Tribological performance of Ni₃Al matrix self-lubricating composites containing Ag prepared by laser melting deposition

The chemical compositions of Ni₃Al matrix powders are 79.51 nickel (Ni), 8.1 aluminum (Al), 5.23 chromium (Cr), 7.02 molybdenum (Mo), 0.13 zirconium (Zr) and 0.01 boron (B) (wt%). Ni₃Al matrix powders and Ag component in a weight ratio of 10:1 were homogeneously mixed together by vibration milling using a vibration frequency of 45 Hz and a vibration time of 10 h.

NAS can be obtained by sintering the mixture directly, using a DR Sinter® SPS3.20 apparatus. In order to meet the condition of LMD, the mixture must be processed as pre-alloyed powders firstly [18]. Subsequently, the pre-alloyed powders were melted to prepare NAL by a 8060 laser melting deposition system. The parameters of SPS and LMD processing are listed in tables 1 and 2, respectively. After SPS and LMD processing, NAS and NAL were obtained. In addition, the surface layers of NAS and NAL were removed by grinding, and then the samples were polished mechanically down to 1200 grit.

The detailed method for measuring the hardness and density of sample has been described elsewhere [19]. The tests of hardness and density were carried out six times for reducing random errors, and the average values of six results were the test results. The end results are given in table 3.

In line with ASTM G99-05 standard [20], the sliding wear tests of contact pairs of sample/Si₃N₄ ball were implemented on a ball-on-disk device. The experimental parameters are shown in the table 4. The friction coefficient was measured and recorded automatically. Detailed calculation procedure of wear rate of sample could be obtained in the previous study [21]. The wear rate was calculated as follows:

$$\begin{eqnarray}&&{\rm{W}}=\displaystyle \frac{{\rm{V}}}{{\rm{F}}\times {\rm{L}}}\end{eqnarray} \tag{ 1 }$$

Where V, F and L represent worn volume, test load and sliding distance, respectively. For the purpose of ensuring repeatability, the wear tests were carried out three times and the average values were acquired.

Phase identifications of NAS and NAL were performed by a x-ray diffraction (XRD). Morphologies of wear scars were characterized using a electron probe micro-analysis (EPMA). For obtaining the undestroyed cross sections of worn scars, the method of cooling fracture was adopted. Samples were incised down, and the residual thickness was about 1–1.5 mm. The incised samples were cooled for 50 min using liquid nitrogen. After cooling, samples were broken by the low shearing force under dust-free environment. Field emission scanning electron microscope (FESEM) was utilized to analyse the microstructures of cross-sections of wear scars.

In order to understand the influence of different processing routes on the composition of materials, the XRD spectra of NAS and NAL are plotted in figure 1. Only the typical peaks corresponding to the Ni₃Al and Ag phases are detected in NAS and NAL, showing that the phase compositions of samples are not affected by the processing technology. But compared with NAS, the diffraction peaks belonging to NAL show lower intensity and wider peak broadening. These phenomena can be explained by the fact that the grain refinement occurs in NAL during LMD process, which is coherent with the results reported by other researchers [22, 23]. The grain size is dependent on the cooling rate and changes inversely with it [24]. The cooling rates of SPS and LMD are within the range of 270–280 and 10³–10⁸ K s⁻¹, respectively [25–27]. On these bases, the difference in XRD patterns of NAS and NAL is reasonable. In addition, it is widely accepted that fine grain makes great contribution to the enhancement of material hardness by virtue of its strengthening effect. Therefore, the disparity in the grain size of NAS and NAL causes a different hardness level, where an increased hardness of 5.20 GPa is obtained for NAL in comparison to NAS (5.02 GPa).

Zoom In Zoom Out Reset image size

The effect of sliding speed on tribological performance of NAS and NAL is investigated in the range of 0.2–1.1 m s⁻¹, and the result is provided in figure 2. In general, the friction coefficients (0.28–0.36) and wear rates (4.6–5.9 × 10⁻⁵ mm³ N⁻¹ m⁻¹) of NAL is lower at all sliding speeds to different extents as compared to those of NAS. The friction coefficients (0.32–0.42) and wear rates (5.4–7.0 × 10⁻⁵ mm³ N⁻¹ m⁻¹) of NAS decrease with the increase of sliding speed until the minimum value is obtained at 0.8 m s⁻¹ and then increase with further increasing of sliding speed. By contrast, the friction coefficients and wear rates of NAL reduce progressively as the speed is increased from 0.2 to 1.1 m s⁻¹. It should not be ignored that the friction coefficients and wear rates of NAL are decreased by 33.3% and 48.6% at 1.1 m s⁻¹, if compared to those of NAS.

Zoom In Zoom Out Reset image size

In order to have a deep understanding of the worn mechanisms of NAS and NAL, the morphologies of worn surfaces of NAS and NAL obtained by using various processing technology are provided in figure 3. At a lower test speed of 0.5 m s⁻¹, the roughnesses of worn surfaces of NAS and NAL are relatively higher with evidence of observing some wear debris, resulting in higher friction coefficient and larger material loss. In addition, there are some slight grooves inlaying in the worn surfaces, indicating that the wear mechanisms of NAS and NAL are mainly guided by mild plowing wear. As the further enhanced sliding speed of 0.8 m s⁻¹ is properly settled, the worn surfaces of NAS and NAL are covered by the complete and compact lubricant layers, leading to the improvement of friction and wear performance. As the test speed increases from 0.8 to 1.1 m s⁻¹, a remarkable difference can be found by observing the morphologies of worn surfaces of NAS and NAL, which is responsible for the distinction of tribological behavior of NAS and NAL at 1.1 m s⁻¹ (see figure 2). Severe spalling phenomenon is noticed on the worn surface of NAS, which is considered to be the typical characteristics of peeling wear. For NAL, the smooth worn surface of NAL is aligned with the improved wear behavior.

Zoom In Zoom Out Reset image size

As Ag is lightly extruded to the worn surface and then plays the lubrication role during the sliding process, the variations in tribological performance and worn surface morphologies of NAS and NAL are believed to be relevant to the content of Ag. Figure 4 displays the elemental compositions of different worn surfaces. When comparing area B with D (also comparing area A with C), the increase in the content of Ag confirms the enrichment of a large number of Ag on the worn surface. Due to the low shear strength of soft metal Ag and the existence of the lubricant layer containing Ag, the excellent tribological properties of NAS and NAL are obtained at 0.8 m s⁻¹. Several interesting observations can be acquired in figures 4(e) and (f). For the same worn surface, the noteworthy difference of Ag content on worn surface of NAS can be explained by the local exfoliation of lubricant layer containing Ag and the occurrence of naked substrate when the speed is 1.1 m s⁻¹. However, the worn surface of NAL is still covered by the complete and compact lubricant layer with massive Ag (see figures 3(f) and 4(g)) after the reciprocating sliding tests. The frictional layer is easily damaged by the high sliding friction force due to the low strength of frictional layer with massive Ag [28]. According to the above mentioned phenomena and facts, it is reasonable to deduce that the tribolayer of NAS is prone to be peeled off and the tribolayer of NAL is not easily damaged. The reason for these differences is discussed and analyzed in detail, as presented in the next section.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The differences in cross-sections of wear scars between NAS and NAL are shown in figure 5. It is apparent to find the clear delamination morphologies in the cross-sections of NAS and NAL. However, the observation of cross-section of NAS indicates the existence of some voids, cracks and dislocation phenomena. According to the delamination theory of wear [29], dislocations begin to accumulate in the early stage and then produce voids and defects in the limited range of the subsurface during the progress of sliding. With the extension of sliding time, holes and defects are elongated due to sliding shear deformation and eventually produce cracks parallel to the worn surface. When the cracks reach to the critical length, the materials between the crack and the surface undergo shear deformation, and finally be peeled off. The spalling of friction layer containing Ag (see figure 5(a)) and exposure of naked substrate (see figure 3(e)) cause the reduction of lubricant particles in the worn interface (see figure 4(f)), leading to the sharp deterioration of tribological properties of NAS at 1.1 m s⁻¹.

Zoom In Zoom Out Reset image size

For NAL fabricated by LMD, there were almost free of any voids and cracks in cross-section of worn surface. In addition, the interface bonding between tribolayer and substrate is compact. These differences between NAS and NAL can be attributed to the distinction in manufacturing processes. LMD is the a process of layer by layer deposition and thus each layer is comprised of the freshly deposited materials and the previously solidified materials. This distinguishing feature is particularly significant for obtaining the stable bonding between layer and layer, which is also propitious to markedly improved the good interface bonding capability between tribolayer and substrate during the sliding wear process. In addition, the high hardness of NAL (see table 3) gives a better mechanical support to the tribolayer formed during the sliding friction process. Furthermore, the stability of the tribolayer with a great deal of Ag is favorable for the reduction of friction coefficient and wear rate, which is in line with test results (see figure 2).