Effects of Laser Scanning Speed on Microstructure, Microhardness, and Corrosion Behavior of Laser Cladding Ni45 Coatings

School of Materials Science and Engineering, Jiangsu University of Science and Technology, Zhenjiang 212003, China CAS Key Laboratory of Nuclear Materials and Safety Assessment, Institute of Metal Research, Chinese Academy of Sciences, Shenyang 110016, China Guangdong Key Laboratory for Technology and Application of Metal Toughening, Guangdong Institute of Materials and Processing, Guangzhou 510650, China


Introduction
Cladding technologies are usually employed to modify the surface structure and properties of alloys [1,2], among which laser cladding exhibits some promising advantages [3][4][5]. First, it is a simple, green, economic, and efficient process, which can make the surface coating have better corrosion resistance and wear resistance than the substrate [6,7]. Secondly, a strong metallurgical bond can form between the coating and substrate [4,8,9], and the size of the heat-affected zone (HAZ) is usually small [4,10,11]. Lastly, it is relatively easy to control the process parameters and to be automated [4,12,13]. Liu et al. [5] developed a composite coating by laser cladding on Ti-6Al-4V alloy with superior wear resistance. e results showed that the average microhardness and wear loss of the coating were twice larger and 10∼30% lower than those of the substrate, respectively. Fesharaki et al. [14] reported better metallurgical bonding of Inconel 625 coating prepared by laser cladding than that developed by Tungsten Inert Gas Welding (TIG) cladding. Tanigawa et al. [15] found that the HAZ of laser cladding prepared Ni-Cr-Si-B alloy coating on C45 carbon steel can be shrunken by employing smaller particles.
Nickel-based superalloys are widely used in the laser cladding process due to their superior mechanical properties and wear resistance [16][17][18]. Chen et al. [3] found that the microhardness of the Ni-Cr-B-Si composite coatings increased with increasing laser scanning speed (5,10,15, and 20 mm/s). Sun et al. [19] studied the effects of NbC on the wear properties of Ni45 coating and reported that the microhardness and wear resistance were significantly improved by the addition of NbC. However, these works are mainly focused on the hardness and wear resistance with very little attention on corrosion behavior. Keeping in mind that corrosion is usually one of the reasons for the reduction in the service life of nickel-based alloys, it is, therefore, worthwhile to evaluate the corrosion resistance of nickelbased alloy coatings prepared by laser cladding, especially to clarify the correlation between corrosion resistance and microstructure.
For this purpose, three laser cladded nickel-based alloy coatings on carbon steel substrate were prepared, adjusting the microstructure by controlling the laser scanning speed (4 mm/s, 6 mm/s, and 10 mm/s). e effects of laser scanning speed on the microstructure, microhardness, and corrosion behavior were investigated by optical microscopy (OM), scanning electron microscopy (SEM), X-ray diffraction (XRD), microhardness, and electrochemical measurements. e evolution of microhardness and corrosion behavior with laser scanning speed was explained in terms of microstructure.

Preparation of Ni45 Coatings.
e substrate material used in this study was medium carbon steel AISI 1045 (0.42∼0.50 wt.% C, 0.17∼0.37 wt.% Si, 0.50∼0.80 wt.% Mn, 0.25 wt.% Cr, ≤0.25 wt.% Cr, ≤0.30 wt.% Ni, ≤0.25 wt.% Cu, ≤0.035 wt.% P and Fe balance) with dimensions 110 mm × 100 mm × 10 mm. e surface of the substrate was ground with 150# sandpaper and cleaned with acetone. Ni45 powder (0.34 wt.% C, 2.0 wt.% B, 4.1 wt.% Si, 12.5 wt.% Cr, 6.6 wt.% Fe and Ni balance) obtained from Beijing AMC Powder Metallurgy Technology Co., Ltd. (Beijing, China), was employed as the cladding material, and its SEM morphology is shown in Figure 1. Before laser cladding, the alloy powder was baked in a vacuum environment (to remove moisture and avoid oxidation) and then ground in a mortar for at least 5 minutes to guarantee uniformity in powder mixing. CO 2 laser with Siemens 820C computer numerical control (CNC) system was used in preparing the coating. e schematic of the laser cladding process employed in this work is shown in Figure 2, and the operating parameters are listed in Table 1. Before the experiment, the cladding powder and bonder were preplaced on the surface of AISI 1045 steel by a self-made powder paving device. After the laser cladding, the samples were cut into coupons with dimensions 10 mm × 10 mm × 5 mm. en, the samples were cleaned and inlaid for subsequent microstructure observation and property testing.

Microstructure Observation and Microhardness
Measurement.
e crystal structures of the Ni45 coatings were investigated by an X-ray diffractometer (XRD, Bruker D8 Advance) with the diffraction angle (2θ) ranging from 10°to 90°and a Cu-Ka radiation at 40 kV and 35 mA at a step size of 0.02°and a scan rate of 6°/min. e microstructures of the as-prepared coatings were characterized by an optical microscope (OM, KEYENCE-VHX-900E) and scanning electron microscope (SEM, XL30-FEG ESEM) with energy dispersive spectroscope (EDS, Oxford INCA) after the samples were mechanically polished and etched in the 3 wt.% HNO 3 solution. e microhardness distribution along the cross section of the Ni45 coatings was measured using HDX-1000 microhardness tester with a load of 300 g and a dwell time of 15 s.    e electrochemical corrosion behavior of the as-prepared coatings was assessed using CS350 electrochemical workstation (Wuhan Corrtest Instruments Corp., Ltd., Wuhan, China) in 0.5 M NaCl solution and 0.5 M NaCl + 0.05 M H 2 SO 4 solution. All the electrochemical tests were conducted using a three-electrode electrochemical cell composed of a saturated calomel reference electrode (SCE), a platinum counter electrode, and a coated sample working electrode [20]. Preceding the electrochemical tests was the monitoring of the open circuit potential (OCP) for 1 h to achieve a relatively stable state. e electrochemical impedance spectroscopy (EIS) tests were performed at OCP with a sinusoidal potential perturbation of 10 mV in a frequency range from 10 5 Hz to 10 −2 Hz. e potentiodynamic polarization curves were measured by sweeping the potential from −0.6 V SCE to 2.0 V SCE at a scanning rate of 0.5 mV/s and terminated when the current density of 50 mA/cm 2 was reached. e electrochemical tests were conducted at 25 ± 1°C in the air without stirring, and triplicate measurements were done to ensure repeatability. Figure 3 shows the surface appearance of Ni45 coatings after laser cladding processes. e coating material was melted sufficiently, and continuous coatings were formed on the substrate. All coatings had a uniform geometric appearance, and the overlaps between adjacent passes were almost parallel to each other. Figure 4 displays the OM of the transition layers between the coatings and substrate. All the interfaces were free of cracks and pores, indicating good bondings between the substrate and coatings. e thickness of the transition layer between the coating and HAZ decreased with increasing laser scanning speed, which resulted from the hindered element diffusion between the coatings and substrate at a shorter heating time during the laser cladding process [8,21,22]. Many columnar grains appeared around the transition layer, due to extreme undercooling at the bottom of the molten pool [23]. Meanwhile, the direction of columnar growth was basically perpendicular to the transition layer along the direction opposite to the thermal flow movement [8,13,23,24]. ese features of columnar grains and microstructures are basically independent of the laser scanning speed. e cross-section microstructures of the asprepared Ni45 coatings were observed by SEM in backscattered electron mode and are presented in Figure 5. As seen in Figures 5(d)-5(f ), more and finer dendrites appeared as the laser scanning speed increased from 4 mm/s to 10 mm/s, demonstrating that the grains tend to grow into finer dendrites at higher laser scanning speed. Also, many precipitation phases were observed in the intergranular areas for all the coatings [25].

Morphology Observation.
According to the solidification theory, the alloy solidification mode mainly includes planar, cellular, dendritic, and equiaxed growth, which depends on the ratio of the temperature gradient (G) to solidification rate (R), i.e., G/R [13,21,24,26,27]. e temperature gradient was the largest at the beginning of the solidification process due to the low temperature of the substrate and the fact that the solidification rate was close to zero. Under this condition, the solidliquid interface is stable, and a layer of planar front growth can form between coatings and substrates, just like the transition layer shown in Figure 3 [13,21,26,27]. With the increase in distance from the bottom of the molten pool, the temperature gradient will decrease [26], and the solidification speed will increase [8,24], resulting in a dramatic decline of G/R. As a result, the solid-liquid interface of the planar grain becomes unstable, promoting the transition of microstructure from plane crystals to cellular crystals [8,21,26]. With further movement of the solid-liquid interface far away from the bottom of the molten pool, the value of G/R becomes much smaller, resulting in the change of crystal growth mode from cellular to dendritic growth [26,27], as shown in Figure 5. At the top of the molten pool, the dendrites transformed into equiaxed crystals due to the further decline of G/R caused by the heat released to the surrounding environment from multiple directions [21,26,27]. Accordingly, increasing the laser scanning speed will affect the microstructure evolution during the solidification process from two aspects. On the one hand, the value of G/R will decrease, which accelerates the transformation from planar crystals to cellular crystals and dendrites consequently [26,28]. On the other hand, the grains nucleate abundantly due to the lower heat input and the more rapid solidification, resulting in the refinement of grains [13,21,22,26,29]. As a result, more and finer dendrites can be observed in the coatings prepared at higher laser scanning speed, as shown in Figure 5. Figure 6 exhibits the XRD patterns of the as-prepared Ni45 coatings. e characteristic peaks correspond to the c-(Ni, Fe), FeNi 3 , and M 23 (C,B) 6 , respectively. ereinto, the intensity of M 23 (C,B) 6 peak showed the most noticeable difference among the three kinds of coatings, which decreased with the increase in laser scanning speed and even disappeared at 10 mm/s. Kesavan et al. [30] studied the microstructure characterization of a nickel-base hard-faced coating and reported that the microstructure of the deposits can be divided into dendrite and interdendritic regions. e dendrite region was composed of the c-nickel phase, while the interdendritic region was composed of Cr-rich carbide and boride precipitates that are needle-, blocky-, floret-, and fine spherical-shaped. Liu et al. [31] investigated the microstructure of NiCoCrAlY coating deposited on cast iron using multilayer laser cladding. ey described the dendrite in NiCoCrAlY alloy coating as the c-(Fe,Ni) solid solution, and the phases in the interdendritic region as M 7 C 3 and M 23 C 6 carbides, where M represents Cr, Fe, or Ni.

XRD and EDS Measurements.
It can be inferred that the dendrite region was composed of c-phase, whereas the black particle phases in the interdendritic region were M 23 (C,B) 6 precipitations. Meanwhile, the intensity of M 23 (C,B) 6 peak was not noticeably observed in the XRD pattern of C2 and C3 coatings, which reflects that the volume fraction of M 23 (C,B) 6 precipitations decreased with increasing laser scanning speed [32].
is can be attributed to the high cooling rate induced by the high laser scanning speed [3,28]. e inhibition of M 23 (C,B) 6 , together with the refined grain ( Figure 5), could have effects on the microhardness and corrosion resistance of the Ni45 coatings [13,30], which will be discussed in the following sections. e elemental content of the dendrites in the C3 coating measured by EDS is listed in Figure 7. Only the contents of Fe, Cr, Ni, and Si were detected in the C3 coating because of the poor sensitivity of EDS in detecting light elements [13,33]. Compared with the original Ni45 powders, the content of element Si, Cr, and Ni changed less after the laser cladding process, but a remarkable increase in the Fe content was observed. It indicates the element diffusion occurring between the substrate and the coating during the laser cladding process [34] and, therefore, demonstrating the formation of the transition layer discussed in Section 3.1.  Figure 8 shows the microhardness distribution along the cross section of the as-prepared Ni45 coatings, in which the height and width of the rectangle represent the microhardness and the depth of different regions, respectively. e microhardness of the Ni45 coating ranges from 313.8 HV to 556.8 HV, which is significantly higher than that of the substrate. With increasing laser scanning speed, the microhardness of the coating decreased, while that of the substrate remained relatively unchanged. On the one hand, the higher laser scanning speed inhibited the formation and growth of borides and carbides, i.e., M 23 (C,B) 6 in Figure 6, which weakened the effect of precipitation strengthening [12,13]. On the other hand, the grains were refined due to the higher cooling speed (Figure 5), leading to the improvement of microhardness [3,13,22]. Based on these two aspects, precipitation strengthening should be the dominant factor determining the microhardness of Ni45 coatings studied in this work [13]. Besides, the as-prepared coatings can be further divided into two parts based on the obvious step change of coatings microhardness: the top and the bottom coatings. e microhardness of the top coatings was lower than that of the bottom when the laser scanning speed was 4 mm/s, while the relationship reversed at 6 mm/s and 10 mm/s. During the solidification process, the solidification phase transition occurred first at the bottom and then advanced to the top [28]. With continuous heating from the postsolidified metal droplets, the bottom microstructure grew coarsely [21,23], whereas the top microstructure solidified rapidly and had no time to grow more coarsely due to the cooling effects of the external environments [21,26]. Meanwhile, at low laser scanning speed (4 mm/s), there was enough time for the growth of carbides and borides, especially at the bottom, resulting in a higher microhardness of bottom coatings [3,28]. e precipitation strengthening, in other words, should be the key factor determining the microhardness distribution inside the coating region at low scanning speed (4 mm/s). e higher the laser scanning speed, the fewer the amount of the precipitate of carbides and borides. As a result, the main Journal of Chemistry strengthening mechanism influencing microhardness distribution could change from precipitation strengthening to grain refinement strengthening, leading to a higher microhardness of top coatings at high scanning speeds (6 mm/s and 10 mm/s).

Microhardness Measurements.
In conclusion, whether the strengthening mechanism is precipitation strengthening or grain refinement strengthening, the coating microhardness can be improved effectively. For the whole coatings, the precipitation strengthening made more contribution to enhancing the microhardness of the coatings compared with the grain refinement strengthening. For hardness distribution inside the coating region, the controlling strengthening mechanism depended on the laser scanning speed. When the laser scanning speed was as low as 4 mm/s, the microhardness distribution was mainly influenced by the precipitation strengthening, whereas the grain refinement strengthening was the key factor determining the microhardness distribution as the laser scanning speed was increased to 6 mm/s and 10 mm/s.

Electrochemical Corrosion Behavior.
e potentiodynamic polarization curves of the Ni45 coatings in 0.5 M NaCl and 0.5 M NaCl + 0.5 M H 2 SO 4 solution are displayed in Figure 9. e values of corrosion current density (i corr ) and corrosion potential (E corr ) based on Tafel extrapolation analysis are listed in Table 2. In both two solutions, the anodic current densities of all the coatings increased with the applied potential without exhibiting any active-to-passive transition. In 0.5 M NaCl solution, the corrosion potentials shifted nobly while the corrosion current densities decreased with the increase in laser scanning speed. It proves that the corrosion resistance of coatings in 0.5 M NaCl solution is improved by increasing the laser scanning speed [20,28,34,35]. In contrast, lower corrosion potentials and higher corrosion current densities were observed for all coatings in 0.5 M NaCl + 0.5 M H 2 SO 4 solution, implying a poor corrosion resistance in the more acidic environment. Moreover, the laser scanning speed seems to have no obvious effects on the corrosion resistance of Ni45 coatings in 0.5 M NaCl + 0.5 M H 2 SO 4 solution, as shown in Figure 9(b) and Table 2. Figure 10 depicts the Nyquist plots of the as-prepared Ni45 coatings in 0.5 M NaCl and 0.5 M NaCl + 0.5 M H 2 SO 4 solutions. In 0.5 M NaCl solution (Figure 10(a)), one apparent unfinished capacitive impedance arc can be identified for all coatings. e radius of the capacitive impedance arc increased with the laser scanning speed, indicating the   Journal of Chemistry improvement in corrosion resistance [26,28,[36][37][38][39]. However, the characteristics of Nyquist plots in 0.5 M NaCl + 0.5 M H 2 SO 4 solution ( Figure 10(b)) are different from those in 0.5 M NaCl solution (Figure 10(a)), which exhibited capacitive impedance arcs with smaller radius but fewer discrepancies among all three coatings, suggesting deteriorated but undifferentiated corrosion resistance. e EIS results are consistent with those of potentiodynamic polarization tests (Figure 9 and Table 2). e above effects of laser scanning speed on corrosion resistance should be related to the microstructure evolution of Ni45 coatings. It has been reported that the formation of Cr-rich precipitated phases can cause Cr depletion in the matrix around these precipitates, deteriorating the corrosion resistance in these regions [40][41][42][43]. Lei et al. [26] studied the corrosion resistance properties of carbon fiber (CF) reinforced Ni-based composite coating by laser cladding. e results showed that the addition of CFs decreased the amount of M 7 C 3 and M 23 C 6 carbides, restraining the formation of Cr-depleted regions at the grain boundaries, thus improving the corrosion resistance of Ni-based alloy composite coatings. Accordingly, the schematic diagrams illustrating the corrosion mechanisms of Ni45 coatings in 0.5 M NaCl and 0.5 M NaCl + 0.5 M H 2 SO 4 solutions are exhibited in Figures 11(a) and 11(b), respectively. In 0.5 M NaCl solution (Figure 11(a)), the precipitate and coatings acted as the cathodes due to their nobler potential, while the Cr-depleted zone around the precipitates with lower potential acted as the anode [40,42]. In this case, the Crdepleted zone dissolved faster than the other zones because of the corrosion microcell, and some precipitates fell off from the matrix. When the laser scanning speed was low, there was relatively sufficient time for the reaction between Cr and C element, and more M 23 (C,B) 6 formed and grew ( Figure 6) leading to severer depletion of Cr around the precipitates. In contrast, less and even no M 23 (C,B) 6 precipitates formed at higher laser scanning speeds owing to the faster cooling rate ( Figure 6). As a result, the Cr-depleted zone and the corresponding corrosion microcell were diminished, which inhibited the corrosion process to some extent. Nonetheless, this inhibition effect of laser scanning speed on corrosion resistance did not exhibit tangible influence in the 0.5 M NaCl + 0.5 M H 2 SO 4 solution, as shown in Figure 11(b). It is expected that both the anodic and cathodic reactions will accelerate in strong acidic solutions. In this case, all the regions on the surface, including the matrix, precipitates, and the Cr-depleted zones, dissolved rapidly, implying that the coating surface corroded uniformly layer by layer (Figure 11(b)). As a consequence, the effect of corrosion microcell on the local falloff of precipitates was eliminated. erefore, the laser scanning speed had less effect on the corrosion resistance of Ni45 coatings. e mechanism shown in Figure 11(a) can be further verified by the variation trend of E corr and i corr with laser scanning speed ( Figure 9 and Table 2), which is explained by the means of the mixed potential theory as illustrated in Figure 12 [44,45]. According to the analyses in Figure 11, the effect of laser scanning speed on corrosion resistance was dominated by selective dissolution of Cr-depleted zone, which suggests that the anodic rather than the cathodic      reaction rate should differ at various laser scanning speeds. It correlates well with the fact that the difference in anodic curves was more noticeable compared with that in the cathodic curves in 0.5 mol/L NaCl solution as shown in Figure 9(a). erefore, only one oblique line (c 1 ) is used to represent the unchanged cathodic curve, while three lines (a 1 , a 2 , and a 3 ) are used to represent the evolved anodic curves at various laser scanning speeds in Figure 12. e point A, the intersection of cathodic curve c 1 and anodic curve a 1 , is assumed to be the electrochemical corrosion state of C1 coating (laser scanning speed of 4 mm/s), at which the corrosion potential and corrosion current density are E corr1 and i corr1 , respectively. On increasing the laser scanning speed to 6 mm/s (C2 coating) and 10 mm/s (C3 coating), less amount of M 23 (C,B) 6 precipitates formed and the Cr-depleted zones were diminished, consequently leading to inhibited anodic reactions. As a result, the anodic curve will move left along with the axis of current density, from a 1 to a 2 and a 3 , whose intersections are point B (E corr2 and i corr2 ) and point C (E corr3 and i corr3 ), respectively. Evidently, E corr1 < E corr2 < E corr3 , while i corr1 > i corr2 > i corr3 . e orders for corrosion potential and corrosion current density are identical to those identified in Table 2. erefore, it should be reasonable to conclude that the improved corrosion resistance of Ni45 coatings in 0.5 M NaCl solution is attributed to the inhibited formation of M 23 (C,B) 6 precipitates at higher laser scanning speeds.

Conclusions
Ni45 coatings with good interface bonding have been prepared by the laser cladding process. e effects of laser scanning speed (4 mm/s, 6 mm/s, and 10 mm/s) on the microstructure, microhardness, and corrosion resistance of the as-prepared Ni45 coatings are clarified and explained: (1) e thickness of the transition layer between the coating and HAZ decreased the transformation from planar crystals to dendrites, and the grain refinements in the coatings were promoted. e formation and growth of M 23 (C,B) 6 precipitates in the coatings were inhibited with increasing laser scanning speed due to lower thermal input and faster cooling rate. (2) e microhardness of all the coatings presents a downward trend with the increase in laser scanning speed, which is attributed to the weakened precipitation strengthening resulting from the reduced amount of M 23 (C,B) 6 precipitates. Inside the coating, two distinct layers were identified with different microhardness. e position of the layer with the highest microhardness depended on the laser scanning speed. e competition between precipitation strengthening and grain refinement strengthening accounted for the microhardness distribution in the cross section of the coatings at various laser scanning speeds. (3) e laser scanning speed exhibited a promotion effect on the corrosion resistance of the as-prepared Ni45 coatings in 0.5 M NaCl solution, whereas no effects could be identified in 0.5 M NaCl + 0.5 M H 2 SO 4 solution. e former is ascribed to the inhibited formation of Cr-depleted zones around the M 23 (C,B) 6 precipitates at higher laser scanning speeds. In contrast, this inhibition effect was invalid in the presence of 0.5 M H 2 SO 4 because of the highly corrosive feature of the acid solution.
Data Availability e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.