Effect of Copper Alloying on Electro-Catalytic Activity ... Effect of Copper Alloying on Electro-Catalytic Activity of Nickel for Ethanol Oxidation in Alkaline Media

In this research, the electro-catalytic activity of nickel-copper (Ni-Cu) alloy towards oxidation of ethanol and its possible redox process were investigated in alkaline solution. For this purpose, cyclic voltammetry, chronoamperometry and electrochemical impedance spectroscopy techniques were employed. According to the cyclic voltammetry studies, Ni-Cu alloy compared to pure nickel can demonstrate a significantly higher response for ethanol oxidation. So, the enhance-ment of the anodic peak current corresponding to the oxidation of nickel hydroxide was accompanied with attenuated cathodic current in the presence of ethanol. The anodic peak currents have a linearly dependence on the square root of scan rate which is the characteristic of diffusion-controlled processes. Based on the chronoamperometry measurements, the reaction exhibited a Cottrellian behavior and the diffusion coefficient of ethanol was found to be 1.26 × 10 –5 cm 2 s –1 . The impedance spectroscopy declared electro-catalytic behavior of Ni-Cu electrode for oxidation of ethanol and showed that the charge transfer resistance decreases by increasing the ethanol concentration.


Introduction
Recently, direct ethanol fuel cells (DEFCs) have attracted much interest for different applications. 1,2 The reason is that they can provide convenient operation, storage and distribution. However, DEFCs require further development compared to hydrogen based fuel cells. 3 One of the unresolved problems of DEFCs is slow anodic rate of ethanol oxidation. 4 In this respect, considerable researches have been devoted to study ethanol electro-oxidation at high pH values. Utilizing alkaline solution has many advantages in fuel cells. For example, it can enhance fuel cell efficiency, reduce corrosion, enable the application of many electrode materials, promote efficiency of the processes occured at both anode and cathode, eliminate sensitivity to the surface structure and decrease poisoning effects. [5][6][7][8] In electrochemical oxidation of ethanol, selection of an appropriate material for the anode is very crucial for obtaining an electro-catalyst of high efficiency. Some studies have reported a significant increase in fuel utilization and power density through optimizing different factors related to fuel cells. 9,10 For the electrode, different materials have been employed to catalyze the electrochemical oxidation of ethanol. 11,12 One of the well-established electrode materials is nickel which poses proper surface oxidation properties. Nickel and its complexes have been commonly applied in electro-catalysis to proceed both anodic and cathodic reactions in water electrolysis and organic synthesis. [13][14][15][16][17] One of the remarkable applications of nickel is catalyzing the ethanol oxidation. Several studies regarding to electro-oxidation of alcohols on nickel electrode have been reported. [18][19][20] For obtaining a synergistic electro-catalytic system, the nickel alloys specifically nickel-copper can be used. [21][22][23] In addition, alloy electrodes compared to other electro-catalysts can offer further advantages such as long-term stability and ease of preparation. The crystal structures of pure Ni and Cu metals are similar and they possess face-centered cubic structures with similar lattice parameters. Therefore, it is possible to make a wide range of Ni-Cu alloy ratios. [24][25][26][27] The aim of present work is to study the electrochemical oxidation of ethanol on Ni-Cu alloy (70/30) electrode in alkaline solution and compare its catalytic activity with pure Ni electrode via electrochemical techniques of cyclic voltammetry, chronoamperometry and impedance spectroscopy.

Experimental
All chemicals used in this work [Sodium hydroxide and ethanol in analytical grade] were of Merck origin (Germany) and used without further purification. Double distilled water was employed to prepare the solutions. Nickel and nickel-copper alloy were prepared from Rooyingran Sanaat Company. The electrochemical measurements were carried out in a one-compartment three-electrode cell powered by a Metrohm-Autolab potentiostat/ galvanostat (model 12/30/302, The Netherlands). The disk of Ni-Cu alloy (70/30) of 1 cm 2 area was employed as working electrode. Before each measurement, the electrode was polished with emery paper of 1000 grit and rinsed in double distilled water and acetone. Counter and reference electrodes were platinum and KCl-saturated Ag/ AgCl electrode (Metrohm, The Netherlands), respectively. All the electrochemical studies were done at 22 ± 1 °C.
The electrochemical impedance experiments were carried out in a frequency range of 100 kHz to 10 mHz and modulation amplitude of 10 mV with respect to open circuit potential. The experimental impedance spectroscopy data were fitted to the proposed equivalent circuit using a home written least square software. This software was programmed according to the method of Marquardt for optimization of functions and Macdonald weighting for the real and imaginary parts of the experimental impedance data. 28,29 Figure 1 displays consecutive cyclic voltammograms of the Ni-Cu electrode in 1 M NaOH solution recorded after 50 cycles at scan rate of 100 mV s -1 . Upon the first scan, a pair of redox peak is observed at 463 and 295 mV which is attributed to the Ni 2+ /Ni 3+ redox couple. In the subsequent cycles, the anodic and cathodic peaks shift to the negative direction and stabilize at 378 and 310 mV, respectively. This behavior is consistent with the data reported in previous electrochemical studies related to formation of the nickel oxide film on the surface and inter-conversion of α-Ni(OH) 2 and ß-phases, conversion of Ni(OH) 2 to NiOOH and enrichment of Ni 3+ species around the surface of the electrode. 30,31 During the next cycles, a negative shift of the anodic peak and its stabilization are evident, which refer to higher potential energies required for NiO-OH nucleation. In the first cycle, the baseline current is higher which corresponds to the oxidation of Ni to Ni 2+ . An increase in peak currents with the number of cycles demonstrates a continuous enrichment of electrode's surface by accessible Ni 2+ and Ni 3+ electro-active species. Figure 2 shows the cyclic voltammograms of the Ni, Cu and Ni-Cu electrode in 1 M NaOH solution recorded at scan rate of 10 mV s -1 . As can be seen, the oxidation behavior of Ni and Ni-Cu electrode is the same and the Ni 2+ /Ni 3+ redox couple is observed. No considerable anodic peak for Cu 3+ is shown in cyclic voltammogram of Cu electrode, but the cathodic peak is clear in inset of Figure 2.  Figure 3a illustrates the typical cyclic voltammograms of a Ni-Cu electrode at various scan rates (2-2500 mV s -1 ) in 1 M NaOH solution. Figure 3b shows that the anodic peak currents increase proportional to the lower scan rate values (2-200 mV s -1 ). This behavior is expected for the electrochemical activity of redox couples that their voltammetric responses are sensitive to the low concentration of surface-confined electro-active species. 32 In this process, only the nickel oxide layer produced on electrode surface participates in the redox reaction. 33 Surface coverage of the redox species, G * , can be calculated according to the slope of the lines shown in Figure 3b: 33 (1) where I p , n, and ν is peak current, electron transfer number and potential scan rate, respectively. The G * value is calculated to be about 8.15 × 10 -8 mol cm -2 which is related to the presence of ca. 80 monolayers of surface species on Ni-Cu electrode. Figure 3c presents the proportionality of anodic peak current to square root values of higher scan rates (350-2500 mV s -1 ). It shows that the oxidation reaction is a diffusion-controlled process at higher scan rate 33 and the reaction is limited by diffusion of hydroxide ion from bulk of solution to the electrode surface according to following equation:

Results and Discussion
(2) trode is high for ethanol oxidation, but it is at higher anodic over-potentials. As it can be seen, the measured anodic potentials of Ni-Cu and Ni electrodes are the same, but the anodic peak current of Ni-Cu is higher than that in Ni electrode. Consequently, the high electro-catalytic activity of Cu electrode is responsible for electro-catalytic activity of Ni-Cu electrode.    In the positive sweep, the measured anodic current is proportional to the bulk concentration of ethanol. An increase in ethanol concentration up to 0.6 M caused a linear increase in the anodic current (Figure 5b). Based on these evidences, catalytic electro-oxidation of ethanol on Ni-Cu electrode is confirmed. Zhang et al. investigated the ethanol oxidation on Ni-B amorphous alloy nanoparticles modified nanoporous Cu in alkaline medium. 34 They reported that ethanol oxidation at the Ni-B modified porous Cu electrode had a negative shift of 52 mV in the onset oxidation potential and the oxidation peak current increased in comparison with the bulk Ni. Kakaei and Marzang studied the electro-catalytic activity of nitrogen doped reduced graphene oxide with Ni-Co nanoparticles for ethanol oxidation in alkaline media. 35 The fabricated alloy electro-catlyst exhibited a remark-able electro-catalytic activity and high stability for the ethanol oxidation reaction in comparison with fabricated Ni and Co.
The decrease in cathodic current that occurs in the negative potential scan verifies the involvement of ethanol in the rate determining step. It also indicates that the process is incapable of reducing all high-valence nickel species that have formed in the anodic half cycle. Along with the oxidation of Ni 2+ species to Ni 3+ , the adsorbed ethanol molecules oxidize on the surface at higher over-potentials. The products or intermediates of the reaction poison the electrode surface and reduces the number of available sites for ethanol adsorption. Consequently, the anodic current approaches a maximum in the positive potential scan and then the overall rate of ethanol oxidation decreases. Electro-catalytic oxidation of ethanol also continues in the early stages of the cathodic half cycle and the current tends to a maximum since some active sites for adsorption of ethanol regenerate due to removal of the adsorbed intermediates and products. Figure 6 shows the cyclic voltammograms of Ni-Cu electrode in the presence of 0.5 M ethanol at various potential scan rates (2-350 mV s -1 ) and also the scan rate proportionality of the anodic peak current. The anodic peak potential appears at 0.5 V as a result of ethanol oxidation on the nickel active sites. The variation of anodic peak current vs. the square root of scan rate values shows a linear relationship which represents that the oxidation of ethanol on Ni-Cu electrode is controlled by diffusion of ethanol species from solution to the redox sites accessible on the electrode's surface (Figure 6b). Although, the cathodic peak of Ni 3+ reduction is negligible at low scan rates, but it appears at higher scan rates. This observation implies that the electro-oxidation of nickel species is much faster than catalytic oxidation of ethanol at higher scan rates. Therefore, oxidation of ethanol on nickel surface is a slow process.
At higher scan rates, the oxidation peak of Ni(OH) 2 to NiOOH rises at potentials that are considerably more negative than the potential of ethanol oxidation (≈ 370 mV). This peak is insignificant at low scan rates but enhances at higher scan rate values. Figure 6c illustrates a linear dependency of the current peak on the scan rate which proposes the presence of surface-confined electro-active species. According to the high current density observed in the presence of ethanol and also the appearance of two oxidation peaks for Ni 2+ and ethanol, it is appeared that one of the anodic current can be attributed to the oxidation of ethanol by NiOOH since the NiOOH reduction peak disappears during the negative scan. The other one can be assigned to the direct electro-oxidation of ethanol on the surface of the oxide layer.
The following redox reaction of the nickel species is expected: while ethanol oxidizes on the alloy surface through the following reaction 23,36,37 : In Eqs. (4) and (5), ethanol oxidation occurs through the reduction of NiOOH sites. The Ni 3+ sites can be regenerated by the power source or via direct electro-oxidation on Ni 3+ oxide surface as follows: Ni 3+ -intermediate → Ni 3+ -product + e (7)  In Eqs. (6) and (7), Ni 3+ provides an active site for direct ethanol oxidation. Also, Eqs. (3), (6) and (7) refer to two anodic peaks of ethanol and Ni 2+ oxidation. Figure 7a displays the chronoamperograms recorded for Ni-Cu in 1 M NaOH solution containing (0.1-0.8 M) ethanol at the potential step of 500 mV. Figure 7b identifies a linear behavior of the net current changes (after elimination of the background current) vs. the inverse square root of time. So, a diffusion-controlled process is dominant in this electrochemical process. The diffusion coefficient of ethanol is calculated 1.26 × 10 -5 cm 2 s -1 by substitution of the line slope of Figure 7b into Cottrell equation 33 (Eq. 8).
The catalytic rate constant of ethanol oxidation on Ni-Cu alloy is evaluated according to Eq. (9) which is introduced by Pariente et al.: 38 (9) Where I catal and I d stand for Ni-Cu electrode currents in the presence and absence of ethanol, respectively. y = kCt is the related error function in which k is the catalytic rate constant, C is the bulk concentration of ethanol and t is the elapsed time. When y〉1.5, erf (y 1/2 ) equals to unity approximately and Eq. (9) simplifies to:    Figure 9 shows the equivalent circuit in the presence of ethanol which is compatible with Nyquist diagram. The capacitor C should be replaced with a constant phase element (CPE), Q, in the equivalent circuit. The CPE behavior is due to the microscopic roughness of the electrode that causes an inhomogeneous distribution in the solution resistance and the double-layer capacitance. 41,42 In Figure  9, R s , Q dl and R ct are solution resistance, a constant phase element related to the double layer capacitance and charge transfer resistance, respectively. Also, Q ads and R ads are the electrical elements corresponding to the adsorption of reaction intermediates.
The experimental data were fitted to the equivalent circuit to obtain the circuit elements. Table 1 represents the equivalent circuit parameters for the impedance spectra of ethanol electro-oxidation. The charge transfer resistance value decreases as the ethanol concentration increases. It approves the catalytic activity of Ni-Cu alloy for ethanol oxidation. n 1 and n 2 are the constant phase element exponent and show the extent of difference between capacitance and the constant phase element.

Conclusions
In this study, the electro-oxidation of ethanol was investigated on the nickel oxide film formed on the surface of Ni-Cu alloy electrode in an alkaline solution. The results confirmed that Ni-Cu alloy is electro-catalytically active for ethanol oxidation at a potential of ≈ 500 mV. The catalytic response obtained for electro-oxidation of ethanol on Ni-Cu electrode is greater than the one observed for pure nickel. The anodic peak currents for ethanol oxidation are linearly proportional to square root of scan rates. The chronoamperograms revealed the dominance of a diffusion-controlled process in presence of ethanol species. The diffusion coefficient of ethanol was calculated 1.26 × 10 -5 cm 2 s -1 . The charge transfer resistance for various concentrations of ethanol was evaluated and highlighted an effective catalytic activity of Ni-Cu alloy for oxidation of ethanol in alkaline solutions.