Every covalent bond has some ionic character and every ionic bond some covalent character. I can understand why a completely ionic bond is an ideal situation. But completely covalent bonds can exist(?).

Take the case of $\ce{H_2}$ or any other diatomic molecule between identical atoms. How can these have any ionic character?

  • 4
    $\begingroup$ It is common to see diatomic Bromine act ionically when being used as an electrophile. $\endgroup$ Commented Jun 1, 2015 at 13:40
  • 3
    $\begingroup$ As a combination of $\ce{H+}$ and $\ce{H-}$. The problem with this question is that ‘covalent bond’ and ‘ionic bonds’ — heck, actually ‘bonds’ in general — is a lot more theoretical of a construct than it might look. It does a good job of explaining a lot of things, but it remains a concept to explain reality; it is not reality. The closest to reality we can get is the calculation of molecular orbitals. Through mathematical tricks, we can derive ‘bonds’; that’s all. $\endgroup$
    – Jan
    Commented Jun 1, 2015 at 13:48
  • 1
    $\begingroup$ @Alizter: But Bromine behaves ionically in a very specific chemical environment. The real question is if isolated bromine atoms have any ionic tendencies? $\endgroup$
    – Gerard
    Commented Jun 1, 2015 at 14:13
  • 2
    $\begingroup$ @Jan: I understand what you mean. Bonds are a theoretical construct. Yet theories are self-consistent (reasonably), and the fact that covalent bonds have ionic character is a statement the theory makes. I want to know if this statement is correct, why it is correct, and the reason behind stating it like this. The fact that bonds are theoretical should not arise in this discussion because it is a discussion purely about the theory itself. $\endgroup$
    – Gerard
    Commented Jun 1, 2015 at 14:16
  • $\begingroup$ If you have symmetric molecule you should think about about metallic not ionic character:.en.wikipedia.org/wiki/Metallic_hydrogen $\endgroup$
    – Mithoron
    Commented Jun 1, 2015 at 14:35

1 Answer 1


According to Pauling's famous The Nature of the Chemical Bond , 3rd edition, at page 73:

In the hydrogen molecule a quantum-mechanical treatment has shown that the two ionic structures $\ce{H+H-}$ and $\ce{H- H+}$ enter into resonance with the extreme covalent structure $\ce{H-H}$ only to a small extent, each ionic structure contributing only about 2 percent to the normal state of the molecule.

He goes on to state that other diatomics such as chlorine should have even less ionic character.

So all molecules, even homonuclear diatoms, have some ionic character.

For further information on diatomic hydrogen's ionic character see section 1.5 of the book and The Normal State of the Hydrogen Molecule Journal of Chemical Physics vol. 1, pages 593-596.

See also Overlaps of Trial Functions for the Hydrogen Molecule. II. Covalent and Ionic Character of H2 J. Chem. Phys. 23, 176

  • 1
    $\begingroup$ In other words, the bond is not something static - no stick gluing together two balls. In the end, the bonding is a quantum phenomenon, and has to be treated as such to get a precise picture of what's actually happening. The older bonding theory is just an approximation of the actual interaction. $\endgroup$
    – Luaan
    Commented Jun 2, 2015 at 6:16
  • 1
    $\begingroup$ @DavePhD when you apply hannay smith equation,the % ionic character will be zero(by applying pauling equation too),doesn't that mean homonuclear diatomic molecules are perfectly covalent??(since % ionic character is 0) $\endgroup$ Commented Oct 9, 2016 at 15:51
  • 1
    $\begingroup$ @VidyanshuMishra See the two definitions of "ionic character" discussed in the following reference: digital.library.unt.edu/ark:/67531/metadc228342/m1/34 By the first definition, homonuclear diatomic molecules have ionic character, while by the second definition they do not. $\endgroup$
    – DavePhD
    Commented Oct 11, 2016 at 14:20
  • 1
    $\begingroup$ Check your first link. It sends me to wrong pages of the book. $\endgroup$ Commented May 9, 2020 at 20:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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