Nitrate removal from synthetic and real groundwater by electrocoagulation: effect of operating parameters and electrolytes

ABSTRACT The present study elaborates the electrocoagulation process for the removal of nitrates from synthetic and real groundwater. Electrocoagulation treatment has a positive effect on the quality of groundwater in general and the removal of nitrate in particular, which is one of the most potent pollutants found in groundwater. Moreover, it is supposed that nitrate removal mechanisms are likely to be dependent on operational parameters like applied current, electrode material, initial concentration of nitrates, and pH of the solution, etc. Therefore, in the present study various process parameters have been optimised in terms of nitrate removal efficiency. Experiments were performed using batch process at different initial concentration of nitrates (100–500 mg/L), pH (6.0–12), stirring speed (100–500 rpm), inter-electrode distance (0.5–2 cm) and electrolysis time (30–180 min). Further, the effect of co-existing ions using KNO3, Ca(NO3)2 and Mg(NO3)2 in the presence of NaCl and Na2SO4 was investigated. Maximum 98%, Nitrate removal efficiency was obtained at inter-electrode distance 1 cm, agitation speed 300 rpm, electrolyte concentration of 1.1688 (g/L) NaCl, current 1.5 A and time 180 min for initial nitrate concentration of 100 (mg/L). Isotherm and kinetics models have been studied and it was observed from the present investigation that the process follows pseudo-second-order kinetics. The Freundlich isotherm model simulations match satisfactorily with the experimental observations. Further, optimised parameters were used to remove nitrates from real groundwater and 92.5% removal efficiency was attained.


Introduction
In recent years, extensive data collected by many researchers divulge that contamination of water resources due to nitrates is a worldwide environmental issue.In many regions of India, including Delhi, according to Central Ground Water Board of India [1], 100-800 (mg/ L) nitrates are present in groundwater.Nitrate is a colourless, odourless, and tasteless anion that has a fatal effect on different creatures when found at high concentrations in water.The presence of even less than 10 (mg/L) nitrates in water may cause methemoglobinemia (blue baby syndrome) which can be fatal for neonates [2,3].Methaemoglobin in the red blood cells is formed due to the reduction of nitrate to nitrite in the stomach of infants, and the nitrite formed binds to haemoglobin.Due to the presence of methaemoglobin, the oxygen-carrying capacity of blood reduces, which turns the body of infant's blue, hence the name blue baby syndrome.Nitrates may also be responsible for high blood pressure, thyroid malfunctioning, cancer-causing and mutagenic health problem.Current pieces of evidence advocate that exposure to nitrate in water used for drinking purposes can modify human thyroid gland function by hindering iodide uptake, leading to alteration in thyroid hormone functions [4].Moreover, nitrates can lead to eutrophication in water bodies, and strongly influence aquatic life [2,5].
Based on the severe impacts of nitrate on human health and the environment, drinking water regulations are becoming increasingly stringent worldwide.World Health Organization and Bureau of Indian Standard [4,6] have set 50 and 45 (mg/L) as an acceptable limit of nitrates in drinking water, respectively, which indicates that nitrate is a matter of concern that should be treated well.
Nitrates removal from water is an emergent research area due to incremental usage of groundwater for drinking and agricultural purposes.Properties like high solubility and adsorption make nitrate removal a difficult task.Nitrate ions, being very stable and extremely soluble have less ability for precipitation.Therefore, nitrate removal using lime softening, oxidation and coagulation is difficult [7,8].Biological treatment methods are described to be efficient in the literature for the removal of nitrates from groundwater and surface water.Reduction of nitrates into nitrogen gas using biological denitrification is considered an eco-friendly and economically feasible method [9][10][11].However, prolonged reaction time and high-temperature sensitivity are the significant limitations of biological treatment.Other alternative advanced techniques proposed by researchers for removal of nitrates are ion exchange, reverse osmosis and electrodialysis [12][13][14][15].Ion-exchange methods are feasible, however, limitations due to the existence of competing anions and concentrated brine generation inhibit its applicability at a commercial scale.Membrane fouling and operating cost are the major drawbacks of reverse osmosis and electrodialysis processes [16].Another primary concern associated with the aforementioned physicochemical processes is the significant amount of brine generation, which needs to be circumvented by alternative reduction processes.
Among the numerous techniques, an extensive literature survey indicated varied applicability and advantages of the electrocoagulation (EC) process for nitrate removal from water.Recently, EC has emerged as a superior treatment technique for nitrate removal because of safe operation, simple process conditions, flexibility and costeffectiveness [17].Many authors have also reported nitrates removal using electrocoagulation [18].Investigated electroreduction and electrocoagulation for the removal of nitrate from water.Authors reported that a pH of 5.0-7.0 and energy consumption of 1 × 10 −3 kWh/g was required to reach acceptable nitrates concentration in electroreduction.In electrocoagulation, the same level was achieved in the pH of 9.0-11.0with energy consumption of 0.5 × 10 −4 (kWh/g).Emamjomeh et al. [19] studied nitrate removal efficiency (NRE) by varying electrolysis time, pH, initial concentration, and current rate.Ninety-three percentage nitrate removal was obtained at 100 (mg NO 3 − /L) at 40 min and pH of 9-11.Also they identified a linear relationship between the electrolysis time and the initial nitrate concentration.They suggested that denitrification using EC process could be used as an effective primary process for the water treatment facilities.Dash and Chaudhari [20] studied electrochemical denitrification of groundwater using aluminium, graphite, iron and titanium as electrode materials.They reported that the removal was only 8% for the graphite electrode while aluminium, iron and titanium electrodes showed a good reduction of 70-97%.Lakshmi et al. [21] employed aluminium alloys as the anode material, and they resulted in high removal efficiency (>90%).Also of few applications of EC for removing nitrates from different wastewater by various researchers has been summarised in Table 1.
It can be noticed from Table 1 that high nitrates removal was obtained using EC however, most of the work has been done on synthetic solutions only.
Therefore, the present study focused on treatment of synthetic as well as real groundwater.The specific objectives of this study were to explore (1) the treatment of water contaminated by nitrate ions by the application of the EC process (2) evaluate the performance of EC in presence of other physicochemical parameters (presence of other ions) (3) to evaluate the performance of EC system on real groundwater samples collected from wells in South-West, Delhi, India.
Brief description of EC process is discussed in the subsequent section.

Electrocoagulation
The EC process is based on the application of an electric current (or potential) between two electrodes, and in-situ generation of coagulant.The process comprises of -(i) oxidation of electrode (sacrificial) to form coagulants (ii) destabilisation of contaminants (iii) flocs formation due to aggregation of destabilised phases [37].Aluminium and iron electrodes are commonly used electrodes for the generation of coagulating species.The reactions which occur during EC using aluminium electrodes are as follows: Oxidation and reduction reactions The overall reaction during electrolysis: EC is a competent technique in nowadays for removal of variety of pollutants due to adsorption on hydroxide is 100 times more as compare to chemical coagulation [38].Also, relatively large flocs are formed containing less bound water and having more stability.They can easily separated by filtration.EC process requires simple equipment and can be designed for different capacities of the water treatment plant.The possibility of the generation of secondary pollutants is also very less as there is no requirement for chemical addition.

Synthetic nitrate solution preparation
Potassium nitrate (KNO 3 ) was used in the present study for the preparation of synthetic nitrate solutions.Synthetic solutions of different initial concentrations ranging from 100 to 500 (mg/L) were prepared by dissolving the required amount of KNO 3 in ultra-pure water to avoid any impurities for nitrate detection.Two different quality aluminium strips of size 15 cm × 2.5 cm × 0.3 cm were purchased from local vendors.The strips (Henceforth named Grade 1 and Grade 2) were characterised for purity (quality) which was determined using energy-dispersive X-ray (EDX) analysis (FEI Company, Netherland).A laboratory digital DC power supply (Keithley, China) with a voltage range from 0 to 30 V and current ranged from 0 to 3 A, was used.Magnetic stirrer stirring speed from 0 to 1200 rpm (Labman, Shahdara, New Delhi, India) was used for maintaining of uniform concentration in the reactor.The batch EC experiments were done by taking 400 ml of synthetic groundwater sample in a glass beaker of 500 ml capacity as shown in Figure 1(a).Aluminium sheets with submerged area 28.75 cm 2 (actual size 15 cm × 2.5 cm × 0.3 cm) were used as electrodes, which were rinsed with 0.01 N HCl solution to remove any depositions.The initial and final weight of electrodes was noted for each run.
During EC, at regular time intervals, samples were taken out, centrifuged using a centrifuge (Remi, R-4 C, India) and filtered through 0.45 µm filters (Whatman 45 µm, China).After completion of each experiment, the sludge was separated, and the supernatant was used to analyse nitrates content.Lastly, nitrate removal efficiency was evaluated using equation (2).
Where Y o and Y f = Initial and final nitrate concentration at time 't' in mg/L.

Parametric studies
Parametric studies for nitrate removal using EC were performed because the effectiveness of EC process depends on many operational parameters, such as conductivity, electrode, pH, current density, electrode spacing, agitation speed, duration and concentration of pollutant.In this study, various influencing parameters effect has been optimised in terms of nitrate removal efficiency.Experiments were performed using batch EC at different initial concentration of NO 3 − (100-500 mg/L), pH (6.0-12.0),stirring speed (100-500 rpm), inter-electrode distance (0.5-2.0 cm) and electrolysis time (30-180 min).Further, the effect of co-existing ions using KNO 3 , Ca(NO 3 ) 2 and Mg(NO 3 ) 2 in the presence of NaCl and Na 2 SO 4 was investigated.Reproducibility of the experimental results was found in the range of ± 2%.

Adsorption capacity determination
Nitrate adsorbed capacity was calculated using following eq (3): Where, q e is the amount of nitrate adsorbed at equilibrium (mg/g) C o is the initial concentration of nitrate (mg/L) C e is the equilibrium concentration of nitrate (mg/L) V is Volume of the synthetic solution (L) m is the amount of electrode dissolved in the solution (g)

Equilibrium and kinetic study
Equilibrium studies are vital for the design and scaling up of the treatment process.This study will help to evaluate the theoretical adsorption mechanism and diffusion of adsorbates.In the present study, Langmuir [39] and Freundlich [40] were applied to understand the interaction between adsorbate and adsorbent.The Langmuir and Freundlich isotherms are expressed by linear equations (4a) and (4b), respectively, and separation factor was calculated by equation (4c) Where C e is the equilibrium adsorbate concentration (mg/L), q e is the amount of nitrate adsorbed at equilibrium (mg/g) q m is the amount of monolayer adsorption capacity (mg/g) b is Langmuir constant related to the free energy of adsorption (L/mg).
K f is the Freundlich constant ((mg/g)/(mg/L) 1/n ) n is favourability of the sorption process and n > 1 represent favourable adsorption condition.
Further, kinetic study performed using pseudo-first and pseudo-second-order kinetic models to check rate of adsorption and adsorption principles between adsorbent and adsorbed.The generalised form of Pseudo-first and second order is expressed by equations (4d) and (4e) respectively.Where q t is the amounts of nitrate adsorbed (mg/g) at time t (min) k 1 is the pseudo-first-order rate constant (L/min).k 2 is the pseudo-second-order rate constant (g/mg min)

Effect of electrodes quality
In an EC process, the selection of electrode material is one of the significant parameters.The electrode material should not be toxic to human beings and the environment.As reported by various authors, the advantages like low cost, readily available and non-toxic makes aluminium the most frequently used electrode material.It is also being reported as very effective in the EC process [41].
In EC process, the main pathway for pollutant removal is adsorption on metal oxide (aluminium hydroxide).Formation of Al(OH) 3 depends on electrode quality as well as on current supplied.As the quality of electrodes is crucial for electrocoagulation, it was checked initially.Therefore, to study the effect of electrode quality on NRE using aluminium electrodes (henceforth called grade 1 and grade 2) experiments were conducted at 1.1688 (g/L) of NaCl concentration, 7 pH, voltage 10 V and electrolysis time of 20 to 180 min.From the results, shown in Figure 1(b), it was observed that NRE increased with time for both the electrodes.The higher nitrate removal efficiency of 95% using grade 2 as compared to grade 1 (34%) was achieved and no significant removal was observed above 180 min.
Further, grade 1 and grade 2 were characterised using SEM and EDX, and the results are given in supplementary Fig. S1 and Fig. S2.It can be seen from Fig. S1 and Fig. S2. that 69.49 and 84.77% of aluminium content was present in grade 1 and grade 2 electrodes, respectively.This result attributed to higher purity, higher aluminium content and smooth surface present in grade 2 as compared to grade 1 electrodes.It was substantiated that grade 2 generated more flocs for adsorption of nitrates as compared to grade 1 due to more aluminium content [42].Therefore, further EC experiments were performed using grade 2 aluminium electrodes.

Effect of initial pH
pH of the solution severely affects the dissolution of electrodes and speciation of coagulants in EC processes.pH plays a crucial role during precipitation and coagulation of the pollutant as reported by various researchers [43].Therefore, in the present study, the effect of initial pH (6)(7)(8)(9)(10)(11)(12) on NRE % and final pH values with time (20-180 min) were examined.The pH adjustment was done by adding 0.01 M NaOH or 0.01 M HCl. Figure 2 It can be seen from Figure 2(a), for initial pH 6, the increase in pH during EC is quite slow (~80 min to reach pH 9).In contrast, for initial pH (7)(8)(9), pH increased rapidly to reach a value of 9.2 within 40 min and a final value of 9.8-10 in ~100 min.The increase in pH is due to increased hydroxide ion (OH − ) concentration, which is generated by water reduction at the cathode during EC process (eq 1b).For initial pH 10, not much change was noticed.At initial pH 11 and 12, the final pH value decreased to 9.8-10.0with time (~100 min.).This pH value (9.8-10.0)corresponds to the value of buffering pH of Al(OH) 3/ Al(OH) 4 -mixture.Under severe alkaline conditions, the formation of Al(OH) 4 − complex is the primary reason for a decrease in pH [44].Al(OH) 3/ Al(OH) 4 − mixture buffering effect makes the final pH more or less around 10.0, whatever the initial conditions.The relative stability of pH after 100 min., could be due to insoluble Al(OH) 3 flocs formation at buffering pH.From Figure 2(b), it can be observed that the nitrate removal efficiency increases with an increase in time and with initial pH values 6-12.According to Mollah et al. [37], the stability of the generate hydroxide species is influenced by the initial pH of the solution.In this study, it was found that low NRE was obtained at pH 6 as compared to pH values 7-12 because cationic monomeric species Al 3+  .
However, these species often are ignored, especially in dilute solutions due to no effect on the overall speciation.A similar observation was reported by [45,46].Therefore, in the present study, only mononuclear and polymeric hydrolysed species adequately predict Al(OH) 3 precipitation leading to higher nitrate removal efficiency.Also, above the neutral pH, produced aluminium flocs have lower solubility.The nitrate removal efficiency was not much affected by increasing the initial pH above 7.
Literature indicated that a pH range of 7-11 is suitable for nitrate removal using EC and the findings from Figure 2(a) and Figure 2(b) shows that the results are well agreement with the literature [47,48].Therefore, pH 7 was taken as optimum for all further experiments.These results are favourable for the industrial development of the technique as, except for very high or very low pH value, there will be no requirement to adjust the initial pH to attain maximum nitrate removal for groundwater.

Effect of current
Current is an imperative parameter in the EC process because it affects the dosage rate of coagulant, rate of bubble production, size and growth of the flocs which further affect the removal efficiency and the operating cost of the EC process.The present study elaborates the effect of current on removal efficiency when it is varied from 0.5 to 2.0 A. The results are shown in Figure 2(c).Figure 2(c) reveals that NRE increased with increasing current from 0.5 A to 1.5 A, which is consistent with Faraday's law.According to the faraday law, the higher dissolution of the anode generates more Al 3+ ions, leading to an increase in the formation of Al(OH) 3 flocs.Also, the bubble generation rate increases and bubble size decreases with an increase in the current, leading to an increase in removal efficiency by flotation [49].Further, an increase in current from 1.5 A to 2.0 A did not lead to an increase in removal efficiency.This is because rising gas bubbles turbulence enhances the mass transfer rate at the electrodes resulting in a decrease in concentration polarisation, which slows down the rate of cathodic and anodic reactions and may lead to anode passivation especially at high currents.Other investigators observed a similar effect of current density on removal efficiency [50].So, 1.5 A was selected as the optimum current for further investigations.

Effect of inter-electrode distance
In the design of the electrochemical cell, inter-electrode distance is one of the vital parameters.When an operational cost optimisation is needed, the inter-electrode distance and effective surface area of electrodes are significant variables [51].In our study, the effect of inter-electrode distance was investigated between 0.5 and 2 cm.From the results shown in Figure 2(d), it can be inferred that with an increase in interelectrode distance from 0.5 cm to 1 cm, the NRE increased from 80 to 98.5%.At an interelectrode distance of 0.5 cm, the low removal efficiency was observed because of degradation of generated aluminium hydroxides flocs by collision with each other and due to high electrostatic attraction inadequate mass transport was there for the EC process [52].However, at 1 cm inter-electrode distance, NRE increased.The slower movement of the generated ions due to a decrease in the electrostatic effect provides sufficient time for the generated aluminium hydroxide to form sufficient flocs, increasing the removal efficiency.Further, the NRE reduced from 98.5% to 81% at 180 min, on increasing electrode distance from 1 to 2 cm.This reduction in efficiency was observed because the travel time of ions increased with an increase in the distance, leading to an insufficient electrostatic attraction resulting in less formation of flocs needed to coagulate pollutants.Results of this study agreed with the literature [53,54].Hence, it can be concluded that a shorter gap favours minimisation of the potential drop, which lead to a higher current density and subsequently result in increased NRE.In this study, all further experiments were carried at 1 cm inter-electrode distance as it was found to be the optimum distance between the electrodes for high removal efficiency.

Effect of stirring speed
In the EC process, uniform conditions are maintained by stirring.Also, by stirring the solution, velocity is imparted for the movement of the generated ions and evades the concentration gradient formation in the electrolysis cell.Hence, the stirring speed must be optimised.For the present study, experiments were performed to investigate the effect of stirring speed on nitrate removal efficiency.Initially, almost no separation was observed when the experiment was performed without agitation.The stirring speed was varied from 100 to 500 rpm for further experiments and the results are depicted in Figure 2(e).
From the results, as given in Figure 2(e), an increase in NRE was observed (50.6 to 95.17%) when agitation speed increased from 100 to 300 rpm.With an increase in agitation speed, the generated ions movement increases, leading to higher interaction among the ions.A higher interaction among ions results in the formation of flocs needed to coagulate the pollutant [55].Moreover, precipitation becomes easier as lighter particles float on the surface of the solution with an increase in agitation speed.However, a further increase in the agitation speed from 300 to 500 rpm, leads to a decrease in the NRE because the flocs get degraded and adsorbed nitrates get desorbed into the solution [56].Further, higher agitation speed consumes more energy resulting in a higher cost of operation.Therefore, in the present study, a moderate agitation speed of 300 rpm, which had a considerably positive effect on the removal efficiency, was taken as the optimum value and all further experiments were performed at 300 rpm.

Effect of electrolysis time
The rate of dissolution of Al 3+ ions is determined by electrolysis time along with the current applied.To study the effect of electrolysis time on the nitrate removal efficiency, the experiments were conducted for different electrolysis times (30-180 min) for 100 (mg/ L) nitrate concentration and the results are depicted in Figure 2(f).It was found that the nitrate removal efficiency increased significantly (30-98%) with an increase in the electrolysis time from 30 min to 180 min.The removal efficiency depends on the concentration of metal ions and electrolysis time.The production rate of Al 3+ ions from the aluminium electrodes can be increased by increasing electrolysis time.The results obtained are in line with those reported in the literature [56].180 min was found to be the optimum electrolysis time at which 98% nitrate removal efficiency was obtained.

Effect of initial concentration
To evaluate the initial concentration effect on the performance of EC for nitrate removal, a wide range of initial concentrations (100 to 500 mg/L) were prepared.The operating conditions were (pH: 7.0, electrode spacing: 1 cm, NaCl conc.: 1.1688 (g/L), rpm: 300, Time: 180 min).The results, as depicted in Figure 2(g) show that with an increase in initial nitrate concentration (100 to 500 mg/L), NRE decreased (95 to 68%).
The predominant pathway is the adsorption on the freshly produced metallic hydroxide flocs for the nitrate removal by EC, given in eq (5a-5d) [57,58].

Overall reaction is:
Al 3+ would also be consumed in water to form Al(OH) 3 precipitate according to equation 1c.
According to Faraday's law, constant amounts of coagulating ions are produced from the sacrificial anode for the same current density and electrolysis time.
From Figure 2(g), it is evident that at the beginning of the EC process there is a sharp rate of removal, and afterwards (after ~60 min), the slope of the curve decreases.This is probably because of an increase in activation polarisation via adsorption on the electrodes at the beginning with a consequent decrease in the rate of aluminium dissolution at the anode.It was found that for all the experiments performed in the studied concentration range (100-500 mg/L), the amount of sludge produced was nearly the same.As per faraday law, at fixed current density and electrolysis time.a constant amount of flocs of aluminium hydroxide (coagulants) were produced in the nitrate solution.
The flocs produced are sufficient to adsorb nitrates, bring about a higher NRE, at low concentrations.However, with an increase in initial nitrate concentration (100-500 mg/L), the decrease in removal efficiency can be ascribed to the fact that, increased NO 3 − ion concentration would block the adsorption sites of Al(OH) 3 rapidly and decreases its ability to adsorb more NO 3 − ion.Also, at higher initial concentration, the number of flocs produced was insufficient to adsorb all the nitrate ions, causing a reduction in the nitrate removal efficiency.Similar results were obtained by [59].

Effect of co-existing cations on nitrate removal
Normally, groundwater contains high concentrations of co-existing ions.The presence of these ions, their type and concentration, play significant roles in denitrification using EC process.Anions and cations might influence the adsorption of nitrates negatively or positively.Therefore, it is essential to compute the effects of the coexisting ions on the nitrate removal in EC process.In this work, efforts have been made to elucidate the effects of various cations, generally present in groundwater, on denitrification using EC process.The three most common cations, namely potassium, calcium and magnesium present in groundwater, were tested to discern their effects on nitrates removal efficiency.
The experiments were performed using salts of potassium, calcium and magnesium at initial NO 3 − concentration: 100 (mg/L), pH: 7, for 180 min.The results are shown in Figure 2(h).
It was observed from Figure 2(h) that nitrate removal efficiency for all the three ions increased with time, the highest being in the case of potassium ions.A minor reduction was observed in the case of calcium and magnesium ions.The possible reason may be the interaction of these ions with metal hydroxides surfaces and their competition with nitrates for adsorption.Similar results were reported by [60] in simultaneous electrooxidation and electro-coagulation processes for the removal of arsenite.

Electrode passivation
One of the major limitations of EC technologies is the high electrical energy consumption as compared to conventional methods [61,62].Further, passive oxide film formation on the anode surface is a major problem in EC [63].Therefore, more energy is consumed due oxide layer formation.To decrease energy consumption, the passivation effect has to be reduced on electrodes.Some researchers have suggested the following remedies that can reduce the passivation on electrodes (i) NaCl as electrolyte [64], (ii) Polarity change of electrodes [65], (iii) configuration of electrode connection [64], (iv) Use of alternating current [66], and (v) Agitation using ultrasonication [67].Therefore, in the present study, efforts have been made to reduce the passivation effect by performing the experiments on different concentrations of NaCl and Na 2 SO 4 , and change of polarity from 20 to 180 min.

Effect of NaCl and Na 2 SO 4
An essential parameter in the EC process is the conductivity of the solution.The removal efficiency of the pollutant and operating cost are directly related to solution conductivity as suggested by many researchers [68,69].Also, Zhang et al. [70] reported that, during electrolysis, the resistance between anode and cathode is directly influenced by the initial conductivity in solution, and indirectly changes a photovoltaic output current.Hence, the flow of electric current in the process of EC, water must have some minimum conductivity.Therefore, the conductivity of low conductivity water must be adjusted by adding salts such as sodium chloride or sodium sulphate in sufficient amounts.
In the present study, NaCl and Na 2 SO 4 were used as supportive electrolytes due to ease of availability and low toxicity.Experiments were conducted at different concentrations of NaCl and Na 2 SO 4 and results were investigated.Effects of NaCl, Na 2 SO 4 and a combination of NaCl and Na 2 SO 4 on NRE are presented in Figure 3(a).
As shown in Figure 3(a), NRE increased (77.86% to 95.23%) with an increase in NaCl concentration from 0.5844 to 1.1688 (g/L).On further increasing NaCl concentration to 1.7532 (g/L), not much enhancement in NRE (95.5%) was observed.Results infer that the addition of NaCl improves the EC performance but diminishes beyond a certain NaCl concentration in the solution.The possible reason is that Cl − ions from NaCl can remove the passivation layer on aluminium electrodes and leads to more production of aluminium hydroxide due to enhanced anodic dissolution.Also, during EC in the presence of Cl − ions indirect electrochemical oxidation occurs as represented by eq.6(a-c).It can be seen from eq. 6 (a-c) that at the anode, the Cl − ion is discharged to generate Cl 2 , which dissolves in the solution immediately.Also, chloride ion scavenges oxygen to form OCl − ions at the electrode, which can oxidise the pollutants effectively [71].
Conversely, NRE decreased with increasing Na 2 SO 4 concentration from 1.42 to 4.32 (g/ L) (66.65% to 18.59%).SO 4 −2 ions in supporting electrolytes have the potential to form or strengthen a shielding layer on the metal electrode surface, which hinders the localised corrosion of aluminium electrodes, leading to lower performance of EC process because of lower current efficiency.Further, a combination of NaCl and Na 2 SO 4 was used to see the combined effect of the electrolyte by taking 0.5844 (g/L) NaCl and Na 2 SO 4 concentrations varied from 1.42 to 4.26 (g/L).It was found that NRE increased when NaCl was combined with Na 2 SO 4 concentration.It can be attributed to chloride ions, that detrimental effect of sulphates on EC efficiency was thwarted by Cl − ions from NaCl, which are more efficient and prevent the inhibition caused by sulphate ions [72].It was concluded that the highest removal efficiency (~95%) was obtained by adding NaCl (0.02) as compared to Na 2 SO 4 and any combination of NaCl and Na 2 SO 4 .So, for further experiments, 0.02 M NaCl was used as the optimum dosage.Similar results were reported by Singh and Ramesh [73], for removal of RB25 by EC and the dye removal were 97.7% and 70.0% for NaCl and Na 2 SO 4 respectively.Further, the Effect of pH change during EC process was checked and results are shown in Figure 3(b) for Na 2 SO 4 and NaCl, respectively.It can be observed from Figure 3(b) that pH plays a major role in the presence of Na 2 SO 4 and NaCl electrolyte.In Figure 3(a) for Na 2 SO 4 , pH of the solution stabilised at pH 11, which facilitates the dissolution of Al(OH) 3 into anionic hydroxy aluminium species such as Al(OH) 4 and Al(OH) 5 2 and the formation of these species make flocculation and electron precipitation difficult to occur.While in the case of NaCl, pH remains in the range between 9.8 and 10 throughout the experiment.In this pH range, amorphous Al(OH) 3 with large surface area is precipitated, which is beneficial for coagulation by adsorption.

Effect of polarity change
Many authors stated that the current density affects excessive oxygen evolution as well as heat generation [43,60].Additionally, constant current density can also help in reducing the passivation effect throughout the experiment.Constant current density can be maintained by the polarity reversal as reported by [74,75].In the present study effect of polarity change has been investigated by reversing the anode and cathode after every 20 min and the results are shown in Figure 3(c).It can be observed from Figure 3(c) that nitrate removal was much higher after polarity change as compare to the no polarity change experiment.In case of no polarity change and polarity change experiment, 98% nitrate removal obtained in 180 min and 120 min, respectively.These results indicated that the polarity change can improve the removal and avoid the occurrence of an oxide passive layer on electrodes.These results are very promising and show that the passivation can be avoided by polarity change with a high performance of EC.

Equilibrium study
Equilibrium studies were performed using eq.4(a-c) and the results are given in Table 2.The results suggested that Freundlich model was the best fit for the present system (R 2 > 0.999 and K F = 16.8) and n > 1 indicated a favourable adsorption process.These results confirm that the adsorption takes place via chemical interactions [76].

Kinetic study
In kinetic study, experimental values were fitted in eq.4(d-e) and the results are shown in Table 2. From Table 2, rate constant K 1 0.07 (min −1 ) and K 2 0.00005.(g/mg.min)were obtained.Low value of K 2 , q (cal) = 434 (mg/g), and (R 2 0.99), suggested that the mechanism is chemisorption of nitrate molecules on aluminium hydroxide.Therefore, pseudosecond-order is considered the best fit model.These findings are supported by other findings of literature [77,78].

Validity test
The validity of the kinetic models is tested and presented in Table 2. From Table 2, the smallest values were obtained in pseudo-second-order model (chi-square (χ 2 ) = 8.9 × 10 −8 and sum of error square (SES) = 2.06 × 10 −7 ).Hence, the best-fit model for nitrate adsorption in the present study was pseudo-second-order.

Treatment of real groundwater
A few groundwater samples from south-west Delhi were collected in bulk due to ease of sampling.South-west Delhi district is one of the eleven administrative districts of NCT of Delhi, India.It is situated between latitude ranging from 28.40ʹ to 28.29ʹ and longitude The groundwater samples from different sites of south-west Delhi predominantly Dwarka and nearby areas were collected using grab sampling from sources such as borewells, private and government hand pumps in years 2015-18.All sampling sites were selected in a view to cover the entire area of the region selected.The water was flushed for 10 to 15 minutes before sample collection to avoid inactive water and to get a precise scenario of groundwater in the aquifer.
The collected samples were characterised for various parameters, such as conductivity, TDS, hardness, chlorides, sulphates and nitrates, etc., as shown in Table 3.
It was noticed that most of the quality parameters exceeded the BIS [6] limits which result in unsuitable drinking water.Further, preliminary investigations revealed higher nitrate content in these samples (106-231 mg/L) which can result in the eutrophication of aqua systems and also reduce the quality of drinking water.The actual groundwater samples were treated using the EC process at optimised values of spacing between electrodes (1 cm), current density (0.0625 A/cm 2 ) and electrolysis time (approx.180 min) (keeping few parameters like rpm constant at 300).A maximum of 92.5% removal efficiency was obtained for the initial concentration of 106 (mg/L) nitrates in real groundwater.Further economic feasibility of EC process was determined using energy consumption and operating cost at industrial electricity price of $0.08/kWh, current = 1.5 A, electrolysis time = 3 hr, voltage = 10 V, volume = 0.4 L. The operating cost was evaluated using eq.( 7a) and (7b) as reported by [79,80].The energy consumption and operating cost were found to be 0.1125 kWh and 0.009 ($/L), respectively.

Conclusions
In the present study, the experiments were performed to optimise the batch EC process parameters for the removal of nitrates from synthetic and real groundwater samples.The results are very promising for the industrial application of EC.It was observed that
(a) demonstrate the influence of initial pH on final pH values after each experiment and Figure 2(b) demonstrated the effect of initial pH on NRE % with time.

Table 1 .
Application of electrocoagulation for the removal of nitrates.
and Al(OH) 2 + predominate at acidic pH and limited flocs are generated for coagulation.Maximum 95% NRE was obtained above pH 7 because during EC process dimeric, trimeric and polynuclear or polymeric hydrolysis products of Al can also form such as Al 2

Table 2 .
Isotherm and kinetic study.50ʹ and 77.14ʹ.South-west district has a varied character with Kapashera sub-division as predominantly rural, Dwarka sub-division as mostly urban and Najafgarh sub-divisions as a mix of both urban and rural population.

Table 3 .
Characterisation of real groundwater samples., current density, electrolysis time, agitation speed and electrolyte concentration are the dominant factors in determining the efficiency of the EC process.Maximum 98% nitrate removal efficiency was obtained in 180 min.Minor changes were observed in NRE in the presence of different anions like potassium, calcium and magnesium.Further, Freundlich model shows favourable adsorption and pseudo-second-order considered the best fit model for adsorption of nitrate.It can be concluded from the present study that EC process can remove nitrates efficiently, for any initial concentration and reaches below BIS, 2012 guideline value.It is a significant technology that can be used for water treatment and can be employed at larger scales. spacing