Why does silicon not react with dioxygen?

I'm doing an organic lab experiment. I'm heating up silicon, carbon, and sulfur individually then placing them in pure $\ce{O2}$. I don't really know about carbon chemistry but I thought the reaction might be related to the electronegativity of silicon being lower than that of sulfur/carbon causing the bond to be more polar or something?


Elemental sulfur commonly occurs in form of $\ce{S8}$, with relatively low melting point and, more importantly, relatively low bond energies. So, it easily breaks.

Carbon commonly occurs as graphite or structures composed of graphite-like fragments, but usually something we call carbon contains up to 10% of other elements, meaning it can produce some combustible gases on heating.

However, more importantly, both elements mentioned above produce gaseous oxides on oxidation, that leave the surface of the reacting compound, exposing it to further reaction.

Silicon, on the other hand, produces solid oxide on heating, that prevents further reaction. It is still possible to make a very fine dust of silicon to react with oxygen, but it is not an easy feat. For reaction to occur to bulk silicon, the products must be removed from the surface. Given that, bulk silicon will not give silicon oxide, but it can react with chlorine in reasonable conditions, giving volatile silicon tetrachloride.

Why silicon forms solid oxide with high melting point is another matter, with answer considerably more obscure. The simplest approach to rationalize it would be to consider atomic radius of carbon, sulfur and silicon and consider bonding in alternative structures for this compounds.

Carbon dioxide is an oxide with small central atom, that can form $\pi$-bonds effectively with oxide, so it does not form polymeric structure.

Sulfur dioxide is an oxide of larger atom, but it has a lone pair on central atom, so polymeric form would be considerably hindered. (on contrary, sulfur trioxide forms polymeric forms, but still has gaseous form which is only marginally less stable).

Silicon dioxide in monomeric form has two $\pi$-bonds, and this is uncharacteristic for third row elements, and silicon is significantly larger than carbon, having enough space for four oxygens around it. So, it easily and effectively forms 3-d network of $\ce{SiO4}$ units with shared oxygens, that is quite hard to melt and vaporize, and, once formed on the surface of silicon, it prevents further reaction.

Still, formation of silicon dioxide films on surface of bulk monocrystalline silicon is essential for microelectronics.


In order for a reaction to occur, bonds in the reactants have to break so that they can form new bonds that comprise the products.

Did you use amorphous carbon? Was it monocrystalline silicon?

The article on amorphous carbon says that it is stabilized by dangling pi bonds with hydrogen, suggesting a bond that is easier to break and start an exothermic reaction that will provide further energy to break other, stronger bonds. The silicon lacks this.

Did you try graphite? My guess is that it would not react either, nor would diamond.

The sulfur probably reacted due to the weaker $\ce{S-S}$ bonds.

BTW, electronegativity isn't a very good explanation for much other than bond polarity.

  • $\begingroup$ At Chernobyl the graphite moderator did light on fire, didn't it? $\endgroup$ – Abel Friedman Oct 19 '14 at 18:01
  • $\begingroup$ How hot was that compared to OP's lab experiment? $\endgroup$ – Brinn Belyea Oct 19 '14 at 18:11
  • $\begingroup$ He didn't say. But the OP had pure oxygen on hand, that helps combustion, and once started, the heat generated from combustion helps sustain the reaction. $\endgroup$ – Abel Friedman Oct 19 '14 at 18:15
  • $\begingroup$ Both graphite and diamond will burn in oxygen if you heat them enough (and you don't need extreme temperatures). $\endgroup$ – matt_black Apr 12 '18 at 12:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.