Effect of UV/chlorine treatment on photophysical and photochemical properties of dissolved organic matter

Yangjian Zhou a , b , 1, Fangyuan Cheng a , 1, Dongyang He a, Ya-nan Zhang a , ∗, Jiao Qu a , ∗, Xin Yang b, Jingwen Chen c, Willie J.G.M. Peijnenburg d , e

Abstract

Dissolved organic matter (DOM) is a ubiquitous component in effluents, DOM discharged with an effluent can affect the composition and properties of natural DOM in the receiving waters. As the photophysical and photochemical properties of effluent DOM can be changed by wastewater treatment processes, the effect of UV/chlorine treatment on the photophysical and photochemical properties of DOM was inves- tigated using Suwannee River fulvic acid (SRFA) and Suwannee River natural organic matter (SRNOM) as representatives. Results showed that the absorbance of the two DOM was significantly decreased. The evolution trends of three representative photophysical parameters upon increase of chlorine dosages were observed. Also, a decrease in DOM aromaticity, molecular weight and electron-donating capacity was ob- served upon increasing chlorine dosage. Quantum yields of excited triplet state of DOM ( 3 DOM ∗), singlet oxygen ( 1 O 2 ) and hydroxyl radicals ( ·OH) first decreases and then increased in the UV/chlorine systems upon increasing chlorine dosages due to the different reaction pathways of the two DOM. Moreover, 3 DOM ∗ can not only be regarded as a “controller” of other reactive intermediates, but also effectively promote the photodegradation of bezafibrate, which is classified as a persistent organic contaminant. This study gives deep insights into effects of UV/chlorine on the photophysical and photochemical prop- erties of DOM, and is helpful for understanding the dynamic roles of DOM in the photodegradation of micropollutants.

Keywords:
Dissolved organic matter
UV/chlorine
Photophysical properties
Photochemical properties Photodegradation

1. Introduction

Dissolved organic matter (DOM), which is among the most complex molecular mixtures of biotically and abiotically degraded biomolecules known, plays amongst others an important role in the photochemical processes in the aquatic environment ( Zark and Dittmar, ; Zhang et al., 2018a ). The photophysical and pho- tochemical properties of DOM isolated from natural waters and their roles in the photodegradation of emerging pollutions are of great interest and have been investigated in many previously the effluent of wastewater treatment plants (WWTPs), which refers to NOM that has been modified during wastewater treatment, also received much attention as EfOM can alter the photoreactivity of natural DOM in the receiving waters ( Zhang et al., 2014 ).
The photophysical properties of DOM are mainly assessed using the absorption ability of light ( McKay et al., 2018 ), including spe- cific ultraviolet absorbance (e.g., at 254 nm, SUVA 254 ), the ratio of DOM absorbance at 254 to 365 nm ( E 2 / E 3 ratio), and slope charac- teristics ( S , nm −1 ) of the absorbance spectrum ( Twardowski et al., 20 04 ; Weishaar et al., 20 03 ). It was reported that the opti- cal properties of EfOM differ when compared with natural DOM ( Bodhipaksha et al., 2015 ), and of seawater DOM when compared with freshwater DOM ( Wang et al., 2019 ) due to the differences in chemical composition were reported. Thus, the DOM with differ- ent sources or undergo any chemical treatment that may induce the degradation of DOM are with different photophysical proper- ties ( Laszakovits et al., 2020 ; Wang et al., 2019 ).
Upon irradiation, the ground state DOM reaches its excited singlet state ( 1 DOM ∗), and then undergoes intersystem crossings (ISC) via spin orbital couplings to the triplet excited state ( 3 DOM ∗) ( McNeill and Canonica, 2016 ). Both 1 DOM ∗ and 3 DOM ∗ can fall back to the ground state of DOM via fluorescence/phosphorescence emission and nonradiative transition, which are the photophysical processes of DOM ( McNeill and Canonica, 2016 ). Besides, 3 DOM ∗ can take part in the redox reaction of organic pollutants via en- ergy transfer or electron transfer, and is an important precursor of reactive oxygen species (ROS) such as singlet oxygen ( 1 O 2 ) and the hydroxyl radical ( ·OH) ( Dalrymple et al., 2010; Dong and Rosario- Ortiz, 2012 ). These reactive species can initiate the indirect photol- ysis of various emerging pollutants in natural water ( Apell et al., 2019 ; Ge et al., 2019 ; Zhou et al., 2019c ). Furthermore, DOM can also inhibit the photodegradation of micropollutants via either light screening or quenching effect ( Janssen et al., 2014 ).
DOM from different sources has been proven to possess dif- ferent photochemical properties ( Lee et al., 2013 ; Wang et al., 2020 ). The 3 DOM ∗, 1 O 2 , and ·OH formation quantum yields of EfOM are higher compared with natural DOM ( Bodhipaksha et al., 2015 ; Zhang et al., 2014 ), and the 3 DOM ∗ formation quantum yields of seawater DOM are higher compared with freshwater DOM ( Wang et al., 2019 ). This leads to different effects of DOM isolated from different waters on the photodegradation of micropollutants ( Wang et al., 2019 ). Although many studies have investigated the properties of different DOM and their effects on the photodegrada- tion of micropollutants, little is known about the effects of sewage treatment technologies on the properties of DOM of the effluents.
UV irradiation, and sometimes UV/chlorine, is frequently used for the disinfection in WWTPs, and UV/chlorine based advanced oxidation processes (AOPs) were recently developed for the re- moval of recalcitrant micropollutants in wastewater ( Fang et al., 2014 ). Various highly reactive electrophilic radicals, such as ·OH and reactive chlorine species (RCS) are produced during the UV/chlorine treatment ( Guo et al., 2017; Zhou et al., 2019b ), and they play important roles in the degradation of micropollutants and in the inactivation of water-borne pathogens ( Duan et al., 2018 ; Xiang et al., 2016 ).
UV/chlorine treatment was proven to degrade humin acid ( Gao et al., 2019 ), and induced reactions in the aliphatic, olefinic and aromatic components of DOM ( Varanasi et al., 2018 ; Zhang and Parker, 2018 ). Among which, the reaction rate con- stant of chlorine radial (Cl ·) with aromatic structures reaches up to 10 10 M −1 s −1 level via single electron transfer or chlorine addi- tion ( Varanasi et al., 2018 ). Furthermore, it has been demonstrated that ozone treatment can change the molecular structure of DOM, thereby significantly affecting the photophysical and photochemi- cal properties of DOM ( Leresche et al., 2019 ). For instance, an in- crease in the quantum yields of 1 O 2 and fluorescence after ozone treatment were observed ( Leresche et al., 2019 ). Thus, it can be hypothesized that the synergistic effect of UV irradiation and reac- tive species induced changes in the composition of DOM can lead to a significant variation of its properties. However, the influence of UV/chlorine treatment on the properties of DOM is still unclear.
In this study, the effect of UV/chlorine treatment, a common step in sewage treatment, on the photophysical and photochemi- cal properties of DOM was investigated. Suwanee River fulvic acid (SRFA) and Suwanee River natural organic matter (SRNOM) were selected as DOM isolates. The optical properties and the appar- ent quantum yields of 3 DOM ∗, 1 O 2 , and ·OH of the two types of DOM were determined during the UV/chlorine treatment. Besides, to investigate the effect of treated DOM on the photodegradation of micropollutants, bezafibrate (BZF) was selected as a target con- taminant, which is a widely used lipid regulator and is frequently detected in wastewater, in sewage effluents and in natural water ( Zhou et al., 2019c ). The contribution of different reactive species generated from DOM treated by the UV/chlorine was evaluated.

2. Materials and methods

2.1. Chemicals

BZF (98%), sorbic acid (SA, 99%), sodium azide (NaN 3 , 98%), 2,4,6-trimethylphenol (TMP, 98%), furfuryl alcohol (FFA, 98%), iso- propanol (IPA, 99%), benzene (99%), phenol (PhOH, 99%), ni- trobenzene (NB, 98%), benzoic acid (BA, 98%), benzophenone (BP, 98%), hydroquinone (HQ, 98%), 4-hydroxybenzophenone (4- hBP, 98%), 4–chloro-4 -hydroxybenzophenone (4-Cl-4hBP, 98%), 2- chlorohydroquinone (2-Cl-HQ, 95%), p -benzoquinone ( p -BQ, 99%), sodium perchlorate (NaClO 4 ) and sodium hypochlorite solution (available chlorine 5%) were purchased from J&K Scientific Ltd. (Beijing, China); sodium dihydrogen phosphate (NaH 2 PO 4 ) and dis- odium hydrogen phosphate (Na 2 HPO 4 ) purchased from Tianjin Damao Chemical (Tianjin, China). Organic solvents used in this study with chromatographical purity were obtained from TEDIA (Fairfield, OH, USA). Suwannee River fulvic acid Standard II (SRFA, 2S101F) and Suwannee River NOM (SRNOM, 2R101N) were pur- chased from the International Humic Substances Society. Ultrapure water (18.2 M ) was obtained from a purification system pro- duced by Chengdu Ultrapure Technology Co., Ltd. (Chengdu, China).

2.2. UV/chlorine treatment experiments

UV/chlorine photochemical experiments were conducted at least in triplicate inside an OCRS-PX32T merry-go-round photore- actor (Kaifeng Hxsei Science Instrument Factory, China) equipped with a motorized turntable. A 500 W Hg lamp (254 nm) was used as the UV light source, and the light intensity at λ = 254 nm was measured to be (1.45 ± 0.07) mW cm −2 using a UV-B dual- channel ultraviolet radiation meter (Photoelectric Instrument Fac- tory of Beijing Normal University, China). During the photochemi- cal experiment, a condensation device was used, and the reactor was kept ventilated to control the temperature of the solutions steady during UV irradiation.
SRNOM and SRFA were added at an initial concentration of 10 mg L −1 , and different dosages of chlorine (0, 25, 50, 75, 100, 125, 150, 200 μM) were added. Phosphate buffer solution (PBS) was used to control the solution pH at 7.0. UV irradiation exper- iments were performed in the photochemical reactor for 10 min, and some quartz tubes were covered with aluminum foil as dark controls. NaClO 4 was used to control the ionic strength in some experiments. After UV/chlorine treatment, the photolysis solution was stored in the dark at room temperature. The residual chlorine was detected using the N,N–diethyl-1,4-phenylenediamine sulfate (DPD) method ( China, HJ 586 −2010 2010 ). DPD can react with free chlorine and generate colored compounds which show significant UV absorbance at 515 nm. The solutions were used in the subse- quent experiments until its residual chlorine was less than 0.05 mg L −1 .
Under UV/chlorine treatment, NB and BA were used as the probe chemicals to measure the steady-state concentrations of ·OH and Cl · ([ ·OH] SS and [Cl ·] SS ), respectively ( Varanasi et al., 2018 ), as detailed in the Text S1 in the supplementary information (SI). However, kinetic models are used to calculated the steady-state concentrations of ClO · and Cl 2 ·− ([ClO ·] SS and [Cl 2 ·−] SS ) (Table S1). This was needed because of lack of the probe compounds. Version 6.80 of the Kintecus software is a compiler to model and regress/fit/optimize the reactions ( Ianni, 2019 ). This software re- quires only the formula for the reaction kinetics and a reaction rate constant to fit the steady-state concentration of reactive radicals in the system (Table S1).

2.3. Spectroscopic measurements

During UV/chlorine treatment, samples containing SRFA or SRNOM were withdrawn and the absorption spectra of SRFA and SRNOM were determined using a Ultraviolet-visible Spectropho- tometer (Hitachi-U2900, Japan). Based on the results of the UV– visible absorption spectrum, the photophysical properties, includ- ing E 2 / E 3 , spectral slope ( S ), special UV absorption (SUVA 254 ), and wavelength-average specific absorption coefficient (SUVA avg ) of the two DOM were calculated; the detailed definition and calculation methods are described in Text S2.

2.4. Determination of photochemical properties of SRFA and SRNOM during UV/chlorine treatment

The photochemical properties of SRFA, SRNOM, SRFA after UV/chlorine treatment (SRFA-UV/Cl) and SRNOM after UV/chlorine treatment (SRNOM- UV/Cl) were determined in an XPA-7 merry- go-round photoreactor (Xujiang Electromechanical Plant, Nanjing, China) with 500 W Hg lamp equipped with 290 nm filters as the simulated sunlight. The light intensity at the surface of the quartz tubes was detected with a TriOS-RAMSES spectroradiometer (TriOS GmbH, Germany), and the result is shown in Fig. S1. TMP, FFA and benzene were used as the probe chemicals to measure the ap- parent quantum yields and steady-state concentrations of excited triplet DOM ( 3 DOM ∗), 1 O 2 and ·OH, respectively ( Zhou et al., 2018 ). Apparent quantum yields of reactive intermediates ( RIs ) were cal- culated by the following equation ( Zhao et al., 2020 ; Zhou et al., 2018 ): where R RIs is the formation rates of specific RIs (M s −1 ), and as can be calculated from (Text S3), [X] is concentration of a specific photosensitizer (M). k a ( λ) is the specific rate of light absorption (s −1 ), and can be calculated from Eq. (2) . The RIs values were calculated from 290 to 400 nm due to the relative insignificance of wavelengths > 400 nm. I p is the incident intensity (Einstein s −1 cm −2 ), εX ( λ) is the molar absorption coefficient of the photosensi- tizer (M −1 cm −1 ), z is the optical path (cm), α( λ) is the unit ab- sorbance of the background matrix (cm −1 ).
The details of the calculation methods of the steady-state con- centrations of 3 DOM ∗ ([ 3 DOM ∗] SS ), 1 O 2 ([ 1 O 2 ] SS ) and ·OH ([ ·OH] SS ) are shown in the Text S3.

2.5. Photodegradation of BZF

Considering the change of DOM concentration under UV/chlorine treatment, we used the same total organic car- bon (TOC) concentration of DOM as a control to study the change of photoreactivity and its effect on the photodegradation of BZF. In order to investigate the effect of excited triplet, 1 O 2 , and ·OH, quenching experiments were carried out with SA (2 mM) as excited triplet quencher, NaN 3 (2 mM) as 1 O 2 and ·OH quencher, and IPA (20 mM) as ·OH quencher ( Zhou et al., 2018 ).

2.6. Analytical methods

A Shimadzu LC-20A HPLC system (Shimadzu, Kyoto, Japan) with a UV–Vis detector and an Ultimate TM AQ-C18 column (250 mm × 4.6 mm, 5 mm, Welch Materials, Maryland, USA) was employed for the quantification of NB, BA , FFA , TMP, PhOH and BZF, as detailed in the Table S2. Dissolved organic carbon (DOC) concentrations of SRFA and SRNOM before and after UV/chlorine were measured with a Shimadzu TOC-50 0 0 analyzer using high- temperature combustion.

3. Results and discussion

3.1. Photophysical properties of SRFA and SRNOM before and after UV/chlorine treatment

UV–vis spectra of SRFA and SRNOM before and after UV/chlorine treatment with different chlorine dosages (0–200 μM) were determined, and the results are shown in Fig. 1 . The ab- sorbance intensity of SRFA and SRNOM decreased after UV irra- diation ( Fig. 1 A and C, Cl-0 vs. initial), indicating that UV-induced degradation can weakens the light absorption ability of the two DOM. The addition of chlorine further weakened the absorption ability and the absorbance intensity of SRFA and SRNOM decreased with increasing chlorine dosage ( Fig. 1 A and C), This can lead to a decreased light screening effect on the photodegradation of mi- cropollutants. The absorbance of SRFA and SRNOM showed no ob- vious difference during UV/chlorine treatment in the dark control (Fig. S2), implying that UV irradiation is essential for the degra- dation of SRFA and SRNOM. The effect of ionic strength, as de- termined by adding NaClO 4 under UV irradiation, was found not to significantly change the absorbance of the two types of DOM. This effect is opposite to the observations in the presence sodium hypochlorite (Fig. S2), which indicates that the decrease of light absorption of the two DOM is ionic strength-independent.
As the changes of the absorbance were quite small, the linear differential absorbance (DA) spectra were used, as commonly done in previous studies ( Korshin et al., 2007 ; Yan et al., 2014 ). The in- tensity of DA was calculated (detailed method is shown in Text S4) and the results showed that the DA for SRFA and SRNOM signifi- cantly increased with the increase of chlorine dosage ( Fig. 1 B and D). Previous studies reported that reactive radicals (e.g., Cl · and ·OH) generated during UV/chlorine treatment have a high reaction rate constant with DOM (~ 10 4 L mg −1 s −1 ) ( Guo et al., 2017 ), and induce the degradation of DOM ( Varanasi et al., 2018 ). This could lead to the changes of its light absorption ability. Thus, the steady-state concentrations of reactive radicals (RCS and ·OH) in the UV/chlorine treatment system with different chlorine dosages were calculated by the kinetic model according to the methods de- scribed in Text S1 and Table S1 ( Guo et al., 2017 ). The steady-state concentrations of RCS and ·OH increased with the increase of chlo- rine dosage during the UV/chlorine treatment (Fig. S3). This is the reason for the observed decrease of light absorbance of the two DOM. The influence of UV/chlorine treatment on the absorbance of SRFA is stronger compared to SRNOM ( Fig. 1 B and D). This is supposed to be attributed to the different reactivity of SRNOM and SRFA with these reactive species.
Three representative parameters ( E 2 / E 3 , SUVA 254 and S ) were calculated to further investigate the effect of UV/chlorine treat- ment on the photophysical properties of SRFA and SRNOM (Text S2). The results showed that E 2 / E 3 and S of SRFA and SRNOM increased with the increase of chlorine dosage, while SUVA 254 of the two DOM decreased with the increase of chlorine dosage under UV irradiation ( Fig. 2 ). These results indicated that the aromaticity and the molecular weight of DOM were reduced. Aromatic structures or other functional groups that have high reactivity with RCS or ·OH (e.g., olefinic structure) are easily oxidized by these reactive species, leading to functional group cleavage, and eventual conver- sion into low molecular-weight ring-opened compounds or min- eralized products ( Fang et al., 2014 ; Zhang and Parker, 2018 ). The aromaticity and molecular weight of SRFA and SRNOM were calcu- lated based on the method described in Text S2. The aromaticity of SRFA and SRNOM were found to decrease from 3.93% to 3.80% and 4.09% to 3.74%, respectively as the chlorine dosage increased from 0 to 200 μM (Table S3), whereas the molecular weight of SRFA and SRNOM decreased from 2.63 to 1.10 kDa and from 2.53 to 0.68 kDa, respectively (Table S4).
It has been reported that E 2 / E 3 is correlated inversely with the electron-donating capacity of DOM ( Helms et al., 2008 ; Sharpless et al., 2014 ). The shift in E 2 / E 3 suggested that UV/chlorine treatment decreased the electron-donating capacity of SRFA and SRNOM. Moreover, the growth rates of E 2 / E 3 and S and the reduction rate of SUVA 254 is faster for SRNOM than for SRFA ( Fig. 2 ). This indicates that the changes of aromaticity and molecular weight for SRNOM were more than those of SRFA during UV/chlorine treatment. This can also be confirmed by the calcu- lated removal rates of aromaticity and molecular weight for the two DOM after UV/chlorine treatment (chlorine dosage is 200 μM) as the removal rates were respectively 2.59 and 1.26 times higher than for SRFA (Table S3; Table S4). The predominant reason for this phenomenon is that SRNOM has a stronger electron donating ca- pacity compared with SRFA ( Walpen et al., 2016 ), and the RCS have higher reactivity with substances possessing stronger electron do- nating capacity ( Varanasi et al., 2018 ).

3.2. Effect of UV/chlorine treatment on 3 DOM ∗ generation from SRFA and SRNOM

The quantum yields of 3 DOM from SRFA and SRNOM were determined to be (4.66 ± 0.10) × 10 −3 and (4.07 ± 0.16) × 10 −3 , respectively ( Table 1 ), which is in the same order of magnitude as reported in a previous study ( Wang et al., 2019 ). After UV irradiation, the 3DOM ∗ of SRFA and SRNOM de- creased to be (3.84 ± 0.06) × 10 −3 and (3.79 ± 0.10) × 10 −3 , respectively ( Fig. 3 ). These results indicated that the photosensi- tizing chromophores of the two DOM were damaged during the UV irradiation treatment. The 3DOM ∗ of SRFA-UV/Cl increased was observed upon increasing chlorine dosages. The 3DOM ∗ of SRFA- UV/Cl was higher than the 3DOM ∗ of SRFA at relatively high chlo- rine dosages of 75 μM and more ( Fig. 3 ). The different effect on the 3DOM ∗ of SRFA in the absence and presence of chlorine is attributed to the different reaction pathways of the two reaction processes: (1) UV irradiation mainly induced the bond-cleavage reaction of DOM or a small amount of reactive oxygen species (e.g., 1 O 2 and ·OH) produced by DOM caused self-sensitized pho- tolysis ( Sharpless et al., 2014 ). (2) In the presence of chlorine, a large amount of RCS and ·OH were generated, which could react with DOM by an addition reaction (hydroxyl or chlorine addition), single electron transfer or bond breaking ( Varanasi et al., 2018 ; Zhang et al., 2018b ), resulting in the generation of more complex functional groups (e.g., aromatic ketones or quinones) ( Duan et al., 2018 ).
At lower chlorine dosages (25–50 μM), the values of 3DOM ∗ of SRFA-UV/Cl are lower compared with those of SRFA, and higher than the 3DOM ∗ of SRFA after UV irradiation ( Fig. 3 ). This is at- tributed to low concentrations of RCS and ·OH in the presence of low chlorine dosages (Fig. S3). Thus, the UV-induced photodegra- dation accounts for a large proportion of the overall reactivity observed. Meanwhile, at low chlorine dosages, ·OH or RCS pre- dominantly induce addition reactions when oxidizing SRFA. Cl · is highly reactive in inducing addition reactions with organic mat- ters ( k ≈ 10 8 –10 9 M −1 s −1 ) ( Gilbert et al., 1988 ; Lei et al., 2019 ), whereas previous studies have shown that the oxidation of aro- matic substances by ·OH occurs firstly by means of hydroxyl ad- dition, followed by electron transfer forming quinones, and finally cleavage of the benzene ring ( Zhang et al., 2016 ). The addition re- actions induced by low concentration ·OH or RCS in the reduction of 3DOM ∗.
Five low molecular DOM analogs (BP, HQ, 4-hBP, 4-Cl-4hBP, and 2-Cl-HQ, with molecular structures shown in Table S5) were se- lected to investigate the effect of hydroxyl or chlorine addition on the 3DOM ∗ of DOM. BP and HQ contain ketone and phenol functional groups and these chemicals were selected since ketones and phenols are the main chromophores of DOM ( McNeill and Canonica, 2016 ). And 4-hBP, 4-Cl-4hBP, and 2-Cl-HQ are corre- sponding hydroxylated and/or chlorinated products of BP and HQ. The 3DOM ∗-values of BP and HQ were (10.50 ± 2.10) × 10 −3 and (0.174 ± 0.010) × 10 −3 , respectively, while the values of 3DOM ∗ of 4-hBP, 4-Cl-4hBP and 2-Cl-HQ were (0.036 ± 0.002) × 10 −3 , (0.033 ± 0.002) × 10 −3 and (0.145 ± 0.015) × 10 −3 , respectively ( Table 1 ). The 3DOM ∗ of hydroxylation or/and chlorination prod- ucts are lower than those of BP and HQ. Thus, it can be concluded that the 3DOM ∗ of DOM is significantly decreased after hydroxyl or chlorine addition.
As the chlorine dosage was increased to 75–200 μM, the 3DOM ∗ values of SRFA-UV/Cl gradually increased, and were higher than that of SRFA. This is attributed to the single-electron transfer reactions induced by RCS or ·OH at high concentrations (Fig. S3). The phenolic structures undergo electron transfer reactions, which leads to the formation of the semiquinone/quinone radical, which then converts to the corresponding semiquinones/quinones in the presence of molecular oxygen ( Rodríguez and von Gunten, 2020 ; Zhang et al., 2016 ). The hydroxyl groups could also be added to the alkyl functionality of DOM, and then undergo further oxidation reactions or dehydration of two adjacent hydroxyl groups, which leads to the formation of ketones with conjugated π bonds during UV/chlorine treatment ( Duan et al., 2018 ). Both quinones and ketones could generate high-energy excited triplets with average energies of approximately 231 kJ mol −1 and 285 kJ mol −1 , respec- tively ( Zhou et al., 2019a ).
The p -BQ (molecular structure is shown in Table S5), a quinone product of HQ oxidation was used to assess discuss the effect of a quinone functionality on 3DOM ∗. The results showed that the 3DOM ∗ of HQ ((0.174 ± 0.010) × 10 −3 ) is lower than that of p – BQ ((0.221 ± 0.025) × 10 −3 ) ( Table 1 ). Furthermore, the values of 3DOM ∗ of BP determined here are in general quite high when compared with other substances. As a consequence, the formation of quinone and ketone functional groups is the causes for the in- crease of 3DOM ∗ during UV/chlorine treatment with high chlorine dosage.
The trend in the variation of 3DOM ∗ for SRNOM upon increase of the chlorine dosage under UV/chlorine treatment is similar to the trend for SRFA ( Fig. 3 ). However, upon increasing the chlo- rine dosage, SRNOM-UV/Cl has a higher value of 3DOM ∗ than SRFA-UV/Cl due to the structural differences of SRNOM and SRFA. The steady-state concentration of 3 DOM ∗ ([ 3 DOM ∗] SS ) produced by SRFA-UV/Cl and SRNOM-UV/Cl decreased with the increase of chlo- rine dosage during UV/chlorine treatment (Fig. S4), This is contrary to the variation trends of the variation of 3DOM ∗ for SRFA-UV/Cl and SRNOM-UV/Cl. This finding is attributed to the absorbance changes of SRFA and SRNOM at λ > 290 nm during UV/chlorine treatment. With the chlorine dosage increasing from 0 to 200 μM, SUVA avg of SRFA and SRNOM decreased from 5.12 to 1.42 L mgC −1 m −1 and from 3.46 to 0.72 L mgC −1 m −1 , respectively (Table S6). The SUVA avg reduction of more than 3 times compared to the ini- tial DOM led to the reduction of [ 3 DOM ∗] SS in SRFA-UV/Cl and SRNOM-UV/Cl solutions.

3.3. Effect of UV/chlorine treatment on 1 O 2 and ·OH generation from SRFA and SRNOM

The quantum yields of 1 O 2 ( 1O2 ) produced by SRFA and SRNOM decreased at first and then increased with the increase of chlorine dosage under UV/chlorine treatment ( Fig. 4 A). These variation trends are coincident with that of the 3DOM ∗ for the two types of DOM. Meanwhile, with the increase of chlorine dosage, the 3DOM ∗ and the 1O2 of SRFA-UV/Cl and SRNOM-UV/Cl showed a positively linear correlation ( R 2 > 0.975) (Fig. S5A and B). This is attributed to generation of 1 O 2 through energy transfer reactions between dissolved oxygen and 3 DOM ∗, i.e., 3 DOM ∗ is the precursor of 1 O 2 ( McNeill and Canonica, 2016 ). In a previous study, the 1O2 from different sources of DOM was also related to the 3DOM ∗ ( McKay et al., 2017). Moreover, in case of SRNOM-UV/Cl (0.206) was lower than the slope in case of SRFA-UV/Cl (0.233) (Fig. S5A and B). The slope is reported to be positively correlated to the transformation ability from 3 DOM ∗ to 1 O 2 ( McKay et al., 2017 ). Similarly, the positively linear correlations between [ 1 O 2 ] SS and [ 3 DOM ∗] SS produced by SRFA-UV/Cl and SRNOM-UV/Cl were observed and the slope for SRFA-UV/Cl was higher compared with the slope of SRNOM-UV/Cl (Fig. S6A and B). Thus, the excited triplet states of SRNOM-UV/Cl were more difficult to transform to 1 O 2 compared with the excited triple states of SRFA-UV/Cl after UV/chlorine treatment.
The 1O2 -values for the selected low molecule DOM analogs were calculated and the results showed that the 1O2 for BP ((8.48 ± 0.09) × 10 −2 ) containing a ketone functionality and for p -BQ ((2.92 ± 0.01) × 10 −2 ) with a quinone functional group were much higher than the value of 1O2 for the other analogs, which is in accordance with the results observed in previous lit- erature ( Leresche et al., 2019 ; McNeill and Canonica, 2016 ). Thus, the formation of quinone and ketone functional groups during the UV/chlorine treatment are the reason for the increase of the 1O2 from SRFA and SRNOM.
For ·OH, an increase was observed of the quantum yields of ·OH ( ·OH ) for SRFA-UV/Cl and SRNOM-UV/Cl ( Fig. 4 B) and a decrease in [ ·OH] SS of these two DOM with increasing chlorine dosage dur- ing UV/chlorine treatment (Fig. S4). Furthermore, the obtained re- sults showed that the OH and [ ·OH] SS were positively linear cor- related with 3DOM ∗ and· [ 3 DOM ∗] SS , respectively (Fig. S5C and D; Fig. S6C and D). However, values of the correlation coefficients ( R 2 ) of 3DOM ∗ with ·OH for SRFA-UV/Cl and SRNOM-UV/Cl were 0.893 and 0.760, respectively, both of which were lower than the values of 3DOM ∗ with 1O2 (0.985 for SRFA-UV/Cl and 0.975 for SRNOM-UV/Cl).
Previous studies had reported that 3 DOM ∗ is not a direct pre- cursor of ·OH, but 3 DOM ∗ could be used as a precursor for low- energy hydroxylators ( McKay et al., 2017 ). The generation path- ways of ·OH from DOM are complicated under light irradiation. The superoxide radical anion (O 2 ·−) is firstly generated during elec- tron transfer reactions between 3 DOM ∗ and dissolved oxygen, and subsequently undergoes dismutation to form hydrogen peroxide (H 2 O 2 ) and eventually ·OH ( Vione et al., 2014 )
Furthermore, ·OH could also be produced by the reaction of 3 DOM ∗ with the hydroxide ion (OH −) in water, which only oc- cur for DOM containing a small amount of structures such as quinones, as the excited quinones have a higher reduction poten- tial ( E 0 (quinones) > 2.19 V > E 0 (OH −) = 1.77 V) ( McNeill and Canonica, 2016 ). As can be seen from Table 1 , the OH of p – ((3.17 ± 0.12) × 10 −5 ). Thus, the generation of quinone groups in the two DOM could lead to the increase of ·OH .

3.4. Correlation between photophysical properties and photochemical properties

As shown in Fig. 5 , in the UV/chlorine system with different chlorine dosages, the RIs of SRFA-UV/Cl and SRNOM-UV/Cl had a good linear correlation with E 2 / E 3 , S , and SUVA 254 , among which E 2 / E 3 and S were positively linear correlated with the RIs (0.788 < R 2 < 0.918), whereas SUVA 254 was negatively linear correlated with the RIs (0.727 < R 2 < 0.952). These findings implied that the DOM with lower molecular weight, aromaticity and electron- donating capacity has the higher RIs . Previous studies also sug- gested that the photochemical properties of DOM were related to its molecular weight, aromaticity, and electron-donating capacity ( Dalrymple et al., 2010 ; McKay et al., 2017 ). This is attributed to the intramolecular interactions of DOM (e.g., energy transfer or electron transfer) ( Ma et al., 2010 ; Sharpless and Blough, 2014 ), as the closely-connected electron-donating structures and electron- accepting structures of DOM have higher ability to absorb light and produce photochemical reactive species ( McKay et al., 2016 ). Thus, the UV/chlorine treatment induced changes of molecular weight, aromaticity or electron-donating capacity of DOM can influence the generation of the photo-chemically reactive species from DOM.
Overall, the trends in the variation of the photophysical and photochemical properties of SRFA and SRNOM were similar in the UV/chlorine treatment with different chlorine dosages. How- ever, the range of variation of all the parameters, including E 2 / E 3 , S , SUVA 254 , and RIs of SRNOM was wider than in case of SRFA ( Figs. 2 ; 3 ; 4 ). SRFA is the fulvic acid component isolated from Suwanee River, while SRNOM is consisted of fulvic acid, humic acid and other organic matters isolated from Suwanee River ( Green et al., 2015 ).Thus, these results indicated that RCS and ·OH produced in the UV/chlorine system could not only af- fect the photophysical and photochemical properties of the fulvic acid component but also other components of DOM (e.g., humic acid).

3.5. Effect of SRFA-UV/Cl and SRNOM-UV/Cl on photodegradation of BZF under simulated sunlight irradiation

The DOM released from WWTPs or from wastewater treated by UV/chlorine can alter the photoreactivity of natural DOM in the re- ceiving waters, and further influence the photodegradation of or- ganic pollutants in natural water ( Zhang et al., 2014 ). Thus, the effect of SRFA-UV/Cl and SRNOM-UV/Cl on the photodegradation of BZF under simulated sunlight irradiation was investigated. The photodegradation rate constant ( k obs ) of BZF in phosphate buffer solution (PBS, pH = 7.0) was determined to be (2.51 ± 0.15) × 10 −4 min −1 ( Fig. 6 ). The extremely low molar absorptivity of BZF at wavelengths > 290 nm was the reason for its persistence to pho- todegradation (Fig. S7). However, faster photodegradation was ob- served for BZF in solutions containing SRFA, SRNOM, SRFA-UV/Cl and SRNOM-UV/Cl, and the photodegradation of BZF was more effectively promoted in the presence of SRFA-UV/Cl and SRNOM- UV/Cl ( Fig. 6 ). It is indicated that DOM-induced indirect pho- todegradation is an important degradation pathway of BZF in sur- face water, and the DOM after UV/chlorine treatment plays more important roles in the photochemical degradation of BZF, which may affect the environmental persistence of BZF in the receiving water.
The role of reactive 3 DOM ∗, 1 O 2 , and ·OH in the photodegrada- tion of BZF sensitized by SRFA, SRNOM, SRFA-UV/Cl and SRNOM- UV/Cl was investigated, and the results are shown in Fig. 6 . Ad- dition of SA obviously inhibited the photodegradation of BZF, im- plying that the excited triplet states of these DOM play important roles in the photodegradation of BZF. NaN 3 and IPA significantly inhibited the photodegradation of BZF, while the effect of IPA is much weaker compared with that of SA and NaN 3 . Consequently, these DOM predominantly promote the photodegradation of BZF through the generation of their excited triplet state, followed by 1 O 2 , whereas ·OH plays a minor role in the photodegradation of BZF.

4. Conclusion

This study evaluated the effect of a practical wastewater treat- ment process, UV/chlorine treatment, on the photophysical and photochemistry properties of DOM. The reactive species generated in the UV/chlorine system were found to be able to significantly decrease the aromaticity, molecular weight, and electron-donating capacity of SRFA and SRNOM. This leads to the inhibitory effects on the absorbance abilities of the two DOM. At low chlorine dosages, the RIs of the two DOM was decreased due Bezafibrate to the hydroxylation and/or chlorination reactions. On the contrary, the RIs was in- creased at high chlorine dosages due to the formation of aromatic ketone or quinone functionalities. In addition, DOM with lower molecular weight, aromaticity and electron-donating capacity have the highest values of RIs . The DOM treated by UV/chlorine more effectively promoted the photodegradation of BZF and the promo- tional effect is mainly attributed to the contribution of the ex- cited triplet states, followed by reaction with 1 O 2 and ·OH. The results of this study indicated that the photophysical and photo- chemical properties of DOM are changed during UV/chlorine treat- ment, which further affects the role of DOM in the photodegra- dation of micropollutants and may further alter the properties of natural DOM.

References

Apell, J.N. , Pflug, N.C. , McNeill, K. , 2019. Photodegradation of fludioxonil and other pyrroles: the importance of indirect photodegradation for understanding en- vironmental fate and photoproduct formation. Environ. Sci. Technol. 53 (19), 11240–11250 .
Bodhipaksha, L.C. , Sharpless, C.M. , Chin, Y. , Sander, M. , Langston, W.K. , MacKay, A .A . , 2015. Triplet photochemistry of effluent and natural organic matter in whole water and isolates from effluent-receiving rivers. Environ. Sci. Technol. 49 (6), 3453–3463 .
China, HJ 586 −2010 (Dec 1, 2010).
Dalrymple, R.M. , Carfagno, A.K. , Sharpless, C.M. , 2010. Correlations between dis- solved organic matter optical properties and quantum yields of singlet oxygen and hydrogen peroxide. Environ. Sci. Technol. 44 (15), 5824–5829 .
Dong, M.M. , Rosario-Ortiz, F.L. , 2012. Photochemical formation of hydroxyl radical from effluent organic matter. Environ. Sci. Technol. 46 (7), 3788–3794 .
Duan, X. , Sanan, T. , Cruz, A. , He, X. , Kong, M. , Dionysiou, D.D. , 2018. Susceptibility of the algal toxin microcystin-LR to UV/chlorine process: comparison with chlori- nation. Environ. Sci. Technol. 52 (15), 8252–8262 .
Fang, J. , Fu, Y. , Shang, C. , 2014. The roles of reactive species in micropollutant degra- dation in the UV/free chlorine system. Environ. Sci. Technol. 48 (3), 1859–1868 .
Gao, Z. , Lin, Y. , Xu, B. , Xia, Y. , Hu, C. , Zhang, T. , Cao, T. , Chu, W. , Gao, N. , 2019. Ef- fect of UV wavelength on humic acid degradation and disinfection by-product formation during the UV/chlorine process. Water Res 154, 199–209 .
Ge, L. , Zhang, P. , Halsall, C. , Li, Y. , Chen, C. , Li, J. , Sun, H. , Yao, Z. , 2019. The impor- tance of reactive oxygen species on the aqueous phototransformation of sulfon- amide antibiotics: kinetics, pathways, and comparisons with direct photolysis. Water Res 149, 243–250 .
Gilbert, B.C. , Stell, J.K. , Peet, W.J. , Radford, K.J. , 1988. Generation and reactions of the chlorine atom in aqueous solution. J. Chem. Soc. 84 (10), 3319–3330 .
Green, N.W. , McInnis, D. , Hertkorn, N. , Maurice, P.A. , Perdue, E.M. , 2015. Suwannee river natural organic matter: isolation of the 2R101N reference sample by re- verse osmosis. Environ. Eng. Sci. 32 (1), 38–44 .
Guo, K. , Wu, Z. , Shang, C. , Yao, B. , Hou, S. , Yang, X. , Song, W. , Fang, J. , 2017. Radical chemistry and structural relationships of PPCP degradation by UV/chlorine treatment in simulated drinking water. Environ. Sci. Technol. 51 (18), 10431–10439 .
Helms, J.R. , Stubbins, A. , Ritchie, J.D. , Minor, E.C. , Kieber, D.J. , Mopper, K. , 2008. Absorption spectral slopes and slope ratios as indicators of molecular weight, source, and photobleaching of chromophoric dissolved organic matter. Limnol. Oceanogr. 53, 955–969 .
Janssen, E.M.L. , Erickson, P.R. , McNeill, K. , 2014. Dual roles of dissolved organic mat- ter as sensitizer and quencher in the photooxidation of tryptophan. Environ. Sci. Technol. 48 (9), 4916–4924 .
Korshin, G.V. , Benjamin, M.M. , Chang, H. , Gallard, H. , 2007. Examination of NOM chlorination reactions by conventional and stop-flow differential absorbance spectroscopy. Environ. Sci. Technol. 41 (8), 2776–2781 .
Laszakovits, J.R. , Somogyi, A. , MacKay, A.A. , 2020. Chemical alterations of dis- solved organic matter by permanganate oxidation. Environ. Sci. Technol. 54 (6), 3256–3266 .
Lee, E. , Glover, C.M. , Rosario-Ortiz, F.L. , 2013. Photochemical formation of hydroxyl radical from effluent organic matter: role of composition. Environ. Sci. Technol. 47 (21), 12073–12080 .
Lei, Y. , Cheng, S. , Luo, N. , Yang, X. , An, T. , 2019. Rate constants and mechanisms of the reactions of Cl · and Cl 2 ·– with trace organic contaminants. Environ. Sci. Technol. 53 (19), 11170–11182 .
Leresche, F. , McKay, G. , Kurtz, T. , von Gunten, U. , Canonica, S. , Rosario-Ortiz, F.L. , 2019. Effects of ozone on the photochemical and photophysical properties of dissolved organic matter. Environ. Sci. Technol. 53 (10), 5622–5632 .
Ma, J. , del Vecchio, R. , Golanoski, K.S. , Boyle, E.S. , Blough, N.V. , 2010. Optical proper- ties of humic substances and CDOM: effects of borohydride reduction. Environ. Sci. Technol. 44 (14), 5395–5402 .
McKay, G. , Couch, K.D. , Mezyk, S.P. , Rosario-Ortiz, F.L. , 2016. Investigation of the cou- pled effects of molecular weight and charge-transfer interactions on the optical and photochemical properties of dissolved organic matter. Environ. Sci. Technol. 50 (15), 8093–8102 .
McKay, G. , Huang, W. , Romera-Castillo, C. , Crouch, J.E. , Rosario-Ortiz, F.L. , Jaffé, R. , 2017. Predicting reactive intermediate quantum yields from dissolved organic matter photolysis using optical properties and antioxidant capacity. Environ. Sci. Technol. 51 (10), 5404–5413 .
McKay, G. , Korak, J.A. , Erickson, P.R. , Latch, D.E. , McNeill, K. , Rosario-Ortiz, F.L. , 2018. The case against charge transfer interactions in dissolved organic matter photo- physics. Environ. Sci. Technol. 52 (2), 406–414 .
McNeill, K. , Canonica, S. , 2016. Triplet state dissolved organic matter in aquatic pho- tochemistry: reaction mechanisms, substrate scope, and photophysical proper- ties. Environ. Sci.: Process. Impacts 18 (11), 1381–1399 .
Mostafa, S. , Rosario-Ortiz, F.L. , 2013. Singlet oxygen formation from wastewater or- ganic matter. Environ. Sci. Technol. 47 (15), 8179–8186 .
Rodríguez, E.M. , von Gunten, U. , 2020. Generation of hydroxyl radical during chlo- rination of hydroxyphenols and natural organic matter extracts. Water Res 117, 115691 .
Sharpless, C.M. , Aeschbacher, M. , Page, S.E. , Wenk, J. , Sander, M. , McNeill, K. , 2014. Photooxidation-induced changes in optical, electrochemical, and photo- chemical properties of humic substances. Environ. Sci. Technol. 4 8 (5), 26 88– 2696 .
Sharpless, C.M. , Blough, N.V. , 2014. The importance of chargetransfer interactions in determining chromophoric dissolved organic matter (CDOM) optical and photo- chemical properties. Environ. Sci.: Process. Impacts 16 (4), 654–671 .
Twardowski, M.S. , Boss, E. , Sullivan, J.M. , Donaghay, P.L. , 2004. Modeling the spectral shape of absorption by chromophoric dissolved organic matter. Mar. Chem. 89 (1 −4), 69–88 .
Varanasi, L. , Coscarelli, E. , Khaksari, M. , Mazzoleni, L.R. , Minakata, D. , 2018. Trans- formations of dissolved organic matter induced by UV photolysis, hydroxyl radi- cals, chlorine radicals, and sulfate radicals in aqueous-phase UV-based advanced oxidation processes. Water Res. 135, 22–30 .
Vione, D. , Minella, M. , Maurino, V. , Minero, C. , 2014. Indirect photochemistry in sun- lit surface waters: photoinduced production of reactive transient species. Chem. Eur. J. 20 (34), 10590–10606 .
Walpen, N. , Schroth, M.H. , Sander, M. , 2016. Quantification of phenolic antioxidant moieties in dissolved organic matter by flow-injection analysis with electro- chemical detection. Environ. Sci. Technol. 50 (12), 6423–6432 .
Wang, H. , Zhou, H. , Ma, J. , Nie, J. , Yan, S. , Song, W. , 2020. Triplet photochemistry of dissolved black carbon and its effects on the photochemical formation of reac- tive oxygen species. Environ. Sci. Technol. 54 (8), 4 903–4 911 .
Wang, J. , Chen, J. , Qiao, X. , Wang, Y. , Cai, X. , Zhou, C. , Zhang, Y. , Ding, G. , 2018. DOM from mariculture ponds exhibits higher reactivity on photodegradation of sul- fonamide antibiotics than from offshore seawaters. Water Res. 144, 365–372 .
Wang, J. , Chen, J. , Qiao, X. , Zhang, Y. , Uddin, M. , Guo, Z. , 2019. Disparate effects of DOM extracted from coastal seawaters and freshwaters on photodegradation of 2,4-dihydroxybenzophenone. Water Res. 151, 280–287 .
Weishaar, J.L. , Aiken, G.R. , Bergamaschi, B.A. , Fram, M.S. , Fujii, R. , Mopper, K. , 2003. Evaluation of specific ultraviolet absorbance as an indicator of the chemical composition and reactivity of dissolved organic carbon. Environ. Sci. Technol. 37 (20), 4702–4708 .
Xiang, Y. , Fang, J. , Shang, C. , 2016. Kinetics and pathways of ibuprofen degradation by the UV/chlorine advanced oxidation process. Water Res. 90, 301–308 .
Yan, M. , Korshin, G.V. , Chang, H. , 2014. Examination of disinfection by-product (DBP) formation in source waters: a study using log-transformed differential spectra. Water Res. 50, 179–188 .
Zark, M. , Dittmar, T. , 2018. Universal molecular structures in natural dissolved or- ganic matter. Nat. Commun. 9 (1), 3178 .
Zhang, D. , Yan, S. , Song, W. , 2014. Photochemically induced formation of reactive oxygen species (ROS) from effluent organic matter. Environ. Sci. Technol. 48 (21), 12645–12653 .
Zhang, K. , Parker, K.M. , 2018. Halogen radical oxidants in natural and engineered aquatic systems. Environ. Sci. Technol. 52 (17), 9579–9594 .
Zhang, T. , Cheng, L. , Ma, L. , Meng, F. , Arnold, R.G. , Saez, A.E. , 2016. Modeling the ox- idation of phenolic compounds by hydrogen peroxide photolysis. Chemosphere 161, 349–357 .
Zhang, Y. , Wang, J. , Chen, J. , Zhou, C. , Xie, Q. , 2018a. Phototransformation of 2,3-di- bromopropyl-2,4,6-tribromophenyl ether (DPTE) in natural waters: important roles of dissolved organic matter and chloride ion. Environ. Sci. Technol. 52 (18), 10490–10499 .
Zhang, Y. , Zhou, Y. , Qu, J. , Chen, J. , Zhao, J. , Lu, Y. , Li, C. , Xie, Q. , Peijnenburg, W.J.G.M. , 2018b. Unveiling the important roles of coexisting contaminants on photochem- ical transformations of pharmaceuticals: fibrate drugs as a case study. J. Hazard Mater. 358, 216–221 .
Zhao, J. , Zhou, Y. , Li, C. , Xie, Q. , Chen, J. , Chen, G. , Peijnenburg, W.J.G.M. , Zhang, Y. , Qu, J. , 2020. Development of a quantitative structure-activity relationship model for mechanistic interpretation and quantum yield prediction of singlet oxygen generation from dissolved organic matter. Sci. Total Environ. 712, 136450 .
Zhou, C. , Chen, J. , Xie, H. , Zhang, Y. , Li, Y. , Wang, Y. , Xie, Q. , Zhang, S. , 2018. Modeling photodegradation kinetics of organic micropollutants in water bodies: a case of the yellow river estuary. J. Hazard. Mater. 349, 60–67 .
Zhou, H. , Yan, S. , Lian, L. , Song, W. , 2019a. Triplet-state photochemistry of dissolved organic matter: triplet-state energy distribution and surface electric charge con- ditions. Environ. Sci. Technol. 53 (5), 2482–2490 .
Zhou, S. , Zhang, W. , Sun, J. , Zhu, S. , Li, K. , Meng, X. , Luo, J. , Shi, Z. , Zhou, D. , Critten- den, J.C. , 2019b. Oxidation mechanisms of the UV/free chlorine process: kinetic modeling and quantitative structure activity relationships. Environ. Sci. Technol. 53 (8), 4335–4345 .
Zhou, Y. , Zhao, J. , Zhang, Y. , Qu, J. , Li, C. , Qin, W. , Zhao, Y. , Chen, J. , Peijnen- burg, W.J.G.M. , 2019c. Trace amounts of fenofibrate acid sensitize the pho- todegradation of bezafibrate in effluents: mechanisms, degradation pathways, and toxicity evaluation. Chemosphere 235, 900–907 .