Modeling and Optimization of Argon-Activated Electrohydraulic Plasma Discharge Process for p-Nitrophenol Remediation

Abstract

This study presents a statistical modelling and optimization of an argon-activated electrohydraulic plasma discharge (EHPD) process for the degradation and mineralization of p-nitrophenol (p-NP) in water. The EHPD reactor design incorporated dual dielectric plates to initiate plasma discharge through a central orifice. A fractional factorial design (FFD) was first employed to screen four operating variables, including argon flow rate, pH, applied power, and persulfate dosage, on the p-NP degradation efficiency and energy yield, revealing argon flow rate and applied power as two identified, significant process factors. These were then further optimized using a central composite design (CCD) and response surface methodology (RSM), with the optimal operating condition found to be 2.73 L/min and 128.6 W for argon flow rate and applied power, respectively. Under the optimal operating conditions, 10 min treatment of 50 mg/L p-NP achieved a degradation efficiency of 94.2% and 75.8% total organic carbon (TOC) removal, along with a first-order reaction rate constant of 0.296 min⁻¹ and an energy efficiency of 0.22 g/kWh. The reaction mechanism for p-NP degradation by EHPD was proposed and confirmed with optical emission spectroscopy and radical scavengers. The optimized EHPD process proved both effective and energy-efficient in treating p-nitrophenol, highlighting its potential as a scalable and sustainable plasma-based technology for eliminating persistent organic pollutants and promoting greener water treatment practices.

1. Introduction

p-Nitrophenol (p-NP) is a widely utilized chemical in the production of pesticides, explosives, dyes, and herbicides [1,2,3,4,5]. Due to its complex lipophilic structure, p-NP is prone to bioaccumulation. When released into wastewater treatment plants, conventional biological treatment processes are often ineffective in degrading this compound efficiently. Moreover, toxicological studies on p-NP have demonstrated its potential mutagenicity and toxicity; as a result, it is classified as a top priority pollutant by the USEPA [6,7,8]. Given the current limitations in treatment methods, there is a clear need for the development of new and advanced technologies to enhance the removal of p-NP from wastewater.

Advanced oxidation processes (AOPs), such as photocatalysis, ozonation, Fenton and photo-Fenton, have been widely studied for the removal of persistent organic pollutants. While these techniques can achieve significant degradation, they often require continuous chemical dosing, pH adjustment, or external energy inputs such as UV light, which limit their large-scale application and sustainability. In contrast, electrical plasma discharge is one of the most efficient technologies because of its inimitable characteristics of producing reactive oxygen species (ROS) and reactive nitrogen species (RNS) without the need of any chemicals or external light source for organic pollutant degradation [9]. When the electrical plasma discharge happens in liquid–gas interaction, both physical and chemical effects, such as ultraviolet (UV) radiation, shock waves and cavitation, and the formation of reactive radicals, ions, and charged particles, all come into play in the degradation of organic pollutants [10].

Despite these advantages, the application of plasma-based processes for p-NP treatment remains limited and fragmented. K. Shang et al. [11] reported the degradation of p-NP by dielectric barrier discharge (DBD) plasma in the presence of persulfate (PS) and Fe²⁺. Results showed that PS could be synergistically activated by DBD plasma and a small quantity of Fe²⁺. The degradation efficiency of p-NP at an initial concentration of 5 mg/L reached 34.8% by 50 min plasma treatment alone at 17 kV, while with the addition of 36 μM of Fe²⁺ and 2.5 mM of PS in p-NP solution, the degradation efficiency was increased to 81.1%. The energy yield was 99.4 mg/kWh for plasma alone and 229.5 mg/kWh with plasma/PS/Fe²⁺. The radical scavenging experiment showed that ^•OH and SO₄⁻ radicals contributed to degradation of p-NP. Another study reported by C. Zhao et al. [5] showed a microwave atmospheric pressure plasma jet to degrade p-NP in wastewater. Completed degradation of p-NP at an initial concentration of 100 mg/L was achieved in 12 min of treatment. TOC removal efficiency was 57.6%, and the energy yield was 0.12 g/kWh. These cases show significant differences in degradation efficiency, mineralization, and energy yield, as well as variability and limitations of plasma-based processes, emphasizing the need for systematic optimization to enhance performance and sustainability.

In this study, argon (Ar)-activated electrohydraulic plasma discharge (EHPD) was used for degradation of p-NP in water. The influence of operation parameters, such as pH, gas flow rate, applied power, and persulfate concentration, on p-NP degradation were investigated. Statistical experimental designs, such as fractional factorial design and response surface methodology (RSM), were employed to determine the optimal running conditions. First, a 2⁴⁻¹ factorial design was used to evaluate the relative significance of various factors that influenced the plasma degradation of p-NP [12,13]. Then, a central composite design (CCD) with response surface methodology (RSM) was employed to further evaluate the significant factors for modeling and optimization in terms of p-NP removal efficiency and energy yield. With CCD design, various variables and their interactions were tested simultaneously with a minimum number of experimental trials needed to fit a quadratic surface and to determine optimal levels of process variables [14,15].

2. Materials and Methods

p-Nitrophenol purchased from TCI Chemicals Pvt. Ltd. (Tokyo, Japan) was used as the target pollutant. Other chemicals, such as sodium sulfate, ascorbic acid, 2-propanol (C₃H₈O), and sodium nitrate (NaNO₃), purchased from Fisher Scientific (Eugene, OR) were used for conductivity adjustment and radical scavenging.

2.1. Experimental Setup and Improved Reactor Design

The setup of the continuous-flow electrohydraulic plasma discharge (EHPD) process was similar to that of our previous work on methylene blue remediation [16]. The body of the reactor was made of polycarbonate material. To produce plasma discharge in the solution continuously passing the reactor, stainless-steel bushings served electrodes that were connected to high voltage and ground to make a full electrical circuit, separated by dielectric plates with a small orifice in the center (0.8 mm) as shown in Figure 1. Improved design of the reactor was achieved via the addition of a second dielectric plate for better plasma discharge and electric safety. In comparison to the most-studied designs, i.e., pin–pin and pin–plate electrohydraulic discharge reactors, this design achieved a unique ability to establish a complete discharge through the conducting channel of the orifice rather than between the electrodes. The orifice in the middle of the dielectric plate allowed electrons to concentrate near the opening, and the discharge current could be kept in the plasma phase as mobile electrons, allowing for better mass transfer and breakdown of substrate molecules in contact with the plasma discharge. This continuous-flow design could also significantly improve the throughput capacity of the reactor, thus potentially reducing reactor size and operating costs at the commercial level [17,18].

In this study, a high-voltage transformer (output: 12 kV, Plasma Technics, Inc., Racine, WI, USA) connected to an alternating current (AC) power supply was used to power the EHPD reactor. A Variac variable voltage regulator (model#: TDGC2-2KM, ISE Inc., Cleveland, OH) was used to vary the voltage and power applied. The applied voltage between the top electrode and the ground electrode was measured using a high-voltage probe (Tektronix P6015A) connected to an oscilloscope (Tektronix TBS1052B, Beaverton, OR, USA). The amount of power applied to the entire system was determined using a wattmeter. A total volume of 100 mL of p-NP solution at an initial concentration 50 mg/L was used for each experimental run. Sodium sulfate was used to adjust the initial conductivity (65 µS/cm) of the solution. The prepared p-Nitrophenol solution was introduced to a three-neck flask connected to a condenser tube and pumped through the EHPD reactor through a peristaltic pump for treatment. The treated solution was then directed back to the flask to accomplish a continuous mixing and circulating operation. A peristaltic pump (Masterflex L/S 7523-60, Vernon Hills, IL, USA) was controlling the flow rate of the prepared p-NP solutions to be continuously passed through the reactor for treatment. Argon was used for the initiation of plasma discharge, and the power was turned on once the solution filled the discharge region. Each treatment lasted 10 min, after which the reactor system was flushed twice with distilled water to clean the system. The initial and final pH and conductivity of p-NP was measured using conductivity and pH probes with a Hach HQ 440D Multi-Meter.

2.2. Experimental Design

Minitab 20.2.0 version (Minitab Inc., State college, PA, USA) and Design Expert 13 (StatEase, Inc., St. Paul, MN, USA) were both used in this study for factor screening and optimization. The experimental design started with screening the main effects of the process variables by performing a fractional factorial design (FFD) with 2^n−k experiments, where n was the number of factors to study at low and high levels (−1, +1), and k was the number of steps to minimize the experimental design [19]. The purpose of the screening experiments was to identify a subset of critical factors of significance, which could be further examined in the second experimental design, i.e., the central composite design (CCD) coupled with response surface methodology (RSM), to determine the optimal values of these factors [14,15].

Four process control variables were selected for FFD (pH, argon flow rate, power applied, and persulfate; levels shown in

Table 1. This table shows 2⁴⁻¹ factorial design factors and their chosen levels.

Factor	Low Level (−1)	High Level (+1)
Gas flow rate (L/min)	0.4	1
pH	5	9
Applied power (W)	200	300
Persulfate conc. (mg/L)	50	250

), and their effects on p-NP degradation efficiency and energy yield were set as dependent variables. Once the significant independent variables were determined, a CCD/RSM was used to determine the optimal conditions of EHPD with respect to the selected, significant independent variables to establish a quadratic response surface for each dependent variable.

2.3. Analytical Methods

Degradation of p-NP by the EHPD treatment was monitored at 1 min intervals up to 10 min using a UV–Vis spectrophotometer (Synergy HT, BioTek Instruments Inc., Winooski, USA) at 317 nm. Each experiment was repeated twice and was carried out at atmospheric pressure and ambient temperature. The degradation efficiency of p-NP was measured as follows:

n (%) = (C₀ − C) × 100/C₀

(1)

In the equation, C₀ = initial concentration, and C = final concentration.

The energy yield for p-NP degradation was calculated using Equation (3). The yield value was expressed as the amount of pollutant converted divided by the energy input required at n % conversion of the pollutant [20]:

Y(g/kWh) = (C₀ × V₀ × n % × 1/100)/Pt

(2)

where C₀ was the initial concentration of the pollutant in mg/L, V₀ was volume of treated solution in L, n % was degradation efficiency at time t, P was average power dissipated in the discharge (kW), and t was the treatment time in hour.

Total organic carbon was measured using a TOC kit and analyzed using a Hach DR3900 laboratory spectrophotometer.

TOC (%) = (TOC₀ − TOC_t) × 100/TOC₀

(3)

where TOC₀ = TOC of initial solution, and TOC_t = TOC of the solution at time t.

2.4. Radical Scavenging Experiments

To determine the role of active species produced during the EHPD process in p-NP degradation, ascorbic acid (3 mmol/L), sodium nitrate (3 mmol/L), and 10 mL diluted 2-propanol (100 mg/L) in distilled water were added to the initial p-NP solution of 50 mg/L to inhibit superoxide radical (^•O₂⁻), aqueous electrons (e⁻_aq), and hydroxyl radical (^•OH), respectively [16]. To compare with the scavenger-free experiment for p-NP degradation, the optimum conditions from the CCD were used for all radical scavenging tests.

3. Results and Discussion

3.1. Operating Factor Screening by Fractional Factorial Design and Analysis

In this study, a 2⁴⁻¹ FFD was used to identify key variables affecting the degradation of p-NP using EHPD. Experiments were conducted at different points within experimental space to systemically assess the impact of four process variables, i.e., pH, argon gas flow rate, applied power, and persulfate concentration. A two-level, half-factorial design with four factors required a total of eight experimental runs, with a duplicate of corner points included for validation as shown in Table S1, with responses being the degradation efficiency and energy yield. The significance of each analyzed input variable was shown in a Pareto chart (Figure 2a,b), and the analysis of variance (ANOVA) results (Table S2) provide clear insights into the key factors influencing the degradation of p-NP using EHPD. The ANOVA results showed that the argon flow rate and applied power were the most significant factors affecting both degradation efficiency and energy yield, as indicated by their extremely low p-values (p < 0.0001) and high F-values. Persulfate concentration also exhibits statistical significance, particularly for degradation efficiency (p = 0.0024), though its impact is less stated than that of argon flow rate and power. In addition, the interaction between argon flow rate and power (AC) is highly significant (p < 0.0001), emphasizing the necessity of optimizing these two parameters together to maximize performance. In contrast, pH has no significant effect on either response (p > 0.05), indicating that within the tested range, pH variations do not substantially influence degradation efficiency or energy yield. A study reported by A. Khourshidi [21] found that ozone recirculation in the DBD reactor increased the degradation efficiency of p-nitrophenol, which was attributed to an increased release of reactive oxygen species in the DBD–O₃ system, whereas the process performance was reported to have minimal effect on the DBD-O₃ system by variations in pH levels, initial pollutant concentrations, or the presence of co-existing substances in comparison with DBD system.

The Pareto charts highlight these findings, illustrating that argon flow rate and power were the most significant standardized effects on both responses. Furthermore, the predicted vs. actual plots illustrated in Figure S1a,b, demonstrated strong model reliability, as the data points align closely with the diagonal reference line, confirming the predictive accuracy of the statistical model. Figure S2a,b illustrated the four main effects on degradation efficiency and energy yield, which clearly showed that for both responses, the larger the vertical line, the greater the difference was when changing from level −1 to +1. It was observed that the effects of argon and power were significantly higher than the other factors. Therefore, argon and power were selected as the important main effects for p-NP degradation using EHPD for optimization.

3.2. Optimization of p-NP Degradation by EHPD

According to the FFD, argon and power were found to be the most significant factors. Therefore, argon and power were chosen as control variables for EHPD to be optimized using CCD with the aid of Design Expert 11 and Minitab version 20.2.0. In CCD, a second-order quadratic model, Equation (4), was developed by varying each independent variable at five levels (−α, −1, 0, +1 and +α) to fit the responses (a total of thirteen experiments were needed). Table S3 showed the combinations of controlling variables that were examined, as well as the results achieved. Different levels of argon flow rate and applied power were used in the experimental runs as determined by the CCD model.

$$Y={\beta }_0+{\beta }_1{x}_1+{\beta }_2{x}_2+{\beta }_{12}{x}_1{x}_2+{\beta }_{11}{x}_1^2+{\beta }_{22}{x}_2^2$$

(4)

where Y: the predicted response, x₁ and x₂: independent variables$,{\beta }_0$: the offset term$,{\beta }_1$ and ${\beta }_2$: linear coefficients, ${\beta }_{11}$ and ${\beta }_{22}$: the squared coefficients, and ${\beta }_{12}$: the interaction coefficient.

3.3. Degradation Efficiency and Energy Yield Model

Response surface models were established based on the responses (degradation efficiency and energy yield) obtained for the designated sets of experiments in Table S3. The goodness of fit according to ANOVA analysis was used to validate the models, and the results are shown in

Table 2. ANOVA for quadratic models of degradation efficiency and energy yield.

Source	Sum of Squares	df	Mean Square	F-Value	p-Value
Response 1: degradation efficiency
Model	233.77	5	46.75	28.75	0.0002	Significant
A-argon	93.92	1	93.92	57.76	0.0001
B-power	122.62	1	122.62	75.41	<0.0001
AB	13.32	1	13.32	8.19	0.0243
A2	2.31	1	2.31	1.42	0.2723
B2	2.11	1	2.11	1.30	0.2919
Residual	11.38	7	1.63
Lack of fit	7.93	3	2.64	3.06	0.1541	Not significant
Pure error	3.45	4	0.8636
Cor total	245.15	12
Response 2: energy yield
Model	0.0159	5	0.0032	792.25	<0.0001	Significant
A-argon	0.0002	1	0.0002	54.29	0.0002
B-power	0.0142	1	0.0142	3536.83	<0.0001
AB	0.0001	1	0.0001	15.91	0.0053
A2	6.545 × 10−6	1	6.545 × 10−6	1.63	0.2428
B2	0.0014	1	0.0014	340.42	<0.0001
Residual	0.0000	7	4.022 × 10−6
Lack of fit	0.0000	3	7.642 × 10−6	5.84	0.0606	Not significant
Pure error	5.232 × 10−6	4	1.308 × 10−6
Cor total	0.016	12

. The degradation efficiency model (Equation (5)) had an overall p-value of 0.0002, and the energy yield model (Equation (6)) had an overall p-value of less than 0.0001, suggesting that both quadratic models were statistically significant. The F values of both models were 28.75 and 792.25 for degradation efficiency and energy yield, respectively, which also implied that the models were significant. Also, argon, power, and argon*power were found to be significant model terms in the degradation efficiency model. For the energy yield model, argon, power, argon*power, and power² were significant.

Moreover, the lack of fit values for both models was insignificant (0.1541 and 0.0606, respectively), providing evidence that both models were highly significant. As shown in Figure S4a,b, the observed degradation efficiency and energy yield values ranged from 83.06 to 96.61% and 0.08 to 0.212 g/kWh, respectively, and were in good agreement with the predicted values.

Degradation efficiency (%)	=	54.85 + 14.01 Argon + 0.1449 Power −0.900 Argon × Argon- 0.000098 Power × Power −0.0304 Argon × Power	(5)
Energy yield (g/kWh)	=	0.33639 + 0.02638 Argon − 0.001578 Power −0.00152 Argon × Argon + 0.000002 Power × Power −0.000067 Argon × Power	(6)

The goodness of fit could also be evaluated using R² value. In both responses, the R² values (0.9536 and 0.9982) were in reasonable agreement with the adjusted R² values (0.9204 and 0.9970). Also, Adeq precision was used to verify the model’s validity by measuring the signal-to-noise (S/N) ratio (Adeq precision should be greater than 4) [22,23]. In this study, the Adeq precision values were 17.0 and 87.5 for degradation efficiency and energy yield responses, indicating an adequate signal. In addition to all these parameters, the normality plots were used to provide further evidence and validation for the regression models. As shown in Figure S3a,b, the normality plot of externally studentized residuals denoted that the majority of color points representing degradation efficiency and energy yield were in a narrow range on a normal probability line, while the minorly significant points deviated from the normal line. Both models confirmed the normal distribution of the plots. The regression models showed strong statistical reliability as indicated by high R² values, adequate Adeq precision, and normal residuals. They accurately predict degradation efficiency and energy yield under controlled lab conditions. However, real wastewater systems are more complex, with multiple variables potentially affecting outcomes and limiting model accuracy at larger scales. Thus, further validation in pilot-scale and real wastewater scenarios is needed to confirm broader applicability.

3.4. Graphical Interpretation and Optimization of Operating Parameters

Figure 3a,b represent the contour and surface plots to view the effects of argon and power and their interactions on degradation efficiency and energy yield. These figures were used to define the optimal value of each variable towards maximizing degradation efficiency and energy yield of the EHPD process. As shown in Figure 3a, the degradation efficiency was enhanced with an increase in either argon flow rate or power. The impact of argon flow rate was more prominent than that of power. It was also noticed that at high power (331.1 W) and mid-range argon flow rate (1.6 L/min), the degradation efficiency reached maximum (96.61%) within 10 min of treatment. Past research showed that argon flow rate was an important parameter that could influence species concentrations when the plasma was initiated [24,25]. A high argon flow rate could also produce larger bubbles in the solution. Thus, the plasma composition, as well as the concentrations of numerous species produced during plasma discharge, was all directly related to argon flow rate. As shown in Figure 3b, to achieve maximal energy yield, a combination of low power and high argon flow rate was required. When a lower power supply was used, energy loss due to heat dissipation could be minimized, thus improving energy yield.

Optimization of influencing factors is required to determine the optimal values of the operating variables to achieve the best p-NP degradation efficiency and energy yield. The response surface optimization was carried out using the Minitab software 20.2.0 version. maximal percentage degradation and energy yield were 92.73% and 0.212 g/kWh, respectively, under the optimal operating conditions, i.e., argon flow rate of 2.73 L/min and power of 128.6 W (Figure 3c). To validate the models for satisfactory prediction of the maximal degradation and energy yield, an additional experiment was conducted under the optimized conditions, which produced the maximal degradation efficiency and energy yield of 94.23% and 0.22 g/kWh, respectively. These results showed that RSM was a reliable technique to establish an accurate model for obtaining maximal degradation and energy yield in this investigation. The optimization process showed the importance of balancing degradation efficiency with energy yield. Increased power input resulted in faster p-NP degradation by generating more reactive species but also led to reduced energy efficiency due to heat dissipation and greater energy losses. In contrast, lower power improved energy yield yet limited degradation efficiency. The determined optimal operating window incorporated both satisfactory degradation performance and energy efficiency, supporting the sustainable use of the EHPD process.

4. Mechanistic Study of p-NP Degradation by EHPD

The emission spectra of the plasma discharge were recorded to better understand the nature of active species that contributed to degradation of p-nitrophenol. Figure 4a presented the optical emission spectra of the argon-activated EHPD for p-NP (50 mg/L) degradation measured under optimal operating conditions. During plasma treatment, argon atomic (Ar I) system lines ranging from 690 to 965 nm were detected in the OES spectra, corresponding to radiative de-excitations from the 4p and 4p′ to 4s energy levels [5,24]. The emission line observed at 309.6 nm was assigned to •OH radical species, specifically related to the A²Σ⁺ (v″ = 0)–X²Π(v″ = 0) transition [26]. Additionally, atomic oxygen species (OI at 777.03 and 844.13 nm) and hydrogen lines of the Balmer series, including Hα (at 655.92 nm) and H_β (at 486.16 nm), were identified [27]. The generation of •OH radicals and hydrogen species during plasma treatment is likely due to high-energy electrons colliding with Ar atoms to produce excited Ar, which subsequently reacts with water to form •OH and H radicals, as described in Equations (7) and (8). Furthermore, •OH and H radicals may also result from the dissociation, ionization, and excitation of water molecules (Equations (9)–(13)). The formation of atomic oxygen is presumed to arise from the dissociation of oxygen molecules, as illustrated in Equation (14) [5,26,27,28].

e⁻ + Ar → Ar^∗ + e⁻

(7)

Ar^∗ + H₂O → Ar + ^•OH + H

(8)

e⁻ + H₂O → ^•OH + ^•H + e⁻

(9)

e⁻ + H₂O → 2e⁻+ H₂O⁺

(10)

H₂O⁺ + H₂O → ^•OH + H₃O⁺

(11)

e⁻ + H₂O → H₂O^∗ + e⁻

(12)

H₂O^* + H₂O → H₂O + ^•H + ^•OH

(13)

e⁻ + O₂ → O^• + O^• + e⁻

(14)

^•O + H₂O → 2^•OH

(15)

To validate their contribution to p-NP degradation, radical scavenger experiments were conducted at an initial p-NP concentration of 50 mg/L under identical operating conditions, i.e., argon flow rate of 2.73 L/min and power of 128.6 W, to detect aqueous electrons, superoxide, and hydroxyl radicals. The results were compared to those obtained in the absence of scavengers to determine the main active species involved. To further establish the influence of radical scavengers on p-NP degradation efficiency, degradation kinetics were investigated as shown in Figure 4b. During the entire treatment, 2-propanol had a significant effect on p-NP degradation, with the degradation rate constant of 0.0848, evincing the production of ^•OH radicals in the argon-initiated EHPD process as the main active species for p-NP degradation. The degradation rate constant was reduced to 0.1129 with the addition of ascorbic acid, clearly suggesting a drop in degradation percentage as compared to scavenger-free trials. The data revealed that superoxide radicals were the second-most active species in degradation. When NaNO₃ was added, the degradation rate constant was 0.2116, showing a minor influence on the p-NP degradation, which meant that hydrated electrons had a minimal impact on p-NP degradation in the argon initiated EHPD process. Moreover, it is important to note that the scavengers were introduced at concentrations designed to quench specific radicals while allowing measurable degradation. This approach aimed to demonstrate the relative contribution of each species without inhibiting the overall plasma process. The combined findings from OES data and scavenger experiments clearly indicate that hydroxyl radicals (•OH) serve as the primary oxidizing agents within the argon-activated EHPD system, while superoxide radicals play a secondary role and hydrated electrons have only a minimal impact.

4.1. Proposed p-NP Degradation Pathway

To identify the p-NP degradation intermediates, 50 mg/L p-NP initial concentration was treated under optimal conditions by EHPD, and then the intermediate products were analyzed using UPLC-MS/MS. The total chromatogram is shown in Figure S5. The intermediate products were hydroxylated compounds based on ortho- and para-positions of the OH⁻, which had high density and were readily attacked by the ^•OH radicals [29]. The findings suggested that p-NP was first oxidized to p-nitrocatechol (m/z = 154) by electrophilic reaction, which was then transformed by ^•OH radicals into p-nitropyrogallol (m/z = 170). The aromatic ring-opening products were further oxidized into aliphatic acids such as formic acid, acetic acid, or oxalic acid [30]. These aliphatic acids were eventually mineralized to NO₂⁻, NO₃⁻, H₂O, and CO₂ via the degradation pathway shown in Figure 5. In addition to ^•OH radicals, active species (such as ^•O and ^•H) and physical effects (such as heat and UV) might also participate in the process.

Many studies reported a similar mechanism for degradation of p-NP through oxidation by ^•OH radicals through electrophilic attack [4,5,30,31,32]. Wang et al. [30], reported that ^•OH radicals might attack the -NO₂ group in p-NP molecules due to the relatively long C—N bond, which was the longest bond in the molecule and thus was prone to breakup to form phenol, which was then oxidized into hydroquinone, benzoquinone, and catechol. Additional reactions between these intermediates and •OH radicals could result in ring cleavage and the formation of an aliphatic compound.

The results from ion chromatography for 50 mg/L p-NP initial concentration treated under optimal running conditions showed that nitrate, nitrite, and sulfate were present in low concentrations, which were produced during decomposition of p-NP molecules. For a 10 min treatment, a relatively high concentration of sulfate was obtained (44.9 mg/L), while much lower concentrations of nitrate and nitrite were seen (14.4 mg/L and 1.4 mg/L). The high sulfate concentration was due to the addition of sodium sulfate to adjust the initial conductivity of the p-NP aqueous solution. To further validate the extent of mineralization, TOC analysis was conducted under the same optimal conditions (50 mg/L p-NP). After 10 min of treatment, a TOC removal rate of 75.68% was observed, indicating that the EHPD process not only transformed p-NP into intermediate products but also accomplished significant mineralization to CO₂ and smaller organic compounds. This high level of mineralization demonstrates the efficiency of the process beyond mere degradation [5,11,32].

Optical emission spectroscopy identified the active species involved in the argon-activated EHPD process, which included argon atoms, ^•OH radicals, atomic oxygen species, and hydrogen lines of the Balmer series. The radical scavenging experiment confirmed that the main active species responsible for p-NP degradation was ^•OH radicals. The intermediates and inorganic ions involved in p-NP degradation by the EHPD process were identified using UPLC-MS/MS and ion chromatography, based on which the possible pathway for p-NP degradation was revealed to be electrophilic reactions by ^•OH radicals and oxidation and mineralization of nitroaromatic rings into inorganic ions (NOx⁻) or compounds such as carbon dioxide.

4.2. Comparative Studies for p-NP Degradation by Nonthermal Plasma Processes

The argon-activated EHPD achieved 94.23% degradation in just under 10 min for an initial p-NP concentration of 50 mg/L at 128.6 W, resulting in an energy yield of 0.22 g/kWh. In comparison, other plasma processes reported in the literature display different balances between efficiency, energy consumption, and chemical requirements (

Table 3. Comparison of the argon-activated EHPD on p-NP degradation with other plasma processes.

Methods	Initial p-NP Concentration (mg/L)	Time (min)	Degradation Efficiency (n%)	Discharge Power	Energy Yield (g/kWh)	Reference
DBD plasma	5	50	34.38	4.2 W	0.1	[11]
Plasma + persulfate	5	50	63.6	4.2 W	0.18	[11]
Plasma + persulfate + Fe2+	5	50	81.1	4.2 W	0.23	[11]
Microwave plasma	100	12	100	380 W	0.07	[5]
EHPD	50	10	94.23	128.6 W	0.22	This study

). DBD plasma alone reached 34.38% degradation after 50 min, while adding persulfate and Fe²⁺ increased removal to 81.1% but required external chemical inputs and had a similar energy yield to EHPD (0.23 g/kWh). Microwave plasma achieved complete degradation in 12 min at higher initial concentrations, but its energy yield was lower (0.07 g/kWh) due to increased energy consumption (380 W). These results indicate that the EHPD process combines rapid degradation and relatively high energy efficiency without requiring chemical additives. With short treatment times, chemical-free operation, and competitive energy performance, the EHPD system provides an alternative approach to existing plasma-based remediation methods.

5. Conclusions

This study presents the optimization and modeling of the argon-activated electrohydraulic plasma discharge (EHPD) process to achieve efficient degradation of p-nitrophenol in aqueous solution. Kinetic and statistical modeling, such as fractional factorial design and response surface methodology, were employed to identify reactor conditions that maximized degradation efficiency per unit of energy and argon, significantly reducing operational costs and greenhouse gas emissions. Through the fractional factorial design (FFD), argon flow rate and applied power were found to be the significant factors of the EHPD process for p-NP degradation, which were then optimized using CCD coupled with response surface methodology (RSM) with respect to two response variables, i.e., p-NP degradation efficiency and energy yield. The optimal values of argon flow rate and applied power were found to be 2.73 L/min and 128.6 W, which achieved the maximal degradation efficiency of 92.73% and energy yield of 0.212 g/kWh. The R² values of the two response models (degradation efficiency and energy yield) were 0.9536 and 0.9982, respectively, which were close to the adjusted R² of 0.9204 and 0.9970. Additional experiments were carried out under optimized conditions to validate the models. For 10 min treatment, the maximal degradation efficiency and energy yield at 50 mg/L initial concentration were 94.23% and 0.22 g/kWh, respectively, indicating that the model accuracy was satisfactory. The EHPD process under optimal running conditions achieved effective and efficient p-NP decomposition and higher or comparable energy yield than competitive plasma techniques and was therefore deemed as a promising technology for treating p-NP wastewater and other pollutants, enhancing sustainability by efficiently mineralizing the toxic pollutant without the addition of chemicals. Additionally, a TOC removal rate of 75.68% was achieved, and ion chromatography detected low concentrations of nitrate and nitrite byproducts, suggesting that the EHPD process resulted in effective mineralization with limited secondary byproduct formation. These findings indicated that EHPD, with its modular and continuous-flow design, holds significant potential as a compact and sustainable technology for scalable, decentralized wastewater treatment. While the models showed reliability in laboratory settings, additional validation with real wastewater and pilot-scale studies is required to determine practical applicability. Future research should also investigate the kinetics and fundamental mechanisms involved in the rapid and effective p-NP degradation achieved by EHPD.