Determination of the Applicability of the Tungsten-Containing Material As Lowcost Electrodes for Reverse Electrodialysis

Electrodialysis, especially reverse, is a promising method of water desalination, concentration of solutions, extraction of valuable components from waste and rinse water, and power generation. The main problem is the search for low-cost universal anode-cathode materials. The work aims to determine the possibility of using the VNZh90 superalloy (5 % Ni, 5 % Fe, 90 % W) and the electroplated Ni-W alloy as a universal cathode-anode material for reverse electrodialysis. The crystal structure of the Ni-W alloy was studied by X-ray diffraction analysis; the morphology was studied by scanning electron microscopy. The anodic behavior of both alloys was studied by voltammetry in 6 % HCl in a saturated NaCl solution. The high passivity of the VNZh90 superalloy was revealed. On the repeated anodic curve, the current density of the passivation plateau decreased 2.8 times and was 37 mA/dm2. This indicates that the use of the VNZh90 superalloy is promising as a universal cathode-anode material of a reverse electrodialyzer. The phenomenon of significant passivation for the Ni-W alloy was also revealed. The primary curve of the alloy showed two dissolution peaks and a well-defined passivation plateau. Probably, the first peak corresponded to a more active phase with a low W content. This was confirmed by the absence of the first peak on the repeated anodic curve and the identity of the passivation plateaus of the primary and repeated curves. The passivation current density was 209 mA/dm2. These data also indicate the possibility and prospects of using the electroplated Ni-W alloy as a universal cathode-anode material of a reverse electrodialyzer after optimizing the composition and deposition method of the alloy, as well as reducing the wear rate.


Introduction
The Sustainable Development Strategy is an effective direction for the development of civilization. Sustainable development is based on the following basic principles: a) recycling or reuse of materials; b) maximum use of clean energy or energy derived from renewable sources; c) prevention of environmental pollution. Electricity and drinking water are among the most problematic resources [1]. In many countries, drinking water is obtained from excessively saline or sea water by the method of electrodialysis [2]. However, in the presence of two and polyvalent metal cations in water, precipitation is formed during electrodialysis, both on membranes and in electrode spaces. To prevent this process, as well as to conduct electrodialysis of concentrated solutions, reverse electrodialysis is used [3,4]. This method is promising and widely used for various purposes. With the help of reverse electrodialysis, it is possible to obtain drinking water [2], Reverse electrodialysis can be used to extract valuable components from wastewater, for example, Ni 2+ from galvanic wastewater [21].
In addition, electrodialysis can be used for the targeted production of metal hydroxides. In [22,23], electrodialysis was used to obtain electrochemically highly reactive Ni(OH) 2 . The authors showed that the use of electrodialysis production of nickel hydroxides is promising compared to other chemical (decomposition [24], two-stage high-temperature synthesis [25], homogeneous template deposition [26,27], deposition at high [28,29], and low [30,31] supersaturation) and electrochemical methods (synthesis in a slot diaphragm electrolyzer [32,33], including using a template [34]). It may be promising to replace the chemical synthesis of layered double hydroxides (LDH) (for example, Ni-Al LDH [35] and Zn-Al LDH [36,37]) with the electrochemical synthesis in an electrodialyzer.
Although the promising method of reverse electrodialysis has many advantages and is widely used in technology, it has some disadvantages. In particular, the possibility of reproduction of microorganisms on ion-exchange membranes [38]. However, the most significant disadvantage is the problem of electrode materials. In reverse electrodialysis, the processes occurring on the electrodes change periodically. In this case, the anodes (oxidation processes occur) become cathodes (reduction processes occur) and vice versa. It should be noted that in the anodic mode, the electrodes must be insoluble. Few materials are known to operate in both the cathodic and anodic modes without the participation of electrode materials in the reaction. These are mainly precious metals of the platinum group. Therefore, the identification of low-cost materials that can work as a universal electrode (both anode and cathode) in a reverse electrodialyzer is very promising.

Literature review and problem statement
The problem of electrode materials for reverse electrodialysis is quite acute. Most of the publications generally ignore the issue of electrodes, focusing on ion-exchange membranes [11,17]. The electrode materials described in the literature can be basically divided into two groups. The first group consists of mixed oxide electrodes. In [39], titanium electrodes coated with mixed Ru-Ir oxides were used. This electrode was used in dilute and saturated chloride solutions, similar to the ORTA electrode for electrolysis for the production of chlorine and alkali. However, it is known that ORTA electrodes are used only as anodes. In the cathodic mode, the oxides of ruthenium and titanium are reduced to metals, and after that, this electrode is no longer suitable for operation as an anode. Accordingly, during cathodic polarization on all electrodes, which are mixed oxide coatings on a titanium base, the oxides will be reduced to free metals both directly and by liberated ad-hydrogen atoms. As a result, during anodic polarization, the titanium base of the electrode will be passivated, and the electrode performance will sharply decrease. Therefore, mixed oxide electrodes can be used in reverse electrodialysis only if the anode period is short. The second group is metal electrodes based on platinoids. In [40], the authors used electrodes in the form of a niobium grid coated with Pt. These electrodes are absolutely inert and can effectively work both as a cathode and as an anode. However, these electrodes are very expensive, which significantly reduces the possibility of using large-size reverse electrodialysis units.
It should be noted that the search for low-cost universal electrode materials should be carried out precisely among metal electrodes. It is well known that any metal electrodes can effectively work as cathodes. However, in the case of monometallic electrodes, only noble metals exhibit the properties of an insoluble anode in any (acidic and alkaline) media, especially in the presence of strong depassivators. However, it is promising to study the anodic behavior of binary metal systems, in which one component is insoluble in an acidic medium and the other in an alkaline one. The most promising is the Ni-W system: Ni is strongly passivated in neutral and alkaline media, and W is passivated in acidic ones. Since the melting points of Ni and W are very different, this metal system can be obtained in two ways: 1) formation of a Ni-W superalloy by powder metallurgy; 2) electrochemical production of Ni-W alloy. Superalloys are not true alloys but are a composite material (similar to organic-organic [41] and inorganic-inorganic composites [42]), consisting of particles of a solid component (W or WC) and particles of a binder metal (most often Ni or Co) [43,44]. When pressed with heating, the binder metal is partially melted, connecting the particles of the solid component. In [45,46], a method was developed for processing various superalloys by selective dissolution of the binder metal in an acidic electrolyte, followed by regeneration of the binder metal powder [47,48]. It was shown that WC-based superalloys are easily processed. At the same time, a very high passivity of the VNZh90 superalloy (90 % W, 5 % Ni, 5 % Fe) was revealed. The high passivity of this alloy was confirmed during anodic oxidation in alkaline-ammonium solutions [49,50] both at constant [51] and alternating current [52]. However, to confirm the possibility of using the VNZh90 alloy as a universal anode-cathode material of a reverse electrodialyzer, it is necessary to additionally study the anodic behavior of this alloy in the corresponding solutions.
Another promising material for a universal electrode is electroplated Ni-W alloy [53,54]. Depending on the electrolyte and deposition conditions, coatings of various structures are formed [55], including nanocrystalline [56,57] and amorphous [58] deposits. Ni-W coatings are used to form microdevices [59], and also as highly hard [60], magnetic (Ni-Fe-W) [61] and catalytically active [62] materials. In [63,64], the authors studied the corrosion properties of the electroplated alloy and showed high corrosion resistance. However, to confirm the possibility of using the Ni-W alloy as a universal anode-cathode material of a reverse electrodialyzer, it is necessary to additionally study the anodic behavior of this alloy in the corresponding solutions.

The aim and objectives of the study
The study aims to determine the possibility of using the VNZh90 superalloy and the electroplated Ni-W alloy as a universal cathode-anode material for reverse electrodialysis.
To achieve the aim, the following objectives were set: -to study the anodic behavior of the VNZh90 alloy in a solution corresponding to the electrolyte of the electrode space of the electrodialyzer; -to study the characteristics and anodic behavior of the electroplated Ni-W alloy in a solution corresponding to the electrolyte of the electrode space of the electrodialyzer.

1. Description of the sample of the investigated superalloy
The VNZh90 alloy of the following composition was used for the study: 90 % W (solid component), 5 % Ni, 5 % Fe (binder metal). The investigated sample of the VNZh90 alloy is the penetrator for armor-piercing munitions. The

2. Methods for studying the structure and morphology of electroplated Ni-W alloy
The crystal structure was studied by X-ray diffraction analysis (XRD) using a DRON-3 diffractometer (Russia) (Cu-Kα radiation, angle range 20-100° 2θ, scanning rate 0.1°/s). The morphology of the coating was studied using a 106-I scanning electron microscope (SEM) (SELMI, Ukraine).

3. Method for studying anodic behavior
Anodic polarization characteristics were measured using a specially designed cell. As a working electrode, we used a VNZh90 superalloy sample (to supply current to the sample, a copper wire was wound and placed in specially prepared tubular PVC containers for isolation from the electrolyte) and a Ni-W alloy sample. Mesh platinum was used as a counter electrode. Reference electrode -Ag/AgCl (sat.). To assess the anodic behavior of W-containing materials, the most aggressive solution was chosen as the electrolyte, simulating the electrolyte of the anodic chamber of the electrodialyzer -a 6 % (wt.) HCl solution in a saturated NaCl solution. Polarization curves were recorded in a potentiodynamic mode using an Ellins P-8 potentiostat (Russia) with an anodic sweep from a stationary potential to a potential of +2,000 mV (for VNZh90 superalloy) and +1,000 mV (for electroplated Ni-W coating) with a sweep rate of 5 mV/s. Anodic polarization curves were recorded twice.

1. Results of studying the anodic behavior of the VNZh90 alloy
Anodic polarization curves are shown in Fig. 1. On the primary anodic curve of the VNZh90 alloy, two dissolution peaks are observed, passing into the passivation plateau with a limiting current density of 100 A/dm 2 .
The repeated curve shows a sharp decrease in both the currents of the dissolution peaks and the limiting current of the passivation plateau (up to 34 A/dm 2 ).

2. Results of studying the characteristics and anodic behavior of the Ni-W alloy
The SEM images of the Ni-W coating surface (Fig. 2) show that the deposited coating is porous and has spherical inclusions in its structure, possibly just a nickel-tungsten alloy or metallic tungsten. Moreover, these particles have a submicron size or are nanoscale. For XRD, the Ni-W coating was applied to a steel base. Therefore, the diffractogram of the obtained alloy (Fig. 3) shows clear peaks of ferrite (the main component of the steel base). The diffractogram shows clearly pronounced peaks of the Ni-W alloy.
Anodic polarization curves of the Ni-W alloy are shown in Fig. 4. The results of the analysis of the digital characteristics of the polarization curves of the VNZh90 superalloy and the electroplated Ni-W alloy are given in Table 1.
The primary polarization curve of the electroplated Ni-W alloy is in the form of the classic anode curve of the passivated metal. The curve shows two dissolution peaks, after which the formation of a passivation plateau with a limiting current of 270-280 mA/dm 2 is observed. The passivation plateau is observed starting from a potential of +180 mV. There are no sharp dissolution peaks on the repeated polarization curve; the passivation plateau, starting from a potential of +180 mV, practically coincides with the passivation plateau of the primary curve.

Discussion of the results of studying the characteristics of W-containing materials
The XRD data confirmed the deposition of the Ni-W alloy [53,54]; no individual peaks of W were found on the diffractogram (Fig. 2). It should be noted that the resulting Ni-W alloy had a fairly high crystallinity. The data of scanning electron microscopy revealed the imperfection of the coating, the presence of pores in it (Fig. 3, c, d), and damage ( Fig. 3, a, b). It should be concluded that it is necessary to optimize the deposition process. In addition, a prerequisite for using this coating as a universal anode-cathode material is its deposition to a strongly passivated substrate, in particular, titanium, zirconium, or niobium.
It was revealed that the polarization curves of the VNZh90 superalloy and the electroplated Ni-W alloy had a shape characteristic of the curve of passivated materials, corresponding to the literature data [45,46].
It was found that for both alloys, the anodic polarization curves for the fresh surface of the alloy (primary curve) and the repeated polarization curves were fundamentally different (Fig. 1, 4).
For the VNZh90 alloy, the primary polarization curve (Fig. 1) had two anodic dissolution peaks, after which the dissolution current density sharply decreased, forming an indistinct passivation plateau with a constantly decreasing limiting current density. The first peak corresponded to the dissolution of Fe, the second peak -Ni. A sharp decrease in the dissolution current density allowed us to conclude that the alloy was significantly passivated. The repeated curves showed only one diffuse dissolution peak and a clearly defined passivation plateau. The dissolution peak potential on the repeated curve characterized the dissolution of Ni, while the dissolution peak of Fe was not observed. The current density of the Ni dissolution peak decreased from 306 mA/dm 2 (primary curve) to 59 mA/dm 2 (repeated curve), i.e. 5.18 times ( Table 1). The limiting current density of the passivation plateau for the repeated polarization curve was 2.8 times lower than for the primary curve (Table 1). All these data indicate very significant passivation of the VNZh90 alloy. This can be explained by the fact that in a tungsten superalloy (which is VNZh90), W is able to dissolve in nickel, which leads to additional significant passivation of nickel even in the presence of strong depassivators. At the same time, tungsten is practically insoluble in Fe. When recording the primary polarization curve (Fig. 1), two peaks were observed and dissolution of nickel and iron occurred. As a result, the surface of the VNZh90 superalloy was depleted by the more active Fe-phase, and the proportion of the W solid component and passivated Ni increased. So, only the Ni dissolution peak was present on the repeated polarization curve (Fig. 1), the peak current densities and passivation plateaus were very significantly reduced. These data fully correlate with the data presented in [46] obtained in a slightly different electrolyte. Thus, the anodic behavior of the VNZh90 superalloy clearly indicates the possibility of using it as a universal cathode-anode material. The disadvantage of this material is that  this alloy cannot undergo mechanical processing and must immediately be formed as a finished product. In addition, this alloy is still not cheap enough. However, the cost of the VNZh90 alloy can be reduced if it is made from recycled metal powders. Besides, when using this alloy as an anode, there is no need to impose high requirements for its mechanical characteristics. A possible direction of development can also be the manufacture and study of the anode characteristics of an alloy with a reduced tungsten content. For the electroplated Ni-W alloy, the primary polarization curve (Fig. 4) also had two anodic dissolution peaks, after which a clearly pronounced passivation plateau was observed. The presence of two peaks probably indicates that the coating with the Ni-W alloy is likely to have a non-uniform distribution of the tungsten content over the depth and the upper thin layer, most likely, contains either pure Ni or Ni-W alloy with a minimum tungsten content. This component is much more active in the presence of a large number of chloride ions as strong depassivators and dissolves more actively. This is indicated by the 2.38 times higher current density of the first peak compared to the second one (Table 1). When the primary polarization curve is recorded, the active components dissolve and a passivation plateau is formed. This coating structure is confirmed by the repeated curve, in which the first dissolution peak of the active Ni-phase is absent, but the second peak, although weakly expressed, is present. And most importantly, the passivation plateaus of the primary and repeated curves are practically identical both in the potential interval and in the value of the limiting passivation current density. The data obtained also clearly indicate significant passivation of the electroplated Ni-W alloy and confirm the possibility of using this alloy as a universal cathode-anode material for reverse electrodialysis. However, it should be noted that the Ni-W alloy is characterized by significantly higher current densities compared to the VNZh90 superalloys: for the dissolution peak of 3,580 mA/dm 2 (Ni-W) and 306 mA/dm 2 (VNZh90), for the passivation plateau of 275 mA/dm 2 (Ni-W) and 93 mA/dm 2 (VNZh90) ( Table 1). For the repeated polarization curve, the limiting current density of the electroplated Ni-W alloy is 209 mA/dm 2 compared to 37 mA/dm 2 (for the VNZh90 superalloy), i.e. 5.65 times higher (Table 1). That is, the service life of such an anode will be as many times lower. The advantage of an electroplated alloy is its significantly low cost, the possibility of applying it to substrates of any shape, and ease of production. The disadvantages include a high wear rate (dissolution). However, it should be noted that in this work, the alloy was deposited from only one electrolyte and in one production mode. There is reason to believe that in further studies, when optimizing the type of electrolyte and its composition, deposition parameters, and the W content in the alloy, the dissolution rate can be significantly reduced.
It should also be noted that the work has shown the possibility of using the VNZh90 superalloy and the electroplated Ni-W alloy as a universal cathode-anode material for a reverse electrodialyzer. However, for industrial use, it is necessary to carry out long-term (endurance) galvanostatic tests of these materials in the anodic mode in various electrolytes to clarify the wear rate and potential stability.

Conclusions
1. The anodic behavior of the VNZh90 superalloy in a 6 % (wt.) HCl solution in a saturated NaCl solution has been studied. The anodic polarization curve has a shape typical for passivated metals. Strong passivation of the alloy is shown: upon the repeated recording of the anodic curve, the current density of the passivation plateau (indicating the dissolution rate of the alloy) has decreased 2.8 times and is 37 mA/dm 2 . Thus, the prospects of using the VNZh90 superalloy as a universal cathode-anode material of a reverse electrodialyzer have been shown.
2. The Ni-W alloy has been electroplated. The formation of Ni-W alloy has been confirmed by X-ray diffraction analysis. The anodic behavior of this alloy in a 6 % (wt.) HCl solution in a saturated NaCl solution has been studied. The anodic polarization curve also has a shape typical for passivated metals. On the anodic polarization curve of the alloy with a fresh surface, two dissolution peaks and a clearly pronounced passivation plateau have been revealed. A well-grounded hypothesis has been put forward about the uneven distribution of W in the surface layer of the alloy: the first peak corresponds to a more active phase with a low W content. The hypothesis has been confirmed by the fact that on the repeated anodic curve, the first peak is absent due to the dissolution of the active phase, and the passivation plateaus of the primary and repeated curves coincide. A significant passivity of the alloy has been shown: the passivation current density is 209 mA/dm 2 . These data have also indicated the possibility and prospects of using the electroplated Ni-W alloy as a universal cathode-anode material of a reverse electrodialyzer. It is recommended to apply a coating on the surface of highly passivated metals (Ti, Mo), and to optimize the composition and deposition method of the alloy, as well as reduce the wear rate.