Naming Ionic Compounds

Every ionic compound has exactly one correct systematic name. Chemists agreed on this naming system for a clear reason. Any formula — from table salt to a lab reagent bottle — can be written or read without ambiguity. The rules rest on two ideas already familiar from Ions and Ionic Charge. Cations and anions combine so their charges cancel. Each ion carries its own name that never changes from compound to compound. This article works through binary ionic compounds (two elements), compounds containing Polyatomic Ions, and the trickier case of transition metals with variable charge. In that case, a Roman numeral must be included. The same logic solves every case: identify the ions, find their charges, balance them, name them in order. It is the same logic used later to build names for Naming and Formulas of Acids.

Key takeaways

Ionic compound names always list the cation first, anion second.
Monatomic anions take the suffix -ide (chlorine → chloride).
Metals with only one possible charge (Groups 1, 2, Al, Zn, Ag) need no Roman numeral.
Transition and post-transition metals with variable charge need a Roman numeral showing the cation's charge in that specific compound.
Polyatomic Ions keep their memorized names unchanged; only the subscripts in the formula change, never the ion's internal atom ratio.
Key idea: the name encodes the identity and charge of each ion; the formula's subscripts are recovered by making total positive and negative charge cancel exactly.

The core insight

A correct ionic name specifies the cation, the anion, and — whenever the cation's charge is not fixed by its group position — a Roman numeral stating that charge explicitly.

The physical picture — prose + figures

An ionic compound is a repeating three-dimensional array of cations and anions held together by electrostatic attraction, not a collection of discrete molecules. Because there is no single "molecule" of NaCl, the formula NaCl represents a ratio — one Na⁺ for every one Cl⁻ — and the name must communicate that ratio through charge bookkeeping rather than through molecular structure. The naming task splits cleanly into two independent jobs. First, name the cation: if it is a main-group metal ion with one fixed charge, use the element name unchanged (sodium, calcium, aluminum). If it is a transition metal or post-transition metal capable of more than one charge, append a Roman numeral in parentheses equal to that charge in the compound at hand. Second, name the anion: if it is monatomic, drop the element's ending and add -ide (chlorine → chloride, oxygen → oxide); if it is polyatomic, use its memorized name unchanged (sulfate, nitrate, phosphate). The subscripts in the formula are never part of the name. Fe₂O₃ is "iron(III) oxide," not "diiron trioxide" — that Greek-prefix style belongs to molecular (covalent) compounds, a different naming system entirely. The Roman numeral already tells you, indirectly, how many oxide ions must appear to balance the charge.

Figure

Cation and anion charges must sum to zero: Al³⁺ paired with O²⁻ requires a 2:3 ratio

Figure

Iron forms two common cations; the Roman numeral removes the ambiguity that the name "iron oxide" alone would leave

The core relations

The entire subscript-finding procedure reduces to one balance condition and one reduction rule.

n_{+} \times q_{+} + n_{-} \times q_{-} = 0

where:

$n_{+}$ = number of cations in the formula unit
$q_{+}$ = charge on the cation (a positive number, e.g. +2)
$n_{-}$ = number of anions in the formula unit
$q_{-}$ = charge on the anion (a negative number, e.g. −3)

\frac{n_{+}}{n_{-}} = \frac{|q_{-}|}{|q_{+}|} \quad \text{(reduced to lowest whole-number ratio)}

where:

$|q_{-}|$ , $|q_{+}|$ = magnitudes of anion and cation charge
the ratio is always simplified to the smallest integers, e.g. 4∶6 → 2∶3

For a variable-charge metal, the Roman numeral $N$ satisfies:

N = |q_{+}| \quad \text{such that } n_{+}(N) + n_{-}q_{-} = 0

where:

$N$ is determined after the anion's identity and charge are known, since the Roman numeral describes the metal's charge in that specific compound only.

Derivation AFinding the formula and name from ion charges

Goal. Given a cation and an anion, produce both the correct formula (with reduced subscripts) and the correct name, including a Roman numeral when needed.

Step 1. Identify each ion and its charge. Look up or deduce each ion's charge from its group (main-group metals) or from the specific compound's context (transition metals). For example, calcium is always Ca²⁺; iron may be Fe²⁺ or Fe³⁺ depending on the compound.

Step 2. Write the "crossed charges" ratio. Take the magnitude of the cation's charge and use it as the anion's subscript; take the magnitude of the anion's charge and use it as the cation's subscript. This is a mechanical way of solving the balance equation:

n_{+} = |q_{-}|, \qquad n_{-} = |q_{+}|

Step 3. Reduce the subscripts to lowest whole-number terms. If $n_+$ and $n_-$ share a common factor, divide both by it. For calcium (2+) and oxide (2−), crossing charges gives Ca₂O₂, which reduces to CaO. Skipping this step is the single most common formula-writing error.

Step 4. Confirm the net charge is exactly zero.

n_{+}q_{+} + n_{-}q_{-} = 0

This check catches any arithmetic slip from Steps 2–3 before the name is written.

Step 5. Assign the Roman numeral if the cation's charge is not fixed. If the cation belongs to a family with only one possible charge (Group 1 → 1+, Group 2 → 2+, Al³⁺, Zn²⁺, Ag⁺), no Roman numeral is used. Otherwise, back-calculate the cation's charge in this compound from the anion's known charge and subscript, then write that value as a Roman numeral in parentheses directly after the metal's name, with no space before the parenthesis.

Step 6. Assemble the name: cation name (+ Roman numeral if needed), then anion name. Monatomic anions get the -ide ending; polyatomic anions keep their standard name unchanged. The subscripts themselves never appear as prefixes in the ionic-compound name.

Worked examples

Example 1 — Tin(IV) with sulfide (asymmetric charges, needs reduction and a Roman numeral). Tin commonly forms Sn²⁺ or Sn⁴⁺. Suppose the compound in hand is confirmed by experiment to contain Sn⁴⁺, and the anion is sulfide, S²⁻. Step 2 (cross charges):

n_{+} = |q_{-}| = 2, \qquad n_{-} = |q_{+}| = 4

giving the trial formula Sn₂S₄. Step 3 (reduce): both subscripts share a factor of 2:

\frac{n_{+}}{n_{-}} = \frac{2}{4} = \frac{1}{2}

so the reduced formula is SnS₂. Step 4 (check charge balance):

(1)(+4) + (2)(-2) = 4 - 4 = 0 \checkmark

Step 5 (Roman numeral): the cation charge used was 4+, so the numeral is (IV). Step 6 (name): tin(IV) sulfide. As a limiting-case check, compare with the other tin sulfide: if instead Sn²⁺ pairs with S²⁻, crossing charges gives Sn₂S₂, reducing to SnS — tin(II) sulfide — a genuinely different compound with a different formula and different Roman numeral, exactly as the variable-charge system is designed to distinguish.

Example 2 — Ammonium phosphate (two different polyatomic ions, non-1:1 ratio). Combine ammonium, NH₄⁺, with phosphate, PO₄³⁻. Step 2 (cross charges):

n_{+} = |q_{-}| = 3, \qquad n_{-} = |q_{+}| = 1

giving (NH₄)₃(PO₄)₁, written (NH₄)₃PO₄. Step 3 (reduce): 3 and 1 share no common factor greater than 1, so no reduction is possible; the formula stands as (NH₄)₃PO₄. Step 4 (check):

3(+1) + 1(-3) = 3 - 3 = 0 \checkmark

Step 5: ammonium's charge (1+) is fixed by the ion itself — no Roman numeral applies to polyatomic cations. Step 6 (name): ammonium phosphate. Notice the parentheses around NH₄ are required whenever a polyatomic ion's subscript exceeds 1, purely as a formula-writing convention; they carry no separate meaning in the spoken or written name.

Limiting cases & common traps

Situation	What to watch for
Fixed-charge metal (Group 1, 2, Al, Zn, Ag)	Never add a Roman numeral — "sodium(I) chloride" is wrong
Subscripts reduce (e.g., Sn₂S₂)	Always simplify to lowest terms before naming
Polyatomic ion appears more than once	Enclose it in parentheses in the formula; name is still unchanged
Same metal, different anion charge	Roman numeral can differ compound to compound (FeO vs Fe₂O₃)
Hydrated ionic compounds	Add "hydrate" prefix with a Greek numeral for water molecules — a separate rule layered on top of the base name

The chemist behind it

The Roman-numeral (Stock) system is named for the German chemist Alfred Stock, who proposed it in the early twentieth century to resolve exactly the ambiguity transition metals create. Before his convention, chemists relied on Latin-derived suffixes — ferrous for Fe²⁺ and ferric for Fe³⁺, cuprous for Cu⁺ and cupric for Cu²⁺ — which required memorizing a separate pair of Latin roots for every variable-charge metal. Stock's proposal replaced this patchwork with a single rule: state the charge directly as a Roman numeral. The older Latin names still appear occasionally on reagent labels and in older literature, but the Stock system is now the standard taught and used internationally, precisely because it generalizes to any metal without new vocabulary.

Origins · 1 minAlfred Stock and how this was discovered

Alfred Stock

1876–1946 · German inorganic chemist, pioneer of boron and silicon hydride chemistry, proposed the Roman-numeral oxidation-state naming convention now standard in IUPAC nomenclature

Ask the tutor: How do I name a compound when both the cation and the anion are polyatomic ions, such as ammonium sulfate versus ammonium sulfite?

Why do some metals need a Roman numeral and others never do?

Main-group metals in Groups 1 and 2, along with a few common ions like Al³⁺, Zn²⁺, and Ag⁺, lose a fixed number of electrons determined by their position in the periodic table, so they only ever form one charge. Transition metals and several post-transition metals (Sn, Pb, Bi) can lose different numbers of electrons depending on the compound, so their charge must be stated explicitly each time using the Roman numeral, as described in Ions and Ionic Charge.

Do I ever write a subscript of 1 in an ionic formula?

No. A subscript of 1 is implied and never written, matching the convention used for Chemical Formulas generally. NaCl means one Na⁺ and one Cl⁻; you never write Na₁Cl₁.

How is naming an ionic compound different from naming a molecular (covalent) compound?

Ionic names never use Greek numerical prefixes (mono-, di-, tri-) because the formula's ratio is fully determined by charge balance and therefore doesn't need to be stated separately. Molecular compounds, which involve no charge balancing, instead use those prefixes explicitly — compare "carbon dioxide" (CO₂, covalent, prefix "di-" stated) to "iron(III) oxide" (Fe₂O₃, ionic, no prefixes, Roman numeral instead).

What happens when a polyatomic ion itself contains a subscript needing multiplication?

You enclose the entire polyatomic ion in parentheses and add the multiplying subscript outside, as in Ca(OH)₂ or (NH₄)₂SO₄. This shows unambiguously that the subscript 2 multiplies the whole ion, not just the last atom in its formula.

Where does this naming system stop applying?

Compounds formally treated as acids, where an H⁺ pairs with an anion and the whole substance is dissolved or considered in its acidic form, follow a distinct nomenclature layered on top of these same anion names, covered separately in Naming and Formulas of Acids.