Fabrication of a hydrophobic, electromagnetic interference shielding and corrosion-resistant wood composite via deposition with Ni–Mo–P alloy coating

Abstract

A hydrophobic, electromagnetic shielding and corrosion-resistant wood-based composite was prepared via electroless deposition of a Ni–Mo–P ternary alloy on birch veneers. The effect of MoO₄²⁻ concentration on the structure and properties of plated veneers was studied. Samples were characterized by scanning electron microscopy-energy dispersive spectroscopy, X-ray diffractometry and X-ray photoelectron spectroscopy. The water contact angles, surface resistivity, electromagnetic shielding effectiveness, corrosion resistance and adhesive strength were measured. Scanning electron microscopy images showed that the birch veneer surfaces were covered with dense, continuous and uniform coatings, and the special wood structure still existed. X-ray diffractometry, energy-dispersive spectroscopy and X-ray photoelectron spectroscopy results showed that the coatings obtained from this experiment were crystalline, and consisted mainly of Ni, Mo and P. Birch veneers plated with crystalline Ni_91.71–Mo_7.07–P_1.22 films exhibited good hydrophobic properties with a water contact angle of 144°, high electrical conductivity with surface resistivity of 208 mΩ cm⁻² and electromagnetic shielding effectiveness above 45 dB from 9 kHz to 1.5 GHz and excellent corrosion resistance with a polarization resistance of 8244 Ω cm⁻². Adhesion test results showed that the Ni–Mo–P films adhered firmly to the wood surface. This study offers a new pathway for fabricating multifunctional wood-based composites.

---

1. Introduction

In recent years, electromagnetic waves have become the fourth most severe source of public pollution after water, noise and air pollution.¹ Electromagnetic interference (EMI) is a well-known problem that results in interception and the malfunction of electronic instruments in commercial, medical, aircraft and military systems.² It also harms the human body, which has led to the desire for a healthy environment, particularly for communication and work.^3,4 Consequently, it is necessary to develop effective and useful EMI shielding materials. Wood is a natural and renewable material that is often used for decoration because of its sound insulation, humidity and temperature-controlling performance. Dry wood is a poor conductor of electricity, which limits its application for EMI shielding. To make wood conductive, electroless plating is used to prepare wood-based composite via metal or alloy coating deposition on the wood surface. In the past two decades, many previous studies have focused on electroless deposition of Ni–P or Cu on wood veneers, and these coatings have been found to exhibit good conductivity and high electromagnetic shielding.^5–12 However, the corrosion resistance of Ni or Cu deposits is poor.

Recently, a ternary system of electroless plating Ni–X–P (X = Cu, Co, W, Mo, Fe) coating has been developed to enhance the combined properties of binary alloy (Ni–P) to meet stricter requirements.¹³ In our previous studies, Ni–Cu–P alloy has been deposited on wood surface to obtain corrosion-resistant wood-based composite for long-term utilization,¹⁴ and Ni–Fe–P alloy deposition endowed wood magnetic responsiveness.¹⁵ Among the ternary alloys, Ni–Mo–P coatings have excellent wear resistance, thermal stability and corrosion resistance. Qin et al.¹⁶ reported that the addition of Mo improved the catalytic activity of amorphous Ni–P catalysts. Amani et al.¹⁷ verified that the Ni–Mo–P coatings showed a higher corrosion resistance compared with Ni–P films. Wu et al.¹⁸ and Chou et al.¹⁹ demonstrated that an increase in Mo content in Ni–Mo–P deposits decreased the electrical resistivity because the coating crystallinity increased. To date, Ni–Mo–P ternary alloys have been plated successfully on many substrates, including alumina ceramic, copper and mild steel panels. However, deposition on wood surfaces has not been reported.

In this work, Ni–Mo–P ternary alloys were deposited on birch veneer surfaces to prepare multifunctional wood-based composite. The effect of MoO₄²⁻ concentration on chemical composition, crystal structure and surface resistivity of Ni–Mo–P deposits was studied. Samples were characterized by water contact angle, electromagnetic shielding effectiveness, corrosion resistance and adhesive strength.

2. Experimental

2.1 Materials and method

Birch veneers (5 cm × 5 cm with 0.6 mm in thickness) were used as substrates for electroless deposition of Ni–Mo–P ternary alloys in this study. The bath composition and operating condition of electroless plating are given in Table 1. Sodium sulfate and sodium molybdate were used as precursor salts of nickel and molybdenum, respectively. Sodium hypophosphite was a reducing agent and the source of P. Sodium citrate and sodium acetate were used as complexing and buffering agent, respectively.

Table 1 Compositions and operating condition of electroless Ni–Mo–P plating

Chemical	Concentration (g L^−1)
NiSO4·6H2O	32.5
Na2MoO4·2H2O	0.6
NaH2PO2·2H2O	24
Na3C6H5O7·2H2O	27.5
CH3COONa·3H2O	12.5
Temperature	85 °C
pH	9.5

---

The preparing process and schematic illustration for the electroless deposition of Ni–Mo–P alloy on wood veneer are shown in Fig. 1. Prior to the plating, wood samples were treated through a series of processes. First, samples were immersed in APTHS ((HO)₃Si(CH₂)₃NH₂) solution at room temperature for 8 min to obtain a self-assembled monolayer of APTHS on the wood surface, then dried in hot air. Activating with a 0.2 g L⁻¹ PdCl₂ solution at room temperature for 10 min, after that, the samples were immersed in 2 g L⁻¹ NaH₂PO₂ solution at 40 °C for 15 min. Finally, the activated veneers were placed into the plating solution (pH was adjusted with ammonia water). After plating, the products were thoroughly rinsed in running water and dried.

Fig. 1 Preparation process and schematic illustration of electroless deposition of Ni–Mo–P film on wood veneer.

2.2 Characterizations

Scanning electron microscope (SEM, model Quanta 200, America FEI Company) with the energy dispersive spectrometer (EDS) attachment were adopted to observe the surface morphology and determine the elemental composition of the deposits. X-ray diffraction (XRD, model D/MAX-3B, Rigaku) was employed to analyze the crystal structure of the coatings. The composition and chemical states of deposits surface were measured by X-ray photoelectron spectroscopy (XPS, model PHI 5700 ESCA System, America Physical Electronic Company). Water contact angle (WCA) of plated veneers was measured with a contact angle analyzer (JC2000C, Beijing Code Tong Technology Co., Ltd) at room temperature with a droplet volume of 5 μL. Corrosion measurement of plated veneers was tested in 3.5 wt% NaCl solution at room temperature with a CHI660D-type electrochemical work station. Corrosion parameters (E_corr, I_corr, R_p) were calculated from the potentiodynamic polarization curve with a conventional three-electrode system.

2.3 Surface resistivity test

The surface resistivity of the plated birch veneers was determined with a smart low direct-current resistance tester (YD2511A-type) according to the Standard (GJB-2604-96) of Chinese National Military. Due to the anisotropic structure of wood, the plated birch veneers were also anisotropic in electrical conductivity. Thus, all resistances were determined by averaging of 20 measurements on each sample surface. The average surface resistivity was calculated according to the formula,

R_s (Ω cm⁻²) = R/S (1)

where R (Ω) is the resistance value of the sample, S (cm²) is the area of current trough plated veneer.

2.4 Shielding effectiveness measurement

The shielding effectiveness (SE) of plated veneers was evaluated by using an Agilent E4402B spectrum analyzer based on the Chinese Industrial Standard SJ20524-95. The eqn (2) was used to calculate the SE value,

SE (dB) = −10 × lg(P_out/P_in) (2)

where P_out and P_in are incident and transmitted power, respectively.

2.5 Adhesion test

The wood samples (5 cm × 5 cm with 1 cm in thickness) were used for adhesion measurement. A tensile test was adopted to measure the adhesion between Ni–Mo–P coating and wood substrate by using a vertical pulling method. The schematic illustration of tensile test was used as in Li et al.²⁰ The adhesion values were then measured with a Shimadzu AG-10TA-type testing machine.

3. Results and discussion

3.1 Effect of MoO₄²⁻ concentration in plating solution

3.1.1 Composition of plated veneers. Electroless deposition of Ni–Mo–P ternary alloys involves some anodic and cathodic reactions, which can be written as:²¹

H₂PO₂⁻ → (HPO₂⁻) ads + (H) ads (3)

(HPO₂⁻) ads + OH⁻ → H₂PO₃ + 2e⁻ (4)

Ni²⁺ + 2e⁻ → Ni (5)

H₂PO₂⁻ + 2H⁺ + e⁻ → P + 2H₂O (6)

(H) ads + e⁻ + H⁺ → H₂ (7)

MoO₄²⁻ + [NiL]²⁺ + 2H₂O + 2e⁻ → [NiLMoO₂]²⁺ + 4OH⁻ (8)

[NiLMoO₂]²⁺ + 2H₂O + 4e⁻ → Mo (s) + [NiL]²⁺ + 4OH⁻ (9)

where L is a ligand.

The MoO₄²⁻ concentration in the plating solution is important for Ni–Mo–P coating deposition. As shown in Fig. 2, the Mo content in the Ni–Mo–P coatings increased from 3.57 wt% to 7.56 wt% with the MoO₄²⁻ concentration increasing from 0.2 g L⁻¹ to 0.6 g L⁻¹, whereas the P content in the deposits decreased from 4.52 wt% to 1.01 wt%. According to eqn (8) and (9), the increase in MoO₄²⁻ concentration promoted the reduction of Mo, and generated abundant hydroxide ions, which were detrimental to P reduction according to eqn (6). These results prove the competition between Mo and P in the co-deposition of Ni–Mo–P ternary alloys.

Fig. 2 Variation in weight percent of Mo and P in Ni–Mo–P coatings with MoO₄²⁻ concentration.

3.1.2 Microstructure of plated veneers. Fig. 3 shows the effect of MoO₄²⁻ concentration on Ni–Mo–P coating microstructure. The characteristic peak of cellulose in birch exists at 2θ = 21.72°.²² The curves at 2θ = 50.92° and 75.15° are ascribed to Ni (200) and Ni (220). A sharp diffraction peak at 2θ = 43.92° is likely to be a (111) reflection of crystalline Ni, which indicates the microcrystalline structure of the Ni–Mo–P deposits, and the peak intensity gradually increased as the MoO₄²⁻ concentration increased. This result can be explained by the change of P and Mo content in the coatings. It has been reported that the amount of P has an important effect on deposit structure, and in general, the low P content (1–5 wt%) coatings were microcrystalline.^23,24 In addition, Wu et al. demonstrated that Ni–Mo–P deposit crystallinity would increase with increasing Mo content in the coatings.¹⁸ With an increase in MoO₄²⁻ concentration from 0.2 g L⁻¹ to 1.0 g L⁻¹, the P and Mo contents in the deposits decreased and increased, respectively (Fig. 2), so the crystallinity of the Ni–Mo–P coatings increased.

Fig. 3 X-ray diffraction patterns of N–Mo–P films deposited at different MoO₄²⁻ concentrations: (a) 0.2 g L⁻¹, (b) 0.4 g L⁻¹, (c) 0.6 g L⁻¹, (d) 0.8 g L⁻¹ and (e) 1.0 g L⁻¹.

3.1.3 Surface resistivity of plated veneer. The surface resistivity of Ni–Mo–P coatings depends mainly on the coating structure, composition and thickness.²⁵ As shown in Fig. 4, as the MoO₄²⁻ concentration increased from 0.2 g L⁻¹ to 0.8 g L⁻¹, the surface resistivity decreased from 354.69 mΩ cm⁻² to 208.14 mΩ cm⁻². This is because, with increasing MoO₄²⁻ concentration, the coating crystallinity increased (Fig. 3) and the internal defects of these deposits decreased, so the resistivity reduced. When the MoO₄²⁻ concentration was at 1.0 g L⁻¹, the surface resistivity showed an inconspicuous increase in Fig. 4. This may be because of the narrow adjustment scale in MoO₄²⁻ concentration. As shown in Fig. 5a, the dark bulb indicates that the resistance of originally dried birch veneer was infinite because of its inherent insulation. However, the dazzling light in Fig. 5b implies that the Ni–Mo–P plated veneer possessed an excellent electro-conductivity.

Fig. 4 Effect of MoO₄²⁻ concentration on surface resistivity of Ni–Mo–P deposits.

Fig. 5 Conductive images of (a) original birch veneer and (b) Ni–Mo–P plated veneer.

3.2 Surface morphology of Ni–Mo–P plated veneers

The morphology of Ni–Mo–P plated veneer is shown in Fig. 6. The entire birch veneer surface is covered by a continuous and compact coating with an obvious metallic sheen (Fig. 6a–c). The basic wood surface structures, including pores, still exist and were not filled by the film. Fig. 6d and e shows the SEM images of Ni–Mo–P coatings with different elemental content. The coating became smoother as the Mo content increases. It has been reported that an increase in MoO₄²⁻ concentration reduces the deposition rate of Ni–Mo–P alloy because the MoO₄²⁻ inhibits H₂PO₂²⁻ oxidation and impedes Ni catalysis.²⁵ Lu et al.²⁶ explained that MoO₄²⁻ inhibited the deposition rate of electroless plating Ni–Mo–P coatings by the adsorption of molybdate anions on the substrate surface or on the growing Ni layer. Therefore, with increase in MoO₄²⁻ concentration, the Ni–Mo–P deposition rate decreased, which resulted in a smoother coating with increasing Mo content.

Fig. 6 SEM images of plated veneer surface: (a)–(c) Ni_91.73Mo_5.82P_2.45 plated veneer with 100, 500 and 1000 times magnification, respectively, (d) Ni_91.91Mo_3.57P_4.52, (e) Ni_91.73Mo_5.82P_2.45 and (f) Ni_91.71Mo_7.07P_1.22.

Fig. 7 High resolution XPS spectra of Ni–Mo–P-coated on birch veneer.

3.3 X-ray photoelectron spectroscopy analysis

The chemical valence states of Ni–Mo–P deposits were determined by X-ray photoelectron spectroscopy (XPS). Fig. 7 shows an XPS wide spectrum of the plated veneer. Ni, O, C, Mo and P peaks exist in the deposits. Fig. 8 also shows the chemical states of Ni and Mo in the as-plated coating after laser treatment in argon. The fitted peaks of Ni 2p_3/2 on the coating are displayed in Fig. 8a. The peak of metallic Ni 2p_3/2 appeared at 853.48 eV. The NiOOH 2p_3/2 spectra showed a main peak at 856.80 eV, with a shake-up satellite peak at 861.28 eV. The fitted peaks of Mo 3d on the coating are shown in Fig. 8b. The metallic Mo 3d_5/2 spectra revealed a peak at 228.80 eV and metallic Mo 3d_3/2 appeared at 231.68 eV. The oxide Mo peak positions are located at 233.22 eV (3d_5/2) and 236.22 eV (3d_3/2), which were identified as MoO₃.²⁷ XPS analyses indicated that the coating is composed of metallic and hydroxides/oxides of Ni and Mo.

Fig. 8 High resolution XPS Ni 2p and Mo 3d spectra of electroless deposited Ni–Mo–P alloys: (a) Ni 2p, (b) Mo 3d.

3.4 Shielding effectiveness of Ni–Mo–P plated veneers

The electromagnetic shielding effectiveness of the pristine birch veneer (Fig. 9a) fluctuated around 0 dB, which indicates that it has no shielding effectiveness. However, the shielding effectiveness of all samples exceeded 45 dB from 9 kHz to 1.5 GHz, and Ni–Mo–P plated veneer showed a slightly lower shielding effectiveness compared with Ni–P-coated veneer (Fig. 9b–e). This phenomenon is due to the content of P in Ni–P coating is lower than that in Ni–Mo–P coating. The shielding effectiveness of Ni–Mo–P-coated veneers increased as the Mo content in the deposit increased (Fig. 9 b–d). According to Schelkunoff's theory, materials with better conductivity have a higher shielding effectiveness.¹⁴ This result is attributed to a decline in surface resistivity with increase in Mo content (see Fig. 4). In general, the necessary condition for effective material shielding is that its electromagnetic shielding effectiveness is above 30 dB. Therefore, the Ni–Mo–P plated veneers are utility materials for EMI shielding.

Fig. 9 Electromagnetic shielding effectiveness of pristine veneer (a) and plated veneers (b) Ni–P, (c) Ni_91.91–Mo_3.57–P_4.52, (d) Ni_91.73–Mo_5.82–P_2.45 and (e) Ni_91.71–Mo_7.07–P_1.22.

3.5 Corrosion resistance of Ni–Mo–P plated veneers

The potentiodynamic polarization curves of Ni–P and Ni–Mo–P deposits in 3.5 wt% NaCl solution are plotted in Fig. 10. The cathode reaction in the polarization curves corresponded to the oxygen reduction, and the anodic polarization curve is the most significant characteristic related to corrosion resistance.²⁸ The corrosion parameters of tested veneers are recorded in Table 2. Larger values of corrosion potential (E_corr) and smaller values of corrosion current density (I_corr) result in higher sample corrosion resistance.²⁹ From Fig. 10, it can be seen that with increase in Mo content in the deposit, the corrosion potential shift increased and the corrosion current density decreased, thus, the anti-corrosive ability of Ni–Mo–P plated veneers increased. This occurs because of the smoother and more compact coating surface with increase in Mo content. It can also be observed that Ni–Mo–P-coated veneers showed a more positive corrosion potential shift and lower corrosion current density than the Ni–P coating. Moreover, the polarization resistance of Ni–Mo–P-coated veneers was above 3000 Ω cm⁻², which is higher than the 2608.1 Ω cm⁻² of the Ni–P coating (in Table 2). The Ni–Mo–P coatings possess a better corrosion resistance than the Ni–P coatings in sodium chloride solution. This behavior can be attributed to the incorporation of Mo, which increases the electrode potential of Ni–Mo–P ternary alloys because of its higher standard electrode potential (−0.20 V) than that of Ni (−0.25 V).

Fig. 10 Tafel plots of plated veneers in 3.5 wt% NaCl solution: (a) Ni–P, (b) Ni_91.91–Mo_3.57–P_4.52, (c) Ni_91.73–Mo_5.82–P_2.45 and (d) Ni_91.71–Mo_7.07–P_1.22.

Table 2 Corrosion parameters of wood veneers coated with Ni–P and Ni–Mo–P alloys

Sample	Ecorr (V)	Icorr (A cm^−2)	Rp (Ω cm^−2)
Ni–P	−0.4030	1.545 × 10^−05	2608.1
Ni91.91–Mo3.57–P4.52	−0.3560	1.011 × 10^−05	3197.5
Ni91.73–Mo5.82–P2.45	−0.3150	7.743 × 10^−06	5099.0
Ni91.71–Mo7.07–P1.22	−0.3010	4.522 × 10^−06	8244.0

---

3.6 Surface wettability of Ni–Mo–P plated veneers

The water contact angles (WAC) of the film and photographs of a water droplet on the veneer are shown in Fig. 11. The original birch veneer was hydrophilic with a WAC of 83° (Fig. 11a), whereas the Ni–Mo–P plated veneers were hydrophobic with over 130° WAC (Fig. 11b–d). When the Mo content was 7.07 wt%, the WAC of plated veneer reached 144°, which is close to the superhydrophobic state. Therefore, the Ni–Mo–P coatings obtained by electroless deposition on the wood surface prove the hydrophobic property of original birch veneer. To examine the surface repellency of Ni–Mo–P-coated veneer, 12 kinds of different pH value liquids were dropped on the plated veneer surface. Fig. 12 shows that these water droplets on the plated birch veneer still exhibited a spherical shape. The Ni–Mo–P coating therefore repels deionized water and corrosive liquids, such as acidic, alkaline and some aqueous salt solutions.

Fig. 11 Photographs and WAC of a water droplet on (a) original birch veneer, (b)–(d) Ni–Mo–P plated veneer.

Fig. 12 Photograph of water droplet for different pH values on the Ni–Mo–P-coated veneer.

3.7 Adhesion of Ni–Mo–P coatings to wood

The results of the adhesive tests between the Ni–Mo–P coating and the wood substrate are listed in Fig. 13. Samples a and b were mostly damaged in the adhesive layer and were slightly damaged in the wood substrate, but samples c and d were damaged in the adhesive layer only. This phenomenon indicates that the bonding strength of the coating and wood was higher than that of the wood itself or the wood and the adhesive layer. The data in Fig. 13 do not represent the adhesive strength between the coating and wood surface completely and only show the strength of the adhesive layer or wood substrate, but the adhesive strength between the coating and wood surface was higher than 1.7 MPa. Therefore, the Ni–Mo–P coatings were bonded firmly to the wood surface.

Fig. 13 Test results for adhesion between Ni–Mo–P coating and wood substrate.

4. Conclusions

Multifunctional wood-based composite with hydrophobic, electromagnetic shielding and corrosion-resistant properties was obtained successfully by electroless deposition of Ni–Mo–P ternary alloy on birch veneers. The Mo content in the coating affects the plated veneer performance. X-ray diffraction, energy dispersive spectroscopy and XPS results indicated that the Ni–Mo–P films were microcrystalline and elemental Mo in the coating existed as Mo⁰ and MoO₃. SEM images showed that the surface veneers were covered by a continuous and uniform coating. The Ni_91.71–Mo_7.07–P_1.22 coatings improved the hydrophobicity of the wood surface significantly with a 144° WAC. The coatings were bonded strongly to the birch veneer. The surface resistivity of the Ni_91.71–Mo_7.07–P_1.22 plated birch veneer reached 208 mΩ cm⁻², the electromagnetic shielding effectiveness reached 45–65 dB from 9 kHz to 1.5 GHz and the polarization resistances reached 8244 Ω cm⁻² in 3.5 wt% NaCl solution. This work provides a low-cost and well-controlled method to prepare multifunctional wood materials.

Acknowledgements

The research was supported by Province in Heilongjiang outstanding youth science fund (JC201301) and Fundamental Research Funds for the Central Universities (2572014EB02-01).

References

1. S. X. Jiang and R. H. Guo, Surf. Coat. Technol., 2011, 205, 4274–4279.
2. S. Yang, K. Lozano, A. Lomeli, H. D. Foltz and R. Jones, Composites, Part A, 2005, 36, 691–697.
3. C. S. Zhang, Q. Q. Ni and S. Y. Fu, Compos. Sci. Technol., 2007, 67, 2973–2980.
4. K. S. Chou, K. C. Huang and Z. H. Shih, J. Appl. Polym. Sci., 2005, 97, 128–135.
5. C. Nagasawa, Y. Kumagi, K. Urabe and S. Shinagawa, J. Porous Mater., 1999, 6, 247–254.
6. L. J. Wang, J. Li and Y. X. Liu, J. For. Res., 2005, 16, 233–236.
7. L. J. Wang and J. Li, BioResources, 2013, 8, 3414–3425.
8. L. J. Wang and H. B. Liu, BioResources, 2011, 6, 2035–2044.
9. L. J. Wang, J. Li and H. B. Liu, Wood Sci. Technol., 2011, 45, 161–167.
10. H. Liu, J. Li and L. Wang, Appl. Surf. Sci., 2010, 257, 1325–1330.
11. L. L. Sun, J. Li and L. J. Wang, Wood Sci. Technol., 2012, 46, 1061–1071.
12. J. Qin, M. M. Zhang and G. J. Zhao, Spectrosc. Spectral Anal., 2015, 35, 1253–1257.
13. Y. Yang, J. N. Balaraju, S. C. Chong, H. Xu, C. Liu, V. V. Silberschmidt and Z. Chen, J. Alloys Compd., 2013, 565, 11–16.
14. B. Hui, J. Li and L. J. Wang, BioResources, 2013, 8, 6097–6110.
15. C. H. Shi, L. Wang and L. J. Wang, J. Mater. Sci.: Mater. Electron., 2015, 26, 7096–7103.
16. Z. Z. Qin, Z. L. Liu and Y. H. Wang, Chem. Eng. Commun., 2014, 201, 338–351.
17. A. E. Fetohi, R. M. A. Hameed and K. M. El-Khatib, J. Power Sources, 2013, 240, 589–597.
18. Y. Wu, C. C. Wan and Y. Y. Wang, J. Electron. Mater., 2005, 34, 541–550.
19. Y. H. Chou, Y. Sung and Y. M. Liu, Surf. Coat. Technol., 2009, 203, 1020–1026.
20. J. Li, L. J. Wang and H. B. Liu, Surf. Coat. Technol., 2010, 204, 1200–1205.
21. Y. H. Chou, Y. Sung and C. Y. Bai, J. Electrochem. Soc., 2008, 155, 551–557.
22. B. Hui, J. Li, Q. Zhao, T. Q. Liang and L. J. Wang, BioResources, 2014, 9, 2949–2959.
23. H. Liu, F. Viejo, R. X. Gou, S. Glenday and Z. Liu, Surf. Coat. Technol., 2010, 204, 1549–1555.
24. Y. Gao, Z. J. Zheng and M. Zhu, Mater. Sci. Eng., A, 2004, 381, 98–103.
25. D. L. Liu, Z. G. Yang and C. Zhang, Mater. Sci. Eng., B, 2010, 166, 67–75.
26. G. J. Lu and G. Zangari, J. Electrochem. Soc., 2003, 150, 777–786.
27. A. Kocijan, C. Donik and M. Jenko, Corros. Sci., 2007, 49, 2083–2098.
28. B. Zaid, D. Saidi, A. Benzaid and S. Hadji, Corros. Sci., 2008, 50, 1841–1847.
29. R. Liu, J. Yao, Q. Zhang, M. X. Yao and R. Collier, Mater. Des., 2015, 78, 95–106.

---