Pure Covalent Bonding: Understanding its Role in Atomic Bonds

[hivesaeed4]

Hi,

If two atoms of the same element form a bond together, would the bond be purely covalent or would it have a small ionic character as well ? If it does have an ionic character please explain why.

Thanks

 

[AlonV]

Well pure covalent is a term used to describe little or no difference between the electronegativity of the atoms forming a bond.Since the atoms are identical,the net dipole moment is 0,so the bond is described as purely covalent.
p.s.bonds in some compunds may vary a lot.Hence Aluminium halides.AlF3 is ionic,AlCl3 is ionic as a solid and covalent as a gas,while AlI3 is covalent.

 

[uby]

The bond would generally be considered purely covalent, though you'd need to evaluate the local environment (i.e., ALL the bonds around each member of the bonding pair) to ascertain whether the bond can be polarized and thus have some degree of ionic character.

 

[jdy07]

In "The Nature of the Chemical Bond," Pauling says that the ionic structures of H2 (H-(1)H+(2) and H+(1)H-(2)) each contribute about 2% to stabilization energy of the H2 molecular bond. I don't know enough about his calculations to know how those percentages were determined and if someone could add that information, I'd really like to know. However, the short approximate reason is that all resonance structures must be used in determining the stabilization energy of the molecular bonds. And so even seemingly "totally covalent" molecules have a bit of ionic character.

I hope this answers your question and isn't totally outdated.

 

[Einstein Mcfly]

In "The Nature of the Chemical Bond," Pauling says that the ionic structures of H2 (H-(1)H+(2) and H+(1)H-(2)) each contribute about 2% to stabilization energy of the H2 molecular bond. I don't know enough about his calculations to know how those percentages were determined and if someone could add that information, I'd really like to know. However, the short approximate reason is that all resonance structures must be used in determining the stabilization energy of the molecular bonds. And so even seemingly "totally covalent" molecules have a bit of ionic character.

I hope this answers your question and isn't totally outdated.

You're talking about "valence bond theory" which is an alternative to molecular orbital theory that looks at the energy of a molecule as the sum of different "resonance structures" which put positive or negative charges at various spots just as you describe for H2. It's interesting, though because it's not as easy to program into a computer, it's been pretty much left behind by molecular orbital theory. Some folks though are still into it. See
https://www.amazon.com/dp/B001772WWO/?tag=pfamazon01-20
for an example.

 

[jdy07]

Thanks for the clarification EM. I didn't realize that was still valence bond theory. It feels different from the valence bond theory that was taught way back in O-chem, and I didn't realize it could be used in calculations for H2.

On the subject of covalence, in inorganic compounds, the covalence of a metal-ligand bond can be measured by the hyperfine coupling of an unpaired spin with the metal center. Is there anyway to measure the covalence in the bonds of a diamagnetic, organic compound? Or are these bonds generally considered to be totally covalent?

 

[DrDu]

The problem is that it is quite hard to define what "covalent" really means for two atoms at bonding distance. E.g. in the example of the valence bond treatment of the H2 molecule the contribution of ionic structures depends strongly on the choice of the atomic orbitals. If these are chosen to be orthogonal to each other, consideration of the covalent structure only won't lead to bonding and bonding is only obtained when resonance with ionic structures is included.
A more rigorous definition of "non-ionic" bond is possible on the basis of Bader's "atoms in molecule" definition:
http://en.wikipedia.org/wiki/Atoms_in_molecules

 

[Einstein Mcfly]

The problem is that it is quite hard to define what "covalent" really means for two atoms at bonding distance. E.g. in the example of the valence bond treatment of the H2 molecule the contribution of ionic structures depends strongly on the choice of the atomic orbitals. If these are chosen to be orthogonal to each other, consideration of the covalent structure only won't lead to bonding and bonding is only obtained when resonance with ionic structures is included.
A more rigorous definition of "non-ionic" bond is possible on the basis of Bader's "atoms in molecule" definition:
http://en.wikipedia.org/wiki/Atoms_in_molecules

We should be clear though that "atomic charge" is a phenomenological concept that we have from wanting to make simple rules about "electronegativities" and such, not a rigorous quantum mechanical entity. In the end, it's all arbitrary. Obviously things like Mulliken which are hilariously basis set dependent are pretty much useless (I suspect we only hear about them because you can make one interesting point with them and the H2 molecule in first semester P-chem) but even things like Bader analysis can have issues jiving with "chemical experience". See: J Comput Chem 25: 189–210, 2004 for a review that swears that only spatial integration (Vornoi or Hirshfeld) are reliable. You know, since there's never a paper saying "all ways work fine, whatever you'd prefer".