The reduction process of nitrate at copper based electrodes was investigated. The cyclic voltammetry (CV) studies allowed us to establish the specific parameters concerning the electrodeposition of the individual metals and their alloys. It was demonstrated that the products resulting from electrochemical nitrate reduction (ENR) in alkaline media can be detected by cyclic hydrodynamic voltammetry (CHV) and square wave voltammetry (SWV) techniques at Cu and CuSn plated Pt electrodes. Moreover, using SWV ammonium can be electrochemically detected with good accuracy. An enhancement of the electrocatalytic activity of Cu by alloying it with Sn was observed. The reduction of nitrate was investigated in an engineering laboratory scale flow reactor under different operating conditions. On the two investigated types of cathode materials (Cu and CuSn), the concentration of nitrate was reduced electrochemically to the maximum permissible limit (50 mg/L) with a energy consumption in the range of 2 – 16 kWh/kg NaNO3 at a CuSn cathode.
Introduction
The removal of nitrate from waste water is a subject of intense research. The approach in our work is based on an electrochemical technique using efficient and selective electrocatalysts for nitrate reduction. The main advantages of the electrochemical approach are its versatility and simplicity in use, its low cost and its environmental friendliness. Therefore, electrochemical techniques must be developed to reduce adequately and specifically nitrate ions to harmless products such as nitrogen. The nitrate reduction mechanism is very complex and many intermediate species can be generated during the reduction. Consequently, a large number of cathode materials have been studied under various conditions. For a fast and simple detection of the products resulting from electrochemical nitrate reduction (ENR) square wave voltammetry (SWV) and cyclic hydrodynamic voltammetry (CHV) with rotating ring-disk electrode (RRDE) represent versatile analytical tools for the detection of reaction intermediates. It is known that copper and copper-tin alloys exibit electrocatalytic activity towards nitrate reduction [1, 2, 3, 4, 5, 6] and therefore they were used in our studies. The group of Polatides et al. [7] reported that, by using a CuSn electrode at very negative potentials (E = -2 V vs. Ag/AgCl), nitrate can be removed up to 97%, with a selectivity of 35% for N2 as a final product. The interest in CuSn coatings increased during the last years due to their better corrosion and mechanical properties compared to pure copper or tin coatings. In general it is a major challenge to develop new methods for nitrate removal from waste waters in which biological alternative cannot be applied. Our work has been strongly influenced by the target application with a strong focus on the cathode material as one crucial parameter.
Experimental Section
Reagents
Sodium nitrate (Merck), sodium nitrite (Merck), copper (II) sulphate pentahydrate (Grüssing), sodium hydroxide (Merck), tin sulphate (Merck), copper (II) sulphate pentahydrate (Grüssing) and sulphuric acid (Merck) were of analytical grade and used as received without further purification. All solutions were prepared with 18.2 MΩ cm (250C) Millipore Milli-Q® water. For all measurements on nitrate reduction 1 M NaOH was used as supporting electrolyte. All the measurements were performed in the presence of air. The pH value of the solution were recorded by a pH/Cond 340i (WTW, Germany) pH-meter.
Electrode materials
In order to study the influence of deposition parameters and bath composition on the electrodeposition of copper, tin and copper-tin alloys, potentiostatic and galvanostatic experiments were performed. Most of the characterization measurements have been chosen to gain further insight on the range of electroplating conditions in which adherent and uniform coatings can be produced. A number of techniques, including scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDX) were used to provide the information on these coatings development processes.
Experimental setups
Voltammetric measurements were carried out using a BioLogic Potentiostat/Galvanostat model VSP (France). A standard glass cell with three electrodes was used. The counter electrode (CE) was an oxygen evolving dimensionally stable anode (Ti/IrO2-Pt) and a Ag/AgCl/KClsat electrode was used as reference electrode (RE).
An electrochemical filter-press micro flow reactor (ElectroCell A/S, Denmark) was used for the electroreduction of nitrates. In both compartments a volume of 300 mL of solution was circulated. The cathode and anode chambers were separated by a Nafion® 324 (DuPont) cation selective – membrane (separator). A model electrolyte (0,1 M NaNO3, 1M NaOH) was used in the cathodic compartment. The anolyte was 1.0 M NaOH. The anodic and cathodic compartments were of equal volumes (10 mL). Sealing of the cell was achieved by using rubber frames, specially designed for the reactor. The geometrical areas of the cathode and the anode were 10 cm2 each and the distance between the two electrodes was 1 cm. The electrolytes were circulated through the cell compartments with a peristaltic pump (TC type, with two channels, Medorex, Germany) with a monitored flow rate. The determination of nitrate and nitrite was performed by ion chromatography (Dionex DX 100 with an anion column for AG 14A/AS14 A) after appropriate dilution. All measurements were performed at room temperature.
Experimental procedure
Model solutions containing nitrate ions were filled in the cathodic compartment of the flow cell. By applying direct current to the electrodes, nitrate ions move towards the cathode by passing through a cationic selective membrane. In this way, we eliminated possible interferences such as nitrite oxidation at the anode. In order to analyse the nitrate and nitrite concentrations samples were taken from the cathode compartment at specific time intervals with a syringe. The total time of the experiments was 24 hours. The experiments were conducted in galvanostatic operation mode (Cu: -20 and -40 mA·cm-2; CuSn: -25 and -50 mA·cm-2).
Results and Discussion
1. Preparation and characterization of Cu and CuSn materials
One aim of this study was to develop a simple, environmentally friendly cupric-stannous plating solution containing a minimum of components that can be electroplated at reasonable rates at room temperature and can be used for ENR. The results discussed in [8] allowed us to establish the specific parameters concerning the electrodeposition of the individual metals and their alloys. Potentiostatic and galvanostatic experiments were performed to study the influence of parameters on the electrodeposition of Cu, Sn and CuSn alloys. SEM images taken for CuSn deposits indicate significant changes in their surface morphology with changing electrodeposition parameters (Fig. 1). As can be seen in Figure 1 (A and B) the CuSn deposits have a uniform surface morphology when galvanostatic deposition is used.
Fig. 1: SEM micrographs showing the surface morphologies of CuSn films obtained at i = 80 mA/cm2 (A), i = 120 mA/cm2 (B), E = -0.5 V (C) and E = -0.6 V (D) with stirring (2000 rpm)
For the potentiostatic deposition at E = -0.5 V vs. Ag/AgCl (Fig. 1C) the surface morphology of the deposits is more rough. This roughness could be related to the fact that, at E = -0.5V the deposition of tin just starts. This assumption is in correlation with the CV studies presented in [8]. At more cathodic potentials (Fig. 1D) more uniform and dense deposits are obtained. Moreover, the SEM measurements showed that the CuSn deposits are more uniform if potentiostatic deposition is applied. Analyzing the CuSn deposits obtained under galvanostatic and potentiostatic conditions via EDX showed that the highest tin percentage in the deposit (73%) was obtained at E = -0.6 V (Tab. 1).
Tab 1: Correlation between potential and CuSn content for the obtained deposits
An electrode material with high Sn content was obtained. In the following the properties of the electrode material will be discussed for electrochemical reduction process.
2. Detection of electroactive products
After synthesis and characterization of copper, tin and copper-tin alloy, the copper and copper-tin electrode material were chosen as cathode materials for the reduction of nitrate from model nitrate solutions. The objective of this part was to establish the optimal conditions for the detection of the electroactive species resulting from ENR. For a fast and simple detection of the products resulting from ENR, square wave voltammetry (SWV) and cyclic hydrodynamic voltammetry (CHV) with rotating ring-disk electrode (RRDE) represent versatile analytical tools. In order to evaluate the possibility of electrochemical detection of the electroactive products resulting from ENR in alkaline media, we performed preliminary studies in mono-component solutions using the standard addition method. Measurements were performed with the disk electrode disconnected; the detection of the individual species was carried out at the Pt ring electrode. In mono-component solutions, well defined oxidation peaks of nitrite, hydroxylamine and ammonia (see Fig. 2, 3, 4) were recorded using both techniques. Figures 2, 3, 4-A and B, which we report here for clarity, were published and discussed in our previous papers [9, 10]. A comparative study on the influence of Cu and CuSn alloys on ENR and the stability of the electrode materials (Fig. 5 A and B) in the presence and absence of nitrate was discussed in our previous paper [10].
Figure 2 A and B show the cyclic and square wave voltammograms recorded at the Pt ring electrode in 1 M Na2SO4 in the presence of different concentrations of NO2–. For clarity, figures present only the positive potential ranges from the corresponding anodic scan.
Fig. 2: Dependence of the anodic peak current on the NO2– concentration by CHV (A) and SWV (B) Experimental conditions: Start potential = -1.5 V and end potential = 2.0 V, ω = 1000 rpm, disk electrode disconnected; CHV: v = 500 mV s-1; SWV: wave amplitude = 50 mV; wave period = 10 ms; wave increment = 5 mV.
In mono-component solutions, depending on the concentration, the oxidation peak potential of hydroxylamine was between +1.1 and +1.3 V for CHV (Fig. 3A) and between +1.0 and +1.1 V for SWV (Fig. 3B), respectively.
Fig. 3: Anodic peak currents dependence with NH2OH concentration by CHV (A) and SWV (B) at Pt ring electrode. Experimental conditions: see Figure 1.
Anodic sweeps recorded in the presence of different concentrations of ammonium are presented in Figure 4. At pH 11 the concentration ratio between NH3 and NH4+ is 40 [11]. Thus, ammonium can be electrochemically detected via ammonia oxidation.
Comparing hydroxylamine and nitrite oxidation, we observed only a relatively small difference (from 100 to 200 mV) between the peak potentials evaluated by CHV and SWV. Taking into account that the hydroxylamine oxidation process is faster at the Pt electrode, the decrease of this difference supports the previous explanation. SWV (Fig. 4B) measurements revealed well defined peaks for ammonia oxidation at ring potentials between +1.2 and +1.5 V, demonstrating that SWV is better suited for the electrochemical detection of ammonia. In conclusion, three different peak potentials could be identified for the oxidation reactions: +0.8 V for NO2–, +1.0 V for NH2OH and approximately +1.3 V for ammonia.
Fig. 4: Anodic peak current dependence with NH4+ concentration by CHV (A) and SWV (B) at Pt ring electrode.
Figure 5 presents details of the currents recorded at the Pt ring electrode during the anodic scans for two different disk potentials: -1.3V for Cu and -1.5V for CuSn electrode, respectively. Measurements were performed by recording eight successive cycles without refreshing of the Cu and CuSn layers. It can be observed that for the selected potentials, every first cycle indicates a maximum electrocatalytic activity, which decreases rapidly in the following sweeps. Moreover, after the first cycle, the peak potentials for the currents at the Pt ring electrode shift to more positive values. This indicates a blocking of the disk surface. The decrease of the copper electrode electroactivity after the first cycle may be related to the adsorption of hydrogen and nitrate reduction products. Based on the above discussed results and by reactivating the surface of the Cu and the CuSn electrode before each experiment, we started a detailed study concerning the influence of the polarization potential of the electrode on the ENR product composition. Only the anodic portions of every first cycle are presented (Fig. 6). A decrease of the disk potential to more negative potentials causes an increase in the recorded ring currents. Depending on the disk potential, the currents recorded at the ring electrode correspond to the oxidation of different intermediate or final products of nitrate reduction.
Fig. 5: Current signals at the Pt ring electrode at different potentials of the Cu (A) and CuSn (B) disk electrodes: (A) -1.3 V and (B) – 1.5 V in 2 g/L NO3– solution containing 1 M Na2SO4 (pH 11) as supporting electrolyte
In the case of the CuSn electrode (Fig. 6) the electrocatalytic activity of the electrode material is improved. Two distinct signals (peaks) are observed. Based on the results obtained in mono-component solutions, the peak at ca. 1.0 V can be attributed to the oxidation of nitrite (see Fig. 2) and hydroxylamine (see Fig. 3). For disk potentials more negative than -0.5 V, a clear signal, corresponding to ammonia oxidation, can be observed at ring potentials starting at 1.3 V. Bouzek et al. [12] and Gootzen et al. [13] concluded that from all possible products that are formed during the nitrate reduction only nitrite and hydroxylamine are sensitive towards oxidation at the platinum ring electrode and other possible products like NH4+, N2 and N2O are not susceptible towards oxidation at the platinum ring electrode. Contrarily, Endo et al. [14] studied the oxidation of ammonia on a disk electrode and concluded from RRDE experiments that two kinds of intermediates involved in the anodic oxidation of ammonia on platinum could be detected in situ; one can then be reduced (probably NOx) and another can be further oxidized (such as NH2OH).
Fig. 6: Influence of the polarization potential on the oxidation currents recorded by SWV at a Cu (A) and CuSn (B) disk electrode. Experimental conditions: Start potential = -1.5 V and end potential = 2.0 V, ω = 1000 rpm, wave amplitude = 50 mV; wave period = 10 ms; wave increment = 5 mV. The Cu and CuSn plated disk electrode was polarised at different constant potentials as indicated in the legends
3. Direct reduction in electrochemical flow reactor
As a part of the evaluation of electrode materials, voltammetric experiments were performed on two electrode materials. Linear sweep voltammetry (LSV) measurements were performed at a Cu and CuSn (Cu27Sn73) electrodes in the absence and presence of nitrate (Fig. 7 A and B). The differences between the two voltammograms were attributed to the electrocatalytic effects specific of each electrode material. The potential was swept in the cathodic range until hydrogen evolution occurred.
Fig. 7: Linear sweep voltammograms at a Cu (A) and CuSn (B) electrode in 1M NaOH (black line) in presence of 0.1M NaNO3 (red line). Scan rate: 10 mV s-1.
A comparison between the two electrode materials, shows an enhancement of the nitrate reduction when CuSn is used. This is explained by the faster start of the electroreduction process (−0.7 V in Fig. 7B vs. −0. 9 V in Fig. 7A). In order to analyse the reaction associated with the peaks, measurements involving prolonged electrolysis were performed at the current densities: (A) -20 mA cm-2 at a Cu electrode and (B) -25 mA cm-2 at a CuSn electrode (Fig. 8).
Fig. 8: Evolution of the concentration of NO3– and NO2– vs charge. Experimental conditions: 0.1 M NaNO3 in 1 M NaOH electrolyte solution; (A) -20 mA cm-2 at a Cu electrode and (B) -25 mA cm-2 at a CuSn electrode.
After 24 h of electrolysis, at the Cu cathode the final concentration of nitrate reaches 1 g/L and it is totally eliminated at the CuSn cathode. The nitrite concentration increases at the Cu electrode and after that, is further diminished under the Maximum Permissible Contaminant Level (MPCL) of 50 mg/L after 24 h of electrolysis.
ENR was further investigated at even higher current densities in order to establish optimized working parameters. For this purpose, prolonged electrolyses were carried out at -40 mA cm-2 at a Cu electrode and -50 mA cm-2 and CuSn electrode, results presented in [8]. The obtained results show that at the same charge (~ 0.29·10-4 C·cm-2) and flow rate (200 mL·min-1), for the Cu electrode, a period of 18 h is sufficient to reduce the nitrate below the MPCL, while for the CuSn electrode, after 14 h the ENR is completed. Thus, we can conclude that at higher current density values, the ENR is more effective at the CuSn electrode.
For cost evaluation of the ENR process, is important to identify the most efficient electrode material. Figure 9 A and B show the concentration decrease of nitrate at Cu and CuSn electrodes, respectively, obtained at a flow rate of 200 ml min-1, from a solution containing 0.1 M NaNO3 in 1 M NaOH, at three current densities. In order to reach the MPCL, the prolonged electrolyses need to perform a conversion degree for nitrate of at least 99.5%.
Fig. 9: Concentration profile of nitrate vs. time at a Cu (A) and CuSn (B) electrode. Experimental conditions: flow rate of 200 mL·min-1; electrolyte: 0.1 M NaNO3 + 1 M NaOH.
At the Cu electrode the MPCL was reached at -40 mA·cm-2 after 20 h. On the other hand, for the CuSn electrode, at -25 mA·cm-2, the MPCL was reached after 22 h, while at the highest current density (-50 mA·cm-2) the MPCL was reached after 18 h.
The key parameters: energy consumption (Ws) and current efficiency (rF) which describe the ENR process were evaluated. All the values were calculated at a flow rate of 200 mL·min-1 and until the MPCL was reached (Fig. 10).
Fig. 10: Energy consumption (Ws) and current efficiency (rF) values for the ENR at Cu and CuSn electrodes, obtained in synthetic nitrate solutions.
The obtained energy consumption (Ws) values are about the same on both electrode materials and increase almost linearly with the increase of the current density, from 2 kWh/kg NO3– to 16 kWh/kg NO3–. The calculated Ws values (at higher current densities) are in accordance with those obtained by Katsouranos et al. [15], who succeeded to remove nitrate from an alkaline medium at an energy consumption value of 16.5 kWh/kg NaNO3 at a Sn electrode. Similarly, Reyter et al. [16] studied the electroreduction of nitrate at a Cu electrode and obtained an energy consumption value of 25 kWh/kg NaNO3. The current efficiency values decrease during the electrolysis by ca. 10%. Our results are in accordance with those of Reyter et al. [16] who observed a decrease of the current efficiency with the time. For the Cu material after 30 min of electrolysis a current efficiency of 30% was obtained, with a decrease to 20% after 180 min electrolysis.
Conclusions
The reduction process of nitrate at copper based electrodes was investigated. The CV studies allowed us to establish the specific parameters concerning the electrodeposition of the individual metals and their alloys. It was demonstrated that the products resulting from ENR in alkaline media can be detected by the CHV and SWV technique at Cu and CuSn plated Pt electrodes. Moreover, using SWV ammonium can be electrochemically detected with good accuracy. An enhancement of the electrocatalytic activity of Cu by alloying it with Sn was observed. On the two investigated types of cathode materials (Cu and CuSn), the concentration of nitrate was reduced electrochemically to the maximum permissible limit (50 mg/L) with a energy consumption in the range of 2 – 16 kWh/kg NaNO3 at a CuSn cathode. In our test the energy consumption are comparable and even smaller compared to values reported in literature. An identical quantity of nitrate, was reduced at the CuSn cathode below the permissible limit (99.5% conversion), in 18 h while at the Cu cathode the process needed 20 h. The current efficiency is about 35% corresponding to a conversion of 99.5%. Comparisons between these two electrodes indicate definite advantages in using the CuSn alloy for nitrate and nitrite reduction. Electrochemical removal of nitrate and nitrite has been demonstrated in a laboratory scale flow reactor under different operating conditions. In the used flow reactor the nitrate/nitrite reduction efficiency was improved with an increase in the current density and operation of the cell in a divided configuration.
Acknowledgements
The financial support during the PhD program of F. M. Cuibus through SOPHRD/6/1.5/S/3/ID 5216 is gratefully acknowledged. The Research Grant for Doctoral Candidates and Young Academics and Scientists (Forschungs-/ Studienstipendium) DAAD is greatly appreciated for the research stay at the TU Ilmenau. A travel grant of the “Frauenförderung der TU Ilmenau” is greatly appreciated.
References
- Badea, G. E., Electrocatalytic reduction of nitrate on copper electrode in alkaline solution. Electrochimica Acta 2009, 54, (3), 996-1001.
- Bouzek, K.; Paidar, M.; Sadílková, A.; Bergmann, H., Electrochemical reduction of nitrate in weakly alkaline solutions. Journal of Applied Electrochemistry 2001, 31, (11), 1185-1193.
- Mácová, Z.; Bouzek, K.; Šerák, J., Electrocatalytic activity of copper alloys for NO3- Reduction in a weakly alkaline solution: Part 2: Copper-tin. Journal of Applied Electrochemistry 2007, 37, (5), 557-566.
- Piao, S.; Kayama, Y.; Nakano, Y.; Nakata, K.; Yoshinaga, Y.; Shimazu, K., Nitrate reduction on tin-modified rhodium, ruthenium, and iridium electrodes. Journal of Electroanalytical Chemistry 2009, 629, (1-2), 110-116.
- Katsounaros, I.; Ipsakis, D.; Polatides, C.; Kyriacou, G., Efficient electrochemical reduction of nitrate to nitrogen on tin cathode at very high cathodic potentials. Electrochimica Acta 2006, 52, (3), 1329-1338.
- Polatides, C.; Kyriacou, G., Electrochemical reduction of nitrate ion on various cathodes – Reaction kinetics on bronze cathode. Journal of Applied Electrochemistry 2005, 35, (5), 421-427.
- Polatides, C.; Kyriacou, G., Electrochemical reduction of nitrate ion on various cathodes – Reaction kinetics on bronze cathode, Journal of Applied Electrochemistry 2005, 35, (5), 421-427.
- F. M. Cuibus, Electrochemical removal of nitrate from waste water, PhD Thesis 2012, 1-128.
- F. M. Bălaj, F. Imre-Lucaci, S. A. Dorneanu, P. Ilea, Detection of electroactive products resulted from electrochemical nitrate reduction in alakaline media, Stud. Univ.“Babeş-Bolyai”, Chemia, LIV, Sp. Iss., 2009, 1, 128 – 134.
- F. M. Cuibus, A. Ispas, A. Bund, P. Ilea, Square wave voltammetric detection of electroactive products resulting from nitrate reduction in alkaline media, Journal of Electroanalytical Chemistry 2012, 675, 32-40.
- Kelter, P.; Mosher, M.; Schott, A., Chemistry: The practical Science, Media Enhanced Edition. Cengage Learning Inc., USA, 2009.
- Bouzek, K.; Paidar, M.; Sadílková, A.; Bergmann, H., Electrochemical reduction of nitrate in weakly alkaline solutions. Journal of Applied Electrochemistry 2001, 31, (11), 1185-1193.
- Gootzen, J. F. E.; Lefferts, L.; Van Veen, J. A. R., Electrocatalytic nitrate reduction on palladium based catalysts activated with germanium. Applied Catalysis A: General 1999, 188, (1-2), 127-136.
- Endo, K.; Katayama, Y.; Miura, T., A rotating disk electrode study on the ammonia oxidation. Electrochimica Acta 2005, 50, (11), 2181-2185.
- Katsounaros, I.; Dortsiou, M.; Kyriacou, G., Electrochemical reduction of nitrate and nitrite in simulated liquid nuclear wastes. Journal of Hazardous Materials 2009, 171, (1-3), 323-327.
- Reyter, D.; Bélanger, D.; Roué, L., Optimization of the cathode material for nitrate removal by a paired electrolysis process. Journal of Hazardous Materials 2011, 192, (2), 507-513.
![]() PDF Version of the article |
![]() Epub Version of the article |
![]() Flash Version of the article |
[qr-code size=”2″] |