when electrons become molecular bonds

[lemur]

I suppose it is the case that molecular atoms are never without bonds to other atoms in molecules. Nevertheless, it seems to be the case that in forming such a bond, electrons must shift from one configuration around an atom to another. My question is during such a shift, do characteristics of the pre-bond configuration carry over into the behavior of the bond(s) themselves? Also, since the bonds seem to be the result of electron-sharing between atoms, how does the formation and behavior of the bond affect the other electrons that do not "participate" in the bond?

[csmyth3025]

When you say "molecular atoms" I'm guessing that you mean those elements that normally occur in molecular form - O₂ being one example. There are other elements, such as argon (a so-called "noble gas"), that naturally occur in monoatomic form (individual atoms).

 

There are several types of molecular bonds - covalent bonds and ionic bonds being the most well known from high school chemistry classes. Wikipedia has an article on molecular bonds that provides short descriptions and links to more in-depth articles. It can be found here: http://en.wikipedia....nding_in_solids

 

Chris

[lemur]

When you say "molecular atoms" I'm guessing that you mean those elements that normally occur in molecular form - O₂ being one example. There are other elements, such as argon (a so-called "noble gas"), that naturally occur in monoatomic form (individual atoms).

 

There are several types of molecular bonds - covalent bonds and ionic bonds being the most well known from high school chemistry classes. Wikipedia has an article on molecular bonds that provides short descriptions and links to more in-depth articles. It can be found here: http://en.wikipedia....nding_in_solids

 

Chris

I don't mean a classification of certain types of atoms/elements. I mean atoms that are bonded with other atoms in molecules. I.e. I'm talking about a situation, not a type. I'll check out your link and see if it gives me some insight, thanks.

[mississippichem]

My question is during such a shift, do characteristics of the pre-bond configuration carry over into the behavior of the bond(s) themselves? Also, since the bonds seem to be the result of electron-sharing between atoms, how does the formation and behavior of the bond affect the other electrons that do not "participate" in the bond?

 

The characteristics of the molecular orbitals come directly from the atomic orbitals that formed them. We use something called Slater determinants [read some about matricies] to calculate these molecular orbitals from their constituent atomic orbitals. This is called LCAO, linear combination of atomic orbitals. The behavior sometimes depends on whether or not we treat the atomic orbitals as one electron "spin-orbitals" or two electron "Gaussian-orbitals".

 

As far as the molecular electrons affecting the non-bonding electrons: that is a very complex topic that I will hopefully learn more about in grad school. I do know of one treatment called dynamic pair correlation that I know roughly how to work with; it involves perturbation theory.

 

Qualitatively, remember that fermions do not like to share space and neither do like charges. Also remember that the entire set of orbitals for a species exists in a dynamic set. Changing the energy level of one electron will produce effects on all the other orbitals and the geometry of the molecule as a whole (see Ligand Field Theory). Also, in some rotational conformations, there is a degree of "mixing" between orbitals. So sometimes, when observing a radiative transition we can't resolve which orbital the photon came from; or we might not even observe the transition at all if the mixing is just right.

Edited May 6, 2011 by mississippichem

[ajb]

We use something called Slater determinants [read some about matricies] to calculate these molecular orbitals from their constituent atomic orbitals.

 

From what I can gather the Slater determinant is a way of expressing the wedge or antisymmetrised product in terms of the determinant of a matrix (up to a normalisation). Fermionic wave functions must be antisymmetric and so Slater determinants are a convenient way of ensuring the correct symmetry properties when building multiparticles states from single particle states.

 

The term Slater determinant is not one I have come across until recently.

Edited May 6, 2011 by ajb

[mississippichem]

From what I can gather the Slater determinant is a way of expressing the wedge or antisymmetrised product in terms of the determinant of a matrix (up to a normalisation). Fermionic wave functions must be antisymmetric and so Slater determinants are a convenient way of ensuring the correct symmetry properties when building multiparticles states from single particle states.

 

The term Slater determinant is not one I have come across until recently.

 

 

Yes, they are common practice in Quantum Chemistry circles. I'm not sure if they are so common in physics circles. They are nice because it does ensure the anti-symmetry and it can give you a simple [math] \Psi = \phi _{1} \phi _{2} \phi _{3}... [/math] in some cases which is well appreciated. This of course only holds if you neglect the non-adiabtic coupling terms and use the much nicer [math] \mathcal{H} ^{0} [/math] instead of the full Hamiltonian.

 

I think it also allows you to treat the electron-electron repulsion function as being pair-wise additive. Which lends itself to separation of variables.

Edited May 6, 2011 by mississippichem

[ajb]

This of course only holds if you neglect the non-adiabtic coupling terms and use the much nicer [math] \mathcal{H} ^{0} [/math] instead of the full Hamiltonian.

 

Of course, interactions are going to prevent you from building up multiparticle states in this way. However, I imagine it is a reasonable approximation to ignore such interactions in many circumstances.

[mississippichem]

Of course, interactions are going to prevent you from building up multiparticle states in this way. However, I imagine it is a reasonable approximation to ignore such interactions in many circumstances.

 

I think I was editing while you posted. Here's what I meant to add:

 

I think it also allows you to treat the electron-electron repulsion function as being (EDIT: not pair-wise additive)pair-wise additive. Which lends itself to separation of variables.

 

It works like charm on cento-symmetric molecules with even degeneracy. Whenever we have a molecule that allows certain bond-rotations, and therefore a high degree of mixing or uneven degeneracy. we have to bring the non-adiabatic coupling terms back in .

Edited May 6, 2011 by mississippichem