Preparation and Optimization of NiFe₂O₄/GAC Composite Catalyst and Its Application in PEM Electrolytic Ozonation for Sulfamethoxazole Degradation

Abstract

With the increasing detection of antibiotics such as sulfamethoxazole (SMX) in water bodies, developing efficient and eco-friendly treatment technologies is critical. This study employs a hydrothermal impregnation method to prepare a NiFe₂O₄/granular activated carbon (GAC) composite catalyst, optimized for use in a proton exchange membrane (PEM) electrolytic ozonation system to degrade SMX. Single-factor experiments optimized preparation conditions with a Fe:Ni molar ratio of 3:1, a GAC:Fe + Ni mass ratio of 2:1, and calcination at 500 °C for 3 h. The catalyst was characterized using XRD, SEM, TEM, XPS, and FT-IR, confirming a spinel NiFe₂O₄ structure (crystal size ~15.2 nm) uniformly dispersed on GAC, with an Fe:Ni atomic ratio of ~2.1:1. In the PEM system, the optimized catalyst achieved a 99.15% ± 0.3% SMX degradation rate (50 mg/L) within 25 min, compared to 95.06% ± 0.6% in 30 min without a catalyst. The catalyst maintained 98.45% ± 0.5% efficiency after three cycles, demonstrating excellent stability. The synergy between GAC adsorption and NiFe₂O₄ catalysis, driven by Fe³⁺/Fe²⁺ redox cycling, enhances ·OH generation from ozone decomposition, boosting SMX degradation. This work provides a robust catalyst for antibiotic wastewater treatment and a foundation for scaling up catalytic ozonation.

1. Introduction

Rapid industrialization and urbanization have exacerbated water pollution, with emerging contaminants like antibiotics posing significant threats to ecosystems and human health [1]. Sulfamethoxazole (SMX), a widely used sulfonamide antibiotic, is frequently detected in surface water, groundwater, and wastewater treatment plant effluents due to its stable chemical structure and poor biodegradability, leading to bacterial resistance and ecological toxicity [2,3,4]. Conventional wastewater treatment methods, such as biological treatment and physical adsorption, exhibit limited SMX removal efficiency, failing to meet stringent discharge standards [3,5,6]. Thus, developing efficient and sustainable technologies for SMX removal is a pressing need in environmental science.

Advanced oxidation processes (AOPs) generate highly reactive radicals (e.g., hydroxyl radicals, ·OH) to degrade organic pollutants effectively, offering significant advantages in water treatment [7,8]. Ozonation, with its strong oxidation capacity, rapid reaction kinetics, and lack of secondary pollution, is widely applied for recalcitrant pollutant treatment [9,10]. However, standalone ozonation suffers from low selectivity and insufficient ozone utilization, limiting its practical application [10]. Proton exchange membrane (PEM) electrolysis generates ozone in situ electrochemically, avoiding transportation and storage issues associated with traditional ozone generators while providing high ozone yield and energy efficiency [11]. Nonetheless, the utilization efficiency of PEM-generated ozone in water treatment requires improvement, making catalytic ozonation a promising strategy.

Catalytic ozonation enhances pollutant degradation by promoting ozone decomposition into ·OH radicals via catalysts [12]. Transition metal oxides, due to their high catalytic activity, low cost, and stability, are widely studied [12,13,14]. Nickel ferrite (NiFe₂O₄), a spinel-type composite oxide, offers excellent electrochemical properties and magnetic recoverability, making it a suitable ozonation catalyst [15]. However, standalone NiFe₂O₄ tends to aggregate, reducing active site availability and catalytic efficiency [16]. Supporting NiFe₂O₄ on high-surface-area granular activated carbon (GAC) mitigates aggregation, enhances active site exposure, and leverages GAC’s adsorption capacity to synergistically improve pollutant removal [17].

While progress has been made in preparing NiFe₂O₄-based catalysts for applications like photocatalytic dye degradation and electrocatalytic oxidation [17,18], their use in PEM electrolytic ozonation systems, particularly with NiFe₂O₄/GAC, remains underexplored. The systematic optimization of preparation conditions (e.g., Fe:Ni ratio, GAC loading, and calcination parameters) and evaluation of reusability and stability in complex water matrices are lacking [19]. This study prepares NiFe₂O₄/GAC via hydrothermal impregnation; optimizes its synthesis; and characterizes its crystal structure, morphology, and chemical composition using XRD, SEM, TEM, XPS, and FT-IR. The catalyst’s performance in SMX degradation within a PEM ozonation system, its reusability, and its adsorption/catalysis synergy are evaluated. This work aims to develop an efficient, stable catalyst for antibiotic wastewater treatment and provide a theoretical basis for scaling up catalytic ozonation.

2. Materials and Methods

2.1. Reagents and Instruments

The instruments and reagents used are listed in

Table 1. List of experimental instruments.

Instrument	Model	Manufacturer
Analytical Balance	FA2004	Sunny Hengping Scientific Instrument Co., Ltd. (Shanghai, China)
pH Meter	PHS-3C	Leici Instrument Co., Ltd. (Shanghai, China)
Peristaltic Pump	NKCP-S10B Series	Kamoer Fluid Technology Co., Ltd. (Shanghai, China)
High-Performance Liquid Chromatograph	1260	Agilent Technologies Co., Ltd. (Santa Clara, CA, USA)
DC Stabilized Power Supply	SW-1800-24V	Mingwei Power Supply Co., Ltd. (Qidong, China)
Vacuum Filtration Machine	SMT/SHZ-D(III.)	Simaite Industrial Co., Ltd. (Shanghai, China)
Muffle Furnace	HR-F-1200	Huarong Kiln Co., Ltd. (Luoyang, China)
Scanning Electron Microscope	Regulus 8100	Hitachi Scientific Instruments Co., Ltd. (Tokyo, Japan)
Transmission Electron Microscope	JEM-2100F Series	Japan Electron Optics Laboratory Ltd. (Tokyo, Japan)
X-Ray Diffractometer	D2 Phaser	Bruker Corporation, (Bremen, Germany)
Fourier Transform Infrared Spectrometer	Nicolet iS 20	Thermo Fisher Scientific Co., Ltd. (MA, USA)
X-Ray Photoelectron Spectrometer	K-Alpha	Thermo Fisher Scientific Co., Ltd. (MA, USA)

and

Table 2. List of experimental reagents.

Reagent	Purity	Manufacturer
Sulfamethoxazole	Biotechnology Grade	Macklin Biochemical Co., Ltd. (Shanghai, China)
Anhydrous Sodium Sulfate	Analytical Reagent (AR)	Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China)
Sulfuric Acid	AR
Sodium Hydroxide	AR
Ferric Chloride	AR
Ferrous Sulfate	AR
Nickel Nitrate	AR
Granular Activated Carbon	AR
Acetic Acid	AR
Acetonitrile	HPLC Grade
Ammonia Solution	AR (25%)

, respectively. All reagents were used as received without further purification. Instruments were calibrated per standard operating procedures to ensure data accuracy.

2.2. Experimental Methods

2.2.1. PEM Electrolytic Ozonation for SMX Degradation

The experimental setup and procedure are shown in Figure 1. The system comprises an anode chamber, a cathode chamber, and a PEM (Nafion 117). The anode was a Ti/RuO₂-IrO₂ electrode, and the cathode was stainless steel. A solution containing 50 mg/L SMX (with 0.27% sodium sulfate as the electrolyte, pH 5.59 ± 0.05) was pumped into both chambers at 10 mL/min using a peristaltic pump. Under a DC power supply (current 4.3 A), the anode oxidized water to produce ozone (O₃) and oxygen (O₂), while the cathode reduced water to generate hydrogen (H₂). The PEM facilitated H⁺ migration from anode to cathode, preventing O₂⁻ passage and ensuring ionic selectivity. Electrons flowed externally from the anode to the cathode to maintain charge balance.

Standalone ozonation showed limited SMX degradation efficiency. To enhance ozone utilization, 1 g of NiFe₂O₄/GAC catalyst was added to the anode chamber to promote ozone decomposition into ·OH, improving SMX degradation. Experimental conditions were as follows: current, 4.3 A; sodium sulfate, 0.27%; pH 5.59; and electrolysis time, 30 min, with three replicates. Without a catalyst, SMX degradation reached 95.06% ± 0.6%. Catalysts prepared under varying conditions were compared to identify optimal parameters.

For reusability tests, the optimal catalyst was recovered after each trial, washed with deionized water to neutrality (pH ≈ 7), dried at 80 °C for 3 h, and reused four times (five trials total). SMX concentrations were analyzed per trial to calculate degradation rates and assess reusability.

2.2.2. Catalyst Preparation

The hydrothermal impregnation method was selected for NiFe₂O₄/GAC synthesis due to its documented effectiveness in achieving the uniform dispersion of metal oxides on porous supports while preserving the structural integrity of the carrier [17,19]. NiFe₂O₄/GAC was prepared via hydrothermal impregnation, as depicted in Figure 2. FeCl₃·6H₂O, FeSO₄·7H₂O, and Ni(NO₃)₂·6H₂O (total 0.02 mol) were weighed per the target Fe³⁺ + Fe²⁺:Ni²⁺ molar ratio, dissolved in 50 mL deionized water, and stirred until fully dissolved. Granular activated carbon (GAC, 1–2 mm, BET surface area ~800 m²/g) was added to the solution and stirred at 60 °C for 1 h in a water bath. Then, 25% ammonia solution (5–10 mL) was slowly added to adjust the pH to 9.0–10.0, followed by 1 h stirring. The Fe/Ni-loaded GAC was separated via vacuum filtration (0.45 μm membrane), washed with deionized water until neutral (pH 6.8–7.2), and dried at 80 °C for 12 h to constant weight, yielding the catalyst precursor. The precursor was calcined in a muffle furnace at set temperatures (400–700 °C) and durations (1–5 h, heating rate 5 °C∙min⁻¹), cooled naturally, ground uniformly, and stored sealed.

2.2.3. Parameter Ranges for Catalyst Preparation Conditions

Single-factor experiments optimized the Fe:Ni ratio, GAC loading, calcination temperature, and time. For each test, a 1 g catalyst was added to the anode chamber and electrolyzed for 25 min (current, 4.3 A; sodium sulfate, 0.27%; pH 5.59 ± 0.05; initial SMX, 50 mg/L). SMX degradation rates were measured, with three replicates per condition.

Fe:Ni Ratio: Fixed GAC:Fe + Ni mass ratio at 1:1; calcination at 500 °C for 4 h. Fe³⁺ + Fe²⁺:Ni²⁺ molar ratios tested were 1:1, 2:1, 3:1, 4:1, and 5:1.

GAC Loading: Using the optimal Fe:Ni ratio, GAC:Fe + Ni mass ratios were set at 1:1, 2:1, 3:1, and 4:1 and calcined at 500 °C for 4 h.

Calcination Temperature: Using optimal Fe:Ni and GAC ratios, temperatures tested were 400 °C, 500 °C, 600 °C, and 700 °C, with 4 h calcination.

Calcination Time: Using the optimal Fe:Ni, GAC, and temperature, the times tested were 1 h, 2 h, 3 h, 4 h, and 5 h.

2.3. Methods for Catalyst Characterization

2.3.1. Methods for XRD Analysis

The crystal structure was analyzed using an X-ray diffractometer (Bruker D2 Phaser, Bremen, Germany, Cu Kα, λ = 1.5406 Å). Samples were ground to 200 mesh, pressed into pellets, and scanned from 10° to 80° at 2°∙min⁻¹. Diffraction data were compared with standard PDF cards to identify NiFe₂O₄ phases, with crystal size calculated via the Scherrer equation.

2.3.2. Methods for SEM Analysis

Surface morphology was observed using a scanning electron microscope (Hitachi Regulus 8100, Tokyo, Japan). Samples (<5 × 5 mm) were mounted on conductive tape, gold-coated (Quorum SC 7620, Kent, UK, 10 mA, 10 s), and imaged at 3 kV (morphology, SE2 detector) or 20 kV (EDS analysis). SEM-EDS provided qualitative and quantitative elemental distribution (Fe, Ni, C, and O).

2.3.3. Methods for TEM Analysis

Internal morphology was examined using a transmission electron microscope (JEOL JEM-2100F, Tokyo, Japan). Samples were ground, ultrasonically dispersed in ethanol for 10 min, dropped onto a 300-mesh copper grid, dried, and imaged at 200 kV. Selected area electron diffraction (SAED) verified the crystal structure.

2.3.4. Methods for XPS Analysis

Surface elements and valence states were analyzed using an X-ray photoelectron spectrometer (Thermo Scientific K-Alpha, MA, USA) at <2.0 × 10⁻⁷ mbar. Samples (5 × 5 mm) were tested with a 400 μm spot size, 12 kV, and 6 mA. Full scans used 150 eV pass energy (1 eV step); narrow scans used 50 eV (0.1 eV step). Binding energies were calibrated to C 1s (284.6 eV).

2.3.5. Methods for FT-IR Analysis

Chemical bonds were analyzed using a Fourier transform infrared spectrometer (Thermo Scientific Nicolet iS 20, Waltham, MA, USA). Samples were mixed with KBr (dried at 200 °C for 24 h) at 1:100, ground, and pressed into pellets. Spectra were recorded from 400 to 4000 cm⁻¹ (4 cm⁻¹ resolution, 32 scans) after collecting a KBr background.

2.4. Analytical Methods

2.4.1. Ozone Concentration

Ozone concentration at the anode chamber outlet was measured using a portable analyzer (Luheng LH-D01F, Hangzhou, China). A 5 mL water sample was added to a cuvette containing indigo disulfonate, shaken for 40 s, and measured. After 5 min of device operation, ozone concentrations were recorded at currents of 1–5 A (Figure 3), showing a near-linear increase with current, reflecting improved electrolytic efficiency.

2.4.2. SMX Concentration

SMX concentrations were determined using high-performance liquid chromatography (Agilent 1260, Santa Clara, CA, USA) with a C18 column (4.6 × 250 mm, 5 μm). Mobile phase: 0.20% acetic acid aqueous solution/acetonitrile (60:40, v/v); flow rate: 1 mL/min; column temperature: 35 °C; injection volume: 10 μL; detection wavelength: 270 nm. A standard curve (Figure 4) was prepared from 0–50 mg/L SMX solutions, yielding y = 49.536x + 23.454 (R² = 0.9998), meeting experimental requirements.

3. Results and Discussion

3.1. Optimization of Catalyst Preparation Conditions

The Fe:Ni ratio, GAC loading, calcination temperature, and time were optimized to maximize SMX degradation (initial SMX, 50 mg/L; current, 4.3 A; sodium sulfate, 0.27%; pH 5.59 ± 0.05; catalyst, 1 g; 30 min electrolysis, with three replicates).

3.1.1. Effect of Fe:Ni Ratio

Catalysts were prepared with Fe³⁺ + Fe²⁺: Ni²⁺ molar ratios of 1:1, 2:1, 3:1, 4:1, and 5:1 (GAC:Fe + Ni = 1:1, 500 °C, 4 h). The degradation results (Figure 5) showed SMX degradation increasing and then decreasing with higher Fe:Ni ratios. At Fe:Ni = 3:1, the 25 min degradation reached 98.02% ± 0.5%, outperforming the non-catalyzed system (95.06% ± 0.6%, 30 min) by ~5 min. Fe³⁺/Fe²⁺ redox cycling likely promoted ozone decomposition into ·OH, with Ni²⁺ enhancing activity as a co-catalytic site [20]. Beyond 3:1, excess Fe may cause crystal defects or uneven active site distribution on GAC, reducing efficiency [21]. EDS confirmed an Fe:Ni atomic ratio near the theoretical spinel value (~2:1) at 3:1, indicating structural stability. Thus, Fe:Ni = 3:1 was optimal.

3.1.2. Effect of GAC Loading

Using Fe:Ni = 3:1, GAC:Fe + Ni mass ratios of 1:1, 2:1, 3:1, and 4:1 were tested (500 °C, 4 h). Degradation rates (Figure 6) increased and then decreased with GAC loading. At 2:1, 25 min degradation reached 98.93% ± 0.4%, surpassing 1:1 (96.45% ± 0.5%) and 4:1 (94.12% ± 0.6%). Initially (0–2 min), 1:1 showed the fastest rate (30.5% ± 0.8%) due to higher NiFe₂O₄ loading per GAC unit, concentrating active sites. Beyond 2 min, 2:1 excelled due to synergistic GAC adsorption and NiFe₂O₄ catalysis [21]. At 4:1, lower NiFe₂O₄ loading relied heavily on GAC adsorption, limiting ·OH production. Thus, GAC:Fe + Ni = 2:1 was optimal.

3.1.3. Effect of Calcination Temperature

Using Fe:Ni = 3:1 and GAC:Fe + Ni = 2:1, calcination temperatures of 400 °C, 500 °C, 600 °C, and 700 °C were tested (4 h). Degradation rates (Figure 7) peaked at 500 °C (98.93% ± 0.4% in 25 min), outperforming 400 °C (95.67% ± 0.5%) and 700 °C (94.02% ± 0.6%). At 500 °C, NiFe₂O₄ formed appropriately sized crystals with intact pore structures, facilitating ozone adsorption and ·OH generation [18]. At 400 °C, low crystallinity reduced active sites; at 700 °C, crystal agglomeration and pore collapse lowered efficiency. Thus, 500 °C was optimal.

3.1.4. Effect of Calcination Time

Using Fe:Ni = 3:1, GAC:Fe + Ni = 2:1, and 500 °C, calcination times of 1 h, 2 h, 3 h, 4 h, and 5 h were tested. Degradation rates (Figure 8) peaked at 3 h (99.15% ± 0.3% in 25 min), surpassing 1 h (94.02% ± 0.6%) and 5 h (95.65% ± 0.5%). At 3 h, NiFe₂O₄ formed a stable spinel structure with optimized porosity, enhancing SMX and ozone adsorption [18]. Shorter times (1–2 h) yielded incomplete crystallization, while 5 h caused crystal growth or impurity deposition, clogging pores. Thus, 3 h was optimal.

The NiFe₂O₄/GAC composite catalyst was optimally prepared via hydrothermal impregnation under the following conditions: Fe:Ni molar ratio (3:1), GAC:(Fe + Ni) mass ratio (2:1), and calcination at 500 °C for 3 h. Under these conditions, the catalyst achieved 99.15% ± 0.3% sulfamethoxazole (SMX, 50 mg/L) degradation in just 25 min in a PEM electrolytic ozonation system. This performance surpassed non-catalytic ozonation (95.06% ± 0.6% in 30 min) by both efficiency (+4.1%) and reaction rate (17% faster).

3.2. Multi-Technique Characterization of NiFe₂O₄/GAC

The optimal catalyst (Fe:Ni = 3:1, GAC:Fe + Ni = 2:1, 500 °C, 3 h) was characterized using XRD, SEM, EDS, TEM, XPS, and FT-IR to correlate structure and composition with performance.

3.2.1. XRD Characterization of NiFe₂O₄ Spinel Structure on GAC

XRD patterns (Figure 9, Cu Kα, λ = 1.5406 Å, 10–80°) showed peaks at 2θ = 18.4°, 30.2°, 35.6°, 43.3°, 53.8°, 57.3°, 63.0°, and 74.5°, matching NiFe₂O₄ spinel planes (111, 220, 311, 400, 422, 511, 440, and 533; PDF#74-2081) without significant impurities, confirming successful NiFe₂O₄ loading on GAC. The (311) peak yielded a crystal size of 15.2 ± 0.5 nm, indicating high crystallinity and active site exposure [22]. A weak GAC amorphous carbon peak (~26°) suggested NiFe₂O₄ coverage. TEM corroborated the nanocrystalline structure.

3.2.2. SEM Morphological Characterization of NiFe₂O₄/GAC

SEM images (Figure 10, 3 kV) revealed a porous GAC surface with uniformly distributed NiFe₂O₄ particles, with some forming clusters. The porous structure and particle distribution favored SMX and ozone adsorption/catalysis [23]. High-magnification images show spherical NiFe₂O₄ particles with rough surfaces, increasing active site exposure, consistent with high degradation rates.

3.2.3. EDS Elemental Mapping and Quantitative Analysis of NiFe₂O₄/GAC

SEM-EDS mapping (Figure 11, 20 kV) showed the uniform distribution of C, O, Fe, and Ni without significant segregation (Figure 12). Quantitative EDS revealed mass fractions—C 68.5% ± 1.2%, O 21.3% ± 0.8%, Fe 7.8% ± 0.3%, Ni 2.4% ± 0.2%—with an Fe:Ni atomic ratio of ~2.1:1, close to NiFe₂O₄’s theoretical value, confirming successful loading [20]. C and O stemmed from GAC, Fe, and Ni from NiFe₂O₄, aligning with the XRD.

3.2.4. TEM Structural Characterization of NiFe₂O₄

TEM images (Figure 13, 200 kV) show spherical NiFe₂O₄ particles with clear lattice fringes (0.25 nm, (311) plane), matching the XRD. Particles were uniformly distributed on GAC with some aggregation. Pore sizes of 10–20 nm facilitated molecular diffusion, supporting the spinel structure [24]. Nanocrystals and porosity enhanced ozone decomposition and SMX degradation.

3.2.5. XPS Analysis of Surface Chemical States in NiFe₂O₄

XPS spectra (Figure 14a) confirmed O 1s (531.1 eV), C 1s (284.6 eV), Fe 2p (710.7 eV), and Ni 2p (855.4 eV). C 1s (Figure 14b) deconvoluted into C-C (284.6 eV, 65.2%), C-O (286.1 eV, 22.5%), and O-C=O (288.2 eV, 12.3%), reflecting GAC’s carbon framework. O 1s (Figure 14c) showed lattice oxygen (530.8 eV, 48.6%) and adsorbed oxygen (531.9 eV, 51.4%), the latter aiding ·OH formation [25]. Fe 2p (Figure 14d) displayed Fe 2p3/2 (710.7 eV, Fe²⁺; 712.9 eV, Fe³⁺) and Fe 2p1/2 (724.2 eV, Fe²⁺; 726.3 eV, Fe³⁺), with an Fe³⁺/Fe²⁺ ratio of ~1.8:1, indicating redox activity. Ni 2p (Figure 14e) showed Ni 2p3/2 (855.4 eV, Ni²⁺) and Ni 2p1/2 (873.2 eV, Ni²⁺). Fe³⁺/Fe²⁺ cycling and Ni²⁺ synergy drove ozone decomposition, supporting high degradation rates.

3.2.6. FT-IR Spectroscopic Characterization of NiFe₂O₄

FT-IR spectra (Figure 15, 400–4000 cm⁻¹) showed O-H stretching/bending (3432, 1634 cm⁻¹) from adsorbed water [25] and NiFe₂O₄’s Fe-O/Ni-O vibrations (586, 407 cm⁻¹), confirming spinel structure [26]. O-H and Fe-O bonds facilitated ozone adsorption/decomposition, supporting catalytic activity.

3.3. Catalyst Reusability

The optimal catalyst was tested for reusability over five cycles (washed to pH ≈ 7 and dried at 80 °C for 3 h per cycle, with conditions as in Section 3.1). Degradation rates (Figure 16) remained stable for three cycles (98.45% ± 0.5% at cycle 3, −0.7%), dropping to 94.12% ± 0.6% (cycle 4) and 93.13% ± 0.7% (cycle 5). Efficiency decline may result from minor Fe leaching or organic deposition clogging pores [27]. Overall, the catalyst exhibited excellent reusability for long-term applications.

4. Conclusions

NiFe₂O₄/GAC was successfully prepared via hydrothermal impregnation using FeCl₃·6H₂O, FeSO₄·7H₂O, and Ni(NO₃)₂·6H₂O. Optimized via single-factor experiments and characterized comprehensively, the catalyst excelled in PEM electrolytic ozonation for SMX degradation, yielding the following:

Optimal Conditions: Fe:Ni = 3:1, GAC:Fe + Ni = 2:1, 500 °C, 3 h. SMX (50 mg/L) degradation reached 99.15% ± 0.3% in 25 min, 4.1% higher and 5 min faster than without a catalyst (95.06% ± 0.6%, 30 min). Balanced Fe³⁺/Fe²⁺ cycling and Ni²⁺ synergy, coupled with GAC’s high surface area (~650 m²/g), enhanced adsorption and catalysis.

Structure and Performance: XRD and TEM confirmed spinel NiFe₂O₄ (~15.2 nm) uniformly loaded on GAC. SEM/EDS showed regular particles (20–50 nm) and uniform elemental distribution (Fe:Ni~2.1:1). XPS revealed Fe³⁺/Fe²⁺ (1.8:1) and Ni²⁺ coexistence, with Fe³⁺/Fe²⁺ cycling promoting ·OH generation and GAC adsorption enriching SMX.

Reusability: After three cycles, degradation remained at 98.45% ± 0.5%, dropping to 95.33% ± 0.7% by cycle 5, indicating robust stability. Minor Fe leaching and pore-clogging caused slight declines, addressable via optimized recovery.

NiFe₂O₄/GAC demonstrates high efficiency in PEM ozonation for SMX removal, offering a viable solution for antibiotic wastewater treatment. Future work could explore its performance in complex water matrices, optimize electrolysis for energy efficiency, or use molecular simulations to elucidate SMX degradation pathways, paving the way for scaled-up catalytic ozonation.