The electrochemistry of nanostructured Ni–W alloys

This work reports on the features that Ni–W nanostructured alloys, electrodeposited on carbon steel by different current pulse programs, may present depending on their surface morphology and surface composition. The Ni–W nanostructured coating, with a cauliflower structure, lack of fragility, and high WO3/W surface composition ratio, is a stable electrode to catalyze hydrogen evolution reaction, exceeding bulk and electrodeposited Ni catalytic activity. Also, the nanostructured alloys must have a low WO3/W surface composition ratio for Ni and its oxides to provide protection and improve corrosion resistance in sulfate media.

Prediction of the electrochemical behavior of nanocrystalline materials from the properties of their coarsegrained and amorphous analogs is not straightforward because of the differences in their surface characteristics. The interdependence of the parameters mentioned before is thus confirmed [27].
In our previous work, we selected a neutral electrolyte containing sulfate anions to demonstrate that Ni-W coatings protected the carbon steel substrate from pitting [11,28]. We interpreted the electrochemical results taking into account the contribution of the coating components [11].
Our research is currently focused on studying how surface morphology and composition of current pulse formed Ni-W alloys (particularly concentration ratios between metallic components and their oxides) affect corrosion resistance and hydrogen evolution reaction (HER) catalysis in the aforementioned electrolyte. Bulk and electrodeposited Ni are also presented for the sake of comparison. polished with sandpaper, decreasing in grit size from 80 to 2500, followed by 0.3 μm alumina powder and finally rinsed with twice-distilled water.

Experimental
The electrodeposited coatings were obtained galvanostatically by pulse electroplating using a Solartron system 12508W, which includes a frequency response analyzer 1250 and an electrochemical interface 1287. The pulse scheme consisted of an "on" time (t on ) during which a cathodic current of 70 mA cm −2 was applied and an "off" time (t off ) during which a zero current was applied, where t0t on 0t off , with 5×10 −3 s≤t≤ 120 s. The deposition time (t dep ) was 60 min.
The plating bath contained 0.06 M NiSO 4 ·6H 2 O, 0.14 M Na 2 WO 4 ·2H 2 O, 0.5 M Na 3 C 6 H 5 O 7 ·2H 2 O, 0.5 M NH 4 Cl, and 0.15 M NaBr (pH09.5). A fresh plating bath was prepared for each experiment using chemical pure reagents and twicedistilled water. During the plating, the solution was gently stirred with a magnetic stirrer at 65°C and degassed with purified nitrogen. Nickel coatings were obtained from the same bath, but in absence of the tungstate salt.
Deposit morphology was analyzed using an FEI, Quanta 200 scanning electron microscope (SEM). Surface composition was evaluated by X-ray photoelectron spectroscopy (XPS) using a Mg Kα source (1,253.6 eV) from XR50, Specs GmbH and a hemispherical electron energy analyzer from PHOIBOS 100, Specs GmbH. Composition in depth was determined by argon ion sputtering (3 keV) at an average rate of 0.5-0.6 nmmin −1 .
The electrochemical behavior of the Ni-W and Ni deposits and bulk Ni was studied with a Zahner IM6e potentiostat/ galvanostat using a potentiodynamic linear polarization technique between preset cathodic (E s,c ) and anodic (E s,a ) switching potentials from the open circuit potential, at a potential scan rate (v) of 2×10 −3 V s −1 . A standard threeelectrode cell was used. A large area Pt sheet counter electrode and a saturated calomel reference electrode (SCE) were used. All potentials in the text were referred to the SCE. Measurements were carried out in a still phosphateborate buffer (0.1 M KH 2 PO 4 +0.05 M Na 2 B 4 O 7 ) pH8.00, with the addition of 1 M Na 2 SO 4 at ambient temperature. Experiments were made under purified N 2 gas saturation.
The effect of oxygen on the open circuit potential (E ocp ) of the Ni-W alloys was evaluated after 1 h of immersion in the electrolyte. Corrosion current density values were obtained using the polarization resistance method [29] after 1 h and 24 h of immersion in the test solution.

Results and discussion
XPS spectra were obtained to focus on the surface chemistry of the W-Ni alloy electrodeposited on carbon steel by different current pulse programs. Typical spectra for these alloys (t 0 120 s, j070 mAcm −2 , t dep 060 min) are shown in Fig. 1 before (a) and after (b) sputtering with Ar + . The spectra show the presence of Ni, W, C, and O and the complete absence of Fe indicating that the Ni-W coating covers the carbon steel substrate completely. In fact, the thickness of these coatings is between 10 and 20 μm [11].
Data in Fig. 1 also show that the amount of C and O decreases while the Ni and W content markedly increases after sputtering the sample. The carbon 1s component at 283.4 eV (Fig. 1) suggests the presence of some tungsten carbide (WC) [11,24,30]. However, the presence of tungsten carbide could also result from implantation of carbon during the Ar + ion etching of the coating [26]. The presence of O is related to metal oxides on the coating surface. A similar behavior is observed for samples grown for t values ranging from 5×10 −3 to 120 s.
A more detailed analysis of the Ni and W species in the electrodeposited alloy is shown in Fig. 2 where we plotted the Ni 2p and W 4f signals for different sputtering times. The highresolution spectra reveal that the alloy surface (t00 min) consists of a mixture of Ni oxides (NiO and NiOOH) and W oxide (WO 3 ). As the sputtering time increases, the amount of Ni and W oxides decreases while the metallic Ni and W content markedly increases. Note that the O content is reduced to a non-significant and constant level during the sputtering process. The metallic oxides are completely removed for t045 min.
The XPS data in Table 1 show that the W/Ni ratio for the bulk alloy (sputtering time 45 min), (W/Ni) b , remains nearly constant (≈0.3-0.4) irrespective of t. In contrast, the XPS data for the alloy surface (t00 min) reveal more differences in the oxide composition with the electrodeposition conditions (Fig. 3a-c). In fact, the XPS data show that the WO 3 /W ratio at the surface, (WO 3 /W) s , decreases from ≈0.57 to 0.23 as t is decreased. This is an important point since the electrochemical properties of the alloy are determined by its surface composition. The influence of t on the surface structure of the Ni-W alloys may be inferred from the SEM images (Fig. 4) that reveal the presence of a fractal cauliflower-like structure formed by nanometer-sized units irrespective of t. It means that the real surface area is too large with pores and channels interconnecting the structure.
Next, we present data about alloy electrochemical behavior to show its relation with surface chemistry and structure. Typical polarization curves for the Ni-W alloys recorded in the −0.7/0.5 V potential range and in the absence of O 2 are shown in Fig. 5. All the alloys exhibit the hydrogen evolution reaction (HER) in the cathodic branch and a limiting current in the anodic branch corresponding to the alloy dissolution (j corr ) through the complex passive film composed by NiO, NiOOH, and WO 3 . HER on the Ni-W alloy is clearly enhanced in overpotentials (η) and current density (j H ) when compared with bulk Ni (Fig. 5a, b). In fact, the onset potential for HER on the Ni-W alloy surfaces is shifted to more positive potentials than that of bulk Ni. It can be argued that this enhancement could be produced by the presence of a certain amount of WC that is known to be an efficient HER catalyst [31]. However, our data show very similar η values for HER on electrodeposited Ni samples and Ni-W alloys prepared under comparable conditions (t 05 × 10 −3 s), thus discarding any possible effect of WC (Fig. 5a). This is not surprising as electrodeposited Ni is a nanostructured material and a clear increase in HER has been reported on this kind of Ni surfaces [32]. In fact, nanostructured materials exhibit different properties of their bulk phases due to the quantum effects related to the shape and size. Therefore, it can be concluded that the nanostructured cauliflower structure of the alloy surface contributes to HER enhancement.
We also observe lower η values, and then a clear HER enhancement (Fig. 5c) on the Ni-W alloys surfaces richer in WO 3 (t 010 and 120 s, see Table 1). In relation to this subject, Savadogo et al. [33] have studied the HER using nickel electrodes with phosphotungstic acid. It has been  demonstrated that the presence of W in the form of WO 3 in the polyoxometalates enhanced the electrocatalytic activity for hydrogen evolution [33,34]. Pt, when supported on tungsten trioxide, also showed electrocatalytic activity for HER due to the synergism towards reactions in an acid solution involving hydrogen atoms [35]. As previously shown, electrocatalytic activity changes markedly with the W/Ni ratio, being W/Ni00.1 the best atomic ratio [36]. Our results, obtained from an almost constant W/Ni ratio (0.3/ 0.4) and similar surface structure, demonstrate that the WO 3 content at the surface is the key parameter controlling the HER electrocatalysis and reducing η about 0.1 V (Fig. 5c).
From the previous results, we can conclude that there are two aspects involved in the HER enhancement observed in the electrodeposited Ni-W alloys on steel: a nanostructured material and the chemical composition of the surface. They explain the catalytic activity on HER observed for WO 3 nanoparticles supported on C in acid and alkaline solutions [37].
However, coatings obtained at t0120 s are fragile while those prepared at t010 s exhibit good mechanical properties [11]. Therefore, the best condition related to HER electrocatalysis and Ni-W mechanical properties is obtained at t010 s. A nanostructured alloy with the optimum (WO 3 /W) s surface composition ratio is obtained in this condition.
When the anodic branch of polarization curves is analyzed (Fig. 5a, b), we observe that the nanostructured Ni-W alloys exhibit much larger dissolution current values than those of bulk Ni because of the presence of W and large  amounts of grain boundaries that make the film more defective. The tendency towards reaching a limiting current is due to the initial formation of protective NiO and NiOOH. This is in agreement with what was reported by other authors who demonstrated that at the same pH, Ni electrode passivity increases progressively as E s,a becomes more positive [38,39]. In contrast, surface W and WO 3 tend to dissolve, giving tungstate ions [40][41][42] during the anodic potential sweep with the corresponding Ni oxide surface enrichment.
Only at E s,a more positive than 0.2 V passivating tungsten oxides are known to be formed [41]. Other authors proposed other interpretations because they mainly used acid solutions containing chloride ions [4, 6, 8, 10-14, 21, 25]. Besides, few authors highlight the effect that the surface crystallographic condition exerts [13,14]. The effect of oxygen in the open circuit potential value (E ocp ) of the alloys is shown in Fig. 6. The initial values lie in the potential windows where alloy dissolution takes place at 30<j corr <100 μAcm −2 as deduced from the polarization curves shown in Fig. 5c.
However, the dissolution rate is markedly reduced after 24-h immersion in the test solution as shown in Table 2. The largest change is observed in the deposits obtained at t05× 10 −3 s since the (WO 3 /W) s ratio is smaller than in those obtained at higher t (Table 1). Hence, the enrichment and subsequent growth of Ni oxides is facilitated. The characteristic of the surface structure already shown and analyzed through SEM images (Fig. 4a) should be added.
Finally, even though the anodic branches of the polarization curves (Fig. 5a, b) indicate that bulk and electrodeposited Ni responses are better than those of the Ni-W deposits, the latter present larger values of hardness [5,11], a valuable characteristic for technological applications.
From what has been analyzed above, we may infer that through the previous selected current pulse programs to obtain electrodeposited Ni-W alloys, we can handle the surface state (morphology and composition ratio among Ni, W, and their corresponding oxides), according to the desired application for the product since not any Ni-W coating, even if nanostructured, can be adequate to catalyze HER or to withstand corrosion in solutions containing aggressive ions.
These results are a good starting point to research the electrocatalytic activity for HER that these nanostructured Ni-W alloys present in NaOH solutions to simulate industrial electrolysis conditions and to evaluate corrosion resistance in solutions containing other aggressive anions (sulfide and chloride).

Conclusions
Ni-W alloys electrodeposited on carbon steel by different current pulse programs were obtained.
All the Ni-W alloy surfaces exhibited a fractal cauliflowerlike structure irrespective of t. It means that the surface has a large real surface area approaching a volume with a large number of pores and channels. However, the coatings obtained at low frequencies showed brittleness.
The surface composition ratio, (WO 3 /W) s , of the Ni-W electrodeposits defines if the alloys may act as HER catalyzers or be corrosion resistant in neutral media containing sulfate ions. It was possible to demonstrate that the Ni-W nanostructured coating, with a large effective area, lack of fragility, and higher (WO 3 /W) s ratio, offers an optimal condition to work as a stable electrode to catalyze HER, thus exceeding the catalytic activity that bulk and electrodeposited Ni have.
To improve the corrosion resistance in neutral media containing sulfate anions, a lower (WO 3 /W) s ratio is needed so that Ni and its oxides may provide protection. This situation is obtained for coatings obtained at low t.