Michael's argument seems to be perfect. In short, you can say that Poly-Si has
got larger active surface area to react with due to presence of grain
boundaries.


Michael Larsson  wrote:
  Hi Jay,

At a guess, I would say this is predominantly due to the presence of
grain boundaries. These are regions of high(er) energy, compared to
the rest of the xtal structure, meaning GBs are more active/reactive,
and more prone to oxidation (along with other chemical reactions with
Si). O2 diffusion rates are probably also higher along GBs, due to the
disorder in the regular xtal lattice, promoting more rapid oxidation.
I would say that increased diffusion rates at GBs are the main reason
for the increased oxidation rates

Following the thermodynamic argument: GBs introduce disorder to the
xtal structure, increasing its energetic state. Thermodynamics shows
that all things in nature tend towards a state of lower energy, thus
the poly-Si will strive to do this at any opportunity (and more so
than SC Si). Oxidation is a mechanism by which the energy state of
many systems are reduced (e.g. rust in metals).