Listen

The presence of toxic algae, such as Microcystis aeruginosa (MA), in drinking water treatment plants (DWTPs) would contribute to algal organic matter (AOM) as precursors toward disinfection by-products (DBPs). Electrocoagulation–flotation (ECF) has shown promising performance in minimizing algal cells from water and dissolved AOM. This study aimed to investigate the effect of current density (CD) and pH on alumina (Al)-based ECF for removing MA cell and DBPs precursors from cyanobacteria-laden water. The performance of Al-based ECF was evaluated at various CD and pH conditions within 20 min. In addition, the total halogenated DBPs formation of the treated suspension after ECF was quantified. At pH 8, the ECF process with 5 mA/cm2 exhibits the most significant reductions in MA cell and soluble AOM, accounting for 97 and 56%, respectively. Additionally, the precursors of trihalomethanes (THMs) and haloketones (HKs) can be effectively removed with flotation despite their significant release at EC. The tremendous reduction of humic acid-like (HAL) substances in extracellular organic matter (EOM) fraction by ECF leads to the minimized THMs formation potential. In summary, Al-based ECF at pH 8 is effective to remove cyanobacteria and minimize the precursors of regulated THMs along with an insignificant reduction in regulated haloacetic acids (HAAs) precursors.

Listen

  • Microcystis aeruginosa is most effectively removed by Al-based ECF at pH 8 for 5 mA/cm2.

  • Bound EOM excretes significantly by EC but markedly reduces along with algal cells after ECF.

  • Reducing HAL substances in EOM leads to the minimized formation of halogenated DBPs.

  • THMs precursors reduce mostly with an implicit decrease in HAAs precursors in the ECF process.

Graphical Abstract

Graphical Abstract
Graphical Abstract
View largeDownload slide
Graphical Abstract
Graphical Abstract
View largeDownload slide
algal organic matter, cyanobacteria, disinfection by-products, electrocoagulation
Listen

Numerous sources of drinking water have been proven to experience rapid algal blooming during specific seasons (Putri et al. 2020), especially for eutrophicated reservoirs or lakes. Pollution of nutrients, such as phosphate and nitrogen, from agricultural activity, domestic sewage, or industrial wastewater can significantly accelerate algae proliferation. In that case, the presence of toxic cyanobacteria deteriorates the coagulation–sedimentation performance of drinking water treatment plants (DWTPs), leading to clogged filters, bad odors, and poor quality water (Wu et al. 2021), then ultimately harming human health because it can cause cancer risk, birth defects, and antibiotic resistance (Richardson & Ternes 2018). In addition, cyanobacteria-laden water contains varied algal organic matter (AOM), including extracellular organic matter (EOM) and intracellular organic matter (IOM). EOM consists of bound extracellular organic matter (bEOM) that adheres to the surface of algal cells as well as dissolved extracellular organic matter (dEOM) secreted from cells to algal suspension (Tang et al. 2017). Either IOM or EOM is well recognized to be the precursor of disinfection byproducts (DBPs), such as trihalomethanes (THMs), haloacetic acids (HAAs), haloacetonitriles (HANs), and haloketones (HKs), in a treatment train with post-chlorination followed by coagulation–sedimentation and filtration (Lin et al. 2022), especially at high chlorine dosage where it produces much higher DBPs (Huang et al. 2021). These challenges can heighten the burden of operational units in drinking water treatment. Thus, the minimization of algal cells with proper pre-treatment approaches in the safe drinking water supply is warranted for cyanobacteria-impact DWTPs.

Chemical coagulation has been widely used to improve the separation of algal cells in raw waters for DWTPs with a high removal efficiency of up to 93% (Ma et al. 2022). Despite the excellent efficiency in algal removal by coagulation has been proven, it highly relies on the characteristics of algal species (Zhao et al. 2019). For instance, as extremely high algal density occurs in raw water (e.g., >106 cells/mL), more chemical dosage is required to destabilize cells and form big flocs during coagulation. Moreover, the coagulation is subjected to a long mixing process (e.g., flocculation) and excessive coagulant dosage, which would lead to significant sludge generation (Sillanpää et al. 2018). Meanwhile, the alumina (Al) hydrates would dominantly influence the floc formation and algal cell removal by the coagulation–sedimentation process (Lin et al. 2021), and thus the removal of algal cells from water depends on the formation of dense flocs due to the low density of algae.

Al-based electrocoagulation–flotation (ECF) is well known to be a promising algal cell separation process because their light cells can be effectively floated in ECF. It requires a small footprint, fast reaction and concentrated sludge. Several studies have exhibited more than 95% algal cell removal with ECF toward cyanobacteria-laden water (An et al. 2019). It also has been proven that natural organic matter (NOM) or AOM can be removed by Al-based ECF (Chellam & Sari 2016; Meetiyagoda & Fujino 2020), accompanying a pronounced reduction in dissolved organic carbon (DOC) and specific UV absorbance (UV254) in comparison with that by chemical coagulation (Lin et al. 2022). Basically, the performance of the ECF process is highly dependent on pH and current densities (CDs) (Bagastyo et al. 2022). The pH level of Al-based ECF predominately influences the formation of released Al hydrates that rule coagulation behaviors. At acidic pH, charge neutralization would predominately happen due to the occurrence of abundant monomeric Al, while insoluble Al hydroxides would form tremendously at neutral or weak alkaline pH conditions in response to sweep flocculation (Wang et al. 2020). On the other hand, the CD applied for ECF is theoretically proportional to the released amounts of Al hydrates. However, it has been demonstrated that the increase in CD would cause slight damage to MA cell structure and simultaneously result in the release of IOM or EOM (Wang et al. 2018). Then, the release of EOM would impair electrocoagulation–flocculation performance for algal cell removal (Fuente et al. 2019). As a result, the removal of EOM as well as the minimization of DBPs precursors are well correlated during the ECF process. Up to date, limited studies have investigated the operating factors influencing the relationship between EOM removal along with MA cells by Al-based ECF and the corresponding DBPs formation potential (DBPFP) for drinking water treatment. The investigations into favorable mechanisms toward the separation of microcells and EOM reduction for DBPs precursors minimization during ECF toward cyanobacteria-laden water are warranted.

This study aimed to investigate the performance of the Al-based ECF process at various pH and CD conditions for the minimization of DBPs precursors in cyanobacteria-laden water. The amounts of released Al were determined in theoretical and studied conditions during EC in the absence of algal samples. The algal samples were prepared with tailor-cultured Microcystis aeruginosa (MA) suspensions at bench scale, and the MA cells were counted and characterized by a flow cytometer (FCM). The variations in MA-derived AOM during ECF were evaluated in terms of dEOM and bEOM by fluorescent excitation–emission matrix (F-EEM) spectra measurement.

Listen

Algae sample cultivation

Listen

The MA suspensions were cultured in 500 mL of Erlenmeyer glass with BG-11 as a medium following a previous study (Lin et al. 2022). The automated flask shaker with illuminance under the ratio of 8:16 (dark/light cycle) was utilized at room temperature (25 °C). The MA suspensions were taken for a test while the desired condition was obtained. The diameter of MA algae was measured by a digital microscope (Digital microscope-CX31, Olympus, Japan) with ranges from 5 to 6 μm, as seen in Supplementary Material, Figure S1. For each test, the MA suspensions were prepared with a DOC concentration of 3 mg/L and a total algal density of 1 × 106 cells/mL.

Protocol of Al-based ECF test

Listen
The Al-based ECF was carried out in a batch trial using a cubical plexiglass reactor with a working volume of 2.4 L, as illustrated in Figure 1. The ECF reactor was equipped with standard jar test agitation and FlocCAM (Jar FlocCAM, Durasens, USA). An alumina plate was employed as both anode and cathode with 50 cm2 of an active area. The inter-gap electrode was 2 cm connected to the DC supply. In this study, the background electrolyte concentration of ECF was prepared with NaClO4 of 0.01 M as well as 0.1 N NaOH and 0.1 N HCl were prepared as working solutions for pH adjustment. The ECF process comprises electrocoagulation and flotation phases, wherein 10 min electrocoagulation was performed by mixing at 100 rpm, followed by 10 min flotation. The electricity and agitator were switched off after electrocoagulation to allow the remaining bubbles to bring residuals to the top layer of the reactor during flotation.
Figure 1
A schematic illustration of ECF reactor.
View largeDownload slide

A schematic illustration of ECF reactor.

Figure 1
A schematic illustration of ECF reactor.
View largeDownload slide

A schematic illustration of ECF reactor.

Close modal

Analytical methods

Listen
For the performance evaluation on ECF, the samples for each test were withdrawn at 10 cm below the water surface before and after EC, as well as flotation to be filtered through 0.45 μm of mixed cellulose ester membrane (Thermo Scientific, USA). Al concentration of each sample during 10 min EC reaction process was determined by inductively coupled plasma-optical emission spectroscopy (ICP-OES) (ICPE 9820, Shimadzu, Japan). The zeta potential of MA suspensions before and after EC treatment was measured by a zeta analyzer (Zetasizer Nano ZS90, Malvern, UK). The size and fractal dimension of MA suspensions after EC treatment was determined by FlocCAM observation following the previous study (Lin et al. 2021). The turbidity was measured by a turbid meter (2100P, Hach, USA). In addition, the algal cell density before and after each test was measured by an FCM (Cyto Flex, Beckman Coulter, USA). The FCM was carried out with red fluorescence intensity as PC 5.5-A at ≥ 670 nm, and green fluorescence intensity as FITC-A was collected at 533 nm. The DOC was quantified using a total organic carbon analyzer (TOC-L, Shimadzu, Japan). Fluorescence spectroscopy (RF-6000, Shimadzu, Japan) was used to measure the fluorophore of DOC samples through F-EEM spectra analysis. The four fluorophore regions, soluble microbial product-like (SMPL), aromatic protein-like (APL), humic acid-like (HAL), and fulvic acid-like (FAL) substances, were categorized and their average fluorescent intensity (AFI) was determined (Lin & Ika 2022). The bEOM and dEOM extraction methods were modified by a previous study (Tang et al. 2017), with 15 min centrifugation using a high-speed centrifuge (DM0636, DLAB, USA) at 4,000 rpm for each sample. After that, the supernatant was filtered and analyzed as dEOM. Furthermore, the pellet was then resuspended by adding an NaCl solution of 0.6%. Finally, the samples were centrifuged for 30 min at 6,000 rpm, and the filtered supernatant was used as bEOM. All the experimental data of DOC, EEM, and DBPFPs were calculated in reduction rate ratio by the following equation:
(1)
where N0 represents the original value before EC, and Nt represents the value obtained after EC or after floatation, respectively.
During the EC process, the Al is released following Faraday's law. Thus, the theoretical amounts of Al released could be counted following Equation (2). Furthermore, several oxidation–reduction mechanisms could occur during ECF at anode, cathode, and solution, as shown in Table 1.
(2)
where I refers to current (A), M is alumina atomic mass (26.98 g/mol), t represents operational time (s), z is the number of electrons involved in the reactions (z = 3), F is Faraday's constant (96,486 C/mol), and V is the volume of the treated sample (L).
Table 1

Theoretical reaction pathways in the ECF process

Reaction siteReactionsEquation
Anode Metal dissolution
 		(3) 
Passivation electrode
 		(4) 
Cathode Acidic condition
 		(5) 
Neutral and alkaline condition
 		(6) 
Solution  (7) 
 (8) 
 (9) 
Reaction siteReactionsEquation
Anode Metal dissolution
 		(3) 
Passivation electrode
 		(4) 
Cathode Acidic condition
 		(5) 
Neutral and alkaline condition
 		(6) 
Solution  (7) 
 (8) 
 (9) 
View Large

For the analysis of DBPFP, four classes in DBP comprise five of THMs substances (trihalomethane [TCM], bromodichloromethane [BDCM], bromoform [TBM], trichloronitromethane [TCNM], and dibromochloromethane [DBCM]); four of HANs substances (dichloroacetonitrile [DCAN], trichloroacetonitrile [TCAN], bromochloroacetonitrile [BCAN], and dibromoacetonitrile [DBAN]); nine of HAAs substances (monochloroacetic acid [MCAA], monobromoacetic acid [MBAA], dichloroacetic acid [DCAA], trichloroacetic acid [TCAA], bromochloroacetic acid [BCAA], bromodichloroacetic acid [BDCAA], dibromochloroacetic acid [DBAA], chlorodibromoacetic acid [CDBAA], and tribromoacetic acid [TBAA]), and two of HK substances (dicloroketone – DCK and trichloroethane – TCK) were analyzed after 7 days of chlorination with NaOCl at room temperature (25 ± 1 °C) in a darkroom. In the chlorination process, the concentration ratio between Cl2 and DOC is fixed at 5. THMs, HANs, and HKs were quantified after the liquid/liquid extraction method following the U.S. EPA 551.1 (U.S. EPA 1995). A 30 mL chlorinated sample was transferred to 40 mL transparent glass vials containing 10 g of sodium sulfate. Thereafter, DBPs were extracted with 3 mL of methyl test-butyl ether (MTBE) before analysis. HAAs were analyzed by liquid/liquid extraction following the modified U.S. EPA method 552.3 (U.S. EPA 2003). Before extraction, the sample was firstly acidified with 1.5 mL of concentrated H2SO4 to adjust the sample pH to <0.5. A 30 mL chlorinated sample was transferred to 40 mL transparent glass vials containing 13.5 g of sodium sulfate. After that, DBPs were extracted by 3 mL of MTBE. The MTBE layer (upper layer) was transferred to 10 mL vials containing 2 mL acid–methanol mixture (10% H2SO4 in methanol) and placed in a heating chamber at 55 °C for 2 h. After cooling, 5 mL of Na2SO4 was added, and the bottom layer was taken and added with 2 mL of NaHCO3. The vials were shaken for 2 min, and then the MTBE layer (upper layer) was takeout and transferred into 1.5 mL vials. All the extracted DBPs were measured with a gas chromatography–electron capture detector (GC–ECD) (Nexis GC-2030, Shimadzu, Japan). This instrument was equipped with a 30 m × 0.25 mm × 0.25 μm column (DB-1701, Agilent, USA). The method detection limit (MDL) for each DBPs compound was described in our previous study (Lin & Ika 2022), which ranges from 0.01 to 0.23 μg/L.

Listen

Effect of CD on MA cell separation

Listen
The effect of CD on MA cell separation during ECF was first investigated under pH 7 to confirm the optimal CD. As seen in Figure 2(a), the cell density of MA decreases markedly during ECF where the ratio of MA cell separation increases insignificantly up to 23% after 10 min EC at 5 mA/cm2 due to incomplete cell destabilization. Nevertheless, the ratio of cell separation significantly increase after flotation in the ECF process, especially at 5 mA/cm2 where the cell separation ratio reaches to a maximum of around 97%. It is followed by 3 and 10 mA/cm2 accounting for 75.3 and 88.5%, respectively. The difference in the cell separation ratio can be explained by the released Al concentration. The higher CD is applied, the higher amounts of Al release, as seen in Figure 2(b). This leads to the marked difference in zeta potential variations with reaction time, as illustrated in Figure 2(c). For ECF with 3 mA/cm2, the lowest amounts of Al released contribute to the lowest cell separation efficiency, even though it occurs at charge neutralization where zeta potential is very close to zero. In contrast, the EC favors sweep flocculation predominately induced by Al(OH)3 precipitates at 5 mA/cm2 where the zeta potential ranges from 2 to 5 mV, causing a significant improvement in cell separation. However, the excessive amounts of Al released at 10 mA/cm2 lower the cell separation efficiency due to pronounced charge reversal (Figure 2(c)). The results of effective cell removal are similar to previous studies (An et al. 2019; Yamaguchi et al. 2019). It is worth noting that overhighly applied CD for this study could materialize a thin oxide layer known as Al2O3 (Hashim et al. 2019), as indicated by Equation (4) in Table 1. In such a condition, it would inhibit the Al release from the surface of the electrode plate. Thus, the lower amounts of released Al happens compared to theoretical calculation, as seen in Figure 2(b), which would worsen the performance of ECF for cell removal. Based on the findings on the effect of CD on MA cell separation, the following ECF tests were conducted at 5 mA/cm2 to perform effectively.
Figure 2
Variations in (a) cell density and cell separation ratio; (b) released Al (dashed line represents theoretical Al concentration); and (c) zeta potential with CD during ECF at neutral pH.
View largeDownload slide

Variations in (a) cell density and cell separation ratio; (b) released Al (dashed line represents theoretical Al concentration); and (c) zeta potential with CD during ECF at neutral pH.

Figure 2
Variations in (a) cell density and cell separation ratio; (b) released Al (dashed line represents theoretical Al concentration); and (c) zeta potential with CD during ECF at neutral pH.
View largeDownload slide

Variations in (a) cell density and cell separation ratio; (b) released Al (dashed line represents theoretical Al concentration); and (c) zeta potential with CD during ECF at neutral pH.

Close modal

Effect of pH on MA cell separation

Listen
Figure 3 shows the variations in MA cell separation ratio in the ECF process at various pH conditions. The pH adjustment shows a profound effect on cell separation at 5 mA/cm2, as shown in Figure 3(a). The cell separation ratio only increases a little after 10 min EC as the pH rises from pH 5 to 8. The cell separation ratio reaches around 30% at pH 8, while a lower cell separation efficiency is found at pH 5 and 7, accounting for 20 and 18%, respectively. However, the cell separation ratio increases substantially after flotation in the ECF process at various pH conditions. The cell separation ratio can be improved to 97.5% at pH 8, followed by 89.3% at pH 7. Unlike the pronounced improvements in cell separation at higher pH, the lowest cell separation ratio occurred at pH 5 is only around 61%. In theory, pH plays an important role in Al species during ECF, as indicated in Equations (8)–(10) in Table 1. At acidic conditions (pH < 6), the majority of soluble monomeric alumina, such as and ⁠, would form. While increasing pH to neutral or weak alkaline conditions, Al ions would transform into several kinds of hydrates such as polymeric alumina (e.g., ⁠) and insoluble products such as Al(OH)3 (Lin et al. 2021). Therefore, Al(OH)3 could form tremendously at pH 7 and 8, even though the released Al is less than that at pH 5 (Figure 3(b)). However, the largest floc size with a smaller fractal dimension is observed after EC at pH 8, as shown in Figure 3(c). These results indicate the pronounced cell removal by EC at pH 8 potentially due to sweep flocculation with lower zeta potentials (+5.3 mV in maximum) after 10 min EC, as shown in Figure 3(d), which corresponds to the variations in turbidity (Supplementary Material, Figure S2). Sweep flocculation dominates MA cell destabilization once large amounts of Al(OH)3 precipitate at alkaline pH, which induces effective cell removal (Lin et al. 2022). However, the extremely high positive zeta potential (+21 mV) occurs at pH 5 after EC insignificantly destabilize MA cells due to strong repulsion forces between cells, resulting in ineffective cell separation. A previous study has confirmed that the coagulation mechanism in the ECF process is dependent on pH variations (Bagastyo et al. 2022). At acidic pH, strong charge neutralization occurs in the state of dominant monomeric Al products, leading to a charge reversal effect on coagulation. Thus, the least MA cell removal after EC at pH 5 is found in this study.
Figure 3
Variations in (a) cell density and cell separation ratio during ECF; (b) released Al; (c) floc size and fractal dimension; and (d) zeta potential with various pH during EC at 5 mA/cm2.
View largeDownload slide

Variations in (a) cell density and cell separation ratio during ECF; (b) released Al; (c) floc size and fractal dimension; and (d) zeta potential with various pH during EC at 5 mA/cm2.

Figure 3
Variations in (a) cell density and cell separation ratio during ECF; (b) released Al; (c) floc size and fractal dimension; and (d) zeta potential with various pH during EC at 5 mA/cm2.
View largeDownload slide

Variations in (a) cell density and cell separation ratio during ECF; (b) released Al; (c) floc size and fractal dimension; and (d) zeta potential with various pH during EC at 5 mA/cm2.

Close modal
To further verify the cell separation behaviors in the ECF process at various pH conditions, the fluorescent contours of FCM were studied. The FCM analysis was conducted with FITC-A and PC 5.5-A channel light-scattering survey for this study. Figure 4 illustrates the scattering dots of FCM toward the integrity of MA cells during ECF in the pH range from 5 to 8. It shows that the MA cells are perfectly united at various pH conditions before ECF treatment. After EC, large amounts of scattering dots for MA cells are significantly displaced. The MA cells are likely destroyed as they undergo the stress during EC, resulting in a fragmented scattering of dots area and wide intensity (Menezes et al. 2021). In that case, the fragmented scattering dots area at pH 5 presents the highest intensity. It reveals that the MA cells have been significantly ruptured, thus lowering the intensity of FITC-A. A previous study has reported that the intensity at FITC-A is highly correlated with lipids within the MA cells. Accordingly, the ruptured cells make lipids disperse, resulting in reduced intensity (Roostaei et al. 2018). Nonetheless, the fluorescent intensities of cells in various pH conditions drop mostly after flotation in the ECF process owing to effective cell separation. For instance, the density of scattering dots toward MA cells at pH 8 is approximately imperceptible, which echoes the most pronounced cell separation, as shown in Figure 3(a). In contrast, it exhibits a high and wide scattering dots area at pH 5 due to the majority of remaining MA cells. In summary, increasing CD and pH is beneficial for MA cell reduction, but pH has a more pronounced effect than CD in terms of MA cell separation, especially at alkaline pH for this study.
Figure 4
Flow cytometer analysis on MA cells at various pH values in a favorable CD (5 mA/cm2) during the ECF process. The scattering dots were detected in red fluorescence (PC 5.5-A-670 nm; y-axis) and green fluorescence (FITC-A-533 nm; x-axis) channels. Please refer to the online version of this paper to see this figure in colour: http://dx.doi.org/10.2166/aqua.2022.065.
View largeDownload slide

Flow cytometer analysis on MA cells at various pH values in a favorable CD (5 mA/cm2) during the ECF process. The scattering dots were detected in red fluorescence (PC 5.5-A-670 nm; y-axis) and green fluorescence (FITC-A-533 nm; x-axis) channels. Please refer to the online version of this paper to see this figure in colour: http://dx.doi.org/10.2166/aqua.2022.065.

Figure 4
Flow cytometer analysis on MA cells at various pH values in a favorable CD (5 mA/cm2) during the ECF process. The scattering dots were detected in red fluorescence (PC 5.5-A-670 nm; y-axis) and green fluorescence (FITC-A-533 nm; x-axis) channels. Please refer to the online version of this paper to see this figure in colour: http://dx.doi.org/10.2166/aqua.2022.065.
View largeDownload slide

Flow cytometer analysis on MA cells at various pH values in a favorable CD (5 mA/cm2) during the ECF process. The scattering dots were detected in red fluorescence (PC 5.5-A-670 nm; y-axis) and green fluorescence (FITC-A-533 nm; x-axis) channels. Please refer to the online version of this paper to see this figure in colour: http://dx.doi.org/10.2166/aqua.2022.065.

Close modal

Changes in fluorescent AOM characteristics

Listen
The variations of fluorescent AOM characteristics were further investigated in the ECF process at the optimum condition (i.e., pH 8 and 5 mA/cm2) to understand the mechanisms of EC toward the AOM removal. Figure 5 shows the variations in DOC and AFI reduction ratios during ECF. The ratio of DOC reduction reaches −21.9% after EC, and the subsequent flotation process improves the DOC reduction up to 56.14%, as shown in Figure 5(a). After EC, several dissolved organics are released into the suspensions due to electric current passing through the algal cells, inducing ions inside the cells to move randomly (Hoseinzadeh et al. 2020). The lower DOC removal is caused due to the fact that some of the AOM in MA suspensions are more hydrophilic (Cao et al. 2020).
Figure 5
Variations of (a) DOC and (b) fluorophore spectra before and after the ECF process (pH: 8; CD: 5 mA/cm2).
View largeDownload slide

Variations of (a) DOC and (b) fluorophore spectra before and after the ECF process (pH: 8; CD: 5 mA/cm2).

Figure 5
Variations of (a) DOC and (b) fluorophore spectra before and after the ECF process (pH: 8; CD: 5 mA/cm2).
View largeDownload slide

Variations of (a) DOC and (b) fluorophore spectra before and after the ECF process (pH: 8; CD: 5 mA/cm2).

Close modal

Four regions of fluorophore spectra were investigated to confirm their contributions in AOM reduction after 10 min EC, as depicted in Figure 5(b) and Supplementary Material, Figure S3. It shows that the fraction of AOM mostly originated from SMPL substances, which have a two-fold higher intensity than others (i.e., HAL, FAL, and APL substances). As a result, the most pronounced release of SMPL substances appears, accounting for −17.7% reduction. While the negative reduction ratio of HAL and FAL substances is only about half compared to that of SMPL and APL substances. This indicates that lower amounts of AOM release from HAL and FAL fractions. However, the AFI of all AOM reduces subsequently after flotation. The most effective reduction in AFI occurs for SMPL substances, accounting for around 60%, followed by APL, FAL, and HAL substances. It has been reported that SMPL substances are composited of abundant polysaccharides (Xie et al. 2019), and Al coagulation is favorable for the removal of polysaccharides at around pH 8 (Naceradska et al. 2019). Thus, the most pronounced reduction in SMPL substances appears for Al-based ECF at pH 8.

On the other hand, it is important to understand that EOM fractions change in terms of dEOM (i.e., dissolved organics) and bEOM (i.e., organics bonds to MA cells) during ECF at pH 8 using a CD of 5 mA/cm2. The DOC variations in terms of dEOM and bEOM were investigated, as shown in Figure 6. It shows that the DOC reduction ratio of both dEOM and bEOM increases up to −50 and −70% after EC ahead of flotation, respectively. In other words, the quantity of bEOM increases much more than that of dEOM. High DOC release from MA cells could be associated with the stirring and electrifying effect on the increased bEOM and dEOM substances during EC (Wang et al. 2018). Regardless of the DOC release, bEOM is much easier to be removed than dEOM after 10 min flotation in the ECF process. The mechanisms of bEOM and dEOM removal could be further explained by investigating a variety of AFI on each fluorescent AOM, as shown in Supplementary Material, Figure S4. It exhibits markedly different AFI changes in four fluorescent AOMs for both dEOM and bEOM in the ECF process. Before ECF, there are no differences in AFI of all four AOMs between bEOM and dEOM, as illustrated in Figure 7. After EC, there is a negative reduction in AFI of bEOM, particularly in SMPL substances with almost 100% negative reduction. In other words, the amounts of bEOM significantly increase in terms of SMPL fraction after EC. This phenomenon is caused potentially due to the occurrence of high stress on MA cells during the EC process. Thus, the MA cells could further secrete bEOM and the amounts of SMPL substances increase twice after EC, as shown in Figure 7(a). Nonetheless, the decrease in AFI of bEOM fractions is substantially improved after subsequent flotation, especially for HAL and FAL substances, accounting for around 70% AFI reduction compared to that before EC. In the ECF process, enormous gases would generate firstly and then increase the collision frequency between the released Al and MA cells or AOM to form flocs by sweep flocculation. At such a condition, the remaining bubbles in subsequent flotation are effective to float flocs with the adsorbed EOM, especially for bEOM separation. A previous study has elaborated that bEOM substances attached to MA cells could be removed simultaneously together with MA cell separation (Tang et al. 2017). Therefore, the fluorescent bEOM fractions are effectively reduced along with MA cell removal by ECF. Unlike bEOM, the negative reduction in dEOM after EC is less than 30% only in terms of SMPL and APL substances, and most fluorescent AOM fractions are ineffectively removed by subsequent flotation except for the fractional HAL substances. For this study, the dEOM substances with high hydrophilicity are negatively charged in MA suspensions (Cao et al. 2020). Thus, it is difficult to destabilize the dEOM by EC, which subsequently leads to ineffective dEOM removal by flotation. The changes in AFI of AOM fractions have suggested that ECF is much more effective in removing bEOM over dEOM, and the SMPL fractions are mostly removed, even though the bEOM concentration raises much more than dEOM at EC.
Figure 6
The DOC reduction ratio of dEOM and bEOM during ECF toward MA suspensions (pH: 8; CD: 5 mA/cm2).
View largeDownload slide

The DOC reduction ratio of dEOM and bEOM during ECF toward MA suspensions (pH: 8; CD: 5 mA/cm2).

Figure 6
The DOC reduction ratio of dEOM and bEOM during ECF toward MA suspensions (pH: 8; CD: 5 mA/cm2).
View largeDownload slide

The DOC reduction ratio of dEOM and bEOM during ECF toward MA suspensions (pH: 8; CD: 5 mA/cm2).

Close modal
Figure 7
The fluorescent intensity and reduction ratio of dEOM and bEOM in terms of (a) SMPL; (b) APL; (c) HAL; and (d) FAL substances during ECF toward MA suspensions at optimum conditions (pH: 8; CD: 5 mA/cm2).
View largeDownload slide

The fluorescent intensity and reduction ratio of dEOM and bEOM in terms of (a) SMPL; (b) APL; (c) HAL; and (d) FAL substances during ECF toward MA suspensions at optimum conditions (pH: 8; CD: 5 mA/cm2).

Figure 7
The fluorescent intensity and reduction ratio of dEOM and bEOM in terms of (a) SMPL; (b) APL; (c) HAL; and (d) FAL substances during ECF toward MA suspensions at optimum conditions (pH: 8; CD: 5 mA/cm2).
View largeDownload slide

The fluorescent intensity and reduction ratio of dEOM and bEOM in terms of (a) SMPL; (b) APL; (c) HAL; and (d) FAL substances during ECF toward MA suspensions at optimum conditions (pH: 8; CD: 5 mA/cm2).

Close modal

Alleviation of halogenated DBPs formation

Listen
The alleviation of halogenated DBPs precursors can be significantly affected by coagulation with Al hydrates (Lin & Ika 2020). To further verify the efficiency in removing the precursors of regulated and unregulated DBPs by Al-based ECF for drinking water treatment, the halogenated DBPFP after ECF was evaluated. It has exhibited that TCM, MCAA, DCAA, TCAA, DCAN, TCAN, DCK, and TCK were only found for this study, as seen in Figure S5, Supplementary Material. The reduction in the precursors of TCM and TCK after ECF is observed followed by other DBPs. In addition, brominated DBPs were not formed since bromide does not exist in MA suspensions for this study. As shown in Figure 8(a), total DBPFP would increase to 60 μg/L from 48 μg/L first after EC and then decrease to 40 μg/L after subsequent flotation. In this case, the precursors of THMs and HKs mostly increase due to the occurrence of stressed MA cells by EC, which leads to their higher DBPFP (Wang et al. 2018). In other words, the negative reduction in the precursors of THM and HK is 35 and 50% after EC, respectively. After subsequent flotation, the amounts of all DBPs precursors decrease to a minimum at which the precursors of THMs are removed up to 23%, followed by HANs, HKs, and HAAs, as illustrated in Figure 8(b). However, limited reductions in HANs, HKs, and HAAs are found at around 10%. However, the released precursors of HKs and THMs after EC are still effectively removed by subsequent flotation, especially for TCK and TCM. Although SMPL substances are dominant in MA suspensions, they preferentially contribute HAAs formation instead of THMs formation during disinfection (Hong et al. 2008). It has been proven that HAL substances are the dominant precursors of THMs and they can be significantly removed up to 70% by sweep flocculation in Al-based coagulation, as evidenced in Figure 7(c), due to their larger molecular weight and hydrophobic features (Lin et al. 2022). In contrast, the precursors of HAAs and HANs, such as SMPL and APL substances, are insignificantly destabilized by coagulation owing to their high hydrophilicity (Hua et al. 2018), as evidenced in Figure 7(a) and 7(b). The findings on the halogenated DBPs formation during ECF have indicated that the ECF process is effective to minimize DBPFP of MA suspensions, especially for the alleviation of THMs and HKs formation, regardless of the occurrence of increased AOM-derived DBPs’ precursors after EC.
Figure 8
Variations in total DBPFP and its reduction during ECF in terms of (a) total DBPFP and (b) DBPFP reduction ratio (pH: 8; CD:5 mA/cm2; Cl2:DOC = 5:1).
View largeDownload slide

Variations in total DBPFP and its reduction during ECF in terms of (a) total DBPFP and (b) DBPFP reduction ratio (pH: 8; CD:5 mA/cm2; Cl2:DOC = 5:1).

Figure 8
Variations in total DBPFP and its reduction during ECF in terms of (a) total DBPFP and (b) DBPFP reduction ratio (pH: 8; CD:5 mA/cm2; Cl2:DOC = 5:1).
View largeDownload slide

Variations in total DBPFP and its reduction during ECF in terms of (a) total DBPFP and (b) DBPFP reduction ratio (pH: 8; CD:5 mA/cm2; Cl2:DOC = 5:1).

Close modal
Listen

The Al-based ECF process is highly effective to remove MA cells together with the minimization of halogenated DBPs precursors. It is found that the pH effect is more pronounced than the CD effect in terms of MA cell separation efficiency and soluble AOM removal, accounting for 97.5 and 56%, respectively, at pH 8 using a mild CD of 5 mA/cm2. This significant reduction of MA cells is induced by strong sweep flocculation. At such a condition, the MA cells rupture significantly after EC, resulting in a pronounced increase in SMPL and APL substances, and subsequently, they are effectively removed after flotation. Likewise, bEOM can be markedly excreted from the MA cells after EC, but the increased bEOM is tremendously removed by flotation along with effective cell separation. In contrast, the amounts of dEOM increase after cells rupture by EC, but there are only limited improvements in dEOM removal after subsequent flotation. Owing to the decomposition of MA cells after EC, the precursors of THMs and HKs dominantly increase, but the subsequent flotation is able to significantly remove all DBPs precursors to a minimum, particularly for THMs and HKs precursors (i.e., HAL substances). It is concluded that ECF is a promising technique for highly efficient cyanobacteria separation, which predominately contributes to a favorable alleviation of the regulated THMs formation potential, accompanying the least mitigation of HAAs formation potential in drinking water treatment.

Listen

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

Listen

The authors declare there is no conflict.

An
J.
,
Li
N.
,
Wang
S.
,
Liao
C.
,
Zhou
L.
,
Li
T.
,
Wang
X.
 &
Feng
Y.
2019
A novel electro-coagulation-Fenton for energy efficient cyanobacteria and cyanotoxins removal without chemical addition
.
J. Hazard. Mater.
365
,
650
–
658
.
https://doi.org/10.1016/j.jhazmat.2018.11.058
.
Google Scholar
Crossref
Search ADS
PubMed
 
Bagastyo
A. Y.
,
Sidik
F.
,
Anggrainy
A. D.
,
Lin
J. L.
 &
Nurhayati
E.
2022
The performance of electrocoagulation process in removing organic and nitrogenous compounds from landfill leachate in a three-compartment reactor
.
J. Ecol. Eng.
23
,
235
–
245
.
https://doi.org/10.12911/22998993/145290
.
Google Scholar
Crossref
Search ADS
 
Cao
B.
,
Zhang
Y.
,
Shi
Y.
,
Ren
R.
,
Wu
H.
,
Zhang
W.
,
Wang
D.
,
Zhang
T.
 &
Xiong
J.
2020
Extracellular organic matter (EOM) distribution characteristic in algae electro-dewatering process
.
J. Environ. Manage.
265
,
110541
.
https://doi.org/10.1016/j.jenvman.2020.110541
.
Google Scholar
Crossref
Search ADS
PubMed
 
Fuente
A. d. l.
,
Muro-Pastor
A. M.
,
Merchán
F.
,
Madrid
F.
,
Pérez-Martínez
J. I.
 &
Undabeytia
T.
2019
Electrocoagulation/flocculation of cyanobacteria from surface waters
.
J. Clean. Prod.
238
.
https://doi.org/10.1016/j.jclepro.2019.117964
Google Scholar
 
Hashim
K. S.
,
Al Khaddar
R.
,
Jasim
N.
,
Shaw
A.
,
Phipps
D.
,
Kot
P.
,
Pedrola
M. O.
,
Alattabi
A. W.
,
Abdulredha
M.
 &
Alawsh
R.
2019
Electrocoagulation as a green technology for phosphate removal from river water
.
Sep. Purif. Technol.
210
,
135
–
144
.
https://doi.org/10.1016/j.seppur.2018.07.056
.
Google Scholar
Crossref
Search ADS
 
Hong
H. C.
,
Mazumder
A.
,
Wong
M. H.
 &
Liang
Y.
2008
Yield of trihalomethanes and haloacetic acids upon chlorinating algal cells, and its prediction via algal cellular biochemical composition
.
Water Res.
42
,
4941
–
4948
.
https://doi.org/10.1016/j.watres.2008.09.019
.
Google Scholar
Crossref
Search ADS
PubMed
 
Hoseinzadeh
E.
,
Wei
C.
,
Farzadkia
M.
 &
Rezaee
A.
2020
Effects of low frequency-low voltage alternating electric current on apoptosis progression in bioelectrical reactor biofilm
.
Front. Bioeng. Biotechnol.
8
.
https://doi.org/10.3389/fbioe.2020.00002
.
Google Scholar
 
Hua
L. C.
,
Lin
J. L.
,
Syue
M. Y.
,
Huang
C.
 &
Chen
P. C.
2018
Optical properties of algogenic organic matter within the growth period of Chlorella sp. and predicting their disinfection by-product formation
.
Sci. Total Environ.
621
,
1467
–
1474
.
https://doi.org/10.1016/j.scitotenv.2017.10.082
.
Google Scholar
Crossref
Search ADS
PubMed
 
Huang
Y.
,
Ding
S.
,
Li
L.
,
Liao
Q.
,
Chu
W.
 &
Li
H.
2021
Ultrasound-enhanced coagulation for microcystis aeruginosa removal and disinfection by-product control during subsequent chlorination
.
Water Res.
201
,
117334
.
https://doi.org/10.1016/j.watres.2021.117334
.
Google Scholar
Crossref
Search ADS
PubMed
 
Lin
J. L.
,
Nugrayanti
M. S.
,
Ika
A. R.
 &
Karangan
A.
2021
Removal of Microcystis aeruginosa by oxidation-assisted coagulation: effect of algogenic organic matter fraction changes on algae destabilization with Al hydrates
.
J. Water Process Eng.
42
,
102142
.
https://doi.org/10.1016/j.jwpe.2021.102142
.
Google Scholar
Crossref
Search ADS
 
Lin
J. L.
,
Nugrayanti
M. S.
 &
Karangan
A.
2022
Effect of Al hydrates on minimization of disinfection-by-products precursors by coagulation with intensified pre-oxidation towards cyanobacteria-laden water
.
Sci. Total Environ.
810
,
152251
.
https://doi.org/10.1016/j.scitotenv.2021.152251
.
Google Scholar
Crossref
Search ADS
PubMed
 
Ma
J.
,
Zhang
R.
,
Xia
W.
,
Kong
Y.
,
Nie
Y.
,
Zhou
Y.
 &
Zhang
C.
2022
Coagulation performance of Al/Fe based covalently bonded composite coagulants for algae removal
.
Sep. Purif. Technol.
285
,
120401
.
https://doi.org/10.1016/j.seppur.2021.120401
.
Google Scholar
Crossref
Search ADS
 
Menezes
I.
,
Maxwell-McQueeney
D.
,
Capelo-Neto
J.
,
Pestana
C. J.
,
Edwards
C.
 &
Lawton
L. A.
2021
Oxidative stress in the cyanobacterium Microcystis aeruginosa PCC 7813: comparison of different analytical cell stress detection assays
.
Chemosphere
269
.
https://doi.org/10.1016/j.chemosphere.2020.128766
.
Google Scholar
 
Naceradska
J.
,
Pivokonska
L.
 &
Pivokonsky
M.
2019
On the importance of pH value in coagulation
.
J. Water Supply Res. Technol. – AQUA
68
,
222
–
230
.
https://doi.org/10.2166/aqua.2019.155
.
Google Scholar
Crossref
Search ADS
 
Putri, M. S. A., Lin, J. L., Hsieh, L. H. C., Zafirah, Y., Andhikaputra, G. & Wang, Y. C. 2020 Influencing factors analysis of Taiwan eutrophicated reservoirs. Water (Switzerland) 12, 1–16. https://doi.org/10.3390/W12051325.
Richardson
S. D.
 &
Ternes
T. A.
2018
Water analysis: emerging contaminants and current issues
.
Anal. Chem.
90
,
398
–
428
.
https://doi.org/10.1021/acs.analchem.7b04577
.
Google Scholar
Crossref
Search ADS
PubMed
 
Roostaei
J.
,
Zhang
Y.
,
Gopalakrishnan
K.
 &
Ochocki
A. J.
2018
Mixotrophic microalgae biofilm: a novel algae cultivation strategy for improved productivity and cost-efficiency of biofuel feedstock production
.
Sci. Rep.
8
,
1
–
10
.
https://doi.org/10.1038/s41598-018-31016-1
.
Google Scholar
Crossref
Search ADS
PubMed
 
Sillanpää
M.
,
Ncibi
M. C.
,
Matilainen
A.
 &
Vepsäläinen
M.
2018
Removal of natural organic matter in drinking water treatment by coagulation: a comprehensive review
.
Chemosphere
190
,
54
–
71
.
https://doi.org/10.1016/j.chemosphere.2017.09.113
.
Google Scholar
Crossref
Search ADS
PubMed
 
Tang
X.
,
Zheng
H.
,
Gao
B.
,
Zhao
C.
,
Liu
B.
,
Chen
W.
 &
Guo
J.
2017
Interactions of specific extracellular organic matter and polyaluminum chloride and their roles in the algae-polluted water treatment
.
J. Hazard. Mater.
332
,
1
–
9
.
https://doi.org/10.1016/j.jhazmat.2017.02.060
.
Google Scholar
Crossref
Search ADS
PubMed
 
U.S. EPA
1995
Standard Method 551.1: Determination of Chlorination Disinfection Byproducts, Chlorinated Solvents, and Halogenated Pesticides/Herbicides in Drinking Water by Liquid-Liquid Microextraction, Derivatization, and Gas Chromatography with Electron Capture Detec
.
Environmental Monitoring and System Laboratory, Cincinnati, OH, USA
.
U.S. EPA
2003
Standard Method 552.3: Determination of Haloacetic Acids and Dalapon in Drinking Water by Liquid-Liquid Microextraction, Derivatization, and Gas Chromatography with Electron Capture Detection
.
Environmental Monitoring and System Laboratory, Cincinnati, OH, USA.
Wang
X.
,
Xiang
P.
,
Zhang
Y.
,
Wan
Y.
 &
Lian
H.
2018
The inhibition of Microcystis aeruginosa by electrochemical oxidation using boron-doped diamond electrode
.
Environ. Sci. Pollut. Res.
25
,
20631
–
20639
.
https://doi.org/10.1007/s11356-018-1977-3
.
Google Scholar
Crossref
Search ADS
 
Wang
X.
,
Xu
H.
 &
Wang
D.
2020
Mechanism of fluoride removal by AlCl3 and Al13: the role of aluminum speciation
.
J. Hazard. Mater.
398
,
122987
.
https://doi.org/10.1016/j.jhazmat.2020.122987
.
Google Scholar
Crossref
Search ADS
PubMed
 
Wu
T.
,
Zhu
G.
,
Zhu
M.
,
Xu
H.
,
Yang
J.
 &
Zhao
X.
2021
Effects of algae proliferation and density current on the vertical distribution of odor compounds in drinking water reservoirs in summer
.
Environ. Pollut.
288
,
117683
.
https://doi.org/10.1016/j.envpol.2021.117683
.
Google Scholar
Crossref
Search ADS
PubMed
 
Xie
W. M.
,
Qiao
L. L.
,
Fang
F.
,
Li
W. W.
,
Meng
H.
,
Wang
G. X.
 &
Zhang
L. M.
2019
Dynamic characteristics of soluble microbial products in a granular sludge reactor
.
J. Clean. Prod.
212
,
576
–
581
.
https://doi.org/10.1016/j.jclepro.2018.12.082
.
Google Scholar
Crossref
Search ADS
 
Yamaguchi
A.
,
Kobayashi
M.
 &
Adachi
Y.
2019
Yield stress of mixed suspension of silica particles and lysozymes: the effect of zeta potential and adsorbed amount
.
Colloids Surf. A Physicochem. Eng. Asp.
578
,
123575
.
https://doi.org/10.1016/j.colsurfa.2019.123575
.
Google Scholar
Crossref
Search ADS
 
Zhao
Z.
,
Sun
W.
,
Ray
M. B.
,
Ray
A. K.
,
Huang
T.
 &
Chen
J.
2019
Optimization and modeling of coagulation-flocculation to remove algae and organic matter from surface water by response surface methodology
.
Front. Environ. Sci. Eng.
13
.
https://doi.org/10.1007/s11783-019-1159-7
.
Google Scholar
 
© 2022 The Authors
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).