w a t e r r e s e a r c h 6 6 ( 2 0 1 4 ) 1 4 0 e1 4 8

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/watres

Photobleaching-induced changes in photosensitizing properties of dissolved organic matter Xi-Zhi Niu a, Chao Liu a, Leo Gutierrez a,b, Jean-Philippe Croue a,* a

Water Desalination and Reuse Center, King Abdullah University of Science and Technology (KAUST), Thuwal 23955-6900, Saudi Arabia b Facultad del Mar y Medio Ambiente, Universidad del Pacifico, Guayaquil, Ecuador

article info

abstract

Article history:

Photosensitizing properties of different dissolved organic matter (DOM) were investigated

Received 4 February 2014

according to their performance in singlet oxygen (1O2), triplet state of DOM (3DOM*), and

Received in revised form

hydroxyl radical ($OH) productions. The photobleaching of DOM solutions after irradiation

26 June 2014

was characterized by fluorescence excitation-emission matrix and UVeVis spectroscopy.

Accepted 18 August 2014

The photosensitizing properties of pre-irradiated DOM solutions were changed in a sun-

Available online 27 August 2014

light simulator. The performance of DOMs in photosensitized degradation of several contaminants was investigated. For a 20 h exposure, the observed degradation rate con-

Keywords:

stant (kobs) of some contaminants decreased as a function of exposure time, and highly

Dissolved organic matter

depended on the properties of both DOM and contaminant. Degradation of contaminants

Photosensitization

with lower kobs was more susceptible to DOM photobleaching-induced decrease in kobs.

DOM characterization

Under the current experimental conditions, the photobleaching-induced decrease of DOM

Photobleaching

photo-reactivity in contaminant degradation was mainly attributed to indirect photo-

Contaminant degradation

transformation of DOM caused by the interactions between photo-inductive DOM moieties and photochemically-produced reactive species. Reactive contaminants can inhibit DOM indirect photobleaching by scavenging reactive species, photosensitized degradation of these contaminants exhibited a stable kobs as a result. This is the first study to report DOM photobleaching-induced changes in the simultaneous DOM photosensitized degradation of contaminants and the inhibitory effect of reactive contaminants on DOM photobleaching. © 2014 Elsevier Ltd. All rights reserved.

1.

Introduction

The presence of dissolved organic matter (DOM) was found to significantly affect photodegradation of different aquatic components (Canonica et al., 1995; Latch et al., 2003; Remucal and McNeill, 2011). This photo-induced process has been attributed to photochemically-produced reactive species, including singlet oxygen (1O2), triplet state of DOM (3DOM*),

* Corresponding author. Tel.: þ966 2 8082984. ). E-mail address: [email protected] (J.-P. Croue http://dx.doi.org/10.1016/j.watres.2014.08.017 0043-1354/© 2014 Elsevier Ltd. All rights reserved.

hydroxyl radical ($OH), etc. (Cooper et al., 1989; Chin et al., 2004; Zhan et al., 2006; Canonica et al., 2006; Barbieri et al., 2008; Wu and Linden, 2010; Xu et al., 2011; Wang et al., 2012). Although these reactive species are produced at low concentrations (Zepp et al., 1977; Zepp et al., 1985) and are very shortlived (Zepp et al., 1985; Sharpless, 2012), they are important in the transformation of refractory contaminants that are insufficiently removed by conventional wastewater treatment plants. In addition, this DOM photosensitized chemical

w a t e r r e s e a r c h 6 6 ( 2 0 1 4 ) 1 4 0 e1 4 8

degradation was found to follow pseudo-first-order kinetics (Chin et al., 2004; Zhan et al., 2006; Canonica et al., 2006; Barbieri et al., 2008; Wu and Linden, 2010; Xu et al., 2011; Wang et al., 2012). Interestingly, even though these reactive species were originally thought to enhance the photodegradation of contaminants, some recent work have revealed possible inhibitory effects of DOM on the photosensitized degradation of contaminants by light screening, reactive species scavenging, and reaction intermediates scavenging (Canonica and Laubscher, 2008; Wenk et al., 2011). DOM can also undergo phototransformation as it is subjected to irradiation in the aquatic system. Photobleaching of DOM has been reported in numerous investigations, and can be originated from the direct destruction of DOM molecules by the UltravioleteVisible (UVeVis) light or from the reactions of DOM molecules with different reactive species. Del Vecchio and Blough (2002) studied the change of DOM before and after irradiation by UVeVis spectroscopy and fluorescence excitation emission matrix (FEEM). Results from this investigation indicated a clear DOM change. Loiselle et al. (2012) found that 3DOM* is a major reactive species that promotes the photodegradation of DOM. Brinkmann et al. (2003) investigated the photo-bleaching of hydrophobic and transphilic fractions of DOM and found that the hydrophobic fraction was not significantly changed, however, the transphilic fraction was more susceptible to irradiation. The possible role of singlet oxygen in DOM transformation has also been previously proposed (Cory et al., 2010), where lignin-like materials of DOM constituents were observed to increase with higher oxygen content, and with subsequent release of H2O2. Consequently, long exposures of DOM solutions under sunlight irradiation might change the kinetics of the reactions. Nevertheless, despite extensive research in the field, the effect of DOM photobleaching on the concurrent DOM photo-induced transformation of contaminants remains as an important gap in knowledge. In the present work, the photosensitizing properties of various DOMs before and after exposure to irradiation in a sunlight simulator were studied. Steady-state concentrations of 1O2 and $OH were quantified with specific probe compounds (Haag et al., 1984; Kochany and Bolton, 1992). 3DOM* produced in these processes was characterized with 2,4,6trimethylphenol (TMP) as a target molecule (Canonica et al., 1995). DOM solutions were characterized with fluorescence (Fluorescence Excitation Emission Matrix) and UVeVis spectroscopy. To investigate the possible role of DOM photobleaching on the indirect photodegradation of contaminants, four target molecules (i.e., ibuprofen, bisphenol A, acetaminophen, and cimetidine) were selected.

2.

Materials and methods

2.1.

Reagents and DOM isolates

Reagents were used as received: sodium phosphate monobasic/sodium phosphate diabasic (99þ% SigmaeAldrich); furfuryl alcohol (FFA, 98% Acros Organics); phenol (99.5þ% SigmaeAldrich); 2,4,6-trimethylphenol (TMP, 99% SigmaeAldrich); sodium nitrite (97þ% SigmaeAldrich); phosphoric

141

acid (85þ% in water, SigmaeAldrich); sodium bicarbonate (100% Fisher Chemical); pyridine (PYR, 99þ% Fluka Analytical); p-nitroanisole (PNA, 98þ% SigmaeAldrich); bisphenol A (99þ% SigmaeAldrich); cimetidine (99% Fluka Analytical); acetaminophen (98þ% SigmaeAldrich); ibuprofen (98þ% SigmaeAldrich). Phosphate buffer solution was prepared by titrating 5 mM Na2HPO4 with 5 mM NaH2PO4, both dissolved in ultrapure water (18.2 MU cm, Milli-Q, Millipore). All chemical stock solutions were prepared with the same buffer. Three purified natural organic matter (NOM) isolates (i.e., hydrophobic acid NOM fractions also called fulvic acids), Suwannee River hydrophobic acid fraction (SR-HPO), Beaufort River hydrophobic acid fraction (BF-HPO), and South Platte River hydrophobic acid fraction (SP-HPO), previously isolated according to the method reported by Leenheer et al. (2000) were selected for this work. In addition, two effluent organic matter (EfOM) fractions (Jeddah wastewater hydrophobic fraction (JWW-HPO) and transphilic fraction (JWW-TPI)) obtained from Jeddah wastewater Treatment Plant (JWWTP) , 2012) using a slightly different protocol (Zheng and Croue were selected.

2.2.

Experimental setup for sunlight simulator

Photo-degradation experiments with purified DOM isolates and were performed using a Suntest XLS þ sunlight simulator (ATLAS, USA) equipped with a xenon arc lamp and a filter (Atlas MTS, Cat. 56052372) to cut off the irradiance below 320 nm. Additional filters (Newport, FSQ-WG320) were placed on top of the reactors to completely avoid the UVB portion of the spectra responsible for possible direct photolysis of probes or contaminants. The solar simulator intensity was set to 400 W/m2 integrated over a wavelength range of 320e800 nm, and was maintained constant throughout the duration of all the experiments. All reactors (10 mL pyrex reactors) used were painted black to prevent light reflection and were submerged in a circulating water bath set at 25  C. An automatic pump (ColeeParmer Masterflex) was used to provide fresh room temperature water (21  C) into the water bath at constant time interval of 2 h. All sample solutions were stirred using magnetic stir bars set to 150 rpm, for maintaining the homogeneity of the solution. The power spectrum of the light source was measured by a self-built testing system. The light was collected using a Newport 77558 two-in-one coupled fused silica fiber probe with an input diameter of 1.4 mm and a transmittance range of 260e2500 nm. The fiber was connected to a NewPort 77529 Fiber matcher, which adjusted the fiber output beam to match the monochromator. The selection of detection wavelength was achieved by a Newport Cornestone 260 monochromator controlled by an NI 488.2 GPIB card. The light output power was measured using 2936C power meter equipped with an 818-UV photodetector. The photodetector was connected close to the output of monochromator for minimizing environmental noise. The data collection and processing program was developed using Labview. The spectral resolution was fixed at 0.5 nm in the measurement. The power spectrum (Fig. 1) indicated that the sunlight simulator (400 W/m2) closely mimicked the natural sunlight at King Abdullah

142

w a t e r r e s e a r c h 6 6 ( 2 0 1 4 ) 1 4 0 e1 4 8

Fig. 1 e Power Spectrum of SUNTEST and natural sunlight at KAUST (location: in front of KAUST cafeteria, Time: 3 pm, 22-May-2013).

University of Science and Technology (KAUST), Thuwal, Saudi Arabia (22 N, 39 E) in terms of spectral distribution. Both natural sunlight and simulated light exhibited low portions of UVB and below. Applied glass filter showed a high transparency in the wavelength range of 320 nm and above while cutting off wavelength range of 320 nm and below.

2.3.

Analytical methods

A High Performance Liquid Chromatography instrument (HPLC, Waters, USA) equipped with a UV-Detector (Waters, USA) was used for the quantification of contaminants, furfuryl alcohol, phenol, and TMP. An XDB-C18 column (Agilent Technologies) was used to separate the contaminants. Detailed HPLC methods are presented in the supplementary materials (Table S1). The total organic carbon (TOC) measurement of raw effluent and reconstituted DOM solutions was performed using a TOC Analyzer (Shimadzu, Japan) with a calibration range of 1e3 mg C/L. Duplicates were conducted for every measurement. A spectrofluorometer (HORIBA FluoroMax-4) was used to study the possible change of DOM solutions during irradiation experiments at excitation wavelength of 240e500 nm and at emission wavelength of 290e550 nm in a 1 cm quartz cell. A 2550 UVeVis spectrometer (Shimadzu, Japan) was used for absorbance measurements of contaminants and DOM solutions.

2.4.

Experimental procedures

L, pH ¼ 7.3) was studied and compared to JWW-HPO and JWWTPI (TOC ¼ 4 mg C/L, pH 7.3), for verifying the reliability of reconstituted EfOM (JWW-HPO and JWW-TPI were originally isolated from JWWTP effluent). Filtered solutions (adjusted to TOC ¼ 4.0 mg C/L) of the water samples were added to the reactors along with 10 mM TMP. Solutions were irradiated in the solar simulator. Samples (600 mL) were collected at regular time intervals in 2 mL amber glass vials and stored in the dark until analysis. The decay of TMP was analyzed by HPLC. Dark controls were also included for all chemical probe experiments to confirm that other degradation pathways (e.g., hydrolysis and non-photolytic oxidation reactions) were not responsible for the probe compound decay. The steady-state concentrations of 1O2 and $OH in irradiated DOM solutions were indirectly measured by monitoring the decay of probe compounds furfuryl alcohol (FFA) and phenol, respectively (Haag et al., 1984; Kochany and Bolton, 1992). The rate constants of the probe compounds with reactive species (i.e., 1O2 and $OH) has been previously documented (1.2  108 M1 s1 for FFA and 1.4  1010 M1 s1 for phenol, respectively) (Haag et al., 1984; Kochany and Bolton, 1992). Initial concentrations of probe compounds in the reactors were 10 mM for phenol and 40 mM for FFA. Reaction solutions were buffered to pH 8.0 with 5 mM phosphate buffer and then irradiated in the solar simulator under the same conditions as described for photo-degradation experiments. Aliquots (600 mL) were sampled at regular intervals and transferred to 2 mL amber glass vials with a crimp seal. Probe compounds were immediately analyzed after sampling by HPLC.

2.4.2.

DOM solutions irradiation experiments

4 mg C/L of BF-HPO, JWW-HPO, and JWW-TPI solutions (i.e., previously adjusted to pH 8.0 with 5 mM phosphate buffer) were irradiated for 12 h at a wavelength range of 300e800 nm. Their performance in photosensitized TMP degradation, and photochemical yields of 1O2 and $OH after pre-irradiation were then investigated.

2.4.3.

Photosensitized degradation of contaminants

The photodegradation of different contaminants (acetaminophen, ibuprofen, bisphenol-A, and cimetidine) at an initial concentration of 10 mM was studied in reactors containing purified DOM isolates (20 mg C/L), and buffered with 5 mM phosphate (pH ¼ 8.0). All reactors contained an initial sample volume of 8 mL. Sample aliquots (600 mL) were collected at different time intervals, and immediately analyzed by HPLC.

3.

Results and discussion

3.1. Production of reactive species from photosensitization of DOM

2.4.1. Quantification of reactive species generated by the photosensitization of DOM

3.1.1.

2,4,6-Trimethylphenol (TMP) was used as a probe compound to investigate the production of 3DOM* (Canonica et al., 1995; Canonica and Freiburghaus, 2001). Solutions were prepared with 5 mM phosphate buffer (pH ¼ 8.0), 10 mM TMP, and 4 or 20 mg C/L DOM. In addition, JWWTP effluent (TOC ¼ 4.3 mg C/

As the energy source of 1O2 (Haag and Hoigne, 1986), possible precursor of $OH (Dong and Rosario-Ortiz, 2012), and being reactive itself, 3DOM* is a very important reactive species in the photosensitization processes in irradiated DOM solutions. 3 DOM* can react with probes either via energy transfer (e.g.,

Quantification of 3DOM* with 2,4,6-trimethylphenol

143

w a t e r r e s e a r c h 6 6 ( 2 0 1 4 ) 1 4 0 e1 4 8

sorbic acid (Grebel et al., 2011)) or via electron transfer (e.g., TMP (Canonica et al., 1995)). However, a probe compound to exclusively detect all 3DOM* is not available to the current state of knowledge. To investigate the reaction properties of 3 DOM*, TMP was selected because it is predominantly transformed by 3DOM* in these photosensitized processes (Canonica et al., 1995). TMP degradation rate constant (kobs, TMP) of 4 mg C/L JWW-TPI and JWW-HPO hybrid (pH 7.3, 1:2 ratio, similar to the isolation recovery) showed a high similarity with that of raw effluent, demonstrating the reliability of the purified isolates in phosphate buffer (Fig. S1). Table 1 shows kobs, TMP in the presence of different DOMs isolated from natural waters (i.e., SR-HPO, SP-HPO, and BFHPO) and wastewater effluent (JWW-HPO and JWW-TPI). These results indicate that the DOMs photosensitized degradation efficiencies of TMP highly depended on their origins. The range of kobs, TMP values obtained for the different DOM solutions are comparable to those reported in previous investigations (Canonica and Freiburghaus, 2001). Similar experiments conducted at higher DOM concentrations (20 mg C/L TOC), specific UV Absorbance (SUVA) and light screening correction factor (CF) due to DOM absorbance are shown in Table 1. The correction factor is defined as the ratio of light absorbed at optically thin conditions over the light absorbed at optically thick conditions (Kohn et al., 2007). This value is used to evaluate the attenuation of irradiation intensity due to light screening by DOM. The method for CF calculation was also described in the previous work by Kohn et al. (2007). The kTMP (value at optically thin conditions, where kTMP ¼ kobs, TMP  CF) at 4 and 20 mg C/L TOC concentrations followed the same trend (in terms of relative value), i.e., an increase in TOC was followed by an increase in kTMP. However, this increase was not directly proportional to TOC concentration, which can be ascribed to a possible increase in internal quenching of 3DOM* and to the increased quenching rate constant in the solutions as described in Eq-1: ½3 DOM*ss; i ¼

ai ki þ ki;TMP  ½TMP

(1)

where [3DOM*]ss,i (M) is the steady-state concentration of triplet state of DOMi, ai is a zero-order formation rate constant of a specific 3DOM*i (M s1), ki is the first-order quenching rate constant of the solution before adding TMP, ki, TMP is the second-order quenching rate constant between TMP and

3 DOM*i, and ki, TMP  [TMP] is the quenching rate constant of TMP which is proportional to TMP concentration. The increase of DOM concentration will increase ki, consequently explaining the deviation in proportionality.

3.1.2. Quantification of 1O2 and ·OH in DOM photosensitization processes FFA and phenol were used as probe compounds for determining steady-state concentrations of 1O2 and $OH respectively. The reaction of probe compound with corresponding reactive species under a pseudo-first-order kinetics is presented in Eq-2: 0

kobs ¼ ½RSSS  kq

(2)

1

where k’obs (s ) is the observed rate constant of the probe compound reacting with the corresponding reactive species, [RS]SS (M) refers to the steady-state concentration of the reactive species, and kq (M1 s1) is the reaction rate constant between the probe compound and the reactive species. [RS]SS can be obtained with a known kq (1.2  108 M1 s1 for FFA, Haag and Hoigne, 1986; 1.4  1010 M1 s1 for phenol, Kochany and Bolton, 1992), while the k’obs is measured from the experiments. Steady-state concentrations of 1O2 and $OH are listed for 4 mg C/L and 20 mg C/L TOC concentrations in Table 1. BF-HPO showed significantly higher 1O2 and $OH yield than other DOM sensitizers at 4 mg C/L and 20 mg C/L TOC. Compared to JWW-TPI, JWW-HPO showed higher 1O2 yield, but lower $OH yield. Increasing TOC enhanced the production of both reactive species, however, a proportional enhancement of reactive species production with TOC concentration was not observed (i.e., similar to the case of TMP degradation).

3.2.

Characterizing photobleaching of DOMs

DOM molecules during irradiation can undergo photolysis or be subjected to photochemically-produced reactive species, which partially agrees with kobs,TMP (TOC ¼ 20 mg C/ L) < 5  kobs,TMP (TOC ¼ 4 mg C/L). Previous studies have also investigated changes in DOM characteristics during irradiation (Del Vecchio and Blough, 2002; White et al., 2003; Brinkmann et al., 2003; Loiselle et al., 2012; Sharpless et al., 2014). Understanding the impact of these changes on the photochemical properties of DOMs will help to predict

Table 1 e Steady-state concentrations of 1O2, ·OH, and kobs,TMP in different DOM solutions. 1

13

O2 (10

M)

$OH (1016 M) kobs,TMP (h1) Correction factor SUVAc a b c

TOCb

JWW-HPO

JWW-TPI

BF-HPO

SR-HPO

SP-HPO

4 20a 4 20a 4 20a 4 20

0.61 ± 0.1 3.6 ± 0.3 1.7 ± 0.1 3.2 ± 0.2 0.08 ± 0 0.15 ± 0.2 1.03 1.29 3.3

0.34 ± 0.3 2.3 ± 0.1 2.0 ± 0.3 4.1 ± 0.1 0.13 ± 0.01 0.22 ± 0.01 1.02 1.16 2.1

0.94 ± 0.04 4.2 ± 0.2 3.1 ± 0.1 7.5 ± 0.3 0.15 ± 0.01 0.15 ± 0.05 1.06 1.43 3.8

ND 1.9 ± 0 ND 4.1 ± 0 0.09 ± 0.01 0.12 ± 0.01 1.07 1.58 4.6

ND 1.85 ± 0.05 ND 6.4 ± 0.2 0.14 ± 0.01 0.24 ± 0.02 1.03 1.22 2.9

Values for 20 mg C/L are presented before correction factor calculation. ND: not detected. Unit: mg C/L. Unit: L mg C1 m1.

144

w a t e r r e s e a r c h 6 6 ( 2 0 1 4 ) 1 4 0 e1 4 8

possible influence of DOM photobleaching on photosensitized contaminants degradation. Therefore, UV-254 (UV absorbance at 254 nm) and FEEM was used to investigate the influence of pre-irradiation on the photochemical properties of JWW-HPO, JWW-TPI, and BF-HPO solutions and to reveal possible changes in DOM characteristics. FEEM signals indicated the presence of fulvic-like chromophores, humic-like chromophores, and protein-like chromophores in DOM isolates according to the characterization protocols described elsewhere (Chen et al., 2003). FEEM signal distribution of JWW-TPI resembled that of the transphilic fraction of wastewater treatment plants effluent (91st Ave. WWTP effluent, Phoenix, AZ) reported by Chen et al. (2003). However, different from the hydrophobic fraction reported in the same work, FEEM signal distribution of JWW-HPO showed more predominant signal in humic-like zone. FEEM of SR-HPO analyzed by Chen et al. (2003) resembled that of the current work. Peak data point selections (Baker, 2001), single-scan emission spectra and volumetric integration (Chen et al., 2003) were previously described and used for the FEEM analysis. In this study, maximum values (peak point data) of every zone (i.e., fulvic-like, humic-like, and protein-like chromophores) were selected at each area and listed in Table 2. Peak shifts were observed between non-irradiated and irradiated samples due to photodegradation of chromophores. Peak positions of non-irradiated samples were selected as comparison sites on FEEM for both solutions before and after irradiation. A Significant decrease in fulvic-like and humiclike chromophore signals was observed in all pre-irradiated solutions, however, protein-like chromophore signals were not significantly changed (Table 2). TMP, FFA, and phenol were then used to quantify the potential change in DOM-photosensitizing properties in preirradiated samples, by examining kobs,TMP, 1O2 and $OH yields respectively (Fig. 2). The kobs,TMP decreased in preirradiated JWW-TPI and BF-HPO solutions, however, no significant change was observed in JWW-HPO. This result indicates that JWW-HPO after irradiation remained stable in terms of 3DOM* properties as characterized by TMP. As aforementioned, JWW-HPO (hydrophobic fraction of a WWTP effluent collected after biological treatment) is less reactive than BF-HPO (hydrophobic DOM fraction isolated from river

Fig. 2 e Comparison of photochemical parameters between irradiated samples and non-irradiated samples. P on the vertical axis refers to the tested photochemical parameters of the solutions (kobs,TMP, [·OH]ss, [1O2]ss), the horizontal dash line represents Pirradiated/PNon-irradiated ¼ 1 (pH ¼ 8.0, TOC ¼ 4 mg C/L, irradiation time: 12 h, irradiation wavelength: 300e800 nm).

water) (Table 1). Humic-like chromophores present in river water DOM mainly originate from terrestrial sources (i.e., , 2003) and are charlignin derivatives) (Leenheer and Croue acterized as strong photosensitizers (Table 1) and subjected to photobleaching. Contrariwise, aromatic moieties present in humic-like structures, i.e., HPO fraction, originating from microbial synthesis and potentially from wastewater, appeared to be less sensitive to solar irradiation than terrestrial HPO fractions. The photo-induced properties of JWW-TPI, a fraction that incorporates smaller and more photoreactive molecules than JWW-HPO, were more impacted by preirradiation. The capabilities of JWW-HPO, JWW-TPI, and BF-HPO solutions for producing 1O2 and $OH showed significant decrease due to pre-irradiation. The 1O2 and $OH yield in irradiated

Table 2 e UV-254 and FEEM characteristics (maximum values) of DOM solutions before and after pre-irradiation. JWW-HPO lex/lemb FEEM

a

Humic-like Fulvic-like Protein-like

UV-254c

Before After Before After Before After Before After

330/435 240/420 275/305

Before: before irradiation, after: after irradiation. a Unit: cps. b Unit: nm. c UV absorbance measured at 254 nm, unit: cm1.

JWW-TPI

Intensity 2.61  1.25  5.32  3.65  2.74  2.50  0.138 0.129

5

10 105 105 105 105 105

BF-HPO

lex/lem

Intensity

lex/lem

Intensity

330/410

5

320/440

1.79  1.59  4.06  3.77  1.10  1.10  0.169 0.147

245/435 280/310

2.69  1.61  4.84  3.47  2.52  2.59  0.091 0.082

10 105 105 105 105 105

245/445 275/305

105 105 105 105 105 105

w a t e r r e s e a r c h 6 6 ( 2 0 1 4 ) 1 4 0 e1 4 8

JWW-HPO samples decreased, however, 3DOM* probed by TMP was stable (Fig. 2). The indirect quantification of 1O2 and $OH can be affected by the change of DOM hydrophobicity, the difference in hydrophobicity would determine the absorptive characteristic between DOM molecule and probe compounds (Latch and McNeill, 2006). Therefore, a possible cause of this observation is that the molecular properties of JWW-HPO were modified during the irradiation process (i.e., SUVA), which consequently changed the bimolecular absorptive behavior between probe compound and JWW-HPO. Change in the hydrophobicity of DOM would then have an impact on the 1 O2 and $OH yield obtained by the indirect quantification protocol, even when no significant change was observed for 3 JWW-HPO*. Another possible cause is that some 3DOM* cannot be probed by TMP, but are responsible for the production of 1O2 and $OH. A decrease of this class of 3DOM* thus will not be fully revealed by TMP.

3.3. Role of DOM photobleaching on photosensitized degradation of contaminants The photosensitized degradation of selected contaminants was investigated within an irradiation time range of 20 h (Fig. 3). The degradation rates of some contaminants followed pseudo-first order kinetics, however, in other specific cases, the instant kobs decreased after longer irradiation time (i.e., after 8 h) (Fig. 3). This result indicates that this phenomenon is selective and depends on the properties of both DOM and contaminant. This implies that a decrease in photodegradation capacity was introduced in the irradiation procedure, however, it was unknown whether this decline was

145

induced by directly or indirectly photodegraded DOM chromophores. Photobleaching of DOM consists of direct and indirect bleaching of the chromophores, where indirect photobleaching is the destruction of chromophores by photochemically-produced reactive species and was reported to be a smaller contributor to DOM photobleaching compared with direct transformation (Del Vecchio and Blough, 2002). The photosensitized degradation of cimetidine in the presence of different DOMs exhibited well-fitting pseudo-firstorder kinetics (Fig. 3-a), and all the UVeVis spectra of applied contaminants (10 mM) showed insignificant absorbance within the irradiation wavelength range (Fig. S2). This result indicates that DOM direct photobleaching did not seem to impact the performance in photosensitized contaminant degradation. JWW-HPO showed well-fitting pseudo-first-order photochemical degradation kinetics with all different contaminants, suggesting that its photochemical properties have not been changed during the time of irradiation. In addition, JWW-TPI showed instability under irradiation. Both results are consistent with the photochemical parameters analyzed in Fig. 2). Photosensitized processes, including formation and scavenging of reactive species, reactions between reactive species and contaminants, and reactions between reactive species and DOM molecules, are schematically described in Fig. 4. If kobs is compared among different contaminants photodegraded in a same DOM solution, the instable kobs only occurred in less reactive contaminants. For instance, as of JWW-TPI, the photodegradation of acetaminophen, ibuprofen, and bispenol-A exhibited instable kobs, while the much more reactive cimetidine showed a stable pseudo-first

Fig. 3 e Photosensitized degradation of contaminants in DOM solutions: (a) cimetidine (concentrations of cimetidine photosensitized by SP-HPO and BF-HPO exceeded lower detection limits of UV detector after 20 h); (b) ibuprofen; (c) bisphenol-A; (d) acetaminophen (pH ¼ 8.0, TOC ¼ 20 mg C/L, contaminants concentration ¼ 10 mM). Dash curves were plotted to show the samples with observed change in kinetics over irradiation time.

146

w a t e r r e s e a r c h 6 6 ( 2 0 1 4 ) 1 4 0 e1 4 8

Fig. 4 e Schematic pathways of contaminants and DOM interaction in photosensitized processes (a refers to the formation rate of 3DOM*; kRS refers to the reaction rate constant of DOM with reactive species (3DOM*, ·OH, and 1 O2); kq refers to the rate constant of contaminants with reactive species; ks refer to the solutions quenching rate constants for reactive species; the red broken lines correspond to DOM photobleaching processes, and the blue solid lines correspond to DOM photosensitized contaminant degradation processes). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

order kinetics (Fig. 3 and Table S2). kobs hereby can be considered as an overall quenching capacity parameter of the contaminant towards reactive species (kq in Fig. 4 and eq-3). Lower kobs indicates inferior quenching capacity, therefore, reactive species can then react with DOM molecules to induce a change in DOM photochemical properties (process denoted by kRS in Fig. 4). On the contrary, if the quenching capacity of these contaminants are high enough (e.g. cimetidine), they can readily function as reactive species scavenger and protect the DOM molecules from reactive species attack (Fig. 4). The reaction between DOM and reactive species is presented in eq-3 (derivation process is available in the SI): RDOM ¼

kRS  ai   kRS þ ks þ kq  contaminant

(3)

where RDOM refers to the reaction rate of a specific reactive species (RS) towards DOM (M s1), kRS refers to the first-order reaction rate constant of reactive species with DOM, ks refers to the first-order quenching rate constant from other components of the solution, ai is the zero-order formation rate constant of the reactive species, and kq refers to the secondorder rate constant of this reactive species with a specific contaminant. A high kq will result in a low RDOM of the DOM with reactive species of concern which reduces the possibility of this DOM of being transformed, and vice versa. If a specific DOM is susceptible to a particular reactive species (high kRS), coexistence of contaminants with higher kq will ‘protect’ the DOM

molecule by quenching this reactive species, which is then reflected by a stable kobs (photosensitized contaminant degradation) as a feedback (Fig. 4). To evaluate the above-discussed DOM ‘protection’ effect from concomitant reactive contaminants in the photosensitized degradation processes, JWW-TPI was selected to undergo photodegradation in the sunlight simulator under same experimental condition in the presence of cimetidine (10 mM) or ibuprofen (10 mM), respectively. A comparison of their FEEM mappings (Fig. 5) showed significant attenuation after 20 h irradiation, especially the humic-like and fulvic-like chromophores, however, protein-like chromophores were not considerably changed, which agrees with the DOM solution analysis presented in Table 2. The loss in signals in this system was classified into three groups: chromophores that were only indirectly transformed and were preserved by reactive contaminants (C-1); chromophores that were only indirectly transformed but were not preserved by reactive contaminants (C-2); and directly photodegradable chromophores, either transformed by direct or indirect photolysis (C3). C-2 presents because quenchers of RS cannot exclusively scavenge all RS produced in the matrix. This ‘unsheltered’ group should be considerably less abundant compared with C-1 and C-3 on the ground that cimetidine has high kobs (Fig. 3 and Table S2) and high reactivity with 1O2 and $OH (Latch et al., 2003). The existence of reactive quenching chemicals, cimetidine and ibuprofen, restrained the attenuation of FEEM signals. However, JWW-TPI solutions irradiated with these two contaminants still exhibited significant decay in these signals, which suggests that the ‘unprotected’ chromophores (C-2 and C-3) of DOM accounted for much more significant loss in FEEM signals. Cimetidine and ibuprofen protected JWW-TPI (mainly humic-like and fulvic-like chromophores) by competing the reactive species and by reducing the reaction rate constant between reactive species and JWW-TPI (Fig. 4 and Eq-3). Cimetidine and ibuprofen were unlikely to interfere with the direct photolysis of JWW-TPI as the UVeVis absorbance in the range of 320e800 nm was insignificant (Fig. S2). Moreover, the more reactive cimetidine exhibited advantageous inhibiting effect on the indirect photobleaching of JWW-TPI (Fig. 5c). As a result, the rate constant (kobs) of cemitidine did not decrease because the functional groups (belonging to C-1) were preserved. Likewise, ibuprofen, with a much lower kobs, showed a decrease in rate constant (Fig. 5b and Fig. 4b). The current results demonstrated the inhibitory effect of reactive chemicals in the DOM solutions on DOM indirect photobleaching, which simultaneously guaranteed the sustainability of the photo-inductive functional groups (belonging to C-1) from DOM molecules. The signal loss observed from Fig. 5a to c is ascribed to the destruction of chromophores that were not protected by cimetidine (C-2 and C-3). These functional groups may not be responsible for the photosensitized degradation of contaminants since the loss of these fractions did not change the photosensitized degradation kinetics. The photo-inductive functional groups (e.g., ketone groups (Canonica et al., 1995)) are probably more transformed by reacting with photochemically-produced reactive species. (C-1 and signal differences among Fig. 5b, c, and d).

w a t e r r e s e a r c h 6 6 ( 2 0 1 4 ) 1 4 0 e1 4 8

147

Fig. 5 e Fluorescence Excitation-Emission Matrix signals of JWW-TPI solutions (Fluorescence signal unit: £ 105 cps, TOC ¼ 20 mg C/L, irradiation time: 20 h, irradiation wavelength: 320e800 nm; (a) JWW-TPI solution before irradiation; (b) JWW-TPI solution after 20 h irradiation with existence of ibuprofen; (c) JWW-TPI solution after 20 h irradiation with existence of cimetidine; (d) JWW-TPI solution after 20 h irradiation; P, F, and H in Figure a refer to responding signals for protein-like, fulvic-like, and humic-like chromophores (according to protocols by Chen et al. (2003)). Note: signal intensity scale of (a) was different from that of (b), (c), and (d).

4.

Conclusions

1) The photosensitizing properties of DOM studied in this investigation were dependent on their origins and fractions (hydrophobics or transphilics). The increase of DOM concentration also increases the internal quenching of reactive species. 2) Photobleaching of DOM degraded some fractions, resulting in a decreased production of 1O2, $OH, and 3DOM*. The photosensitizing properties of transphilic fraction of Jeddah wastewater effluent organic matter (JWW-TPI) were more susceptible to photobleaching. 3) Photosensitized degradation of contaminants in the presence of DOM was also impacted by photobleaching of DOM, especially for contaminants with lower kobs. DOM functional groups responsible for the sensitized degradation of contaminants were mainly eliminated by photochemically produced reactive species. Functional groups that are not preserved by reactive contaminants, most probably directly photodegradable groups, did not seem to be responsible for the sensitized degradation of contaminants. However, destruction of these groups accounted for major losses in FEEM signals of photobleached DOMs. Contaminants with high kobs during their photosensitized degradation with DOMs inhibited the phototransformation of photosensitizable groups in DOM molecules by quenching the reactive species, photosensitized degradation of these contaminants exhibited a stable kobs as a result.

Acknowledgment We acknowledge the designer of the power spectrum measurement system Mr. Chao Shen from Photonics Lab at KAUST and Hanting Wang, Shahid Rosado, and Ofelia Romero at UIUC for their kind help in experimental operations. We also would like to acknowledge Water Desalination and Reuse Center of KAUST.

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.watres.2014.08.017.

references

Baker, A., 2001. Fluorescence excitationeemission matrix characterization of some sewage-impacted rivers. Environ. Sci. Technol. 35 (5), 948e953. Barbieri, Y., Massad, W.A., Dı´az, D.J., Sanz, J., Amat-Guerri, F., Garcı´a, N.A., 2008. Photodegradation of bisphenol A and related compounds under natural-like conditions in the presence of riboflavin: kinetics, mechanism and photoproducts. Chemosphere 73 (4), 564e571. Brinkmann, T., Sartorius, D., Frimmel, F.H., 2003. Photobleaching of humic rich dissolved organic matter. Aquat. Sci. 65 (4), 415e424.

148

w a t e r r e s e a r c h 6 6 ( 2 0 1 4 ) 1 4 0 e1 4 8

Canonica, S., Jans, U., Stemmler, K., Hoigne, J., 1995. Transformation kinetics of phenols in water: photosensitization by dissolved natural organic material and aromatic ketones. Environ. Sci. Technol. 29 (7), 1822e1831. Canonica, S., Freiburghaus, M., 2001. Electron-rich phenols for probing the photochemical reactivity of freshwaters. Environ. Sci. Technol. 35 (4), 690e695. Canonica, S., Hellrung, B., Muller, P., Wirz, J., 2006. Aqueous oxidation of phenylurea herbicides by triplet aromatic ketones. Environ. Sci. Technol. 40 (21), 6636e6641. Canonica, S., Laubscher, H.U., 2008. Inhibitory effect of dissolved organic matter on triplet-induced oxidation of aquatic contaminants. Photochem. Photobiol. Sci. 7 (5), 547e551. Chen, W., Westerhoff, P., Leenheer, J.A., Booksh, K., 2003. Fluorescence excitation-emission matrix regional integration to quantify spectra for dissolved organic matter. Environ. Sci. Technol. 37 (24), 5701e5710. Chin, Y.P., Miller, P.L., Zeng, L.K., Cawley, K., Weavers, L.K., 2004. Photosensitized degradation of bisphenol a by dissolved organic matter. Environ. Sci. Technol. 38 (22), 5888e5894. Cooper, W.J., Zika, R.G., Petasne, R.G., Fischer, A.M., 1989. Sunlightinduced photochemistry of humic substances in natural waters: major reactive species. Adv. Chem. Ser. 219, 333e362. Cory, R.M., McNeill, K., Cotner, J.P., Amado, A., Purcell, J.M., Marshall, A.G., 2010. Singlet oxygen in the coupled photochemical and biochemical oxidation of dissolved organic matter. Environ. Sci. Technol. 44 (10), 3683e3689. Del Vecchio, R., Blough, N.V., 2002. Photobleaching of chromophoric dissolved organic matter in natural waters: kinetics and modeling. Mar. Chem. 78 (4), 231e253. Dong, M.M., Rosario-Ortiz, F.L., 2012. Photochemical formation of hydroxyl radical from effluent organic matter. Environ. Sci. Technol. 46 (7), 3788e3794. Grebel, J.E., Pignatello, J.J., Mitch, W.A., 2011. Sorbic acid as a quantitative probe for the formation, scavenging and steadystate concentrations of the triplet-excited state of organic compounds. Water Res. 45 (19), 6535e6544. Haag, W.R., Hoigne, J., Gassman, E., Braun, A.M., 1984. Singlet oxygen in surface waters 1. Furfuryl alcohol as a trapping agent. Chemosphere 13 (5e6), 631e640. Haag, W.R., Hoigne, J., 1986. Singlet oxygen in surface waters .3. Photochemical formation and steady-state concentrations in various types of waters. Environ. Sci. Technol. 20 (4), 341e348. Kochany, J., Bolton, J.R., 1992. Mechanism of photodegradation of aqueous organic Pollutants 2. Measurement of the primary rate constants for reaction of OH radicals with benzene and some halobenzenes using an Epr Spin-Trapping method following the photolysis of H2O2. Environ. Sci. Technol. 26 (2), 262e265. Kohn, T., Grandbois, M., McNeill, K., Nelson, K.L., 2007. Association with natural organic matter enhances the sunlight-mediated inactivation of MS2 coliphage by singlet oxygen. Environ. Sci. Technol. 41 (13), 4626e4632. Latch, D.E., Stender, B.L., Packer, J.L., Arnold, W.A., McNeill, K., 2003. Photochemical fate of pharmaceuticals in the environment: cimetidine and ranitidine. Environ. Sci. Technol. 37 (15), 3342e3350. Latch, D.E., McNeill, K., 2006. Microheterogeneity of singlet oxygen distributions in irradiated humic acid solutions. Science 311 (5768), 1743e1747.

, J.P., Benjamin, M., Korshin, G.V., Leenheer, J.A., Croue Hwang, C.J., Bruchet, A., Aiken, G., 2000. Comprehensive isolation of natural organic matter from water for spectral characterizations and reactivity testing. In: Barrett, S.E., Krasner, S.W., Amy, G.L. (Eds.), Natural Organic Matter and Disinfection By-products: Characterization and Control in Drinking Water, ACS Symposium Series 761. American Chemical Society, pp. 68e83. , J.P., 2003. Peer reviewed: characterizing Leenheer, J.A., Croue aquatic dissolved organic matter. Environ. Sci. Technol. 37 (1), 18Ae26A. Loiselle, S., Vione, D., Minero, C., Maurino, V., Tognazzi, A., Dattilo, A.M., Rossi, C., Bracchini, L., 2012. Chemical and optical phototransformation of dissolved organic matter. Water Res. 46 (10), 3197e3207. Remucal, C.K., McNeill, K., 2011. Photosensitized Amino acid degradation in the presence of riboflavin and its derivatives. Environ. Sci. Technol. 45 (12), 5230e5237. Sharpless, C.M., 2012. Lifetimes of triplet dissolved natural organic matter (DOM) and the effect of NaBH4 reduction on singlet oxygen quantum yields: implications for DOM photophysics. Environ. Sci. Technol. 46 (8), 4466e4473. Sharpless, C.M., Aeschbacher, M., Page, S.E., Wenk, J., Sander, M., McNeill, K., 2014. Photooxidation-induced changes in optical, electrochemical and photochemical properties of humic substances. Environ. Sci. Technol. 48 (5), 2688e2696. Wang, L., Xu, H.M., Cooper, W.J., Song, W.H., 2012. Photochemical fate of beta-blockers in NOM enriched waters. Sci. Total Environ. 426, 289e295. Wenk, J., von Gunten, U., Canonica, S., 2011. Effect of dissolved organic matter on the transformation of contaminants induced by excited triplet states and the hydroxyl radical. Environ. Sci. Technol. 45 (4), 1334e1340. White, E.M., Vaughan, P.P., Zepp, R.G., 2003. Role of the photoFenton reaction in the production of hydroxyl radicals and photobleaching of colored dissolved organic matter in a coastal river of the southeastern United States. Aquatic Sci. e Res. across Boundaries 65 (4), 402e414. Wu, C., Linden, K.G., 2010. Phototransformation of selected organophosphorus pesticides: roles of hydroxyl and carbonate radicals. Water Res. 44 (12), 3585e3594. Xu, H.M., Cooper, W.J., Jung, J., Song, W.H., 2011. Photosensitized degradation of amoxicillin in natural organic matter isolate solutions. Water Res. 45 (2), 632e638. Zepp, R.G., Wolfe, N.L., Baughman, G.L., Hollis, R.C., 1977. Singlet oxygen in natural-waters. Nature 267 (5610), 421e423. Zepp, R.G., Schlotzhauer, P.F., Sink, R.M., 1985. Photosensitized transformations involving electronic-energy transfer in natural-waters e role of humic substances. Environ. Sci. Technol. 19 (1), 74e81. Zhan, M., Yang, X., Xian, Q., Kong, L., 2006. Photosensitized degradation of bisphenol A involving reactive oxygen species in the presence of humic substances. Chemosphere 63 (3), 378e386. , J.P., 2012. Contribution of different effluent Zheng, X., Croue organic matter fractions to membrane fouling in ultrafiltration of treated domestic wastewater. J. Water Reuse Desalination 2 (4), 204e209.