Influence of the size of spraying powders on the microstructure and corrosion resistance of Fe-based amorphous coating

C. Zhang,R.Q. Guo,Y. Yang,Y. Wu,L. Liu

DOI: https://doi.org/10.1016/j.electacta.2011.05.020

IF: 6.6

2011-01-01

Electrochimica Acta

Abstract:The Fe-based amorphous coatings with the composition of Fe 48 Cr 15 Mo 14 C 15 B 6 Y 2 were successfully sprayed on mild steel substrate by the high velocity oxygen fuel (HVOF) spraying process with different feedstock powder sizes (i.e., powder A: −33 + 20 μm, powder B: −45 + 33 μm, powder C: −55 + 45 μm). The coatings were characterized for its morphology, microstructure and thermal stability by using X-ray diffraction (XRD), differential scanning calorimetry (DSC), scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The corrosion behavior of the coatings in 3.5 wt% NaCl solution was studied with potentiodynamic and potentiostatic polarization test. It was found that the particle size of the feedstock powders had a significant influence on microstructure and corrosion resistance of the resultant coatings. The coatings sprayed with the finest powders show the most compact structure; while the coating with the coarser powders exhibits a better corrosion resistance. It is found that the corrosion resistance of the coatings is closely related to the wetting behavior which is affected by the oxygen content and the roughness of coatings. The coatings with hydrophobicity exhibit a better corrosion. The present result demonstrates that the amorphous coatings with hydrophobicity and excellent corrosion resistant are promising for industrial application in marine environment. Keywords Amorphous coating Microstructure Corrosion resistance Hydrophobicity 1 Introduction An increasing interest has been paid in Fe-based amorphous coating in recent years because of their combinations of excellent properties including exceptional hardness, superior corrosion and wear resistances, and relatively low material cost [1–6] . Such characteristics make the Fe-based amorphous coating well suited for industrial applications in aggressive environments [6–8] . Various methods have been used to fabricate Fe-based amorphous coating, such as plasma spraying [4] , kinetic spraying [9] , flame spraying [10] and spark plasma sintered [11] and high velocity oxygen fuel (HVOF) [6–8,12,13] . Among them, HVOF spray process is superior to others because coatings prepared by HVOF usually exhibit excellent quality with low porosity, high hardness, high adhesion and compression stress [14,15] . The principle of HVOF is that the fuel (mostly using kerosene) and oxygen with a desired mass ratio combust to create a high temperature and high pressure exhaust gas, which then passes through a narrow nozzle to obtain a high speed of gas flow. In front of the nozzle, powders are introduced radially, thereby experiencing a high acceleration to supersonic velocities; upon impacting the substrate, the powders spread out thinly to form a well-bonded dense coating. The detailed process of HVOF was described in Ref. [16] . Liu et al. [6] has successfully prepared dense and hard FeCrMoCBY coatings with almost fully amorphous structure by HVOF, which exhibit the wear resistance several times higher than electroplated Cr and Ni-based amorphous coating. Regarding the corrosion behavior of amorphous coatings, Farmer et al. and Branagan et al. [8,17] reported that Fe 49.7 Cr 17.7 Mn 7.4 Mo 1.6 W 1.6 B 15.2 C 3.8 Si 2.4 amorphous coating prepared by HVOF is more resistant to pitting than nickel-based alloy (C-22) coating in seawater solutions and the salt fog environment. They argued that the excellent corrosion resistance of the Fe-based amorphous coating studied is attributed to the structural and compositional effect. First, amorphous alloys with homogenous structure at atomic level and without crystal defects such as grain boundaries and dislocations could highly reduce the sensitive sites for pitting initiations; Secondly, the Fe-based amorphous coatings containing a large amount of valve elements, such as Cr, Mo and W, could easily form a dense and stable passive film to prevent the coatings from Cl − attacks [18] . On the other hand, the spraying parameters also have an important effect on the structure and performance of amorphous coatings. Zhou et al. [12] studied the effect of spraying parameters, i.e., ratio of fuel and deposition rate, on the structure and corrosion resistance of the coatings with the same composition as ours. The result revealed that optimal spraying parameters could improve the corrosion resistance of Fe-based amorphous coatings due to the reduction of porosity and increase in the amorphous fraction. In addition to the spray parameters stated above, the feedstock powder size also plays a crucial role in the manufacturing and quality of coating as it affects the deposition efficiency, porosity, oxygen content, and eventually the properties of the coatings [19] . Farmer et al. [8] found that powder size has an effect on the corrosion performance of Fe 49.7 Cr 17.7 Mn 7.4 Mo 1.6 W 1.6 B 15.2 C 3.8 Si 2.4 amorphous coating in seawater. For example, the passive film of the coating prepared with coarse powders is more stable than that of the coating prepared with fine powders. However, the mechanism for such an effect has not been understood so far. In the present work, the influence of the size of spraying powders on the microstructure and corrosion resistance in 3.5 wt% NaCl solution of Fe 48 Cr 15 Mo 14 Y 2 C 15 B 6 amorphous coatings was studied. It is found that the structure and corrosion performance of the coatings is closely related to the particle size of the spraying powders. 2 Experimental procedure Fe 48 Mo 14 Cr 15 Y 2 C 15 B 6 powders were produced by high pressure Ar gas atomized, then sieved into three kinds of size ranges: i.e., powder A: −33 + 20 μm; powder B: −45 + 33 μm; powder C: −55 + 45 μm. The mild steel (0.45 wt% C) was selected as substrate with a size of 60 mm × 40 mm × 5 mm. The coatings using three particle sizes were fabricated by HVOF thermal spray system under the identical spraying condition. The detailed spraying parameters are the following: kerosene and oxygen flow are 22 L h −1 and 32 m 3 h −1 , respectively; the spraying distance is 350 mm and the powder feed is 30 g min −1 . All the spraying experiments were conducted in open air. In addition, an amorphous rod with 2 mm diameter with the same nominal composition was prepared by copper mould casting [20] for comparative study. The microstructure of the powders and coatings (named as coating A, B and C for powder A, B and C, respectively) were characterized by scanning electron microscopy (SEM, Quanta 200) coupled with energy dispersive spectroscopy (EDS) and X-ray diffraction (XRD, χ’Pert PRO). The oxygen content in powders and coatings was measured using an oxygen determinator (Leco Ro-416DR). Thermal behavior for the powders and corresponding coatings was investigated by differential scanning calorimetry (DSC, Netzsch Sta 449 C) in a flow of argon atmosphere with a heating rate of 0.33 °C s −1 . Further details of the microstructure of the coatings were examined by transmission electron microscopy (TEM, Jeol-2010). The hardness of the coatings was measured using a micro-indenter with the load of 100 g and a dwell time of 15 s. At least 10 measurements were done for each sample to ensure the data repeatability. The porosity of the coatings was evaluated by analyzing SEM micrographs with the Image Pro-Plus 6.0 software. The corrosion properties of the coatings were evaluated by electrochemical measurements. Prior to the corrosion test, all samples were wet ground up to 2000 SiC paper and polished to mirror finish, followed by ultrasonically cleaning in acetone, washed in distilled water and dried in warm air. Potentiodynamic polarization was performed at room temperature in a three electrode cell with platinum net counter-electrode and saturated calomel reference electrode (SCE). Specimens for corrosion test were closely sealed with epoxy resin, leaving only an end-surface (with a surface area of about 1 cm 2 ) exposed for testing. The electrochemical behavior was characterized by recording a Tafel plot at a potential sweep rate of 0.5 mV s −1 from −100 mV to 1500 mV ( vs. OCP) in 3.5 wt% NaCl aqueous solution open to air after immersing the specimens for an hour, when the open-circuit potential became almost steady. Each test was repeated three to five times for repeatability and reliability. In addition, potentiostatic test was applied to investigate the stability of passive film under the potential of 300 mV (in the region of passivation) for 10 h and the potential of 1500 mV (in the region of transpassive dissolution) for 500 s of the coatings in 3.5 wt% NaCl solution. After polarization, the corroded surface of the sample was examined by SEM. The wetting behavior of the amorphous coating was investigated through the contact angle (CA) measurements with distilled water using an optical contact-angle meter at ambient temperature. All coating samples were wet ground to 2000 SiC paper and polished using 1-μm suspension of diamond in water, followed by ultrasonically cleaning in alcohol and dried in warm air. For CA measurements, 2 μL water droplets were dropped carefully on to the polished surfaces with a syringe at a very low flow rate from an outlet of a vertical thin tube. The outlet is set parallel to the surface, 2 mm above it. A typical droplet grows slowly and reaches the dry surface within a minute. At least, ten measurements were done for each sample to ensure the data reproducibility. 3 Results 3.1 Characterization of powders and as-sprayed coatings The morphology of the powders with three dimensional ranges is shown in Fig. 1 (a)–(c) . It can be seen that all powders are spherical with smooth surface, indicative of good fluidity. Fig. 1 (d) shows the XRD patterns of powders A, B and C, demonstrating a fully amorphous structure of all powders with different sizes. The DSC curves of the three types of powders are shown in Fig. 1 (e), which display very similar thermal reaction behaviors with three exothermic signals at almost the same temperature ranges from 630 to 750 °C. The onset temperature ( T x ) for crystallization and the total exothermic enthalpy (Δ H x , total ) of the exothermic reactions for the three powders are listed in Table 1 . It can be seen that the values of T x and (Δ H x , total ) are almost identical for the three powders, demonstrated that the powders A, B and C have almost the same structure and amorphous fraction although they have quite different particle size. Fig. 2 (a) and (b) shows the XRD patterns and DSC curves, respectively, of three coatings sprayed from powders A to C and the as-cast rod. The appearance of only broad diffraction hump at the angle of 2 θ = 43° and the absence of any Bragg peaks associated with crystalline phases indicate that the three coatings prepared from different powders as well as the rod are basically amorphous within the resolution limit of XRD. In the DSC curves, it is found that the three coatings demonstrate a very similar thermal behavior as to the spraying powders, i.e., three exothermic crystallization peaks at almost the same temperature ranges with nearly identical crystallization temperature around 630 °C (see Table 1 ). The amorphous rod, however, exhibits a distinct glass transition event at 582 °C, which was not observed for the powders and coatings. In addition, the three exothermic peaks of the rod are slightly shifted to lower or higher temperature regions as compared with the coatings and powders. This difference may result from the difference in the preparation process between rod and powders. It can be seen from Table 1 that (Δ H x , total ) increases with the increase of feedstock powder size, demonstrating that the volume fraction of amorphous phase is higher for the coatings with the coarser powders due to boundary effect, i.e. the coating with the finer powders could possess more interfaces where oxidation most likely occurs and led to the reduction of the crystallization enthalpy. The structure of the cross-section and the surface morphology of different Fe-based amorphous coatings are shown in Fig. 3 . It is found that all the coatings are well-bonded to substrate. Almost a fully dense coating with porosity of about 0.05% was obtained for powder A ( Fig. 3 (a)), which has the smallest size. With the increase of powder size, the porosity of coating increases to 1.9% for powder B and 4.2% for powder C. A few big pores with diameters of about 10–30 μm can be occasionally observed in coating C ( Fig. 3 (c)). The surface morphology reveals that powders C were not fully melted during spraying ( Fig. 3 (f)). This is different from the coating with powder A, which were fully melted with “pancake” morphology (see Fig. 3 (d)). The results imply that the surface morphology of the coatings can be tailored by using different powder sizes. The oxygen content in the three coatings and the as-received powders were examined using an oxygen determinator, which indicated an oxygen content of about 0.07 wt% in all powders, 0.64 wt% in coating A, 0.47 wt% in coating B, and 0.40 wt% in coating C, respectively. Although the coatings contain much more oxygen than the powders, the oxygen content in coating C with the largest particle size is significantly lower than coating A. This difference is due to the fact that the smaller particle size with larger specific surface area leads to higher degree of oxidation during spraying. By calculating the specific surface area of powders A (taking the average diameter of 26 μm) and powders C (taking the average diameter of 50 μm), it is found that the ratio of surface area between powders A and C is 1.92, which is very close to the ratio of oxygen content in coatings A and C (i.e., 1.6). Fig. 4 shows a magnified SEM image of the cross-section structure of coating A, which clearly shows a typical lamellar structure of coatings. The line analysis of composition indicated that the interface between two pancakes is enriched with oxygen. It will be shown in later section that the enrichment of oxygen at the interface plays a crucial role in corrosion resistance of the coatings. To get detailed microstructures of the coatings, TEM observations on coating A and C were carried out. Fig. 5 (a) shows a typical bright field image, involving the matrix and interface between pancakes, of coating A. It can be seen that the matrix is a fully amorphous structure, as indicated by the diffuse diffraction ring in the selected area diffraction (SAD) pattern (see the inset on the top-left corner of Fig. 5 (a)). However, the interface is crystallized with grain size of about 30–100 nm. Fig. 5 (b) shows the dark field image from spot 1 of the SAD pattern (the inset of the lower-right corner in Fig. 5 (a)) of the interface. EDS revealed that the crystallized grains mainly contain 45.06 at% O and 34.15 at% Fe as well as some Cr and Mo. The result implies that the interface of coating A was highly oxidized during spraying and led to the formation of (Fe, M)-oxides (M = Cr, Mo). It is worthy of mentioning that the real value of oxygen may be even higher than 45.06 at% due to the fact that EDX usually underestimates the content of light elements, such as oxygen. In contrast, different microstructures at interfaces were observed in coating C, where the interface exhibits mainly amorphous structure although partial crystallization could be occasionally detected in some other interfaces. EDS shows that the interface mainly contains Fe, O and some Cr, indicating iron oxides but with an amorphous structure are the main phases at the interfaces in coating C. However, the oxygen content at the interface in coating C is significantly lower than that in coating A. Finally, the micro-hardness of the three coatings prepared with different powder sizes was measured using a microindenter, which is also listed in Table 1 . It was found that all coatings have the hardness of about 1000 H V . 3.2 Corrosion behavior 3.2.1 Potentiodynamic polarization test The corrosion behavior of the amorphous coatings was examined with potendiodynamic polarization in 3.5 wt% NaCl solution at 25 °C in open air. The polarization curves of the three coatings in comparison with the as-cast rod sample are shown in Fig. 6 . It can be seen that all the coatings were spontaneously passivated with a wide passivation region of more than 1000 mV and a very low passivation current density on the order of 0.1–1 A m −2 , implying a superior passive film protection, although their corrosion resistance is inferior to the fully amorphous rod, due to the combined effect of oxidation and porosity in coatings. For comparison, the carbon steel substrate was also tested in the same condition, which shows an immediate pitting once the scan goes to the anode polarization range. Table 2 summarizes some important electrochemical parameters attained from the potentiodynamic polarization curves, such as the corrosion potential ( E corr ), corrosion current density ( i corr ), transpassive potential ( E tr ), passivation region ( E pass ) and passivation current density ( i pass ). The polarization resistance ( R p ), which reflects the degree of corrosion reaction, is calculated from the Stern–Geary Equation [21] by fitting the polarization resistance constants for anodic and cathodic reactions in the Tafel polarization curve and is also listed in the table. Among the three coatings, coating C prepared from the coarsest powders exhibits better corrosion resistance than the others, as evidenced by the lowest i corr (2.3 × 10 −2 A m −2 ) and i pass (0.18 A m −2 ). In addition, the polarization resistance ( R p ) for coating C is about 1.8 times higher than that of coating B and 3 times higher than coating A. To clarify the mechanism of corrosion for the coatings, the surface morphology of coating A after breakdown of the passive film was investigated by SEM. Fig. 7 shows the image of corrosive damage when the coating was potendiodynamic scanned to the point where transpassive dissolution just started to occur. It can be clearly seen that the oxidized boundary between particles was preferentially attacked, indicating that the oxidized phase at interface promotes corrosion, which is in good agreement with the previous results. 3.2.2 Potentiostatic polarization test To reveal the details during passivation and its breakdown, potentiostatic polarization measurements were performed. Fig. 8 (a) and (c) shows the typical current density ( di / ds )–time ( t ) curves of the three coatings at a constant applied potential of 300 mV ( vs. SCE, in the passive region) for 10 h and 1500 mV ( vs. SCE, in the transpassive dissolution region) for 500 s in 3.5 wt% NaCl solution, respectively. As can be seen, the passive film on the coating C is more protective and stable than coatings A and B in both passive and transpassive dissolution regions, as indicated by the lower values of current density for coating C in the whole period under the two conditions. The results are in good agreement with potentiodynamic polarization tests. Concerning the details of the potentiostatic polarization behavior under the applied potential of 300 mV ( Fig. 8 (a)), it is found that the current density displays a continuous decay with time for the three coatings, reflecting the formation and growth of the passive layers [22–24] . The curve for coating C is almost linear with a gentle slope in the time regime less than 10 3 s, and a slope of around −0.4 beyond that, indicating a diffusion mechanism for the formation of the passive film in coating C [22] . In contrast, the curve for coating A decays at a steep slope of around −0.5 in less than 10 3 s, and then transfers into a nearly constant level up to 10 4 s, indicating a balance between growth and dissolution of the passive film at the middle stage. Finally, it decays at a slope of around −0.75, demonstrating a high-field mechanism for coating A at the final stage [22,24] . Fig. 8 (b) is the enlargement of the current ( i ) − time ( t ) curve of coating A from the time scale of 5 × 10 3 to 1 × 10 4 s, which clearly shows a few current transients with heights of 3.4–8.3 μA in the potentiostatic polarization. Generally, the current transient represents the nucleation, growth and repassivation of pits [25–27] . These current transients indicate the occurrence of metastable pits during the process of potentiostatic polarization in coating A, as characterized by a sharp rise in current followed by a slow decay to the level well below the pre-transient current background for each of the transient signals. The fast rise in current refers to the growth of pit, while the decay of current represents the repassivation of the pits. However, for coating C, no significant current transients were detected under the detection limits in the whole process of potentiostatic polarization, indicating that such metastable pits as formed in coating A were hardly formed in this case. As far as the applied potential of 1500 mV (in the transpassive dissolution region) is concerned (see Fig. 8 (c)), the current density of three coatings remains at a high level above 10 A m −2 , reflecting the occurrence of transpassive dissolution [28,29] in all coatings. Similar to the case in the passive region, the current density is again the lowest for coating C, but the highest (>400 A m −2 ) for coating A, demonstrating that the damaged area on coating C is less than on coatings A and B. The increase in the current density with polarization time for the coatings implies the increase in corroded area. On the other hand, the slope of around 0 for the curves of all coatings at 1500 mV in the initial period of time could be attributed to the evolution of O 2 [22] . After a period of 10 s, the slopes of the three coatings all became positive which reveals the strong dissolution of the passive films. Fig. 9 (a) and (b) shows the morphologies of coating A and C after potentiostatic polarization at 1500 mV for 500 s. Consistent with the above results, a much stronger attack occurred leaving clear witness corrosion mark for coating A as compared to coating C, further demonstrating that coating C prepared with larger particles has a better corrosion resistance. 4 Discussions The investigation of the effect of the particle size of the feedstock powders reported here indicates that, the major change in microstructure for as-deposited coatings is associated with the melted state and oxidation of powders during spraying. The microstructure of HVOF coating is largely dependent on the temperature and velocity of in-flight particles during spraying. At the same spraying parameters, the thermal energy of the flame is fixed; increase in the powder particle size will lead to a decrease in the particle velocity and temperature at impact [30] . Thus powders A and B with smaller dimensions got well-melted, while powders C just got partially melted at the surface. As a result, the densest coating A was obtained by using the finest powders. The electrochemical results illustrate that the size of the feedstock powders also had a significant effect on the corrosion resistance of the coatings in 3.5 wt% NaCl solution. The coating prepared with coarser powders and larger porosity exhibits superior corrosion resistance than the coating prepared with finer powders and lower porosity. The corrosion resistance in coating is actually a very complex system associated with porosity, crystallization, oxidation and surface wettability, etc. However, the present results indicate that all coatings have a similar amorphous structure without significant crystallization, and contrary to anticipation, the porosity in coatings seems to have a reverse effect on corrosion resistance in the present study, as coating C with the highest porosity (4.2%) exhibits the best corrosion resistance. In order to identify the mechanism for the difference in corrosion resistance of various coatings, the surface wettability of the coatings was examined, as it has been reported previously that the wettability has a significantly effect on the corrosion resistance of bulk metallic glass (BMG) [31,32] . The wettability of the coatings is usually characterized by contact angle (CA) with water. The measurement yielded an average contact angle of 64 ± 8°, 90 ± 3° and 100 ± 2° for coating A, B and C, respectively. This indicates that coating A is hydrophilic, coating C is hydrophobic and coating B is in between. Fig. 10 shows the correlation between the contact angle and the polarization resistance ( R p ), demonstrating clearly that larger contact angle (i.e., higher hydrophobicity) leads to larger polarization resistance. Similar phenomenon was reported recently for a Zr-based BMG in simulated body fluid [31] . The question then arises as to why coating C has a higher contact angle than the other two coatings made from the finer powders. Since the nominal composition of the three coatings is the same, the most likely causes could be the difference in porosity and oxygen content between coating C and coatings A and B. Coating C has more micro-scale pores in the polished surface than coating A and B (see Fig. 3 ), and the contact angle could be enhanced as a results of the effect of the as-trapped superhydrophobic vapor pockets. This result implies that the solid/liquid contact mode in the coatings is somehow close to the Cassie's state, in which the surface roughness (here, the roughness is associated with the porosity and size of pores of the coatings) reduces the wettability and increases the contact angle [33] . However, if the pore is big enough to allow the absorption of water solution, the corrosion resistance would be deteriorated. Further work is essentially required for the relationship between the pore size and hydrophobicity of the coatings. Another factor affecting contact angle in the present study is probably the oxygen content. It is well know that surface oxidation usually reduces the water contact angle on metal substrate [34] . The experimental result revealed that coating C involves the lowest oxygen content (0.4 wt%) at the interface of non-melted particles. The iron and chromium oxides are both hydrophilic due to generation of a hydroxide-containing compound (or hydroxyl group) more or less on the surface of such metal oxides [35] . Therefore, the smaller amount of oxygen content in coating C could also lead to a larger contact angle and higher hydrophobicity. Regarding the relation between hydrophobicity and corrosion resistance of the amorphous coatings, it has been previously reported that hydrophobicity can enhance the corrosion resistance in a Zr-based metallic glass [31] . As the hydrophobic surface usually has a lower surface energy, which reduces the contact area of solid surface in contact with the water solution. As a result, the probability of Cl − ion attacks on metal surface could be significantly reduced. On the other hand, the reduction of wettability can also promote the repassivation of the pits that were previously formed. The mechanism of the enhancement of corrosion resistance by forming hydrophobic (or superhydrophobic) film has been reported on Cu-, Al-metals and Mg alloy [36–38] . It should be noted that the formation of oxides not only reduces the hydrophobicity, but also deteriorate directly the corrosion resistance themselves. Because the metal oxides formed can hinder the formation of passive film, and even lead to diffusion channels for electrolyte to cause inner corrosion. It has been previously reported that the oxide contours in thermally sprayed coating are prone to be the favorable sites for microgalvanic and/or microcrevice corrosion [39] , leading to an easy penetration of chloride ions and inner corrosion of the coatings. This is actually what was observed in the present study, as shown in Fig. 7 where it is shown clearly that corrosion always occurred along the oxide-rich interfaces between pancakes in coating A with more oxygen content. Similar phenomenon was also found in our previous work [39] . 5 Conclusions Fe-based coatings with almost fully amorphous structure have been successfully fabricated by HVOF process. The particle size of the feedstock powders has a significant effect on the microstructure and corrosion resistance of the coatings. The finer powders led to a denser coating, while the coarser powders caused a relatively loose coating, but better corrosion resistance. These seemingly paradoxical variations in corrosion resistance are due to the differences in wetting behavior and oxygen content among different coatings. Coating C prepared with the coarse powders exhibits a contact angle of 100 ± 2°, representing hydrophobic character, which can prevent the coating from ion contact and thus lead to better corrosion resistance, while coating A made from the finest powders exhibits a hydrophilic property and relatively poorer corrosion resistance. The wettability of the coating is closely related to the porosity and oxygen content. The higher porosity in a certain range (i.e. higher roughness) and less oxygen content increase the contact angle, and therefore better corrosion resistance. On the other hand, the high oxygen content, which results in the formation of oxides at the interfaces, could also degrade the corrosion resistance itself due to the fact that these phases can affect the formation of passive film and even cause inner corrosion of attached ions. The present work provides a unique perspective on amorphous coatings with hydrophobicity and excellent corrosion resistance, for promising industrial applications in marine environment. Acknowledgments This work was financially supported by the National Fundamental Research Program of China (Grant No. 2007CB613908 ) and National Natural Science Foundation of China (Grant No. 50871042 ). This work was also partially supported by National Magnetic Confinement Fusion Program , China ( 2010GB109000 ) and the internal funding from Huazhong University of Science and Technology ( 2010ZD003 ). C. Zhang thanks N. Li and Q. Chen for the preparation of the amorphous rods. Authors are also grateful to the Analytical and Testing Center, Huazhong University of Science & Technology for technical assistances. References [1] K. Kishitake H. Era F. Otsubo J. Therm. Spray Technol. 5 1996 476 [2] F. Otsubo H. Era K. Kishitake J. Therm. Spray Technol. 9 2000 494 [3] F. Otsubo K. Kishitake Mater. Trans. JIM 46 2005 80 [4] S. Kumar J.S. Kim H.J. Kim C.H. Lee J. Alloys Compd. 475 2009 L9 [5] J.B. Cheng X.B. Liang B.S. Xu Y.X. Wu J. Mater. Sci. Technol. 25 2009 687 [6] X.Q. Liu Y.G. Zheng X.C. Chang W.L. Hou J.Q. Wang Z. Tang A. Burgess J. Alloys Compd. 484 2009 300 [7] J.C. Farmer J.-S. Choi C.K. Saw J.J. Haslam S.D. Day P.D. Hailey T. Lian R.B. Rebak J.H. Perepezko J.H. Payer D.J. Branagan B. Beardsley A. D’amato L.F. Aprigliano Metall. Mater. Trans. A 40 2009 1289 [8] J.C. Farmer J.J. Haslam S.D. Day T. Lian C.K. Saw P.D. Hailey J.-S. Choi R.B. Rebak N. Yang J.H. Payer J.H. Perepezko K. Hildal E.J. Lavernia L. Ajdelsztajn D.J. Branagan E.J. Buffa L.F. Aprigliano J. Mater. Res. 22 2007 2297 [9] S.H. Yoon J.H. Kim G. Bae B.D. Kim C.H. Lee J. Alloys Compd. 509 2011 347 [10] J. Voyer J. Therm. Spray Technol. 19 2010 1013 [11] A. Singh S.R. Bakshi A. Agarwal S.P. Harimkar Mater. Sci. Eng. A 527 2010 5000 [12] Z. Zhou L. Wang F.C. Wang H.F. Zhang Y.B. Liu S.H. Xu Surf. Coat. Technol. 204 2009 563 [13] D. Zois A. Lekatou M. Vardavoulias J. Alloys Compd. 504 2010 S283 [14] J. Kawakita T. Fukushima S. Kuroda T. Kodama Corros. Sci. 44 2002 2561 [15] M.M. Verdian K. Raeissi M. Salehi Corros. Sci. 52 2010 1052 [16] L. Pawlowski The Science and Engineering of Thermal Spray Coating, vol. 2 2008 John Wiley & Sons Ltd. England ch 3 [17] D.J. Branagan M.C. Marshall B.E. Meachan L.F. Apriliano R. Bayler E.J. Lemieux T. Newbauer F.J. Martin J.C. Farmer J.J. Haslan S.D. Day Thermal Spray Conference 2006 2006 ASM International Materials Park, OH e-proceeding [18] C. Suryanarayana A. Inoue Bulk Metallic Glasses, vol. 1 2010 Taylor & Francis Group Boca Raton ch 7 [19] T Kairet M. Degrez F. Campana J. -P. Janssen J. Therm. Spray Technnol. 16 2007 610 [20] S.J. Pang T. Zhang K. Asami A. Inoue Acta. Mater. 50 2002 489 [21] ASTM Standard G59-97, 2009, Standard Test Method for Conducting Potentiodynamic Polarization Resistance Measurements, ASTM International, West Conshohocken, PA, 2009, doi:10.1520/G0059-97R09, www.astm.org . [22] S.J. Pang C.H. Shek T. Zhang K. Asami A. Inoue Corros. Sci. 48 2006 625 [23] J.W. Schultze M.M. Lohrengel D. Ross Electrochim. Acta 28 1983 973 [24] A. Gebert V. Haehnel E.S. Park D.H. Kim L. Schultz Electrochim. Acta 53 2008 3403 [25] G.T. Burstein P.C. Postorius S.P. Mattin Corros. Sci. 35 1993 57 [26] A.M. Riley D.B. Wells D.E. Williams Corros. Sci. 32 1991 1307 [27] M.H. Moayed R.C. Newman Corros. Sci. 48 2006 1004 [28] S.J. Pang T. Zhang K. Asami A. Inoue Corros. Sci. 44 2002 1847 [29] J. Jayaraj K.B. Kim H.S. Ahn E. Fleury Mater. Sci. Eng. A 449–451 2007 517 [30] L. Pawlowski The Science and Engineering of Thermal Spray Coating, vol. 2 2008 John Wiley & Sons, Ltd. England ch 5 [31] Y.B. Wang H.F. Li Y.F. Zheng S.C. Wei M. Li Appl. Phys. Lett. 96 2010 251909 [32] K. Zhao K.S. Liu J.F. Li W.H. Wang L. Jiang Scr. Mater. 60 2009 225 [33] X.J. Feng L. Jiang Adv. Mater. 18 2006 3063 [34] K.T. Hong H. Imadojemu R.L. Webb Exp. Therm. Fluid Sci. 8 1994 279 [35] H.-J. Schlinsog, The Continental Group, Inc. Stamford, Conn, Assignee. US patent 4292095 (September 1981). [36] T. Liu Y.S. Yin S.G. Chen X.T. Chang S. Cheng Electrochim. Acta 52 2007 3709 [37] Y.S. Yin T. Liu S.G. Chen T. Liu S. Cheng Appl. Surf. Sci. 255 2008 2978 [38] J. Wang D.D. Li Q. Liu X. Yi Y. Zhang X.Y. Jing M.L. Zhang Electrochim. Acta 55 2010 6897 [39] R.Q. Guo C. Zhang Q. Chen Y. Yang N. Li L. Liu Corros. Sci. 53 2011 2351