Phototransformation of halophenolic disinfection byproducts in receiving seawater: Kinetics, products, and toxicity

Highlights

• Photolysis rates of halophenolic DBP analogues followed iodo- > bromo- > chloro-.

• A QSAR model was developed for the photolysis of halophenolic DBPs.

• Iodophenolic DBPs phototransformed to the bromo-, chloro- and hydroxy-counterparts.

• The hydroxy-counterparts (halo(hydro)quinone) were more toxic than the parent DBPs.

• Further phototransformation of halo(hydro)quinone was a detoxification process.

Abstract

Flushing toilet with seawater is an effective method for preserving freshwater resources, but it introduces iodide and bromide ions into domestic wastewater. During chlorine disinfection, iodide and bromide ions in the saline wastewater effluent lead to the formation of iodinated and brominated aromatic disinfection byproducts (DBPs). Examples of aromatic DBPs include iodophenolic, bromophenolic and chlorophenolic compounds, which generally display substantially higher toxicity than haloaliphatic DBPs. This paper presented for the first time the rates of phototransformation of 21 newly identified halophenolic DBPs in seawater, the receiving waterbody of the wastewater effluent. The phototransformation rate constants (k) were in the range from 7.75 × 10⁻⁴ to 4.62 × 10⁻¹ h⁻¹, which gave half-lives of 1.5–895 h. A quantitative structure−activity relationship was established for the phototransformation of halophenolic DBPs as $\log k=-0.0100\times \text{Δ}{G}_f^0+5.7528\times \log MW+0.3686\times pKa-19.1607$, where ΔG_f⁰ is standard Gibbs formation energy, MW is molecular weight, and pK_a is dissociation constant. This model well predicted the k values of halophenolic DBPs. Among the tested DBPs, 2,4,6-triiodophenol and 2,6-diiodo-4-nitrophenol were found to exhibit relatively high risks on marine organisms, based on toxicity indices and half-lives. In seawater, the two DBPs underwent photonucleophilic substitutions by bromide, chloride and hydroxide ions, resulting in the conversion to their bromophenolic and chlorophenolic counterparts (which are less toxic than the parent iodophenolic DBPs) and to their hydroxyphenolic counterparts (iodo(hydro)quinones, which are more toxic than the parent iodophenolic DBPs). The formed iodo(hydro)quinones further transformed to hydroxyl-iodo(hydro)quinones, which have lower toxicity than the parent compounds.

Introduction

Flushing toilet with seawater is an effective method to conserve freshwater. This has been practiced in Hong Kong, Marshall Islands, Avalon, and Kiribati (Boehm et al., 2009; Yang et al., 2015; Liu et al., 2017). Consequently, the domestic wastewaters contain relatively high concentrations of inorganic ions such as iodide and bromide. In Hong Kong, the iodide and bromide ions in wastewater effluents have been found in the ranges of 30–60 μg/L and 20–31 mg/L, respectively (Gong and Zhang, 2013; Yang et al., 2015; Liu et al., 2017; Li et al., 2018; Gong et al., 2018). Chlorination of water, rich in bromide and iodide ions, generates a suite of brominated, iodinated, and chlorinated disinfection byproducts (DBPs) (Richardson et al., 2007; Agus et al., 2009; Song et al., 2010; Criquet et al., 2012; Tang et al., 2012; Roccaro et al., 2013; Bond et al., 2014; Hua et al., 2015; Zhu and Zhang, 2016; Sharma et al., 2017; Li and Mitch, 2018; Richardson and Postigo, 2018; Yan et al., 2016, 2018; Zhang et al., 2018; Gao et al., 2018; Jiang et al., 2018). There has been a growing concern regarding brominated and iodinated DBPs due to their substantially higher toxicity than that induced by the chlorinated counterparts (Echigo et al., 2004; Richardson et al., 2007; Dad et al., 2013; Yang and Zhang, 2013; Liu and Zhang, 2014; Sharma et al., 2014). In recent years, different groups of halophenolic DBPs have been identified in chlorinated wastewater effluents, including 5-halosalicylic acids, 4-halophenols, 2,4-dihalophenols, 2,6-dihalophenols, 2,4,6-trihalophenols, 2,6-dihalo-4-nitrophenols, 3,5-dihalo-4-hydroxybenzaldehydes, 3,5-dihalo-4-hydroxybenzoic acids, and 2,5-dibromohydroquinone (Yang and Zhang, 2016). Toxicity of twenty halophenolic DBPs and five haloaliphatic DBPs has been evaluated by measuring the growth inhibition to the marine alga Tetraselmis marina and the developmental toxicity to the marine polychaete Platynereis dumerilii (Liu and Zhang, 2014; Yang and Zhang, 2013). The results revealed that halophenolic DBPs generally induced dozens to hundreds of times higher toxicity than haloaliphatic DBPs. Moreover, of the halophenolic DBPs tested, 2,4,6-triiodophenol exhibited the highest growth inhibition to the marine alga; 2,6-diiodo-4-nitrophenol and 2,4,6-triiodophenol were two of the most toxic DBPs to the marine polychaete (Liu and Zhang, 2014; Yang and Zhang, 2013).

Chlorinated saline wastewater effluents containing DBPs are continuously discharged into seawater (the ultimate receiving water body), and consequently halophenolic DBPs (especially iodophenolic and bromophenolic ones) might chronically do harm to marine species (Yang et al., 2015). Fortunately, the solar irradiation could transform most of toxic DBPs to less toxic products, causing a decrease in the toxicity of wastewater effluents (mixtures of all DBPs) (Liu et al., 2017; Lv et al., 2017). However, some toxic halogenated DBPs were likely persistent in receiving seawater (Fig. S1 in the Supplementary Information). Although great progress has been made to understand phototransformation of DBPs from the mixture point of view, information on individual halophenolic DBPs is still lacking.

Recently, the quantitative structure‒activity relationship (QSAR) approach has been increasingly applied in studies of emerging water contaminants to establish relationships between experimental (including chemical and toxicological) observations and physicochemical properties of the molecules (Yang and Zhang, 2013; Liu and Zhang, 2014; Xiao et al., 2015; Jin et al., 2015; Borhani et al., 2016; Wang et al., 2018). This approach enables prediction of properties on the assumption that compounds with similar structures behave alike and that the property differences are attributable to enthalpy changes caused by different types and numbers of functional groups (Chen, 2011). Different QSAR models have been developed for the hydrolysis of DBPs (Wang et al., 2018; Yu and Reckhow, 2015; Chen, 2011; Glezer et al., 1999). Currently, a large number of DBPs (especially the toxic halophenolic ones) in chlorinated wastewater effluents are still unknown, and might be gradually identified and confirmed in future studies. It is important to develop a QSAR model for phototransformation of halopenolic DBPs, enabling prediction of the stability of halophenolic DBPs which were not included in this study.

Accordingly, the present paper aimed to: (i) investigate the phototransformation kinetics of various groups of halophenolic DBPs; (ii) develop a QSAR model for the phototransformation kinetics of halophenolic DBPs; (iii) delineate the phototransformation mechanisms of two selected iodophenolic DBPs, 2,4,6-triiodophenol and 2,6-diiodo-4-nitrophenol (which are of relatively high risks to marine organisms as shown later in the Results and Discussion) in seawater by identifying products and transformation pathways; and (iv) evaluate toxicity variations of the two iodophenolic DBPs during transformation against the marine polychaete P. dumerilii. This species has been successfully used in measuring the comparative toxicity of various DBPs, wastewater effluent and drinking water samples (Yang and Zhang, 2013; Yang et al., 2015; Liu et al., 2015, 2017; Li et al., 2017; Jiang et al., 2017; Han et al., 2017; Han and Zhang, 2018).