Abatement of Nitrophenol in Aqueous Solution by HOCl and UV/HOCl Processes: Kinetics, Mechanisms, and Formation of Chlorinated Nitrogenous Byproducts

Abstract

HOCl and UV activated HOCl (UV/HOCl) have been applied for water disinfection and abatement of organic contaminants. However, the production of toxic byproducts in the HOCl and UV/HOCl treatment should be scrutinized. This contribution comparatively investigated the elimination of 4-nitrophenol and the generation of chlorinated byproducts in HOCl and UV/HOCl treatment processes. 61.4% of 4-nitrophenol was removed by UV/HOCl in 5 min with HOCl dose of 60 μM, significantly higher than that by UV (3.3%) or HOCl alone (32.0%). Radical quenching test showed that HO^• and Cl^• played important roles in UV/HOCl process. 2-Chloro-4-nitrophenol and 2,6-dichloro-4-nitrophenol were generated consecutively in HOCl process; but their formation was less in the UV/HOCl process. Trichloronitromethane (TCNM) was only found in the UV/HOCl process, and its production increased with increasing HOCl dosage. Besides chlorinated products hydroxylated and dinitrated products were also identified in the UV/HOCl process. Transformation pathways involving electrophilic substitution, hydroxylation, and nitration were proposed for 4-nitrophenol transformation in the UV/HOCl process. Wastewater matrix could significantly promote the transformation of 4-nitrophenol to 2-chloro-4-nitrophenol in UV/HOCl process. Results of this study are helpful to advance the understanding of the transformation of nitrophenolic compounds and assess the formation potential of chlorinated byproducts in HOCl and UV/HOCl disinfection processes.

1. Introduction

Chlorination is extensively applied for disinfection in drinking water treatment worldwide [1,2]. However, chlorine alone does not effectively inactivate chlorine-resistant pathogens, such as Giardia and Cryptosporidium [3,4,5]. Ultraviolet (UV) light has a potential to modify the nucleic structure of microorganisms, thus reducing the risk of water-borne pathogens [6,7]. The combination of UV and chlorine (UV/HOCl) has attracted extensive attention due to its efficient disinfection performance and propensity to decompose organic contaminants [8,9].

The photolysis of hypochlorous acid/hypochlorite ion (HOCl/OCl⁻) can directly generate reactive oxidizing species including chlorine (Cl^•) and hydroxyl radicals (HO^•) (Equation (1)) [10]. As a non-selective oxidant, HO^• can react with various contaminants at nearly diffusion-controlled rates (10⁸–10¹⁰ M⁻¹ s⁻¹) [11]. In contrast, Cl^• prefers to interact with substances containing aromatic rings and electron-rich moieties selectively, such as phenols, benzoic acids, toluenes, and anilines, with second rate constants of 1.2–2.5 × 10¹⁰ M⁻¹ s⁻¹ [12,13]. HO^• and Cl^• also react with Cl⁻ or HOCl/OCl⁻ to generate other secondary radicals (i.e., ClO^• and Cl₂^•−) (Equations (2)–(6)) [14,15,16]. These reactive species make UV/HOCl a promising strategy for the elimination of a wide spectrum of micropollutants [17,18]. It should be noted that, the quantum yields for HOCl/OCl⁻ photolysis are greater than 1.2 by UV₂₅₄ irradiation (pH 4–10) due to the chain reactions [10]. The quantum yields are also highly dependent on the irradiation wavelength, solution pH, and concentration of HOCl/OCl⁻ [19,20,21].

HOCl/OCl⁻ + hv → HO^•/O^•− + Cl^•

(1)

Cl^• + Cl⁻ ↔ Cl₂^•− k₊ = 6.5 × 10⁹ M⁻¹s⁻¹, k₋ = 1.1 × 10⁵ s⁻¹

(2)

HOCl + HO^• → ClO^• + H₂O k = 2.0 × 10⁹ M⁻¹s⁻¹

(3)

OCl⁻ + HO^• → ClO^• + OH⁻ k = 8.8 × 10⁹ M⁻¹s⁻¹

(4)

HOCl + Cl^• → ClO^• + H⁺ + Cl⁻ k = 3.0 × 10⁹ M⁻¹s⁻¹

(5)

OCl⁻ + Cl^• → ClO^• + Cl⁻ k = 8.2 × 10⁹ M⁻¹s⁻¹

(6)

The UV/hydrogen peroxide (UV/H₂O₂) process is the conventional UV-based advanced oxidation process (AOP) [18]. However, under UV₂₅₄ irradiation, the molar absorption coefficient (ε) and quantum yield (Φ) of H₂O₂ (${\epsilon }_{{\mathrm{H}}_2{\mathrm{O}}_2}$ = 19.6 M⁻¹ cm⁻¹, ${\Phi }_{{\mathrm{H}}_2{\mathrm{O}}_2}$ = 1.0) are relatively lower than HOCl/OCl⁻ (${\epsilon }_{\mathrm{HOCl}}$ = 59 M⁻¹ cm⁻¹, ${\epsilon }_{{\mathrm{OCl}}^{–}}$= 66 M⁻¹ cm⁻¹, ${\Phi }_{\mathrm{HOCl}/{\mathrm{OCl}}^{–}}$ greater than 1.2) [18,22]. In the UV/H₂O₂ process, only 5–10% of the dosed H₂O₂ is consumed and the H₂O₂ residual is required to be removed by further process in drink water treatment, which causes an increase in cost [18,23]. The UV/persulfate (UV/PS) process is also an alternative AOP, which generates sulfate radicals (SO₄^•−) and HO^• [18,23]. However, UV/PS is sensitive to pH, chloride, and inorganic carbon [23]. This sensitivity can affect its treatment efficiency depending on the reactivity of radicals with individual contaminants [23]. Therefore, compared with these two UV-based AOPs, UV/HOCl is a more cost-effective process [18].

HOCl can transform numerous organic and inorganic micropollutants typically encountered in drinking water and wastewater (e.g., ferrous, arsenic, nitrite, phenols, pesticides, drugs, etc.) [24,25]. Phenols are toxic organic compounds widely used as raw materials for synthesis of pesticides, dyes, drugs, plastics, and antioxidants [1]. Since HOCl is not as strong an oxidant to mineralize organic micropollutants, potentially harmful chlorinated byproducts may be formed by electrophilic substitution [25,26,27]. Electrophilic substitution is one of the most important mechanisms accounting for the transformation of phenols during chlorination [28]. As a consequence of the ortho/para orientation of the hydroxyl group, the chlorination of phenol occurs through consecutive substitution at the 2, 4, and 6 positions, resulting in the generation of mono-, di-, and tri-chlorophenols [25]. However, during UV/HOCl treatment, the transformation rate of phenol is faster than chlorination alone, which is ascribed to the joint action of reactive radicals (e.g., HO^• and Cl^•) and HOCl [29]. In addition to electrophilic substitution by HOCl, phenol also undergoes fast reaction with reactive chlorine species (e.g., Cl^• and Cl₂^•−) [13,30]. Cl^• can extract hydrogen atom from phenol to form phenoxyl radical, which subsequently undergoes Cl^• addition generating chlorophenols [13]. Also, Cl₂^•− reacts moderately fast with phenol by single electron transfer or hydrogen abstraction to generate phenoxyl radical [30]. Chlorophenols are toxic and difficult to biodegrade and are lethal to a variety of organisms [31]. These chemicals have been listed as priority organic contaminants [32], and their use and production are restricted [31]. Therefore, the formation of chlorophenols in the processes of HOCl and UV/HOCl is a matter of serious concern.

The disinfection byproducts (DBPs) generated by the interaction between disinfectants and organic contaminants, or natural organic matter (NOM) have been a major concern of water safety [33,34,35]. Compared with the commonly studied DBPs (e.g., trihalomethanes, THMs, and haloacetic acids, HAAs), the unregulated DBPs, in particular those nitrogenous disinfection byproducts (N-DBPs), have been less studied although they are of higher carcinogenicity and mutagenicity [36,37]. As a significant group of N-DBPs, halonitromethanes (HNMs) have attracted great consideration because of their high genotoxicity and cytotoxicity [38,39]. Among the studied HNMs, trichloronitromethane (TCNM) comprises more than 50% proportion of total HNMs formed during drinking water treatment [40,41]. Therefore, TCNM has been recognized as a surrogate of HNMs to study the formation potential and regulation strategy of HNMs [41]. The amount of TCNM in disinfected water was reported to be 0.2–0.5 μg L⁻¹ [42]. However, reports on the potential and mechanism of TCNM formation in UV/HOCl processes are limited [43]. In addition, halonitrophenols are a newly identified group of nitrogen-containing aromatic halogenated DBPs [16,44,45]. Toxicological studies have indicated that aromatic halogenated DBPs generally induced higher developmental toxicity and growth inhibition than the commonly known aliphatic halogenated DBPs [44,45]. A study on the constitution of the DBP mixture from chlorination of bromide-rich raw water indicated that aromatic fractions accounted for 49–67%, which dominated the developmental toxicity of chlorinated water samples [44]. Because studies indicated that the one- and two-carbon-atom DBPs of current interest accounted for only ~16% of disinfected water cytotoxicity, there is a need to identify toxicity drivers within the poorly characterized higher-molecular-weight (more than two carbon atoms) DBP fraction, e.g., halonitrophenols [46]. However, reports on the formation potential and mechanisms of halonitrophenols and TCNM in UV/HOCl processes are limited [43]. In consequence, it is still essential to investigate the formation potential of halonitrophenols and TCNM during the degradation of substituted phenols in the UV/HOCl process in order to assure the safety of drinking water.

4-Nitrophenol is a nitro-substituted phenol with high toxicity and poor biodegradability [47]. It is extensively found in the industrial effluent of pesticides and synthetic dyes [47]. 4-Nitrophenol has endocrine disrupting effect, and its biological toxicity is greater than that of 2- and 3-nitrophenol [48,49]. In consequence, 4-nitrophenol can cause serious harmful effect on human health and ecosystem. This chemical is also ranked as a priority pollutant by the U.S. EPA with a range from 1 to 20 ppb for its maximum allowable concentrations [50]. As a result, it is advisable to develop efficacious water treatment methods to remove 4-nitrophenol from aqueous solution. When the water containing 4-nitrophenol is disinfected by HOCl and UV/HOCl processes, hydroxylated and chlorinated intermediates are likely formed under the electrophilic attack of HOCl, HO^•, and Cl^• [28,51,52]. Further oxidation of these intermediates may give rise to N-DBPs through ring-opening. However, the transformation mechanism of 4-nitrophenol in the HOCl and UV/HOCl processes, and the formation potential of N-DBPs have been barely reported, so further investigations are essentially needed. In real water, the efficiency of HOCl and UV/HOCl processes is influenced by water matrix, e.g., dissolved organic matter (DOM), alkalinity, halides, and ammonia [18,53]. DOM significantly restrains the degradation of some micropollutants by UV/HOCl process through scavenging radicals (e.g., HO^• and Cl^•), consuming the oxidant HOCl/OCl⁻, and filtering UV light [18,53]. Carbonate radicals (CO₃^•−) can be produced from the scavenging of HO^• and Cl^• by HCO₃⁻ or CO₃²⁻ [18]. Due to the scavenging effect of (bi)carbonate and overall lower reactivity of CO₃^•− with micropollutants, the presence of HCO₃⁻ and CO₃²⁻ is usually disadvantageous for micropollutant destruction [54]. In addition, halides (e.g., Cl⁻ and Br⁻) affect UV/HOCl process negatively or positively depending on the structures of targeted pollutants [18,55]. Therefore, effects of wastewater matrix on the transformation of 4-nitrophenol during the HOCl and UV/HOCl processes also warrant investigation.

The goals of this work are: (1) to investigate the transformation kinetics of 4-nitrophenol by HOCl and UV/HOCl processes and identify chlorinated intermediates; (2) to evaluate the formation potential of TCNM in HOCl and UV/HOCl processes; (3) to propose the degradation mechanisms and pathways of 4-nitrophenol in HOCl and UV/HOCl processes; (4) to investigate effects of wastewater matrix on the transformation of 4-nitrophenol during the HOCl and UV/HOCl processes.

2. Materials and Methods

2.1. Reagents and Materials

Chemicals, suppliers, and purities are provided in Text S1 of Supplementary Materials.

2.2. Chlorination Experiments

50 mL aqueous solution involving 10 μM 4-nitrophenol was buffered to neutral condition with 10 mM phosphate buffer. Note that the initial concentration of 4-nitrophenol (10 μM) was higher than the environmental concentrations, just for the purpose to amplify the reactions for better detection of transformation products. 10 μM HOCl was spiked into the reaction vials to initiate the chlorination reaction. The vials were retained in the dark at 20 °C. A control experiment without HOCl was carried out concurrently. 1 mL sample was collected from each vial at pre-set time intervals and quenched promptly with 20 μL 1 M Na₂S₂O₃. Samples were kept at 4 °C in the dark and analyzed within 24 h.

2.3. UV/HOCl Experiments

The UV/HOCl experiments were conducted in a BL-GHX-V photoreactor (Shanghai Bilon Instrument Co., Ltd., Shanghai, China) coupled with an ozone-free, low-pressure mercury lamp (15 W) emitting predominantly 254 nm monochromatic light (UV₂₅₄). The photoreactor was turned on for 15 min to achieve lamp emission stabilization in advance. The average UV fluence rate was measured to be 8.7 × 10⁻⁷ Einstein L⁻¹ s⁻¹ by a chemical actinometer [56]. Experimental conditions of the UV/HOCl treatment were the same as chlorination except that reaction solutions were transferred into quartz vials and exposed to UV₂₅₄ irradiation. The effect of HOCl dosage (ranged from 40 to 100 μM) on the elimination of 4-nitrophenol during the UV/HOCl process was investigated. A control experiment without HOCl (direct photolysis) was carried out concurrently. To assess the contribution of the free radicals to 4-nitrophenol elimination by UV/HOCl, an experiment was carried out by spiking 5 mM TBA as radical quencher under identical operating conditions to those in the UV/HOCl treatment.

2.4. HPLC Analyses

The concentrations of 4-nitrophenol and its chlorinated byproducts in the samples were determined using a Hitachi L-2000 high performance liquid chromatograph (HPLC) with an L-2455 diode array detector (DAD). Separation was achieved by an Agilent Zorbax Eclipse Plus C18 reverse phase column (5 μm, 4.6 mm × 250 mm). The injection volume was 10 μL. The elution was isocratic with a mixture of 65% methanol (with 0.1% formic acid) and 35% water (with 0.1% formic acid) at a flow rate of 1 mL min⁻¹. The detection wavelength of 4-nitrophenol, 2-chloro-4-nitrophenol, and 2,6-dichloro-4-nitrophenol was 312 nm, and the corresponding retention times were 4.7, 7.0, 10.5 min, respectively.

2.5. Quantification of TCNM

The formation potential of TCNM in the HOCl and UV/HOCl processes was tested in a set of vials. The reaction solution was 50 mL, containing 10 μM 4-nitrophenol and 60 μM HOCl. Other reaction conditions were identical to those as mentioned above. At pre-set time points, reaction solutions were quenched with 1 mL 1 mM Na₂SO₃. The quenched reaction solutions were refrigerated at 4 °C until further treatment and analysis.

The concentration of TCNM was measured according to the U.S. EPA standard method 551.1 [57]. TCNM was extracted by methyl tert-butyl ether (MtBE) and analyzed using an Agilent 7890 gas chromatograph (GC) equipped with an electron capture detector (ECD). A DB5-fused silica capillary column (1.5 μm, 0.53 mm × 30 m) was used for separation. The analytical parameters were as follows: inlet temperature of 220 °C; detector temperature of 280 °C; detector current of 1 nA. The heating procedure was described as follows: the initial temperature of 35 °C kept for 6 min; raising from 35 to 100 °C at a speed of 10 °C min⁻¹ and kept 100 °C for 2 min; rising from 100 to 200 °C at a speed of 20 °C min⁻¹ and kept 200 °C for 2 min, with a total of 27.5 min.

2.6. Identification of Intermediate Products

To identify intermediates, two aqueous solutions (50 mL, pH 7.0) containing 10 μM 4-nitrophenol and 60 μM HOCl were irradiated for 30 min and then concentrated by solid phase extraction (SPE) using hydrophilic-lipophilic balance (HLB) cartridges. Details of the procedures of SPE are provided in Text S2 of Supplementary Materials.

The SPE-concentrated samples were examined by a high-resolution mass spectrometer (HRMS, Triple TOF 5600+ MS/MS, AB Sciex, Boston, MA, USA) coupled with an HPLC (Shimadzu, Kyoto, Japan) to determine intermediate products. Analytical separation was accomplished on a Kinetex LC C18 column (2.6 μm, 100 Å, 100 mm × 2.1 mm, i.d.). The sample injection volume was 20 µL. The elution flow rate was 0.3 mL min⁻¹ with a mixture of eluent A (H₂O plus 0.1% formic acid) and eluent B (MeOH plus 0.1% formic acid), employing a linear gradient as follows: 10% B, 0–1 min; 10% to 90% B, 1–13 min, and 90% kept for 5 min; 90% to 10%, 18–18.1 min, and 10% kept for 1.9 min. Mass analyses were carried out under negative electrospray ionization (ESI(–)) in the TOF MS full scan (m/z 50–1000) mode. Operating parameters of ESI were set as: ion spray voltage floating of −4.5 kV; source temperature of 550 °C; gas I and gas II of 65 psi; curtain gas of 35 psi.

2.7. Evaluate the Influence of Wastewater

Wastewater was also used to evaluate the influences of water constituents on the transformation of 4-nitrophenol during the HOCl and UV/HOCl processes. Wastewater was obtained from a secondary sedimentation tank of a municipal sewage treatment plant (Nanjing city, China). The water sample was filtered through 0.22 μm cellulose acetate membrane before using. Water quality parameters including TOC, Abs₂₅₄, nitrate, nitrite, pH, and typical inorganic ions are shown in Table S1 of Supplementary Materials. Similar chlorination and UV/HOCl experiments were conducted in wastewater as mentioned above, except that no phosphate buffer was used.

3. Results and Discussions

3.1. Degradation Kinetics

The degradation kinetics of 4-nitrophenol by UV irradiation, chlorination alone, and UV/HOCl are illustrated in Figure 1a. No loss of 4-nitrophenol was observed under dark without HOCl. Direct photolysis of 4-nitrophenol was negligible (~3.3%) within 5 min upon UV₂₅₄ irradiation, which is in agreement with the report of Zhao et al. [47]. The photostability of 4-nitrophenol may be ascribed to the low quantum yield of the excited state of 4-nitrophenol, which probably stems from the stabilization effect of the −NO₂ group as a strong para electron-acceptor substituent [47,58]. Chlorination resulted in approximately 32.0% removal of 4-nitrophenol over a course of 5 min. Notably, the elimination of 4-nitrophenol during UV/HOCl process was appreciably enhanced. A degradation efficiency of 61.4% was accomplished within 5 min by UV/HOCl. The enhanced removal efficiency of 4-nitrophenol in UV/HOCl process indicates that, except for HOCl, there are other reactive species responsible for the elimination of 4-nitrophenol. HOCl and OCl⁻ can produce reactive free radicals (e.g., HO^• and Cl^•) under UV irradiation, which were likely involved in the elimination of 4-nitrophenol. In order to determine whether free radicals contributed to the elimination of 4-nitrophenol, TBA was spiked into the UV/HOCl system as a radical quencher for HO^• and Cl^• [59]. TBA reacts quickly with HO^• and Cl^• with the second-order rate constants of 1.9 × 10⁹ and 6.0 × 10⁸ M⁻¹ s⁻¹, respectively [60]. As shown in Figure 1a, the presence of 5 mM TBA appreciably depressed the elimination of 4-nitrophenol, and the removal efficiency decreased from 61.4% to 33.6%. Therefore, it can be inferred that HO^• and Cl^• significantly contributed to the removal of 4-nitrophenol in the UV/HOCl process. The fact that 4-nitrophenol could still degrade in the UV/HOCl process with 5 mM TBA indicates that HOCl per se also played a significant role.

The elimination of 4-nitrophenol by UV/HOCl could be well fitted by pseudo-first-order kinetics (Equation (7)). As HOCl dosage increased from 40 to 100 μM, the observed pseudo-first-order rate constant of 4-nitrophenol (k_obs) increased linearly ($R^2=0.99$) from 0.137$\pm$0.016 to 0.416$\pm$0.073 min⁻¹ (Figure 1b). This result shows that the elimination of 4-nitrophenol by UV/HOCl was positively affected by the amount of HOCl. With the increase of HOCl dose, the concentration of reactive radicals produced by HOCl/OCl⁻ photolysis increased, which accelerated the degradation of 4-nitrophenol. Similar result was observed by Guo et al. in the decomposition of chlortoluron by UV/HOCl process [59].

$$-\ln \frac{{[4-\mathrm{Nitrophenol}]}_{\mathrm{t}}}{{[4-\mathrm{Nitrophenol}]}_0}=k_{\mathrm{obs}}\mathrm{t}$$

(7)

3.2. Formation of Chlorinated Nitrophenols

Electron-rich phenols undergo electrophilic substitution with HOCl [25]. The generation of chlorinated products should be scrutinized because they are generally more toxic than their parent compounds [61,62]. In this study, two halogenated intermediates, 2-chloro-4-nitrophenol and 2,6-dichloro-4-nitrophenol, were identified by HPLC during the chlorination of 4-nitrophenol. 2-Chloro-4-nitrophenol was observed to be the main byproduct, and its concentration gradually increased to 5.3 μM with prolonged reaction time within 15 min (Figure 2a). The formation of 2,6-dichloro-4-nitrophenol was observed after 5 min, consistent with the speculation that dichlorophenols are derived from further chlorination of the monochlorophenols [25]. After reaction for 15 min, the concentration of 2,6-dichloro-4-nitrophenol accumulated to 0.63 μM. The mass balance calculation showed that the sum of these two halogenated products accounted for ~92% of 4-nitrophenol, indicating that chlorination is the primary transformation mechanism of 4-nitrophenol in the presence of HOCl alone.

Interestingly, when UV/HOCl was applied for treatment of 4-nitrophenol, only 2-chloro-4-nitrophenol was identified by HPLC (Figure 2b). At HOCl dosage of 60 μM, the concentration of 2-chloro-4-nitrophenol reached the maximum (1.15 μM) within 2 min, afterwards it gradually decreased to 0.35 μM at 15 min. Obviously, the evolution pattern of 2-chloro-4-nitrophenol in UV/HOCl system was disparate to that in HOCl system. The hump shape of 2-chloro-4-nitrophenol profile in UV/HOCl process indicated that reactive radicals (e.g., HO^• and Cl^•) could promote further transformation of 2-chloro-4-nitrophenol. This hypothesis was further confirmed by radical quenching experiments. As shown in Figure 2b, when 5 mM TBA was spiked into UV/HOCl system, the concentration of 2-chloro-4-nitrophenol increased continuously as the irradiation time was extended, reaching 2.8 μM at 15 min. This observation confirms again that 2-chloro-4-nitrophenol is predominantly generated by electrophilic substitution between HOCl and 4-nitrophenol (Equation (8)), whereas further transformation of 2-chloro-4-nitrophenol requires the contribution of reactive radicals (Equation (9)). Notably, the amount of 2-chloro-4-nitrophenol produced in the UV/HOCl process with 5 mM TBA was significantly lower than that in the direct chlorination process (2.8 vs. 5.3 μM). This may be due to the fact that a fraction of HOCl was photochemically decomposed under UV₂₅₄ radiation, leading to a decrease in the concentration of free chlorine capable of halogenating 4-nitrophenol.

$$\text{4-Nitrophenol}+\mathrm{HOCl}\text{→} \text{2-Chloro-4-nitrophenol}$$

(8)

$$\text{2-Chloro-4-nitrophenol}+{\mathrm{HO}}^{•}/{\mathrm{Cl}}^{•}\to \mathrm{Products}$$

(9)

In addition, the concentration of HOCl remarkably affected the generation of 2-chloro-4-nitrophenol in UV/HOCl process (Figure 3). As the dosage of HOCl rose from 40 to 100 μM, the maximum amount of 2-chloro-4-nitrophenol elevated from 0.91 to 1.82 μM. It should be noted that the increase of HOCl dosage will also increase the concentration of reactive radicals during UV/HOCl process. Therefore, although the concentration of 2-chloro-4-nitrophenol raised rapidly at the inception phase of the reaction, it decreased rapidly thereafter due to the oxidation by more reactive radicals. For instance, when the dosage of HOCl was 80 μM, the resulting 2-chloro-4-nitrophenol was almost completely converted within 15 min. HO^• and Cl^• are highly oxidative species (with redox potentials of 2.8 and 2.4 V, respectively) that are capable of oxidizing many organic pollutants. Therefore, we speculate that 2-chloro-4-nitrophenol generated in the UV/HOCl process may undergo further oxidation (e.g., ring-opening cleavage) under the attack of HO^• and Cl^•, resulting in the production of low-molecular-weight organic compounds such as N-DBPs.

Based on Equations (8) and (9), the generation and transformation of 2-chloro-4-nitrophenol by UV/HOCl can be depicted as Equation (10).

$$\frac{\mathrm{d}\left[\text{2-Chloro-4-nitrophenol}\right]}{\mathrm{dt}}=k_1\left[\text{4-Nitrophenol}\right]-k_2\left[\text{2-Chloro-4-nitrophenol}\right]$$

(10)

where [4-Nitrophenol] and [2-Chloro-4-nitrophenol] are the concentrations of 4-nitrophenol and 2-chloro-4-nitrophenol, respectively; $k_1$ and $k_2$ are the pseudo-first-order rate constants for the generation and transformation of 2-chloro-4-nitrophenol, respectively. Because the degradation of 4-nitrophenol obeys pseudo-first-order kinetics (as Equation (7)), the concentration of 4-nitrophenol can be calculated as Equation (11). Therefore, Equation (10) can be modified as Equation (12). With rearrangement of Equation (12), the generation and transformation of 2-chloro-4-nitrophenol in UV/HOCl process can be described by a consecutive kinetic model (Equation (13)).

$$\left[\text{4-Nitrophenol}\right]={\left[\text{4-Nitrophenol}\right]}_0{\mathrm{e}}^{-k_1\mathrm{t}}$$

(11)

$$\frac{\mathrm{d}\left[\text{2-Chloro-4-nitrophenol}\right]}{\mathrm{dt}}=k_1{\left[\text{4-Nitrophenol}\right]}_0{\mathrm{e}}^{-k_1\mathrm{t}}-k_2\left[\text{2-Chloro-4-nitrophenol}\right]$$

(12)

$$\left[\text{2-Chloro-4-nitrophenol}\right]=\frac{k_1{\left[\text{4-Nitrophenol}\right]}_0}{k_2-k_1}\left({\mathrm{e}}^{-k_1\mathrm{t}}-{\mathrm{e}}^{-k_2\mathrm{t}}\right)$$

(13)

The kinetic model was used for fitting experimental data by least-square nonlinear regression to determine the $k_1$ and $k_2$ values of 2-chloro-4-nitrophenol under different HOCl dosages (

Table 1. k₁ and k₂ values obtained from the experimental data fitted by the kinetic model at different HOCl dosage.

Concentration of HOCl (μM)	k1 (min−1)	k2 (min−1)
40	1.801	0.024
60	1.872	0.102
80	0.868	0.271
100	1.409	0.201

). Results suggest that 2-chloro-4-nitrophenol had higher generation and lower transformation rate constants in the presence of low concentration of HOCl. In contrast, lower generation and higher transformation rate constants of 2-chloro-4-nitrophenol were observed with high concentration of HOCl. These results suggest that higher dosage of HOCl could result in higher concentrations of radicals, which was favorable for further transformation of 2-chloro-4-nitrophenol.

3.3. Formation Potential of TCNM

By GC analysis, we quantified the formation potential of TCNM during the UV/HOCl process, as shown in Figure 4a. When the HOCl dosage was set at 60 μM, the formation of TCNM was not observed within the first 5 min. After 5 min of reaction, TCNM concentration increased remarkably, accompanied by a steady decline of 2-chloro-4-nitrophenol concentration (see Figure 2b). This phenomenon indicates that the generation of TCNM is a result of further transformation of 2-chloro-4-nitrophenol. Studies have shown that nitrophenols (e.g., 3-nitrophenol) can be good precursors of TCNM during chlorine disinfection [63]. These precursors undergo chlorination and subsequent ring-opening reactions, resulting in the generation of TCNM and other N-DBPs. With the reaction time reaching 30 min, the yield of TCNM gradually increased to 5.8 nM. Therefore, the reactive radicals generated by photolysis of HOCl play a crucial role in opening of the aromatic rings of chlorinated intermediates to generate TCNM. Note that no production of TCNM was observed in the presence of HOCl alone, possibly owing to the relatively weak oxidation ability of HOCl (E⁰ = 1.48 V) [64]. In addition, due to the electron-withdrawing effect of Cl substituent [65], the reactivity of 2-chloro-4-nitrophenol with HOCl is less as compared to the parent 4-nitrophenol. Therefore, it is not easy for HOCl alone to oxidatively cleave the benzene ring as compared to more reactive radicals. The lack of TCNM is also consistent with the fact that electrophilic substitution was the predominant mechanism accounting for 4-nitrophenol transformation in the presence of HOCl alone (Figure 2a).

It is also found that the formation potential of TCNM during the UV/HOCl process correlated well with the concentration of HOCl. As shown in Figure 4b, with the increase of HOCl dosage from 40 to 100 μM, the yield of TCNM increased by 5.7-fold (from 2.4 to 13.6 nM) after 30 min reaction. The increase of HOCl concentration led to the generation of 2-chloro-4-nitrophenol in the early stage, and also accelerated the degradation of 2-chloro-4-nitrophenol due to the formation of more reactive radicals. Therefore, the formation potential of TCNM increased significantly with the increase of HOCl dosage. Due to the high genotoxicity and cytotoxicity of TCNM [38,39], its formation needs to be controlled and regulated during the UV/HOCl process for removing 4-nitrophenol from wastewater.

3.4. Intermediate Products

The maximum concentration of 2-chloro-4-nitrophenol accounted for only 30% of transformed 4-nitrophenol in UV/HOCl system. Therefore, other intermediate products might also be generated. The reaction solutions of 4-nitrophenol treated by UV/HOCl for 30 min were concentrated by SPE and then examined by HRMS for identification of those intermediates escaped from HPLC analyses. The structural elucidation of intermediates was according to: (1) the measured mass with error less than 10 ppm of the theoretical accurate mass based on the elemental composition; (2) the fragmentation pattern of the molecular ions; (3) the isotopic distribution endowed by chlorine atoms (i.e., isotopic abundance ratio of ³⁵Cl and ³⁷Cl). Based on the above tactics, we preliminarily identified 5 intermediates of 4-nitrophenol (see Figure S1 and Table S2 of Supplementary Materials). Among them, 2-chloro-4-nitrophenol and 2,6-dichloro-4-nitrophenol are chlorinated products of 4-nitrophenol (note that 2,6-dichloro-4-nitrophenol was produced at a low concentration during the UV/HOCl process and could not be directly identified and quantified by HPLC). In addition, two nitrated products (i.e., 2,4-dinitrophenol, 2-chloro-4,6-dinitrophenol) and one hydroxylated product (e.g., 4-nitrocatechol) were identified. Based on the identified products and radical chemistry, the transformation pathways of 4-nitrophenol in the UV/HOCl process are described in Scheme 1. Detailed mechanisms accounting for the transformation of 4-nitrophenol by UV/HOCl process are illustrated as follows.

3.5. Transformation Mechanisms and Pathways

It is well known that HOCl reacts with electron-rich phenols to form chlorinated products via electrophilic substitution [25]. Substitution reactions usually proceed in the ortho or para position of the phenolic hydroxyl substituent due to electronic resonance effects [25]. Since the para position of 4-nitrophenol is occupied by the nitro group, one of its ortho position should be substituted by Cl to form 2-chloro-4-nitrophenol. Because 2-chloro-4-nitrophenol also has an unsubstituted ortho position, it is likely subjected to HOCl electrophilic attack again and generates 2,6-dichloro-4-nitrophenol sequentially. Meanwhile, a fraction of HOCl is decomposed into HO^• and Cl^• under UV₂₅₄ irradiation. Under the attack of reactive radicals, ring-opening reactions of 2-chloro-4-nitrophenol and 2,6-dichloro-4-nitrophenol occur, resulting in the generation of TCNM (see Scheme 1).

In addition, 4-nitrophenol undergoes electrophilic substitution by HO^• attack, leading to the generation of hydroxylated products. HO^• attacks the ortho position of 4-nitrophenol to produce 4-nitrocatechol [51]. Under further attack of HO^•, the C-N bond of 4-nitrocatechol is broken, producing 1,2,4-trihydroxybenzene and nitrogen dioxide radical (NO₂^•). A similar phenomenon of NO₂^• production has previously been reported in the oxidation of nitroaromatic compounds by HO^• attack [66]. These hydroxylated products undergo ring-opening reactions under further attack of HO^• to form oxygen-containing aliphatic compounds, which are eventually mineralized into H₂O and CO₂ [67]. When HO^• directly attacks the nitro position, 4-nitrophenol is converted to hydroquinone [51]. Further reaction of hydroquinone with HO^• also leads to the formation of 1,2,4-trihydroxybenzene, which is eventually mineralized.

In many advanced oxidation processes, NO₂^• is regarded as a potent nitrating agent that leads to the generation of nitrated byproducts [68,69,70]. NO₂^• can be formed by many pathways, including the photolysis of NO₃⁻, decomposition of peroxynitrous acid (HOONO), oxidation of NO₂⁻, etc. [68,71]. In this study, NO₂^• may be derived either from the denitration of 4-nitrocatechol or 4-nitrophenol under the attack of HO^•/Cl^•, or through HO^•/Cl^• oxidation of NO₂⁻ converted from nitro substituent. As an electrophilic nitrating reagent, NO₂^• can react with electron-rich compounds to generate nitrated byproducts through hydrogen atom abstraction, electrophilic addition or substitution [71,72]. For instance, 4-nitrophenol can undergo electrophilic substitution reaction with NO₂^• to form 2,4-dinitrophenol [73]. Similarly, 2-chloro-4-nitrophenol can be converted to 2-chloro-4,6-dinitrophenol under electrophilic attack of NO₂^•. All these nitrated byproducts can undergo further ring-opening and oxidation reactions under the attack of HO^•/Cl^•, which finally forms TCNM and other small molecular compounds.

3.6. Effects of Wastewater Matrix

As expected, the wastewater matrix slightly inhibited the transformation of 4-nitrophenol by HOCl process as compared to Milli-Q water (Figure 5a). This is likely due to the consumption of HOCl by reducing species (e.g., Fe²⁺, S²⁻) and effluent organic matter (EfOM) as they are ubiquitously present in wastewater.

Like in Milli-Q water, the elimination efficiency of 4-nitrophenol in wastewater by UV/HOCl process was significantly higher than that by HOCl process. However, the conversion of 4-nitrophenol by UV/HOCl process was increased in wastewater as compared to Milli-Q water. This unusual phenomenon may be associated with the indirect photolysis of 4-nitrophenol caused by some wastewater components (e.g., EfOM and NO₃⁻) under UV₂₅₄ irradiation. These components have photosensitizing activity and can generate transient reactive species upon radiation, such as triplet excited state (³EfOM^*) and HO^• [74,75]. Note that EfOM may quench a fraction of HOCl and HO^•/Cl^•, thus decreasing the removal efficiency of target compounds [76,77]. However, we cannot rule out the possibility that the transient reactive species mentioned above (e.g., ³EfOM^*) may also promote the decay of organic compounds.

In both wastewater and Milli-Q water, the generation of 2-chloro-4-nitrophenol in HOCl treatment showed a continuously increasing trend (Figure 5b), suggesting that HOCl alone could not convert 2-chloro-4-nitrophenol efficiently. The concentration of 2-chloro-4-nitrophenol in wastewater was 4.63 μM in 15 min, slightly lower than that in Milli-Q water (5.31 μM). In UV/HOCl process, however, the concentration of 2-chloro-4-nitrophenol in both water matrices showed a first increasing and then decreasing pattern. This suggest that reactive radicals such as HO^• and Cl^• are important to further convert 2-chloro-4-nitrophenol. Most noticeably, the maximum concentration of 2-chloro-4-nitrophenol in wastewater was 4.80 μM, much higher than that in Milli-Q water (1.15 μM). This observation might be attributed to the light screening effect of wastewater constituents that partially inhibited the photolysis of HOCl [10,78]. The reduced decomposition of HOCl was conducive for chlorination of 4-nitrophenol to generate 2-chloro-4-nitrophenol. Overall, the effects of wastewater matrix appear to be complex and further investigation is still necessary.

4. Conclusions

Results showed that the elimination efficiency of 4-nitrophenol by UV/HOCl was significantly higher than that by chlorination alone. Radical quenching experiments confirmed the important roles of HO^• and Cl^• in the removal of 4-nitrophenol during UV/HOCl process. Electrophilic substitution is the predominant reaction mechanism for 4-nitrophenol transformation in HOCl process, which quantitatively results in the formation of 2-chloro-4-nitrophenol and 2,6-dichloro-4-nitrophenol. In UV/HOCl process, the chlorinated product 2-chloro-4-nitrophenol can undergo further transformation upon the attack of HO^•/Cl^•. Other intermediates, including 4-nitrocatechol, 2,4-dinitrophenol, and 2-chloro-4,6-dinitrophenol were identified by HRMS. Among them, chlorinated nitrophenolic derivatives can undergo further ring-opening and oxidation reactions under the attack of reactive radicals, generating TCNM and other small molecular organic compounds. The formation of chlorinated nitrophenols and TCNM in HOCl and UV/HOCl disinfection processes should be scrutinized because of their carcinogenicity and mutagenicity. The wastewater matrix could significantly promote the transformation of 4-nitrophenol to 2-chloro-4-nitrophenol in UV/HOCl process, which might be related with the formation of transient reactive species and light screening effect of wastewater constituents.