Parameters optimization for eutrophic lake water treatment by a novel process of iron-carbon micro- electrolysis coupled with catalytic ozonation using response surface methodology

A novel process of iron-carbon micro-electrolysis (ICME) coupled with catalytic ozonation (CO) for treatment of eutrophic lake water was developed. A series of batch experiments with ICME alone and CO alone was designed to investigate the effects of process parameters, such as initial pH, dose of Fe-C, time of micro-electrolysis, ozone flux, dose of TiO2/activated carbon (TiO2/AC), and time of ozonation, on the removal rates of total nitrogen (TN), total phosphorus (TP), CODMn and Chl-a. The process parameters were optimized using response surface methodology. The results showed that initial pH, dose of Fe-C and ozone flux had significant effects on removal of TN, TP, CODMn and Chl-a. Within the range of selected operating conditions, the optimized values of initial pH, dose of Fe-C, time of micro-electrolysis, ozone flux, dose of TiO2/AC, and time of ozonation were 3.8, 13.7 g/L, 29.6 min, 3.19 L/min, 294.74 mg/L and 106.73 min, respectively. Furthermore, ICME alone had significant advantages in TP and CODMn removal and CO alone favored TN and Chl-a. Under the optimal process conditions, the final removal rates of TN, TP, CODMn, and Chl-a by the hybrid ICME-CO process reached 75.33%, 86.29%, 94.42% and 97.57%, respectively. The present research provides a new alternative technology with promise for treatment of eutrophic lake water.


INTRODUCTION
According to the latest survey, over 64% of the global lakes are eutrophic (Zhang et al. ). Lake eutrophication occurs when the abundance of potential nutrient elements (including nitrogen (N) and phosphorus (P)) required by aquatic plants are significantly increased, and the photosynthesis rate of lake aquatic organisms greatly increases (Gao et al. ). The eutrophication of water bodies creates many problems: (1) the rapid propagation of algae leads to insufficient light in the water column and (2) a sharp drop in dissolved oxygen. These conditions can result in an unpleasant odor emanating from water that contains widespread deadly aquatic organisms and the occurrence of algal blooms or red tides (Gao et al. ). These present enormous problems for the environment and human health. The 'Taihu Lake Cyanobacteria Pollution Incident' in China in 2007 is one example (Chen et al. ). Hence, the treatment of eutrophic lake water is urgently needed.
In addition to containing high levels of N and P, eutrophic lake water typically contains high concentrations of organic compounds and chlorophyll-a (Chl-a). A variety of physical (Nguyen et al. ), biological (Wu et al. ) and chemical techniques (Huang et al. ) has been used for eutrophic lake water treatment in order to improve the quality of the aquatic ecosystem. Physical techniques, such as adsorption (Qin et al. ) and aeration (Nguyen et al. ), have had excellent effects in pollutant removal, but the main removal targets are N/P for adsorption and organic compounds for aeration, respectively. These techniques lack the ability to remove multiple contaminants as a whole. Biological techniques, such as those involving the addition of microorganisms (Wu et al. ), and the cultivation of aquatic plants (Hu et al. ) in water, can control total nitrogen (TN) and total phosphorus (TP) within a certain range under suitable conditions, but their low removal rates of organics and Chl-a and potential risks of overpopulation of the introduced organisms are obvious disadvantages. The addition of algae removers, such as magnesium-based oxygen (Huang et al. ) and copper sulfate (Tsai et al. ), into the water body is the common chemical technique. These removers act on the algal cell wall with sulfur-containing groups to destroy algae growth.
These removers have prominent removal effects on Chl-a but exhibit poor removal of nutrients and cause secondary pollution. The target compounds for removal by the above techniques for eutrophic lake water treatment are either nutrients, such as N and P, or Chl-a, and could not eliminate the harm of eutrophic water. Therefore, to realize the complete treatment of eutrophic lake water, it is necessary to identify and develop a more efficient process.
In iron-carbon micro-electrolysis (ICME), also known as internal electrolysis, microprimary cells are formed by iron wastewater (Lin et al. ), due to its advantages of low complexity, ease of operation, low raw material cost, short processing time, and low secondary pollution from the reaction products, among others. Furthermore, the Fe in iron filings can react with dissolved phosphate to form iron phosphate precipitates, such as Fe 3 (PO 4 ) 2 and FePO 4 , indicating that ICME should be a suitable technology for eutrophic water treatment. The main advantage of ICME is to convert refractory organic matter into easy degradable organic matter, but it does not provide the extensive and complete removal of organic matter. In contrast, ozone oxidation is more effective for deep removal of organic matter, but it typically exhibits low ozone utilization (Lu et al. ). To overcome this drawback, ozone oxidation can be used in combination with a catalytic material to generate a strong oxidizing hydroxyl radical to enhance the method's oxidation efficiency, a process known as catalytic ozonation (CO) (Wang & Chen ). Therefore, following pretreatment with ICME the addition of CO can theoretically achieve complete removal of multiple contaminants in eutrophic water.
However, to date, little research on eutrophic water treatment via the coupling of ICME with CO has been reported.
Hence, investigation of the optimal parameters for this novel combination process for the complete treatment of pollutants in eutrophic water is warranted.
In this study, a novel process of ICME coupled with CO (ICME-CO) was constructed for treatment of eutrophic lake water, ICME and CO were used as the pretreatment and advanced treatment of ICME-CO, respectively. A series of batch experiments with ICME alone and CO alone was designed to investigate the effects of process parameters, such as initial pH, dose of Fe-C, time of micro-electrolysis, ozone flux, dose of TiO 2 /AC, and time of ozonation, on the removal rates of TN, TP, COD Mn and Chl-a, respectively. Moreover, the above process parameters were further optimized using response surface methodology (RSM) constructed by central composite design. Moreover, the hybrid ICME-CO process was further used in actual eutrophic lake water treatment based on the optimized process parameters.
The present research provides a new alternative technology with promise for the treatment of eutrophic lake water.

Materials
In this study, simulated and actual eutrophic lake water were prepared and collected, respectively. The simulated eutrophic lake water was taken from Ruoyu Lake in Changzhou University, and ammonium chloride (National Pharmaceutical Group, China), potassium phosphate (National Pharmaceutical, China), and self-cultivated algae were added into it to control the concentrations of TN, TP, COD Mn and Chl-a within a certain range, which were used for single-factor and response surface analysis with the purpose of determining the optimal process parameters. The actual eutrophic lake water was taken from

Experimental design
Experimental device of ICME-CO Batch experiments of ICME alone and CO alone were used for eutrophic lake water treatment. The hybrid ICME-CO process used ICME as a pretreatment process and CO as an advanced treatment process. Figure 1 shows the experimental diagram set-up.
For the ICME process alone (Figure 1(a)), 1,000 mL of simulated eutrophic lake water was accurately measured in a large beaker, and the pH adjusted to the required pH value. The appropriate amounts of iron filings and granular activated carbon (volume ratio is 1:1) were added to the water and the mixture was stirred for an extended period of time. After the required reaction time for microelectrolysis, the water sample was left to undergo static settlement, and the TN, TP, COD Mn and Chl-a contents in the beaker were determined. The removal rate of each component was calculated using Equation (1): where C t was the pollutant concentration after the reaction, mg/L; and C 0 was the initial concentration of pollutant, mg/ L. For the individual CO process (Figure 1(b)), the target flux ozone (30 mg O 3 /L) prepared by an ozone generator was added into the simulated eutrophic lake water by an aerator. Furthermore, a certain amount of TiO 2 /AC composite catalyst was also added into the water. After a given ozonation reaction time, the TiO 2 /AC in the water settled out and TN, TP, COD Mn and Chl-a contents were determined as in the above method. Conversely, the CO inflow was the ICME effluent for the treatment of actual eutrophic lake water by the hybrid ICME-CO process, and the other conditions were the same.

Single-factor experiment
To optimize the reaction conditions of the hybrid ICME-CO process for the treatment of eutrophic lake water, singlefactor experiments with ICME alone and CO alone were designed to investigate their effects on the removal rates of TN, TP, COD Mn and Chl-a in eutrophic lake water. For the ICME process, the initial pH, dose of Fe-C and time of micro-electrolysis were the main factors; for CO, ozone flux, dose of TiO 2 /AC and time of ozonation were investigated. The detailed experimental conditions for the treatment of simulated eutrophic lake water by the above two processes are listed in Table 2.

Experimental design using RSM
RSM was used to optimize the parameters of the ICME and CO processes and to determine the final optimal conditions for the hybrid ICME-CO process. Based on the single-factor experiments with ICME and CO, a set of 3 × 3 response surface analysis experiments was designed using Design-Expert 11.0 software. Details of the test factors and level design of the individual ICME and CO processes are presented in Table 3. Furthermore, according to the above scheme, 34 combinations of initial conditions were used for the individual ICME and CO processes. Additionally, analysis of variance (ANOVA) was used to analyze the suitability of the regression models at the 95% confidence level. The P-value and F-value were used to assess the significance of the variables, and a model with a P-value less than 0.05 and a large F-value was considered significant (Nayak & Vyas ).

Analysis methods
The TN, TP, COD Mn and Chl-a concentrations of all water samples in this study were determined using a UV-visible

RESULTS AND DISCUSSION
Single-factor analysis However, the removal of TP in this type of system depends mainly on the adsorption of granular Fe/C.  increased amounts of the catalyst is associated with increased processing cost. Thus, it was necessary to identify a reasonable amount of catalyst. The effects of time of ozonation were similar to those of ozone flux, and the removal rate of each index was maximized at a treatment time of 90 min (Figure 2(f)).
However, when the total amount of pollutants was held constant, extending the reaction time beyond this point did not continuously enhance the reaction effect.

CONFLICT OF INTEREST
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Hankun Zhang: Funding acquisition.

CREDIT AUTHORSHIP CONTRIBUTION STATEMENT
Qiuya Zhang: Resources.

DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
First received 14 September 2020; accepted in revised form 4 March 2021. Available online 17 March 2021