Abstract

The objective of this work was to investigate electrochemical removal of nitrate from a high salinity waste stream generated by Donnan dialysis. Donnan dialysis for nitrate removal is a promising technique. It produces a distinctive composition of a high salinity waste stream of NaCl or Na2SO4 that requires a viable disposal method. The waste stream has the full anionic composition of contaminated groundwater, but the only cation is sodium. Experiments were conducted in a batch system setup. A copper cathode was chosen over brass, aluminum and graphite cathodes. A dimensionally stable anode, Ti/PbO2, was selected over a Ti/Pt anode. Electrochemical denitrification of high salinity Donnan dialysis nitrate wastes was successfully achieved, with different behavior exhibited in high salinity NaCl solution than in high salinity Na2SO4 solution. NaCl inhibited nitrate removal at high salinities while Na2SO4 did not. The maximum removals after 4 h operation in the high salinity wastes were 69 and 87% for the NaCl and Na2SO4 solutions respectively.

INTRODUCTION

In recent years the quality of potable water has been negatively impacted by agricultural, commercial and industrial activities and has experienced rising nitrate (NO3) levels. High levels of nitrate in drinking water can cause negative health effects for humans such as methemoglobinemia, a disorder that primarily effects fetuses and young infants under the age of 6 months (Fewtrell 2004). The United States Environmental Protection Agency has set a maximum allowable limit of nitrate in drinking water at 45 mg/L and the World Health Organization as well as the European Union has set the limit at 50 mg/L. Despite this preventative measure, reports of groundwater contaminated with nitrate proliferate worldwide (Mas-Pla & Menció 2019).

Current nitrate removal technologies have significant drawbacks. Physicochemical removal of nitrate, which includes techniques such as ion exchange, reverse osmosis, and electrodialysis, do not destroy the nitrate, but rather transfer it into a waste stream that requires disposal (Duca & Koper 2012). In biological denitrification, bacteria and microorganisms are used to enzymatically convert nitrate into nitrogen gas. However, the process is very sensitive to the operating conditions as salinity, temperature and solution pH influence the ability of bacteria to remove nitrate (Matejua et al. 1992).

A promising method for nitrate removal is electrochemical denitrification, where electrical energy provides the driving force for the reduction of nitrate through a series of electron transfers, as shown in Equations (1)–(4). Electrochemical removal of nitrate has the advantages of elimination of nitrate, no chemical addition, no production of sludge or brine, and simplistic operation (Xu et al. 2018). However, some disadvantages include the relatively slow rate of nitrate reduction and the production of by-products, such as nitrite (NO2) and ammonia (NH3) as shown in the following reactions: 
formula
(1)
 
formula
(2)
 
formula
(3)
 
formula
(4)

A key design parameter in electrochemical denitrification systems is the electrode pair used for the reduction of nitrate. Multiple types of electrodes have been studied, including pure monometallic metals, metal alloys, and surface modified metals, among others (Paidar 1999; Peel et al. 2003; Polatides & Kyriacou 2005; Prasanna & Chaudhari 2005; Dortsiou et al. 2009; Katsounaros et al. 2009; Li et al. 2009a, 2009b; Su et al. 2017; Beltrame et al. 2018; Zhang et al. 2018; Gao et al. 2019; Rao et al. 2019; Ye et al. 2019). When choosing an electrode, other characteristics besides the electrode catalytic activity are considered. These secondary electrode characteristics include electrode stability (corrosion resistance), durability (ability to run for long times, not to be poisoned), toxicity (in the event of leaching), and cost (Garcia-Segura et al. 2018).

Fundamental studies on electrochemical denitrification focus on the removal of nitrate from a solution, with a supporting electrolyte added to enhance the conductivity of the solution (Polatides & Kyriacou 2005; Li et al. 2009a, 2009b). Electrochemical denitrification has also been shown to be successful in nitrate contaminated solutions with different ionic compositions, such as those found in the waste stream of ion exchange regeneration (Paidar 1999; Dortsiou et al. 2009), low-level nuclear waste solutions (Katsounaros et al. 2009) and groundwater (Peel et al. 2003; Prasanna & Chaudhari 2005).

A solution composition not investigated for chemical denitrification is that generated by a Donnan dialysis nitrate removal process. Donnan dialysis is a physicochemical technique that employs concentration gradient ratios to induce separation of an ionic species. In Donnan dialysis for nitrate removal, the feed compartment contains the nitrate contaminated water and the receiver compartment contains a high salinity NaCl or Na2SO4 stripping solution. An anion exchange membrane separates the two compartments. The high concentration ratio drives the anions in the stripping solution to the feed compartment. Due to electroneutrality, nitrate is transported to the receiver compartment, thereby generating a high salinity nitrate contaminated waste stream (Hasson et al. 2014). This high salinity waste stream has the same anionic composition as the contaminated water, but the only cation is sodium.

The role of chloride and sulfate on nitrate electro-reduction has been investigated in multiple studies with no consensus reached. While it has been reported that chloride ions interfere with the reduction of nitrate when using a titanium cathode anode pair (Prasanna & Chaudhari 2005), one study concluded that chloride improved the reduction of nitrate with a graphite felt cathode and dimensionally stable anode (DSA) (Jing et al. 2015). Similar disagreements are found with the role of sulfate, with one study showing that it retards nitrate reduction (Birdja et al. 2014) when using a palladium electrode, while another found an improvement in the overall removal of nitrate (Huang et al. 2013) when using a brass cathode and DSA anode. The effect of each anion is dependent on the electrode pair employed in the nitrate electro-reduction.

The objective of this research was to investigate electrochemical removal of nitrate from a high salinity waste stream, generated by a Donnan dialysis process, in a batch system. Initially, a suitable electrode pair was chosen, after which the system was characterized, and the role of temperature was evaluated. Kinetic rate constants were also obtained for a variety of conditions. Subsequently, the effects of high salinity chloride and high salinity sulfate on the nitrate removal were examined in solutions simulating a waste stream of Donnan dialysis.

MATERIAL AND METHODS

Electrochemical system

Two different systems were used in this study. Both systems consisted of undivided electrochemical cells provided with two electrodes immersed in the aqueous solution (plate electrode cell setup). A schematic of the electrochemical system is shown in Figure 1. A DC power supply (0–5 A, 0–30 V) was connected to the electrodes, providing a constant current for the nitrate removal. A magnetic stir bar was placed at the bottom of the cylindrical reactor to ensure uniform mixing. The characteristics of the electrolytic cells are presented in Table 1.

Table 1

Characteristics of the electrolytic cells

Electrolytic cell 
Cathodes Copper, brass, aluminum and graphite Copper 
Anode Ti/PbO2 DSA Ti/PbO2 DSA and Pt/Ti 
Electrode shape Rod Rectangular 
Electrode dimensions ∅︀ = 1.5 cm, L = 22 cm 40 × 5 cm × 1.5 mm 
Inter-electrode distance (cm) 2.0 0.8 
Immersed area (cm294 350 
Solution volume (mL) 400 1,150 
Electrolytic cell 
Cathodes Copper, brass, aluminum and graphite Copper 
Anode Ti/PbO2 DSA Ti/PbO2 DSA and Pt/Ti 
Electrode shape Rod Rectangular 
Electrode dimensions ∅︀ = 1.5 cm, L = 22 cm 40 × 5 cm × 1.5 mm 
Inter-electrode distance (cm) 2.0 0.8 
Immersed area (cm294 350 
Solution volume (mL) 400 1,150 
Figure 1

Electrochemical cell setup.

Figure 1

Electrochemical cell setup.

For cathode selection, a 0.5 L plastic cylindrical reactor was used as the electrolytic cell. Circular rod electrodes of copper, brass, aluminum and graphite were used as cathodes. In all experiments, Ti/PbO2 DSA was used while the cathodes were varied. All experiments were conducted at a current density (j) of 21.8 mA/cm2 for 3.5 h. For all other parameters tested, a rectangular cathode (copper) and anode were placed in a glass cylindrical reactor holding a volume of 1.15 L. All experiments were conducted with a current density of 10 mA/cm2, excluding the experiments for anodic selection, which were conducted with a current density of 7.5 mA/cm2. Each experiment lasted 3 h, unless otherwise stated.

Feed water quality

In the experiments conducted for electrode selection and for testing the effect of temperature on nitrate removal, a 1.5 mM nitrate solution was used, prepared by dissolving analytical grade NaNO3 in deionized (DI) water containing either 10 mM NaCl or 10 mM Na2SO4. Nitrate removal tests, on the solution denoted as SG1, were performed with a composition simulating the nitrate contaminated groundwater, found in Rehovot, Israel (Jensen et al. 2012). Finally, nitrate electro-removal was examined with simulated Donnan waste streams, denoted as SG2. In this solution only the anions of the SG1 solution were present; i.e. the cations calcium and magnesium were absent. As explained in the Introduction section, in Donnan dialysis with an anion exchange membrane, only anions are transported through the membrane to the receiver compartment which contains the actuating high salinity NaCl/Na2SO4 solution. The compositions of SG1 and SG2 solutions are listed in Table 2. These solutions were prepared by dissolving analytical grade NaNO3, MgSO4 · 7H2O, NaHCO3, CaCl2 · 2H2O, NaCl and Na2SO4 in DI water. In order to mimic the Donnan waste stream, additional amounts of NaCl or Na2SO4 were added to the SG2 composition. The sum of the total Cl and SO42− salinity is reported.

Table 2

Feed solutions simulating nitrate contaminated groundwater (SG1) and Donnan waste stream (SG2)

 Solution (mM)
 
Ion SG1 SG2 
NO3 1.5 1.5 
Na+ 4.7 11.8–101.2 
Ca2+ 3.0 0.0 
Mg2+ 1.0 0.0 
Cl 5.5 5.5–55.0 
SO42− 1.0 1.0–50.1 
HCO3 3.8 3.8 
 Solution (mM)
 
Ion SG1 SG2 
NO3 1.5 1.5 
Na+ 4.7 11.8–101.2 
Ca2+ 3.0 0.0 
Mg2+ 1.0 0.0 
Cl 5.5 5.5–55.0 
SO42− 1.0 1.0–50.1 
HCO3 3.8 3.8 

Analytical methods

Nitrate concentrations were determined using adorption at a wavelength of 220 nm (UV-Vis spectrophotomer, Thermo Scientific Evolution 201). Solution turbidity (NTU) was measured by a Hach Turbidimeter 2100P. Magnesium and calcium compounds were quantified using complexometric EDTA (ethylenediaminetetraacetic acid) titration (Standard Methods 2340C and 3500B) (APHA 2005).

RESULTS AND DISCUSSION

Prior to examining the removal of nitrate from the Donnan waste stream, the electrochemical system configuration was characterized. This was done using simple solutions consisting of 1.5 mM nitrate and 10 mM NaCl or 10 mM Na2SO4 in DI water.

Cathode selection

The four cathodes studied were graphite, aluminum, copper and brass. Figure 2 displays the nitrate removal after 3.5 h as a function of cathode material in both 10 mM Na2SO4 and 10 mM NaCl solutions. Copper and brass had much greater nitrate removal compared to graphite and aluminum. Copper exhibited nitrate removal of 88 and 75% in Na2SO4 and NaCl respectively, while brass had a 91% removal in Na2SO4 and 86% removal in NaCl. Aluminum performed intermediately, with 68% removal in Na2SO4 and 59% removal in NaCl. The graphite cathode achieved 32% removal in Na2SO4 and 10% removal in NaCl. Graphite is a non-metal and a poor electron donor (Prasanna & Chaudhari 2005), and therefore has poor capacity for transfer of electrons to reduce nitrate.

Figure 2

Nitrate removal by different cathodes after 3.5 h (anode: DSA; j = 21.8 mA/cm2; [NO3]0 = 1.5 mM; 10 mM NaCl or Na2SO4).

Figure 2

Nitrate removal by different cathodes after 3.5 h (anode: DSA; j = 21.8 mA/cm2; [NO3]0 = 1.5 mM; 10 mM NaCl or Na2SO4).

Copper exhibited very high reduction activity in cyclic voltammetry studies when compared to other pure metals such as nickel, platinum, graphite, gold and silver (Dima et al. 2003). This has been theorized to occur by the unique electro-catalytic abilities of copper to inhibit the hydrogen evolution reaction; the hydrogen reaction competes with the nitrate reduction reaction on the cathodic surface, thereby allowing nitrate to be more easily reduced. This explanation can also be applied to brass, as it is an alloy made of copper and zinc.

When comparing the removal between the two solutions, it is evident that there is more removal in the Na2SO4 solutions than in the NaCl solutions. This can be explained by the evolution of the oxidizing hypochlorite ion (Equations (5)–(7)) in NaCl solutions (Xu et al. 2018). The hypochlorite ion can react with nitrite to reform nitrate (Equation (8)), resulting in less removal. 
formula
(5)
 
formula
(6)
 
formula
(7)
 
formula
(8)
In addition to the nitrate removal efficiency, the specific energy consumption and turbidity were also investigated. It is well established that the specific energy consumption depends on factors such as the composition of the solution, which affects the conductivity and resistance of the cell, as well as the identity and the configuration of the electrodes (Katsounaros et al. 2009; Xu et al. 2018). The specific energy consumption (E, kWh/mmol NO3) is defined as: 
formula
(9)
where Ut is the potential (V), I is the current (A), t is the time (h), is the current efficiency and ΔCNO3 is the change in nitrate concentration (mmol).

Figure 3 shows the difference in the specific energy consumption among the four cathodes after 3.5 h. The current efficiency is assumed to be constant for all the cathodes at the same current density. Graphite, which does not transfer electricity well, had the lowest removal and the highest specific energy consumption (0.62 and 0.13 kWh/mmol NO3 in Na2SO4 and NaCl respectively) out of all the electrodes tested. It was therefore concluded that graphite is not a suitable electrode for nitrate removal. The other three electrodes had similar energy consumptions of about 0.1 kWh/mmol NO3.

Figure 3

Specific energy consumption for different cathodes after 3.5 h (anode: DSA; j = 21.8 mA/cm2; [NO3]0 = 1.5 mM; 10 mM NaCl or Na2SO4).

Figure 3

Specific energy consumption for different cathodes after 3.5 h (anode: DSA; j = 21.8 mA/cm2; [NO3]0 = 1.5 mM; 10 mM NaCl or Na2SO4).

With all electrodes, the specific energy consumption was higher in NaCl solutions than in Na2SO4 solutions. This is due to the higher removal efficiencies in Na2SO4 solutions than in NaCl solutions. Another reason is that Na2SO4 is more conductive than NaCl on a molar basis (Lobo 1989); therefore, equal molar amounts of the same salts result in a lower voltage for the Na2SO4 solutions and hence lower power consumption.

When aluminum was employed as a cathode, high turbidity was obtained (400 and 850 NTU in Na2SO4 and NaCl solution, respectively), due to the dissolution of the aluminum; a phenomenon documented in the literature (Govindan et al. 2015). The dissolution of the cathode in electrochemical denitrification requires periodic cathode replacement as well as a potential post-treatment step for the sludge (Garcia-Segura et al. 2017). With the three other electrodes, turbidity was minimal (<1 NTU).

Based on these results, copper and brass cathodes were the best candidates for this research. However, brass compositions differ from supplier to supplier. The copper in brass can range from 55 to 95%, and can contain trace amounts of other elements in addition to zinc, such as lead, tin and iron. Additionally, brass alloys change their composition over time in electrochemical processes due to leaching, which can significantly impact reduction kinetics (Garcia-Segura et al. 2018). Therefore, this electrochemical denitrification investigation was based on copper cathodes.

Anode selection

The performance of two anodic materials, DSAs Ti/PbO2 and Ti/Pt, was compared for nitrate removal with a copper cathode. Both these DSA anodes have been studied for nitrate electro-reduction (Li et al. 2009b; Malinovic et al. 2015; Rao et al. 2019), yet their performance was not compared. Figure 4 shows the nitrate removal efficiencies for each of the above anodes paired with copper. In the 10 mM Na2SO4 solution, the nitrate removal was 84% with the Ti/PbO2 DSA anode but only 27% with the Ti/Pt DSA anode. Similarly, in 10 mM NaCl the nitrate removal was 66% with the Ti/PbO2 anode in comparison to 25% with the Ti/Pt anode. Since the nitrate removal was greater in all solutions with the Ti/PbO2 anode, it was chosen as the anode for this investigation. Higher nitrate removal efficiencies in a Cu-Ti/PbO2 system than a Cu-Ti/Pt system may be explained by the high oxidation potential of the Ti/Pt anode, which re-oxidizes the nitrite to nitrate when used with copper (Yu & Kupferle 2007; Abeygunawardhana et al. 2015).

Figure 4

Nitrate removal using different anodes after 3 h (cathode: copper; j = 7.5 mA/cm2; [NO3]0 = 1.5 mM; 10 mM NaCl or Na2SO4).

Figure 4

Nitrate removal using different anodes after 3 h (cathode: copper; j = 7.5 mA/cm2; [NO3]0 = 1.5 mM; 10 mM NaCl or Na2SO4).

Electrochemical cell characterization

Characterization of the Cu-DSA system was conducted by determination of the geometrical factor (Rg, 1/cm) and the system resistance (Rs, Ω). This was accomplished by measurements carried out with four different NaCl solutions (0.3–1.4 g/L) in the rectangular electrode batch system. The total voltage of an electrochemical system is expressed as: 
formula
(10)
where VT and V0 are the total and initial intrinisic voltage, respectively, I is the current (A), and RT is the total resistance (Ω).
The intrinsic voltage and the total resistance of a solution are obtained by a plot of the voltage against the current. The total resistance is expressed as: 
formula
(11)
where K is the conductivity of the solution (S/cm).

Therefore, a plot of RT against 1/K yields a straight line with a slope of the geometrical factor and an intercept of the system resistance. The practical importance of the geometrical factor is that it is a design parameter related to the energy consumption of the electrochemical cell (Hasson et al. 2008).

For rectangular electrodes (Hine 1985), the resistance of the solution, RSolution (Ω), is directly proportional to the distance between the electrodes, l (cm), and inversely proportional to the electrode surface area, S (cm2): 
formula
(12)
The RSolution for any cell geometry may also be expressed as: 
formula
(13)

Table 3 summarizes the specific conductance, total resistance, intrinsic voltage, geometrical factor and system resistance obtained for the rectangular electrode Cu-DSA system. The intrinsic voltage of the system is 2.43–2.58 V, the geometrical factor is 0.0034 1/cm, and the system resistance is 0.6119 Ω. Values of the calculated resistance matched well with the theoretical resistance; it was therefore possible to neglect the edge effects of the electrodes (Hasson et al. 2008).

Table 3

Experimental data obtained for the copper cathode batch system

K (μS/cm) Rt (ΩV0 (V) Rg (1/cm) Rs (Ω
647 5.86 2.58 0.0034 0.6119 
1,210 3.49 2.52 
1,940 2.41 2.46 
2,650 1.83 2.43 
K (μS/cm) Rt (ΩV0 (V) Rg (1/cm) Rs (Ω
647 5.86 2.58 0.0034 0.6119 
1,210 3.49 2.52 
1,940 2.41 2.46 
2,650 1.83 2.43 

Effect of temperature

When the electrical energy input is greater than the enthalpy of the reduction reaction, an electrochemical reactor heats up by ohmic heating, and the temperature in the reactor rises (Li et al. 2009c). Results of experiments examining the effect of temperature on nitrate removal are summarized in Figure 5. In experiments conducted without control the reactor temperature reached 38–43 °C. The solution temperature in the controlled reactor was held between 16 and 22 °C. As seen, nitrate was removed more efficiently without the temperature control in both NaCl and Na2SO4 saline solutions. Higher temperatures increase the nitrate diffusion rate from the bulk solution to the electrodes and increase its adsorption strength, resulting in higher removal efficiencies (Huang et al. 2013).

Figure 5

Effect of temperature on nitrate removal in 10 mM (a) NaCl solution and (b) Na2SO4 solution.

Figure 5

Effect of temperature on nitrate removal in 10 mM (a) NaCl solution and (b) Na2SO4 solution.

Kinetics of nitrate electrochemical removal

Kinetic studies of a process provide insight into the rate of the reaction and help ascertain key design parameters in a scaled-up reactor, such as the hydraulic retention time (Prasanna & Chaudhari 2005). The kinetics of the nitrate removal and the effect of temperature on the kinetic parameters were examined by analyzing the experimental data, according to the generalized rate law: 
formula
(14)
where k is the rate constant and n is the reaction order.

Table 4 summarizes the different kinetic expressions for the rate law and its integrated form. The rate constants for zero, first and second reaction order were obtained and the corresponding rate law was plotted with the experimental data for the first 1.5 h. An example of the kinetic analysis in a 10 mM Na2SO4 solution is shown in Figure 6. It is evident that the first order rate law best describes the removal of nitrate, a conclusion reached in other studies as well (Prasanna & Chaudhari 2005; Katsounaros & Kyriacou 2008).

Table 4

Rate laws and their integrated forms for different reaction orders (Levenspiel 1999)

Reaction order (nRate law Integrated rate law Plot 
   
   
   
Reaction order (nRate law Integrated rate law Plot 
   
   
   
Figure 6

Experimental and predicted nitrate removal according to zero, first and second rate laws (j = 10 mA/cm2; [NO3]0 = 1.5 mM; 10 mM Na2SO4).

Figure 6

Experimental and predicted nitrate removal according to zero, first and second rate laws (j = 10 mA/cm2; [NO3]0 = 1.5 mM; 10 mM Na2SO4).

Table 5 shows the effects of temperature on the first order rate constant in 10 mM Na2SO4 and 10 mM NaCl solutions. The rate constant increases with the temperature and leads to higher nitrate removal efficiencies, as previously described. The rate constant is also lower in NaCl solutions as compared with their corresponding Na2SO4 solutions under similar conditions; this result is consistent with the data and is accounted for by the oxidizing effect of the generated hypochlorite ion (Equations (5)–(8)).

Table 5

First order rate constant at varying temperatures ([NO3]0 = 1.5 mM)

Salinity (10 mM) Temperature (°C) Rate constant (1/h) R2 
Na2SO4 37 1.07 0.9978 
17 0.60 0.9922 
NaCl 40 0.71 0.9978 
20 0.42 0.9925 
Salinity (10 mM) Temperature (°C) Rate constant (1/h) R2 
Na2SO4 37 1.07 0.9978 
17 0.60 0.9922 
NaCl 40 0.71 0.9978 
20 0.42 0.9925 

Synthetic groundwater (SG1)

Figure 7 displays the removal of nitrate as a function of time in synthetic groundwater SG1 (i.e. with full ionic composition). A total of 74% of nitrate was removed in 3 h; however, a scaling layer deposited on the cathode. The scale layer was composed of calcium and magnesium compounds with 2.6 mmol Ca2+/g of scale and 0.84 mmol Mg2+/g of scale – concentrations which correspond well with the ionic composition of SG1. Such scale deposits can hinder the application of electrochemical nitrate removal from groundwater. The evolution of a scale layer with time can block nitrate active sites on the cathode, thereby disabling nitrate removal in a continuous operation mode. This hindrance does not apply to the nitrate waste stream from a Donnan dialysis process, since it does not contain calcium or magnesium cations.

Figure 7

Removal of nitrate in SG1 solution (j = 10 mA/cm2).

Figure 7

Removal of nitrate in SG1 solution (j = 10 mA/cm2).

High salinity Donnan waste stream (SG2)

Variable salinities of Na2SO4 or NaCl can be used in the stripping solution for Donnan dialysis. Hence, SG2 was tested with 5, 15, 30, and 55 mM NaCl and 1, 11, 26, and 51 mM Na2SO4. SG2 with no additional salinity was also tested. Figure 8 shows the effect of a high salinity NaCl solution on nitrate removal. The NaCl concentration increased the rate of nitrate removal following the trend 15 > 30 > 5 > 55 mM. The maximum nitrate removal, achieved with 15 mM of NaCl, was 89% after 4 h. At chloride concentrations over 15 mM, less nitrate was removed. This can be attributed to competitive adsorption between nitrate and chloride on the cathode surface. As both anions compete for active sites on the cathodic surface, higher concentrations of chloride result in less nitrate removal (Katsounaros & Kyriacou 2007). In the low salinity of 5 mM NaCl, only 78% nitrate was removed after 4 h. The reduced rate of nitrate removal by increased voltage (15.0 versus 6.5 V in 5 and 15 mM NaCl respectively) may have resulted from competitive side reaction of hydrogen evolution (Lacasa et al. 2012).

Figure 8

Normalized nitrate concentration at varying NaCl salinities (j = 10 mA/cm2; SG2).

Figure 8

Normalized nitrate concentration at varying NaCl salinities (j = 10 mA/cm2; SG2).

The effect of high Na2SO4 salinity on nitrate removal is displayed in Figure 9. Higher salinities resulted in higher nitrate removal after 4 h. The nitrate removal increased from 78% in 1 mM Na2SO4 to 87% in 51 mM Na2SO4. An increase in nitrate removal with additional sulfate was also observed by Huang et al. (2013) when using a brass-DSA system. Nitrate removal initially increased with the increase of Na2SO4 salinity, for reasons previously explained for high salinity NaCl. The low salinity solution of 1 mM Na2SO4 exhibited the high voltage of 15.2 V compared to an average voltage of 6.0 V for the higher salinities solutions of 11, 26, and 51 mM Na2SO4, resulting in unwanted side reactions. However, once the voltage was stabilized at 11 mM Na2SO4, salinity increase augmented the nitrate removal. This is due to the cationic catalysis mechanism explained by Katsounaros & Kyriacou (2007). When electro-reduction begins, positively charged cations migrate to the cathode. The cations facilitate nitrate reduction by forming neutral ion pairs with nitrate that prevent the nitrate ions from being repelled by the negatively charged electrode. Increased amounts of Na2SO4 provide additional sodium cations which augment nitrate reduction at low dosages and overcome suppression of the nitrate reduction caused by SO42− anions.

Figure 9

Nitrate removal after 4 h at varying Na2SO4 salinities (j = 10 mA/cm2; SG2).

Figure 9

Nitrate removal after 4 h at varying Na2SO4 salinities (j = 10 mA/cm2; SG2).

CONCLUSIONS

Results of this study demonstrate the feasibility of nitrate removal from a high salinity Donnan waste stream by electrochemical techniques using a copper-Ti/PbO2 DSA electrode pair. While electrochemical denitrification with nitrate contaminated groundwater resulted in a cathodic scaling layer, Donnan dialysis specific ionic composition prevented this process from occurring. Nitrate removal in a low salinity Donnan dialysis waste stream was 78% after 4 h. At high NaCl salinity, nitrate removal was inhibited with only 69% removal after 4 h. However, high Na2SO4 salinity enhanced nitrate removal and a maximum removal of 87% was achieved.

ACKNOWLEDGEMENTS

The authors wish to acknowledge with thanks the funding support of the Israel Science Foundation (Grant number 1766/17).

REFERENCES

REFERENCES
Abeygunawardhana
P.
,
Nanayakkara
N.
&
Vithanage
M.
2015
Development and optimization of Ti/Cu cathode and Ti/IrO2 anode for electrochemical denitrification
.
Desalin. Water Treat.
57
(
40
),
19025
19037
.
APHA
2005
Standard Methods for the Examination of Water and Wastewater
,
21st edn
.
American Public Health Association/American Water Works Association/Water Environment Federation
,
Washington, DC, USA
.
Beltrame
T. B.
,
Coelho
V.
,
Marder
L.
,
Ferreira
J. Z.
,
Marchesini
F. A.
&
Bernardes
A. M.
2018
Effect of operational parameters and Pd/In catalyst in the reduction of nitrate using copper electrode
.
Environ. Technol.
39
(
22
),
2835
2847
.
Birdja
Y. Y.
,
Yang
J.
&
Koper
M. T. M.
2014
Electrocatalytic reduction of nitrate on tin-modified palladium electrodes
.
Electrochim. Acta
140
,
518
524
.
Dortsiou
M.
,
Katsounaros
I.
,
Polatides
C.
&
Kyriacou
G.
2009
Electrochemical removal of nitrate from the spent regenerant solution of the ion exchange
.
Desalination
248
(
1–3
),
923
930
.
Duca
M.
&
Koper
M. T. M.
2012
Powering denitrification: the perspectives of electrocatalytic nitrate reduction
.
Energy Environ. Sci.
5
(
12
),
9726
9742
.
Gao
W. C.
,
Gao
L. L.
,
Meng
J.
,
Li
D.
,
Guan
Y. Y.
,
Cui
L.
,
Shen
X. J.
&
Liang
J. Y.
2019
Preparation of a novel Cu-Sn-Bi cathode and performance on nitrate electroreduction
.
Water Sci. Technol.
79
(
1
),
198
206
.
Garcia-Segura
S.
,
Lanzarini-Lopes
M.
,
Hristovski
K.
&
Westerhoff
P.
2018
Electrocatalytic reduction of nitrate: fundamentals to full-scale water treatment applications
.
Appl. Catal. B: Environ.
236
,
546
568
.
Govindan
K.
,
Noel
M.
&
Mohan
R.
2015
Removal of nitrate ion from water by electrochemical approaches
.
J. Water Process Eng.
6
,
58
63
.
Hasson
D.
,
Lumelsky
V.
,
Greenberg
G.
,
Pinhas
Y.
&
Semiat
R.
2008
Development of the electrochemical scale removal technique for desalination applications
.
Desalination
230
(
1–3
),
329
342
.
Hasson
D.
,
Beck
A.
,
Fingerma
F.
,
Tachman
C.
,
Shemer
H.
&
Semiat
R.
2014
Simple model for characterizing a Donnan dialysis process
.
Ind. Eng. Chem. Res.
53
(
14
),
6094
6102
.
Hine
F.
1985
Electrode Processes and Electrochemical Engineering
.
Plenum Publishing Co
,
New York, USA
.
Huang
W.
,
Li
M.
,
Zhang
B.
,
Feng
C.
,
Lei
X.
&
Xu
B.
2013
Influence of operating conditions on electrochemical reduction of nitrate in groundwater
.
Water Environ. Res.
85
(
3
),
224
231
.
Jensen
V. B.
,
Darby
J. L.
,
Seidel
C.
&
Gorman
C.
2012
Drinking Water Treatment for Nitrate
.
Technical Report 6
.
Prepared for California State Water Resources Control Board
.
Jing
D.
,
Li
W.
,
Zhao
Q.-L.
,
Wang
K.
,
Zheng
Z.
&
Gao
Y.-Z.
2015
Electroreduction of nitrate in water: role of cathode and cell configuration
.
Chem. Eng. J.
271
,
252
259
.
Katsounaros
I.
,
Dortsiou
M.
&
Kyriacou
G.
2009
Electrochemical reduction of nitrate and nitrite in simulated liquid nuclear wastes
.
J. Hazard. Mater.
171
(
1–3
),
323
327
.
Lacasa
E.
,
Llanos
J.
,
Cañizares
P.
&
Rodrigo
M. A.
2012
Electrochemical denitrification with chlorides using DSA and BDD anodes
.
Chem. Eng. J.
184
,
66
71
.
Levenspiel
O.
1999
Chemical Reaction Engineering
,
3rd edn
.
Wiley
,
New York, USA
.
Li
M.
,
Feng
C.
,
Zhang
Z.
,
Shen
Z.
&
Sugiura
N.
2009b
Electrochemical reduction of nitrate using various anodes and Cu/Zn cathode
.
Electrochem. Commun.
11
(
10
),
1853
1856
.
Li
M.
,
Feng
C.
,
Zhang
Z.
,
Lei
X.
,
Chen
R.
,
Yang
Y.
&
Sugiura
N.
2009c
Simultaneous reduction of nitrate and oxidation of by-products electrochemical method
.
J. Hazard. Mater.
171
(
1–3
),
724
730
.
Lobo
V. M. M.
1989
Handbook of Electrolyte Solutions, Parts A and B
.
Elsevier
,
Amsterdam, the Netherlands
.
Malinovic
B.
,
Pavlović
M. G.
&
Halilović
N.
2015
Electrochemical removal of nitrate from wastewater using copper cathode
.
J. Environ. Prot. Ecol.
16
,
1273
1281
.
Matejua
V.
,
Cizinska
S.
,
Krejci
J.
&
Janoch
T.
1992
Biological water denitrification - a review
.
Enzyme Microb. Technol.
14
(
3
),
170
183
.
Peel
J. W.
,
Reddy
K. J.
,
Sullivan
B. P.
&
Bowen
J. M.
2003
Electrocatalytic reduction of nitrate in water
.
Water Res.
37
(
10
),
2512
2519
.
Prasanna
D. B.
&
Chaudhari
S.
2005
Electrochemical denitrification of simulated ground water
.
Water Res.
39
(
17
),
4065
4072
.
Rao
X.
,
Shao
X.
,
Xu
J.
,
Yi
J.
,
Qiao
J.
,
Li
Q.
,
Wang
H.
,
Chien
M.
,
Inoue
C.
,
Liu
Y.
&
Zhang
J.
2019
Efficient nitrate removal from water using selected cathodes and Ti/PbO2 anode: experimental study and mechanism verification
.
Sep. Purif. Technol.
216
,
158
165
.
Su
L. H.
,
Li
K.
,
Zhang
H. B.
,
Fan
M. H.
,
Ying
D. W.
,
Sun
T. H.
,
Wang
Y. L.
&
Jia
J. P.
2017
Electrochemical nitrate reduction by using a novel CO3O4/Ti cathode
.
Water Res.
120
,
1
11
.
Xu
D.
,
Li
Y.
,
Yin
L.
,
Ji
Y.
,
Niu
J.
&
Yu
Y.
2018
Electrochemical removal of nitrate in industrial wastewater
.
Front. Environ. Sci. Eng.
12
(
1
),
9
23
.
Yu
J.
&
Kupferle
M. J.
2007
Two-stage sequential electrochemical treatment of nitrate brine wastes
.
Water Air Soil Pollut.
8
(
3–4
),
379
385
.
Zhang
Y. M.
,
Zhao
Y. L.
,
Chen
Z.
,
Wang
L. Q.
,
Wu
P. P.
&
Wang
F.
2018
Electrochemical reduction of nitrate via Cu/Ni composite cathode paired with Ir-Ru/Ti anode: high efficiency and N2 selectivity
.
Electrochim. Acta
291
,
151
160
.