Answers are grounded in this paper and cite their sources.
Selective growth of chemical-vapor deposited selective tungsten, on thermally grown titanium silicide thin films, on silicon substrates, has been characterized. The nucleation of selective tungsten is found to be strongly affected by the surface chemical composition of the silicide. A thin layer of titanium nitride, formed on the silicide surface during the silicidation of sputtered titanium in nitrogen ambient, does not allow selective tungsten nucleation and has to be chemically removed to observe growth of the tungsten film. Localized titanium oxide regions on the silicide surface lead to adhesion problems between tungsten film and the silicide underlayer resulting in tungsten film blistering and bubbles formation. The regions of silicide where these bubbles form show accumulation of fluorine from the WF6 reduction at the interface. In regions of good adhesion and normal growth characteristics, no titanium oxide was found, almost all the silicide was consumed during the growth of selective tungsten and no fluorine accumulation at the interface was observed. These results indicate that it is possible to grow selective tungsten on titanium silicide barrier over silicon substrate provided that the surface of silicide is free of any titanium oxide or nitride layers. Under such conditions, the silicide layer acts as a sacrificial barrier allowing selective tungsten growth without resulting in any silicon consumption or encroachment of tungsten into the silicon substrate.
No takes yet. Share an insight, caveat, or question.
Chen et al. (1987) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: