Kinetic and mechanistic insights into the degradation of clofibric acid in saline wastewater by Co²⁺/PMS process: a modeling and theoretical study

Abstract

Recently, the degradation of non-chlorinated organic pollutants in saline pharmaceutical wastewater by SO₄˙⁻-based advanced oxidation processes (AOPs) has received widespread attention. However, little is known about the oxidation of chlorinated compounds in SO₄˙⁻-based AOPs. This study chose clofibric acid (CA) as a chlorinated pollutant model; the oxidation kinetics and mechanistic pathway were explored in the Co²⁺/peroxymonosulfate (PMS) system. Notably, a high removal efficiency (81.0%) but low mineralization rate (9.15%) of CA within 120 min were observed at pH 3.0 during Co²⁺/PMS treatment. Exogenic Cl⁻ had a dual effect (inhibitory then promoting) on CA degradation. Several undesirable chlorinated by-products were formed in the Co²⁺/PMS system. This demonstrated endogenic chlorine and exogenic Cl⁻ both reacted with SO₄˙⁻ to generate chlorine radicals, which participated in the dechlorination and rechlorination of CA and its by-products. Furthermore, SO₄˙⁻ was the dominant species responsible for CA degradation at low Cl⁻ concentrations (≤1 mM), whereas Cl₂˙⁻ was the predominant radical at [Cl⁻]₀ > 1 mM. A possible degradation pathway of CA was proposed. Our findings suggested that chlorinated compounds in highly saline pharmaceutical wastewater will be more resistant and deserve more attention.

---

1 Introduction

It is well known that pharmaceutical industries produce a large amount of toxic and hazardous wastewater, characterized by high biological oxygen demand (BOD), chemical oxygen demand (COD), salinity, and toxicity.¹ Pharmaceutical wastewater has severe adverse impacts on aquatic environments and requires pretreatment before discharge.^2,3 Clofibric acid (CA), a stable pharmaceutical contaminant with potential ecosystem and human health risks, is persistent and has been commonly detected in aquatic environments.^4,5 Due to its complex structure, CA removal was limited to traditional biological degradation.⁶ Currently, advanced oxidation processes (AOPs) show stronger oxidation abilities to refractory contaminants and have been extensively applied in CA degradation, such as UV/O₃,⁷ UV/TiO₂,⁸ and UV/O₃/peroxydisulfate (PDS).⁹ Among these methods, SO₄˙⁻-based AOPs have garnered substantial scientific interest.⁹

Typically, SO₄˙⁻ is extremely reactive towards a wide range of organic compounds with redox potentials between 2.5–3.1 V. Furthermore, SO₄˙⁻ tends to react with organic compounds via one-electron transfer rather than addition or H abstraction, which are less influenced by water matrix components.⁹ Peroxymonosulfate (PMS) is an efficient precursor of SO₄˙⁻ and activated by transition metals,¹⁰ UV radiation,¹¹ heat,¹² ultrasound,¹³ and others. Among those methods, Co²⁺ is the most efficient transition-metal ion for PMS activation, leading to radical-induced contaminant oxidation.¹⁴ Co²⁺/PMS has been universally recognized as a SO₄˙⁻-based system via one-electron oxidation.¹⁵ However, recent studies indicated a two-electron transfer pathway involving a non-radical route (i.e., Co⁴⁺) occurred at pH = 3.0 in the Co²⁺/PMS system.^16,17 To date, the main reactive oxygen species (ROS), involving radical and non-radical pathways, and their contributions to CA degradation affected by water constituents (i.e., inorganic ions) in the Co²⁺/PMS system were overlooked.

Chloride ion (Cl⁻) is ubiquitous in industrial wastewater, whereas the impacts of exogenic Cl⁻ (inorganic form from a water matrix) in oxidations of different chlorinated organic compounds in the Co²⁺/PMS system are still controversial due to the complicated reaction process. For example, Anipsitakis et al.¹⁸ reported a facilitating effect of low dosage Cl⁻ on the degradation of 2,4-dichlorophenol in the Co²⁺/PMS system. Chan and Chu et al.¹⁹ found the presence of Cl⁻ (0–6.8 mM) significantly restrained the removal efficiency and reduced the reaction rate of atrazine. A dual impact (inhibitory then promoting) of Cl⁻ (0–500 mM) was reported in the oxidative degradation of 4-chloro-2-nitrophenol by Co²⁺/PMS.²⁰ Hence, the concentration of exogenic Cl⁻, as well as substrate properties might be significant in chlorinated organic compound oxidations in AOPs. Currently, some groups have investigated the influence of exogenic Cl⁻ on the degradation efficiency of CA in AOPs.^9,21 Qin et al.⁹ pointed out that degradation of CA in the UV/O₃/PDS system dropped slightly as the Cl⁻ concentration increased from 0 to 10 mM. Tang et al.²¹ reported that a Cl⁻ concentration increase (0–2 mM) promoted CA degradation in the UV/Cl⁻ system. However, the effect of exogenic Cl⁻ on CA degradation in the Co²⁺/PMS system was not thoroughly examined, especially under highly saline conditions.

In addition to the effect of exogenic Cl⁻ ions in highly saline wastewater on CA removal, the significance of endogenic Cl⁻ (organic form from the contaminant itself) on the degradation process of CA was also neglected. Endogenic Cl⁻ from chlorinated organic pollutants played an important role in dehalogenation and halogenation processes. Dechlorination was observed during chlorobenzene oxidation with SO₄˙⁻.^22,23 The chlorine atoms released from the benzene ring participated in de novo formation of chlorinated by-products by SO₄˙⁻,²⁴ which led to the formation of polychlorinated aromatics during 2,4,6-trichlorophenol (TCP) oxidation.²⁵ However, previous studies have mainly focused on the oxidation efficiency and degradation pathways of CA in AOPs,^7–9 and little attention has been paid to the dechlorination and rechlorination of endogenic Cl⁻ during CA oxidation.

This study aims to: (1) evaluate the influence of exogenic Cl⁻ on the degradation kinetics of CA; (2) explore the transformation of endogenic chlorine of CA during Co²⁺/PMS treatment; (3) investigate the conversion of reactive oxygen species and their contributions to CA degradation; (4) explore the oxidation mechanisms and propose a possible degradation pathway of CA. This work hopes to provide valuable information for studying the role of exogenic Cl⁻ in the oxidation of chlorinated compounds in pharmaceutical wastewater.

2 Experimental

2.1 Materials

Clofibric acid (C₁₀H₁₁ClO₃, >99%), acid orange 7 (AO7, C₁₆H₁₁N₂O₄SNa, >98%), and carbamazepine (CBZ, C₁₅H₁₂N₂O, 98%) were obtained from Acros Organics. Oxone® ([2KHSO₅·KHSO₄·K₂SO₄] salt, 95%), L-histidine and hexamethyldisilazane, pyridine, and chlorotrimethylsilane were obtained from Sigma-Aldrich. HPLC grade acetonitrile and methanol were supplied by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Methyl phenyl sulfoxide (PMSO) and methyl phenyl sulfone (PMSO₂) were obtained from Shanghai Rhawn Chemical Technology Co., Ltd. CoSO₄·7H₂O, NaNO₂, NaCl, NaOH, and H₂SO₄ were of analytical grade, and used without further purification. Milli-Q water (Millipore, France, >18.2 MΩ cm) was used to prepare all reagents and solvents.

2.2 Experimental procedure

All reactions were performed in a 50 mL glass beaker and magnetically stirred (300 rpm) at room temperature (25 ± 2 °C). Reactions were initiated by mixing a specific concentration of PMS into a solution containing CA, CoSO₄·7H₂O, or other chemicals without pH control. While the effect of pH on the removal efficiency of CA was also studied, NaOH or H₂SO₄ adjusted the pH of the mixed solution to the desired level. Samples were taken at predetermined times for analysis and quenched with methanol (for HPLC) or NaNO₂ (for GC-MS and TOC). All tests were conducted in duplicate to ensure reproducibility and the data are shown as averages with error bars (±5%).

2.3 Analytical methods

Residual CA and CBZ concentrations were measured by a Waters e2695 with 2489 UV/Vis detector using an Atlantis® T3 Waters column (4.6 mm × 250 mm, 5 μm) at 227 and 210 nm, respectively. The mobile phase for CA detection contained 0.5% phosphoric acid/methanol (15/85 (v/v)) at a flow rate of 1.0 mL min⁻¹. The sample injection volume was 20 μL. AO7 concentrations were measured by a Hitachi F-4600 spectrophotometer (Hitachi, Japan) at 485 nm. PMS concentrations were detected spectrophotometrically by iodometric titration at 352 nm.²⁶ A Shimadzu TOC analyzer (TOC-VCPH, Shimadzu, Japan) measured the total organic carbon (TOC). The intermediates of CA were analyzed by GC-MS, (Agilent 8860-GC/5977B-MSD, USA) and the detailed operational parameters are provided in Text S1.†

The oxidation of clofibric acid under various conditions was modeled using pseudo-first-order kinetics eqn (1)

C/C₀ = e^−kt (1)

where C₀ and C are the concentrations of CA at t = 0 and t min, respectively, k is the pseudo-first-order rate constant (min⁻¹), and t is the reaction time.

2.4 Kinetic modeling

Kinetic modeling of the Co²⁺/PMS system was performed by Kintecus 6.51 (https://www.kintecus.com). All radical reactions used as input for kinetic modeling were listed in Table S1.† The formations and transformations of reactive species such as the hydroxyl, sulfate, and chlorine radicals were simulated according to the reactions in Table S1.† Most reaction rate constants were obtained from previous studies,^27–29 while some unknown reaction rate constants were calculated by fitting the traces of CA degradation curves by taking these unknown rate constants as adjustable. Due to negligible production at pH 3.0, reactions of chlorite/chlorate were not included in our model.³⁰ Unknown or poorly-defined parameters were optimized to accurately simulate species distributions by using a numerical procedure associated with Kintecus programs.

3 Results and discussion

3.1 CA degradation

3.1.1 Effect of Co²⁺ concentration. Co²⁺ played an important role as a catalyst in the Co²⁺/PMS system. The impacts of catalyst levels on CA degradation were studied and the degradation kinetics and impacts on the rate constant k are shown in Fig. S1a† and 1a, respectively. No CA oxidation was observed without Co²⁺ addition (Fig. S1a†), which suggested CA was not oxidized by PMS itself. Furthermore, CA degradation rates increased significantly with Co²⁺ concentration increased (Fig. 1a). The reaction rate constant k increased from 0 to 3.36 × 10⁻² min⁻¹ as Co²⁺ increased from 0 to 0.3 mM. Dramatic degradation rates and oxidation efficiencies were detected at [Co²⁺]₀ < 0.02 mM, whereas no obvious enhancement was obtained when the Co²⁺ concentration increased from 0.02 to 0.3 mM. This may be due to Co²⁺ inducing rapid PMS decomposition with the highest efficiency to generate SO₄˙⁻ (eqn (2)). The regeneration of Co²⁺ with excess PMS (E⁰(Co³⁺/Co²⁺) = 1.45 V) may be a vital step to continue catalysis at low Co²⁺ levels (≤0.02 mM) (eqn (3)).²⁸ Moreover, the non-radical Co⁴⁺ could also exist in the Co²⁺/PMS system at pH 3.0 (as seen in eqn (4)) and more Co⁴⁺ may form at higher Co²⁺ dosages.¹⁶ Co⁴⁺ could react with the target pollutant to regenerate Co²⁺ via eqn (5) and participate in the next oxidation cycle. Notably, the maximum degradation rate appeared when Co²⁺ concentration increased to 0.3 mM, revealing that 0.5 mM PMS was not completely consumed and continued to produce radicals. Xue et al. also reported a similar phenomenon during thiamphenicol (TAP) and florfenicol (FFC) in the Co²⁺/PMS system.³¹ In this study, the optimal dosage of Co²⁺ was 0.01 mM with an appropriate degradation efficiency (81.0%) and a high rate constant k (3.21 × 10⁻² min⁻¹).

HSO₅⁻ + Co²⁺ → Co³⁺ + OH⁻ + SO₄˙⁻ (2)

HSO₅⁻ + Co³⁺ → Co²⁺ + H⁺ + SO₅˙⁻ (3)

HSO₅⁻ + Co²⁺ → Co^IVO²⁺ + H⁺ + SO₄²⁻, k = 2.05 M⁻¹ s⁻¹ (4)

Co⁴⁺ + pollutant → Co²⁺ + products (5)

Fig. 1 Effect of initial (a) Co²⁺ concentration, (b) PMS concentration, (c) pH, and (d) Cl⁻ concentration on the rate constants k of CA. Experimental conditions: [CA]₀ = 0.1 mM, (a, b, d) pH = 3.0; (a, c, d) [PMS]₀ = 0.5 mM; (b, c, d) [Co²⁺]₀ = 0.01 mM.

3.1.2 Effect of PMS concentration. The impact of initial PMS concentration on CA degradation in the Co²⁺/PMS system was studied. As illustrated in Fig. 1b, at PMS concentrations between 0–10 mM, the rate constant of CA degradation increased from 0 to 6.56 × 10⁻² min⁻¹. Higher levels of PMS accelerated the formation of reactive oxygen species, and the degradation efficiency of CA was proportional to the oxidant dosages under the same catalyst dosage (eqn (2) or eqn (4)). The rate constant increased rapidly when PMS was 0–1 mM, while only a slight improvement was observed at PMS concentrations above 1 mM. Wang et al.¹⁰ revealed excessive PMS acted as a radical scavenger, which accelerated the conversion of SO₄˙⁻ to less reactive SO₅˙⁻ (eqn (6)). Increasing the PMS concentration to 10 mM, the low concentration of Co²⁺ limited ROS production, which led to the slight degradation rate enhancement. Notably, the degradation efficiency of CA increased gradually when [PMS]₀ < 0.5 mM, whereas no evident depletion of CA was obtained at PMS concentrations between 0.5–10 mM (Fig. S1b†). Therefore, 0.5 mM was selected as the optimal PMS concentration.

HSO₅˙⁻ + SO₄˙⁻ → SO₅˙⁻ + HSO₄˙⁻ (6)

3.1.3 Effect of pH. The solution pH played a vital role in organic degradation as well as ROS formation in the SO₄˙⁻-based AOP system.^31,32 Notably, the CA concentration was not detected by HPLC under neutral or alkaline conditions (Fig. S2†). Due to the strong oxidation capability of PMS under alkaline conditions,²⁸ the impact of initial pH 3.0 to 6.0 was examined. As shown in Fig. 1c, the rate constants at pH levels of 3.0, 4.0, 5.0, and 6.0 were 1.91 × 10⁻², 2.13 × 10⁻², 2.27 × 10⁻², and 2.12 × 10⁻² min⁻¹, respectively. Fig. 1c indicated the rate constant increased slightly at pH levels between 3.0–5.0 in the Co²⁺/PMS system. It is worth noting that HSO₅⁻ and Co²⁺ were the dominant forms of PMS and CoSO₄·7H₂O at pH < 6.0.¹⁰ Additionally, higher pH levels assisted the formation of CoOH⁺ via Co²⁺ hydrolysis (eqn (7)) and was the predominant species responsible for PMS decomposition (eqn (8)).³³ The rate constant dropped slightly as the pH increased from 5.0 to 6.0, possibly due to HPLC measurement errors as the pH approached neutral or alkaline conditions (Fig. S2†). A similar phenomenon was obtained for the oxidation of TCP²⁵ and FFC.³¹ Although the rate constant optimized at pH = 5.0, the degradation efficiency (78.0%) decreased slightly after 120 min as compared to pH 3.0 (81.0%). Thus, an initial pH = 3.0 without adjustment was applied in the following experiments.

Co²⁺ + H₂O → CoOH⁺ + H⁺ (7)

CoOH⁺ + HSO₅⁻ → CoO + H₂O + SO₄˙⁻ (8)

3.1.4 Effect of Cl⁻ concentration. Pharmaceutical wastewater usually contains many salts, which may influence the oxidation efficiency of pharmaceutical contaminants. Thus, the impact of chloride ions on CA degradation was investigated. As illustrated in Fig. 1d, Cl⁻ had a dual effect (inhibitory then promoting) on the CA degradation. Low Cl⁻ concentrations (≤1 mM) had a negative impact on CA oxidation, but facilitated oxidation when [Cl⁻]₀ > 1 mM (Fig. S3a†). The chloride ion was supposed to scavenge SO₄˙⁻ to generate less reactive chlorine radicals (Cl˙/Cl₂˙⁻), and lead to the inhibitory effect of Cl⁻ on CA degradation.³⁴ Further addition of chloride and formation of chlorine species (HOCl/Cl₂) shown in eqn (9) and (10) may facilitate rapid CA degradation.³⁵ For comparison, removal of AO7 and CBZ under the same conditions was also studied. As shown in Fig. S4,† a similar phenomenon (inhibitory then promoting) was observed and critical points also appeared at low Cl⁻ concentrations. Moreover, the same dual effect of Cl⁻ was observed for 4-chloro-2-nitrophenol removal in saline wastewater by Co²⁺/PMS,²⁰ whereas an opposite dual effect (promoting then inhibitory) was obtained on chlorinated azo dye oxidation in the UV/PS system.³⁶ However, CA degradation efficiency gradually increased with added Cl⁻ in the PMS/Cl⁻ system (Fig. S3b†). The removal rate of CA was 75.1% after 120 min in the presence of 100 mM Cl⁻ by PMS/Cl⁻ but reached 85.4% when Co²⁺ and Cl⁻ coexisted. According to previous studies, PMS can oxidize Cl⁻ directly to generate reactive chlorine species (HClO/Cl₂) to degrade organic compounds.³⁵ However, this dual effect of Cl⁻ was not present in H₂O₂-based AOPs, such as Fenton and photo-Fenton systems,³⁷ because Cl⁻ cannot activate H₂O₂ to oxidize the contaminants.

Cl⁻ + HSO₅⁻ → SO₄²⁻ + HOCl (9)

2Cl⁻ + HSO₅⁻ + H⁺ → SO₄²⁻ + Cl₂ + H₂O (10)

3.2 Mineralization

The mineralization of CA was evaluated by measuring TOC removal efficiency. As depicted in Fig. 2, the mineralization rate increased with prolonged reaction times, while TOC removal in the Co²⁺/PMS system was much higher than in the absence of Co²⁺. TOC removal was 3.09% and 9.15% with 0 mM Cl⁻ in the Co²⁺/PMS system after 30 and 120 min, respectively. However, the mineralization degree of CA was 2.24%, and 1.69% after 120 min at 1 mM, and 100 mM Cl⁻ in the Co²⁺/PMS system, respectively. TOC removal decreased with additional Cl⁻ in the Co²⁺/PMS system. Interestingly, the degradation efficiency of CA improved at high Cl⁻ (300 mM) in the PMS/Cl⁻ system (Fig. S3b†) but the TOC removal rate was lower than 0.2%. In general, the mineralization degree of CA was quite low in the Co²⁺/PMS system. The high degradation efficiency but low mineralization degree of CA suggested the formation of refractory intermediates in a saline environment, especially chlorinated by-products.³⁸

Fig. 2 Mineralization of CA degradation with different Cl⁻ concentrations at different reaction times in the Co²⁺/PMS/Cl⁻ and PMS/Cl⁻ system, respectively. Experimental conditions: [CA]₀ = 0.1 mM, [Co²⁺]₀ = 0.01 mM, [PMS]₀ = 0.5 mM, pH 3.0.

3.3 Free radical identification

SO₄˙⁻ and ˙OH are regarded as the primary radicals yielded in PMS-based systems.^34,39 Hence, quenching experiments were conducted to detect the major ROS during the oxidation of organic pollutants in AOPs. Both ˙OH and SO₄˙⁻ were scavenged by methanol (MeOH) with the rate constants of 9.7 × 10⁸ M⁻¹ s⁻¹ and 2.5 × 10⁷ M⁻¹ s⁻¹, respectively. tert-Butyl alcohol (TBA) reacts with ˙OH (k_{TBA, ˙OH} = (3.8–7.6) × 10⁸ M⁻¹ s⁻¹), but not SO₄˙⁻ .^40,41 In the absence of Cl⁻, as shown in Fig. 3a, 20 mM TBA had almost no influence on CA degradation, while the degradation rate of CA decreased from 81.0% to 66.0% by adding 1 M TBA. In contrast, MeOH inhibited CA degradation. Adding 20 mM MeOH reduced the degradation rate of CA from 81.0% to 34.8%, and CA removal stopped completely when 1 M MeOH was added. This result indicated that SO₄˙⁻ was the primary oxidant in the Co²⁺/PMS system without Cl⁻ addition. In the presence of Cl⁻, as shown in Fig. 3b and c, there was no obvious inhibitory effect on CA degradation in the Co²⁺/PMS system with 20 mM TBA, while CA degradation decreased slightly as TBA increased to 1 M. This indicated ˙OH was not the primary radical because it is difficult to form in an acidic solution (eqn (11)).³⁴ Besides, partial inhibitory on CA removal by adding 20 mM MeOH and complete inhibitory with 1 M MeOH were observed (see Fig. 3b), which suggested SO₄˙⁻ was the primary oxidant in the Co²⁺/PMS system at low Cl⁻ concentration (1 mM). However, in the presence of 100 mM Cl⁻ (shown in Fig. 3c), only partially restrained on CA removal was detected by adding 1 M MeOH. That revealed SO₄˙⁻ was not the dominant oxidant in the presence of 100 mM Cl⁻. Therefore, SO₄˙⁻ was the primary oxidant in the Co²⁺/PMS system at low Cl⁻ levels (≤1 mM), whereas other reactive oxygen species might be predominant at high Cl⁻ concentrations.

SO₄˙⁻ + HO⁻ → ˙OH + SO₄²⁻, k = (6.5 ± 1 × 10⁷) M⁻¹ s⁻¹ (11)

Fig. 3 Effect of radical scavengers on CA degradation in the Co²⁺/PMS system. Experimental conditions: [CA]₀ = 0.1 mM, [Co²⁺]₀ = 0.01 mM, [PMS]₀ = 0.5 mM, pH 3.0, (a) 0 mM Cl⁻; (b) 1 mM Cl⁻; (c) 100 mM Cl⁻.

3.4 Non-radical identification

The low solubility of Co³⁺ (k_sp = 1.6 × 10⁻⁴⁴) limits its presence in solutions at pH 3.0.¹⁷ Hence, although Co³⁺ could exist in the Co²⁺/PMS system, it contributes very little to CA oxidation.

Recent studies confirmed that high-valent cobalt-oxo [Co⁴⁺] species played a significant role in the oxidation of organic pollutants by Co²⁺/PMS.^16,17 PMSO reacts readily with Co⁴⁺ to produce PMSO₂, though SO₄˙⁻ and ˙OH cannot oxidize PMSO to PMSO₂.¹⁷ Hence, the consumption of PMSO and formation of PMSO₂ were conducted to identify the presence of Co⁴⁺. As shown in Fig. 4a, PMSO directly oxidizes to PMSO₂ with η-(PMSO₂) (represents the conversion ratio of PMSO to PMSO₂) approaching 100% by PMS alone. For the PMS/Cl⁻ system, PMSO oxidation increased as the Cl⁻ increased from 1 to 100 mM, but formation of PMSO₂ decreased in the co-presence of PMS and Cl⁻. For example, the η-(PMSO₂) was 93.2% at 1 mM Cl⁻, however, it was only 54.9% at [Cl⁻]₀ = 100 mM. Some reactive chlorine may have formed when Cl⁻ was 100 mM and oxidized PMSO to produce non-PMSO₂ products that lowered the PMSO₂ formation rate. However, the loss rate of PMSO accelerated in the Co²⁺/PMS/Cl⁻ system as compared to the PMS/Cl⁻ system (Fig. 4b). Additional consumption of PMSO after Co²⁺ addition suggested that other oxidants (e.g., Co⁴⁺, SO₄˙⁻ or ˙OH) might transform PMSO. At the same time, the formation of PMSO₂ was higher in the presence of Co²⁺ (Fig. 4b), which indicated formation of Co⁴⁺ and rapid oxidation of PMSO to PMSO₂ (rate constant, 2.0 × 10⁶ M⁻¹ s⁻¹). Additional consumption of PMSO was much higher than PMSO₂ formation when Co²⁺ was involved (Fig. S5†), which confirmed that Co⁴⁺ was not the only reactive oxygen species responsible for PMSO oxidation. Moreover, Liu et al.¹⁷ reported Co⁴⁺, SO₄˙⁻, and ˙OH existed in solution when the PMS/Co²⁺ ratio >10. The PMS/Co²⁺ ratio in this study was 10 and our results agreed with those of Liu et al.¹⁷

Fig. 4 (a) η(PMSO₂), (b) PMSO loss and PMSO₂ corresponding production in Co²⁺/PMS at different Cl⁻ concentrations. Experimental conditions: [PMS]₀ = 0.5 mM, [PMSO]₀ = 0.5 mM, pH 3.0.

As depicted in Fig. S6,† the effect of PMSO on CA degradation was investigated. CA degradation dropped significantly in the Co²⁺/PMS system in the presence of both PMSO and Cl⁻. This confirmed that PMSO may act as a scavenging agent for high-valent cobalt, PMS, and free radicals, which led to the inhibitory degradation of CA in the Co²⁺/PMS system. For the PMS/Cl⁻ system, the degradation efficiency of CA reached 75.1% in 100 mM Cl⁻, while complete inhibition of CA degradation with added PMSO. Chlorine species (HOCl/Cl₂) generated by the reaction of Cl⁻ and PMS could preferentially oxidize PMSO rather than CA.

3.5 Kinetic modeling

These results revealed that multiple ROS, including SO₄˙⁻, ˙OH, and Co⁴⁺, were produced in the Co²⁺/PMS system. Kinetic modeling helped understand the various reactive species and their contributions to the overall degradation (see equations Table S1†). Fig. S7† shows a typical kinetic plot with a fitting degree above 98.0%.

As depicted in Fig. 5a–c, three main reactive species (SO₄˙⁻, ˙OH, and Co⁴⁺) may oxidize CA during Co²⁺/PMS in the absence of Cl⁻. The concentration of SO₄˙⁻ (∼10⁻⁶ M) was approximately nine and four times higher than ˙OH, and Co⁴⁺, respectively. Thus, the contribution of SO₄˙⁻ to CA oxidation reached 99.96%, while the contribution of Co⁴⁺ (∼0.03%) and ˙OH (∼0.01%) were negligible (Fig. 6). This agreed with quenching experiments and further revealed that SO₄˙⁻ plays a primary role in the Co²⁺/PMS system without Cl⁻ addition.

Fig. 5 The estimated concentration of primary reactive species at different Cl⁻ concentrations in the Co²⁺/PMS/Cl⁻ system. Experimental conditions: [CA]₀ = 0.1 mM, [Co²⁺]₀ = 0.01 mM, [PMS]₀ = 0.5 mM, pH 3.0.

Fig. 6 The contribution of primary reactive species to CA degradation at different Cl⁻ concentrations in the Co²⁺/PMS/Cl⁻ system. Experimental conditions: [CA]₀ = 0.1 mM, [Co²⁺]₀ = 0.01 mM, [PMS]₀ = 0.5 mM, pH 3.0.

A rapid adverse impact of Cl⁻ on the yield of SO₄˙⁻ should be related to the quenching reaction of SO₄˙⁻ by Cl⁻ with a high rate constant of 2.7 × 10⁸ M⁻¹ s⁻¹ (eqn (12)). As a result, the contribution of SO₄˙⁻ dropped significantly from 99.96% to 0.28% as Cl⁻ levels increased from 0 to 300 mM. Notably, Cl⁻ facilitated the formation of ˙OH (∼10⁻¹³ M) at low Cl⁻ concentrations (≤1 mM), while the ˙OH levels decreased with additional Cl⁻ (Fig. 5). Overall, the contribution of ˙OH was small regardless of the amount of Cl⁻ present and agreed with quenching test results. It has been reported that ˙OH level was relatively low under acidic conditions.^35,42 However, the Co⁴⁺ concentration was relatively stable (∼10⁻¹⁰ M) as the initial Cl⁻ level increased from 0 to 300 mM. Consequently, the contribution of Co⁴⁺ was less than 0.04%. This was likely due to its low reactivity with CA (k = 2.82 × 10⁻² M⁻¹ s⁻¹) and the transformation of Co⁴⁺ to SO₄˙⁻/˙OH when the molar ratio of PMS/Co²⁺ exceeded 10.

SO₄˙⁻ + Cl⁻ → SO₄²⁻ + Cl˙, k_f = 2.7 × 10⁸ M⁻¹ s⁻¹ (12)

Cl˙ + Cl⁻ → Cl₂˙⁻, k = 6.5 × 10⁹ M⁻¹ s⁻¹ (13)

Cl₂˙⁻ + Cl₂˙⁻ → Cl₂ + 2Cl⁻ (14)

Interestingly, the Cl₂˙⁻ concentration was overwhelming (∼10⁻⁵ M) among all chlorine reactive species and significantly increased with Cl⁻. As a result, the contribution of Cl₂˙⁻ to CA degradation reached 99.44% with 300 mM Cl⁻. The absolute predominance of Cl₂˙⁻ was due to two things—the high rate constant of the reaction between CA and Cl₂˙⁻ (k = 1.41 × 10⁸ M⁻¹ s⁻¹) played a significant role in CA oxidation. Secondly, the transformation of Cl˙ to Cl₂˙⁻ with a high rate constant of 6.5 × 10⁹ M⁻¹ s⁻¹ (eqn (13)) relates to the high concentration of Cl₂˙⁻. Furthermore, generation of Cl˙ increased (0–1 mM) and then decreased (1–300 mM) with added Cl⁻. In the presence of 1 mM Cl⁻, Cl˙ accounted for 23.16% of CA oxidation due to its high CA reactivity (k = 5.5 × 10⁹ M⁻¹ s⁻¹) and high production (∼10⁻⁷ M). Cl₂ concentration had a similar increase as Cl₂˙⁻, due to the conversion of the Cl₂˙⁻ to Cl₂ as shown in eqn (14). The other three reactive species, (HOCl, ClO˙, and ClOH˙⁻) at relatively low concentrations, initially increased, then decreased with added Cl⁻ (Fig. S8†). Consequently, their contributions to CA were relatively low and were neglected. Cl₂˙⁻ should be the dominant reactive species for CA oxidation at [Cl⁻]₀ > 1 mM, while SO₄˙⁻ played a critical role at low Cl⁻ concentrations (≤1 mM) in the Co²⁺/PMS system.

3.6 Mechanism and pathways

GC-MS analyzed the CA intermediates. The primary chlorinated by-products at different Cl⁻ concentrations (i.e., 0, 1, and 100 mM) at 30 and 120 min are shown in Table S2.† The chlorinated by-products increased with added Cl⁻ or reaction times. For example, 11 kinds of chlorinated by-products were detected in the Co²⁺/PMS system without Cl⁻ addition, whereas 22 and 23 refractory chlorinated by-products at 1 mM and 100 mM Cl⁻ were detected, respectively (Table S2†). Based on the identified products and previous studies,^9,43 possible mechanisms and a CA degradation pathway were proposed (Fig. 7).

Fig. 7 A possible CA oxidation pathway in the Co²⁺/PMS system. Experimental conditions: [CA]₀ = 0.1 mM, [Co²⁺]₀ = 0.01 mM, [PMS]₀ = 0.5 mM, pH 3.0.

In the absence of Cl⁻, SO₄˙⁻ was the main reactive species in CA degradation. Thus, CA oxidation could proceed by three different pathways, including C1–O bond breaking, dechlorination, and attack on the C7 position of CA. Firstly, cleavage of the C1–O bond formed 4-chlorophenol (P1). Dechlorination of 4-chlorophenol attacked by SO₄˙⁻ may generate phenol (not detected, which implied in the present study) and the release of chloride ions (shown in Scheme 1a).⁴⁴ Interestingly, chloride ions undergo oxidization by SO₄˙⁻ to produce chlorine radicals that form chlorinated aromatic by-products (P2 and P3) by nucleophilic addition. Peng et al.³⁶ also indicated chlorine atoms of chlorinated azo dyes might be partially released as inorganic ions, which undergo subsequent rechlorination. Alternatively, SO₄˙⁻ initially reacts with CA, which led to CA dechlorination, followed by oxidization of the intermediate by SO₄˙⁻ to generate chlorinated by-products (P4). In the last way, the C7 position of CA was attacked by SO₄˙⁻ followed by demethylation and decarboxylation intermediates, such as 2-(4-chloro-phenoxy)-propionic acid (not detected) and acetic acid-4-chlorophenyl ester (P5), which was further transformed to chlorinated by-products (P6 and P7) through chlorination, respectively. Furthermore, the aromatic ring was gradually cleaved to produce organic acid compounds such as butanedioic acid (P8), methacrylic acid (P9), and glycolic acid (P10). Finally, short-chain alcohol compounds (e.g., glycerol (P11) and ethylene glycol (P12)) formed and eventually oxidized into CO₂ and H₂O. The identification of polychlorinated by-products (P2, P3, and P6) during CA removal without exogenous Cl⁻ addition provided direct evidence for dechlorination and rechlorination of endogenic Cl⁻. Furthermore, the formation of various chlorinated by-products also explains the high degradation efficiency (81.0%) but low mineralization rate (9.15%) within 120 min during CA removal in the absence of Cl⁻.

Scheme 1 The possible pathway of (a) dechlorination and (b) chlorination on CA degradation intermediates.

In the presence of Cl⁻, SO₄˙⁻ was scavenged by Cl⁻ to produce less reactive chlorine radicals (Cl˙/Cl₂˙⁻), which resulted in the reduction of CA oxidation. The inhibition of Cl⁻ is vital at low chloride levels (≤1 mM). Additional Cl⁻ dramatically enhanced CA oxidation, whereas the overall degradation rate was still lower than in the absence of Cl⁻. Hence, Cl₂˙⁻ was the dominant chlorine species contributing to CA degradation under chloride-rich conditions. More refractory polychlorinated by-products (chlorine atom number ≥ 2) were produced with additional Cl⁻, such as 3,4-dichlorophenol, 2,3-dichlorophenol, and 2,5-dichlorophenol. Thus, the mineralization rate dropped as the Cl⁻ concentration increased. The oxidation pathways of CA involving Cl⁻ might include C1–O bond cleavage, chlorination, dechlorination, and rechloridation process. Notably, 2,4,6-trichlorophenol was generated as the chlorinated by-product of CA in the presence of exogenous Cl⁻. It was assumed that chlorination phenol was initiated by Cl₂˙⁻ attack of the degradation product radicals, yielding 2,4,6-trichlorophenol (Scheme 1b). This result further proved that a higher exogenous Cl⁻ concentration produced additional chlorine radicals, and then oxidized CA to produce more chlorinated by-products, consistent with previous reports.^25,31

4 Conclusions

The degradation kinetics and mechanism of CA were studied in the Co²⁺/PMS system. The reaction rates of CA increased with added Co²⁺ and PMS though varied slightly with initial pH. Furthermore, Cl⁻ had a dual effect (inhibitory then promoting) on CA degradation. Intermediate identification by GC-MS revealed that more refractory polychlorinated by-products (chlorine atom number ≥ 2) were produced in the Co²⁺/PMS/Cl⁻ system, which resulted in a decreased CA mineralization rate. According to quenching experiments and kinetic modeling, SO₄˙⁻ (45.7%) played the main role in CA degradation, followed by Cl₂˙⁻ (31.1%) and Cl˙ (23.2%) at low Cl⁻ levels (≤1 mM), whereas Cl₂˙⁻ was the predominant radical when [Cl⁻]₀ ˃ 1 mM. However, the contribution of Co⁴⁺ to CA degradation was relatively low regardless of added Cl⁻. A possible CA degradation pathway in saline wastewater was proposed, including cleavage of the C1–O bond, chlorination, dechlorination, and rechloridation process. Our findings indicated the introduction of exogenic Cl⁻ increased the difficulty in oxidative degradation of CA. Hence, it is imperative to explore other effective technologies for CA treatment in saline wastewater.

Conflicts of interest

There are no conflicts to declare.

Author contributions

Jiale Wang: Conceptualization, Visualization, Writing-original draft, Data curation. Siyi Fan: Data curation, Resources, Data curation. Zhirui Xu: Resources, Validation. Jiaqi Gao: Investigation, Resources. Ying Huang: Conceptualization, Funding acquisition, Software, Writing-review & editing, Supervision. Xubiao Yu: Methodology, Funding acquisition, Supervision. Huihui Gan: Visualization, Investigation, Funding acquisition.

Acknowledgements

This work was supported by Natural Science Foundation of Ningbo (No. 202003N4135), the General Research Project of Zhejiang Provincial Department of Education (No. Y202043966), National Natural Science Foundation of China (No. 41977152, No. 41676104 and No. 52070103), Zhejiang Provincial Natural Science Foundation (No. LGF19E090006), the Commonweal Project of Ningbo Science and Technology Bureau (No. 2019C10024), the fund by Laboratory of Sustainable Urban Drainage of Ningbo University, and the K. C. Wong Magna Fund in Ningbo University.

Notes and references

1. D. Sreekanth, D. Sivaramakrishna, V. Himabindu and Y. Anjaneyulu, Bioresour. Technol., 2009, 100, 2534–2539.
2. K. K. Ng, X. Shi, S. L. Ong, C.-F. Lin and H. Y. Ng, Chem. Eng. J., 2016, 295, 317–325.
3. X. Shi, O. Lefebvre, K. K. Ng and H. Y. Ng, Bioresour. Technol., 2014, 153, 79–86.
4. M. Isidori, A. Nardelli, L. Pascarella, M. Rubino and A. Parrella, Environ. Int., 2007, 33, 635–641.
5. T. J. Runnalls, D. N. Hala and J. P. Sumpter, Aquat. Toxicol., 2007, 84, 111–118.
6. H. Lin, X. Tang, J. Wang, Q. Zeng, H. Chen, W. Ren, J. Sun and H. Zhang, J. Hazard. Mater., 2021, 405, 124204.
7. Y. Wang, H. Y. Li, P. Yi and H. Zhang, J. Hazard. Mater., 2019, 379, 120771.
8. T. E. Doll and F. H. Frimmelm, Water Res., 2004, 38(4), 955–964.
9. W. Qin, Z. Lin, H. Dong, X. Yuan, Z. Qiang, S. Liu and D. Xia, Water Res., 2020, 186, 116336.
10. X. Wang, Z. Wang, Y. Tang, D. Xiao, D. Zhang, Y. Huang, Y. Guo and J. Liu, Chem. Eng. J., 2019, 368, 999–1012.
11. G. Wen, X. Xu, H. Zhu, T. Huang and J. Ma, Chem. Eng. J., 2017, 328, 619–628.
12. Y.-Y. Ahn, J. Choi, M. Kim, M. S. Kim, D. Lee, W. H. Bang, E.-T. Yun, H. Lee, J.-H. Lee and C. Lee, Environ. Sci. Technol., 2021, 55, 5382–5392.
13. N. Li, S. Tang, Y. Rao, J. Qi, Q. Zhang and D. Yuan, Electrochim. Acta, 2019, 298, 59–69.
14. G. P. Anipsitakis and D. D. Dionysiou, Environ. Sci. Technol., 2004, 38, 3705–3712.
15. Z. Zhang and J. O. Edwards, Inorg. Chem., 1992, 31, 3514–3517.
16. Y. Zong, X. Guan, J. Xu, Y. Feng, Y. Mao, L. Xu, H. Chu and D. Wu, Environ. Sci. Technol., 2020, 54, 16231–16239.
17. B. Liu, W. Guo, H. Wang, S. Zheng, Q. Si, Q. Zhao, H. Luo and N. Ren, J. Hazard. Mater., 2021, 416, 125679.
18. G. P. Anipsitakis, D. D. Dionysiou and M. A. Gonzalez, Environ. Sci. Technol., 2006, 40, 1000–1007.
19. K. H. Chan and W. Chu, Water Res., 2009, 43, 2513–2521.
20. J. Zhou, J. Xiao, D. Xiao, Y. Guo, C. Fang, X. Lou, Z. Wang and J. Liu, Chemosphere, 2015, 134, 446–451.
21. Y. Tang, X. Shi, Y. Liu, L. Feng and L. Zhang, R. Soc. Open Sci., 2018, 5, 171372.
22. Y. Sun, J. Zhao, B.-T. Zhang, J. Li, Y. Shi and Y. Zhang, Chem. Eng. J., 2019, 368, 553–563.
23. L. Zhou, Y. Zhang, R. Ying, G. Wang, T. Long, J. Li and Y. Lin, Environ. Sci. Pollut. Res., 2017, 24, 11549–11558.
24. B. Sheng, F. Yang, Y. Huang, Z. Wang, R. Yuan, Y. Guo, X. Lou and J. Liu, Chem. Eng. J., 2020, 381, 122634.
25. L. Xu, R. Yuan, Y. Guo, D. Xiao, Y. Cao, Z. Wang and J. Liu, Chem. Eng. J., 2013, 217, 169–173.
26. C. Liang, C.-F. Huang, N. Mohanty and R. M. Kurakalva, Chemosphere, 2008, 73, 1540–1543.
27. J. E. Grebel, J. J. Pignatello and W. A. Mitch, Environ. Sci. Technol., 2010, 44, 6822–6828.
28. Y.-H. Guan, J. Ma, X.-C. Li, J.-Y. Fang and L.-W. Chen, Environ. Sci. Technol., 2011, 45, 9308–9314.
29. Y. Lei, S. Cheng, N. Luo, X. Yang and T. An, Environ. Sci. Technol., 2019, 53, 11170–11182.
30. M. Jaksić, B. Nikolić, I. Csonka and A. Djordjevic, J. Electrochem. Soc., 1969, 116, 684.
31. Y. Xue, Z. Wang, R. Bush, F. Yang, R. Yuan, J. Liu, N. Smith, M. Huang, R. Dharmarajan and P. Annamalai, Chem. Eng. J., 2021, 415, 129041.
32. Y. Huang, M. Jiang, S. Gao, W. Wang, Z. Liu and R. Yuan, Sep. Purif. Technol., 2022, 120921.
33. G. P. Anipsitakis and D. D. Dionysiou, Environ. Sci. Technol., 2003, 37, 4790–4797.
34. Z. Wang, R. Yuan, Y. Guo, L. Xu and J. Liu, J. Hazard. Mater., 2011, 190, 1083–1087.
35. R. Yuan, S. N. Ramjaun, Z. Wang and J. Liu, J. Hazard. Mater., 2011, 196, 173–179.
36. W. Peng, Y. Fu, L. Wang, Y. Wang, Y. Dong, Y. Huang and Z. Wang, Chin. Chem. Lett., 2021, 32, 2544–2550.
37. Z. Wang, M. Feng, C. Fang, Y. Huang, L. Ai, F. Yang, Y. Xue, W. Liu and J. Liu, RSC Adv., 2017, 7, 12318–12321.
38. C. Fang, X. Zhang, X. Lou, Y. Shi, Y. Tang, D. Huang, J. Liu and Y. Guo, Int. J. Environ. Anal. Chem., 2022, 1–15.
39. F. Yang, Y. Huang, C. Fang, Y. Xue, L. Ai, J. Liu and Z. Wang, Chemosphere, 2018, 199, 84–88.
40. C. Qi, X. Liu, Y. Li, C. Lin, J. Ma, X. Li and H. Zhang, J. Hazard. Mater., 2017, 328, 98–107.
41. J. Miao, J. Sunarso, C. Su, W. Zhou, S. Wang and Z. Shao, Sci. Rep., 2017, 7, 1–10.
42. C. Liang, Z.-S. Wang and C. J. Bruell, Chemosphere, 2007, 66, 106–113.
43. K. Zhu, X. Wang, M. Geng, D. Chen, H. Lin and H. Zhang, Chem. Eng. J., 2019, 374, 1253–1263.
44. C. Cai, X. Duan, X. Xie, S. Kang, C. Liao, J. Dong, Y. Liu, S. Xiang and D. D. Dionysiou, J. Hazard. Mater., 2021, 410, 124604.

---

Footnote

† Electronic supplementary information (ESI) available. See https://doi.org/10.1039/d2ra02673b

---