Prechlorination of algae-laden water: the effects of ammonia on chlorinated disinfection byproduct formation during long-distance transportation

Algal blooms in drinking water sources pose a significant threat to the safety of drinking water supply (Mowe et al. 2015; Treuer et al. 2021; Yin et al. 2023; Zhang et al. 2023; Shi et al. 2024). This is due to the release of odorous substances and algal toxins by algal cells (Murik et al. 2014; Jia et al. 2016; Song 2023). In addition, the significant presence of algae can also lead to incomplete removal during drinking water treatment, resulting in high algae content in the treated water (Oliver 1983; Falconer 1999; Shen et al. 2011; Yu et al. 2023).

To address the challenges posed by an algal bloom, prechlorination is commonly employed as a pretreatment method to enhance coagulation and algae removal efficiency (Sheng et al. 2023; Zheng et al. 2023). However, excessive prechlorination can cause the rupture of algae cells, thereby reducing the coagulation effect (Lin et al. 2018, 2022). Additionally, the intracellular organic matter (IOM) of algal cells is an important precursor of DBPs, so it is crucial to control the degree of prechlorination (Wert & Rosario-Ortiz 2013; Hua et al. 2019). In practical engineering, the transportation time of drinking water from its source to the drinking water treatment plant can vary from several hours to over 10 h. The existence of a large number of algal cells can cause damage to the pipeline (Wang et al. 2023). Therefore, the preoxidation of algal cells at the drinking water intake point is commonly used to alleviate pipeline damage (Tang et al. 2015; Zhou et al. 2017; Sheng et al. 2023). However, preoxidation has been reported to induce the spontaneous and continuous damage of algal cells during water transportation, which would result in the production of a large amount of DBP precursor and the ineffectiveness of coagulation (Steller 1995; Ameisen 2002; Qian et al. 2010; Wert & Rosario-Ortiz 2013; Fan et al. 2016; Qi et al. 2016a; Li et al. 2023). Therefore, it is essential to carefully control the dosage according to the water transportation time to minimize IOM release and improve the efficiency of coagulation-based algal removal processes (Vavilala et al. 2016; Sun et al. 2022).

Surface water pollution globally has reached critical levels, with a progressively increasing trend in content over the years due to various anthropogenic activities (Zhang et al. 2022; Somma et al. 2023; Sun et al. 2023; Zhang et al. 2024). Upon the addition of chlorine to water bodies, an immediate reaction occurs between chlorine and ⁠, resulting in the formation of chloramines (Pan et al. 2023). Consequently, the prechlorination process employed in water treatment facilities could effectively be regarded as a chloramination process. Unlike direct chlorine disinfection, the N-DBPs produced during chloramine disinfection have stronger genotoxicity and cytotoxicity. The protective role of substantially impacts the integrity of algal cells and the generation of chlorination DBPs in long-distance water transportation following prechlorination. Therefore, it is of great significance to pay attention to the impact of high algae water and long-distance transportation on reducing the safety risk of drinking water.

Based on the aforementioned concerns, this study aims to: (1) investigate the effects of transportation time and on algal cell integrity after prechlorination; (2) detect the variation of released AOM induced by prechlorination during simulated transportation processes; and (3) evaluate chlorinated DBP formation from and transportation time aspects.

The algal species used in this study was Microcystis aeruginosa (M. aeruginosa) because of its prevalence in algal blooms and relevance to water quality and treatment challenges in waterworks. M. aeruginosa (strain FACHB905), a cyanobacterium previously described by Shen et al., was obtained from the Wuhan Institute of Hydrobiology, Chinese Academy of Sciences, and cultured in BG-11 medium (Shen & Song 2007). Specific cultivation methods can be found in Supplementary Text S1.

The reagents employed in the experimentation were of analytical or superior purity, and the aqueous solutions used for solution preparation were constituted using Milli-Q ultrapure water. SYTOX green nucleic acid stain was purchased from Invitrogen, USA. Sodium hypochlorite (NaClO, Sigma-Aldrich) stock solution, Na₂SO₃ solution (0.5 g/L), solution (1 g/L), and ascorbic acid solution (4 g/L) were prepared just before the experiments. A stock solution of NaClO (calculated in Cl₂, 1 g/L) was stored at 4 °C in the absence of light, and its effective concentration was determined before each experiment using the N,N′-diethyl-p-phenylenediamine (DPD) colorimetric method. Residual chlorine concentration was quantified by subjecting the prepared DPD and buffer solution mixture to the colorimetric reaction, and the resulting OD₅₁₀ was employed for concentration determination. The residual chlorine concentration was subsequently calculated via a standard curve.

The algal cell stocks utilized in the experiment were derived from cultures that had been expanded for 10–14 days. Simulated water samples were prepared by diluting stabilized M. aeruginosa cultures from the stationary growth phase with original water to achieve a concentration of 1.0 × 10⁶ (cells/mL) (OD₆₈₀ ≈ 0.1). The source water was collected from the Jingmi Diversion Canal (N 40°31′16.53″,E 116°51′54.01″).

The experimental design consisted of four groups, all with a Cl₂ dosage of 1 mg/L. The concentrations were set at 0, 0.05, 0.25, and 0.8 mg/L, encompassing typical ranges of concentrations in surface waters. The corresponding mass ratios were 1:0, 1:0.05, 1:0.25, and 1:0.8, respectively. Among these, ratios of 1:0 and 1:0.05 represent simplified chlorination processes, with the exception that a smaller quantity of chlorine remains in the case of a chlorine-to-nitrogen ratio of 20:1. The ratio of 1:0.25 and 1:0.8 falls within the monochloramine portion of the breakpoint chlorination curve. Each experimental group included three parallel samples.

The experimental procedure is outlined as follows: 800 mL of the simulated water sample was taken in a beaker for each trial. Chlorine (1 mg/L) was added to the algae suspension with various concentrations of ammonia nitrogen. The mixture was rapidly stirred at 250 rpm for 5 min. After oxidation, the stirring speed was reduced to 40 rpm to simulate long-distance pipeline conveyance for 0–480 min. The whole experimental process was conducted in darkness. After different simulated transportation times (60, 120, 180, 240, 360, and 480 min), samples were taken separately and quenched for residual chlorine with sodium thiosulfate. A portion of the quenched sample was used for cell integrity analysis, and another portion of the sample was passed through a 0.45 μm filter membrane for AOM analysis.

The experiments assessing the formation of chlorinated DBPs during the chlorination of the dissolved phase in preoxidized algal suspensions were conducted by the methodology proposed by Qi et al. (2016a). The experiment involving the chlorinated DBPs from the dissolution of preoxidized algal suspensions was conducted in sealed brown glass bottles. The samples were kept in the dark at room temperature for 72 h following chlorine addition, and ascorbic acid was added for quenching. Samples were then collected for subsequent DBP analysis.

Chloroform (TCM), chloral hydrate (CH), dichloroacetone, trichloroacetone, dichloroacetonitrile (DCAN), trichloroacetonitrile (TCAN), dichloronitromethane, and trichloronitromethane (TCNM) were determined by an Agilent 7890B-5977B gas chromatogram and mass spectra (GC/MS) (Agilent, USA) according to the Ministry of Ecology and Environment, PRC method GB/T5750.10-2006. Monochloroacetic acid (MCAA), dichloroacetic acid (DCAA), and trichloroacetic acid (TCAA) were pretreated with an extraction/derivatization procedure using methyl tert-butyl ether and acidic methanol and analyzed by GC according to the Ministry of Ecology and Environment, PRC method GB/T5750.10-2006. Other conditions and the method detection limits can be found in Supplementary Text S2.

Algal cell quantification is achieved by integrating microscopic enumeration and UV–visible spectrophotometry (T6,20-1650-01-1195, Beijing Puxi, China), resulting in the derivation of optical density values at OD₆₈₀.

The method for assessing the damage to algal cells is referenced from the study by Daly et al. (2007). SYTOX Green nucleic acid stain was used to stain algal cells for the determination of cell integrity. Detailed measurement methods can be found in Supplementary Text S3.

The dissolved organic carbon (DOC) concentration in the dissolved phase of algal suspensions was measured with a Shimadzu TOC-V_CPH analyzer. A FlowAccess continuous flow analyzer (SAN⁺⁺, SKALAR, Netherlands) was used to determine the nitrite, nitrate, and concentrations. Dissolved organic nitrogen (DON) concentrations were determined by subtracting the measured nitrite, nitrate, and concentrations from total nitrogen. The total nitrogen was analyzed as nitrate after potassium persulfate oxidation (D'Elia et al. 1977).

Excitation-emission matrix (EEM) spectra were measured using a Cary Eclipse fluorescence spectrophotometer (Varian, Surrey, UK). The scanning wavelength range was set as Ex = 200–450 nm and Em = 250–550 nm, with bandwidths of Ex = 5 nm and Em = 5 nm. Prior to each sample measurement, the signal values of a blank sample with overnight tap water as the background were determined. Subsequently, the signal values obtained from the actual samples were corrected by subtracting the signal values of the blank sample. The regions for dividing the three-dimensional fluorescence spectra are detailed in Supplementary Text S4.

The impact of the ratio and transportation time on N-DBP concentrations was examined, as depicted in Figure 1(b). TCNM, TCAN, and DCAN were identified as the primary chlorinated N-DBPs among those formed. Simulated experiments indicate that the overall production of N-DBPs declines as concentration in the system increases. DCAN was identified as the predominant byproduct, with concentrations of 15.9, 13.3, 9.4, and 9.1 μg/L across four ratios after 480 min of water transport. Moreover, it is evident that the production of DBPs due to IOM released by algal cells during water transport is significantly more remarkable than the initial ammonia nitrogen in the system. Notably, the limitation of the TCAN formation is attributed to its susceptibility to alkaline-catalyzed hydrolysis at pH values above 5.5 (Croue & Reckhow 1989). Furthermore, TCAN formation demonstrated reduced sensitivity to levels due to its unique formation mechanisms involving nitrogen from NH₂Cl or nitrogenous organic compounds. Although EOM is a significant contributor to DBPs, the production of DBPs due to EOM should be the same for the same experimental group after various simulated transportation times. Therefore, this study mainly focuses on the influence of released IOM due to the spontaneous and continuous damage of algal cells during water transportation, which would result in the production of a large amount of DBPs and the ineffectiveness of coagulation.

Figure 1(c) illustrates the impact of simulated transportation time and the ratio on the formation of C-DBPs during the Cl₂ treatment in preoxidized algal suspensions. The results show that CH, TCM, TCAA, MCAA, and DCAA are the predominant chlorinated C-DBPs. For transportation times under 60 min, there is a correlation between the ratio and the quantity of C-DBPs produced; higher ratios lead to increased C-DBP concentrations. C-DBPs exhibited a cumulative increase with the extension of water transportation duration. During extended simulated transportation, from 360 to 480 min, the total C-DBPs showed no significant increase in the group with a ratio of 1:0. Similar to Figure 1(a), the concentration of C-DBPs in the group with a ratio of 1:0 was progressively lower than that in the group with a ratio of 1:0.05 when the water transportation time was 360 min.

In the presence of in water, Cl₂ shows a preference for reacting with (Wang et al. 2022b). During prechlorination disinfection, the presence of in water can mitigate damage to algae (Liu et al. 2020). Previous studies have confirmed that chloramine disinfection has a longer duration compared to chlorine disinfection (Lee & Westerhoff 2009). Interestingly, an increased ratio of damaged cells was observed with extended transportation time, even after the complete quenching of residual chlorine and chloramines. Similar effects have been reported in cases of KMnO₄ preoxidation and prechlorination followed by transportation (Qi et al. 2016a, b). This result can be explained by the programmed cell death (PCD) (Darehshouri et al. 2008; Bramucci & Case 2019), which can lead to the death of algae (Vavilala et al. 2015). PCD can be induced by factors such as oxidative stress, ultraviolet irradiation, nutrient deprivation, salt stress, and heat, ultimately triggering a genetically controlled cellular suicide (Kroemer et al. 1995; Elmore 2007; Wang et al. 2022a). Chemical oxidants can induce oxidative stress in algal cells, leading to their PCD (Ding et al. 2011). These findings reveal that the integrity of algal cells in algal-laden water with presence is influenced not just by the concentration but also by the distance between the water source and the water treatment plant.

As can be seen from Figures 1 and 2, the groups with ratios of 1:0 and 1:0.05 had almost the same rate of algal cell damage, but the two groups showed great differences in DBPs. C-DBPs of the group with a ratio of 1:0.05 were gradually higher than those of the group with 1:0 at water transportation times greater than 360 min. To further investigate the difference, we measured EEM and AOM to analyze the reason for this phenomenon.

As shown in Figure 3(b), the initial DOC value in the prepared algal suspension was determined to be 5.03 mg/L, which might be attributed to natural organic matter and the desorption of the loosely bound AOM on the cells (Xie et al. 2013). Corresponding to Figure 2, it was also found that the group with a ratio of 1:0.05 had higher DOC than the group with a ratio of 1:0 for both the 360 and 480 min water transportation times. These results indicate that the substances adsorbed after the death of the algal cells are dominated by DOC, which includes polysaccharides and large proteins (Aksu & Tezer 2005; Marungrueng & Pavasant 2006).

When the water transportation times are greater than 360 min, the concentration of C-DBPs in the group with a ratio of 1:0 is lower than that in the group with a ratio of 1:0.05. Upon the death of algal cells, the IOM released, including plant pigments and carotenoid-like substances, undergoes decomposition and is transformed into humic-like substances (Yang et al. 2011). Combined with the results of EEM, at the water transportation time is 120 min, the concentration in region V was significantly higher in the group with a ratio of 1:0 than 1:0.05, which indicated that the algal cells in the group with a ratio of 1:0 released more IOM. At 240 min, a significant decrease was observed in the group with a ratio of 1:0, which decreased further at 480 min and was lower than the group with a ratio of 1:0.05, suggesting that the group with a ratio of 1:0 showed a higher adsorption capacity of the damaged algal cells during the water transportation process.

Prior research has demonstrated that dead algal cells serve as effective biosorbents for the removal of organic matter and metal ions from aquatic environments (Areco et al. 2012; Tavana et al. 2020). Algae is one of the biosorbents with multilayer cell walls and surface functional groups, and dead algal cells have stronger adsorption properties than live algal cells (Cheng et al. 2017; Tavana et al. 2020). After the rupture of algal cells, the DOM released is dominated by large molecular proteins, as well as small molecular amino acids (Xie et al. 2013). The less molecular-weight amino acids are not easy to be adsorbed by slightly damaged cells because of the weak attractive force for physical adsorption (Periat et al. 2014). Thus, it can be confirmed that the adsorption of dead algal cells is dominated by large molecular proteins. Algal proteins are an important source of humus-like substances. This confirms that although the group with a ratio of 1:0 had the same rate of algal cell damage as the ratio of 1:0.05, the algal cells of the group with suffered lower oxidative damage and thus were less able to adsorb.

In this study, based on the ratio and transportation time after prechlorination, it was found that the concentration of C-DBPs in the group with a ratio of 1:0 was progressively lower than that in the group with a ratio of 1:0.05 as the duration of the water transportation increased. Further analyses of algal cell integrity and algal cell AOM release revealed two main factors that caused this difference. One was due to the protective effect of a small amount of on the algal cells, resulting in a lower release of AOM in the 1:0.05 group than in the 1:0 group. Secondly, the group with a ratio of 1:0 had a higher degree of damage to the algal cell structure than that with a ratio of 1:0.05 and thus had a higher adsorption effect. Prior to prechlorination of algae-laden water, chlorine dosage and delivery time should be considered in the context of the background concentration of in the water. Further verification is needed to generalize the findings to other algae species and actual water bodies utilizing the methods demonstrated in this study.

This work was supported by the National Natural Science Foundation of China (Grant No. 52170014, 52125003, 51808531).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.