This is not always the case. Sometimes silicon oxide do promote cell adhesion through a protein film attached to the surface. So I think that in your case, this has to do with silicon oxide termination. You could check the hydrophobicity of the silicon oxide (by contact angle measurements) and change it by functionalization (silanization) or plasma treatment. This will affect cell adhesion. 

Normally, on the metal is a native oxide layer (titanium etc.) or other adsorbed molecules (gold, platinum). Therfore, the cells are not adhered directly on the surface of the metal.

Hi Vincent, we observed cells adhering on silicon oxide as a result of protein layer adsorped on silicon/tungsten non-topographical surfaces using a culture medium containing FBS. do you have a credible source I can look at for silicon oxide promoting cell adhesion? what about the impact of Calcium in culture media and saline solution?

thanks Dirk