Figure 3 shows images at different magnifications by SEM of the typical threaded topography of the Ti6Al4V screws and the characteristic surface with cavities and holes heterogeneously dispersed due to the impingement of the particles used in the blasting process. The chemical analysis performed by EDX of a representative area (Table 1) shows the characteristics peaks of Ti, Al and V, together with some proportion of Ca and P. Ca and P came from the hydroxyapatite particles used in the blasting process.

Due to this process, it is unavoidable that the blasting particles induces abrasive pollution on the surface, giving rise to a surface modification not only in the final roughness but also in the chemical composition, that influences the physicochemical properties of the blasted surfaces [25]. To minimize the drastic changes produced by of the blasting process, the posterior acid attack is usually used as a method to smooth titanium surfaces and also to eliminate the residual particles resulting from the blasting process [26]. The use of HCl posterior to the blasting process does not assure that residual hydroxyapatite particles remain incrusted in the surface of the implants. Nevertheless, given the chemical nature of the Ca-P compounds on the surface, its presence could be even considered as beneficial in the osseointegration process.

After the thermal treatment, an oxide scale is grown on the Ti6Al4V screws (Fig. 4). Barranco et al. [27] established that the increase in roughness (at microscale) of specimens due to the thermal treatment at 700 °C for 1 h was not significant. However, the nanoroughness of the surface can be increased about 140 nm due to this treatment [27] (compare Figs. 3 and 4). The thermal oxidation of the Ti6Al4V alloy after oxidation treatment at 700 °C for 1 h gives rise to a modified surface whose composition and crystalline order is changed. Previous XRD results of Ti6Al4V thermally treated at 700 °C for 1 h carried out by the authors and published in Billi et al. [28] revealed diffraction angles assigned to rutile, without any evidence of aluminium oxide. Confidence about the stability of the microstructure during the thermal oxidation treatments was provided in previous works by no significant differences in hardness from 329 to 320 HV0.5 after 700 °C for 1 h [4].