While the octet rule is a useful model that allows you to picture the structure of molecules, it is important to realize that all MOLECULES do not obey the octet rule. The concept serves only as a rule of thumb.


Compounds also exist in which the central atom has more than an octet of electrons. All of the compounds formed from the noble gas elements (Argon on down) are examples.

A very important concept to remember: ONLY Carbon, Nitrogen, Oxygen, and Fluorine, MUST have an octet !

The existence of compounds of noble gas elements was thought to be impossible - because the noble gas atoms already have complete octets. One of the first noble gas compounds to be synthesized was xenon tetrafluoride, XeF4. The electron dot structure for XeF4 has twelve electrons in the valence orbitals of xenon.


There is no direct relationship between the formula of a compound and the shape of its molecules. The shapes of these molecules can be predicted from their Lewis structures, however, with a model developed about 30 years ago, known as the valence-shell electron-pair repulsion (VSEPR) theory.

The VSEPR theory assumes that each atom in a molecule will achieve a geometry that minimizes the repulsion between electrons in the valence shell of that atom. The five compounds shown below can be used to demonstrate how the VSEPR theory can be applied to simple molecules.

















There are only two places in the valence shell of the central atom in CO2 where electrons can be found. Repulsion between these pairs of electrons can be minimized by arranging them so that they point in opposite directions. Thus, the VSEPR theory predicts that CO2 should be a linear molecule, with a 1800 angle between the two C - O double bonds.

There are three places on the central atom in boron trifluoride (BF3) where valence electrons can be found. Repulsion between these electrons can be minimized by arranging them toward the corners of the equilateral triangle. The VSEPR theory, therefore, predicts a trigonal planar geometry for the BF3 molecule, with a F - B - F bond angle of 1200. Also, it is important to note that boron is VERY HAPPY with only six electrons and not eight. This is a "tricky" element that is overlooked easily.

CO2 and BeF3 are both two-dimensional molecules, in which the atoms lie in the same plane. If we place the same restriction on methane (CH4), we should get a square-planer geometry in which the H - C - H bond angle is 900. If we let this system expand into three dimensions - we end up with a tetrahedral molecule in which the H - C - H bond is approximately 1090.

Repulsion between the five pairs of valence electrons on the phosphorus atom PF5 can be minimized by distributing these electrons toward the corners of a trigonal bipyramid. Three of the positions in a trigonal bipyramid are labeled equatorial because they lie along the equator of the molecule. The other two are axial because the they lie along an axis perpendicular to the equatorial plane. The angle between the three equatorial positions is 1200, while the angle between an axial and an equatorial position is 900.

There are six places on the central atom in SF6 where valence electrons can be found. The repulsion between these electrons can be minimized by distributing them toward the corners of an octahedron. The term octahedron literally means "eight sides," but it is the six corners, or vertices, that interest us. To imagine the geometry of an SF6 molecule, locate fluorine atoms on opposite sides of the sulfur atom along the X, Y, and Z axes of an XYZ coordinate system.


The valence electrons on the central atom in both NH3 and H2O should be distributed toward the corners of a tetrahedron, as shown in the figure below. Our goal, however, is not predicting the distribution of valence electrons. It is to use this distribution of electrons to predict the shape of the molecule. Until now, the two have been the same. Once we include nonbonding electrons, that is no longer true.

The VSEPR theory predicts that the valence electrons on the central atoms in ammonia and water will point toward the corners of a tetrahedron. Because we cannot locate the nonbonding electrons with any precision, this prediction cannot be tested directly. But the results of the VSEPR theory can be used to predict the positions of the nuclei in these molecules, which can be tested experimentally. If we focus on the positions of the nuclei in ammonia, we predict that the NH3 molecule should have a shape best described as trigonal pyramidal, with the nitrogen at the top of the pyramid. Water, on the other hand, should have a shape that can be best descried as bent, or angular. Both of these predictions have been shown to be correct, which reinforces our faith in the VSEPR theory.

Predict the shape of the following molecules.







Draw the Lewis Dot Structure and then name the shape of the following molecules, :

2a. HOCl


2b. CCl4


2c. H2CO


2d. SeO2


2e. SF6


2f. PO4-3

Compounds that contain double and triple bonds raise an important point: The geometry around an atom is determined by the number of places in the valence shell of an atom where electrons can be found, not the number of pairs of valence electrons. Consider the Lewis structures of carbon dioxide (CO2) and carbonate (CO32-) ion, for example.


There are four pairs of bonding electrons on the carbon atom in CO2, but only two places where these electrons can be found. (There are electrons in the C=O double bond on the left and electrons in the double bond on the right.) The force of repulsion between these electrons is minimized when the two C=O double bonds are placed on opposite sides of the carbon atom . The VSEPR theory, therefore, predicts that CO2 will be a linear molecule, just like BeF2, with a bond angle of 1800.

The Lewis structure of the carbonate ion also suggests a total of four pairs of valence electrons on the central atom. But these electrons are concentrated in three places: The two C - O single bonds and the C=O double bond. Repulsion between these electrons are minimized when the three oxygen atoms are arranged toward the corners of an equilateral triangle. The CO3-2 ion should therefore have a trigonal-planer geometry, just like BF3, with a 1200 bond angle.


Bond polarities (Polar Bonds) arise from bonds between atoms of different electronegativity. When more complex molecules are considered we must consider the possiblility of molecular polarities that arise from the sums of all of the individual bond polarities.

To do full justice of molecular polarity - one must consider the concept of vectors (mathematical quantities that have both direction and magnitude).

Let's begin by thinking of a polar bond as a VECTOR pointed from the positively charged atom to the negatively charged atom. The size of the vector is proportional to the difference in electronegativity of the two atoms.

If the two atoms are identical, the magnitude ofthe vector is ZERO, and the molecule has a nonpolar bond.


Let's consider molecules with three atoms. We can establish from the Lewis dot symbols and VSEPR that CO2 is a linear molecule. Each of the C - O bonds will have a vector arrow pointing from the carbon to the oxygen. The two vectors should be identical and pointed in exactly opposite directions.

The sum of these two vectors must be zero because the vectors must cancel one another out. Even though the C - O bonds must be polar, the CO2 MOLECULE is NONPOLAR.

- - - -


HCN, hydrogen cyanide, is linear. Since carbon is more electronegative than hydrogen one would expect a vector pointing from H to C. In addition, nitrogen is more electronegative than C so one should expect a bond vector pointing from C to N. The H-C and C-N vectors add to give a total vector pointing from the H to the N.

HCN is a POLAR MOLECULE with the vector moving from the hydrogen to the nitrogen - making the hydrogen end somewhat positive and the nitrogen end is somewhat negative.

- - - -


In contrast, let's examine the case of SO2. We know from the Lewis dot symbol and from VSEPR that this molecule is "bent." Its overall geometry would be considered to be trigonal planar if we considered the lone pair electrons on the Sulfur.

Lone pair electrons are NOT considered when we examine polarity since they have already been taken into account in the electronegativity.

We would predict that there should be polarity vectors pointing from the sulfur to the two oxygens. Since the molecule is bent the vectors will NOT cancel out. Instead they should be added together to give a combined vector that bisects the O-S-O angle and points from the S to a point in-between the two oxygens.

- - - - - - - -