*3.2. Hardness and Roughness Evolutions*

Figure 4 shows the evolution of the hardness and roughness before and after the different SMAT and SMAT + polishing treatments for the Ti-Nb wall. For the polished sample, the maximum hardness was reached for 25% of Nb (Figure 4a). According to the literature [42,43], this increase is due to the presence of a small fraction of ω-phase, especially when the composition is around 25% Nb. The amount of omega phase was increased by the successive remelting and tempering from the successive layer by layer deposition generated by the DED-CLAD process. This tempering promoted the formation of the ω-phase, which explains the increase in the microhardness, as was also observed for TiMo alloys [44].

**Figure 4.** Evolution of (**a**) the hardness (HV0.3) and (**b**) the roughness (Ra) as a function of the Nb alloying amount for the P, S and P + S conditions of the TiNb walls.

After the SMAT modifications, it was first interesting to notice that the hardness evolutions for the S and S + P samples are almost equal (Figure 4a). This meant that the polishing has allowed to remove a few microns to smooth the roughness (Figure 4b) while remaining in the hyperdeformed thickness. The range of hardness increase by shot peening depended slightly on the alloying amount. The maximum increase was obtained due to the appearance of the kink bands in the Ti25Nb, which made it possible to change hardness from 250 to 400 Hv (60% increase). Sadeghpour et al. [45] have recently shown that there is a relation between the presence of omega phase and the formation of kink bands due to the accumulation of simple dislocation and dislocation channels, which correlates with our study. The strong internal misorientation of the 50 to 100% Nb containing samples also increased the hardness but to a lesser extent (15 to 40% increase).

The evolutions of roughness presented in Figure 4b shows that the roughness after SMAT was inversely related to the hardness. When the material was hard, the roughness increase remained moderate. This is because—for a given impact energy—the balls created deeper craters when the material was softer (Figure 4a). Because of the polishing procedure, the P and S + P samples have the same low roughness, independently of the initial hardness. These modifications of hardness versus roughness made it possible to dissociate the effect of the roughness from the other effects (chemistry and SPD induced by SMAT). Indeed, several papers have suggested that the surface roughness plays a major role in cell biology: the higher the roughness, the higher the cell adhesion [46–48].
