Characterization of DOM and disinfection by-products precursors in biological activated carbon filter backwash water from drinking water treatments

Biological activated carbon (BAC) filtration is used as the main advanced drinking water treatment process in China. Nowadays, drinking water resources are deteriorating with contamination by anthropogenic chemicals. Dissolved organic matter (DOM) is ubiquitous in natural aquatic environments and engineered systems, which is a complex mixture of different compounds such as carbohydrates/polysaccharides, amino acids/peptides/proteins, lipids, humic substances, and anthropogenic organic pollutants (Shi et al. 2021; Zhang et al. 2022). BAC filtration can efficiently improve the water quality by removing DOM through adsorption and microorganism biodegradation (Ho et al. 2007; Thuptimdang et al. 2021). It is reported that the turbidity, COD_Mn, NH₃-N, UV₂₅₄, taste, and odor can be further decreased by the BAC filter (Xiang et al. 2013).

In our previous research, DOM is an important DBP precursor (Chu et al. 2011). Most DBPs have potential carcinogenic, teratogenic, and mutagenic toxicity, so it is important to study those (Richardson et al. 2007). BAC filtration also can remove the emerging contaminants, which are recalcitrant to the conventional water treatment, such as pharmaceutical and personal care products (PPCPs), endocrine-disrupting compounds (EDCs), pesticides, and herbicides (Benner et al. 2013; Chys et al. 2017; Fu et al. 2019). The disinfection by-products (DBPs) and their precursors can be controlled by BAC filtration and the performance of the O₃-BAC filter for DBP control was higher than that of the conventional process (Liu et al. 2017; Yu et al. 2021). BAC filtration is extensively applied to ensure drinking water safety.

The BAC filtration treatment requires backwashing at regular intervals to control the growth of the biofilm. A prolonged usage of BAC filter would reduce water quality due to the activated carbon saturation and the biomass overgrowth. Uncontrolled BAC treatment exposes downstream water consumers to the risk of pathogenic and/or infectious diseases associated with microbial breakthroughs into the distribution system (Simpson 2008; Shen et al. 2018). The regular backwash process ensures the normal operation of the BAC filter. As a result, large volumes of water are produced during the backwashing in drinking water treatment plants (DWTPs) (Loret et al. 2013). A common practice in DWTPs is recycling the BAC filter backwash water (BAC FBW) to the raw water with or without pretreatment (Tan et al. 2017). The process of backwashing removes considerable biodegradable organic matter and some decaying micro-organisms from the BAC filter. Most BAC filter backwash water contains abundant contaminants, including suspended solids, natural organic matter (NOM), bacteria, and inorganic metals (e.g., Fe, Mn, and Al) and the quality of backwash water will vary due to the differences in raw water quality and treatment train design (McCormick et al. 2010; Korotta-Gamage & Sathasivan 2017).

While the characteristics of NOM serving as precursors for trihalomethanes (THMs) have been well studied in raw water and sand filter backwash water (Walsh et al. 2008; McCormick et al. 2010; Lin et al. 2017), little is known about the unregulated DBPs in BAC FBW that are recycled to the head of DWTPs. Moreover, there are few studies on the influence of seasonal variation and backwash intervals on NOM in BAC FBW. Therefore, it is necessary to comprehensively monitor the change in backwash water quality of BAC filters and study the change of organic matter in different seasons and backwash intervals. The backwash interval of BAC filters in water treatment plants in Shanghai is 3–5 days in summer and 5–7 days in winter. There is no regulation for the backwash intervals in spring and autumn, so it is necessary to study the backwash water at different backwash intervals in spring or autumn. Finally, it can provide support for the long-term treatment and recycling of backwash water in DWTPs.

In this study, the BAC FBW in a DWTP was monitored over one year. Fluorescence spectroscopy was used to characterize and analyze the chemical composition of DOM in BAC FBW in different seasons and backwash intervals. The 17 DBPs formation potential (DBPs FP) including THMs, haloacetonitriles (HANs), haloacetamides (HAMs), haloacetaldehydes (HALs), and their cytotoxicity of BAC FBW were also investigated. The research offers insight into the DOM characteristics of the BAC filter backwash water in DWTPs.

The backwash parameters of the BAC filter are 4 min of air scouring followed by 8 min water rinse. The backwash water consumption of each BAC filter is about 200 m³. The water production capacity of the water plant is 60,000 m³. There are six BAC filters and the backwash water consumption accounts for 0.4%–0.9% of the water production. The backwash water is firstly discharged into the reuse pool and then recycled to the head of the DWTP.

The sampling control interval of backwash water of BAC filters is to take samples every 1 min during backwashing. The samples were collected in 1 L mild-cleaned plastic bags and transported to the laboratory freezer at 4 °C and processed within 24 h. The samples were filtered through a pre-washed 0.45 μm cellulose acetate membrane filter before analysis.

THMs, HANs, halonitromethanes (HNMs), and HALs mixture standards (Supelco 47904) were supplied from Sigma-Aldrich (St Louis, MO, USA). Methyl tert-butyl ether (MTBE) and anhydrous sodium sulfate were purchased from Aladdin Industrial Inc. (Shanghai, China). All other chemicals with analytical grade were obtained from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). Ultrapure water was produced using the Millipore Milli-Q gradient water purification system (Milli-Q, Merck Millipore, USA).

Samples filtered through 0.45 μm cellulose acetate membrane filter were stored at 4 °C until further analysis for dissolved organic carbon (DOC) and fluorescence excitation–emission matrix (EEM) spectra. The DOC concentration was measured with a TOC/TN analyzer (TOC-VCPH, Shimazu, Japan). Three independent replications were carried out and the average value was presented.

EEM spectra were measured using a fluorescence spectrophotometer (F-2710, Hitachi, Japan). The excitation wavelength (λ_ex) was varied from 220 to 450 nm, and the emission (λ_em) was varied from 220 to 550 nm, each in 5 nm increments. The scan speed was set to 12,000 nm/min.

Seventeen halogenated DBPs FP were analyzed, including four THMs: chloroform (TCM), bromodichloromethane (BDCM), dibromochloromethane (DBCM), and bromoform (TBM); five HANs: trichloroacetonitrile (TCAN), dichloroacetonitrile (DCAN), bromochloroacetonitrile (BCAN), dibromoacetonitrile (DBAN), and tribromoacetonitrile (TBAN); three HNMs: dichloronitromethane (DCNM), trichloronitromethane (TCNM), and tribromonitromethane (TBNM), and five HALs: dichloroacetaldehyde (DCAL), trichloroacetaldehyde (TCAL), bromodichloroacetaldehyde (BDCAL), dibromochloroacetaldehyde (DBCAL), and tribromoacetaldehyde (TBAL).

The DBP FP test method and the dosage of chlorine addition were performed according to the methods detailed in Krasner et al. (2006). Fill a 250 mL triangular flask with a water sample, keep the pH at 7.0 ± 0.2 and add 750 μL sodium hypochlorite solution with 5.2% effective chlorine. Make the content of available chlorine 20 mg/L, place it in a 25 °C water bath to react for three days in a dark place, and control the residual chlorine at 3–5 mg/L at the end of the reaction. Residual chlorine was quenched using excess sodium thiosulfate. Duplicate samples, along with laboratory and field blanks, were extracted for DBP analysis. After that, the water sample was extracted by liquid–liquid extraction and the formation potential of DBP was determined by gas chromatography with an electron capture detector (GC-ECD). DBP was separated via RTX-5MS column (30 m × 0.25 mm × 0.25 μm). The specific analysis conditions of DBPs are shown in Table 2.

The cytotoxicity of DBP was characterized by the ratio of the formation potential of DBP to the LC₅₀ value of corresponding Chinese hamster ovary (CHO) cells (Zeng et al. 2016). After summing, the cumulative cytotoxicity of various DBPs was obtained to explore the effects of different water samples on cytotoxicity. The LC₅₀ value of CHO cells used for calculation was obtained from previous studies (Yang et al. 2013; Wagner & Plewa 2017).

The results showed that the relative abundances of fluorescence components had a direct relationship with seasonal variations. EEM spectra exhibited high intensities in regions I and II, with corresponding area volumes of 20%–30% and 38%–48% of the sum of all the regional volumes, respectively. These two regions represented aromatic protein-like compounds, which was possibly due to the stimulated microbial activity in the BAC filter (Ly et al. 2017). With the change of seasons, the abundance of each region changed significantly. The proportion of regions I and II was the highest in summer, up to 75%, and the lowest was only 52% in winter. Region IV represented the soluble microbial products, including tryptophan-like and biologically related tyrosine-like compounds, which were the lowest in summer and the highest in winter. The dynamic change of fluorescence area was mainly due to the different microbial activities in the BAC filter under the change of temperature consistent with the DOC analysis.

The DOC concentration and DOM composition in the BAC filter backwash water changed with the backwash interval. The backwash interval impacted the pollutant accumulation in the BAC filter bed and the growth of biomass as well. The study on the concentration of DOC in backwash water in different seasons shows that the concentration of DOC in spring and autumn is higher, so the effect of the backwash interval and backwash process in spring was also investigated to analyze the variation of BAC FBW.

TCM and BDCM were presented at the highest concentrations, at 65.3 and 29.2 μg/L, respectively, which was consistent with the research results (Hu et al. 2021). There were bromines in the backwash water, which was the reason to generate BDCM in the backwash water. Among the five HANs, TBAN and DCAN were the most prevalent, at 27.2 and 8.74 μg/L, respectively. DCNM formed to the greatest extent among the three HNMs, at 5.87 μg/L in chlorinated BAC filter backwash water. Among the five HALs, TCAL was present at the highest concentration at 28.03 μg/L in chlorinated BAC FBW. The backwash interval impacted more greatly on the formation potential of THMs, with an increase from 95.1 to 126 μg/L when the backwash interval was prolonged to five days. The variation was due to the enhancement of NOM with the prolonged backwash interval. The impact of backwash interval variation on the formation potential of the other DBPs was not obvious. Some results confirmed the presence of bromide in the raw water of the Jinze reservoir from Tai Lake (Hong et al. 2015; Li et al. 2018). Moreover, brominated DBPs are produced in BAC FBW after chlorine disinfection, so the presence of bromide in BAC FBW should not be underestimated. Therefore, more attention should be paid to the safety risks of brominated DBPs after backwash water reuse.

The fluorescence spectroscopy, DBP formation potential, and cytotoxicity of BAC FBW were investigated during different seasons and backwash intervals. The main conclusions of this study are as follows:

• (1)

  The BAC filter backwash water in a DWTP was monitored over one year. The DOC concentration reached its lowest value of 1.95 mg/L in the summer. Aromatic protein-like compounds were found to be the main component of DOM by FRI analysis in all seasons. Seasonal variations in temperature impacted the composition of DOM due to the microbial activity in the BAC filter.

• (2)

  BAC filter backwash intervals changed the DOM quantity in the backwash water. DOC concentration increased with the prolongation of backwash intervals, but decreased over five days. The DBP formation potentials also increased with the prolonged backwash interval. On the premise that the BAC FBW is reused, it is not advisable to use a long backwash interval, so as to ensure the quality of the backwash water of the BAC filter and not cause deterioration of the effluent of the water plant.

• (3)

  THMs were the main DBP in chlorinated BAC FBW. However, the order of contribution to cytotoxicity was HANs > HALS > THMs > HNMs, respectively. In addition to the regulated THMs, the water treatment plants should also pay attention to HANs when considering the reuse of BAC FBW to avoid the biological risk of effluent quality.

This research was funded by the Natural Science Foundation of Shanghai (21ZR1467300), Science and Technology Innovation Action Plan technical standard project (22DZ2200300), and the National Key Research and Development Program (2021YFC3201303).

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

The authors declare there is no conflict.