Here is a treatment of
Chemical Structure
Determination brought to you by Gardiner Meyers
Here is a quick review of
Lewis Dots
Try this quick Quiz on Lewis Dots Structures.
Have a look at some of the many websites related to Lewis Structures
Basically, we can predict a lot of the chemical structure around us
by keeping in mind two things:
We also must admit that we have pushed our picture of the
one-electron atomic orbitals as the actual orbitals in the molecule a bit too far.
The angular properties predicted by the 'spherical' nodal structure of
s, p, and d orbital simply will not hold for a chemical bond, which is at best only
cylindrically symmetric. In fact, in a molecule the labels s, p, and d are
meaningless, because they represent total angular momentum states. Total angular
momentum of the individual electron is not conserved in a molecule, only its
projection of the bond axis. We expect the actual molecular orbitals to be lobes and
blobs, but the angles between the lobes will be dictated by, you guessed it, electron-
electron repulsion.
Almost all molecules are singlet states, that is all the electrons are paired.
These types of molecules are diamagnetic, because the magnetic moments of the individual electrons
all cancel each other out. This means that all the electrons are either in an bond
pair or they are in a lone pair. Counting the number of electron pairs
around a central atom will determine the angle between the bonds. We expect to count
bond pairs because of the first point, above, that electrons have two flavors and thus
two electrons fit in each orbital. We further expect bonding electrons to arrange
themselves between the bonded nuclei (lie along the bond) because of the second item
above, the interplay between electron-nuclear
attraction and electron-electron repulsion.
Atoms that have
less than an octet will have these bond angles:
Note the tetrahedron is the prefered structure of atoms with an octet and no multiple bonds. Super-octet molecules are possible with elements beyond the 2nd period, and is particularly common for the transition metals (why?)
So, in order to predict Bond Angles, we need a Lewis Structure to start
with. (Lewis dot structures do not in themselves predict molecular geometry, but
merely provide valence electron count)
Examples of these structures occur in common small molecules and ions.
Remember our 2nd Period sequence of hydrides? Perhaps these were the molecules that Lewis was thinking about when he arranged his dots in a 'square'.
Write the Lewis Structures for these examples
The situation of 5 electron blobs about a central atom is the most interesting, because the axial and equatorial positions in the trigonal bipyramidal structure are distinct. This is of some consequence when bonding and lone pairs exist, as in the case below. Note the structure that results minimizes the electron repulsion by keeping the lone pair electrons as far apart as possible.
How do the atomic orbitals accommodate such shapes? The atomic orbitals have to
mix to form the molecule anyway, so it seems reasonable that the orbitals on a single
atom can mix as well. This 'mixing' of the many electron atom orbitals, prior to and
in preparation for chemical bonding, is called hybridization.
Hybridization of s and p
orbitals is what allowed Lewis to put his dots on a square with all four sides
equivalent. Hybridization of d orbitals as well allows for 'super-octet' molecules to
be created
In Summary, the VSEPR shape can be rationalized after the fact with hybrid atomic
orbitals. This is consistent with the Valence Bond picture of chemical bonding.
