How to Read a Chemical Formula

Co and CO are not the same thing. The first is cobalt, a metal. The second is carbon monoxide, a gas that kills people in their sleep. The only difference in the notation is whether the second letter is capitalised, and that single convention carries more weight than any other rule in chemical writing.

Everything else in a formula is built on rules of comparable strictness. Once you know them, a string like (NH₄)₂Cr₂O₇ stops being a wall of characters and becomes a sentence with a subject and a count.

The capital letter starts a new element

An element symbol is one capital letter, optionally followed by one lowercase letter. That is the entire parsing rule, and it is unambiguous by design.

• NI would be nitrogen followed by iodine. Ni is nickel.
• SO is sulfur then oxygen. So is nothing at all.
• NO is nitrogen monoxide. No is nobelium.

This is why handwritten chemistry is marked strictly on capitalisation, and why a formula typed in all caps is genuinely unreadable rather than merely untidy. Symbols come from the element's name in whichever language fixed the convention, which is why a few of them look unrelated to the English word — sodium is Na and iron is Fe, and there are eleven such cases.

A subscript counts the symbol immediately before it

H₂O is two hydrogen atoms and one oxygen. No subscript means one; nobody writes H₂O₁.

The subscript belongs to the symbol on its left and to nothing else. In C₆H₆, the first 6 counts carbon and the second counts hydrogen, and the fact that they are equal is a fact about benzene rather than a rule about formulas.

Brackets multiply everything inside them

This is where most misreadings happen. A subscript after a closing bracket applies to every atom in the group.

Ca₃(PO₄)₂ contains three calcium atoms, two phosphorus atoms and eight oxygen atoms — four oxygens in each phosphate group, times two groups. Read casually it looks like it might contain four.

Brackets exist because the group inside them is a real chemical unit that stays together: a phosphate ion, a nitrate ion, an ammonium ion, a hydroxide ion. Writing Ca₃P₂O₈ would give the same atom count and destroy the information that the compound is built from phosphate ions. The formula is not just an inventory; it is a claim about structure.

Square brackets do a stronger version of the same job. In K₃[Fe(CN)₆] the square brackets enclose a complex ion in which one iron atom is bonded to six cyanide groups, and the round brackets inside count the cyanides. Reading it from the inside out: six carbons, six nitrogens, one iron, three potassiums.

The raised dot means water that is loosely attached

CuSO₄·5H₂O is copper sulfate pentahydrate, and the dot is doing something no other symbol in chemical notation does. It says that five water molecules are incorporated into the crystal structure for every copper sulfate unit — held in the lattice, real and countable, but not bonded the way the sulfate is.

The practical consequence is that the hydrated and anhydrous forms are different substances with different masses and different colours, and that heating drives the water off. If you are counting atoms, the water counts. This is the single most reliable source of a wrong answer in molar mass arithmetic.

Coefficients sit in front and mean something else entirely

In an equation you will see numbers before a formula as well as inside it, and they do not interact.

• H₂O is one water molecule.
• 2H₂O is two water molecules — four hydrogen atoms and two oxygens.
• H₂O₂ is one molecule of hydrogen peroxide, a different compound.

The coefficient scales the whole formula and can be changed when balancing an equation. The subscripts cannot: changing a subscript changes what substance you are talking about, which is why "just add a 2 to the oxygen" is the classic wrong move in balancing.

Superscripts are charge, and the sign comes last

A raised number with a plus or minus is an ionic charge, written with the digit first and the sign second: Fe³⁺, SO₄²⁻, OH⁻. A charge of one is written as a bare sign with no digit.

Charge and count sit on opposite sides of the symbol for exactly this reason. In SO₄²⁻ the subscript 4 counts oxygens and the superscript 2− is the charge on the whole ion. The two numbers have nothing to do with each other, and reading the ion as containing two of anything is a misreading.

A raised number on the left of a symbol is something else again: a mass number. ¹⁴C is the isotope carbon-14, not a compound and not an ion.

The letters in brackets after a formula are physical states

• (s) solid
• (l) liquid
• (g) gas
• (aq) aqueous — dissolved in water

These matter more than they look. NaCl(s) and NaCl(aq) behave completely differently, because in solution the compound has separated into free sodium and chloride ions that can move independently. Half of introductory solution chemistry depends on noticing which one a question means.

The same substance has more than one correct formula

Three different formulas can describe glucose, and each answers a different question.

• CH₂O, the empirical formula, has been divided through: its subscripts are glucose's 6 : 12 : 6 reduced by a factor of six, so the string stays arithmetically honest and goes silent about how big the molecule is.
• C₆H₁₂O₆, the molecular formula, counts one molecule and stops there. Fructose and galactose are written exactly the same way, which is a fair measure of how much a count leaves out.
• HOCH₂(CHOH)₄CHO, a condensed structural formula, is read left to right as a walk along the molecule. Its bracket repeats a group four times down a chain rather than enclosing an ion, and the multiplication rule above applies to it unchanged.

The empirical formula is what an elemental analysis gives you directly, and it cannot distinguish glucose from formaldehyde. Going from one to the other requires knowing the molar mass, and that relationship is a standard exam question.

Reading a name backwards into a formula

Names encode the same information in words, using two systems that are easy to confuse.

Greek prefixes count atoms, and are used for compounds of two nonmetals. Dinitrogen tetroxide is N₂O₄. Carbon monoxide is CO, carbon dioxide is CO₂. The count is stated because these elements combine in several different ratios and none of them is implied by the names alone.

Roman numerals state an oxidation state, and are used where a metal has more than one. Iron(II) chloride is FeCl₂ and iron(III) chloride is FeCl₃ — the numeral gives the charge on the metal, and you work out how many chlorides are needed to balance it. It does not count anything directly, which is the trap. Copper(II) sulfate contains one copper, not two. Oxidation states are worth understanding properly for this reason alone.

Where a metal has only one common state, no numeral is used: sodium chloride is NaCl and needs no qualification.

One formula, read end to end

(NH₄)₂Cr₂O₇ — ammonium dichromate.

Start inside the round brackets: N with no subscript is one nitrogen, H₄ is four hydrogens, and together they make the ammonium group. The 2 outside the bracket doubles it: two nitrogens, eight hydrogens. Then Cr₂ is two chromium atoms and O₇ is seven oxygens, which together form the dichromate ion. Total: 2 N, 8 H, 2 Cr, 7 O.

The bracket also tells you the compound is ionic and built from two ammonium ions and one dichromate ion, which is information that the atom count alone would not give you.

Seven rules, in the order you meet them

Capitalisation separates elements. A subscript counts the symbol before it. A bracket subscript counts everything inside. A dot adds water. A front coefficient scales the whole formula. A right-hand superscript is charge; a left-hand one is a mass number. Letters in round brackets after the formula are the physical state.

Seven rules, no exceptions worth mentioning, and between them they will parse any formula you meet before university.