Sample preparation effects in the surface analysis of V/Ti alloys

Abstract The surface composition of V–21% Ti alloy as revealed by Auger electron spectroscopy becomes depleted in Ti on bombardment with 2‐kV argon ions. Alloy composition changes by preferential sputtering are frequently observed, but what is surprising in this case is that a steady‐state composition is not achieved until after ca. 8 h of sputtering. Experience with other alloy systems has been that steady state is attained in minutes; furthermore, it is difficult to conceive an atomic process which would account for such slow composition changes. The alloys was prepared by monoarc co‐melting of the pure metals, slicing with a slow‐speed diamond saw and mechanical polishing down to 1 μ diamond. The polishing damage will extend several micrometres below the sample surface, which is consistent with the thickness of sample removed by 8 h of ion bombardment. A possible model is provided by radiation induced segregation, where the defects created by ion bombardment of the alloy annihilate at both external and internal surfaces associated with polish damage. At the beginning of the sputtering, the internal surface provides the dominant sink, whereas after the polish damage has been entirely removed this is provided by the external ‘analysed’ surface. This interpretation was tested by preannealing a freshly polished specimen to remove the polish damage before sputtering. In this case, the steady‐state composition was attained in minutes, as predicted by the model. One final question concerns the absence of such effects in other systems, e.g. W/Mo, Cr/Mo. The answer can probably be found in the different room temperature mobilities of the ion‐bombardment‐induced defects in such refractory alloys when compared with V/Ti. This is supported by the composition changes observed on ‘room temperature annealing’ when bombardment of V/Ti is stopped, an effect never observed in the refractory alloys.