VSEPR is a useful and organized summary of experimental results, and allows us to predict the 3-dimensional geometry of a large number of covalent compounds and polyatomic anions. At this level it would also be nice to try to understand how we might view these geometries in terms of the orbitals that are used for bonding by the atoms in these molecules. Two theories have evolved from quantum mechanics which assist the chemists in describing the experimental observations in terms of the atomic or molecular orbitals in a compound. One is Valence Bond Theory (pages 309 - 111) and the other is Molecular Orbital Theory (pages 324 - 335). We will focus our attention on VB Theory. We will not discuss the bonding in these molecules in terms of MO theory.

VB theory is a way of describing the electron pair bonds that occur in covalent compounds. This theory has too important underlying assumptions.

1. An orbital on one atom comes to occupy a portion of the same region of space as an orbital on the other. When this occurs the two orbitals are said to overlap.

2. The total number of electrons in both orbitals is no more than two.

When an orbital, occupied by a valence electron, on an atom overlaps with an orbital containing a valance electron on another atom the electrons in the orbitals begin to move about both atoms. The attraction that each electron experiences from both nuclei pull the atoms together. The strength of the bond depends on the amount of overlap; the greater the overlap of the orbitals the stronger the bond. Also remember the two overlaping orbitals can not hold more than two electrons and their spins must be opposite.

In the formation of the chemical bond in dihydrogen, H₂, two atoms of hydrogen, each with a single valence electron in a 1s orbital approach each other. For hydrogen the valence electron is in a 1s orbital which has spherical symmetry. As they get close the two 1s orbitals overlap with each other forming a chemical bond. Here is a QuickTime movie of the chemical bond in H₂.

Helium atoms will not bond to each other as their 1s orbitals are filled, overlap between two filled 1s orbitals would place four electons in the same region of space. According to our model only two electrons can occupy the same region of space. So He₂ does not exist.

We can understand the bonding in HF using our VB description also(Valence Bond Model). The 1s electron on hydrogen overlaps with the singly occupied 2p orbital on fluorine. Here is a QuickTime movie of the chemical bond in HF. The movie shows the electron configuration for a fluorine and for a hydrogen atom. The fluorine atom has seven valence electrons with an unpaired electron in one of its 2p orbitals. The hydrogen atom has one valence electron in its 1s orbital. When the 2p orbital on the fluorine atom overlaps with the 1s orbital on hydrogen a covalent bond forms. An extension of this approach is found in F₂ where the half-filled 2p orbital on each fluorine atom overlaps to form the covalent bond.

In water the oxygen atom has six valence electrons. Two of the 2p orbital on the oxygen atom are half-filled. These two half-filled orbital could bond to two hydrogen atoms. Here is a QuickTime movie of the chemical bond in H₂O. The movie shows the electron configuration for an oxygen atom and for the hydrogen atoms. The oxygen atom has six valence electrons with an unpaired electron in two of its 2p orbitals. The hydrogen atom has one valence electron in its 1s orbital. When the two 2p orbitals on the oxygen atom overlaps with the 1s orbital on the hydrogen atoms forming two covalent bonds. The interesting observation is the bond angle we would expect from our Valence Bond Model is 90 degrees, but the experimental observation is a bond angle of 105 degrees. To explain the experimental bond angle we will need use different orbitals on the oxygen atom.

Another conflict with the Valence Bond Model is found with ammonia, NH₃. Here is a QuickTime movie of the chemical bond in NH₃. The movie shows the electron configuration for a nitrogen atom and for the hydrogen atoms. The nitrogen atom has five valence electrons with an unpaired electron in each of its 2p orbitals. The hydrogen atom has one valence electron in its 1s orbital. When the three 2p orbitals on the nitrogen atom overlaps with the 1s orbital on the hydrogen atoms forming three covalent bonds. The interesting observation is the bond angle we would expect from our Valence Bond Model is 90 degrees, but the experimental observation is a bond angle of 107.5 degrees. To explain the experimental bond angle we will need use different orbitals on the nitrogen atom.

Finally a new question arises when we look at the electron configuration for carbon. It has four valence electrons, two in the 2s orbital and two electrons in the 2p orbitals. Experimentally the simplest compound containing carbon and hydrogen is methane, CH₄. All four of the C-H bonds are identical in length and all the bond angles are the same. How can we understand this experimental fact with our VB model? The answer is--we can not unless we modify the model. The way we modify the model is to promote a 2s electron into the empty 2p orbital. This promotion would require energy, which is more than made up for by the formation of two additional bonds. With this new 'promoted' electron arrangement on the carbon atom three 1s orbitals on hydrogen would interact with the three singly occupied 2p orbitals and one 1s orbital on hydrogen would interact with the singly occupied 2s orbital. We would expect two different kinds of bonds. Experimentally we only observe one kind of C-H bond, that is, all are the same. To explain this, VB theory postulates that after promoting the 2s electron into the empty 2p orbital a reorganization of the atomic orbitals on carbon occurs to produce four identical orbitals. Here is a QuickTime movie of the promotion and mixing of the atomic orbitals on carbon.

This reorganization of the 2s and three 2p orbitals is called hybridization. The four atomic orbitals on carbon (one 2s and three 2p) hybridize and form four identical orbitals each 25 % 's' character and 75 % 'p' character. These new orbitals are called sp3 hybrid orbitals. They are each singly occupied and can bond with a 1s orbital on hydrogen forming four identical C-H bonds. We already know the geometry of this molecule is tetrahedral. The bond angle between each adjacent hybrid orbital is 109.5 degrees.

The same hybridization can be used to describe the bent geometry in H₂O and the trigonal pyramidal geometry in NH₃. In water two of the sp³ hybrid orbitals contain a lone pair of electrons. In NH₃ one of the hybrid orbitals contains a lone pair of electrons.