Electrochemical Dissolution and Passivation of Cu-Ni Alloys in Sodium Sulphate Aqueous Solution

The electrochemical behaviour of two copper-nickel alloys is studied in 0.1 M Na2SO4 using cyclic voltammetry, potentiodynamic and current time transient techniques. The microstructure of the compounds formed during anodic sweep is characterized using x-ray diffraction analysis. The forward sweep is characterized by the existence of two well-defined potential regions; the selective dissolution and the simultaneous dissolution potential regions. The first potential region consisted in two anodic peaks A1 and A2 corresponding to the formation of Ni(OH)2 and NiO, and Cu2O, respectively. The second potential region consisted in two anodic peaks A3 and A4 corresponding to the formation of CuO and Ni2O3, respectively. It is found that the Cu-Ni system in sulphate solution follows the dissolution-redeposition mechanism of the dissolution of metals from the alloys. The X-ray diffraction analysis confirmed the existence Ni2O3 and CuO on the electrode surface of nickel rich alloy polarized to potential more noble than the potential of A4 and the existence of Cu2O and NiO on the surface of copper-rich alloy polarized to a potential in the A2 region. Atomic Force microscope images show surface enlargement beyond the critical potential Ecrit.


Introduction
The copper-nickel system is an isomorphous one in which only a single type of crystal structure is observed for all ratios of the components [1].Many studies have been conducted of the dealloying of copper-based alloys. The electrochemical behavior of Cu-Ni alloys has been studied by numerous authors in alkaline solutions [2][3][4][5][6][7].It was established that the passivity of arrange of Cu-Ni alloys in 0·1MKOH solution was due to oxide film growth by migration under a high electric field. The rate-controlling parameters were a function of the copper content of the alloy, but largely followed the kinetics of film growth of pure nickel. Analysis by Auger electron spectroscopy of the formed anodic films indicates that the oxide film is predominantly nickel oxide even when the Cu content in the alloy is 70%, suggesting that Cu in the alloy does not affect the film growth process to any significant extent.
The passive layer formed potentio-statically in 1·0 M NaOH on Cu-50 wt-%Ni has also been investigated, using X-ray photoelectron spectroscopy(XPS)and ion scattering spectroscopy(ISS) [6].The thickness and the composition of the sublayers of the passive film, as well as the composition of the underlying metal surface, were found to depend on electrochemical parameters such as potential and time of passivation.
Copper-nickel alloy [1] is an amorphous system where copper and nickel forms one single phase all over the range of composition. Two mechanisms of dissolution are generally proposed for the dissolution of metals from isomorphous alloys: simultaneous dissolution of both components of the alloy followed by redeposition of one component on the surface and selective dissolution of one element from the crystal lattice. Three mechanisms are proposed for the dissolution of metals from alloys below their critical potential Ecrit. For a binary alloy AB, below its critical potential, removal of the less noble component B occurs where the more noble component A is present as a layer that covers the alloy surface, which is then ceased any further removal of the less noble component and the current decays to a very low value and is termed the passivation current denisity ipass. As soon as the critical potential of the alloy is reached, the surface losses its planarity and the current rises sharply to a very high value where both components dissolve simultaneously. The ionizationredepositoin mechanism assumes an initial simultaneous ionization of both components A and B. This step may involve intermediate formation of ad-atoms of the less noble component, B. Alternatively; direct transfer of B ions from kink sites to the electrolyte solution must be considered [9,10]. Irrespective of these details of the initial charge transfer step, the electropositive component A is thought to redeposit on the alloy surface in a consecutive step. It is assumed that an ions may be dissolved from the alloy at underpotentials that are cathodic with respect to the reversible potential EA' that agglomerates on the electrode surface. They have, therefore, the character of an unstable intermediate species and will be plated out on the surface of A' crystals at the same electrode potential.
The volume diffusion mechanism, which was introduced by Pickering and Wagner [11], does not consider an "under potential dissolution" of the electropositive component. Instead, the atoms of the latter are assumed to accumulate as mobile ad-atoms on the electrode surface. It may be expected, therefore, that these ad-atoms will have an increasing tendency to move back to the kink sites and steps and to block any further removal of the less noble component from these positions by the usual mechanism of the metal dissolution. Provided that the anodic over potential is high enough, there is, however, an alternative possibility that the preferential dissolution will proceed from terrace sites (which may be in the vicinity of the kink sites [12]. This process results in the formation of surface vacancies. The major concept of the volume diffusion mechanism is that by the injection of these surface vacancies into the bulk of the alloy, a large vacancy super saturation is developed adjacent to the alloy /electrolyte interface which enhances the diffusivity beyond its thermal value. Repopulating of the electrode surface with Cu atoms via volume diffusion is therefore considered to be possible even at room temperature.
The surface diffusion mechanism considers the nucleation and growth of the crystals of the pure, or almost pure, noble component via a surface diffusion process. Following earlier suggestions by Gerischer [13], this mechanism may be described in detail by a model which, for a binary AB, includes: 1. Removal of both components from steps kink sites and formation of ad-atoms of the more noble component A at electrode potentials E < E c . The ad-atoms may subsequently (a) crystallize via surface diffusion or (b) accumulate at the steps and kink positions, where they are thought to block the removal of the less noble species. However, in contrast to the assumptions that were made with the volume diffusion process, a small but finite steady-state dissolution rate of B atoms is now considered to be possible by exchange of places between A and B atoms.
2. Removal of B atoms from terrace sites in order to account for the increased dissolution rate at E > E c . Since the remaining an atom is expected to form a porous layer of small individual crystallites, this process is thought to operate without transport of B atoms to the electrode surface via volume diffusion.
This present work has used two alloys of different compositions, one being copper rich and the other nickel rich, to elucidate the mechanism of dissolution and the composition of the passive films formed in the Cu-Ni system in 0·1MNa 2 SO 4 solution. The voltammetric profiles of pure copper and pure nickel were included for comparison. The microstructure and composition of the passive film formed during the anodic sweep was characterized by X-ray diffraction analysis. Atomic force microscopy (AFM) was used to examine the alloy surface for surface enlargement associated with the electric field beyond the critical potential E crit .

Experimental
Pure copper and nickel (supplied by Merck 99.99%) and two copper-nickel alloys of the composition Cu-20 wt% Ni and Cu-80 wt% Ni were used in studying the voltammetric behaviour copper-nickel alloys in aerated Na 2 SO 4 solution. The alloys were made by fusion of pure two metals in a graphite crucible at a desired temperature, air cooled and scraped into the desired thickness area. The electrodes were then mounted in a Teflon mount so as only their cross sectional areas were exposed to solution. Before each experiment, the electrodes were mechanically polished starting with fine grained emery papers and followed with alumina paste to obtain a mirror-like electrode surface, then rinsed with acetone and doubly distilled water. Measurements were performed in Na 2 SO 4 solutions prepared from Analar grade chemicals and used without further purification. A conventional electrochemical cell of 100 cm capacity was used for the present work. It was contained three separated compartments two of them were used for fitting both the working and the platinum counter electrodes. The third compartment was used for fitting the reference electrode, which is saturated calomel electrode (SCE).
Cyclic voltammetric, potentiodynamic anodic polarization and potentiostatic current/time measurements were carried out with the help of an EG&G potentiostate\Galvanostate Model 237A controlled by a PC computer using 352 corrosion software.
Morphology of the alloy surface in the potential range beyond the critical potential E crit was maintained by scanning electron micrographs SEM), model XL-200 Philips operated at 15 k eV. The composition of the corrosion products formed during anodic polarization over the alloys surfaces was studied using x-ray diffraction analysis using Philips P. W. Model 1730 diffractometer adopted at 40 kV and 25 mA with Cu-k α radiation and a Ni filter.

Results and Discussion
The electrochemical behaviour of the alloys I and II was studied in 0.1 M Na 2 SO 4 between E c = -1600 mV up to E a = 200 mV and 1000 mV, respectively with sweep rate of 50 mV s -1 and the data is represented graphically in Figure 1. The curves of pure copper and pure nickel were introduced Universal Journal of Chemistry 1(1): 1-6, 2013 3 for comparison. The electrochemical behaviour of pure copper, Fig. 1 (curve 1), studied between E c = -1600 mV and E a = 200 mV and was characterized by the appearance of two anodic peaks A 2 and A 3 , and two cathodic peaks C 2 and C 3 . The anodic peaks A 2 and A 3 are related to the formation of Cu 2 O and CuO, respectively [14]. 2Cu At the same time, the existence of the cathodic peaks C 2 and C 3 was ascribed to the reduction of Cu 2 O to Cu and CuO to Cu 2 O, respectively. Figure 1, curve 2, represents the cyclic voltammetric behaviour of pure nickel [14] studied between E c = -1600 mV and E a = 1000 mV. Two anodic peaks, A 1 and A 4 , appear on sweeping the potential in the anodic direction. These peaks are related to the formation of NiO and Ni 2 O 3 , respectively.
Ni + H 2 O →NiO + 2H + + 2e¯ E = -0.515 V (SCE) A 1 [3] On sweeping the potential in the cathodic direction only one cathodic peak C 4 appears due to the reduction of Ni 2 O 3 to NiO. The cathodic C 1 , which is mainly ascribed to the reduction of NiO to Ni, does not appear due to the transformation (on switching the potential at E a of more noble values) of NiO to species, which are reduced at more negative potential [15].  Figure 1, curves 3 and 4 represent the cyclic voltammetric behaviour of the alloys I and II, respectively. The cyclic voltammograms were characterized by the existence of well-defined potential regions (separated by a certain critical potential E crit ); the subcritical or selective dissolution potential region and the simultaneous dissolution potential region. The zero current potential of the two alloys are situated at potentials more negative than those of pure copper and pure nickel.

The Subcritical Potential Region
It seems that the dissolution of the alloys in this region follow the dissolution -redeposition mechanism where both copper and nickel dissolves initially followed by the redeposition of the more noble component, copper. This region was characterized by the appearance of the anodic peaks A 1 and A 2 . The anodic peak A 1 represent the situation where both of copper and nickel dissolves followed by copper redeposition on the alloy surface and the oxidation of nickel to NiO and/ or Ni(OH) 2 . It is seems that, taking into consideration the molar ratio of nickel in the alloy, the peak current density i p of the peak A 1 was increased on alloying nickel with copper. These results indicate the enhancement of the selective dissolution of nickel from the alloy with increasing the copper content. These results indicate that the dissolution of copper with nickel initially assists the diffusion of the nickel from the bulk of the alloy towards its surface and hence increases the peak current density of the anodic peak A 1 .By thermodynamic reasoning, it can be shown that this process is impossible unless, in accordance with the principles of irreversible thermodynamics, a coupling between the anodic partial reactions occurs, which may reflect an average activity of adsorbed ad-atoms of the electropositive component Cuthat exceeds unity [13]. If this situation prevails, Cu + ions may be dissolved from the alloy at underpotentials that are cathodic with respect to the reversible potential E Cu ' that agglomerates on the electrode surface. They have, therefore, the character of an unstable intermediate species and will be plated out on the surface of Cu crystals at the same electrode potential, which leads to the appearance of the anodic peak A 2 and formation of CuO. Cu x -Ni y + H 2 O →Cu x-1 -Ni y-1 .Cu redep + NiO + 2H + + 2e¯ A 1 [5] 2Cu x-1 -Ni y-1 .Cu redep + H 2 O →2Cu x-1 -Ni y-1 + Cu + + e¯ A 2 [6]

Simultaneous Dissolution Potential Region
This region lies at potential nobler than the critical potential E crit of the alloys, where simultaneous dissolution of both nickel and copper occur. The following observations can be made: (i) in alloy I (low nickel content) the surface losses its planarity and acquires the porous structure as shown by the hysteresis loop of the voltammogram on switching the potential in the cathodic direction and as seen from the SEMmicrograph in Fig. 2.The formation of surface porous structure was reported [16] for an active Ni dissolution in H 2 SO 4 at potentials -/0.25 VAg/AgCl by a step-flow mechanism, followed by the rapid formation of large three-dimensional etch pits, leading to considerable surface roughening.(ii) in alloy II (high nickel content) this region was characterized by the appearance of three anodic peaks A 3 and A 4 corresponding to the formation of CuO and Ni 2 O 3 , respectively.   The complementary relationship between the anodic and cathodic peaks was studied by sweeping the potential from E c = -1600 mV and reversing it at different E a values. Inspection of the data in the voltammogram of alloy I, Figs. 5 and 6 showed that on reversing E a at value more noble than the potential of the anodic peak A 1 two cathodic peaks C 1 and C 2 appear as a result of reduction of Ni(OH) 2 and NiO to Ni respectively. Reversing E a at value more noble than the potential of the anodic peak A 2 three cathodic peaks C 1 , C 2 and C 3 appear. The cathodic peak C 3 is related to the reduction of Cu 2 O to Cu. On reversing E a at value more noble than the potential of the anodic peak A 3 four anodic peaks C 1 , C 2 , C 3 and C 4 appear. The cathodic peak C 4 represents the reduction of CuO to Cu 2 O. On reversion the potential beyond the potential value of the anodic peak A 4 an new cathodic peak, C 5 , appears as a result of the reduction of Ni 2 O 3 to Ni +2 compounds.  Figures. 7 and 8 show the influence of the sweep rate on the electrochemical behaviour of the alloys I and II, respectively. The general shapes of the volammograms remained unchanged with increasing the sweep rate. It seems that, incrasing the scan rate increases the peak current densities i p of the anodic peaks and shifts their peak potentials E p as well as the critical potential E crit to more noble potentials. The voltammograms were used to analyze the dependence of peak current density i p and peak potential E p on the sweep rate. Figure 9 represents the linear dependence of the peak current density i p of the anodic peaks on the root of the scan rate, where lines passing by origin was obtained for the anodic peaks A 2 , A 3 and A 3 (alloy II) or slightly different than zero for the peaks A 1 and A 2 (alloy I) and A 1 (alloy II). The intercepts different than zero indicate that the induction rate is required before the electrode reactions become limited by diffusion. The linear dependence of the peak potential E p of the anodic peaks A 1 , A 2 , A 3 and A 4 of alloy II and A 1 of alloy I vs. log νis given in   In order to obtain more information about the electrochemical behaviour of copper-nickel alloys in Na 2 SO 4 solution, potentiostatic current/time transients were carried out at different anodic step potentials, E s . Figure 11 represents the current transients of alloy II in 0.1 M Na 2 SO 4 . It is seen that the current density decreases rapidly with time at first and then reaches a steady-state value. As the value of the potential is made more positive, both of the instantaneous and steady-state current values increase due to and increase in the thickness of the anodically formed layers. Plotting the current density vs. t -1/2 for the descending part of the current transients gives straight lines passing by the origin, Fig. 12. When E s at -500 mV, these is two slopes as one could be expect, owing to the formation of two different layers, NiO and Ni(OH) 2  layers are diffusion-controlled processes and obey the following equation [17,18] where P = z.c.D 1/2 /π 1/2 , where z is the number of exchanged electrons, c is the concentration of the electrolyte and D is the diffusion coefficient of the diffusing species. i = P / t 1/2

Conclusions
Two alloys of different composition are used to study the electrochemical behaviour of copper-nickel system sodium sulphate solutions using cyclic voltammetric, potentiodynamic anodic polarization and potentiostatic current \time transient techniques.
The forward sweep was characterized by the appearance of two well-defined potential regions separated by the critical potential E crit ; the selective dissolution potential region and the simultaneous dissolution potential region.
The dissolution of the alloys in the first region follow the dissolution -redeposition mechanism where both copper and nickel dissolves initially followed by the redeposition of the more noble component, copper. This region was characterized by the appearance of the anodic peaks A 1 and A 2 where Ni(OH) 2 and NiO, and CuO are formed respectively.
The simultaneous potential region was characterized by surface enlargement and the appearance of two anodic peaks A 3 and A 4 , which are related to the formation of Cu 2 O and Ni 2 O 3 , respectively.
The SEM micrograph showed surface enlargement and pits formation.
The reverse sweep was characterized by the appearance of five catholic peaks C 5 , C 4 , C 3 , C 2 and C 1 , which are related to the reduction of Ni 2 O 3 to Ni +2 componds, CuO to Cu 2 O,Cu 2 O to Cu,NiO to Ni and Ni(OH) 2 to Ni, respectively.
Potentiostatic current/time transient measurements reveal that the formation of Ni(OH) 2 , NiO, Cu 2 O, CuO and Ni 2 O 3 layers involves a nucleation and growth mechanism under diffusion control.