Oxidation number

In reactions between covalent molecules, electrons are shared, not fully transferred, so we keep track with oxidation numbers: the charge an atom would have if all its bonds were fully ionic. The rules:

• An atom in a free element has oxidation number 0 (Na, O₂, P₄, solid sodium).
• A monatomic ion has its own charge (Na⁺ is +1, Cl⁻ is –1).
• Oxygen is usually –2, but –1 in peroxides (such as H₂O₂), –½ in superoxides (such as KO₂), and +2 in OF₂.
• Hydrogen is usually +1, but –1 in metal hydrides (such as NaH).
• Fluorine is always –1. The other halogens are usually –1, but can be positive when bonded to oxygen.
• The oxidation numbers of all the atoms add up to the charge of the species (zero for a neutral molecule).

To find an unknown oxidation number, call it x and use the last rule. For chromium in Cr₂O₇²⁻: 2x + 7(–2) = –2, so 2x = 12, and x = +6.

Now redox can be defined simply: oxidation is an increase in oxidation number; reduction is a decrease.

Stock notation writes a metal's oxidation number in Roman numerals after it: iron(III) oxide is Fe₂O₃ (iron at +3), and manganese(IV) oxide is MnO₂.

Oxidation numbers can come out as fractions. These are averages. In Fe₃O₄, iron averages +8/3, because the compound is really FeO·Fe₂O₃, with one iron at +2 and two at +3. In S₄O₆²⁻, two sulphur atoms are at +5 and two at 0, averaging +2.5.

The highest oxidation number of a main-group element usually equals its group's number of valence electrons: from sodium +1 up to chlorine +7. If a calculation gives an impossible value (above that maximum), the formula must contain a hidden structural feature. Sulphur in H₂SO₅ seems to be +8, which is impossible for sulphur; in fact the molecule contains a peroxide link (–O–O–), whose oxygens are –1, not –2.