A parametric study of titanium silicide formation by rapid thermal processing

A parametric study of titanium silicide formation by rapid thermal processing was conducted to determine the effects of annealing temperature (650 degrees C, 750 degrees C), annealing time (30 s, 60 s), wet etching (no HP dip, With KF dip), sputter etching (no sputter etch, with sputter etch), and annealing ambient (Ar, N-2) on the completeness of conversion of 60 nm Ti on (111)-Si to C54-TiSi2 based on sheet resistance and the uniformity of the sheet resistance measurements across the entire wafer. Statistical analysis of the results showed that temperature, annealing ambient, and sputter etching had the greatest influence. Increasing the temperature and using arson gas instead of nitrogen promoted conversion of the film to C54-TiSi2. On the other hand, sputter etching retarded it. The results also indicated significant interactions among these factors. The best uniformity in sheet resistance was obtained by annealing at 750 degrees C without sputter etching. The different sheet resistance profiles showed gradients that were consistent with expected profile behaviors, arising from temperature variations across the wafer due to the effect of a flowing cold gas and the effects of the wafer edge and flats.