I know that I learned in Chemistry class that a bond is defined to be ionic if the electronegativity difference between the atoms is more then 1.7, half the electronegativity difference between the most electronegative element fluorine and the least electronegative element francium. I think I heard that there's no clear definable distinction between a covalent bond and an ionic bond. I'm not sure it's the case that we can't define one. I believe we could define it by whether there is a localized electron pair at each bond. I know aluminum and oxygen have an electronegativity difference of more than 1.7. However, one web page said Corundum was a covalent network. Then I checked for myself and saw that corundum indeed probably does satisfy my definition of a covalent network. I know that sulfur can hybridize to make 6 covalent bonds with fluorine. Theoretically, could have a substance that's like corundum with the sulfur atoms instead of aluminum atoms and carbon atoms instead of oxygen atoms. Here we would say the sulfur hybridized to form 6 covalent bonds and the carbon is forming the usual 4. I know coordinate covalent bonds are possible also. If that substance could exist as a covalent network with a localized electron pair at each bond, why couldn't it also be a covalent network with a localized electron pair at each bond when you have aluminum and oxygen instead? It's still the exact same total number of protons per unit cell.

My question is

Has it already been researched whether corundum has a localized electron pair at each bond between an aluminum atom and an oxygen atom?

I know it's not completely clear but my questions was written the best way I can possible write it with my abilities and my lack of expertise in quantum mechanics. I can't figure out given how quantum mechanics works what my question would be, and I don't want to use my time on studying quantum mechanics. Maybe even if I did know how quantum mechanics worked, I would still be unable to figure an algorithm for determining what type of answer would satisfy me. When somebody is working on a complicated project that uses a lot of thinking, they can't figure out an algorithmic plan on how they will react to every situation. Rather, they trust themself to use their discretion later on what ever situation arises. A good answer would come from an expert who can function as if they are a future self of me and is good at hunting down existing research that they would use their discretion that they can write an answer about that has a good chance of satisfying me. As long as they write it in a way I can see as a good answer and see why it can be considered one, I think it will satisfy me.

As described in my question Can replacing some of the atoms in a covalent network with that of the element one atomic number higher make it nonstick?, maybe researchers will use an answer to that question to invest in new research. Then if they're going to invest in that research, then an answer to this question that corundum indeed does have a localized electron pair at each bond might further help them incorporate this idea into that idea and add corundum to the list of potential substances that can be treated that way to gain those revolutionary properties. Corundum will have the additional property that it can't combust. Maybe they will make hard highly nonstick nondegradable frying pans out of treated corundum.

  • $\begingroup$ We actually have an earlier question here about different methods of defining a bond order. Some of these methods also provide criteria for distinguishing ionic vs covalent bonds. $\endgroup$
    – Tyberius
    Jan 7, 2021 at 22:58
  • 1
    $\begingroup$ I saw the 4 close votes so I fixed up my question. $\endgroup$
    – Timothy
    Jan 13, 2021 at 22:06

1 Answer 1


Maybe I didn't understand exactly your question, but it seems to me that you'd better take a look at Pauling's rules. Even though they are not always strictly respected, they works in most cases. The stability of a compound involves several factors and one of the most important is the radius ratio.

About the distinction between ionic and covalent bond; again Pauling provided a formula for calculating the ionic and covalent percentage in the bond for binary compounds. Actually a pure covalent bonding can be found (for example in pure elements such as Ga), conversely a pure ionic compound has not yet been found (i.e. a very few percentage of covalent bonding is always present). I don'know the origin of the electronegativity difference > 1.7 value, but this number is meaningles by itself; I guess it refers to some specific binary stoichiometry. In fact you have to consider not only the electronegativity difference, but also how it is shared and this depends on the coordination number and ionic radius ratio. This factors affects the tendency of an atom in a molecule to attract electrons.

  • $\begingroup$ I don't know much about quantum mechanics. I can't understand what you're saying. I do know about orbitals though. The question is a question about orbitals, not the other Pauling's rules if Pauling's rules define the nature of a bond a totally different way than whether there is a localized electron pair at each bond, which I don't know. If you didn't take the course Physical Inorganic Chemistry and don't feel like it and you can easily find somebody else who took it and is an expert on orbitals who is happy to answer this question, I think it would be a good idea to. Or if you want, you $\endgroup$
    – Timothy
    Jan 7, 2021 at 21:51
  • $\begingroup$ could maybe ask another question for clarification of this question of it relies on knowledge I have that you don't. $\endgroup$
    – Timothy
    Jan 7, 2021 at 21:52
  • $\begingroup$ I don't know what is your knowledge, but I find very bizarre your statement "Theoretically, could have a substance that's like corundum with the sulfur atoms instead of aluminum atoms and carbon atoms instead of oxygen atoms" because it neglects at all the fundamentals of crystallochemistry. You immagine a corundum-like S2C3 compound on a wrong basis; theory is good if it can be applied to practice and conforms to the expectations, otherwise remains a mathematical game. there are a lot of compounds isotypic with corundum; I would try to apply them your theory,before predicting odd compunds $\endgroup$
    – gryphys
    Jan 9, 2021 at 7:15
  • $\begingroup$ I know that sulfur can be hypervalent and make 6 bonds. Forget about the existence of corundum. Then you can still predict that structure might be theoretically possible because the sulfur atoms are making 6 bond with the carbon atoms making 4. Maybe it can't. But maybe the aluminum atoms actually are hybridizing like the sulfur atoms in sulfur hexafluoride are. $\endgroup$
    – Timothy
    Jan 10, 2021 at 5:59
  • $\begingroup$ I don't understand. Al3+ and S+6 have the same outer electronic configuration 2p6 but I dont't see how to bond 4 atoms of C to the very high positively charged S6+ ion (whose size is strongly reduced in comparison of Al3+; I know you don't consider this...) $\endgroup$
    – gryphys
    Jan 10, 2021 at 11:00

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .