Microstructure and high temperature sliding wear behaviour of laser remelted NiCrBSi/35(WC–12Ni) composite coating

Plasma spraying process and laser remelting process

Commercially available mixture of 65 wt-% self-fluxing NiCrBSi alloy powders and 35 wt-% nickel clad WC powders (12 wt-%Ni and balance WC), prepared via the mechanical mixing method and provided by Beijing General Research Institute of Mining and Metallurgy (Beijing, China), were used as original feedstock powders. The nominal compositions of the self-fluxing Ni based alloy powders are as below: 15 wt-%Cr, 4·5 wt-%Si, ⩾15 wt-%Fe, 0·8 wt-%C, 3·1 wt-%B and balance Ni. Before spraying, disc shaped 1Cr18Ni9Ti stainless steel substrates with a size of φ45×16 mm were sand blast and cleaned in an ultrasonic bath with acetone so as to ensure good adhesion between substrate and coating. And a commercial APS-2000 air plasma spraying facility (China) was used to deposit NiCrBSi/WC–Ni coating. The spraying parameters used were current of 500 A, voltage of 68 V, powder rate of 20 rev min⁻¹ (∼30 g min⁻¹) and spraying distance of 150 mm, using Ar/H₂ with a ratio of 50∶10 as the source gas. The thickness of the as sprayed coating was ∼300 μm.

Laser remelting was conducted on a commercial 10 kW transverse flow continuous wave CO₂ laser material processing system, equipped with a four-axis computer numerically controlled machine tool for laser material processing. The processing parameters of laser remelting were set at a continuous power of 3000 W, power density of 60 W mm⁻² pieces and beam scanning speed of 5 mm s⁻¹. A 10 mm wide track was performed in successive scans on the as sprayed coating surface with a 3 mm overlap, using Ar gas as the protective gas.

Friction and wear test

The above as sprayed and as remelted specimens were cut into cylindrically shaped specimens with a size of φ24×7·9 mm, and the resulting specimens were mechanically ground with graded emery papers and polished with diamond paste to obtain a mirror finish surface (R_a⩽0·1 μm). Friction and wear tests were carried out in a ball on disc contact configuration using an SRV tribometer (Optimol, Monchengladbach, Germany). Commercially available Si₃N₄ balls (φ10 mm,1700 HV_2N) were used as the upper specimens, and the discs coated with coatings as the lower specimens. Before tribological tests, both Si₃N₄ balls and coating specimens were cleaned in acetone for 10 min. Then, a series of tests were conducted from room temperature (designated at 20°C) to elevated temperature of up to 800°C in air, at an oscillating frequency of 15 Hz, a load of 10 N, an amplitude of 1 mm, a duration time of 10 min and a relative humidity of 20–30%. The friction coefficient curves were recorded automatically with a computer attached to the SRV tester. The wear volume loss was measured using a MicroXAM 3D non-contact surface mapping profiler (ADE Corporation, Westwood, MA, USA). The average of three repeat tests is reported as the final value in this paper. The relative measurement errors for friction coefficients and wear volume losses are ±10 and ±5%.

Characterisation of as deposited coatings and worn surfaces

Vickers microindentation was carried out along a line vertical to the coating surface using an MH-5-VM microhardness tester at a load of 1 N and a dwell time of 10 s. Standard metallographic polishing techniques were used to prepare the cross-section samples. The typical morphologies of original feedstock powders, as deposited coating and worn surfaces of coatings were characterised by means of a JSM-5600LV scanning electron microscope (SEM) equipped with an energy dispersive spectrometer (EDS). The worn surfaces of the counterpart Si₃N₄ balls were also analysed using the abovementioned 3D non-contact surface profiler.

Microstructure and microhardness of coatings

The SEM image of mixed NiCrBSi/WC–Ni powders is shown in Fig. 1a. It can be observed that the NiCrBSi powders prepared by inert gas atomisation possess spherical shape and relatively smooth surface. The reaction sintered WC-Ni powders present an angular and faceted morphology and are well dispersed in NiCrBSi alloy powders. The average particle sizes of NiCrBSi and WC–Ni powders are about 45 and 80 μm respectively. Figure 1b shows the typical surface morphology of plasma sprayed NiCrBSi/WC–Ni composite coating. Typical characteristics of plasma sprayed coating, including some unmelted and half-melted powder particles, intersplat structures, micropores and microcracks, are clearly observed. Meanwhile, those incompletely melted particles still retain the original morphology of NiCrBSi and WC–Ni powders (see the white and black arrows in Fig. 1b).

OPEN IN VIEWER

Figure 1

SEM images of a NiCrBSi/WC–Ni powders and b plasma sprayed NiCrBSi/WC–Ni composite coating

The cross-sectional structures of the as deposited and as remelted NiCrBSi/WC–Ni coatings are shown in Fig. 2. It is observed that, apart from micropores, intersplats and microcracks, grey NiCrBSi phase and white WC and W₂C ceramic phase are also differentiated in the cross-section of plasma sprayed coating (Fig. 2a), according with that reported elsewhere.13 The existence of W₂C can be ascribed to the decomposition of WC ceramic particles, resulting from the high temperature effect of plasma spraying. Comparing Fig. 2b with Fig. 2a, it can be seen that the thickness of the two types of NiCrBSi/WC-Ni coatings is ∼300 μm. Meanwhile, the obvious interfaces between the coatings and substrates reveal that, the both coatings exhibit the typical mechanical bonding mechanism. However, the pores, cracks and lamellar structures in the as sprayed coating are eliminated by means of laser remelting, corresponding to the increase in microhardness for laser remelted coating. Figure 3 shows the microhardness profiles of the two types of coatings, which were measured along the cross-section perpendicular to coating/substrate deposition direction. The plasma sprayed coating has relatively uniform hardness distribution (about 920±75 HV), while the laser remelted coating presents a high and fluctuant hardness of 1100±100 HV, due to its denser microstructure with pore and crack free as well as precipitation during the remelting process (Fig. 2).

OPEN IN VIEWER

Figure 2

Cross-sectional SEM images of a plasma sprayed and b laser remelted NiCrBSi/WC–Ni composite coatings

OPEN IN VIEWER

Figure 3

Cross-sectional microhardness of plasma sprayed and laser remelted NiCrBSi/WC–Ni coatings: plasma sprayed and laser remelted coatings are labelled as PS coating and LR coating respectively

To further observe the microstructure more easily and noticeably, the laser remelted coating was etched with a mixed solution of HCl–HNO₃–H₂O (volume fraction: 1∶10∶10) to preferentially dissolve the matrix, as shown in Fig. 4. It can be found the cross-sectional microstructure of the etched coating can be divided into two main sections including top and bottom layer with different microstructures, because of the effect of gradual temperature variation in the melting pool for laser remelting process, as indicated in Fig. 4b and c. It is seen that since the high density of WC–Ni particles in the as remelted coating tends to sink it from the surface to interface, and therefore, the content of WC increases from the interface to the surface. Also, blocks with an atomic composition of 16·7C–23·8Cr–17·2W–36·9Ni–2·7Fe (at-%, determined by EDS) were observed at the particle boundaries. This indicates that the outer WC particles have melted and diffused into the liquid NiCrBSi alloy and solidified in mixed carbides.7 However, no sharp interface exists between the two regions (Fig. 4a), which is in agreement with the variation of hardness (Fig. 3) and the SEM observation (Fig. 2b), due to the physical discrepancy between NiCrBSi alloy and cermet WC–Ni.

OPEN IN VIEWER

Figure 4

Cross-sectional SEM images of laser remelted NiCrBSi/WC–Ni coating: a overview; b near top zone; c bottom zone

Friction and wear properties of coatings

The variation of friction coefficients for both plasma sprayed and laser remelted coatings sliding against Si₃N₄ ceramic ball from room temperature to 800°C are presented in Fig. 5. It can be seen that the friction coefficients for both coatings varied in a similar manner. Namely, the friction coefficient initially increased up to a maximum from room temperature to 200°C. Then, it decreased with further elevated temperatures from 300 to 600°C. Finally, the friction coefficient increased to some extent and kept a moderate value from 700 to 800°C. This phenomenon is ascribed to the following reasons. At relatively low temperatures (from room temperature to 200°C), the surface protuberance of Si₃N₄ was capable of penetrating into the surface of the coating, leading to abrasive wear accompanied with a high friction coefficient. At an increased temperature of 300–600°C, due to the synergistic effect of high temperature and frictional heating, chrome oxide (Cr₂O₃) with excellent high-temperature lubricating effect was formed on the coating surface and thus resulted in decrease of friction coefficient.21,22 Furthermore, at 700 and 800°C, since the strength of the composite coating further decreased with the elevated temperatures, Cr₂O₃ layer was easily removed from the coatings surface, leading to an increase in the friction coefficient.12,23

OPEN IN VIEWER

Figure 5

Friction coefficients of plasma sprayed and laser remelted NiCrBSi/WC–Ni composite coatings at various temperatures

The variation of wear volume losses for both plasma sprayed and laser remelted coatings sliding against Si₃N₄ ceramic ball from room temperature to 800°C is shown in Fig. 6. The wear volume losses of both coatings increased with elevated temperatures at the whole temperature range, largely attributed to the decrease in the coating strength.12,23 On the other hand, laser remelted coating presented smaller wear volume losses than plasma sprayed coating from room temperature to 600°C, especially, at a temperature of below 400°C. As compared to plasma sprayed coating, the improved wear resistance of the laser remelted coating was mainly due to its dense microstructure and enhanced hardness (Figs. 2 and 3), contributing to its high load carrying capacity.24 However, at further increased temperatures of 700 and 800°C, the plasma sprayed coating exhibited better antiwear property than the laser remelted coating. This is because large residual stress was generated at 700 and 800°C, leading to the decreased cohesion strength and hence severe adhesion wear damage for as remelted coating (Figs. 7c and 8c).12 On the other hand, plasma sprayed coating experienced precipitation hardening at high temperature of 700 and 800°C, contributing to its increased hardness and wear resistance (Fig. 9c).

OPEN IN VIEWER

Figure 6

Wear volume losses of plasma sprayed and laser remelted NiCrBSi/WC–Ni composite coatings at various temperatures

OPEN IN VIEWER

Figure 7

SEM images of wear scars of laser remelted NiCrBSi/WC–Ni composite coating sliding against Si₃N₄ ball at 20°C and elevated temperatures: a 20°C; b 400°C; c 800°C

OPEN IN VIEWER

Figure 8

Three-dimensional surface profile of worn surface of Si₃N₄ ball against laser remelted NiCrBSi/WC–Ni composite coating at 20°C and elevated temperatures: a 20°C; b 400°C; c 800°C

OPEN IN VIEWER

Figure 9

SEM images of wear scars of plasma sprayed NiCrBSi/WC–Ni composite coating sliding against Si₃N₄ ball at 20°C and elevated temperatures

Analysis of wear mechanism

The wear mechanisms for both coatings were mainly dependent on their different structures and the synergistic effect of test temperature and frictional heat. Figures 7 and 9 show the typical worn surface morphologies of the coatings sliding against Si₃N₄ balls at various temperatures. SEM observation revealed that the worn surface of the plasma sprayed coating at room temperature and 400°C presented some signs of plastic deformation, shallow scratch grooves and microcracks (Fig. 9a and b). This was attributed to the existence of some incompletely melted powder particles, micropores and microcracks (Figs. 1b and 2a), leading to crack initiation and propagation resulting from high contact stress and compressive stress. Besides, during the sliding wear test, the hard WC particles in the wear debris as well as the wear debris adhering to the Si₃N₄ ball caused the obvious scratch grooves on the worn coating surface. When test temperature was further elevated up to 800°C, obvious scratch grooves and microcutting appear on the worn surface of the plasma sprayed coating. In this case, the existence of thin oxide layers acting as stress raisers in association with the thermal mismatch between the metallic phase and ceramic phase (WC and W₂C) accounts for the failure of the coating.12 On the other hand, some precipitated phase also appeared on the surface of the plasma sprayed coating sliding against ceramic at 800°C, due to enhanced high temperature effect (Fig. 9c). Such precipitation hardening of the plasma sprayed coating corresponds well with its better wear resistance than the laser remelted coating at 700 and 800°C (Fig. 6).

The worn surface of the laser remelted coating was relatively smooth showing some signs of slight peeling and shallow scratch groove, attributed to mild abrasive wear at room temperature (Fig. 7a). And some evidence of slight adhesion and plastic deformation emerged on the worn surface of the laser remelted coating at 400°C (Fig. 7b). When the temperature was further elevated up to 800°C, the laser remelted coating showed signs of severe adhesion, plastic deformation and partial spallation (Fig. 7c), corresponding to its larger wear volume loss as compared with the plasma sprayed coating (Fig. 6). This is because high temperature effect (400 and 800°C) and frictional heating resulted in ‘softening’ of the as tested coating and hence enhanced adhesion wear, which agreed with the worn surface feature of Si₃N₄ ceramic ball (Fig. 8). Namely, the worn surface of Si₃N₄ ball at room temperature was relatively smooth and featured scratch grooves, indicating the occurrence of abrasive wear. At elevated temperatures of 400 and 800°C, wear debris was adhered to the worn surface of Si₃N₄ ball and adhesion wear became more severe, corresponding to the increased adhesion wear damage and wear volume loss.