# What's up with SiO₂ being tetrahedral?

If silicon dioxide tends to form a crystal lattice with four $\ce{O}$'s around a central $\ce{Si}$, why isn't the molecular formula $\ce{SiO4}$ then? I'm confused why it's unique in that its molecular formula doesn't match up with its geometry.

The problem you are having with the formula is based on a misunderstanding about how far you can take molecular and structural formulae.

For compounds that form distinct molecules it is often worth writing the molecular formula in a way that helps you understand the structure of the molecule. But this is a convenience not a generalisation that can apply to all possible compounds.

It doesn't apply to $\ce{SiO2}$ because there is no such thing as an silica molecule. Silica, like many other minerals, is a 3D network of bonds with no discrete molecular components. In silica each Silicon is bonded to 4 Oxygens, but each oxygen is shared with two silicons. This gives the $\ce{SiO2}$ formula. This also explains why mineralogy is harder than chemistry.

Every $\boxed{^{14}\textrm{Si}}$ atom is surrounded by 4 $\boxed{^8\textrm{O}}$ atoms. On the other hand, every $\boxed{^8\textrm{O}}$ atom is surrounded by 2 $\boxed{^{14}\textrm{Si}}$ atoms.

Suppose that the number of $\textrm{Si-O}$ bonds is $n$, then the number of $\boxed{^{14}\textrm{Si}}$ atoms is $n/4$, while the number of $\boxed{^8\textrm{O}}$ atoms is $n/2$. Therefore you arrive at the formula $\textrm{Si}\textrm{O}_2$.

As a short answer, chemical formulae are only really meant to show the proportions between different atoms in a compound, and in general they do not carry any information regarding the way the atoms are bonded. In quartz, there are simply two oxygen atoms for each silicon atom, hence $\ce{SiO_2}$.

• Your answer appears to describe the empirical formula, the simplest whole number ratio of atoms. But not the molecular formula, which confusingly for silica is also $\ce{SiO2}$ according to PubChem. My question is getting at why silica does not follow convention with having a $\ce{SiO4}$ molecular formula since that would be a better description of its tetrahedral structure. Take butane for example. There are 2 carbons for every 5 hydrogens so its empirical formula is $\ce{C2H5}$. But its molecular formula is $\ce{C4H10}$ since that describes its actual structure. Dec 6 '13 at 5:09
• Structure of one molecule that is. Is there not 4 oxygen atoms in one molecule of silica? If not, then why is it tetrahedral? I was taught that the molecular geometry is derived from the molecular formula, at least that's what we use all the time in class. Are you saying molecular formula can't be used across the board like this? Are there are other exceptions? Dec 6 '13 at 5:20
• @reflexiv First of all, notice that if you accept $\ce{SiO2}$ is the empirical formula, then the "molecular formula" (a misnomer since quartz does not form molecules; it is an network solid) cannot be $\ce{SiO4}$, as the latter is not equal to the former times an integer. The following sentence from the Wikipedia article for quartz may also help: "[Quartz] is made up of a continuous framework of $\ce{SiO4}$ silicon–oxygen tetrahedra, with each oxygen being shared between two tetrahedra, giving an overall formula $\ce{SiO2}$." Dec 6 '13 at 10:51
• Also, I reiterate that molecular formulae do not in general carry structural information. You may know the structure of simple molecules from their formula, such as for $\ce{NH3}$, $\ce{H2SO4}$ or $\ce{IF5}$, but you may be hard-pressed to figure out that by $\ce{C_{14}H_{12}O2}$ I am referring to biphenyl-4-ylacetic acid. Sometimes we write out formulae in such a way that helps us figure out the structure. Phosphorous pentoxide is often written as $\ce{P4O_{10}}$ because the latter contains a hint to the structure of each molecule. Dec 6 '13 at 10:58
• Similarly, though ethanol and dimethylether are funcional isomers and therefore have the same molecular formulae ($\ce{C2H6O}$), we often write $\ce{C2H5OH}$ for ethanol and $\ce{H3COCH3}$ or $\ce{O(CH3)_2}$ for dimethylether. However, this is not the general case. Dec 6 '13 at 11:04

Because the oxygens are shared with more than one silicon,so in large crystal structure the ratio of silicon to oxygens will be 1:2 not 1:4

As one oxygen is not wholely solely with one silicon atom but it is bonded with two silicon atoms thus contribution of one oxygen is 1/2 towards one SiO2 molecule.. As one silcon is bonded to 4 oxygen atoms so 4(1/2) = 2 thus contribution of oxygen towards silcondioxide is 2 and we have one silicon so formula becomes SiO2.