Advanced Oxidation of Catechol in Atmospheric Waters

Abstract

Phenols such as catechol emitted during biomass burning and combustion contribute to aqueous secondary organic aerosol (aqSOA) formation, influencing air quality and climate. Here, catechol (QH₂) is used as a surface-active, water miscible model phenol to investigate advanced oxidation in atmospheric waters. We demonstrate that exposure to ozone and Fenton’s reagent (FR; Fe²⁺ + H₂O₂) accelerates catechol oxidation and promotes o-quinone formation. Ozone contributes through direct oxidation, iron redox cycling, and reactive oxygen species generation in the coupled O₃/Fenton system. Experiments were conducted with 200 μM catechol under variable conditions: (i) pH 1.0–9.0; (ii) ozone mixing ratios of 0.4–3.6 ppmv; (iii) presence/absence of FR (1 μM Fe²⁺, 50 μM H₂O₂); and (iv) dimethyl sulfoxide (DMSO) as a hydroxyl radical (HO^•) scavenger. Catechol loss was monitored by UV–visible spectroscopy at λ_max = 276 nm. Product formation was tracked by UV–visible at λ_max = 389 nm (o-quinone) and λ_max = 490 nm (catechol–Fe complex). Kinetics were quantified within an early oxidative window (0–2 min), when H₂O₂ is rapidly consumed and radical production is most active. Under representative conditions (pH 3.0; O₃(g) = 0.4 ppmv; FR present), catechol decayed with a pseudo first order rate constant k′ = 1.69 × 10^–2 min^–1, substantially faster than in the absence of FR. Within the same window, o-quinone reached ∼7.5 μM. These results highlight how coupled ozone–Fenton chemistry can accelerate phenolic oxidation in atmospheric aqueous phases, affecting aqSOA aging and optical properties.

Introduction

Atmospheric waters, including cloud droplets and aerosol-bound moisture, drive key chemical and biological processes that influence the Earth’s environment. Since the industrial revolution, increased emissions of greenhouse gases (GHGs) and volatile organic compounds (VOCs) from urbanization and industrial activities have exacerbated environmental challenges such as air pollution and climate change. In the atmosphere, VOCs such as benzene, toluene, and anisole undergo oxidation by hydroxyl radicals (HO^•) and ozone (O₃), yielding phenols and polyphenols, including catechols. These intermediates engage in interfacial reactions, contributing to the formation of aqueous secondary organic aerosols (aqSOA), which significantly impact both air quality and climate systems. −

Catechol, a key VOC-derived compound, participates in atmospheric processes occurring in atmospheric waters, including cloud droplets and deliquesced aerosols. It readily forms complexes with transition metals such as iron, which is commonly found in atmospheric particles, particularly within cloud droplets. , These complexes participate in accelerated oxidative processes, i.e., via the Fenton reaction, enhancing organic compound transformation. Iron catalysis facilitates the formation of sulfate and organic aerosols, influencing cloud formation, radiative forcing, and overall atmospheric composition. ,

This study aims to unravel the complexity of a system where catechol, iron ions, and oxidizers such as dissolved O₃(aq) and H₂O₂ coexist, driving advanced oxidation processes (AOPs) that include the Fenton reaction. While AOPs have been widely recognized as effective methods for wastewater remediation, , their role in atmospheric chemistry, particularly within aqueous organic aerosol, remains largely unexplored. We hypothesize that metal-mediated AOPs play a significant role under atmospherically relevant conditions. To investigate this, catechol was selected as a molecular polyphenol model pollutant to probe the oxidation dynamics of aqSOA. This approach reveals a previously underexamined pathway that generates reactive semiquinone radicals, potentially influencing air quality and climate models. Given the transient nature of hydroxyl radicals (HO^•), their formation was monitored using dimethyl sulfoxide (DMSO), chosen for its stability, solubility, and lack of interference with metal chemistry. These findings provide valuable insight into organic aerosol transformation mechanisms and highlight the necessity of incorporating metal-mediated AOPs into atmospheric chemistry models.

Materials and Methods

Chemicals

All the chemicals used in this work were of analytical grade and their purity per certificate of analysis is provided next. Catechol (Sigma-Aldrich, 99.9%), DMSO (Alfa Aesar, ≥99.9%); ferrous sulfate heptahydrate (Sigma-Aldrich, ≥99%); hydrogen peroxide (Acros, 30.0%); NaOH (Amresco, ACS grade), and HCl (EMD Millipore, ACS grade). All the reagents were prepared with ultrapure deionized water (18.2 MΩ·cm, ELGA PURELAB flex, Veolia).

Reactor Setup

The reaction is conducted in a 60 mL amber glass reactor (Scheme ). Initially, the water is adjusted to a specific pH value, ranging from 1 to 9, to reflect the acidic to neutral conditions typically observed in atmospheric environments (commonly pH 1 to 7) such as aerosols, fog, and clouds, and to include higher pH values for mechanistic investigations. The pH adjustment is done by the dropwise addition of 0.1 M HCl or 0.1 M NaOH under continuous magnetic stirring and placed in the reactor. A calibrated ozone generator (UVP, 97-0067-01) produces a stable concentration of ozone [O₃(g)] from a 0.5 L min^–1 flow of ultrapure air provided with a meter (Aalborg). The ozone generated is bubbled into the open reactor through a medium-coarse sparger (12 mm long and 7 mm wide). By varying the irradiation path length (from 2.54 to 22.86 cm) using a 185 nm mercury lamp, the generator produces ozone concentrations ranging from 0.4 to 3.6 ppmv. The resulting ozone flow controlled by a flow meter (Aalborg) was diverted into the reactor. A flow of 131 mL min^–1 of 0.40 ppmv O₃(g), representative of concentrations typically found in polluted atmospheric plumes or urban environments, is sparged in a typical experiment through a coarse frit (12 mm long, 7 mm wide) into the water for 30 min (Scheme ). This process yields an aqueous ozone concentration of [O₃(aq)] = 4.40 × 10^–9 M, based on the Henry’s law constant, K _H = 0.011 M atm^–1 at 25 °C. Higher concentrations of O₃(aq) and flows were examined to gain mechanistic insight into the reaction pathways.

1. Experimental Setup with Ozone Generator for Advanced Oxidation of Catechol .

^a The flow of pure air is controlled with a flow meter (Aalborg PMR 1-017005) and fed to the ozone generator. The ozone generated is bubbled into the open reactor through a medium-coarse sparger.

For the preparation of an oxidative environment, 1.00 mL of 50.0 μM Fe²⁺ and 2.50 mL of 1.00 mM H₂O₂ are sequentially added to the water saturated with O₃(aq). In this way, hydroxyl radicals (HO^•) are generated by the dark Fenton reaction under acidic to neutral conditions. DMSO is also introduced in selected reactions as a radical scavenger for HO^•. The final concentrations of Fe²⁺ and H₂O₂ in the solution are 1.0 μM and 50 μM, respectively, consistent with atmospherically relevant levels in clouds and aerosols. These values fall within the lower end of the reported 1–1000 μM range for Fe²⁺, , and align with the mean H₂O₂ concentrations of 18–71 μM measured in rainwater.

Catechol Addition to the Reaction Mixture

The final step in the preparation of a typical experiment is the addition of a freshly prepared 650 μM catechol standard to form a 45.0 mL diluted reagent mixture of 200 μM unless noted otherwise. A series of experiments tested the effect of varying parameters, while keeping the initial [catechol] at 50, 100, 200, 300, or 500 μM. These concentrations were selected to investigate the kinetics and mechanisms of catechol’s advanced oxidation, encompassing the range (22–223 μM) reported as relevant to atmospheric waters impacted by biomass burning. Accordingly, the selected experimental conditions were chosen to remain within reported atmospheric ranges while enabling quantitative monitoring of reaction intermediates and kinetics. The other varying parameters were [DMSO] of 5, 10, 25, and 50 μM; O₃(g) molar ratio of 0.40, 1.20, 2.00, 2.80, and 3.60 ppmv, flow rate (F) set to 131, 238.0, 329.0, and 413.0 mL min^–1, and pH adjusted from 1.0–9.0, as tested with a calibrated pH meter (Thermo Scientific).

Control Experiments

To validate the role of each reagent, a series of controls is performed while changing one condition at a time (Table S1, Supporting Information) by using (control A) in the presence of DMSO scavenger, (control B) no Fe²⁺ but with H₂O₂, (control C) no H₂O₂ but with Fe²⁺, (control D) no Fenton reagent (H₂O₂ and Fe²⁺), (control E) no O_3(g) (air only), and (control F) no catechol as shown in Table S1 (Supporting Information).

UV–Visible Analysis

A Thermo Fisher Scientific Evolution 220 UV–visible spectrometer equipped with a 1.00 cm quartz cuvette was used for UV–visible measurements. All UV–visible experiments were performed in triplicate, and reported uncertainties correspond to ±1 standard deviation. Spectra were recorded from 190 to 750 nm with a bandwidth of 1 nm. Aliquots (2.0 mL) taken at 0, 0.33, 0.67, 1, 2, 5, 10, 20, 30, 40, and 60 min were studied. Sampling times refer to aliquot withdrawal from the reactor; aliquots were collected manually with a micropipette at predetermined times and transferred immediately for UV–visible analysis, with spectra typically acquired within 1–2 min of sampling. Quantitative analyses are facilitated with calibration curves of catechol at λ_max = 276 nm and DMSO at λ_max = 209 nm, using the molar absorptivity 2.53 × 10³ and 3.92 × 10³ M^–1cm^–1, respectively. The literature values of the molar absorptivity of o-quinone at λ_max = 389 nm and the catechol-iron­(III) complex at λ_max = 490 nm is 1.37 × 10³, and 4.19 × 10³ M^–1 cm^–1, respectively.

Amperometric Determination of Hydrogen Peroxide

Hydrogen peroxide (H₂O₂) was detected and quantified using a sensor based on a screen-printed carbon electrode (SPCE) modified with Prussian Blue. The SPCE consists of a carbon-paste working and counter electrodes, along with an Ag/AgCl reference electrode. Prussian Blue is electrochemically deposited onto the working electrode using cyclic voltammetry and equimolar solutions of 2.5 mM Ferric nitrate nonahydrate, Fe­(NO₃)₃·9H₂O, (ACS grade, 99.4%, Mallinckrodt AR) and Potassium ferricyanide, K₄[Fe­(CN)₆] (ACS grade, 99.5%, J.T. Baker) in 0.1 M HCl (ACS grade, 36.5%, EMD Milipore Co.) and 0.1 M KCl (ACS grade, 99.4%, VWR) following a previously established procedure. All solutions were prepared using ultrapure deionized water (18.2 MΩ·cm, ELGA). The potential was scanned from +1.1 V to −0.2 V vs Ag/AgCl, 0.1M KCl at 50 mV·s^–1 for five cycles (Supporting Information Figure S1A,B). Modified electrodes were thoroughly rinsed with deionized water and stored in 0.1 M KCl at 4 °C for at least 1 h before use to ensure film stabilization.

In an acidic solution, H₂O₂ can be reduced electrocatalytically under PB-modified SPCEs at 0.00 V vs Ag/AgCl, 0.1 M KCl (Supporting Information Figure S2A). Therefore, H2O2 was amperometrically determined at the fixed potential of 0.00 V vs Ag/AgCl in 0.1 M KCl supporting electrolyte, employing the standard addition method (n = 5). The detection limit (DL) and lower limit of detection (LLD) were also determined by the standard addition method (n = 10) in the absence of an unknown sample (Supporting Information Figure S2B). The amperometric determination proceeds by first establishing a steady-state current for unknown samples (∼40 s), followed by standard additions of 0.5 mL, 100 μM H₂O₂ (ACS grade, 30%, Acros) at 20 s intervals, and extracting the steady-state cathodic currents from each interval (Supporting Information Figure S3). Calibration curves were constructed by plotting the steady-state current response against the added concentration, and unknown concentrations were determined from the x-intercept of the resulting linear regression. Reported values represent the average concentrations of three measurements (n = 3), with error bars representing the standard deviation (±s).

Results and Discussion

Advanced Oxidation of Catechol by Combined Fenton Chemistry and Ozone

The advanced oxidation process (AOP) involving ozone (O₃) and Fenton reagents (FR) proceeds rapidly through a cascade of short-lived reaction intermediates, dominated by HO^• attack on catechol. The high reactivity of this system arises from the catalytic decomposition of H₂O₂ by Fe²⁺, which generates HO^• according to Reactions –. −

$${\mathrm{Fe}}^{2+}+{\mathrm{H}}_2{\mathrm{O}}_2\to {\mathrm{Fe}}^{3+}+{\mathrm{HO}}^{\mathrm{•}}+{\mathrm{HO}}^{-};k_1=76{\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$ R1

$${\mathrm{Fe}}^{2+}+{\mathrm{HO}}^{\mathrm{•}}\to {\mathrm{Fe}}^{3+}+{\mathrm{HO}}^{-};k_2=3\times {10}^8{\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$ R2

In this hybrid O₃/FR system, ozone contributes both direct electrophilic attack on electron-rich aromatic rings and indirect radical-driven pathways initiated by HO^•, while the Fe²⁺/H₂O₂ pair sustains continuous radical production. Figure shows the UV–visible results obtained for 200 μM catechol oxidized with a 1:50 FR ratio (1 μM Fe²⁺: 50 μM H₂O₂) under a continuous 131 mL min^–1 flow of 0.4 ppmv O₃(g) for 1 h. Under these conditions, rapid depletion of catechol and formation of characteristic oxidation products are observed, consistent with HO^•-initiated ring-opening and subsequent formation of low-molecular weight carboxylic acids.

1.

UV–visible absorption spectra of experiment with 200 μM catechol, 50.0 μM H₂O₂ and 1.0 μM Fe²⁺ at pH 3.0 under 131 mL min^–1 flow of 400 ppbv O₃(g).

In Figure , the characteristic absorbance of catechol at 276 nm decreases steadily and accompanied by a simultaneous increase in the absorbance signals associated with o-quinone at 389 nm and the catechol–Fe complex at 490 nm, as shown in the inset. The weak absorption feature near 389 nm observable at time zero reflects a background amount of o-quinone formed during the brief mixing and sampling period preceding acquisition of the first spectrum. Subsequent o-quinone concentrations were determined relative to this initial absorbance by subtracting the background contribution. The ratio of [Fe²⁺]: [H₂O₂] directly influences the flux of HO^• generated during Fenton oxidation, since Fe²⁺ is the primary initiator of H₂O₂ decomposition. In the hybrid O₃/FR system, Fe²⁺ drives classical Fenton chemistry and reacts with ozone to produce additional HO^•, thereby enhancing the overall oxidative capacity of the system as seen in the (Reactions –) below.

$${\mathrm{Fe}}^{2+}+{\mathrm{O}}_3\to {\mathrm{Fe}}^{3+}+{\mathrm{O}}_3^{\mathrm{•}-};{k_3=1.7\times {10}^5\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$ R3

$${\mathrm{O}}_3^{\mathrm{•}-}\rightleftharpoons {\mathrm{O}}_2+{\mathrm{O}}^{\mathrm{•}-};K_4=5.5\times {10}^{-7}\mathrm{M}(\mathrm{a}\mathrm{t}20\mathrm{°C})$$ R4

$${\mathrm{O}}^{\mathrm{•}-}+{\mathrm{H}}^{+}\rightleftharpoons {\mathrm{HO}}^{\mathrm{•}};{\mathrm{p}K}_a=11.8$$ R5

The advanced oxidation of catechol in this system proceeds through coupled radical- and metal-mediated pathways. Reactive oxygen species (ROS) are generated in parallel via Fe²⁺ + H₂O₂ Fenton Reaction , and Fe²⁺ + O₃ Reaction . During the initial oxidative window (0–2 min), rapid H₂O₂ consumption and the strong inhibitory effect of DMSO indicate that HO^•-mediated oxidation is the predominant pathway for catechol transformation. As summarized in Scheme , catechol oxidation generates o-semiquinone radical intermediates that can subsequently undergo additional HO^•-mediated oxidation, Fe­(III)-mediated electron-transfer reactions, and proton-coupled redox transformations leading to o-quinone formation. , Concurrently, Fe³⁺ produced during iron redox cycling coordinates catechol and catecholate species, forming Fe–catechol complexes (Scheme ). , A Fe­(III)-initiated experiment conducted under otherwise identical conditions produced similar spectroscopic features associated with catechol oxidation and Fe-catechol coordination chemistry (Figure S4, Supporting Information), supporting the inclusion of Fe­(III)-mediated pathways in Scheme . Depending on solution pH and Fe³⁺ availability, catechol binds iron in mono-, bis-, or tris-catecholate configurations, as summarized in Scheme . − Chloride ions introduced during pH adjustment may participate in acidic O₃/Fenton chemistry. Chloride can scavenge HO^• radicals to form less reactive chlorine-centered species, such as Cl₂ ^•–, while reactive chlorine species may also form through chloride oxidation by ozone. , However, chloride is expected to contribute only secondarily to catechol transformation under the experimental conditions employed here. In addition to direct catechol oxidation, H₂O₂ and O₃ promote Fe²⁺/Fe³⁺ redox cycling and ROS generation, which sustain the formation of semiquinone intermediates and their subsequent oxidation to o-quinone.

2. Proposed Mechanism for Catechol Oxidation in the Coupled O₃/Fenton System ,

^a H₂O₂ and O₃ promote ROS generation and Fe²⁺/Fe³⁺ redox cycling. Catechol oxidation proceeds through both HO^•-mediated and Fe­(III)-mediated pathways involving catechol-Fe complexes, o-semiquinone radical intermediates, proton-coupled electron-transfer reactions, and subsequent formation of o-quinone. ,

3. Formation of pH Dependent Tris-, Bis-, and Mono-Fe³⁺-Catechol Complexes ,,

As previously mentioned, the concentration of H₂O₂ was monitored by amperometry, given that it is the primary oxidant driving the reaction. The amperometric determination proceeds by establishing a steady-state cathodic current for the reaction mixture, followed by standard additions of 0.5 mL aliquots of 100 μM H₂O₂ at 20 s intervals. Calibration curves were constructed from the resulting steady-state current responses (Figure S3A), and unknown H₂O₂ concentrations were determined from the x-intercept of the linear regression. Based on time-resolved amperometric measurements (Figure S3B), the initial H₂O₂ concentration of 50.0 μM decreased to ∼12.5 μM after 2 min, corresponding to ∼75% depletion. This rapid loss of H₂O₂ demonstrates that Fe²⁺-driven ROS production is confined to the initial 2 min of the O₃/FR process, defining the kinetically relevant oxidative window from which all reported rate constants for catechol loss (k′) and product formation (k ₀) are derived. The corresponding early-time concentration profiles for catechol, oxidation products, iron complexes, and H₂O₂ within this oxidative window are shown in Figure . Consequently, the kinetic analysis focuses on this early oxidative window, where HO^• production is maximal, while slower oxidation pathways involving ozone and Fe-catechol redox chemistry are expected to become relatively more important after substantial H₂O₂ depletion.

2.

Initial kinetics of the experiment shown in Figure , monitored at (black circle) λ = 276 nm for catechol ([Cat] _t/[Cat]₀), (red square) λ = 389 nm for o-quinone, and (blue diamond) λ = 490 nm for the catechol-Fe (Cat-Fe) complex at. The normalized hydrogen peroxide concentration ([H₂O₂] _t /[H₂O₂]₀) is shown for comparison. Experimental conditions: 200 μM catechol, 1.0 μM Fe²⁺, 50.0 μM H₂O₂, pH 3.0, and 131 mL min^–1 flow of 400 ppbv O₃(g).

Within the initial 2 min oxidative window shown in Figure , the kinetics indicate a 3.46 ± 0.06% loss of catechol (6.91 ± 0.12 μM), accompanied by a 75.1 ± 0.5% depletion of H₂O₂ and the formation of 3.74 ± 0.01% o-quinone (7.48 ± 0.01 μM), all relative to the initial catechol concentration. These values are measured explicitly at t = 2 min, corresponding to the period of active Fe²⁺ driven HO^• production. In addition, 0.164 ± 0.001% catechol–Fe complex (0.327 ± 0.002 μM) is formed over the same 2 min interval, as shown in Figure . Because catechol, o-quinone, and catechol-Fe complex were quantified independently by UV–visible methods, these values should not be interpreted as a complete quantitative mass balance.

Kinetic Analysis and Role of Oxidants in Catechol Oxidation

As summarized in Table S1, the reaction of the experiment and controls with catechol in Figure and the controls in Figure S5 (Supporting Information) proceed faster in the presence of FR and O₃. As expected, introducing DMSO substantially suppresses the formation of the oxidation products. It was confirmed that DMSO by itself in the presence of FR (no catechol added) does not interfere in the UV–visible analysis (Figure S6, Supporting Information).

3.

Initial kinetic profiles involving (A) normalized catechol concentration, [Cat] _t /[Cat]₀, monitored at λ = 276 nm (upper portion of panel A, left y-axis), and o-quinone concentration monitored at λ_max = 389 nm (lower portion of panel A, right y-axis) for the experiment (black circle) shown in Figure , and controls A (red square, with 50 μM DMSO), control B (blue triangle, without Fe²⁺), control C (pink star, without H₂O₂), control D (green diamond, without Fe²⁺ and H₂O₂), and control E (purple triangle down, without O₃(g); air only) under 131 mL min^–1 flow of 400 ppbv O₃(g), and (B) concentration of catechol-Fe complex produced at λ_max = 490 nm. Experimental conditions: 200 μM catechol, 1.0 μM Fe²⁺, 50.0 μM H₂O₂, pH 3.0, and 131 mL min^–1 flow of 400 ppbv O₃(g).

Figure informs about synergistic effects that are active during AOP as compared to the individual contribution of reagents. Particularly, ozone in the presence of the FR can enhance the advanced oxidation of catechol by providing the channel of reactions R5 + R6. Catechol possesses two dissociation constants (pK _a1 = 9.45 and pK _a2 = 13.74) , and in the presence of Fe³⁺, the deprotonated state of catechol is stabilized through the thermodynamically favorable formation of a coordination complex, thereby resulting in a drop of the first dissociation constant of catechol (pK _a1) to values of ∼5–8. ,, The concentration of the resulting catechol and its products (o-quinone and catechol-Fe complex) formed at 276, 389, and 490 nm, respectively were determined by subtracting the contributions of the o-quinone at those wavelengths. Catechol concentrations were determined from calibration curves at 276 nm, whereas o-quinone and catechol-Fe complex concentrations were calculated using literature molar absorptivities at 389 and 490 nm, respectively.

The apparent rate constant for catechol oxidation was determined using eqs –. Although catechol oxidation in the O₃/FR system proceeds through both direct ozone reactions and Fenton-generated HO^• radicals, the latter were not quantified independently. Therefore, eq is used as an apparent kinetic framework based on the dissolved ozone concentration, while the experimentally derived k′ reflects the overall oxidation occurring in the coupled O₃/Fenton system. Under pseudo-steady-state conditions, the rate of catechol decay can be expressed as

$$\frac{\mathrm{d}[{\mathrm{Q}\mathrm{H}}_2]}{\mathrm{d}t}=-k[{\mathrm{O}}_3][{\mathrm{Q}\mathrm{H}}_2]$$ 1

Because the experiments are carried out under a continuous flow of O₃(g), the dissolved ozone concentration, [O₃(aq)], is treated as approximately constant during the initial 2 min fitting period. Under these conditions, the second-order rate expression can be simplified to a pseudo-first-order form by treating the dissolved ozone concentration, [O₃(aq)], as approximately constant. The resulting apparent pseudo-first-order rate constant, k′, describes catechol loss within the coupled O₃/Fenton system during the initial oxidative window, when both dissolved ozone and Fenton-generated HO^• radicals contribute to oxidation. This approximation is valid only within the early oxidative window, during which oxidant concentrations remain quasi-steady and competitive scavenging and reagent depletion are minimal; deviations from pseudo-first-order behavior are expected at longer reaction times. Accordingly, the rate law can be expressed as

$$\frac{\mathrm{d}[{\mathrm{Q}\mathrm{H}}_2]}{\mathrm{d}t}=-k^{\prime }[{\mathrm{Q}\mathrm{H}}_2]$$ 2

where k′ is an apparent pseudo-first-order rate constant describing catechol loss in the coupled O₃/Fenton system during the initial oxidative window. Integration of eq yields

$$\ln \frac{{[{\mathrm{Q}\mathrm{H}}_2]}_t}{{[{\mathrm{Q}\mathrm{H}}_2]}_0}=-k^{\prime }t$$ 3

consistent with first-order decay kinetics, where linear regression of ln ([QH₂] _t /[QH₂]₀) versus time gives k′ as the slope. The pseudo-first-order rate constant k′ is reported in units of min^–1 throughout this work unless otherwise specified. The lifetime of catechol (τ) relevant to secondary organic aerosol formation can be estimated using

$$\tau =\frac{1}{k{[{\mathrm{Q}\mathrm{H}}_2]}_0[\mathrm{o}\mathrm{x}\mathrm{i}\mathrm{d}\mathrm{i}\mathrm{z}\mathrm{e}\mathrm{r}]}$$ 4

where [QH₂]₀ is the initial catechol concentration (μM), k is the second-order rate constant derived from the measured k′, and [oxidizer] is the equilibrium concentration of the oxidant (e.g., O₃ or HO^•). The lifetime τ corresponds to the time required for catechol to decrease by 37% and is reported in minutes to maintain consistency with the kinetic analysis. ,

Figure displays the pseudo-first-order rate constants obtained from the initial four concentration points in Figure , corresponding to the 0–2 min time interval, which defines the kinetically relevant early oxidative window. This fitting window coincides with the period of rapid H₂O₂ depletion (∼75% within the first 2 min, Figure S3B), during which Fe²⁺-driven HO^• production remains active and oxidant concentrations are quasi-steady. Concentration–time profiles were constructed from Figure using molar absorptivities reported by Albarran et al., and the resulting data were fitted according to the pseudo-first-order treatment described in eqs –. Within this early interval, the formation of o-quinone is evident in Figure from its monotonic increase over time, while a similar trend is observed for the catechol–Fe complex at 490 nm (Figure ). Rate constants and product yields vary with experimental conditions, with the full O₃/FR system exhibiting the most rapid catechol oxidation and product formation.

4.

Pseudo-first-order rate constants for catechol loss (k′) obtained from experiment shown in Figure and corresponding control systems shown in Figure . Experimental conditions: 200 μM catechol, 1.0 μM Fe²⁺, and 50.0 μM H₂O₂ at pH 3.0 under a 131 mL min^–1 flow of 400 ppbv O₃(g). Control A (with 50 μM DMSO), Control B (without Fe²⁺), Control C (without H₂O₂), Control D (without Fe²⁺ and H₂O₂), and Control E (without O₃(g); air only). Rate constants were derived by fitting concentration–time data within the initial 0–2 min oxidative window.

In Figure , catechol oxidation follows pseudo-first-order kinetics, with the corresponding rate constants (k′, reported in min^–1) obtained by fitting concentration–time data within the same initial 2 min oxidative window for both the experiment and all control systems. Under full O₃/FR conditions of the experiment, the concentration–time profiles shown in Figure yield a pseudo-first-order rate constant for catechol loss (Figure ) of k _QH2+FR+O3 = 1.69 × 10^–2 min^–1 (experiment), which is approximately 2.7-times larger than that obtained in the presence of 50 μM DMSO, k _{QH2+DMSO+FR+O3} = 6.26 × 10^–3 min^–1 (control A). The reduced rate constant in the presence of DMSO reflects efficient scavenging of HO^•, which would otherwise react with catechol and accelerate oxidation. Moreover, Figure S7 (Supporting Information) shows that when increasing the DMSO concentration from 5, 10, 25, to 50 μM the kinetics of the experiment is progressively deaccelerated. Because HO^• are highly reactive and short-lived, their direct detection is challenging; therefore, radical scavengers such as 4-hydroxybenzoic acid, terephthalic acid, and dimethyl sulfoxide (DMSO) are commonly employed to probe HO^• production rates. DMSO is particularly well suited for this purpose due to its high solubility and minimal propensity to form complexes with iron or other metal ions. Similarly, pseudo-first-order rate constants for the remaining control systems were derived from fits to the same initial 0–2 min oxidative window and are summarized in Figure . The corresponding values are k _QH2+H2O2+O3 = 3.28 × 10^–3 min^–1 (control B), k _{QH2+Fe²⁺+O3} = 2.26 × 10^–3 min^–1 (control C), k _QH2+O3 = 1.50 × 10^–3 min^–1 (control D), and k _QH2+DMSO+FR = 5.99 × 10^–3 min^–1 (control E).

According to Pillar et al., the aqueous phase lifetime for catechol loss by hydroxyl radicals at pH 9 is τ_QH2+HO^• = 9.5 h (eq ), based on the rate of k _{QH2+HO ^•} = 1.1 × 10¹⁰ M^–1 s^–1, ,, and a tropospheric HO^• concentration of [HO^•] = 1.6 × 10⁶ radicals cm^–3, corresponding to an aqueous phase HO^• concentration [HO^•(aq)] = 2.65 × 10^‑15 M. This lifetime is reported to be 5.6× greater than catechol loss by HO^• in the gas phase (τ_QH2+HO^• = 1.7 h).

For ozone, the dissolved concentration [O₃(aq)] under continuous O₃(g) flow was estimated by Henry’s law:

$${[{\mathrm{O}}_3(\mathrm{a}\mathrm{q})]=K}_{\mathrm{H},{\mathrm{O}}_3}{C}_{\mathrm{P},{\mathrm{O}}_3}(1-P_{\mathrm{W}})$$ 5

where the Henry’s law constant and vapor pressure of water at 293 K are K _H,O3 = 0.011 M atm^–1 and P _H2O = 0.0228 atm, yielding [O₃(aq)] = 4.40 × 10^–9 M for 0.4 ppmv O₃(g) and 3.96 × 10^–8 M for 3.6 ppmv O₃(g), respectively.

To compare with literature second-order rate constants, the pseudo-first-order constant k′ (min^–1) can be converted to a second-order rate constant k (M^–1 s^–1) using k′ = k [O₃(aq)] and unit conversion from minutes to seconds

$$k=\frac{k^{\prime }}{[{\mathrm{O}}_3(\mathrm{a}\mathrm{q})]}\left(\frac{1\min }{60\mathrm{s}}\right)$$ 6

For example, at 0.4 ppmv O₃ (g), using the measured k′ and [O₃(aq)] = 4.40 × 10^–9 M, the resulting k can be directly calculated from the expression above (with all units tracked explicitly). The resulting value is within one order of magnitude of the literature aqueous rate constant k _QH2+O3 = 5.2 × 10⁵ M^–1 s^–1, differing by a factor of <10; remaining differences likely reflect variation in experimental conditions, particularly pH.

In addition, Figure A shows o-quinone formation at 389 nm. Under O₃/FR conditions, a typical reaction produces 7.5 μM o-quinone after 2 min, which corresponds to 3.8% conversion of the initial 200 μM catechol concentration (7.5/200 = 0.0375) and is 14.3× greater than the yield obtained with O₃ alone.

Finally, Scheme summarizes ozone-initiated radical chemistry: catechol reacts with O₃ to form the ozonide radical anion O₃ ^•–, which is in equilibrium with O^•– (pK _a = 11.8) (equilibrium R4 in Scheme ; with a constant K = 5.5 × 10^–7 M at 20 °C). , Subsequent oxidation depends on in situ HO^• formation via protonation equilibria of O^•– (Reaction ) at pH ≤ 10.8. −

Since Fe²⁺ is oxidized to Fe³⁺ by ozone and H₂O₂, as described in Reactions and , catechol subsequently coordinates Fe³⁺ to form a 1:1 complex. Upon deprotonation, a Fe³⁺-catecholate complex is generated (reaction R5, Scheme ). Because iron undergoes redox cycling, Fe³⁺ is then reduced back to Fe²⁺ via inner-sphere electron transfer from the coordinated ligand, leading to formation of an Fe²⁺-semiquinone complex (reaction R6, Scheme ). This process is followed by release of Fe²⁺ to generate the free o-semiquinone radical (reaction R7, Scheme ). Additional proton-coupled electron-transfer transformations of coordinated semiquinone species are also possible (reactions R8–R10, Scheme ). The free o-semiquinone radical exists in equilibrium with its radical anion form (reaction R11, Scheme ). In addition, HO^• can directly oxidize catechol to form o-semiquinone radicals (reaction R12, Scheme ). These semiquinone intermediates can subsequently undergo oxidation through HO^•-mediated pathways (reaction R13, Scheme ) or Fe­(III)-mediated electron-transfer reactions (reaction R14, Scheme ) to yield o-quinone. Fe­(III)-mediated oxidation of catechol to semiquinone and quinone-type products is consistent with previous Fe­(III)-catechol studies under acidic conditions. , Likewise, ligand-to-Fe charge transfer and subsequent radical-mediated transformations have been proposed as important pathways in Fe­(III)-phenolic systems. Although multiple pathways are mechanistically feasible, the rapid depletion of H₂O₂ and the strong inhibitory effect of DMSO indicate that HO^•-driven oxidation predominates during the initial 0–2 min oxidative window analyzed kinetically, whereas Fe-catechol redox pathways are expected to contribute more significantly during the subsequent slower oxidation regime. Free Fe­(III) hydrolysis species are known to exhibit characteristic UV–visible absorption bands that are distinct from those of Fe­(III)-catechol complexes. , Additionally, at pH 3.0 and under high ligand-to-metal ratios, catechol reacts with Fe³⁺ to form mono-catecholate–Fe³⁺ complexes that absorb in the visible region of the spectrum, with λ_max values near 405 and 520 nm. , At pH 5.0–7.0, Fe³⁺ coordination favors the formation of bis- and tris-catecholate complexes, as summarized in Scheme . − ,

Effect of pH

Operational parameters such as pH, significantly influence AOP because pH affects the availability of ROS, the rate of reactions, and the type of oxidation products formed. To determine this influence, experiments were conducted at different environmentally relevant pH values in order to assess their impact on the production of intermediates, which subsequently determine the formation of HO^•. In Figure S8, experiments involving of 200 μM catechol and FR with ozone were monitored λ_max = 276 nm. The impact of pH on the oxidation of catechol at 276 nm is evident in Figure . A higher oxidative loss of catechol compared is observed at acidic pH values, corresponding to the relevant atmospheric pH of aerosols, fogs, and clouds.

5.

Effect of pH on k′ for the reaction of 200 μM catechol, 1 μM Fe²⁺, 50 μM H₂O₂ under 131 mL min^–1 flow of 400 ppbv O₃(g) at pH 1.0, 1.25, 1.50, 2.0, 2.50, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5 7.0, 8.0, and 9.0. Absorbance was monitored at λ_max = 276 nm, and rate constants (k′, min^–1) were obtained by fitting the initial 2 min oxidative window.

The optimal pH for Fenton reaction is between 1.8–3.0, arange that closely matches the acidity of many atmospheric waters including fog and cloud droplets. At lower pH levels, hydroxyl radicals are easily scavenged by protons and decreases Fe²⁺, resulting to a decreased reaction deficiency. From Figure , the transformation of catechol is more favorable in acidic condition, which is evident in the rate constant at pH 3 to be 1.69 × 10^–2 min^–1 which is 6-times greater than pH 7.0 at k′ = 2.71 × 10^–3 min^–1 and 9-timer higher than pH 9 at k′ = 1.76 × 10^–3 min^–1. According to Salgado et al., the formation of mono-complexes, bis-complexes, and tri-complexes is favored at the acidic, neutral, and basic pHs levels, respectively.

Moreover, when pH increases, the ortho position corresponding to the hydroxyl dissociation group will have a higher electronic density, implying that this position is more prone to the electrophilic substitution of ozone, because the substituting –O– group is more activating than the −OH group the protonation of catechol influences the formation of these oxidations. The UV–visible absorption spectra at different pH values are displayed in Figure S8 (Supporting Information). Although the UV–visible spectra remain broadly similar across the investigated pH range (Figure S8, Supporting Information), the extracted pseudo-first-order rate constants reveal a clear dependence of catechol oxidation on pH.

Effect of Ozone Concentration

Ozone plays a pivotal role as an essential oxidant in the degradation of organic pollutants. An increase in ozone concentration leads to a corresponding increase in the rate of oxidation. Conversely, a decrease in ozone concentration results in slower and less effective oxidative loss within AOPs. The concentration of ozone introduced into the reactor varied at different levels in Figure , specifically at 3.6, 2.8, 2.0, 1.2, and 0.4 ppmv sparged with a constant 131 mL min O₃(g) flow of fixed molar ratio. The corresponding UV–visible absorption spectra for these ozone concentrations are shown in Figure S15 (Supporting Information).

6.

Kinetic profiles for 200 μM catecholoxidation at pH 3.0 with 1.0 μM Fe²⁺, and 50.0 μM H₂O₂ under 131 mL min^–1 flow of O₃(g) yielding [O₃(aq)] of 4.4 nM (black circle), 13.2 nM (red square), 22.0 nM (blue triangle), 30.8 nM (pink star), and 39.6 nM (green diamond). (A) Normalized catechol decay at λ_max = 276 nm with inset showing extracted k′ versus [O₃(aq)]. (B) o-Quinone formation at λ_max = 389 nm with inset showing the rate of formation (k ₀). (C) Catechol-Fe complex formation at λ_max = 490 nm with inset showing k ₀ versus [O₃(aq)]. In all insets, the solid black line represents the linear regression, and the dashed teal lines denote the 95% confidence intervals. All rate constants were obtained by fitting data within the initial 0–2 min oxidative window.

Figure illustrates the relationship between ozone concentration, catechol oxidation and the formation of products. As [O₃(aq)] increases from 4.4 nM to 39.4 nM, k′ increases according to the relationship k′ (min^–1) = 1.57 × 10^–4 nM^–1 min^–1 × [O₃(aq)] + 1.58 × 10^–2 (min^–1) with a coefficient of determination r ² = 0.9740. The formation of o-quinone with k ₀ (min^–1) = 1.63 × 10^–1 min^–1 × [O₃(aq)] + 5.68 μM min^–1 (r ² = 0.9302) and catechol-Fe complex k ₀ (min^–1) = 8.93 × 10^–4 min^–1 × [O₃(aq)] + 1.63 × 10^–1 μM min^–1 (r ² = 0.9695). At 2 min, the transformation of catechol increases with higher aqueous [O₃(aq)] = 3.5% at 4.4 nM to 5% at 39.4 nM. The formation of o-quinone changed from 3.5% to 6.3% of the initial catechol concentration and 0.16% to 0.2% for catechol-Fe complex.

Effect of Ozone Flow Rate

The ozone flow influences oxidation, higher flow rates influence increase ozone mass transfer and the oxidation rate. Figure shows the pseudo-first-order decay of catechol relative to its initial concentration at different flow rates. The rate constants (k′) of the flow rates was determined by fitting the exponential decay function to each data set

7.

Pseudo-first-order decay of normalized catechol concentration ([cat] _t /[cat]₀) monitored at λ_max = 276 nm for experiments with 1.0 μM Fe²⁺, and 50.0 μM H₂O₂ at pH 3.0 under variable O₃(g) flow rates of 131 mL min^–1 (black circle), 238 mL min^–1 (blue triangle), 329 mL min^–1 (pink star), and 413 mL min^–1 (green diamond). The inset shows the dependence of the extracted rate constant (k′, min^–1) on O₃(g) flow rate, including the linear regression (solid black line) and the 95% confidence interval (red dashed lines). All k′ values were obtained by fitting concentration–time data within the initial 0–2 min oxidative window.

The higher the flow rate, the faster the transformation of catechol and this predominantly occurs in the gas-liquid interface. , As the flow increases from 131 to 413.0 mL min^–1, the percentage of loss of catechol from 3.5% to 5%. Similarly, the inset shows a strong correlation (r ² = 0.977) between k′ and ozone flow rate with a linear relationship of k′ (min^–1) = 2.11 × 10^–7 mL^–1 × flow (mL min^–1) + 2.57 × 10^–4 (min^–1).

Effect of Reactant Concentration

The concentration of reactants influences reaction kinetics and oxidation efficiency. At fixed oxidant levels, increasing catechol concentration leads to a higher organic-to-oxidant ratio, which can reduce fractional catechol conversion due to increased competition for a finite flux of HO^•. Accordingly, Figure S11 shows that catechol transformation is highest at 50 μM, followed by 100 and 200 μM, with progressively lower relative oxidation observed at 300 μM and 500 μM. This trend is further reflected in the corresponding pseudo-first-order rate constants shown in Figure S12. Because reported catechol concentrations in atmospheric waters impacted by biomass burning range from 22 to 223 μM, the concentrations investigated here span environmentally relevant conditions up to the upper atmospheric limit.

The corresponding UV–visible spectra for experiments performed at different H₂O₂ concentrations are shown in Figure S13 (Supporting Information), and the associated kinetic profiles are shown in Figure S14. In addition, the concentrations of Fe²⁺ and H₂O₂ play critical roles in determining oxidative efficiency. Fe²⁺ catalyzes H₂O₂ decomposition to generate HO^•, thereby controlling the radical production rate. As shown in Figure S9, an Fe²⁺ concentration of 25 μM yields the highest catechol oxidation and the largest concentrations of o-quinone and catechol–Fe complexes (Figure S10). At higher Fe²⁺ concentrations, self-scavenging reactions increasingly suppress HO^• availability, resulting in diminished oxidative efficiency.

Environmental Implications

The ozone- and Fenton-driven oxidation of catechol illustrates how coupled oxidant systems strongly influence chemical processing within atmospheric aqueous phases, including fog droplets, cloud water, and aerosol-bound aqueous layers. In these environments, catechol is a key precursor to brown carbon, and its oxidative transformation governs aerosol aging, optical properties, and the bleaching of light-absorbing aqueous secondary organic aerosol (aqSOA). As a representative phenolic compound emitted from biomass burning and industrial activities, catechol undergoes multiphase reactions that directly affect its atmospheric lifetime, reactivity, and contribution to aerosol toxicity.

Present results show that the hybrid O₃/FR process substantially accelerates catechol oxidation primarily through hydroxyl-radical production arising from Fe²⁺/H₂O₂ redox cycling during the initial oxidative window, while additional Fe­(III)-mediated and ozone-mediated pathways contribute to subsequent oxidation and product evolution. The efficiency of this transformation is governed by environmentally relevant parameters such as pH, reactant concentrations, and ozone exposure. Acidic conditions (pH ≈ 1.5–3.0), characteristic of deliquesced aerosol but also possible in certain fog and cloud droplets, enhance classical Fenton chemistry and maximize HO^• production, whereas increasing pH reduces iron solubility and slows oxidation. Similarly, variations in ozone concentration and gas–liquid mass transfer regulate the steady-state oxidant flux, influencing catechol residence times and the formation of o-quinone and catechol–Fe complexes within the short, kinetically significant oxidative window. Together, these findings demonstrate that the balance between brown carbon formation and bleaching, the rate of aqSOA aging, and the persistence of phenolic species in the atmosphere are tightly linked to coupled oxidant chemistry in atmospheric aqueous phases.

Glossary

Abbreviations

AOP

advanced oxidation process

aqSOA

aqueous secondary organic aerosol

BrC

brown carbon

Cat

catechol (1,2-dihydroxybenzene)

DMSO

dimethyl sulfoxide

FR

Fenton reagent

HO^•

hydroxyl radical

H₂O₂

hydrogen peroxide

HPLC

high-pressure liquid chromatography

O₃

ozone

O₃(aq)

aqueous ozone

PB

Prussian Blue

QH₂

catechol

ROS

reactive oxygen species

SPCE

screen-printed carbon electrode

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsestair.6c00223.

• Additional experimental methods information with Figures S1–S15 and Table S1 (PDF)

The authors declare no competing financial interest.