrm{COS}=4-1.5\ast \left(\frac{\mathrm{COD}}{{\mathrm{TOC}}_0}\right)$$

(21)

Average oxidation state (AOS)

$$\mathrm{AOS}=4-1.5\ast \left(\frac{\mathrm{COD}}{\mathrm{TOC}}\right)$$

(22)

where TOC is total organic carbon (mg/L) after treatment, and TOC₀ is initial TOC (mg/L) of wastewater. AOS and COS range between +4 for CO₂ and −4 for methane, indicating the most oxidized and most reduced state of carbon [51]. At the optimum conditions, in PFP (pH 3, C_Fe2+: 2.3 g/L, H₂O₂: 27 g/L, R.T.: 30 min), 61% of TOC was removed, and in PEFP (pH 3, C_Fe2+: 27 A/m², H₂O₂: 28 g/L, R.T.: 30 min), 73% of TOC was removed. In PFP, COS and AOS were determined to be 3.64 and 3.09, respectively. In the PFP process, COS and AOS were determined to be 3.77 and 3.17, respectively. High COS values identified that the organic compounds after treatment were comprised mainly of organic acids, and the increase of AOS shows the enhancement of biodegradability [50].

4. Conclusions

In this study, dual criterial optimization, in which the results maximized the COD removal efficiency of denim production wastewater and minimize the operating cost of the PFP and PEFP, was determined by RSM. It was shown that the R², R²_adj, and R²_pre values of the COD removal efficiency response and total operating cost response were appropriate for the quadratic model. In the ANOVA analysis of the model, as suggested by the BBD for COD removal and the total operating cost by Fenton processes, statistical parameters of F-value, p-value, CV, PRESS, and AP were determined in desirable ranges. Additionally, it was found that BBD could be a reliable statistical model to explain the process/operating parameters and to determine the optimal conditions in terms of COD removal and total operating cost.

As a result of ANOVA, the significance of the main factors for the COD removal efficiency by PFP and PEFP was pH > Fe²⁺ concentration/current density > H₂O₂ concentration. The significance of the main factors for the total operating cost was Fe²⁺ concentration > H₂O₂ concentration > pH in PFP, and it was current density > pH > H₂O₂ concentration in PEFP.

At the optimized conditions based on RSM, in PFP (pH 3, C_Fe2+: 2.3 g/L, H₂O₂: 27 g/L, R.T.: 30 min) 85% of COD and 61% of TOC were removed; in PEFP (pH 3, C_Fe2+: 27 A/m², H₂O₂: 28 g/L, R.T.: 30 min), 90% of COD and 73% of TOC were removed. It was determined that both processes were effective in terms of biodegradability.

Total operating cost was obtained as USD 14.62/m³ (USD 4.25/kgCOD_removed) and USD 13.79/m³ (USD 3.73/kgCOD_removed), respectively, in PFP and PEFP. The COD removal efficiency of PFP was 7% lower than PEFP, while the total operating cost of PFP was 6% higher than PEFP (for the cost of USD/m³).

The sludge volume of PFP was higher in a ratio of 85%, and the sludge disposal cost was approximately 2.5 times higher than PEFP. It was determined that the energy requirement of the PFP process was approximately 15% lower than PEFP because of the extra energy requirement of the DC power supply in PEFP. According to the RSM results, in conditions where the COD removal was maximum and the operating cost was minimum, when the total operating cost was considered, it was determined that the operating cost of PEFP per m³ of wastewater was 6% cheaper than the PFP, while the total operating cost of PEFP per kg of removed COD was 14% cheaper than PFP.

The addition of the photovoltaic solar panel to the processes resulted in a 61% and 64% reduction in total operating cost for PFP and PEFP, respectively. This study demonstrated the importance of integrating solar energy into UV-assisted Fenton and electrochemical Fenton processes, reducing total operating costs and preferring these processes.

According to all techno-economical evaluations, PEFP is a technique that produces less sludge, consumes less chemicals, offers lower operating costs, and offers more effective COD and TOC removal compared to the PFP process.