In situ organic Fenton-like catalysis triggered by anodic polymeric intermediates for electrochemical water purification

Significance

Organic Fenton catalysis is one promising water purification process, but it needs ex situ added organic activators and suffers from secondary pollution. Herein, we develop in situ organic Fenton-like catalysis free of secondary pollution by utilizing the redox-active polymeric intermediates generated from electrochemical pollutant degradation. We elucidate its mechanisms with electrochemical, photochemical, and spectrographic approaches. H₂O₂ activation mediated by nonhalogenated aromatic compounds can be regulated from nonradical to radical pathway by anodic potential to refine the catalytic properties in the organic Fenton-like catalysis. This work demonstrates the great potential of in situ organic Fenton-like reaction in green catalysis and provides a conceptual advance to design clean, robust, and cost-effective catalytic systems for water purification.

Abstract

Organic Fenton-like catalysis has been recently developed for water purification, but redox-active compounds have to be ex situ added as oxidant activators, causing secondary pollution problem. Electrochemical oxidation is widely used for pollutant degradation, but suffers from severe electrode fouling caused by high-resistance polymeric intermediates. Herein, we develop an in situ organic Fenton-like catalysis by using the redox-active polymeric intermediates, e.g., benzoquinone, hydroquinone, and quinhydrone, generated in electrochemical pollutant oxidation as H₂O₂ activators. By taking phenol as a target pollutant, we demonstrate that the in situ organic Fenton-like catalysis not only improves pollutant degradation, but also refreshes working electrode with a better catalytic stability. Both ¹O₂ nonradical and ·OH radical are generated in the anodic phenol conversion in the in situ organic Fenton-like catalysis. Our findings might provide a new opportunity to develop a simple, efficient, and cost-effective strategy for electrochemical water purification.

The efficient generation of reactive oxygen species is essential for pollutant degradation in water purification. The metal-mediated Fenton catalysis has been widely used for several decades owning to its high efficiency, low cost, and easy operation (1). However, it has several technical drawbacks to largely limit further applications, e.g., harsh pH, metal-rich sludge, secondary pollution, and poor stability (1). Alternatively, the metal-free Fenton catalysis has recently attracted increasing interests. Redox-active compounds serve as the oxidant activator to decompose pollutants via radical and/or nonradical pathways (2–5). These pathways depend highly on the atomic and electronic structures and molecular configurations of compounds and their molecular interactions with oxidants (6–18). So far organic activators are ex situ introduced and cause secondary pollution, although the performance can be largely improved (2–18). Such an intrinsic drawback greatly restricts its practical applications. Thus, in situ organic Fenton-like catalysis without secondary pollution is greatly desired for clean and safe water purification.

Electrochemical oxidation (EO) at low bias is widely used for pollutant degradation owning to its high current efficiency and low energy consumption, but largely suffers from electrode fouling (19, 20). Such fouling is mainly caused by anodic polymeric intermediates with large molecular size, low geometric polarity, and high structural stability, thus anodic oxidation is thermodynamically terminated at this stage (19, 20). How to remove polymeric intermediates is essential for electrochemical water purification. It is interesting to note that anodic polymeric intermediates usually contain quinonelike moieties (C = O) and persistent organic radicals, as the electrons in nucleophilic C-OH can be readily transferred to generate C-O· and C = O (19, 20). Quinonelike moieties are redox-active because of their high electron density and strong electron-donating properties, thus can serve as the metal ligand and reductant to enhance transition-metal redox cycling, and also be involved in the environmental geochemistry of natural organic matters (21–30). Moreover, quinonelike moieties and persistent organic radicals can directly serve as an organic activator to initiate organic Fenton-like catalysis for environmental remediation (31–40). Thus, these redox-active anodic polymeric intermediates are likely to trigger organic Fenton-like catalysis.

Inspired by above analyses, we constructed and validated in situ organic Fenton-like catalysis for electrochemical water purification at low bias before oxygen evolution (Scheme 1). Phenol, a model chemical widely present in environments, and other typical halogenated and nonhalogenated aromatic compounds were selected as target pollutants. Carbon felt (CF), a model material with high activity and low cost, and other typical dimensionally stable anodes were selected as target electrodes. Reaction systems were named in the form of “EO + ex situ added reagent + cathode,” as their anodes were identical. Pollutant degradation and electrode antifouling performances were evaluated under various conditions. After the major reactive oxygen species were identified using a suite of testing methods, and the potential role of trace transition metals, especially iron and copper, was examined, the possible molecular mechanism of the in situ organic Fenton-like catalysis was proposed.

Scheme 1.

Scheme diagrams of the EO-Ti, EO/H₂O₂-Ti, and EO/O₂-CF systems.

Results and Discussion

Nonspecific Feasibility of the In Situ Organic Fenton-like Catalysis.

Aromatic pollutants can be degraded by direct electron transfer or surface-bound reactive oxygen, accompanied by the generation and accumulation of high-resistance polymeric intermediates (19, 20). Less than 20% of phenol was degraded in the EO-Ti system. The phenol degradation continuously improved with anodic bias, but oxygen evolution began to occur (Scheme 1, Fig. 1 A–F, and SI Appendix, Figs. S1–S3). When 3.0 mM H₂O₂ was ex situ added to construct the EO/H₂O₂-Ti system (Scheme 1), the phenol removal was substantially accelerated, which was not caused by the physical adsorption or atmosphere condition or H₂O₂ itself (Fig. 1 and SI Appendix, Figs. S1–S7). The synergistic effects between EO and H₂O₂ were bias-dependent and pH-insensitive compared to the Fe²⁺/H₂O₂ benchmark (Fig. 1 and SI Appendix, Figs. S8–S10). H₂O₂ dosage played an important role because of the self-scavenging effect and anodic decomposition (SI Appendix, Figs. S11 and S12) (1). Thus, the ex situ dosed H₂O₂ was efficient to refine the EO-Ti system for phenol degradation (Fig. 1 and SI Appendix, Figs. S1–S3).

Fig. 1.

Phenol degradation in the EO-Ti, EO/H₂O₂-Ti, and EO/O₂-CF systems at different external bias: +1.1 V (A), +1.2 V (B), +1.3 V (C), +1.4 V (D), +1.5 V (E), and +1.6 V (F). Testing conditions: phenol (100.0 mg·L⁻¹), bias (+1.1 ∼ +1.6 V/SCE), H₂O₂ (3.0 mM), pH (natural, ∼6.5), Na₂SO₄ (0.1 M), solution volume (100.0 mL), anode (carbon felt, 10.0 cm²), cathode 1 (Ti, 10.0 cm²), cathode 2 (CF, 10.0 cm²), electrode gap (2.0 cm), reference electrode (SCE), O₂ bubbling (10.0 mL min⁻¹), stirring rate (500 rpm), and reaction time (3.0 h).

Electrode fouling is the main bottleneck of EO system at low bias (19, 20). Electron transfer, hydrogen abstraction, and radical complexion were main pathways to generate polymeric intermediates on electrode surface and in bulk solution. Severe electrode fouling and activity decrease were observed in the EO-Ti system, but they were substantially alleviated and even exhibited an improved performance in the EO/H₂O₂-Ti system (Fig. 2 and SI Appendix, Figs. S13–S18). For the first, second, and third runs of the degradation tests in the EO/H₂O₂-Ti system, the Langmuir–Hinshelwood first-order reaction rate constant increased from ∼0.030 to ∼0.080 min⁻¹, but it decreased from ∼0.080 to ∼0.015 min⁻¹ when no H₂O₂ was added for the fourth run (Fig. 2B). In comparison, the EO-Ti system exhibited unstable performance (Fig. 2C and SI Appendix, Figs. S13–S15), but it was largely improved when H₂O₂ was added in the fourth run (Fig. 2D).

Fig. 2.

Cyclic phenol degradation on the CF electrode in the catalytic systems of EO/H₂O₂-Ti (A and B) and EO-Ti (C and D). Testing conditions: phenol (100.0 mg·L⁻¹), bias (+1.4 V/SCE), H₂O₂ (3.0 mM), pH (natural, ∼6.5), Na₂SO₄ (0.1 M), solution volume (100.0 mL), anode (CF, 10.0 cm²), cathode (Ti, 10.0 cm²), electrode gap (2.0 cm), reference electrode (SCE), stirring rate (500 rpm), and reaction time (3.0 h).

Redox-active intermediates with rich quinonelike moieties played governing roles in the in situ organic Fenton-like catalysis. Their generation, accumulation, and regulation from the anodic phenol conversion were of considerable interests. CF was selected due to its good anodic activity at low bias. Key reaction parameters, i.e., anodic bias and pH, were finely regulated for phenol conversion and intermediate distribution (Fig. 1 and SI Appendix, Figs. S3, S9, S11, S12, and S19). The significant change of electrode surface chemistry was essential (SI Appendix, Figs. S16–S18). The CF electrode was deactivated quickly with increasing oxygen content after phenol degradation; the fouled electrode even exhibited a Fenton-like activity (SI Appendix, Fig. S19C). Electrode fouling and antifouling played a governing role in the in situ organic Fenton-like catalysis (Scheme 2).

Scheme 2.

Mechanism of the in situ organic Fenton-like catalysis mediated by the anodic polymeric intermediates generated in the electrochemical phenol oxidation for water purification. The organic Fenton-like system is in situ constructed based on the as-generated redox-active polymeric products from the anodic phenol oxidation and the cathodic H₂O₂ formation from the selective two-electron reduction of dissolved oxygen. Anodic polymeric intermediates might trigger a dual Fenton-like activation mechanism of selective nonradical and nonselective radical pathways for pollutant degradation and electrode antifouling in electrochemical water purification.

In situ organic Fenton-like catalysis was also established in the anodic conversion of other typical aromatic pollutants with and without halo substitution or phenol group in molecular structure. Furthermore, the electrochemical treatment of the real wastewater samples containing phenols and the surface water samples containing natural organic matters (SI Appendix, Table S1 and Figs. S20–S23) was also performed. In addition to CF, other electrodes like PbO₂, Sb-SnO₂, and boron-doped diamond also exhibited similar promoting effects (SI Appendix, Fig. S24). These results demonstrate the substrate- and electrode-nonspecific feasibility of the in situ organic Fenton-like catalysis for electrochemical water purification.

Substrate-Specific Mechanisms of the In Situ Organic Fenton-like Catalysis.

A rapid and exhaustive phenol conversion occurred in the EO/H₂O₂-Ti system (Fig. 1 and SI Appendix, Figs. S1–S3). The first step was the anodic conversion to dihydroxylated rings, e.g., catechol, resorcinol, and hydroquinone (HQ) (Reaction 1), producing intermediates such as ortho- and parabenzoquinone (SI Appendix, Figs. S1, S2, S25, and S26) (41). Dihydroxylated rings could react with their own quinones to generate quinhydrone (QHQ) via radical complexion (SI Appendix, Figs. S25 and S26) (41). These intermediates with a plenty of redox-active quinonelike moieties bound strongly onto the CF electrode to deactivate surface reactive sites (Reaction 1) (19, 20). H₂O₂ could be activated by the anodic intermediates via radical and nonradical pathways to generate reactive species, e.g., ¹O₂, ·OH, surface complexes, etc. (Reactions 2–4, Fig. 3 and SI Appendix, Figs. S27–S29) (42–45). In turn, polymeric intermediates were also self-activated to favor further oxidation with reduced energy barriers (Reactions 5 and 6) (19, 20, 42–45). Such an in situ organic Fenton-like catalysis could not only antifoul electrode, but also improve EO performance (Fig. 2 and Scheme 2).

$$\mathrm{pheno}{\mathrm{l}}_{\mathrm{ads}}-n{e}^{-}\to \mathrm{intermediate}{\mathrm{s}}_{\mathrm{ads}},$$ [1]

$$\mathrm{intermediate}{\mathrm{s}}_{\mathrm{ads}}+{\mathrm{H}}_2{\mathrm{O}}_2\to \mathrm{intermediate}{\mathrm{s}}_{\mathrm{ads}}@{\mathrm{H}}_2{\mathrm{O}}_2\to \mathrm{product}{\mathrm{s}}_{\mathrm{ads}}{+}^1{\mathrm{O}}_2,$$ [2]

$$\mathrm{intermediate}{\mathrm{s}}_{\mathrm{ads}}+{\mathrm{H}}_2{\mathrm{O}}_2\to \mathrm{intermediate}{\mathrm{s}}_{\mathrm{ads}}@{\mathrm{H}}_2{\mathrm{O}}_2\to \mathrm{product}{\mathrm{s}}_{\mathrm{ads}}+·\mathrm{OH},$$ [3]

$$\mathrm{intermediate}{\mathrm{s}}_{\mathrm{ads}}\mathrm{+}{\mathrm{H}}_2{\mathrm{O}}_2\to \mathrm{intermediate}{\mathrm{s}}_{\mathrm{ads}}@{\mathrm{H}}_2{\mathrm{O}}_2\to \mathrm{product}{\mathrm{s}}_{\mathrm{ads}}+2\mathrm{O}{\mathrm{H}}^{-},$$ [4]

$$\mathrm{product}{\mathrm{s}}_{\mathrm{ads}}+·\mathrm{OH}{/}^1{\mathrm{O}}_2\to ···\to \mathrm{C}{\mathrm{O}}_2\mathrm{+}{\mathrm{H}}_2\mathrm{O},$$ [5]

$$\mathrm{product}{\mathrm{s}}_{\mathrm{ads}}-n{e}^{-}\to ···\to \mathrm{C}{\mathrm{O}}_2+{\mathrm{H}}_2\mathrm{O}.$$ [6]

Both radical and nonradical mechanisms were involved in the phenol conversion in the in situ organic Fenton-like catalysis (Fig. 3, Scheme 2, and SI Appendix, Figs. S27–S29). ·OH and ¹O₂ directly affected the degradation of pollutants, and such a decomposition process was substantially inhibited by the added CH₃OH and NaN₃, respectively (SI Appendix, Fig. S30). ¹O₂ was found to be the main reactive oxygen species (Fig. 3). Considering ketone molecular structures and good redox-cycling properties, H₂O₂ might be activated by the intermediates to generate ¹O₂ as the major nonradical pathway in the in situ organic Fenton-like catalysis, via the nucleophilic addition, displacement, and decomposition with the key dioxirane intermediates as the rate-limiting step (SI Appendix, Scheme S1 and Figs. S31 and S32) (8, 9). However, because of the very poor stability and immediate consumption of the as-generated reactive dioxirane intermediate, its molecular structure and aqueous concentration could not be detected with the currently available tandem mass spectrometry and liquid chromatography. ·OH was proven as the minor reactive species in such a system. The role of trace transition metals (especially iron and copper) in either the acid-pretreated CF electrode or the Chelex-pretreated supporting electrolyte in the formation of ·OH was excluded by adding the Fe/Cu chelating agents into the EO/H₂O₂-Ti system in the radical detection and phenol degradation tests (SI Appendix, Figs. S33–S37 and Table S2).

Fig. 3.

ESR and fluorescent tests for ¹O₂ and ·OH generation in organic solutions, benchmark systems, and EO/H₂O₂-Ti in the phenol degradation: TEMP/¹O₂ (A), TEMP/¹O₂/NaN₃ (B), TEMP/¹O₂/D₂O (C), DMPO/·OH (D), DMPO/·CH₃ (E), and 3-CCA/·OH (F). Testing conditions: phenol (100.0 mg·L⁻¹), HQ (100.0 mg·L⁻¹), BQ (100.0 mg·L⁻¹), HQ+BQ (50.0 + 50.0 mg·L⁻¹), H₂O₂ (3.0 mM), peroxydisulfate (PDS) (3.0 mM), BiOCl (0.5 g·L⁻¹), FeCl₂ (10.0 mM), TEMP (100.0 mM), NaN₃ (10.0 mM), D₂O (100.0 mM), DMPO (100.0 mM), DMSO (10.0 mM), 3-CCA (10.0 mM), bias (+1.4 V/SCE), pH (natural, ∼6.5), Na₂SO₄ (0.1 M, 24.0-h Chelex-pretreated), solution volume (100.0 mL), anode (CF, 10.0 cm²), cathode (Ti, 10.0 cm²), electrode gap (2.0 cm), reference electrode (SCE), stirring rate (500 rpm), and reaction time (0.5 h). Sample with 3-CCA as the ·OH chemical probe was excited from 340 to 400 nm, and the resulting fluorescence was measured from 350 to 550 nm, respectively.

The ·OH-mediated organic Fenton-like catalysis was demonstrated to occur mainly between H₂O₂ and halogenated quinoid compounds, i.e., tetrachloro-1,4-benzoquinone (2–4). No halogenated quinoid compound was produced and only non–halo-quinoid intermediates were detected (SI Appendix, Figs. S1, S2, and S19). The promotion of anodic system and the complexity of polymeric intermediates might be responsible for the radical activation of H₂O₂ (SI Appendix, Scheme S2). Under anodic polarization, the atomic and electronic structures and molecular configurations of the non–halo-quinoid intermediates and their molecular interactions with H₂O₂ could be electrochemically regulated. They became more thermodynamically favorable to undergo the nucleophilic addition reaction by H₂O₂ to form the reactive intermediates and trigger the homolytic decomposition reaction on benzene ring to generate ·OH (SI Appendix, Scheme S2) (42–45). The reaction between semiquinone radical and H₂O₂ is spin-restricted and impossible to produce ·OH under usual conditions (2). Thus, in the EO/H₂O₂-Ti system no organic Fenton reaction of the semiquinone-mediated one-electron reduction occurred. Instead, the organic Fenton-like reaction of the quinone-mediated nucleophilic addition and homolytic decomposition took place (SI Appendix, Scheme S2).

Mechanisms of the in situ organic Fenton-like catalysis were substrate-specific in the pollutant conversion with and without molecular halo substitution. With the nonhalogenated substrates, i.e., phenol, atrazine, and roxarsone, ¹O₂ served as the main reactive oxygen species and the nonradical activation of H₂O₂ was the major pathway (SI Appendix, Figs. S30 and S38–S40). In comparison, with the halogenated substrates, i.e., dichlorophenol, tetrachlorphenoxide, dibromophenol, and diiodophenol, ·OH acted as the main reactive oxygen species and the radical activation of H₂O₂ was the major pathway (SI Appendix, Figs. S30 and S38–S40). The anodic promoting effect was also substrate-specific for the ·OH generation in the in situ organic Fenton-like catalysis. A medium amount of ·OH was generated in the anodic degradation of the nonhalogenated phenol, while only a trace amount of ·OH was formed in the anodic degradation of the nonhalogenated atrazine and roxarsone (Fig. 3 and SI Appendix, Figs. S38–S40). Both the radical and nonradical mechanisms in the in situ organic Fenton-like catalysis promoted electrochemical water purification (Fig. 1 and SI Appendix, Figs. S20–S24).

Self-Feeding In Situ Organic Fenton-like Catalysis.

Ex situ added H₂O₂ was rapidly decomposed in the initial 30 min via a nonradical pathway under anodic polarization in the EO/H₂O₂-Ti system, regardless of the presence of phenol or not (Fig. 4 A and B). Thus, a fully self-feeding EO/H₂O₂-Ti system was developed by in situ generating H₂O₂ for sustainable organic Fenton-like catalysis (Scheme 2 and SI Appendix, Figs. S41–S44). The self-feeding EO/O₂-CF system exhibited a superior capacity and a good feasibility as evidenced by a higher phenol removal and a faster degradation rate (Figs. 1 and 3 and SI Appendix, Figs. S45 and S46). A significant two-stage reaction was observed, especially at +1.1 and +1.2 V (Fig. 1 A and B), indicating a sustainable synergistic effect between EO and H₂O₂ at the final stage, which was mainly attributed to the effective accumulation of H₂O₂ (Fig. 4C and Scheme 2). The potential-promoted nucleophilic addition reaction by H₂O₂ to form the reactive dioxirane intermediates and trigger the hemolytic decomposition reaction to generate ·OH occurring on different electrodes with various nonhalogenated aromatic compounds (SI Appendix, Fig. S47 A–F). The potential threshold was highly substrate- and electrode-dependent, and a high anodic activity of working electrode and the simple molecular structure of the nonhalogenated compound could favor ·OH generation (SI Appendix, Fig. S47 G and H).

Fig. 4.

H₂O₂ decomposition and accumulation properties in the EO/H₂O₂-Ti system with and without phenol addition (A and B) and in the self-feeding EO/O₂-CF system (C). Testing conditions: phenol (100.0 and 0.0 mg·L⁻¹), bias (+1.3, +1.4, and +1.5 V/SCE), pH (natural, ∼6.5), Na₂SO₄ (0.1 M), solution volume (100.0 mL), anode (CF, 10.0 cm²), cathode 1 (Ti, 10.0 cm²), cathode 2 (CF, 10.0 cm²), electrode gap (2.0 cm), reference electrode (SCE), O₂ bubbling (10.0 mL·min⁻¹), stirring rate (500 rpm), and reaction time (3.0 h).

For electrochemical water purification, the process selectivity and efficiency are of considerable interest for the decontamination performance and energy consumption (1). Anodic current distribution between pollutant degradation and water decomposition is a direct indicator. The EO/O₂-CF system exhibited a larger net current from pollutant degradation, i_net, indicating its higher process efficiency and reaction selectivity with lower energy consumption (SI Appendix, Fig. S48). The negative i_net in the EO/H₂O₂-Ti system suggests that the anodic H₂O₂ decomposition was the dominant reaction in the initial minutes, if its high aqueous concentration was taken into account (Fig. 4). In principle, the great superiority of the EO/O₂-CF system in the current distribution could well explain its excellent water purification and electrode antifouling performances (Fig. 1).

The self-feeding EO/O₂-CF system exhibited an excellent stability with a sustainable promoting effect on the in situ organic Fenton-like catalysis (Fig. 5 and SI Appendix, Figs. S49–S51). The redox-active HQ, benzoquinone (BQ), and QHQ were mainly responsible for the H₂O₂ activation by serving as an electron mediator, and they were much stronger than phenol for reactive oxygen generation (Fig. 3). Considering their distinct formation, the different reactive oxygen generations could well be explained (SI Appendix, Fig. S52). Furthermore, the guest- and self-modifications of the self-feeding in situ organic Fenton-like catalysis, i.e., nitrogen-doped CF cathode (NCF), exhibited a much higher O₂ reduction, H₂O₂ generation, and phenol removal in the EO/O₂-NCF system (SI Appendix, Figs. S3, S53, and S54).

Fig. 5.

Continuous-wave ESR spectra for TEMP/¹O₂ and DMPO/·OH generation in the phenol degradation in the different systems: EO-Ti (A and D), EO/H₂O₂-Ti (B and E), and EO/O₂-CF (C and F). Testing conditions: phenol (100.0 mg·L⁻¹), H₂O₂ (3.0 mM), TEMP (100.0 mM), DMPO (100.0 mM), bias (+1.4 V/SCE), pH (natural, ∼6.5), Na₂SO₄ (0.1 M, Chelex-pretreated), solution volume (100.0 mL), anode (CF, 10.0 cm²), cathode 1 (Ti, 10.0 cm²), cathode 2 (CF, 10.0 cm²), electrode gap (2.0 cm), reference electrode (SCE), O₂ bubbling (10.0 mL·min⁻¹), stirring rate (500 rpm), and reaction time (3.0 h).

CONCLUSIONS

Redox-active anodic intermediates from phenol conversion were in situ utilized to initiate organic Fenton-like catalysis for electrochemical water purification. Phenolic contaminants are widely present in water environments, but their electrochemical oxidation is thermodynamically terminated by the formation of polymeric intermediates. In situ organic Fenton-like system could provide a new way to resolve this problem. The synergy among electroassisted adsorption, anodic oxidation, and organic Fenton-like endows a high catalytic capacity, low energy consumption, and good cyclic stability. Considering that quinonelike moieties are the important components of natural organic matters with conjugated molecular networks, there might be a great possibility to stabilize EO-generated persistent free radicals for in situ organic Fenton-like catalysis in natural environments.

Experimental and Methods

Construction of the EO-Ti, EO/H₂O₂-Ti, and EO/O₂-CF Systems.

Three anodic systems were established on electrochemical workstation (CHI 760d, Chenhua Co.) (Scheme 1). The EO-Ti system was constructed in a cylindrical three-electrode single-compartment cell with CF, Ti, and saturated calomel electrode (SCE) as working electrode, counterelectrode, and reference electrode, respectively. CF electrode was thermally pretreated in 0.1 M HCl aqueous solution at 100 °C for 2.0 h, then ultrasonically cleaned in methyl alcohol, acetone, and distilled water in sequence. The working and counterelectrodes were both 2.0 × 6.0 cm² in size, with electrode gap of ∼2.0 cm. No additional external resistance or iR compensation was used. The EO/H₂O₂-Ti system was constructed by dosing H₂O₂ into the EO-Ti system, while the EO/O₂-CF system was established based on the EO-Ti system with two changes: 1) replacing Ti cathode by CF cathode to in situ generate H₂O₂ from O₂ reduction, and 2) aerating the supporting electrolyte by continuous O₂ flow under CF cathode.

Phenol Degradation Test.

Phenol degradation tests were carried out in the EO-Ti, EO/H₂O₂-Ti, and EO/O₂-CF systems for comparison. Typically, 100.0 mL Na₂SO₄ aqueous solution (0.1 M) containing 50.0 ∼300.0 mg·L⁻¹ phenol was electrolyzed in 120.0-mL electrochemical cell, which was continuously stirred by a magnetic stirrer at 500 rpm and the applied bias was controlled in the range of +1.1 ∼ +1.7 V (versus SCE). Solution samples of 2.0 mL were regularly collected and rapidly microfiltered for analysis. All degradation tests were carried out in duplicate, and the mean values with SD are presented.

Electron Spin Resonance and Fluorescent Tests.

Reactive oxygen radicals formed by phenol/H₂O₂, intermediates/H₂O₂ and reaction solution were monitored by electron-spin resonance (ESR) with 5,5-dimethyl-1-pyrroline N-oxide (DMPO) and 2,2,6,6-tetramethyl-4-piperidone (TEMP) as spin-trapping agents with and without dimethyl sulfoxide (DMSO) or sodium azide (NaN₃) or deuterium oxide (D₂O). They were also characterized by fluorescence with coumarin-3-carboxylic acid (3-CCA) and terephthalic acid (TPA) as chemical probes. For the ESR tests, the basic system consisted of 100.0 mg·L⁻¹ organics, 3.0 mM H₂O₂, 100.0 mM DMPO or TEMP, 10.0 mM DMSO or NaN₃ or 100.0 mM D₂O. For the fluorescent tests, 10.0 mM 3-CCA or 3.0 mM TPA in Chelex-treated Na₂SO₄ aqueous electrolyte (0.1 M, pH ∼6.5) was used at ambient temperature. ESR spectra were recorded either immediately after interacting with H₂O₂, or at given time intervals on a Bruker spectrometer (A300) with the following settings: center field = 3,512 G, microwave frequency = 9.86 GHz, and power = 6.36 mW. Fluorescence detection was performed on a spectrofluorometer (RF-5301PC, Shimadzu Co.). The samples with 3-CCA and TPA were excited from 340 to 400 nm and at 312 nm, and the resulting fluorescence was measured from 350 to 550 nm and at 426 nm, respectively.

Analysis.

Pollutants were determined by high performance liquid chromatography (HPLC)y (HPLC-1100, Agilent Inc.) with a Hypersil-ODS reversed-phase column and detected at 254 nm using a VWD detector. Mobile phase was a mixture of water and methanol (30:70) delivered at a flow rate of 1 mL·min⁻¹. Mineralization was determined from total organic carbon (TOC) (Vario TOC cube, Elementar Co.). Intermediates were identified by gas chromatography mass spectrometry (GCT Premier, Waters Inc.) and liquid chromatography mass spectrometer (LCMS-2010A, Shimadzu Co.). H₂O₂ was monitored using iodometric titration method. Electrode potential was measured using a commercial multimeter (UT39A, UNIT Inc.).

Data Availability.

All study data are included in the article and SI Appendix.