IUPAC Nomenclature

Read the formula from the left. If it starts with a metal, or with the ammonium ion ${\text{NH}}_4^{+}$, the compound is ionic: name the positive ion, then the negative ion, and never use prefixes, because the charges already fix the ratio. Next ask whether the metal is multivalent — able to form more than one charge. Most transition metals are, and so are tin and lead; silver (always 1+) and zinc (always 2+) are the common exceptions that need no Roman numeral. If the formula contains only nonmetals, the compound is molecular and the atoms share electrons instead of forming ions, so nothing fixes the ratio and prefixes must state it. Finally, a hydrogen compound dissolved in water, such as ${\text{HCl}}_{(aq)}$, is named as an acid (hydrochloric acid) — those rules are covered in the acids and bases unit. Choosing the wrong branch is the source of most naming errors, so make this decision deliberately every time.

1. Name the metal (cation) exactly as it appears on the periodic table.
2. Name the nonmetal (anion) by keeping its root and changing the ending to "-ide".
3. No prefixes (di-, tri-, …) are used — the ratio is implied by the charges, not stated in the name.
4. Examples: $\text{NaCl}$ = sodium chloride. $\text{MgO}$ = magnesium oxide. ${\text{Al}}_2{\text{O}}_3$ = aluminum oxide (still no prefixes — the 2 and 3 are determined by charge balance, not named). ${\text{Li}}_3\text{N}$ = lithium nitride.

Many transition metals (and a few main-group metals like Sn, Pb) can form more than one stable ion charge — e.g. iron can be Fe²⁺ or Fe³⁺. They can lose different numbers of electrons depending on what they react with, so the formula alone does not tell a reader which ion is present unless the name says so. For these, a Roman numeral in brackets after the metal's name states its charge in that specific compound, found by working backward from the anion's known charge. The Roman numeral is the charge of one metal ion, not the number of metal atoms: ${\text{Cu}}_2\text{O}$ is copper(I) oxide even though it contains two copper ions. Going the other way (name to formula), read the charge from the numeral and balance it exactly as you would for any ionic compound.

A polyatomic ion is a tightly bonded group of atoms that carries an overall charge and moves together as a single unit. Their names must be memorized, but a few patterns help. Most negative polyatomic ions contain oxygen and end in "-ate" or "-ite": the "-ate" ion has one more oxygen than the "-ite" ion with the same charge (nitrate ${\text{NO}}_3^{-}$ vs nitrite ${\text{NO}}_2^{-}$; sulfate ${\text{SO}}_4^{2-}$ vs sulfite ${\text{SO}}_3^{2-}$). Adding "hydrogen" to the front adds an ${\text{H}}^{+}$ and raises the charge by one (carbonate ${\text{CO}}_3^{2-}$ becomes hydrogen carbonate ${\text{HCO}}_3^{-}$, the "bicarbonate" in baking soda). In names, polyatomic ions keep their own names — no "-ide" is added, so ${\text{Na}}_2{\text{SO}}_4$ is sodium sulfate. When more than one of a polyatomic ion is needed to balance charge, put brackets around the whole ion before adding the subscript: ${\text{Ca(OH)}}_2$, not ${\text{CaOH}}_2$ (which would mean one O and two H).

For compounds of two nonmetals, Greek prefixes state exactly how many of each atom are present — because, unlike ionic compounds, you cannot predict the ratio from charge alone. Nitrogen and oxygen alone form NO, ${\text{NO}}_2$, ${\text{N}}_2\text{O}$, ${\text{N}}_2{\text{O}}_4$ and ${\text{N}}_2{\text{O}}_5$, so the name has to carry the count. Write the element farther left on the periodic table first, and give the second element an "-ide" ending.