Degradation of roxarsone in a sulfate radical mediated oxidation process and formation of polynitrated by-products

Abstract

Organoarsenicals such as roxarsone (ROX) are extensively utilized in the poultry industry, and land application of poultry litter is an important route by which arsenics are introduced into the environment. In the present study, degradation of ROX and structurally related nitrophenols by a heat activated persulfate (PS) oxidation process, one in situ chemical oxidation process (ISCO), was systematically explored. The effects of temperature, PS dosage, pH and natural water constituents (i.e., fulvic acid, Cl⁻) on the degradation of ROX were investigated. Products analysis by solid phase extraction (SPE) and liquid chromatography-electrospray ionization-triple quadrupole mass spectrometry (LC-ESI-MS/MS) revealed that 2,4-dinitrophenol (2,4-DNP) and 2,4,6-trinitrophenol (2,4,6-TNP) were generated as major intermediate products, suggesting denitration–renitration process occurred during SO₄˙⁻-based oxidation of ROX. Interestingly, the formation of polynitration by-products was further confirmed in heat activated persulfate oxidation of nitrophenols. Formation of inorganic arsenics during ROX degradation was measured by inductively coupled plasma-mass spectrometry (ICP-MS). It was evident that the arsenic substituent of ROX was converted to As(V). On the basis of the intermediate products identified, detailed mechanisms and transformation pathways for ROX oxidation were proposed. Results manifest that heat activated PS oxidation could be an efficient approach to treat ROX contamination. However, post-treatment is necessary for complete removal of As(V) to minimize ecotoxicological risks.

---

Introduction

Organoarsenicals, such as roxarsone (ROX, 3-nitro-4-hydroxyphenylarsonic acid), are widely utilized as feed additives in the broiler poultry industry to promote growth and control coccidial intestinal parasites.¹ Undesirably, organoarsenicals are known to have ecological toxicity and may pose adverse effects to microorganisms, animals, plants, and human beings when they are present in the environment.² Currently, application of organoarsenicals in the livestock industry has been banned in developed countries; however, they are still popular in developing countries such as China and India. ROX and 4-aminophenyl arsonic acid (p-ASA) are the most widely used organoarsenicals in China.³ ROX is poorly metabolized in the animal body with approximately 95% excreted as original form via urine and faces.^2,3 During litter storage and composting, ROX can release inorganic arsenics via biotic and/or abiotic transformation pathways.^4–7 Land application of poultry litters serves as an important route by which arsenics are introduced into the environment.^3,8 Therefore, poultry litter containing organoarsenicals can cause serious arsenic contamination of soil and groundwater.^2,3,9,10 It is imperative to develop effective and sustainable treatment technologies to removal ROX and minimize associated risks.

Advanced oxidation processes (AOPs) depending on highly reactive radicals (e.g., hydroxyl radical (HO˙) and sulfate radical (SO₄˙⁻)) are promising technologies for water treatment and environmental remediation.¹¹ HO˙ is produced by conventional AOPs, such as Fenton, UV/Fenton, UV/H₂O₂ and O₃/H₂O₂, etc.,^11,12 while SO₄˙⁻ can be generated through activating persulfate (PS) or peroxymonosulfate (PMS) by heat, UV radiation, base, transit metals and catalysts.^13–16 The redox potential of SO₄˙⁻ and HO˙ was reported to be 2.5–3.1 and 1.8–2.7 V (depending on the solution pH), respectively, which was much higher than that of other oxidants (e.g., O₃, HOCl, and KMnO₄) commonly applied for water treatment and disinfection.¹⁷ SO₄˙⁻ has a longer lifetime (300 μs vs. 40 μs for HO˙) and is more selective than HO˙ because it reacts with organic contaminants mainly through electron transfer mechanism while HO˙ can participate in a variety of reactions with equal preference.^13,18 In addition, high aqueous solubility and stability of PS and PMS allow these precursors to survive longer and thus be transported to longer distance in subsurface.^13,18 These advantages make SO₄˙⁻-based AOPs (SR-AOPs) a promising in situ chemical oxidation (ISCO) technology, especially for soil and subsurface remediation.^13,18

Among various activation approaches, heat activation bears certain advantages. For example, it requires no additional chemicals, which minimizes the consumption of PS during premixing with catalysts (e.g., Fe²⁺ and Co²⁺) and avoids secondary pollution.¹³ Heat activated process can be easily manipulated by varying temperature.¹⁹ Heating not only accelerates PS decomposition and radical formation, but also enhances the overall reactions according to thermodynamic law.^13,18 However, in practical application, heating large volume of groundwater may be energy-consuming, thus heat activation is often combined with in situ thermal remediation (ISTR).¹³ Heat activated PS process has been demonstrated to be effective to oxidize herbicides,^20,21 industrial chemicals,^22–24 and pharmaceuticals.^25–27 However, ROX degradation in heat activated PS oxidation process has not been investigated so far, although its degradation by HR-AOPs (e.g., photocatalytic oxidation) has been reported recently.^28,29

This study was designed to explore the feasibility of employing heat activated PS process to degrade ROX in aqueous solution. Factors governing ROX degradation, such as temperature, PS concentration, pH, natural organic matter (NOM), and Cl⁻ were investigated. The degradation products were analyzed by SPE-HPLC-MS/MS and ICP-MS. Based on the molecular structures of the intermediates and products, detailed mechanisms and transformation pathways for ROX oxidation were proposed. Particularly, the transformation and fate of nitro and arsenic groups of ROX during heat activated PS oxidation process was elucidated. In addition, the degradation of ROX was compared with several structurally related nitrophenols to elucidate insights into the correlations between molecular structure and reactivity. Interestingly, we observed the formation of more toxic and recalcitrant polynitrated by-products during SO₄˙⁻-based oxidation of ROX and nitrophenols. This finding sheds light on the potentially risks associated with the treatment of nitroaromatic compounds by SO₄˙⁻-based oxidation processes.

Materials and methods

Reagents and materials

All chemicals were of reagent grade or higher purity and used without further purification. Roxarsone (ROX, 99.0%), 2-nitrophenol (2-NP), 2,4-dinitrophenol (2,4-DNP), 2,6-dinitrophenol (2,6-DNP), 2,4,6-trinitrophenol (2,4,6-TNP), 4-hydroxy-3-nitrobenzoic acid (3N4HBA) and 4-chloro-2-nitrophenol (4Cl2NP) were obtained from Sigma-Aldrich (St, Louis, MO). Potassium persulfate (K₂S₂O₈, 99.5%) was purchased from Aladdin (Shanghai, China). HPLC grade methanol and formic acid were purchased from Tedia (Fairfield, OH). All stock solutions were prepared by dissolving the chemical agents in Milli-Q water (resistance of 18 MΩ cm⁻¹) prepared by a Millipore system (Bedford, USA) and stored at 4 °C. Suwannee River Fulvic Acid (SRFA, 2S101F) was obtained from the International Humic Substances Society (IHSS, University of Minnesota, St. Paul, MN). Oasis hydrophilic–lipophilic balance (HLB) solid phase extraction (SPE) cartridges (6 cm³/200 mg, WAT106202) were purchased from Waters Corporation (Milford, MA).

Experimental setup

Batch experiments were conducted in 33 mL screw-cap EPA vials with Teflon septa at predetermined temperature (40–70 °C) controlled by a thermo-stated water bath (Xianou Instrument Manufacture Co., Ltd, Nanjing). Prior to heat activation, appropriate volume of ROX stock solution (1 mM) and PS stock solution (100 mM) were transferred into the vials to achieve a total of 20 mL reaction solution with predefined concentration of reactants (e.g., 50 μM ROX and 2 mM PS). The degradation of nitrophenols, i.e., 2-NP, 3N4HBA, and 4Cl2NP, were performed under identical conditions. Initial pH was adjusted by 0.01 M H₂SO₄ or NaOH to desired value. No buffer was used to avoid potential reactions between buffer species and SO₄˙⁻. Control experiments with substrates alone or substrates with PS at ambient temperature were performed concomitantly to assess the loss of ROX by means other than oxidative degradation. To investigate the influence of SRFA and Cl⁻ on ROX degradation, different amounts of Cl⁻ (0–200 mM) or SRFA (0–10 mg L⁻¹) were added into the reaction solution. Aliquots (0.5 mL) were withdrawn at predetermined time intervals and chilled in an ice bath for 10 min to stop the reaction and kept at 4 °C until analysis. All the experiments were carried out in duplicates to assure accurate data acquisition and error bars in figures represent the standard deviation.

For products identification, 100 mL solution containing 50 μM ROX and 2 mM PS was allowed to react for 7 h at 60 °C before being chilled and enriched using Waters Oasis HLB SPE cartridge. Detailed procedures are given in Text S1, ESI.†

Analytical methods

Concentrations of ROX and nitrophenols were measured by a Hitachi L-2000 high performance liquid chromatography (HPLC, Hitachi) equipped with an L-2455 diode array detector. Detailed operation conditions were provided in Table S1, ESI.† Quantification was based on multipoint standard calibration curves.

ROX degradation products were analyzed using an Agilent 1200 HPLC coupled with a 6410B triple quadrupole mass spectrometer with electrospray ionization source (LC-ESI-MS/MS, Agilent Technologies, USA). Separation was accomplished using an Agilent Zorbax Eclipse Plus C18 column (3.5 μm, 2.1 mm I.D. × 150 mm). Full scans and product ion scans were conducted to determine the quasi-molecular ions and elucidate the structures of major transformation products. Detailed operational conditions and parameters are presented in Text S2, ESI.†

Formation of arsenite and arsenate was quantified using a Perkin-Elmer Flexar ion chromatography (IC) coupled with a Perkin-Elmer NexION 300× quadrupole-based ICP-MS. Separation was carried out using a Hamiton PRP-X100 anion exchange column (4.6 mm I.D. × 250 mm). An isocratic mobile phase comprised of 8.5 mM NH₄NO₃ and 8.5 mM (NH₄)₂HPO₄ (pH 6.0) was used at a flow rate of 1.2 mL min⁻¹. Germanium was added to the post-column solution as the internal standard. ICP-MS was set up in the He gas collision mode to minimize polyatomic interferences. Arsenic species in the samples were identified by comparing their retention times with those of the standards and quantified by using external calibration curves with peak areas.

Solution pH was measured by a combined glass electrode (E-201-C, Leici) connected to a PHS-3CW microprocessor pH/mV meter (BANTE instrument).

Results and discussions

Reaction kinetics and effects of temperature

Decomposition of PS molecule via cleavage of the peroxide bond generates two SO₄˙⁻ radicals when the activation energy is greater than the dissociation energy of peroxide bond (i.e., 34 kcal mol⁻¹).^13,30 This process can be heat-driven, and high temperature increases the rate of activation dramatically, resulting in drastic oxidizing conditions.^30,31 The high temperature also thermodynamically promotes the chemical reactions between reactive species and contaminants.¹³ In the present study, increasing temperature significantly increased the removal of ROX, as illustrated in Fig. 1A. The removal of ROX increased from 10.6% to 98.2% in 8 h as the temperature increased from 40 to 70 °C. Under each temperature investigated, ROX degradation could be fitted by pseudo-first-order kinetic model described by eqn (1).where [ROX] is the concentration of ROX at reaction time t, k_obs is the pseudo-first-order rate constant which can be obtained by linear regression of the plot of ln ([ROX]_t/[ROX]₀) versus t. It was observed that k_obs increased 28-folds when the temperature was elevated from 40 to 70 °C. It is illustrated in Fig. 1B that a linear relationship between ln k_obs and 1/T can be established, suggesting that the temperature-dependence of k_obs can be evaluated using Arrhenius equation as quoted in eqn (2).where A is the pre-exponential factor, E_a is the apparent activation energy, R is the universal gas constant (8.314 J mol⁻¹ K⁻¹), and T is the absolute temperature. Accordingly, the apparent activation energy of ROX degradation was determined to be 102.4 ± 0.6 kJ mol⁻¹. This value was similar to that of trichloroethene (108 ± 3 kJ mol⁻¹),³¹ but comparatively lower than that of other contaminants such as naproxen (155.03 ± 26.4 kJ mol⁻¹) and sulfamethoxazole (119.6 kJ mol⁻¹).^26,27

Fig. 1 (A) Effects of temperature on ROX degradation in heat activated PS oxidation process; (B) plot of ln(k_obs) versus T⁻¹. Experimental conditions: [ROX]₀ = 50 μM, [PS]₀ = 2.0 mM, T = 40–70 °C, pH = 7.0, V = 20 mL, and reaction time = 480 min. Error bar represents the standard deviation of 2 replicates.

Effects of PS concentration

It is generally accepted that reactive radicals (e.g., SO₄˙⁻) generated from heat activated PS are dominant oxidizing species responsible for the degradation of organic contaminants.^13,30 Although excess PS may scavenge SO₄˙⁻ (eqn (3)),¹⁷ high PS dosage was reported to facilitate the degradation of contaminants due to high level of radicals produced.^23,31 It is revealed that increasing the concentration of PS markedly enhanced the degradation of ROX (Fig. 2A), consistent with previous studies.^23,31 When the initial PS concentration was increased from 0.5 to 4.0 mM, k_obs of ROX degradation increased linearly from 0.00225 to 0.0121 min⁻¹. Correspondingly, the half-life time of ROX decreased from 308 to 60 min. It is evident from Fig. 2B that the k_obs is proportional to the concentration of PS. k_obs represents the overall degradation of ROX. It is the sum of the product of second-order rate constant of each oxidizing species (e.g., SO₄˙⁻, HO˙ and S₂O₈²⁻ (E⁰ = 2.01 V)) with respective steady-state concentration (eqn (4)). Thus, there is a linear relationship between PS concentration and the steady-state concentrations of oxidizing species, as proposed in a previous study.³²

SO₄˙⁻ + S₂O₈²⁻ → SO₄²⁻ + S₂O₈˙⁻, k = 1.2 × 10⁶ M⁻¹ s⁻¹ (3)

where [Ox]_i is the steady-state concentration of oxidizing species i; k_i is the second-order rate constant for ROX reaction with species i; α_i is the yield of species i (e.g., SO₄˙⁻) by heat activated PS decomposition at a given temperature.

Fig. 2 (A) Effects of PS concentration on ROX degradation in heat activated PS oxidation process; (B) plot of k_obs versus PS concentration. Experimental conditions: [ROX]₀ = 30 μM, [PS]₀ = 0.5–4 mM, T = 60 °C, pH = 7.0, V = 20 mL, and reaction time = 120 min. Error bar represents the standard deviation of 2 replicates.

Effects of pH

Solution pH plays a complicated role in SR-AOPs because it can influence the formation and transformation of SO₄˙⁻ as well as speciation of the target compounds. For instance, acid can catalytically decompose PS without generation of radicals,³³ while base activates PS to generate SO₄˙⁻.¹⁵ Conversion of SO₄˙⁻ to HO˙ may also occur at basic pH (eqn (5)).^23,31

SO₄˙⁻ + OH⁻ → HO˙ + SO₄²⁻, k = (6.5 ± 1.0) × 10⁷ M⁻¹ s⁻¹ (5)

The effects of pH on the degradation of ROX were shown in Fig. 3. It appeared that k_obs increased gradually with an increase in pH. The removal of ROX after 8 h reaction was 42% and 52% at pH 7 and 10, respectively. The enhanced degradation of ROX at elevated pH can be rationalized by the speciation of ROX as well as the conversion of SO₄˙⁻ to HO˙. ROX has three exchangeable protons located at arsenic and hydroxyl groups, corresponding to pK_a of 3.45, 5.95, 9.15, respectively.³⁴ At pH above pK_a1 (2.56), deprotonation occurs at the arsenic group. The negative charge at the arsenic oxygen can partially delocalize to the aromatic ring thus increases electron density of the ring. Similarly, when pH is above pK_a3 (9.7), deprotonation occurs at phenolic group, which further increases the electron density of the ring by delocalizing the lone pair electrons of phenolic oxygen. Thus, increasing pH facilitated the electrophilic attack of SO₄˙⁻ and/or HO˙, thus promoted the degradation of ROX.

Fig. 3 Pseudo-first-order rate constant for ROX degradation as a function of pH in heat activated persulfate oxidation process. Experimental conditions: pH = 2.5–10, [ROX]₀ = 50 μM, [PS]₀ = 2.0 mM, T = 60 °C, V = 20 mL, reaction time = 480 min. Error bar represents the standard deviation of 2 replicates.

At neutral pH, SO₄˙⁻ has been confirmed to be the major oxidizing species involved in heat activated PS system by radical quenching and electron paramagnetic resonance (EPR) studies.^35,36 When solution pH is greater than 9.0, HO˙ will become dominant.^35,36 Mechanisms for reactions of SO₄˙⁻ and HO˙ with organic compounds are difference. SO₄˙⁻ is known to react with organic compounds primarily through electron transfer,³⁷ whereas HO˙ reacts preferentially via addition to C=C double bonds and abstracting H from C–H, N–H, or O–H bonds.¹¹ Therefore, pH can affect the reactions by altering the composition of oxidizing species in the solution as well as the mechanisms of pollutants degradation. The disparate reaction mechanism is also likely being responsible for the faster degradation of ROX observed at alkaline pH.

Effects of natural water constituents

Chloride (Cl⁻) and NOM are constituents ubiquitously present in natural waters. The presence of these species can decrease the degradation efficiency of contaminants in AOPs by acting as radical scavengers.^38,39 The scavenging of SO₄˙⁻ and/or HO˙ by these constituents may generate secondary reactive species which may be involved in the oxidation of targeted contaminants as well.⁴⁰ Therefore, influence of Cl⁻ and SRFA (as a representative of aquatic NOM) on the degradation of ROX in heat activated PS process were examined in this work. Result shows that Cl⁻ at concentrations of 5 and 10 mM inhibited the degradation of ROX, as shown in Fig. 4A, presumably due to the scavenging of SO₄˙⁻ and HO˙ by Cl⁻ (eqn (6) and (7)). However, it was observed that Cl⁻ at higher concentrations (100 and 200 mM) slightly promoted the degradation of ROX. This may be due to the accumulation of reactive chlorine species (RCS), which offset the loss of SO₄˙⁻ or HO˙ by Cl⁻ scavenging. It was reported that halides can be oxidized to reactive halogen species in SR-AOPs.^41,42 For example, Cl⁻ can be converted to chlorine atom radical (Cl˙) by single electron oxidation of SO₄˙⁻, which initiates chain propagation reactions, leading to the formation of a series of RCS including radical chlorine species (Cl₂˙⁻ and ClOH˙⁻) and free chlorine (Cl₂/HOCl) (eqn (6)–(9)).^22,40,43 These reactive chlorine species are moderate oxidants, but can selectively react with electron-rich compounds such as phenols, leading to the formation of chlorinated compounds (eqn (10) and (11)).^44,45

SO₄˙⁻ + Cl⁻ ↔ SO₄²⁻ + Cl˙, E⁰(SO₄˙⁻/SO₄²⁻) = 2.6 V (6)

Cl˙ + Cl⁻ ↔ Cl₂˙⁻, E⁰(Cl˙/Cl₂˙⁻) = 2.41 V (7)

Cl₂˙⁻ + Cl₂˙⁻ → Cl₂ + 2Cl⁻, E⁰(Cl₂/Cl⁻) = 1.36 V (8)

Cl_2(aq) + H₂O → HOCl + Cl⁻ + H⁺, E⁰(HOCl/Cl⁻) = 1.48 V (9)

R + Cl₂˙⁻ → R–Cl + Cl⁻ (10)

R–H + HOCl → R–Cl + H₂O (11)

Fig. 4 Effects of (A) Cl⁻ and (B) SRFA on ROX degradation in heat activated PS oxidation process. Experimental conditions: [ROX]₀ = 50 μM, [PS]₀ = 2.0 mM, T = 60 °C, pH = 7.0, V = 20 mL, and reaction time = 480 min. Error bar represents the standard deviation of 2 replicates.

Since ROX contains phenolic structure, RCS likely participated in its oxidation. It is presumed that, at low Cl⁻ concentration, the RHS generated was unlikely to compensate the reduced availability of SO₄˙⁻/HO˙ caused by Cl⁻ scavenging, thus showing inhibitory effect. In the presence of high concentration of Cl⁻, however, the steady-state concentration of RCS could be orders of magnitude higher than that of SO₄˙⁻,⁴¹ which could not only offset the negative effect of SO₄˙⁻/HO˙ scavenging, but also promote the degradation of ROX.

The RHS produced in SR-AOPs are reactive with organic compounds in aqueous solution and proceed by H-abstraction, addition to alkenes and alkynes, and one-electron oxidation pathways.¹¹ H-abstraction and one electron oxidation is beneficial to the goal of mineralization, while addition forms halogenated compounds.¹¹ Previous studies have demonstrated that SO₄˙⁻-based oxidation of phenolic compounds in the presence of Cl⁻ produced chlorinated by-products.^40,44,45 In this work, however, no chlorinated by-products were identified by either HPLC or LC-MS/MS analysis, indicating probably that RCS was involved in the oxidation of ROX via H-abstraction and/or one electron oxidation pathways.

NOM was shown to play a negative role in SR-AOPs in previous studies.^27,39,46 Since the electron-rich moieties (e.g., phenolic) in NOM molecules are readily to react with electrophilic radicals including SO₄˙⁻ and HO˙, NOM usually serves as a radical sink in AOPs system.^27,39,46 In the present study, SRFA was found to suppress the degradation of ROX, and such inhibition was become more prominent at higher SRFA concentration (Fig. 4B). These results were consistent with the radical quenching theory.^27,39,46 The interactions between SO₄˙⁻ and NOM has previously been explored by David Gara et al., and formation of H-bonded adducts between SO₄˙⁻ and NOM was supported by density functional theory (DFT).⁴⁷

NOM may also inhibit the oxidation of organic compounds by adsorption and complexation, making targeted contaminants less available for radical attack.^13,30 In addition, it was reported that NOM could act as antioxidants of phenoxyl radicals.^48,49 SO₄˙⁻-based oxidation of phenolic compounds may generate phenoxyl radical.⁴⁷ In the present study, the phenoxyl radical of ROX produced by SO₄˙⁻ attack might be reduced back to ROX by reaction with SRFA molecule. However, this hypothesis warrants further studies.

Intermediate products and transformation mechanisms

Intermediate products generated in heat activated PS oxidation of ROX were analyzed by SPE-LC-MS/MS. The total ion chromatograph (TIC) of the concentrated reaction solution depicts two possible intermediate products (Fig. 5). They were determined to correspond to 2,4-DNP (R_t = 11.5 min) and 2,4,6-TNP (R_t = 18.0 min) by further comparing their retention times and product ion spectra (MS/MS) with those of authentic standards (Fig. S1 and S2, ESI†).

Fig. 5 Total ion chromatograph (TIC) of ROX degradation products concentrated by SPE revealed the formation of two major intermediate products, 2,4-dinitrophenol (2,4-DNP) and 2,4,6-trinitrophenol (2,4,6-TNP). Insets show the mass spectra of the two polynitrated by-products. Detailed SPE procedures were provided in Text S1, ESI.† Chromatographic separation conditions were provided in Text S2, ESI.† MS analysis condition: ESI(+), fragmentor voltage, 135 V; scan range, 50–300 m/z. Structural assignment of the two products were further supported by the product ion scan mass spectra of their authentic standards, which was presented in Fig. S1 and S2, ESI.†

Considering that there is only one nitro group within ROX molecule and no other N-bearing species present in the solution, the formation of 2,4-DNP and 2,4,6-TNP implies that both denitration and renitration processes occurred.⁵⁰ Similar formation of polynitrated products has previously been observed in HO˙-based AOPs.^51–53 For example, Carlos et al. reported the formation of 1,3-dinitrobenzene during the degradation of nitrobenzene in Fenton and Fenton-like processes.^52,53 More recently, Zhang et al. observed the formation of 2,4-DNP during the degradation of 4-nitrophenol in heat and transit metal co-activated persulfate system.⁵⁴ We hypothesized that nitrating agents such as NO₂˙ and NO₃˙ are responsible for the formation of polynitrated aromatics.^51,53,55 In the present study, NO₂˙ or NO₃˙ could be generated by SO₄˙⁻ oxidation of NO₂⁻ or NO₃⁻ (eqn (12) and (13)) which was released from ROX molecule upon SO₄˙⁻ attack.⁵¹ Attack of SO₄˙⁻ on the aromatic ring of ROX could generate NO₂˙ directly as well.⁵⁰ Further oxidation of 2,4-DNP by SO₄˙⁻ and nitrating agents led to the formation of 2,4,6-TNP. Based on the above analysis, pathways of 2,4-DNP and 2,4,6-TNP formation in heat activated persulfate oxidation of ROX can be proposed (Fig. 6).

SO₄˙⁻ + NO₂⁻ → SO₄²⁻ + NO₂˙ k = 8.8 × 10⁸ M⁻¹ s⁻¹ (12)

SO₄˙⁻ + NO₃⁻ → SO₄²⁻ + NO₃˙ k = 2.1 × 10⁰ M⁻¹ s⁻¹ (13)

Fig. 6 Proposed pathways of ROX degradation in heat activated PS oxidation process.

Transformation of arsenic substituent

Inorganic arsenic species generated upon heat activated PS oxidation of ROX was monitored by ICP-MS. Formation of inorganic arsenic was observed in biotic and abiotic degradation of ROX.^28,56 The cleavage of As–C bond occurred during HO˙-mediated degradation of organoarsenicals, generating As(IV) which disproportionated to yield 1 : 1 ratio of As(III) and As(V).⁵⁷ As(III) can be oxidized to As(IV) by donation of an electron to SO₄˙⁻ (eqn (14)).^58–60 Further oxidation of As(IV) to As(V) occurs by reacting with another SO₄˙⁻ or dissolved oxygen, as described in eqn (15) and (16).^57,58 Moreover, HO˙ generated by reaction of SO₄˙⁻ with H₂O can mediate the oxidation of As(III) to As(IV) as shown in eqn (17) and (18).

As(III) + SO₄˙⁻ → As(IV) + SO₄²⁻ (14)

As(IV) + SO₄˙⁻ → As(V) + SO₄²⁻ (15)

As(IV) + O₂ → As(V) + O₂˙⁻ (16)

H₂O + SO₄˙⁻ → HO˙ + H⁺ + SO₄²⁻ (17)

As(III) + HO˙ → As(IV) + HO⁻ (18)

Thus, As(V) appeared to be the ultimate form with respect to the transformation of arsenic in heat activated PS system,^57–59 which was verified by the experimental observation. As seen in Fig. 7, As(V) was the dominant inorganic arsenic species formed during ROX degradation while As(III) was negligible. Mass balance analysis shows that the sum of inorganic As(III) + As(V) accounted for 90% of the removed ROX, indicating that organic arsenic substituent was mostly transformed to inorganic arsenic species, predominantly As(V). Detailed transformation pathway of arsenic is also presented in Fig. 6. Since the mobility and toxicity of As(V) is lower than that of As(III), heat activated PS process is expected to detoxify ROX-containing water or soil. As(V) can be removal from water by adsorption or coagulation.⁶¹ Therefore, post-treatment such as iron oxide adsorption is desired to completely eliminate the toxicity of ROX.

Fig. 7 Evolution of inorganic arsenics during heat activated PS oxidation of ROX. Experimental conditions: [ROX]₀ = 50 μM, [PS]₀ = 2.0 mM, T = 60 °C, pH = 7.0, V = 20 mL, and reaction time = 480 min.

Degradation kinetics and products of structurally related nitrophenols

Reactivity of SO₄˙⁻ with organic compounds strongly depends on the structural characteristics of the target compound, such as type, number, and positions of substituents.^17,37 In order to elucidate more insights of ROX oxidation, the degradation of ROX was compared with that of structurally related compounds, including 2-nitrophenol (2NP), 3-nitro-4-hydroxybenzoic acid (3N4HBA), and 4-chloro-2-nitrophenol (4Cl2NP). These compounds all contain an orth-nitrophenol structure but have different para-substituents. As shown in Fig. 8, their degradation rates follow the order of 2NP > 4Cl2NP > 3N4HBA > ROX. SO₄˙⁻ is an electrophilic reagent that preferentially attacks sites with high electron density.^17,37 The result of Fig. 8 confirms that electron-withdrawn substituents at para site (i.e., –AsO₃H₂, –COOH, –Cl) reduce the electron density of the aromatic ring due to inductive effect, making the ring less favored for SO₄˙⁻ attack.³⁷ The inhibitory effect of the para-substituents follows the order of –AsO₃H₂ > –COOH > –Cl.

Fig. 8 Measured pseudo-first-order rate constant for degradation of ROX and structurally related analogs 2-nitrophenol (2-NP), 3-nitro-4-hydroxybenzoic acid (3N4HBA), and 4-chloro-2-nitrophenol (4Cl2NP) by heat activated persulfate oxidation process. Experimental conditions: [Substrates]₀ = 50 μM, [PS]₀ = 2.0 mM, T = 60 °C, pH = 7.0, V = 20 mL, and reaction time = 480 min. Data was the mean value of duplicates.

Results presented herein clearly showed that the presence of arsenic and nitro substituents in ROX molecule could inactive the aromatic ring, thus exhibiting deleterious effects. Given the fact that 2-NP is substantially more susceptible for SO₄˙⁻ attack than ROX, we assumed that the detachment of arsenic group is the rate-determining step during ROX degradation. To better understand the degradation of nitro-containing compounds by SO₄˙⁻-based oxidation processes, the transformation products of 2-NP, 3N4HBA and 4Cl2NP were analyzed by LC-MS/MS as well. It was noticed that 2-NP, 3N4HBA and 4Cl2NP all led to the formation of polynitrated by-products such as 2,4-dinitrophenol, 2,6-dinitrophenol, and 4-chloro-2,6-dinitrophenol (Fig. S3, ESI†). This suggests that denitration–renitration reactions are prevalent during SO₄˙⁻-based oxidation of nitrophenolic compounds, which was unrecognized before. Polynitrated aromatics are usually more persistent and toxic than mononitrated aromatics, and some of them are listed as the priority pollutants by regulatory agencies.⁶² Thus, particular attention should be paid to their formation during SO₄˙⁻-based oxidation of nitroaromatics.

Conclusions

Heat activated PS oxidation of ROX was systematically investigated in aqueous solution. Experimental results showed that the degradation of ROX followed pseudo-first-order kinetics. The degradation was accelerated by increasing the temperature or PS dosage. ROX could be effectively degraded over the pH range of 2.5 to 10, and pseudo-first-order rate constant (k_obs) increased with increasing pH. The presence of Cl⁻ at low concentration suppressed ROX degradation, while at high concentration promoted the degradation. SRFA inhibited the degradation of ROX by serving as a radical scavenger. 2,4-DNP and 2,4,6-TNP were identified by LC-MS/MS technique as by-products of ROX degradation. Their formation suggests that both denitration and renitration reactions occurred during heat activated PS oxidation of ROX. Such formation of polynitrated by-products was also observed in SO₄˙⁻-based oxidation of other nitrophenolic compounds which was unrecognized before. ICP-MS analysis showed that the organic arsenic was mainly converted to inorganic As(V). Thus, post-treatment of As(V) is required to completely eliminate the risks of ROX.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21607077), Fundamental Research Funds for the Central Universities (KJQN201741), Natural Science Foundation of Jiangsu Province, China (BK20160709). The content of the paper does not necessarily represent the views of the funding agencies.

References

1. K. P. Mangalgiri, A. Adak and L. Blaney, Environ. Int., 2015, 75, 68–80.
2. D. J. Fisher, L. T. Yonkos and K. W. Staver, Environ. Sci. Technol., 2015, 49, 1999–2012.
3. X. Liu, W. Zhang, Y. Hu, E. Hu, X. Xie, L. Wang and H. Cheng, Chemosphere, 2015, 119, 273–281.
4. A. J. Garbarino, J. R. Bednar, D. W. Rutherford, R. S. Beyer and R. L. Wershaw, Environ. Sci. Technol., 2003, 37, 1509–1514.
5. I. Cortinas, J. A. Field, M. Kopplin, J. R. Garbarino, A. Jay Gandolfi and R. Sierra-Alvarez, Environ. Sci. Technol., 2006, 40, 2951–2957.
6. J. F. Stolz, E. Perera, B. Kilonzo, B. Kail, B. Crable, E. Fisher, M. Ranganathan, L. Wormer and P. Basu, Environ. Sci. Technol., 2007, 41, 818–823.
7. A. J. Bednar, J. R. Garbarino, I. Ferrer, D. W. Rutherford, R. L. Wershaw, J. F. Raville and T. R. Wildeman, Sci. Total Environ., 2003, 302, 237–245.
8. D. W. Rutherford, A. J. Bednar, J. R. Garbarino, R. Needham, K. W. Staver and R. L. Wershaw, Environ. Sci. Technol., 2003, 37, 1515–1520.
9. B. P. Jackson, J. C. Seaman and P. M. Bertsch, Chemosphere, 2006, 65, 2028–2034.
10. J. H. Huang, K. N. Hu and B. Decker, Water, Air, Soil Pollut., 2011, 219, 401–415.
11. J. J. Pignatello, E. Oliveros and A. MacKay, Environ. Sci. Technol., 2006, 36, 1–84.
12. M. Klavarioti, D. Mantzavinos and D. Kassinos, Environ. Int., 2009, 35, 402–417.
13. A. Tsitonaki, B. Petri, M. Crimi, H. Mosbæk, R. L. Siegrist and P. L. Bjerg, Crit. Rev. Environ. Sci. Technol., 2010, 40, 55–91.
14. L. W. Matzek and K. E. Carter, Chemosphere, 2016, 151, 178–188.
15. C. Qi, X. Liu, J. Ma, C. Lin, X. Li and H. Zhang, Chemosphere, 2016, 151, 280–288.
16. J. Kang, X. Duan, L. Zhou, H. Sun, M. O. Tadé and S. Wang, Chem. Eng. J., 2016, 288, 399–405.
17. P. Neta, R. E. Huie and A. B. Ross, J. Phys. Chem. Ref. Data, 1988, 17(3), 1027–1284.
18. R. J. Watts and A. L. Teel, Pract. Period. Hazard., Toxic, Radioact. Waste Manage., 2006, 10, 2–9.
19. R. L. Johnson, P. G. Tratnyek and R. O’brien Johnson, Environ. Sci. Technol., 2008, 42, 9350–9356.
20. C. Tan, N. Gao, Y. Deng, N. An and J. Deng, Chem. Eng. J., 2012, 203, 294–300.
21. Y. Ji, C. Dong, D. Kong, J. Lu and Q. Zhou, Chem. Eng. J., 2015, 263, 45–54.
22. C. Liang, C. J. Bruell, M. C. Marley and K. Sperry, Soil Sediment Contam., 2003, 12, 207–228.
23. K. C. Huang, R. A. Couttenye and G. E. Hoag, Chemosphere, 2005, 49, 413–420.
24. H. Hori, Y. Nagaoka, M. Murayama and S. Kutsuna, Environ. Sci. Technol., 2008, 42, 7438–7843.
25. C. Tan, N. Gao, Y. Deng, W. Rong, S. Zhou and N. Lu, Sep. Purif. Technol., 2013, 109, 122–128.
26. A. Ghauch, A. Muthanna Tuqan and N. Kibbi, Chem. Eng. J., 2015, 279, 861–873.
27. Y. Ji, Y. Fan, K. Liu, D. Kong and J. Lu, Water Res., 2015, 87, 1–9.
28. D. Lu, F. Ji, W. Wang, S. Yuan, Z. Hu and T. Chen, Environ. Sci. Pollut. Res., 2014, 21, 8025–8035.
29. S. Zheng, W. Jiang, Y. Cai, D. D. Dionysiou and K. E. O'Shea, Catal. Today, 2014, 224, 83–88.
30. B. G. Petri, R. J. Watts, A. Tsitonaki, M. Crimi, N. R. Thomson and A. L. Teel, In Situ Chemical Oxidation for Groundwater Remediation, 2011,  DOI:10.1007/978-1-4419-7826-4_4.
31. R. H. Waldemer, P. G. Tratnyek, R. L. Johnson and J. T. Nurmi, Environ. Sci. Technol., 2007, 41, 1010–1015.
32. Y. Fan, Y. Ji, D. Kong, J. Lu and Q. Zhou, J. Hazard. Mater., 2015, 300, 39–47.
33. I. M. Kolthoff and I. K. Miller, J. Am. Chem. Soc., 1951, 73, 3055–3059.
34. A. Adak, K. P. Mangalgiri, J. Lee and L. Blaney, Science Direct, 2015, 70, 74–85.
35. C. Liang and H. W. Su, Ind. Eng. Chem. Res., 2009, 48, 5558–5562.
36. G. Fang, D. D. Dionysiou, D. Zhou, Y. Wang, X. Zhu, J. Fan, L. Cang and Y. Wang, Chemosphere, 2013, 90, 1573–1580.
37. P. Neta, V. Madhavan, H. Zemel and R. W. Fessenden, J. Am. Chem. Soc., 1977, 99, 163–164.
38. C. Liang, Z. Wang and N. Mohanty, Sci. Total Environ., 2006, 370, 271–277.
39. H. V. Lutze, S. Birher, I. Rapp, N. Kerlin, R. Bakkour, M. Geisler, C. von Sonntag and T. C. Schmidt, Environ. Sci. Technol., 2015, 49, 1673–1680.
40. R. Yuan, S. N. Ramjaun, Z. Wang and J. Liu, J. Hazard. Mater., 2011, 196, 173–179.
41. Y. Yang, J. J. Pignatello, J. Ma and W. A. Mitch, Environ. Sci. Technol., 2014, 48, 2344–2351.
42. J. Lu, J. Wu, Y. Ji and D. Kong, Water Res., 2015, 78, 1–8.
43. R. Yuan, Z. Wang, Y. Hu, B. Wang and S. Gao, Chemosphere, 2014, 109, 106–112.
44. L. Xu, R. Yuan, Y. Guo, D. Xiao, Y. Cao, Z. Wang and J. Liu, Chem. Eng. J., 2013, 217, 169–173.
45. G. P. Anipsitakis, D. D. Dionysiou and M. A. Gonzalez, Environ. Sci. Technol., 2006, 40, 1000–1007.
46. Y. Guan, J. Ma, Y. Ren, Y. Liu, J. Xiao, L. Lin and C. Zhang, Water Res., 2013, 47, 5431–5438.
47. P. M. David Gara, G. N. Bosio, M. C. Gonzalez, N. Russo, M. del Carmen Michelini, R. Pis Diez and D. O. Mártire, Photochem. Photobiol. Sci., 2009, 8, 992–997.
48. J. Wenk, U. von Gunten and S. Canonica, Environ. Sci. Technol., 2011, 45, 1334–1340.
49. J. Lu, J. Shao, Z. Wang, H. Liu and Q. Huang, Environ. Sci. Technol., 2015, 49, 8850–8857.
50. J. Zhou, J. Xiao, D. Xiao, Y. Guo, C. Fang, X. Lou and Z. Wang, Chemosphere, 2015, 134, 446–451.
51. S. J. Zhang, H. Jiang, M. J. Li, H. Q. Yu, H. Yin and Q. R. Li, Environ. Sci. Technol., 2007, 41, 1977–1982.
52. L. Carlos, D. Fabbri, A. L. Capparelli, A. Bianco Prevot, E. Pramauro and F. S. García Einschlag, Chemosphere, 2008, 72, 952–958.
53. L. Carlos, D. Nichela, J. M. Triszcz, J. I. Felice and F. S. García Einschlag, Chemosphere, 2010, 80, 340–345.
54. M. Zhang, X. Chen, H. Zhou, M. Murugananthan and Y. Zhang, Chem. Eng. J., 2015, 264, 39–47.
55. D. Vione, V. Maurino, C. Minero and E. Pelizzetti, Environ. Sci. Technol., 2005, 39, 1101–1110.
56. S. Zheng, Y. Cai and K. E. O'Shea, J. Photochem. Photobiol., A, 2010, 210, 61–68.
57. T. Xu, P. V. Kamat, S. Joshi, A. M. Mebel, Y. Cai and K. E. O'Shea, J. Phys. Chem., 2007, 111, 7819–7824.
58. B. Neppolian, E. Celik and H. Choi, Environ. Sci. Technol., 2008, 42, 6179–6184.
59. B. Neppolian, A. Doronila and M. Ashokkumar, Water Res., 2010, 44, 3687–3695.
60. L. Zhou, W. Zheng, Y. Ji, J. Zhang, C. Zeng, Y. Zhang, Q. Wang and X. Yang, J. Hazard. Mater., 2013, 263, 422–430.
61. D. Mohan and C. U. Pittman Jr, J. Hazard. Mater., 2007, 142, 1–53.
62. O. Isayev, B. Rasulev, L. Gorb and J. Leszczynski, Mol. Diversity, 2006, 10, 233–245.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra17764f

‡ The authors contributed equally to this research.

---