why does SiO2 form a crystal structure while CO2 forms a linear one? Would it be possible to make a CO2 crystal or SiO2 gas in a lab or is there something preventing that from happening.


  • 2
    $\begingroup$ Apparently, CO2 makes a SiO2 (trydimite) analog at about 400GPa at 1800K. It is discussed in str.llnl.gov/str/Yoo.html. A paper from that group is in Physica Status Solidi 223(2) 427-433 (2001). $\endgroup$
    – Jon Custer
    Jul 11, 2016 at 22:52

1 Answer 1


$\ce{SiO2}$ forms a tetrahedral covalently bonded network , while $\ce{CO2}$ is a strictly molecular compound that forms double bonds between carbon and oxygen. This is because Si is in the next period, so it has 3p orbitals instead of 2p orbitals. Orbitals of a higher shell (greater principal quantum number) are physically bigger, but oxygen still has smaller 2p orbitals, therefore there is less overlap between silicon and oxygen’s p orbitals than there is between oxygen and carbon’s p orbitals. This makes a Si-O π bond much higher in energy, so it does not form. Instead, only a lower energy tetrahedral network of σ bond forms, giving us silica as we know it. This explains why $\ce{SiO2}$ cannot form double-bonded molecules like $\ce{CO2}$. Also, single Si-O bonds are especially stable (466 kJ mol-1) and short. This is attributed to the overlap between oxygen’s p orbitals and silicon’s empty d orbitals. Additionally, silicon is a larger atom than carbon, so the bonding electron pair repulsion between the oxygens bonded to silicon in SiO2 is much lower than what would occur if such a tetrahedrally coordinated CO2 network could exist. Silica’s behaviour can, thus, be explained by enthalpy, MO theory and VSPER.

Crystal $\ce{CO2}$ can definitely be made in a lab, and it is known as dry ice. $\ce{CO2}$ freezes at -56 °C. It boils at -78 °C and you will often see this temperature written under arrows in synthesis because dry ice is the most common method of carrying out reactions at such a low temperature. At atmospheric pressure, the dry ice crystal structure does not, however, resemble network covalent silica, but some high-pressure solid $\ce{CO2}$ phases have network covalent structures. Here is a phase diagram:

CO2 phase diagram, taken from: https://crystallography365.wordpress.com/2014/10/27/solid-carbon-dioxide/

Phase I $\ce{CO2}$, shown in the bottom-left corner, is the most common dry ice crystal structure, and it is molecular, exhibiting a face-centered cubic arrangement with respect to the carbon. The high-pressure Phase V $\ce{CO2}$ (top-right), on the other hand, has a "partially collapsed cristobalite" structure. Isostructural with one of silica's polymorphs, it is non-molecular.

As for gas-phase $\ce{SiO2}$, it would certainly be hard to make, as it boils at close to 3000 °C, presumably for the bond-stability reasons listed above. I haven't read amy literature about its gas-phase structure, but I assume it would be similar to that of $\ce{CO2}$ but extremely unstable.


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