Journal of Materials Science & Technology  2019 , 35 (10): 2144-2155 https://doi.org/10.1016/j.jmst.2019.05.051

Orginal Article

Low-valence ion addition induced more compact passive films on nickel-copper nano-coatings

Quangquan Doae, Hongze Ana, Guozhe Mengab*, Weihua Lib*, Lai-Chang Zhangd, Yangqiu Wanga, Bin Liua, Junyi Wanga, Fuhui Wangac

a Corrosion and Protection Laboratory, Key Laboratory of Superlight Materials and Surface Technology (Harbin Engineering University), Ministry of Education, Harbin, 150001, China
b Corrosion and Protection Institute, School of Chemical Engineering and Technology, Sun Yat-Sen University, Zhuhai, 519082, China
c Corrosion and Protection Division, Shenyang National Laboratory for Material Science, Northeastern University, Shenyang, 110819, China
d School of Engineering, Edith Cowan University, 270 Joondalup Drive, Joondalup, Perth, WA, 6027, Australia
e Shipbuilding Faculty, Viet Nam Maritime University, Haiphong, Viet Nam

Corresponding authors:   * Corresponding authors at: Southern Marine Science and Engineering GuangdongLaboratory (Zhuhai), 519082, China. E-mail addresses: mengguozhe@hrbeu.edu.cn (G. Meng),liweihua3@mail.sysu.edu.cn (W. Li).* Corresponding authors at: Southern Marine Science and Engineering GuangdongLaboratory (Zhuhai), 519082, China. E-mail addresses: mengguozhe@hrbeu.edu.cn (G. Meng),liweihua3@mail.sysu.edu.cn (W. Li).

Received: 2019-02-19

Revised:  2019-03-4

Accepted:  2019-05-24

Online:  2019-10-05

Copyright:  2019 Editorial board of Journal of Materials Science & Technology Copyright reserved, Editorial board of Journal of Materials Science & Technology

More

Abstract

Ni-Cu nano-coatings were prepared by pulsed electroplating technique in the baths containing various amount of boric acid. Their microstructure, morphologies and corrosion resistance were characterized in detail. The addition of boric acid strongly influences on the microstructure of the Ni-Cu coatings. The coating with a grain size of 130 nm, obtained from the bath containing 35 g L-1 boric acid, shows the highest corrosion resistance. This is attributed to the low-valence Cu ion (Cu+) additions in nickel oxide, which could significantly decrease the oxygen ion vacancy density in the passive film to form a more compact passive film. The higher Cu+ additions and the lower diffusivity of point defects (D0) are responsible for the formation of more compact passive film on the coating obtained from the bath with 35 g L-1 boric acid.

Keywords: Passive film ; Point defect ; Corrosion ; Electroplating ; Coating

0

PDF (6440KB) Metadata Metrics Related articles

Cite this article Export EndNote Ris Bibtex

Quangquan Do, Hongze An, Guozhe Meng, Weihua Li, Lai-Chang Zhang, Yangqiu Wang, Bin Liu, Junyi Wang, Fuhui Wang. Low-valence ion addition induced more compact passive films on nickel-copper nano-coatings[J]. Journal of Materials Science & Technology, 2019, 35(10): 2144-2155 https://doi.org/10.1016/j.jmst.2019.05.051

1. Introduction

Thanks to the excellent resistance to corrosion, nickel and nickel alloys are promising alternatives to bulk stainless steels for applications in harsh environment [[1], [2], [3], [4]]. For example, 70Ni-Cu alloy has high corrosion resistance in marine environments [5]. However, the service life expectancy of equipments will be longer for full cycle cost and materials sustainable development, which requires that these alloys have higher corrosion resistance to the environment. In order to enhance the corrosion resistance of passive metals, nanocrystallizations have been applied on pure Ni and pure Cu coatings [[6], [7], [8], [9]]. The findings showed that nanocrystallization could markedly improve the corrosion resistance of the passive metals. Such nanostructures were beneficial to enhancing the electrochemical activity of passive elements to form more perfect film. Recent investigations have shown that passive metals in alloys have superior corrosion resistance. Lv et al. [10] claimed that the corrosion resistance of Ni-Fe nanostructured coating was markedly higher than that of pure Ni electrodeposition coating. Ni-W nanostructured coating showed higher resistant to corrosion than pure Ni coating in borate buffer solution and 3.5 wt% NaCl solution [11]. North and Pryor [12] reported the anti-corrosion superiority of the Cu-Ni alloy over that of pure copper, indicating that the Cu2O formed on the Cu-Ni was more protective than that formed on Cu. They supposed that nickel substitution into the Cu2O film, ignored cation vacancies, and then reduced the cation vacancy concentration and thus improve the corrosion resistance. Then, such an idea occurs to us whether nanocrystallization could enhance the electrochemical corrosion resistance of Ni-Cu alloys and how about the mechanism.

Nanostructured coatings are often prepared by different techniques, including plasma, thermal spraying, chemical and physical vapor deposition [[13], [14], [15], [16]]. Compared with these techniques, electroplating deposition is simple to operate and economical in use [[17], [18], [19]]. Electroplating can be prepared with DC (direct current), PCR (pulse current reverse) and PC (pulse current) techniques [19]. PC or PCR techniques were reported to generate nanostructured coatings with outstanding properties [20]. Gu et al. [21] fabricated a serial of nanostructured Ni films using three electrodeposition modes with different duty cycles and temperatures. The current research group has deposited nanostructured nickel coatings with high density of twins by PC electroplating technique. The results showed that nanostructured twins could greatly enhance their corrosion resistance by modifying the microstructure, semiconducting characteristics of the passive film [22,23].

On the other hand, additives are reported to have significant effect on the grain refinement of metal or alloy coatings. For example, Moti et al. [24] found that the addition of saccharin could well manipulate the grain size of nanocrystalline Ni. Meng et al. [25] reported that phytic acid supplementation is beneficial for the growth of nanostructured twins. Boric acid is often used as a buffer agent to control the pH decrease due to OH- formation in the part-electrolyte interface [[26], [27], [28]], which could markedly affect the electrodeposition process. It was reported that the grain nanocrystallization of an alloy is in favor of the enrichment of passive elements in the passive film by the express access of enormous grain boundaries [29]. However, it is unknown what effect boric acid as an additive will have on the microstructure and the corrosion behavior of Ni-Cu coating by pulse current technique.

Therefore, the objective of this work was to prepare Ni-Cu coating in sulfate bath containing various amount of boric acid by pulse electroplating technique. Then the microstructures of these coatings and the composition of passive films were characterized in detail. The electrochemical corrosion behavior by electrochemical measurements in 0.3 M NaCl solution was investigated. The values of lower diffusivity of point defects (D0) in the passive films formed on the coatings were quantified to evaluate the effects of boric acid on the microstructure and corrosion behavior of Ni-Cu coatings.

2. Experimental

2.1. Preparation of coatings

The anode was a graphite bar with 60 mm in length and 13 mm in diameter. The coatings were deposited on Q420 steel (standard GBT, China) with a chemical composition (wt%): C ≤ 0.18, Si ≤ 0.55, Mn 1.00-1.60, S 0.025, Cr ≤ 0.30, Ni ≤ 0.70, Mo ≤ 0.20 and Fe balance. The steel samples (20 mm × 30 mm × 4 mm) were ground with waterproof abrasive papers up to 2000 grits. The Ni-Cu alloy coatings were made on the steel samples from the various electrolytes, which contained 170 g L-1 NiSO4·6H2O, 45 g L-1 NiCl2·6H2O, 2.5 g L-1 CuSO4·5H2O, 15, 25, 35 or 45 g L-1 H3BO3, 105 g L-1 Na3C6H5O7·2H2O, 0.6 g L-1 C7H5NO3S and 0.2 g L-1 CH3(CH2)11OSO3Na. Electrodeposition was performed by a pulse current with a current density of 6 A dm-2, duty cycle of 40%, temperature at 65 °C and total plating time of 30 min. Before electroplating, the anode and cathode were all degreased in acetone and deionized water with an ultrasonic vibration generator for 20 min.

2.2. Morphology and microstructure characterization of coatings

The surface of sample after electroplating was washed with distilled water and then dried. The phase constituents of the coatings were examined by X-ray diffraction (XRD) in X’Pert Pro with monochromatic CuKα sources. The surface and cross-section morphologies of these coatings were characterized by a SIGMA 500 scanning electron microscope (SEM) in a secondary electron mode.

The coatings were peeled from the steel matrixes after electroplating. Afterward, coatings were thinned by ion milling for transmission electron microscopy (TEM) observation on a JEM 2100 microscope operated at 200 kV.

2.3. Electrochemical measurements

The prepared coatings were washed with distilled water, dried with hot air and connected with conducting wires. Then, they were sealed by silicone rubber (Nanda China) with an exposed area of 10 mm × 10 mm. The measurements of electrochemical were carried out in 0.3 M NaCl solution at 25 ± 1 °C, using a Zahner IM6ex, with a three-electrode cell containing 250 mL test solution, where the working, counter and reference electrodes were Ni-Cu coating, Pt plate (20 mm × 15 mm × 1 mm) and Ag/AgCl (saturated KCl), respectively. Before electrochemical tests, the working electrodes were initially polarized at -1.0 V vs. SHE (standard hydrogen electrode) about 2 min for removing the oxide films formed in the air, and then stabilized for 20 min at open circuit potential (OCP). The potentiodynamic polarization experiments were performed at a scan rate of 0.333 mV/s from -300 mV vs SHE to 1600 mV vs. SHE. The electrochemical impedance spectroscopy (EIS) measurements were carried out by a sinusoidal signal with an amplitude of 10 mV (rms) over a frequency range from 100 kHz to 10 mHz. Before Mott-Schottky (M-S) relationship test, the coatings were pre-polarized for 30 min at 75, 100, 125, 150, 175 and 200 mV vs. SHE, respectively, in the test solution.

2.4. XPS measurements

X-ray photoelectron spectroscopy (XPS) was used for analyzing the coating on an AlKα radiation. Quantitative analyses of XPS spectra were performed by XPSPeak4.1 software on the basic of the fitting of peak areas to estimate the relative atomic ratio of the nickel and copper in passive films on the coatings. A general formulation for determining the atom fraction of any constituent in a passive film, CA, which is given by [30]:

CA = NANi = (IA/SA)/Σ(Ii/Si) (1)

where Ii is the area of the peak generated by element i, Si is its relative sensitivity factor and Ni is its number of moles.

3. Results and discussion

3.1. Surface morphology and microstructure of coatings

Fig. 1 presents the SEM surface morphologies of the Ni-Cu coatings synthesized from the various electrolytes containing 15, 25, 35 and 45 g L-1 boric acid, respectively. As seen in the figure, all the coatings are uniform and exhibit features from velvet flower-like to cauliflower-like morphology with increasing the boric acid concentration in the bath. The cross-section morphologies of the four coatings are shown in Fig. 2. The thicknesses are 23.7 μm, 22.0 μm, 22.5 μm and 26.8 μm for the coatings synthesized from electrolyte with 15, 25, 35 and 45 g L-1 boric acid, respectively. All coatings are lack of visible defects or voids (Fig. 2(a)-(d)). However, the roughness of the sample synthesized from electrolyte with 35 g L-1 boric acid is apparently the largest among the prepared coatings. Fig. 3 shows the corresponding elemental mapping for Ni and Cu composition, indicating the homogeneous distribution of Ni and Cu atoms in the coating. From the energy dispersive X-ray spectroscopy (EDS) results, the Cu content are 1.86 at.%, 3.64 at.%, 3.87 at.%, and 2.51 at.% in the coatings deposited from the bath with 15, 25, 35, and 45 g L-1 boric acid.

Fig. 1.   SEM surface morphologies of Ni-Cu coatings synthesized from the baths containing (a) 15 g L-1, (b) 25 g L-1, (c) 35 g L-1 and (d) 45 g L-1 boric acid.

Fig. 2.   SEM images of cross-sectional morphologies of Ni-Cu coatings synthesized from baths with (a) 15 g L-1, (b) 25 g L-1, (c) 35 g L-1 and (d) 45 g L-1 boric acid.

Fig. 3.   Elemental mapping for element distribution in the coatings synthesized from the baths with (a, b, c) 15 g L-1, (d, e, f) 25 g L-1, (g, h, i) 35 g L-1 and (j, k, l) 45 g L-1 boric acid.

Fig. 4 displays the corresponding TEM images (the insets: the electron diffraction patterns) of the Ni-Cu coatings. The almost continuous diffraction circles of the electron diffraction pattern in the insets show that the Ni-Cu coatings are consisted of fine grains with a face-centered cubic (FCC) structure. The brightest (200) circle indicates that the grain orientation is preferential. Fig. 5 shows the XRD patterns of these coatings, which are comprised of two peaks at 2θ values of 43° and 51°, corresponding to (111) and (200) crystal planes. Of course, it also indicates a strong (200) preferential grain growth orientation. Furthermore, the average grain size of these coatings is closely related to the boric concentration in the bath. Statistical analysis of the grain sizes by ImageJ software (Fig. 6) shows that the average grain size is approximately 200, 140, 130 and 210 nm for the coatings obtained from the bath with 15, 25, 35 and 45 g/L boric acid, respectively. The coating obtained from the bath with 35 g L-1 boric acid has the smallest grain size.

Fig. 4.   TEM images and corresponding diffraction patterns (insets) of the Ni-Cu coatings synthesized from the baths with (a) 15 g L-1, (b) 25 g L-1, (c) 35 g L-1 and (d) 45 g L-1 boric acid. The diffraction patterns in insets indicate the formation of nanostructured single γ phase.

Fig. 5.   XRD patterns for Ni-Cu nanocomposite coatings synthesized from bath with 15, 25, 35 and 45 g L-1 boric acid.

Fig. 6.   Statistical distributions for grain size of the nanocrystals of nickel-copper coatings synthesized from the bath with (a) 15 g L-1, (b) 25 g L-1, (c) 35 g L-1 and (d) 45 g L-1 boric acid.

3.2. Electrochemical corrosion behavior of coatings

Fig. 7 and Table 1 show the potentiodynamic polarization curves and electrochemical parameters of these coatings in 0.3 M NaCl solution. The coating obtained from the solution with 35 g L-1 boric acid shows the highest corrosion potential Ecorr (50 mV), pitting potential Epit (335 mV), and the lowest corrosion current density icorr (0.13 μA cm-2) among all the test coatings. Similarly, the passive current density (ip) of the sample obtained from the solution with 35 g L-1 boric acid (2.52 μA cm-2) is lower than those for samples obtained from solution with 15, 25 and 45 g L-1 boric acid (4.1, 2.7 and 3.2 μA cm-2, respectively). These results demonstrate that the coating synthesized from the bath with 35 g L-1 boric acid has better corrosion resistant than the coatings synthesized from the baths with 15, 25 and 45 g L-1 boric acid.

Fig. 7.   Potentiodynamic polarization curves of the coatings obtained from the baths with various boric concentrations in 0.3 M NaCl solution at 25 ± 1 °C.

Table 1   Electrochemical parameters obtained by fitting potentiodynamic curves in Fig. 7 (icorr and ip are the corrosion and passive current density, respectively; Ecorr and Epit are the corrosion and pitting potential, respectively).

Boric acid (g L-1)Ecorr (mVSHE)icorr (μA cm-2)ip (μA cm-2)Epit (mVSHE)
151.60.614.1196
25450.522.7284
35500.132.5335
45440.553.2195

New window

Fig. 8(a) shows the Nyquist plots; one can see immediately that a single semicircle for these coating in 0.3 M NaCl solution. The Ni-Cu coating obtained from the bath with 35 g L-1 boric acid has larger partial circle diameter, and the size of this semicircle is the biggest, followed by those obtained from 25, 45 and 15 g L-1 boric acid in order. Four coatings exhibit the large phase angle on a wide frequency range. The Ni-Cu coating obtained from the bath with 35 g L-1 boric acid has the highest phase angle in Bode plots and the model of impedance a linear slope, close to -1 in the middle at lowest frequency, indicating that the coating obtained from the bath with 35 g L-1 boric acid presents the best corrosion resistance. As all these specimens show self-passivation behavior (evident from the polarization behavior in Fig. 7), an imperfect passive film connected with the surface of the matrix is often modeled to simulate the interface during corrosion process [23,31]. Hence, the equivalent electrical circuit (Fig. 8(a) inset) is selected to interpret the EIS data. The equivalent circuit includes of solution resistance (RS), constant phase element (CPE) of the passive film capacitance (CPEf), film resistance (Rf), CPE of the double layer capacitance (CPEdl) and charge transfer resistance on the matrix surface (Rt). The fitting results are summarized in Table 2. The corrosion resistance can be determined through the charge transfer resistance Rt of the coating [32,33]. The higher the value of Rt, the lower the rate of the reaction: Mmatrix → Mfilmn+ + ne- (Mmatrix denotes the metal atom in the matrix, Mfilmn+ the cation in the passive film, e electron and n the charge transfer number in the reaction). Fig. 9 shows the Rt values of these coatings obtained from the different boric acid concentration in the bath. One can observe that the coating obtained from the bath with 35 g L-1 boric acid has the highest Rt (111.9 kΩ cm2). This indicates that this coating shows the best corrosion resistance, which is also in line with the results of potentiodynamic polarization measurements (Fig. 7 and Table 1).

Fig. 8.   Nyquist plots of nano-coatings. The inset in shows the corresponding equivalent electrical circuit. RS is the solution resistance, CPEf is CPE which expresses the passive film capacitance and Rfilm is passive film resistance, CPEdl is also the CPE which expresses the double layer capacitance and Rt expresses the charge transfer resistance of Ni-Cu coatings, respectively.

Table 2   Key electrochemical parameters obtained by fitting the EIS data in Fig. 8 (RS is the solution resistance, CPEf is the constant phase element (CPE) of the passive film capacitance, Rfilm is the passive film resistance, CPEdl is the CPE of the double layer capacitance and Rt is the charge transfer resistance on the matrix surface).

Boric acid (g L-1)RS (Ω cm2)CPEf (×10-5 S s-n cm-2)n1Rfilm (kΩ cm2)CPEdl (×10-5 S s-n cm-2)n2Rt (kΩ cm2)
1514.451.7980.95416.253.4910.774523.68
2515.872.0350.960241.421.920.587764.05
3515.511.7340.947252.250.8250.4286111.9
4514.111.680.9620.853.1560.756429.17

New window

Fig. 9.   Relationship between the charge transfer resistance (Rt) of the Ni-Cu nano-coatings and boric concentration (g L-1) in the electrolyte.

Generally, the smoother surface of the sample is, the higher the corrosion resistance. Some studies [[34], [35], [36]] have shown that surface roughness and corrosion resistance are closely related. This is in conflict with our current data: the electrochemical experiments showed that the coating from the bath with 35 g L-1 boric acid possessed the best corrosion resistance; however, Fig. 2 showed that the coating presented the largest roughness. This indicated that the roughness of the surface was not the major factor affecting the corrosion behavior in this work. This may be due to the very small difference among the surface roughness of these coatings.

Boric acid is a useful additive in the electrodeposition, which is often considered as a buffer for the sulfate-based solution. Amblard et al. [37] stated that the competitive growth rates between various nuclei control the orientation of growth crystallites and ultimately the resultant deposit texture. The kinetics of this competition is primarily controlled by the cathodic reaction and a very important parameter that is the solutions pH. Therefore, boric acid is usually added to nickel alloy electroplating solutions (sulfate-based solution) in order to maintain the solution pH during electrodeposition. However, during nickel deposition, the rise in pH results in precipitation of nickel hydroxide on the sample (cathode), thereby passivating the surface and inhibiting further deposition [26,38]. In this work, one can see that the boric acid concentration in the bath significantly alters the microstructure and morphology of the coatings and their corrosion behavior. The coatings obtained from the solution with 15 and 45 g L-1 boric acid display much inferior corrosion resistance properties to of the coatings from 25 and 35 g L-1 boric acid in the bath. So, the coatings obtained from the bath with 25 and 35 g L-1 boric acid are used for further study in the following work.

3.3. Effect of the composition of the passive films on corrosion behavior

During electroplating, a complex between a Ni2+ ion and an anion forms in solution. When the complex is adsorbed to the cathode surface, two consecutive one-electron transfer reactions take place, thereby reducing the concentration of nickel ion and releasing the anion [26]:

Ni2+ + X- ↔ NiX+ (2)

NiX+ + e- ↔ NiXads (3)

NiXads + e- ↔ Ni + X- (4)

Meanwhile, copper ions reduction occurs via two steps in order [27,28].

The first step is the reduction of Cu2+ ions

Cu2+ + H2O + e- → Cu2O + 2H+ (5)

Cu2+ + e- → Cu+(6)

The second step is the reduction of Cu+ ions.

Cu2O + 2H+ + 2e- → Cu + 2H2O (7)

Cu+ + e- → Cu (8)

When they are concurrently deposited to form an alloy coating, they will be oxidized to ions again in the corrosive environment and can be detected by XPS. Fig. 10 shows the XPS survey spectra of the passive films on the coatings after 30 min of passivation at 0.15 VSHE in 0.3 M NaCl solution. There exist identified Ni and Cu peaks for all spectra, but no Cl peak can be observed. It clearly indicates that chloride-free passive films have formed on the Ni-Cu alloy coatings.

Fig. 10.   XPS survey spectra of the passive films formed on the Ni-Cu nanostructured coatings synthesized from baths with (a) 25 g L-1 and (b) 35 g L-1 boric acid after passivation 30 min at 0.15 VSHE in 0.3 M NaCl solution.

Fig. 11 shows the core-level spectra of Ni 2p3/2 for the passive films formed on the coatings after passivation 30 min at 0.15 VSHE in 0.3 M NaCl solution. By computational fitting, the state of the nickel in the passive films was analyzed. NiO, Ni(OH)2, NiO satellite and Ni(OH)2 satellite are identified at 853.6, 855.5, 856.93 and 861.7 eV, respectively. Similarly, CuO and Cu2O are also identified corresponding to the peaks at 933.4 and 932.8 eV, respectively (Fig. 12). A quantitative XPS results information about the ionic state of the nickel and copper are summarized in Table 3. The Cu2O proportion (36.62%) in the Cu peak on the specimen obtained from the bath with 25 g L-1 boric acid is smaller than that of the specimen obtained from the bath with 35 g L-1 boric acid (48.94%). The Ni(OH)2 proportion (21.67%) in the Ni peak on the passive film of the coating from the bath with 35 g L-1 boric acid is smaller than that of the coating from 25 g L-1 boric acid (47.36%). Compared with the data in Fig. 7, Fig. 8, the highest corrosion resistance of the coating from the bath with 35 g L-1 boric acid is highly ascribed to its highest Cu2O content and the lowest Ni(OH)2 content in the passive film. This also well corresponds to the authors’ previous work [39] on the relationship of corrosion resistance and the content of Cu2O and Ni(OH)2 in its passive film which with various of Cu in the Ni-Cu nano-coatings. Recent work has proved that more nickel hydroxides could degrade the corrosion resistance of the passive film form on Ni [10,40]. The effect of Cu2O can be attributed to the addition of lower valence Cu+ to the higher valence Ni2+, leading to the change of the semiconductor behavior and the point defects of the passive film.

Fig. 11.   Core-level spectra of Ni2p3/2 for the passive films formed on the Ni-Cu nanostructured coatings synthesized from baths with (a) 25 g L-1 and (b) 35 g L-1 boric acid after passivation 30 min at 0.15 VSHE in 0.3 M NaCl solution.

Fig. 12.   Core-level spectra of Cu2p3/2 for the passive films formed on the Ni-Cu nanostructured coatings synthesized from baths with (a) 25 g L-1 and (b) 35 g L-1 boric acid after passivation 30 min at 0.15 VSHE in 0.3 M NaCl solution.

Table 3   Atomic percentage (at.%) of nano coatings after passivation 30 min at 0.15 VSHE in 0.3 M NaCl solution.

Boric acid (g L-1)Cu2p3/2Ni2p3/2
CuO (%)Cu2O (%)Ni(OH)2 (%)NiO (%)
2563.3836.6247.3652.64
3551.0648.9421.6778.33

New window

Furthermore, no signal of Cl- (which should correspond to the peak at 198.7 eV) is found in the passive films formed on all the coatings (Fig. 10 or Appendix Fig. A2). This result is in line with the document [41], which indicates that the existence of chloride ions is closely related to the point defect in the passive film.

3.4. Diffusivity of point defects in the passive film

Fig. 13 shows the M-S curves for the passive films formed on the coatings from the bath with 25 and 35 g L-1 boric acid after 30 min passivation at 75, 100, 125, 150, 175 and 200 mV (SHE), respectively, in 0.3 M NaCl solution. Most curves are straight lines with a positive slope (except some part of the curves with a negative slop at higher potential), indicating that these passive films show n-type semiconducting behavior. It is known that the conduction for n-type semiconductor is controlled by its donor concentration. Donor concentrations (Nd) in the passive films can be quantified from the slope of the straight line of M-S plots as follows [31,[42], [43], [44]]:

$\frac{1}{C^{2}}=\frac{2}{εε_{0}qN_{d}} (E-E_{FB}-\frac{kT}{q})$ (9)

where C is the capacitance of passive film semiconductor, ε is the dielectric constant of the passive film (12 [45]), ε0 is the permittivity of vacuum (8.854 × 10-14 F cm-1), q is the electron charge (1.602 × 10-19 C) and Nd is the donor concentration. E is applied potential, EFB is the flat band potential, k is the Boltzmann constant (1.38 × 10-23 J K-1) and T is the absolute temperature.

Fig. 13.   M-S plots of the passive films formed the on Ni-Cu nanostructured coatings synthesized from baths with (a) 25 g L-1 and (b) 35 g L-1 boric acid in 0.3 M NaCl solution after 30 min passivation at 75, 100, 125, 150, 175 and 200 mVSHE.

Generally, donor is always referred to oxygen ion vacancy (point defect) in the passive film. Table 4 summarizes the donor (oxygen ion vacancy) densities of the passive films formed on these two coatings under various potentials. According to the point defect model (PDM) [46], the relation between Nd (donor concentration) and Ef (passive film formed potential) can be theoretically described as follows:

Nd1exp-bEf2 (10)

where ω1, ω2, and b are constants that can be found from the experimental data of the donor concentration Nd. Fig. 14 shows the relation between Nd and Ef of the passive films formed on the synthesized coatings. The fitting curves to the experimental data in the figure agree well with the Eq. (10): Nd = [5.42 × exp(0.016Ef) + 262] × 1019 for the coating obtained from the bath with 25 g L-1 boric acid. Nd = [3.09 × exp(0.016Ef) + 222] × 1019 for the coating obtained from the solution with 35 g L-1 boric acid. This indicates that PDM is suitable to study the passive films formed on these coatings. Sikora et al. [46] indicated that ω2 ω2 in Eq. (10) is related to the diffusivity of the point defects D0 (here, refers to the diffusion coefficient of oxygen ion vacancy) by Eq. (11) based on the Nernst-Plank transport equation:

D0= $\frac{J_{0}}{2kω_{2}}$ =$\frac{i_{ss}RT}{4eFω_{2} ε_{L}}$ (11)

where J0 is the steady state flux of donors, iss is the steady state current density throughout the passive film, which can be determined from the potentiostatic polarization. K = FεL/(RT), F is the Faraday constant, εL the mean electric field strength, R is the gas constant, T is temperature in Kelvin and e is the change of an electron.

Table 4   Carrier densities Nd (×1021 cm-3) of the passive film formed on the Ni-Cu coatings in 0.3 M NaCl solution after 30 min passivation at 75, 100, 125, 150, 175 and 200 mVSHE.

Potential (mVSHE)Concentration of Boric acid (g L-1)
2535
752.812.31
1002.942.4
1252.972.48
1503.142.53
1753.652.78
2003.943.01

New window

Fig. 14.   Nd-Ef curves and their fitting curves of the passive film formed on the Ni-Cu nanostructured coatings synthesized from baths with (a) 25 g L-1 and (b) 35 g L-1 boric acid.

To obtain the iss, potentiostatic polarization tests were performed. Fig. 15 shows the potentiostatic current transients of two passive films. The current seems to have reached a steady state value after 1800s, and all the current transient curves measured at various potentiostatic conditions are similar. From Fig. 15 inset, iss are measured to be 138.5 nA cm-2, 81.16 nA cm-2 for the passive films formed on the coatings from the baths with 25 g L-1, 35 g L-1 boric acid, respectively. The passive film thickness (Lss) is related to Ef and εL by Eq. (12) [46]:

LSS= $\frac{1}{ε_{L}}(1-α)E_{f}+B$ (12)

where α is the polarizability of the film/solution interface (α = 0.5) [22] and B is a constant.

Fig. 15.   Potentiostatic current transients for the passive films formed on the Ni-Cu nanostructured coatings synthesized from baths with (a) 25 g L-1 and (b) 35 g L-1 boric acid, measured by the potential at 75, 100, 125, 150, 175 and 200 mVSHE in 0.3 M NaCl solution. The insets show the steady state current density through passive films formed on these coatings for 1800s at various film formation potentials in 0.3 M NaCl solution.

The thickness of passive film Lss is calculated using the relation [47]:

Lss = QM/(zFAρr) (13)

where Q is the charge spent on passive film formation during anodic scan at various potentials (75, 100, 125, 150, 175 and 200 mV (vs. SHE)), z is the number of electrons interchanged, r is the roughness factor (r = 3.5 [48]), A = 1 cm2, M(NiO) =74.69 g mol-1 and ρ(NiO) =6.8 g cm-3 [49]. Lss (the thickness of passive film) of two coatings is presented as a function of the film formation potential (Ef) in Fig. 16. The two slopes, which represent anodizing constant of the passive film, are 34.775 nm V-1 for the sample from the bath with 25 g L-1 boric acid and 36.262 nm V-1 for the sample from the bath with 35 g L-1 boric acid, respectively (Fig. 16). It is indicated that the thickening of the passive film has been greatly inhibited with the positive increase of the potential for the specimen from the bath with 35 g L-1 boric acid. The εL can also be calculated to be 1.44 × 105 V cm-1 for the coating from 25 g L-1 boric acid and 1.38 × 105 V cm-1 for the coating from 35 g L-1 boric acid.

Fig. 16.   Lss-Ef and their fitting liners of the passive film formed on the Ni-Cu coatings for 1800s at various film formation potentials in 0.3 M NaCl solution.

The diffusivity of point defects D0 is calculated as 1.47 × 10-17 cm2 s-1 for the coating obtained from the bath with 25 g L-1 boric acid and 1.06 × 10-17 cm2 s-1 for the coating synthesized from the bath with 35 g L-1 boric acid. The results indicate that the lower Nd and D0 play an important role in suppressing the mass transfer through the passive film according to Fick’s law of diffusion. In other words, the lower Nd and D0 can inhibit the growth and dissolution of the passive film formed on the sample. This is also in line with the results of potentiodynamic polarization measurements (Fig. 7).

Generally there are three types of point defects in passive film, i.e. cation vacancies, anion vacancies and cation interstitials, which are electron acceptors, electron donors, and electron donors, respectively [50,51]. Ni and Cu have similar cation radii (Pauling radii 0.73 Å for Cu2+, 0.77 Å for Cu+, and 0.69 Å for Ni2+). This indicates that copper ion or vacancies are in favor of substituting nickel ones rather than forming interstitial ions. A recent work [52] demonstrated that a copper vacancy (VCu-) is easily formed in the passive film than a copper interstitial ion due to its lower formation energy. Therefore, a dominance of anion vacancies (that is, oxygen vacancies) is over the other two types of point defects in these passive films. According to PDM, the following reactions would occur at the metal/film (m/f) and film/solution (f/s) interfaces:

Ni→$Ni_{Ni}^{"}$ +$ V_{O}^{2+}$ +2e- (14)

Cu→$Cu_{Ni}^{"}$ +$ V_{O}^{2+}$+2e- (15)

Cu→$Cu_{Ni}^{’}$ +$\frac{1}{2} V_{O}^{2+} $+e- (16)

$Ni_{Ni}^{"}$→$Ni_{Nisolution}^{2+}$+$V_{Ni}^{2-}$ (17)

$Cu_{Ni}^{"}$→$Cu_{solution}^{2+}$ +$V_{Cu}^{2-}$ (18)

2CuNi'→2$Cu_{solution}^{+}$+$V_{Cu}^{-}$ (19)

$V_{O}^{2+}$ +H2O→OO+2H+ (20)

NiO+2H+→Ni$Cu_{solution}^{2+}$ +H2O (21)

In these reaction, NiNi", CuNi" and CuNi' are the metal cation in cation sites on the metal sub-lattice; VNi2-, VCu2- and VCu- are the cation vacancies on the metal sub-lattice; $V_{O}^{2+}$ is the oxygen vacancy on the oxygen sub-lattice; OO is the oxygen anion on the oxygen sub-lattice; $Ni_{Nisolution}^{2+}, $Cu_{solution}^{2+}$ and $Cu_{solution}^{+}$ are the metal cations in the solution. Oxygen vacancies generate at the metal|film interface (m|f) and annihilate at the film|solution interface (f|s), and cation vacancies generate at the f|s interface and annihilate m׀f interface. The number of oxygen vacancies indicates that the reactions (14-16) prevail over the reactions (17-19). This can also be demonstrated by the fact that all Nd values increase with the film formed potentials. However, this is very different from the report that the passive film on pure nickel shows a p-type semiconducting behavior, whose test solution was borate buffer solution (pH 8.5) without or with various Cl- concentrations [41,47,53]. These studies show that the dominance of cation vacancies than oxygen vacancies is due to the lower formation energy of cation vacancy. It seems to conflict with the results in this work. It is noted that their test solution (pH 8.5) was greatly different from that in this work (neutral). It was reported that isoelectric point of NiO was 7.3 [54]. This indicates that the pH 8.5 test solution is actually Lewis base to the NiO film, corresponding to the enhancement of the reaction (20), which would facilitate the annihilation of the oxygen vacancies, thereby leading to the dominance of cation vacancies in the passive film (so the film showed a p-type semiconducting behavior). However, for the solution in the present work, it is Lewis acid to the NiO film, corresponding to the increasing of the reaction (21), which would impede the reactions (17-20), thereby leading to the preponderance of oxygen vacancies over cation vacancies and a n-type semiconducting behavior. Another cause may be attributed to the nanostructured of these coatings (Fig. 4). It has been confirmed that the high density of dislocations near the nano-grain boundaries would promote the reactions (14-16) at m׀f interface [55]. The reactions (14-16) at m׀f interface are electrochemical and Rt is an indicator of the resistance to these reactions. From the Fig. 9, Rt reaches its maximum for the sample from the bath with 35 g L-1 boric acid, indicating that the resistance to the reactions (15-17) at the m׀f interface is the highest, so the generation of $V_{ O }^{2+} V_{ O }^{2+}$ is significantly impeded in this case. This well agrees with the lower Nd and D0 in the passive film on the coating from the bath with 35 g L-1 boric acid. On the other hand, the XPS result (Fig. 12) shows that Cu2+ and Cu+ cations co-exist in the passive film on these nano- coatings from 25 and 35 g L-1 boric acid, indicating that the reactions (15, 16) occur at m/f interface. When Cu+ is added to NiO film, only half of oxygen vacancies is generated compared with reactions (14, 15). This can be proved by the fact that the higher Cu+ content is corresponding to the lower Nd and D0 in the passive film on the coating from the bath with 35 g L-1 boric acid. Therefore, Cu+ additions to nickel oxide can effectively reduce the point defects in the passive film. The less diffusivity of point defects D0 corresponds to the more compact passive film and the more corrosion resistance.

4. Conclusions

The nanostructured Ni-Cu coatings were synthesized by puled electrodeposition technique in sulfate bath containing various amount of boric acid. The corrosion behavior and the diffusivity of point defects (D0) in the passive film were investigated by electrochemical measurements in 0.3 M NaCl solution:

(1) The concentration of boric acid in the bath strongly influences the microstructure of the coatings. The values of their average grain size are 200, 140, 130 and 210 nm for the samples from the bath with 15, 25, 35 and 45 g L-1 boric acid, respectively.

(2) The passive films formed on the coatings exhibited a n-type semiconducting behavior with the donor density that measured by a Mott-Schottky analysis. The donor concentration increased with increasing the film formation potential.

The thickness of the passive films linearly increased with increasing the formation potential.

(3) The higher Cu+ content (48.94%) in the passive film, the lower of the diffusivity of point defects (1.06 × 10-17 cm2 s-1). The higher Cu+, the higher inhibiting the passive film thickness and the more compact passive film formed on Ni-Cu coating.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Nos. 51771061 and 51571067), National Basic Research Program of China (No. 2014CB643301), National Natural Science Foundation of Heilongjiang Province, China (No. E2016022), the Fundamental Research Founds for the Central Universities (No. HEUCFG201838), the Ministry of Science and Technology of China (No. 2012FY113000), Key Laboratory of Superlight Materials and Surface Technology (Harbin Engineering University), Ministry of Education and the Chinese Scholarship Council in conjunction with Harbin Engineering University & the Viet Nam Maritime University.

Appendix A

Fig. A1

Fig. A1   Surface morphology of Ni-Cu nano alloy coatings synthesized from bath with (a)15, (b) 25, (c) 35 and (d) 45 g L-1 boric acid after potentiodynamic polarization.

Fig. A2   The core-level spectra of Cl 2p for the passive films formed on the Ni-Cu nanostructured coatings synthesized from bath with (a) 25 and (b) 35 g L-1 boric acid after passivation 30 min at 0.15 VSHE in 0.3 M NaCl solution.


/