*3.2. Surface Integrity*

#### 3.2.1. Surface Roughness

Figure 5 displays the milled and burnished areas (4 mm × 4 mm) of each target surface in order to provide the qualitative effect of the burnishing operation under the stated operational and tribological conditions.

(**c**) AISI 316 burnished surface—270 N (**d**) S46500 burnished surface—270 N (**e**) S46500 burnished surface—470 N

**Figure 5.** Textured surfaces after milling and burnishing processes at different contact pressures.

It can be observed that the macro-texture conferred by surface milling was more prominent on the austenitic stainless steel than on the martensitic stainless steel. However, under the same burnishing conditions, the texture of the austenitic milled surface disappeared and was replaced by an imprint of the burnishing tool (Figure 5c), showing the extent of the contact area, as well as the magnification of the interaction between the ball and the surface. Regarding the surface finish obtained on the martensitic steel, a softened texture can be seen in comparison with the AISI 316 surface under the same conditions (270 N). It can also be observed that the peaks intensified as the load increased (470 N). Figure 6

summarizes the macro-texture parameters obtained after milling and ball burnishing in order to provide a quantitative description of the analyzed surface modifications.

**Figure 6.** Textured surfaces after milling and burnishing processes at different contact pressures.

Based on the height parameters Sa and Sq, a normal distribution of heights (Sa = 0.8 Sq) was evidenced under milling conditions for both materials. This observation was corroborated by the skewness values (Ssk ~ 0) under the same milling conditions. This tendency varied after ball burnishing. Thus, under the same load conditions, the statistical asymmetry after burnishing on AISI 316 stainless steel showed a mass distribution skewed to below the mean plane (Ssk > 0), while the surface of UNS S46500 stainless steel is skewed to above the main plane (Ssk < 0) in equal proportion. The load increment on the second one slightly varied under this condition. With regard to kurtosis (Sku ~ 3), it was observed that the austenitic surface became a non-abrupt platykurtic condition, whereas the martensitic surface responded with a better fit to a Gaussian distribution despite the load increase.

The S10z parameter provides a practical criterion for the statistical behavior of the height. It reveals that the five-point peak height and five-point pit height were 20% more prominent on the austenitic surface. However, this was reduced by 36% on martensitic stainless steel and only 20% on austenitic stainless steel after ball burnishing. This was not consistent with Figure 5c. Since the austenitic crystal lattice had a higher deformation capability (reflected in the COF value), a redistribution of the surface texture was evidenced. Thus, the new average roughness profile may have been displaced below the level of the initial valleys, leading to a loss of tolerance. On the other hand, increasing the pressure on the martensitic surface led to the generation of a pile-up and consequently an increase in the S10z parameter (~15%). This elucidates the marked differences in the surface roughness depending on the stainless-steel microstructure during the ball-burnishing process.

The directional properties quantified through the Str parameter were shown to be moderately isotropic (Str ~ 0.5) after milling for both surfaces. After ball burnishing, the austenitic surface became directionally anisotropic (Str < 0.3), whereas under the same conditions, the martensitic surface increased its isotropy. At a 470 N load, the martensitic surface became anisotropic.
