Mechanism and Pathway of Atrazine Degradation by Peroxymonosulfate Activated by CoNiFe-Layered Double Hydroxide

Abstract

Advanced oxidation processes (AOPs) based on activated persulfate (PS) are gradually being employed in the treatment of novel pollutants. In this study, an efficient and reliable CoNiFe-layered double hydroxide (LDH) was prepared by a hydrothermal method, which could effectively activate peroxomonosulfate (PMS) and cause free sulfate radical (SO₄^•−) oxidation to decompose atrazine (ATZ). The degradation rate of ATZ was greater than 99% within 60 min at pH 7 when the initial concentration of ATZ was 10 mg·L⁻¹, and the dosages of PMS and activator were 0.6 mM and 80 mg·L⁻¹. The analysis of ATZ degradation confirmed the reusability of the activator and its strong structural stability. The generation of four free radicals was analyzed and confirmed, and the influence on the degradation reaction was SO₄^•− > O₂^•− > ¹O₂ > •OH. The analytical results showed that the metal ions reacted with HSO₅⁻ in PMS to cause an oxidation–reduction cycle change in the valence state of the metal ions and generated the primary factor affecting the degradation reaction—SO₄^•−. Nine degradation intermediates with reduced toxicity were detected and possible ATZ degradation pathways were deduced, thus confirming the activation mechanism of CoNiFe-LDH.

1. Introduction

Pesticides have gradually become an essential part of agricultural production. However, pesticides that are not absorbed by plants will be discharged into the surface water with rainwater or through the soil to groundwater [1]. 2-Chloro-ethylamino-6-isopropylamino-1,3,5-triazine, known as atrazine (ATZ), is a herbicide [2]. ATZ is classified as persistent organic pollutants (POPs) due to the slow photolysis of atrazine under natural conditions and the difficulty of microbial decomposition [3]. ATZ is highly mobile and stable under natural conditions [4]. As a result, it has contaminated many water sources around the world in varying concentrations [5,6,7,8,9]. Atrazine is also an endocrine disruptor, causing potential damage to human organs and even accelerating the growth and metastasis of cancer cells [10].

Nowadays, ATZ can be treated with biological [11,12,13], physical [14,15,16], or chemical oxidation technologies [17,18,19]. However, most of these methods have limitations in terms of their applicability and efficiency in removing or reducing the concentration of toxic and difficult-to-degrade organics in wastewater to below acceptable levels. The development of more efficient and reliable processing methods is crucial for practical applications [20,21]. Sulfate-based advanced oxidation processes (SR-AOPs), utilizing SO₄^•− oxidation, offer cost-effective and powerful degradation of organic pollutants compared to traditional oxidants. SR-AOPs using SO₄^•− oxidation can cost-effectively degrade organic pollutants compared to conventional oxidants [22,23,24]. SO₄^•− generated through the cleavage of O−O bonds in activated peroxomonosulfate (PMS) serves as a potent oxidizing agent, exhibiting a redox potential ranging from 2.5 to 3.1 eV [25]. SO₄^•− demonstrates high selectivity and oxidizing power through electron transfer, facilitating the degradation of pollutant macromolecules into smaller, less toxic entities by cleaving their chemical bonds. SR-AOPs are advantageous due to their long half-life (30−40 s) and broad pH applicability, surpassing other oxidative processes [26,27,28]. However, the degradation efficiency of PMS in direct reactions with pollutants is inferior to that achieved through reactions facilitated by technologies or materials, as these enhance the rapid formation of free radicals. And PMS has an asymmetric molecular structure and is susceptible to activation by transition metals [29].

The extensive utilization of layered double hydroxides (LDHs) as activators for PMS and as carriers of transition metals is due to their cost-effectiveness, extensive specific surface area, and exceptional reactivity [30,31]. Influenced by the octahedral structure of Cu(OH)₆ subjected to the Jahn–Teller effect, Yan et al. [32] chose to introduce non-toxic magnesium ions as covalent cations. They used 0.2 g·L⁻¹ CuMgFe-LDH and 4.0 mmol·L⁻¹ PMS at pH 7.6 to achieve a 93.7% degradation of 0.08 mmol·L⁻¹ ethylbenzene, with minimal Cu leaching (0.095 mg·L⁻¹). The special chemical structure of LDHs enhances their role as metal-based activators, promoting metal ion circulation and low dissolution rates [33]. Transition metal ions like Co²⁺, Fe²⁺, and Ni²⁺ can activate PMS [34], with the introduction of additional metal ions increasing the number of hydroxyl on the surface of the original bimetallic-LDH, accelerating redox cycling, and thus improving activation performance and pollutant degradation efficiency [35].

Therefore, in this study, a CoNiFe-LDH activator was synthesized by a simple hydrothermal method, and the structural characteristics of the resulting material were investigated and compared briefly with hydrotalcite doped with only two metals. The purpose of this work was to compare the changes in the characterization of the material after the addition of a third metal. The stability and reusability of the prepared CoNiFe-LDH were investigated. In addition, free radical scavenging experiments were utilized to identify the types of free radicals as well as the order of strength in which they functioned, the intermediates in the degradation process were discovered, and possible degradation pathways and mechanisms were proposed.

2. Materials and Methods

2.1. Materials

Peroxide sulfate was purchased from Shanghai Maclin Biochemical Technology Co., Ltd. (Shanghai, China), and ATZ was obtained from Shanghai Pedder Pharmaceutical Technology Co., Ltd. (Shanghai, China). All the chemical reagents used in the experiments were of analytical grade and procured from Aladdin Reagent Co., Ltd. (Shanghai, China). Deionized water was used throughout the experiments.

2.2. Preparation

The CoNiFe-LDH activator was prepared through a hydrothermal process [36]. First, 2.170 g of Co(NO₃)₂·6H₂O, 0.730 g of NiCl₂·6H₂O, 2.420 g of Fe(NO₃)₃·9H₂O, and 2.520 g of urea were added to 80 mL of deionized water in a beaker. The mixture was stirred with a magnetic stirrer until the solid particles had completely dissolved, and then poured into a 100 mL stainless steel reactor and placed in an oven, where the hydrothermal reaction proceeded at 120 °C for 20 h. After the reaction, the reactor was naturally cooled to room temperature, the precipitate was centrifuged, and the clear liquid was repeatedly washed to neutrality with deionized water. The obtained solids were dried in an oven at 80 °C for 12 h. Finally, the CoNiFe-LDH activator was obtained as a brown solid. The preparation process for NiFe-LDH and CoFe-LDH was the same as above. The workflow is shown in Figure 1.

2.3. Characterization

The surface morphologies of the prepared materials were analyzed using scanning electron microscopy (SEM, Zeiss Sigma 300, Oberkochen, Germany). CoNiFe-LDH was characterized using X-ray diffraction (XRD, Paneco X’Pert PRO, Almelo, The Netherlands), which can determine the elemental composition and crystal structure of the CoNiFe-LDH activator. X-ray electron spectroscopy (XPS, Thermo Scientific K-Alpha, Waltham, MA, USA) analyzed the elements and the valence state of each element before and after the reaction. The obtained data were fitted and analyzed using XPS PEAK software. Fourier transform infrared spectroscopy (FT-IR, Nicolet 670, Thermo Scientific, Waltham, MA, USA) was used to reveal the possible functional groups in the materials. The specific surface area and pore structures of the CoNiFe-LDH were analyzed using an automatic specific surface area tester (BET, ASP-2460, Micromeritics Corporation, Norcross, GA, USA).

3. Results

3.1. Characterization of CoNiFe-LDH Activator

3.1.1. SEM

Figure 2 presents SEM images depicting the surface morphology of the synthesized CoNiFe-LDH activator. These images show a hierarchical structure similar to that of conventional LDH: consisting of nanosheets with smooth and irregular surfaces stacked on top of each other [37]. Notably, the surface of the CoNiFe-LDH activator is porous, indicating that it has a large specific surface area, which can effectively enhance the activation of PMS to produce more SO₄^•−. The magnified image further elucidates the presence of numerous small monolithic structures and rice-like granular particles on the surface of the activator, formed through interactions between LDH nanoparticles.

3.1.2. XRD

As shown in Figure 3, the XRD spectra of all three materials, CoNiFe-LDH, CoFe-LDH, and NiFe-LDH, show several high and narrow peaks at 2θ = 11.5°, 22.8°, 34.5°, 38.8°, and 59.9°, corresponding to the (003), (006), (012), (015), and (110) crystal planes of LDH, respectively [38]. The (003) peak is the signature diffraction peak of LDH materials [39,40]. These peaks, which can be clearly observed, are the signature diffraction peaks of LDH materials. In addition, the peaks of the (003) and (006) crystal planes of CoNiFe-LDH are somewhat reduced relative to those of CoFe-LDH and NiFe-LDH, possibly reflecting the different atomic radii of Co, Ni, and Fe [41].

3.1.3. XPS

Figure 4 shows the full XPS spectrum and the fine spectra of Co, Ni, and Fe, which show strong diffraction peaks at 284.29, 531.7, 781.3, 710.69, and 855.52 eV corresponding to C 1s, O 1s, Co 2p, Fe 2p, and Ni 2p, respectively. That is, the material was mainly composed of C, O, Co, Fe, and Ni [42]. In the Co 2p map, the two strong peaks at 797.2 eV and 781.3 eV represent Co 2p1/2 and Co 2p3/2, respectively. The intensity difference between the two peaks is 15.9 eV, indicating the presence of Co²⁺ and Co³⁺ in CoNiFe-LDH [43]. In the Fe 2p map, the two sets of strong peaks at 710.9 and 724.3 eV represent Fe 2p3/2 and Fe 2p1/2, respectively, indicating the presence of Fe³⁺ [44]. In the Ni 2p map, the intensity peaks at 856.1 and 873.5 eV represent Ni 2p3/2 and Ni 2p1/2, respectively [45]. In the ternary material CoNiFe-LDH, the primary functionality is attributed to Ni; Fe and Co play supporting roles [46]. The binding energies of Ni 2p, Fe 2p, and Co 2p show a significant increase, which suggests that there are electronic interactions between Ni, Fe, and Co in this material, and the coordinated interaction between these metals has a non-negligible effect on the activation of PMS.

3.1.4. FT-IR

The FT-IR spectra of NiFe-LDH, CoFe-LDH, and CoNiFe-LDH are shown in Figure 5. The FT-IR spectra of the three materials show obvious characteristic peaks at 3480, 1361, and 2340 cm⁻¹, respectively. The characteristic peak at 3480 cm⁻¹ is caused by the stretching vibrations of the hydroxyl groups of the interlayer water molecules [47,48]. The 3480 cm⁻¹ peak is considerably more enhanced in the CoNiFe-LDH spectrum than in the spectra of the NiFe-LDH and CoFe-LDH because the addition of the third metal increases the number of surface hydroxyl [49]. Many characteristic peaks in the 400–800 cm⁻¹ range were caused by the stretching vibrations of M–O and M–OH (M = Co, Ni, Fe), produced by metal oxides and metal hydroxides. The FT-IR results further demonstrate that these metal–oxygen bonds help to facilitate electron transfer between Fe, Ni, and Co, thus favoring the activation of PMS [50].

3.1.5. BET

Figure 6 presents the N₂ adsorption–desorption and pore-size distribution curves for the CoNiFe-LDH activator under isothermal conditions, with detailed void structure data provided in

Table 1. Structural parameters of the CoNiFe-LDH activator.

Material	SBET * (m2·g−1)	Smicro (m2·g−1)	Smeso (m2·g−1)	Vtot (cm2·g−1)	Vmicro (cm2·g−1)	Vmeso (cm2·g−1)
CoNiFe-LDH	31.1249	5.5202	25.6047	0.1370	0.0022	0.1348

* S_BET is the specific surface area of the material, S_micro is the specific surface area of micropores, S_meso is the specific surface area of mesopores, V_tot is the total pore volume, V_micro is the micropore volume, and V_meso is the mesopore volume.

. The adsorption–desorption curve predominantly exhibits a type IV pattern, characterized by convex upward isotherms in the low relative pressure region and a steep rise in the high P/P₀ region. As the relative pressure approaches 1.0, adsorption in macropores leads to a continuous rise in the curve, indicating the formation of a mesoporous multimolecular layer structure. At P/P₀ = 0.9, the appearance of H3-type hysteresis loops suggests an irregular pore structure. The S_micro analysis reveals a substantial S_micro, providing numerous active sites for PMS activation.

3.2. CoNiFe-LDH Activation Performance

3.2.1. ATZ Degradation in PMS/CoNiFe-LDH System

To demonstrate the removal efficacy, the degradation efficiency of the CoNiFe-LDH/PMS system was contrasted with the PMS system alone at 10 mg·L⁻¹ ATZ, 0.6 mM PMS, 80 mg·L⁻¹ CoNiFe-LDH, and pH 7.0, as depicted in Figure 7.

As shown in Figure 7, in the degradation experiment, PMS is a strong oxidizing agent with a high standard reduction potential, but the presence of PMS alone resulted in the removal of only 10.6% of ATZ within 60 min. Under the same conditions, the adsorption experiment where only the activator was added without PMS led to the absorption of merely 6.4% of ATZ within 60 min. Conversely, the degradation rate of CoNiFe-LDH/PMS system was significantly accelerated in the first 30 min, and the removal rate reached 99.9% within 60 min, indicating the effective activation of CoNiFe-LDH on PMS. This enhanced efficacy is likely due to the ability of the activator to rapidly produce a significant amount of free radical SO₄^•− in water, facilitating the complete degradation of ATZ. It further indicates that the activator and PMS had a synergistic effect on degradation.

3.2.2. Effect of Inorganic Anions and Humic Acid

As shown in Figure 8, five ions as well as humic acid (HA) were selected to study their effects on the ATZ removal rate, and the results showed that the five ions and HA had varying degrees of inhibition on the ATZ removal ratio and reaction rate.

As with the results of Inayat et al. [51], the most significant inhibitory effect was caused by CO₃²⁻. The removal rate of ATZ decreased from 99.9% to 67.3% after 60 min of reaction, and the quasi-first-order kinetic constant also decreased from 0.1325 to 0.0184 min⁻¹. This is due to the rapid reaction of CO₃²⁻ with SO₄^•− and hydroxyl radical (•OH) to generate a carbonate radical (CO₃^•−) with low oxidation activity. In addition, the presence of CO₃²⁻ in the solution increases the pH, which will lead to a decrease in the production and quantity of free radicals in the alkaline system when the concentration of CO₃²⁻ is high, ultimately leading to a decrease in the pollutant removal ratio. The specific reaction is shown in Equations (1)–(5).

$${\mathrm{CO}}_3^{2-}+{\mathrm{H}}_2\mathrm{O}\to {\mathrm{OH}}^{-}+\mathrm{H}{\mathrm{CO}}_3^{2-},$$

(1)

$${\mathrm{SO}}_4^{\bullet -}+{\mathrm{CO}}_3^{2-}\to {\mathrm{SO}}_4^{2-}+{\mathrm{CO}}_3^{·-},$$

(2)

$${\mathrm{SO}}_4^{\bullet -}{+\mathrm{HCO}}_3^{2-}\to {\mathrm{SO}}_4^{2-}+\mathrm{H}{\mathrm{CO}}_3^{·},$$

(3)

$$\bullet \mathrm{OH}+{\mathrm{CO}}_3^{2-}\to {\mathrm{OH}}^{-}+{\mathrm{CO}}_3^{·-},$$

(4)

$$\bullet \mathrm{OH}+{\mathrm{H}\mathrm{CO}}_3^{2-}\to {\mathrm{H}}_2\mathrm{O}+{\mathrm{CO}}_3^{·-},$$

(5)

HPO₄²⁻ also inhibits the degradation of ATZ by generating low-activity phosphate radicals through quenching SO₄^•− and •OH. The specific reaction is shown in Equations (6) and (7). Additionally, HPO₄²⁻ forms stable complexes with surface Co²⁺ and Fe³⁺, reducing the activation site on the activator surface and thus weakening the activation of PMS, resulting in a sharp decline in the ATZ removal rate.

$${\mathrm{SO}}_4^{\bullet -}+{\mathrm{HPO}}_4^{2-}\to {\mathrm{SO}}_4^{2-}+{\mathrm{HPO}}_4^{·-},$$

(6)

$$\bullet \mathrm{OH}+{\mathrm{HPO}}_4^{2-}\to {\mathrm{OH}}^{-}+{\mathrm{HPO}}_4^{·-},$$

(7)

NO₃⁻, SO₄²⁻, and Cl⁻ have a relatively weak inhibitory effect on ATZ, and the degradation rate is significantly reduced, which may be because these anions are more inclined to react with free radicals and compete with pollutants to produce new free radicals with weak oxidation potential. The specific reaction is shown in Equation (8). At this time, a small number of anions that do not participate in the reaction are adsorbed on the surface of the activator, occupying the active site of the reaction, thereby reducing the ability of the activator to activate PMS. In addition, the effect of Cl⁻ on ATZ removal is due to the fact that Cl⁻ gradually consumes part of SO₄^•−and •OH to produce active chlorine with lower oxidizing capacities than SO₄^•− and •OH. The specific reaction is shown in Equations (9)–(12). It consumes some free radicals to generate HClO and Cl₂ with lower oxidation abilities than SO₄^•− and •OH. The specific reaction is shown in Equations (13) and (14). However, HClO and Cl₂ also have the oxidation ability to degrade ATZ. Overall, the impact of Cl⁻ on ATZ degradation is not significant.

$${\mathrm{SO}}_4^{\bullet -}+{ \mathrm{NO}}_3^{-}\to {\mathrm{SO}}_4^{2-}+{\mathrm{NO}}_3^{·},$$

(8)

$${\mathrm{SO}}_4^{\bullet -}+{\mathrm{Cl}}^{-}\to {\mathrm{SO}}_4^{2-}+{\mathrm{Cl}}^{·},$$

(9)

$$\bullet \mathrm{OH}+{\mathrm{Cl}}^{-}\to {\mathrm{Cl}+\mathrm{HO}}^{·-},$$

(10)

$${\mathrm{Cl}}^{-}+{\mathrm{Cl}}^{·}\to {\mathrm{Cl}}_2^{·-},$$

(11)

$${2\mathrm{Cl}}^{·}\to {\mathrm{Cl}}_2,$$

(12)

$${2\mathrm{Cl}}^{-}+{\mathrm{HSO}}_5^{-}+{\mathrm{OH}}^{+}\to {{\mathrm{SO}}_4^{2-}+\mathrm{Cl}}_2+{\mathrm{H}}_2\mathrm{O},$$

(13)

$${\mathrm{Cl}}^{-}+{\mathrm{HSO}}_5^{-}\to {\mathrm{SO}}_4^{2-}+\mathrm{HClO},$$

(14)

HA can scavenge free radicals by competing with SO₄^•− and •OH. When phenolic hydroxyl and -COOH in humic acid adsorb onto the surface of the activator, they can hinder the degradation reaction. Additionally, the formation of semianthraquinone radicals by quinones, hydroquinones, and phenols in HA can activate PMS to produce SO₄^•− and •OH [52]. The adsorption of the phenolic hydroxyl groups and -COOH functionalities in HA to the activator surface may occlude the CoNiFe-LDH active sites, thereby impeding the degradation reaction.

3.3. Structural Stability and Reusability

To determine the structural stability of the CoNiFe-LDH activator, the spectra obtained from XPS, XRD and FT-IR after the reaction were compared with those before the reaction. The results are shown in Figure 9. In the XPS spectra of Figure 9a, the position of each diffraction peak of CoNiFe-LDH is not noticeably changed after the reaction; the elemental composition was well maintained after the reaction [53], further demonstrating the good robustness of CoNiFe-LDH to the PMS-activated degradation of ATZ. However, in the FT-IR spectra of Figure 9b, the peak in the 400–800 cm⁻¹ range is weakened because the metal oxides in the material participated in PMS activation during the reaction process. As depicted in Figure 9c, the post-activation XRD spectra closely resemble the pre-activation ones, with no discernible shifts in the positions of the characteristic peaks. Figure 9d as well as Table 1 show that the properties of CoNiFe-LDH do not change much before and after the reaction in the repeated tests.

To assess the reusability of CoNiFe-LDH, three cycles of ATZ degradation tests were performed. As depicted in Figure 10, the removal efficiency slightly decreased with each reuse, remaining high at 98.9% in the first cycle and dropping to 94.2% and 84.2% in subsequent cycles, indicating the effective recyclability of the CoNiFe-LDH activator. Following the reaction, the leaching concentrations of Co, Ni, and Fe were determined to be 0.89, 0.66, and 0.154 μg·L⁻¹, respectively. These values are considerably lower than the permissible limits outlined in the Chinese industrial pollutant discharge standards (GB 25467-2010 [54] and GB 13456-2012 [55]), indicating that they do not exceed the defined thresholds for secondary pollution.

In summary, the reduced ATZ removal rate can be attributed to the loss of metal ions and the decrease in activation groups during the reaction, alongside the reduction in specific surface area and pore size, which diminishes active site availability and the contact with pollutant molecules and PMS, thereby weakening the activation capacity of CoNiFe-LDH for PMS.

3.4. Free Radical Types

The reactive oxygen species produced by the CoNiFe-LDH/PMS system during ATZ degradation were determined through EPR. The four possible reactive oxygen species are •OH, SO₄^•−, ¹O_2, and O₂^•−. The selected trapping agents were 5,5-dimethyl-1-pyrrolin-N-oxide (DMPO) for •OH, SO₄^•−, and O₂^•− detection and 2,2,6,6-tetramethylpiperidine (TEMP) for ¹O₂ detection. The EPR results are presented in Figure 11.

The faint DMPO-•OH peak in the EPR spectra with solo PMS addition (Figure 11a) indicates minimal •OH radical production. When CoNiFe-LDH was added, the characteristic peaks of DMPO-•OH and DMPO-SO₄^•− appeared at a reaction time of 5 min. The notable enhancement of the peak intensities indicate large amounts of SO₄^•− and •OH at this time. After 10 min of reaction, the DMPO-•OH peaks were further enhanced whereas the DMPO-SO₄^•− peaks increased only slightly with no notable overall change. Subsequently, SO₄^•− reacts with water to form more •OH radicals, keeping the SO₄^•− concentration steady but increasing the •OH levels. Figure 11b and Figure 11c show the EPR spectra of DMPO-O₂^•− and TEMP-¹O₂, respectively. Their peak intensities rise with extended reaction time, indicating that O₂^•− and ¹O₂ are vital oxidants in the CoNiFe-LDH/PMS system.

Quenching experiments confirm the role of free radicals in ATZ degradation by this system. In the experiments, tert-butanol (TBA) specifically reacts with •OH (kTBA-•OH= (5.9 ± 0.5) × 10⁸ M⁻¹s⁻¹), methanol (MeOH) with •OH (kMeOH-•OH = 1.9 × 10⁹ M⁻¹s⁻¹) and SO₄^•− (kMeOH-SO₄^•− = 1.1 × 10⁷ M⁻¹s⁻¹), furfuryl alcohol (FFA) with •OH (kFFA-•OH = 1.5 × 10¹⁰ M⁻¹s⁻¹) and ¹O₂ (kFFA-¹O₂ = 1.2 × 10⁸ M⁻¹s⁻¹), and p-benzoquinone (BQ) with O₂^•− (kBQ-O₂^•− = 9.8 × 10⁸ M⁻¹s⁻¹) [56]. The PMS-to-quencher ratio was 1:20, and the results are presented in Figure 12.

Quencher addition drastically reduced ATZ removal and reaction rates. MeOH decreased ATZ removal from 99.9% to 36.1% within 60 min and dropped the reaction rate constant from 0.1325 to 0.0081 min⁻¹. This is because the production of SO₄^•− can not only directly attack ATZ to cause degradation, but also provide support for the generation of other free radicals. Meanwhile, α-H in MeOH can remove SO₄^•− [57]. TBA, BQ, and FFA also inhibited ATZ removal, reducing the •OH, ¹O₂ and O₂^•− reaction rate constants. The key free radicals in ATZ degradation are ranked as SO₄^•− > O₂^•− > ¹O₂ > •OH.

3.5. Mechanisms

To examine the interplay of the metal elements, XPS analysis was conducted before and after the reaction, as depicted in Figure 13.

The high-resolution map of Co 2p (Figure 13a) shows six simulated peaks: two peaks of Co³⁺ at 798.6 and 782.6 eV, two peaks of Co²⁺ at 796.9 and 780.9 eV [58], and two satellite peaks at 802.7 and 786.1 eV. Similarly, the Ni 2p (Figure 13b) and Fe 2p (Figure 13c) maps feature six peaks each, corresponding to Ni³⁺ and Ni²⁺ at 875.2 and 856.8 eV, 873.4 and 855.7 eV [59], and Fe³⁺ and Fe²⁺ at 726.1 and 716.3 eV, 724.1 and 710.4 eV, respectively.

The ratios of metal ions in different valence states were analyzed before and after the reaction. Notably, the metal ion ratios (M(III)/M(II)) of CoNiFe-LDH, i.e., the ratios of Co(III)/Co(II), Ni(III)/Ni(II), and Fe(III)/Fe(II) increased from 1.05, 0.72, and 1.46 to 1.53, 0.92, and 2.40, respectively, post-reaction. This result indicates that the bivalent metal ions on the activator surface reacted with the solution, forming trivalent metal ions as shown in Equation (15) and demonstrating the high efficiency of PMS activation by CoNiFe-LDH. Additionally, the peaks of both bivalent and trivalent metal ions shifted to higher-energy positions, indicating a transition from trivalent to bivalent states through a cycle of M(II) → M(III) → M(II) (M = Co, Ni, Fe). The reactions are summarized by Equations (16)–(19). When all three metals were simultaneously added to the reaction, the complete set of cyclic reactions was further promoted to increase the reaction rate [60,61]. This series of reactions enabled the efficient ATZ degradation by the CoNiFe-LDH/PMS system [62].

$$\mathrm{M}(\mathrm{II})+{\mathrm{HSO}}_5^{-}\to \mathrm{M}(\mathrm{III})+{\mathrm{SO}}_4^{\bullet -}+{\mathrm{OH}}^{-},$$

(15)

$$\mathrm{M}(\mathrm{III})+{\mathrm{HSO}}_5^{-}\to \mathrm{M}(\mathrm{II})+{\mathrm{SO}}_5^{\bullet -}+{\mathrm{H}}^{+},$$

(16)

$${\mathrm{Ni}}^{3+}+{\mathrm{Co}}^{2+}\to {\mathrm{Co}}^{3+}+{\mathrm{Ni}}^{2+},$$

(17)

$${\mathrm{Fe}}^{3+}+{\mathrm{Co}}^{2+}\to {\mathrm{Fe}}^{2+}+{\mathrm{Co}}^{3+},$$

(18)

$${\mathrm{Fe}}^{3+}+{\mathrm{Ni}}^{2+}\to {\mathrm{Ni}}^{3+}+{\mathrm{Fe}}^{2+},$$

(19)

Moreover, CoNiFe-LDH possesses a substantial surface area, thereby offering an abundance of active sites. The ATZ degradation mechanism of the CoNiFe-LDH/PMS system involves Co(II) and Ni(II) on the activator surface reacting with HSO₅⁻ to form SO₄^•−, achieving an electron balance. At this time, the generated SO₄^•− reacts with water or hydroxyl to form •OH. The ATZ macromolecules in the solution are attacked and rapidly decomposed by the active substances SO₄^•− and •OH. Meanwhile, trivalent metals react with HSO₅⁻ and reduce to bivalent metals, producing weakly oxidizing SO₅^•−, which can generate •OH. Fe²⁺ reacts with PMS to produce SO₄^•−, completing an activation cycle. The specific degradation is shown in Figure 14.

The known standard redox potentials of Co(III)/Co(II), Ni(III)/Ni(II), and Fe(III)/Fe(II) are 1.81, 0.48, and 0.77 V, respectively. The redox potential of HSO₅⁻/SO₅^•− is 1.10 V. From a thermodynamic perspective, HSO₅⁻ can feasibly reduce Co(III) to Co(II) and produce a large number of reactive oxygen species (ROS), which can then combine with pollutants to form small-molecule products. However, HSO₅⁻ cannot easily reduce Ni(III) and Fe(III) to Ni(II) and Fe(II), respectively. This suggests that Ni(II) and Fe(II) likely originate from the reduction of Co(II), indicating the cycling of metal ions between valence states.

As the reaction proceeds, the free electrons in the solution combine with dissolved oxygen in water to form O₂^•− and also combine with •OH to form ¹O₂. Finally, ATZ is decomposed into CO₂ and H₂O under the action of the major free radicals SO₄^•−, •OH, ¹O₂, and O₂^•−.

Evidently, Co(III)/Co(II), Ni(III)/Ni(II), and Fe(III)/Fe(II) recycling is essential for achieving efficient ATZ degradation using CoNiFe-LDH on PMS.

3.6. Pathways

In this study, ATZ degradation intermediates were qualitatively analyzed using liquid phase mass spectrometry, and confirmed the degradation products.

The degradation of ATZ involves three processes: dealkylation, alkyl oxidation, and dechlorination [63]. SO₄^•− is an electrophilic oxidizing free radical that preferentially targets electron-rich sites on pollutant molecules. In this study, SO₄^•− attacks the alkyl-amine side chain of ATZ, cleaving the C–N bonds in the isopropyl amine and ethylamine groups to produce dealkylation products. The initial step in dealkylation involves the formation of carbon-centered radicals. SO₄^•− and •OH extract H from the methyl group adjacent to the N in ATZ, creating carbon-centered radicals. These radicals are subsequently oxidized by dissolved oxygen to form peroxide radicals, which undergo further hydrolysis to yield dealkylation products. H-abstraction by SO₄^•−/•OH also leads to the formation of alkyl oxidation products. In addition, Niu et al. [64] showed that the C−Cl bond in the ATZ molecule has a relatively low polarity and is more easily broken than other bonds. Therefore, SO₄^•−/•OH attack initiates the dechlorination hydroxylation reaction to generate intermediate products.

Based on the nine detected intermediates and their inferred mechanisms, two possible ATZ degradation pathways in the CoNiFe-LDH/PMS system were proposed, as shown in Figure 15. According to Hooper et al. [65], if P4 and P5 are the dealkylation products of ATZ, i.e., DEA and DIA, they are less toxic than ATZ, and the hydroxylation products generally do not have toxic effects on aquatic organisms. Therefore, the hazardous effects of atrazine degraded by the CoNiFe-LDH/PMS system will be significantly reduced.

4. Conclusions

In this study, CoNiFe-LDH was prepared through the hydrothermal method as a PMS activator for ATZ degradation in solution. Characterization revealed its multilayer structure with a large surface area and active sites. After adding three metal elements, more •OH were generated on the material surface than after adding two metal elements. The increased number of •OH accelerated the reduction–oxidation cycle among the metals in the system, thus improving the performance of the activator.

The CoNiFe-LDH/PMS system effectively degraded ATZ. When the initial concentration of ATZ was 10 mg·L⁻¹, the PMS and activator dosages were 0.6 mM and 80 mg·L⁻¹, respectively, and the pH was 7.0, 99% or more of the ATZ could be degraded within 60 min.

Cyclic experiments confirmed the excellent reusability and structural stability of the prepared CoNiFe-LDH/PMS system. The degradation rate was 75.9% after three cycles and the leaching of Co, Ni, and Fe remained low at 89, 66, and 154 μg·L⁻¹, respectively. The low removal rate might be explained by the occupation of the activator surface by ATZ and its intermediate products, which decreased the number of active sites or dissolution of metal ions.

The four free radicals SO₄^•−, •OH, ¹O₂, and O₂^•− all play key roles in the ATZ degradation process. Nine degradation intermediates were detected, from which the possible degradation path of ATZ was inferred. Through XPS characterization, the activation mechanism of CoNiFe-LDH was confirmed along with the reduction–oxidation cycle among the three metal ions.