Comparison of Di ﬀ erent Anode Materials to Remove Microcystis aeruginosa Cells Using Electro-Coagulation–Flotation Process at Low Current Inputs

: Cyanobacterial blooms are a threat to the drinking water supply owing to their potential toxicity. Microcystins which are the most widespread cyanotoxins, are mainly produced by Microcystis spp. In this study, we cultured Microcystis aeruginosa cells in BG-11 medium at 25 ◦ C to investigate the e ﬃ ciency of the electro-coagulation–ﬂotation process to remove them. Di ﬀ erent anode materials (Fe, Al, Cu, and Zn) along with a graphite cathode were compared separately in the 10–100 mA current range in a 0.025 M Na 2 SO 4 electrolyte. Turbidity, optical density at 684 nm (OD 684 ), OD 730 , Chl-a concentration, and DOC concentration were analyzed to clarify the mechanism by which M. aeruginosa cells are removed. The Al anode indicated the highest removal e ﬃ ciencies in terms of turbidity (90%), OD 684 and OD 730 (98%), and Chl-a concentration (96%) within 30 min at 4.0 mA / cm 2 current density and the lowest average electrode consumption of 0.120 ± 0.023 g / L. The energy consumption of the Al electrode was 0.80 Wh / L. From these results, we found that Al was the best among the anode materials evaluated to remove M. aeruginosa cells. However, further studies are required to optimize the Al anode in terms of pH, treatment time, electrode distance, and electrolyte concentration to enhance the removal of M. aeruginosa cells.


Introduction
Cyanobacterial blooms are one of the major problems worldwide and they can be harmful to the environment, animals, and human health. As a result of rapid oxygen consumption, hypoxic conditions develop, resulting in plant and animal die-off in water bodies [1]. These blooms are a threat to the drinking water supply owing to their potential toxicity and the release of taste and odor compounds such as geosmin and 2-methyl-isoborneol (2-MIB) [2,3]. Generally, cyanobacteria release a large amount of soluble algogenic organic matter (AOM) to the water such as extracellular organic matter (EOM), which is expelled to the water by living algal cells as a product of their metabolic activities, and intracellular organic matter (IOM), which is secreted upon the rupture of cells caused by endogenous or exogenous factors [4][5][6]. IOM is released as a result of cell lysis when cyanobacteria go through water treatment processes, such as ozonation or chlorination and it is a source of precursor materials for the formation of carcinogenic disinfection by-products (DBPs) combining with chlorine [3,7]. Cyanobacteria have been recognized for decades as a potential source of precursor materials for the formation of DBPs such as trihalomethane (THM) and haloacetic acid (HAA) [3]. The dominant AOMs such as hydrophilic polysaccharides and hydrophobic proteins are At the cathode: In solution: The removal of cyanobacteria through the sedimentation process is challenging because they have a lower density than water owing to gas vacuoles in their cells for regulating the water content [15]. Therefore, the ECF process might be effective and low-cost for cyanobacterial removal.
Only a few studies have focused on the effect of the anode material, mainly Al and/or iron (Fe), on the removal of algae and cyanobacteria by the ECF process. In this research, we compared four anode materials, Fe, Al, copper (Cu), and zinc (Zn), to determine the effect of the anode material on the removal of Microcystis aeruginosa cells.

Preparation of Microcystis aeruginosa Cultures
To prepare new M. aeruginosa cultures, we obtained samples from a M. aeruginosa monoculture maintained in our laboratory. We used BG-11 medium prepared as described by Rippka et al. [18]. Table 1 shows the ingredients and quantities that were used to prepare the BG-11 medium. We used twelve 500 mL Erlenmeyer flasks to culture M. aeruginosa. Before inoculation, flasks, rubber stoppers, and the BG-11 medium were sterilized at 121 • C and 0.12 MPa for 20 min in an autoclave (SP300, Yamato Scientific Co., Ltd., Tokyo, Japan) [19]. Each flask was filled with 250 mL of BG-11 medium and closed with a rubber stopper (SILICONE Sterile Stopper, C-40 joint size, AS ONE Corporation, Osaka, Japan) to maintain a proper aeration for their growth. We maintained the cultures in an incubator (LH-55RDS, NK Systems Limited, Tokyo, Japan) at 25 ± 1 • C temperature providing a 12:12 light and dark period and a 60 µmol m −2 s −1 photon flux of photosynthetically active radiation [20][21][22]. The flasks with cultures were gently shaken about three times a day to ensure equal exposure of M. aeruginosa cells to the light. Figure 1a,b show the appearance of M. aeruginosa cultures at the initial stage and after 14 days of growth, respectively. The initial average pH, turbidity, and optical densities at 684 nm and 730 nm (OD 684 , and OD 730 ) of the culture were 7.40 ± 0.06, 6 ± 1 NTU, 0.0309 ± 0.0012, and 0.0322 ± 0.0018, respectively. Cultures showed an S-shaped growth curve and reached an exponential growth within four days. In the exponential phase of growth (14 days after inoculation), about 2.5 L volume of cultures was harvested and gently stirred to obtain a homogenous sample. Before starting the experiments, the pH and electrical conductivity (EC) of samples were adjusted using 0.1 M NaOH or 0.1 M HCl solution and Na 2 SO 4 , respectively. Table 2 shows the characteristics of M. aeruginosa suspensions prepared for the ECF treatment.  We used twelve 500 mL Erlenmeyer flasks to culture M. aeruginosa. Before inoculation, flasks, rubber stoppers, and the BG-11 medium were sterilized at 121 °C and 0.12 MPa for 20 min in an autoclave (SP300, Yamato Scientific Co., Ltd., Tokyo, Japan) [19]. Each flask was filled with 250 mL of BG-11 medium and closed with a rubber stopper (SILICONE Sterile Stopper, C-40 joint size, AS ONE Corporation, Osaka, Japan) to maintain a proper aeration for their growth. We maintained the cultures in an incubator (LH-55RDS, NK Systems Limited, Tokyo, Japan) at 25 ± 1 °C temperature providing a 12:12 light and dark period and a 60 μmol m −2 s −1 photon flux of photosynthetically active radiation [20][21][22]. The flasks with cultures were gently shaken about three times a day to ensure equal exposure of M. aeruginosa cells to the light. Figure 1a,b show the appearance of M. aeruginosa cultures at the initial stage and after 14 days of growth, respectively. The initial average pH, turbidity, and optical densities at 684 nm and 730 nm (OD684, and OD730) of the culture were 7.40 ± 0.06, 6 ± 1 NTU, 0.0309 ± 0.0012, and 0.0322 ± 0.0018, respectively. Cultures showed an S-shaped growth curve and reached an exponential growth within four days. In the exponential phase of growth (14 days after inoculation), about 2.5 L volume of cultures was harvested and gently stirred to obtain a homogenous sample. Before starting the experiments, the pH and electrical conductivity (EC) of samples were adjusted using 0.1 M NaOH or 0.1 M HCl solution and Na2SO4, respectively. Table 2 shows the characteristics of M. aeruginosa suspensions prepared for the ECF treatment.

Experimental Setup and Electrode Materials
As shown in Figure 2, all the experiments were carried out at room temperature using 200 mL of M. aeruginosa suspension mixed at a speed of 200 rpm using a magnetic stirrer (PC-320, CORNING,

Experimental Setup and Electrode Materials
As shown in Figure 2, all the experiments were carried out at room temperature using 200 mL of M. aeruginosa suspension mixed at a speed of 200 rpm using a magnetic stirrer (PC-320, CORNING, Corning, NY, USA). The electrode distance was maintained at 2 cm for each experiment [13]. The effective areas of the anodes and cathode were 25 cm 2 and 30 cm 2 , respectively. We compared the anode materials along with the graphite cathode separately in the 10 mA-100 mA current range using 0.025 M Water 2020, 12, 3528 4 of 13 Na 2 SO 4 electrolyte. The current density was in the range of 0.4-4 mA/cm 2 . The DC voltage was in the range from 1 to 3.8 V. DC power was supplied by a regulated DC digital power supply unit (PMC35-2A, KIKUSUI, Kanagawa, Japan). After conducting each set of experiments, the electrodes were washed with 0.01 M HCl, thoroughly washed with distilled water to remove residuals on the surface, and then weighed.
Corning, NY, USA). The electrode distance was maintained at 2 cm for each experiment [13]. The effective areas of the anodes and cathode were 25 cm 2 and 30 cm 2 , respectively. We compared the anode materials along with the graphite cathode separately in the 10 mA-100 mA current range using 0.025 M Na2SO4 electrolyte. The current density was in the range of 0.4-4 mA/cm 2 . The DC voltage was in the range from 1 to 3.8 V. DC power was supplied by a regulated DC digital power supply unit (PMC35-2A, KIKUSUI, Kanagawa, Japan). After conducting each set of experiments, the electrodes were washed with 0.01 M HCl, thoroughly washed with distilled water to remove residuals on the surface, and then weighed.
Anode materials and a graphite cathode were purchased from Ajax Scientifc Ltd., Scarborough, ON, Canada. Fe, Al, Cu, and Zn electrodes were selected for the study because these metals are less toxic for humans. All electrodes were rectangular in shape. The details of different anodes are shown in Table 3.

Analytical Methods
All chemicals used in the study were purchased from FUJIFILM Wako Pure Chemical Industries, Ltd., Osaka, Japan. All physicochemical analyses of water samples were carried out in accordance with standard methods for the examination of water and wastewater [23].
Samples for water quality analyses were collected from 2 cm below the surface using a pipette without disturbing the flocs. Turbidity was measured using a laboratory turbidity meter (2100N, HACH Company, Loveland, CO, USA) and pH was measured using a portable digital meter (Orion Anode materials and a graphite cathode were purchased from Ajax Scientifc Ltd., Scarborough, ON, Canada. Fe, Al, Cu, and Zn electrodes were selected for the study because these metals are less toxic for humans. All electrodes were rectangular in shape. The details of different anodes are shown in Table 3.

Analytical Methods
All chemicals used in the study were purchased from FUJIFILM Wako Pure Chemical Industries, Ltd., Osaka, Japan. All physicochemical analyses of water samples were carried out in accordance with standard methods for the examination of water and wastewater [23].
Samples for water quality analyses were collected from 2 cm below the surface using a pipette without disturbing the flocs. Turbidity was measured using a laboratory turbidity meter (2100N, HACH Company, Loveland, CO, USA) and pH was measured using a portable digital meter (Orion 3 star, Thermo Fisher Scientific Co., Ltd., Waltham, MA, USA). EC and total dissolved solids (TDS) were analyzed using a portable conductivity meter (AS710, AS ONE Corporation, Osaka, Japan).
Several studies demonstrated that algae show high absorbance at 684 nm and 730 nm. It was shown that there is a linear relationship between cell count and OD at 684 nm for M. aeruginosa (R 2 > 0.99) [24,25]. A research carried out by Patel et al. [26] also showed a linear relationship between cell count and OD 730 for five cyanobacteria species (R 2 > 0.99). Therefore, in this study, ODs at 684 nm and 730 nm were analyzed using a UV-Vis spectrophotometer (UV-1280, Shimadzu, Kyoto, Japan). Dissolved organic carbon (DOC) concentration was measured for samples filtered through a 0.45 µm filter paper by a method based on the catalytic oxidation using a TOC analyzer (TOC-2300, Hiranuma Sangyo Co., Ltd., Ibaraki, Japan). Fluorescence microscopy images of M. aeruginosa cells were obtained before and after ECF treatment (BZ-X810, KEYENCE Corporation, Osaka, Japan).
The chlorophyll-a (Chl-a) concentration of water samples was measured as described by Zavřel et al. [27]. One milliliter each of sample was dispensed into separate Eppendorf safe-lock tubes and the tubes were centrifuged at 15,000× g for 7 min in a refrigerated microcentrifuge (TOMY MX-105, Tomy Seiko Co., Ltd., Tokyo, Japan). After discarding the supernatant, 1 mL of methanol was added and the samples were incubated at +4 • C for 20 min to extract pigments from the cells. The samples were again centrifuged at 15,000× g for 7 min and the absorbance of the supernatant was measured at wavelengths of 665 and 720 nm using a UV-Vis spectrophotometer (UV-1280, Shimadzu, Kyoto, Japan).
Chl-a concentration was determined using Equation (5).
A 665 : absorbance at a wavelength of 665 nm A 720 : absorbance at a wavelength of 720 nm Energy consumption was calculated in Wh per L using the Equation (6) reported by Ozyonar et al. [16].

Energy consumption
Wh V: Voltage (volt) I: Current (ampere) t: Operation time (hours) v: Volume of the treated solution (L) The dissociation of ions from the anode follows the Faraday's law. The theoretical electrode consumption was calculated in g per L using the Equation (7) given by Ozyonar et al. [16].

Statistical Analyses
All experiments were carried out in triplicate. Statistical analyses were carried out using IBM SPSS Statistics 20.0 software. The removal efficiencies in terms of turbidity, OD 684 , OD 730 , Chl-a concentration, and DOC concentration achieved after the ECF treatment were compared among different anode materials by one-way ANOVA post-hoc Tukey HSD (Honestly significant difference) tests to identify the best anode material. All statistical analyses used a significance level of 5% (p ≤ 0.05).

Comparison of Treated Water Characteristics
The applied current was as low as 10 mA at the start of each experiment and was gradually increased. There were no significant changes observed during the first 30 min of treatment of M.
Water 2020, 12, 3528 6 of 13 aeruginosa suspensions until the applied current reached 100 mA. Table 4 shows the results of treated water after the 30 min treatment and 30 min settling time at 100 mA applied current. The initial pH was adjusted to 7.00 ± 0.01 before the start of each experiment. The highest and lowest average pHs were recorded for the Cu and Fe anodes: 11.46 ± 0.48 and 8.88 ± 0.34, respectively. The initial average EC and TDS were 3893.67 ± 4.73 µS/cm and 3002 ± 2.65 mg/L, respectively, after adding Na 2 SO 4 electrolyte. The treatment carried out using the Al anode showed slight reductions in the average EC and TDS to 3776.67 ± 0.0058 µS/cm and 2890 ± 0.01 mg/L, respectively, but the other anode materials showed slightly higher values than initial values. The Al anode showed the optimal average treated water turbidity, OD 684 , OD 730 , and Chl-a concentration: 15.33 ± 1.53 NTU, 0.0168 ± 0.0047, 0.0164 ± 0.0047, and 0.0063 ± 0.0001 µg/mL, respectively. However, the lowest average DOC concentration of 21.85 ± 2.85 mg/L was obtained when the Cu anode was used. Figure 3 shows the appearance of M. aeruginosa suspensions before and after treatment using different anode materials. As shown in Figure 3c, the ECF treatment carried out using the Al anode showed the clearest water after the treatment. In the case of using the Fe anode (b), the suspension in the beaker turned yellowish-green at the beginning of the ECF treatment and then changed to yellowish-brown with the time. The suspension treated using the Cu anode (d) appeared a light blue at the final stage of the treatment.
Water 2020, 12, x FOR PEER REVIEW 6 of 14

Comparison of Treated Water Characteristics
The applied current was as low as 10 mA at the start of each experiment and was gradually increased. There were no significant changes observed during the first 30 min of treatment of M. aeruginosa suspensions until the applied current reached 100 mA. Table 4 shows the results of treated water after the 30 min treatment and 30 min settling time at 100 mA applied current.
The initial pH was adjusted to 7.00 ± 0.01 before the start of each experiment. The highest and lowest average pHs were recorded for the Cu and Fe anodes: 11.46 ± 0.48 and 8.88 ± 0.34, respectively. The initial average EC and TDS were 3893.67 ± 4.73 μS/cm and 3002 ± 2.65 mg/L, respectively, after adding Na2SO4 electrolyte. The treatment carried out using the Al anode showed slight reductions in the average EC and TDS to 3776.67 ± 0.0058 μS/cm and 2890 ± 0.01 mg/L, respectively, but the other anode materials showed slightly higher values than initial values. The Al anode showed the optimal average treated water turbidity, OD684, OD730, and Chl-a concentration: 15.33 ± 1.53 NTU, 0.0168 ± 0.0047, 0.0164 ± 0.0047, and 0.0063 ± 0.0001 μg/mL, respectively. However, the lowest average DOC concentration of 21.85 ± 2.85 mg/L was obtained when the Cu anode was used.  Figure 3 shows the appearance of M. aeruginosa suspensions before and after treatment using different anode materials. As shown in Figure 3c, the ECF treatment carried out using the Al anode showed the clearest water after the treatment. In the case of using the Fe anode (b), the suspension in the beaker turned yellowish-green at the beginning of the ECF treatment and then changed to yellowish-brown with the time. The suspension treated using the Cu anode (d) appeared a light blue at the final stage of the treatment.     As shown in Figure 5, the ECF treatment carried out using the Al anode showed the clearest water after the treatment and settling process. In contrast, treatments using Fe, Cu and Zn anodes produced yellowish-brown, light blue, and white cloudy effluents, respectively. Final average EC values were higher than initial EC value in suspensions treated with Fe, Cu and Zn. However, the final EC was 3776.67 ± 0.0058 μS/cm for the suspension treated with Al anode. It was slightly lower than initial EC value.   As shown in Figure 5, the ECF treatment carried out using the Al anode showed the clearest water after the treatment and settling process. In contrast, treatments using Fe, Cu and Zn anodes produced yellowish-brown, light blue, and white cloudy effluents, respectively. Final average EC values were higher than initial EC value in suspensions treated with Fe, Cu and Zn. However, the final EC was 3776.67 ± 0.0058 µS/cm for the suspension treated with Al anode. It was slightly lower than initial EC value.  As shown in Figure 5, the ECF treatment carried out using the Al anode showed the clearest water after the treatment and settling process. In contrast, treatments using Fe, Cu and Zn anodes produced yellowish-brown, light blue, and white cloudy effluents, respectively. Final average EC values were higher than initial EC value in suspensions treated with Fe, Cu and Zn. However, the final EC was 3776.67 ± 0.0058 μS/cm for the suspension treated with Al anode. It was slightly lower than initial EC value.     Figure 7 shows the actual and theoretical anode consumptions of different anode materials. The same applied current of 100 mA was used for each experiment using different anodes. The Al and Cu anodes showed the lowest and highest average anode consumptions of 0.120 ± 0.023 and 0.351 ± 0.170 g/L, respectively. However, the theoretical anode consumptions were lower than the actual anode consumptions for Al and Cu anode materials (Figure 7).    Figure 7 shows the actual and theoretical anode consumptions of different anode materials. The same applied current of 100 mA was used for each experiment using different anodes. The Al and Cu anodes showed the lowest and highest average anode consumptions of 0.120 ± 0.023 and 0.351 ± 0.170 g/L, respectively. However, the theoretical anode consumptions were lower than the actual anode consumptions for Al and Cu anode materials (Figure 7).  Figure 7 shows the actual and theoretical anode consumptions of different anode materials. The same applied current of 100 mA was used for each experiment using different anodes. The Al and Cu anodes showed the lowest and highest average anode consumptions of 0.120 ± 0.023 and 0.351 ± 0.170 g/L, respectively. However, the theoretical anode consumptions were lower than the actual anode consumptions for Al and Cu anode materials (Figure 7).

Discussion
In this study, we conducted comparative experiments using Fe, Al, Cu, and Zn as anode materials under the same operational condition for M. aeruginosa removal. We used indirect measurements such as turbidity, OD, Chl-a concentration, and DOC concentration to measure M. aeruginosa concentrations and AOM at before and after treatment. The Al anode showed the highest removal efficiency in terms of all these parameters except DOC concentration ( Figure 6). In this investigation, we started with a current input as low as 10 mA and gradually increased it. There were no any significant changes observed during the first 30 min of treatment in M. aeruginosa suspensions until the applied current reached 100 mA. The Al anode showed the highest removal efficiency of 98% for M. aeruginosa at 100 mA current and 4 mA/cm 2 current density for 30 min treatment. A quite similar result was reported by Gao et al. [15]. They demonstrated that 25 min treatment was adequate for 100% M. aeruginosa removal at 5 mA/cm 2 current density. According to Faraday's law, the amount

Discussion
In this study, we conducted comparative experiments using Fe, Al, Cu, and Zn as anode materials under the same operational condition for M. aeruginosa removal. We used indirect measurements such as turbidity, OD, Chl-a concentration, and DOC concentration to measure M. aeruginosa concentrations and AOM at before and after treatment. The Al anode showed the highest removal efficiency in terms of all these parameters except DOC concentration ( Figure 6). In this investigation, we started with a current input as low as 10 mA and gradually increased it. There were no any significant changes observed during the first 30 min of treatment in M. aeruginosa suspensions until the applied current reached 100 mA. The Al anode showed the highest removal efficiency of 98% for M. aeruginosa at 100 mA current and 4 mA/cm 2 current density for 30 min treatment. A quite similar result was reported by Gao et al. [15]. They demonstrated that 25 min treatment was adequate for 100% M. aeruginosa removal at 5 mA/cm 2 current density. According to Faraday's law, the amount

Discussion
In this study, we conducted comparative experiments using Fe, Al, Cu, and Zn as anode materials under the same operational condition for M. aeruginosa removal. We used indirect measurements such as turbidity, OD, Chl-a concentration, and DOC concentration to measure M. aeruginosa concentrations and AOM at before and after treatment. The Al anode showed the highest removal efficiency in terms of all these parameters except DOC concentration ( Figure 6). In this investigation, we started with a current input as low as 10 mA and gradually increased it. There were no any significant changes observed during the first 30 min of treatment in M. aeruginosa suspensions until the applied current reached 100 mA. The Al anode showed the highest removal efficiency of 98% for M. aeruginosa at 100 mA current and 4 mA/cm 2 current density for 30 min treatment. A quite similar result was reported by Gao et al. [15]. They demonstrated that 25 min treatment was adequate for 100% M. aeruginosa removal at 5 mA/cm 2 current density. According to Faraday's law, the amount of metal ions released from the anode to the solution increases with increasing current input and electrolysis time. These metal ions incorporated with OH − ions promotes the aggregation of cyanobacterial cells to form flocs [28].
Moreover, the density and size of microbubbles increases and decreases with increasing current input, respectively, which promotes flotation [29].
The initial pH is one of the most critical factors affecting the ECF process [17]. In this study, we maintained the initial pH at 7.00 ± 0.01. However, the pH increased to alkaline levels after ECF treatment for all the anode materials investigated. This may be due to OHions generated at the cathode as a result of metal ion oxidation at the anode. A similar result was reported by Bleeke et al. [30] for the treatment of Scenedesmus acuminatus using the same anode materials. The Cu anode showed the highest final average pH of 11.46 ± 0.48 (Table 4). Tumsri et al. [31] also mentioned that the pH increased markedly during electrolysis when using the Cu anode.
According to Gao et al. [15], at the neutral pH of 7, the amount of aluminum hydroxide is much higher than that of other metal hydroxides. Moreover, a higher current efficiency can be generated by Al electrodes than by Fe, Cu, or Zn electrodes. Owing to hydrolysis reactions, Al ions are hydrolyzed and produce various types of Al species such as monomeric species (e.g., Al(OH) 2+ and Al(OH) 2+ ) and polymeric species (e.g., Al 13 (OH) 34 5+ and Al 13 O 4 (OH) 24 7+ ). These cationic products react with OH − ions, which transform into amorphous Al(OH) 3 according to complex precipitation kinetics [17,31]. In this study, the Zn anode showed the lowest electrical consumption (0.675 Wh/L); however, its removal efficiency was lower. The electrical consumption of the Al anode was 0.80 Wh/L ( Figure 8) with respect to the lowest anode consumption of 0.12 ± 0.023 g/L ( Figure 7) at pH 7 for 30 min treatment with 98% M. aeruginosa removal efficiency. Gao et al. [15] reported the electrical consumptions of the Al electrode at different pHs for the complete removal of M. aeruginosa by 75 min treatment. The electrical consumption gradually increased with increasing pH: 0.30 Wh/L from pH 4 to 7, 0.53 Wh/L from pH 7 to 9, and 0.60 Wh/L at the initial pH of 10. An experiment carried out for Chlorella vulgaris showed that the algae removal efficiency was 96-98.1% for 30 min electrolysis at 6-7 initial pH and 2 mA/cm 2 current density. The energy consumption was 0.91 Wh/L [31]. At a lower pH, low amount of Al ions is required to remove algal cells owing to charge neutralization. However, when pH is increased, sweep flocculation dominates and a higher amount of Al ions are required to achieve a similar removal efficiency [15]. It is considered that a high efficiency of cyanobacteria/algae removal could be achieved at a low energy consumption under neutral and acidic pH conditions [13,32].
In this study, we used EC and color as indirect measurements especially to identify the excess metal ion concentrations in treated suspensions. Final EC values were higher than initial EC value in suspensions treated with Fe, Cu and Zn respective to dissolution of excess metal ions. However, the final average EC of suspension treated with Al anode was slightly lower than initial EC value. It is realized that there were lower excess aluminum ions in treated suspension. In the case of the Fe anode, the suspensions in the beaker turned yellowish-green at the beginning of the ECF process and gradually changed to yellowish-brown (Figures 3 and 5). These color changes happen owing to the presence of Fe 2+ and Fe 3+ ions [15]. Higher turbidity and ODs of treated water were obtained when using the Fe anode than when using the Al and Cu anodes (Table 4). Fe at 0.3 mg/L causes turbidity, changes color, and stains laundry and plumbing fixtures. However, there is usually no conspicuous taste of water at concentrations <0.3 mg/L [33]. The suspensions treated using the Cu anode appeared light blue at the final stage of treatment (Figures 3 and 5). It could be due to Cu 2+ species produced during electrolytic dissolution [31]. Cu produces a bitter taste and changes the color of water when its concentration is higher than 5 mg/L, and stains laundry fixtures at higher than 1 mg/L [33]. The Zn anode showed the lowest removal efficiency in terms of turbidity as the final solutions appeared white and cloudy as a result of Zn(OH) 2 formation (Figures 3 and 5). Generally, high pHs enhance the generation of these metal hydroxides [34].
We maintained a 30 min settling time for each experiment and measured the turbidity. During the 30 min settling time, the turbidity decreased for all anodes as a result of the sedimentation of flocs ( Figure 4). However, the reduction in turbidity during the settling time was very small for the Zn and Fe anodes owing to Zn(OH) 2 precipitation, and of the suspension turned yellowish-brown owing to the oxidation of Fe 2+ to Fe 3+ in the presence of oxygen [34,35].
Formation of sludge is a crucial problem as it contains metal ions. Aluminum in sludge can be recovered for subsequent reusing purposes using acidic and alkaline extraction methods. Sulfuric acid (H 2 SO 4 ) is used as the acidification agent in the recovering of the aluminum. This mechanism can also be used to extract other metal ions. Recovery of aluminum in water treatment plant could be a significant approach towards sustainability development [36]. As well as ECF method can use to recover these algae towards the producing of biofuels as a sustainable green energy source.
Regarding the morphology of M. aeruginosa cells before and after ECF treatment, we observed that the cells appeared normal with diameters in the range of 5-10 µm; however, these cells tended to aggregate (Figure 9) after treatment owing to coagulants and the formation of extracellular polysaccharide mucilage around them [37]. In addition, Bakheet et al. [22] reported that more than 30 min electrolysis can disrupt the cell walls and membranes of M. aeruginosa cells, resulting in the release of IOM to the medium [4][5][6]. It is important to remove IOM and EOM precursor materials because they contribute to the formation of carcinogenic DBPs combining with chlorine [3,7]. According to Zhao et al. [38], the EOM and IOM concentrations ranged from very low concentrations to about 100 mg/L depending on the cyanobacterial species. In this research, the initial DOC concentration was 73.85 ± 1.52 mg/L in M. aeruginosa suspensions. The Cu anode showed the highest DOC removal efficiency of 70%, whereas the other anodes did not display a significant removal of DOC ( Figure 6). However, the Cu anode exhibited the highest anode consumption and energy consumption (Figures 7 and 8). According to Zhou et al. [39] Cu ions can interact with proteins, intercellular nucleic acids, enzymes, and metabolites microbial cells after uptake by them. In case of Al anode, we obtained only a 40% DOC removal efficiency. Guo et al. [40] reported a 38.7% DOC removal efficiency from the IOM of M. aeruginosa at 5 mg/L alum dosage. The charge neutralization process involves the elimination of AOM during coagulation owing to the occurrence of extremely negative charges on AOM [6]. However, it is important to carry out these ECF experiments considering the maximum M. aeruginosa removal while preventing the release of IOM via cell lysis [41].

Conclusions
Al was the best anode material among the materials evaluated (i.e., Fe, Cu, and Zn) to remove M. aeruginosa cells at 4.0 mA/cm 2 current density for a 30 min treatment period and a 30 min settling time with the lowest anode consumption. Even though the Cu anode showed the highest DOC removal efficiency, its consumption was high. ECF method with Al anode showed to be a highly effective and rapid method for the removal of M. aeruginosa cells. However, further studies are recommended to optimize the Al anode in terms of pH, treatment time, electrode distance, and electrolyte concentration to obtain the maximum removal of M. aeruginosa cells. Moreover, the residual aluminum ion concentration, the effect of halide ions on anode dissolution, usage of dimensionally stable anodes could be recommended as important studies for the application purposes.