This is an interesting question, particularly considered in the context that Cairns-Smith (1985) even suggested that clays (silicates in solution) may have had some sort of early selection acting on them due to their surface chemistries.
However, there are a number of major problems with Silicon. Some are chemical and some are astrophysical in nature. For example:
- Silicon has a lower electronegativity than carbon and a longer bond length. Silicon can polymerize, but many conformations (such as rings) are highly reactive or unstable.
- Silicon lacks chirality. Since biochemical reactions are very specific this may present a fundamental problem for alien biochemistries.
- We don't see silicon macromolecules in nature. Large carbon molecules are seen in space such a polycyclic aromatic hydrocarbon rings. The largest silicon molecule seen in space is a chain of SiC_3 (and maybe SiC_4).
- On reacting with oxygen (which it does readily) silicon likes to form solids like sand.
- Silicon is much less common than Carbon in the Universe. The Solar abundance of silicon is 1/10 that of carbon, and supernova yields suggest that the silicon abundance may be as low as 1/100 that of carbon during nucleosynthesis in low/intermediate mass stars.
To form complex silicon molecules we would probably need to keep it in an oxygen-free environment and somehow maintain it in solution. One possibility would be to hold it at high pressure and temperature such as in the interiors of planets (think deep hot biosphere theory) but this presents another host of problems for conceivable biochmistries and is very speculative.
Apponi, A.J., McCarthy, M.C., Gottlieb, C.A., & Thaddeus, P. 1999, Journal of Chemical Physics, 111, 3911
Cairns-Smith, A. G. (1985) Seven Clues to the Origin of Life Cambridge University Press, New York, ISBN 0-521-27522-9.
Woosley, S.E., & Weaver, T.A. 1995, Astrophysical Journal Supplement, 101, 181