1. Introduction
Biomaterials engineering is a discipline that associates medical knowledge and engineering science to increase life expectancy. The fabrication of joint replacement prosthesis is part of this discipline, where metals such as ASTM F-1537 cobalt alloys are commonly used due to their good mechanical properties [
1]. However, the damage suffered by a metallic biomaterial in contact with a biological medium causes the deterioration of its physical properties [
2]. Likewise, it has been demonstrated that the release of metallic ions inside the human body leads to mutagenic and carcinogenic effects [
3].
In the last few years, research in metallic biomaterials has focused on improving their surface functionality through modification processes such as deposition of coatings, passivation-oxide layers, etc. [
3]. Laser surface texturing (LST) has excelled in improving the tribological properties of biomedical alloys through reducing friction coefficients and improving load-carrying capacity [
4]. Dimples resulting from LST act as lubricant reservoirs and are used as micro-hydrodynamic bearings [
5,
6]. Moreover, dimples serve as traps for wear debris [
7].
Dashtbozorg et al. [
8] studied the response of metastable S phase formed on a nitrided AISI 316 L stainless steel, which was modified through nanosecond (ns) Yb-doped fiber laser. Different morphological and chemical changes were observed when the pulse duration of the laser was modified. Cracks were seen near the surface of the 15 ns textured samples, whereas complete loss of the S phase was observed within the 220 ns textured samples. Additionally, the formation of localized high chromium, high nitrogen, low iron, and low nickel in the textured superficial region helped confirm the presence of chromium nitrides. In another study, Menci et al. [
9] compared the differences between the textured patterns obtained on a β-Ti alloy using Neodymium-doped yttrium aluminium garnet and Fiber ns lasers. Both conditions can be viable, as the attaching surface in the femoral stem and acetabular cup will stimulate bone growth due to their higher roughness.
Regarding the effect of surface texturing on the tribological properties of biomedical alloys, Zhang et al. [
10] developed a petaloid surface texturing on a Co-Cr-Mo alloy artificial joint through laser pulse ablation. Pin on plate tests indicated that petaloid surface texturing is a practical approach to improve the tribological performance of Co-Cr-Mo alloy joint prostheses because of its lubricant reservoir effect. Using tungsten carbide pins, Salguero et al. [
11] carried out pin-on-flat reciprocating tests at 5 N over Ti6Al4V texturized samples at different pulse energy densities and scanning speeds, with 1.5 mm diameter under lubricated conditions. A reduction of approximately 80% of wear track volume was obtained for lower scanning speeds (˂100 mm/s). The improvement was mainly due to the modification of the alloy by oxidation processes and microstructural changes. Alvarez-Vera et al. evaluated the tribological performance of CoCr microtextured discs against Ultra High Molecular Weight Polyethylene (UHMWPE)cylindrical pins with 3.5 mm diameter, under lubricated sliding conditions using a pin-on-disc tribometer at 10 N. Different laser surface microtextures, including dimple, line, net, and dimple patterns, were selected to generate different hydrodynamic responses. The wear rate and coefficient of friction decreased due to the elastohydrodynamic/hydrodynamic lubrication regime, reducing the wear and surface damage of UHMWPE pins.
Thus, important processing parameters resulting from LST are pulse duration, pulse repetition rate (PRR), maximum laser power, scan speed, process repetitions, etc. By properly selecting the laser parameters, the topography and chemistry of the implant surfaces can be optimized for the desired biomedical application [
9]. In addition, the short nanosecond (10
−9 s (ns)) pulsed laser can induce melt formation and vaporization, which can be exploited differently [
9].
However, further work investigating the correlation and influence of the noted processing parameters in the surface of medical-grade CoCrMo alloy is missing from the literature. Therefore, the aim of the present study was to investigate the morphological, chemical, and tribological changes produced by an ns-pulse fiber laser on the surface of an ASTM F-1537 cobalt alloy when the repetition rate, laser power, and process repetitions are variable. Surfaces were characterized using scanning electron microscopy (SEM). The chemical characterization was evaluated via two spectroscopy techniques: energy-dispersive (EDS) and Raman. The tribological response was evaluated via the ball-on-disc transitory wear test under lubricated conditions using optical 3D measurements and SEM.
4. Discussion
Dimples with a repetition rate of 140 kHz showed molten material on their edge and center, which caused the dimples to be irregular (
Figure 3). Further, dimples created with a repetition rate of 210 kHz reached a more defined circle (
Figure 4). This behavior may be associated with the “plume shielding” effect noted by Allahyari et al. [
21], related to the high repetition rates. Plume shielding occurs in ablative conditions where the laser pulses interact with the ablation cloud, inducing absorption and scattering and reducing the material removal rate.
Additionally, the laser power increase produced higher pulse energy, increasing the amount of molten material on the edges of the dimple, especially on the dimples obtained at 90%P. This large formation of material could negatively affect the tribological performance of the surface by serving as third bodies that could accelerate wear. Notably, an apparent increase in the depth of the dimples could be seen at six and ten repetitions compared to three repetitions.
It could be noticed in the EDS analysis that the intensity of O is similar for each one of the employed frequency values (
Figure 5). Thus, the variation in frequency did not have a relevant impact on the oxidation resulting from the surface texturing. The Si bonds identified in the Raman spectra (
Figure 6) could be associated with the grinding process with SiC sandpaper. The metallic oxides (CoO and MoO
3) resulted from the oxidation during laser texturing. However, Raman spectroscopy did not have the needed sensitivity for identifying more oxides.
The LST increased the roughness of the surface (
Table 6). This may be related to the formation of molten material around the dimples. A wide range of dimple depth was obtained. Higher laser power and process repetitions increased the depth of the dimples. In addition, a wider dimple diameter increased the depth of the dimples. This effect may be attributed to better use of the laser energy with larger dimples. The LST decreased the hardness of the zone adjacent to the dimples slightly (
Table 7). This phenomenon might be due to this zone became a heat-affected zone because of the heat transfer from the dimple. It should be clarified that it was difficult to measure the Vickers indentations on the closest zone to the dimples. It could be expected that the hardness of this zone should have increased because of the rapid solidification of the molten material [
22,
23].
The surface texturing as surface treatment decreased the wear rate of the surface compared to the untreated material (
Figure 7). However, remarkable variations among the wear rates of the different textured samples were noticeable. The wear rates of M1 and M2 samples were the highest compared with the M3 and M4 samples, which were in the group with the lowest wear rates. Notably, the unique difference between M1–M2 and M3–M4 samples was the laser power percentage (
Table 5); the influence of the dimple size in the improvement of the surface tribological properties was clearly seen. According to
Figure 4, a wider internal diameter of dimple was obtained with higher laser power. This effect influenced the wear resistance of the textured surface. These results support the theory put forward by Schneider et al. [
24] that the dimple size plays an essential role in trapping wear debris, storing lubricant, and in the hydrodynamic pressure built by the lubricant film. In the case of the M5–M8 samples, their wear rates were relatively low. This may be due to the dimple size effect noted previously, considering that dimple size in this set of samples was even bigger. Nevertheless, apparent differences among the wear rates of M5–M8 samples could be seen. The convergence of the dimple size with another critical variable, dimple depth, could have influenced the behavior of the wear rates. In agreement with the results reported by Ji et al. [
25], the dimple depth has a critical impact on the pressure distribution applied by the lubricant film. Further, no single textured condition exhibits the highest wear resistance; their optimization depends on the employed tribological parameters [
24]. According to the selected tribological parameters, the M8 sample showed the lowest wear rate in the present study.
In general, the behavior of the COF was stable (
Figure 8). This could be associated with the elastohydrodynamic lubrication between both contact surfaces. The results were consistent with Schneider et al. [
24], who reported similar COF values in a metal (texturized)–alumina contact under lubrication. However, texturing effect in friction was evident because the magnitude of the COF for most textured samples was lower than for the untreated sample. The textured patterns increase the lubricant film between the two surfaces because they serve as lubricant reservoirs [
22]. Importantly, the lubrication regimen could be considered elastohydrodynamic because the two surfaces are not totally conforming [
26]. Further, not all texturing conditions reduced COF. For example, the COF of M4 and M5 samples kept similar values to the COF of the M1 sample. This phenomenon might be due to the increase in rugosity for the M4 and M5 samples (
Table 6) after texturing that increased the number of asperities and raised friction between surfaces [
27]. This influence of the rugosity in the COF could also suggest a mixed lubrication because the sliding speed was not excessively high [
28]. Even though laser texturing increased the rugosity of surfaces, the COF behavior of all samples was nearly constant during the test. Additionally, the sample with the lowest COF did not exhibit the lowest wear rate also. The M2 sample showed one of the lowest COFs, but its wear rate was one of the highest. Moreover, the M8 sample showed the lowest wear rate, and its COF was not the lowest. However, the relation between wear rate and COF was not always inversely proportional for all texturing conditions; the M3 sample exhibited a lower wear rate and a lower COF. It can be inferred that the M3 sample was the condition with the best tribological performance because the two aforementioned variables were among the lowest. The COF does not always keep a directly proportional relationship with the wear rate. The COF is considered a system property as its behavior depends on the employed conditions, whereas the wear rate is a material property [
29].
Wear mechanism of abrasion could be seen, as evidenced by grooves along the motion direction (
Figure 9a). In addition, adhered material could be identified on the wear track (
Figure 9b). This adhesion of material could be related to particles pulled out from the surface due to wear. These resulting particles could have helped to form three-body abrasion, and after been deformed by the cyclic sliding, they adhered on the wear track. A less severe abrasion mechanism could be distinguished on the textured samples (
Figure 10). This may be because the dimples reduced the real contact area between the surfaces. Further, the increase in the hydrodynamic lubrication improved the load-carrying capacity of the surface, which in turn decreased the wear according to some authors [
23,
30]. Additionally, a certain deformation on the edge of the dimples could be identified (
Figure 10a,b). This phenomenon might be due to the increase in the pressure of contact due to the decrease in the contact area, which deformed plastically the edges before the dimples after some load cycles. No evidence of adhesion could be observed, possibly because the dimples serve as traps for the wear debris (
Figure 10c). These particles within the dimples are a mechanism to prevent surface wear because there is no interaction between the debris and the sliding surfaces.