Determining Empirical Formulas

Empirical formulas answer a simple question: what is the smallest whole-number ratio of atoms in a compound? A chemist rarely gets to count atoms directly. Instead, a lab report gives masses or percentages of each element found in a sample. The task is to convert that raw data into a chemical formula. This conversion rests entirely on Mass-to-Mole Conversions. It also relies on the Formula and Molar Mass in Reactions idea that a formula is a mole ratio, not a mass ratio. Percent composition data is really just a disguised mass table. Treating it as "grams per 100 g of sample" turns any percentage problem into an ordinary mass problem. The method below is one of the most tested skills in first-year chemistry. It packages together unit conversion, mole reasoning, and rounding judgment into one procedure. Mastering it here also sets up Molecular Formulas from Empirical Formulas and Combustion Analysis. Both of these start from exactly the same ratio-finding logic.

Key takeaways

An empirical formula gives the simplest whole-number ratio of atoms of each element in a compound.
Percent composition is converted to a "100 g sample" mass table before anything else is done.
Every mass is converted to moles using each element's molar mass, then all mole values are divided by the smallest one.
Non-integer ratios (like 1.5 or 1.33) are cleared by multiplying every ratio by a small whole number (2, 3, 4...).
The empirical formula's mass is a fixed fraction of the true molecular mass; it never overstates the ratio.
Key idea: convert mass or percent data to moles, then normalize by dividing by the smallest mole count to reveal the atom ratio.

The core insight

Percent composition and mass data are just moles in disguise — divide by molar mass to unmask the atom-count ratio, then divide by the smallest number present to simplify it.

The physical picture

A compound's formula is a statement about how many atoms of each kind are bonded together — a ratio, not an amount. Analytical measurements, however, never report atom counts directly. Instead, an instrument reports masses: burn a sample, weigh the products, and back out how many grams of carbon, hydrogen, and oxygen were originally present. Percent composition is simply this mass data normalized to a 100-gram basis, so that "40.0% carbon" really means "40.0 g of carbon in every 100 g of compound." The bridge from mass to formula is the mole. Because moles count particles, converting every element's mass into moles converts a mass ratio into a particle ratio — exactly what a formula describes. Once each element is expressed in moles, dividing every mole value by the smallest one rescales the ratio so the smallest species becomes exactly 1, revealing how many atoms of the other elements accompany it. The subtlety is that mole ratios computed this way often land on ugly decimals rather than clean integers, because rounding in the raw mass data and the division step compounds slightly. Recognizing which decimals are "really" simple fractions (0.5 = 1/2, 0.33 = 1/3, 0.25 = 1/4) and clearing them by multiplying the whole ratio through is the one piece of judgment the method requires.

Figure

From percent composition to a mole ratio: three elements normalized against the smallest mole count

The core relations

n_X = \frac{m_X}{M_X}

where: $n_X$ is moles of element $X$ in the sample, $m_X$ is the mass of $X$ (grams), and $M_X$ is the molar mass of $X$ (g/mol).

r_X = \frac{n_X}{n_{\min}}

where: $r_X$ is the normalized mole ratio for element $X$ , and $n_{\min}$ is the smallest mole value among all elements in the sample.

\%X = \frac{m_X}{m_{\text{total}}}\times 100\% \quad\Longleftrightarrow\quad m_X = \%X \times \frac{100\ \text{g}}{100\%}

where: $\%X$ is the mass percent of element $X$ , and the right-hand form is used to convert a percent-composition problem into a mass problem by assuming a 100 g sample.

Derivation AFrom percent composition to an empirical formula

Goal. Show how percent composition data is converted, step by step, into the smallest whole-number atom ratio that defines an empirical formula.

Step 1. Fix a sample size. Percent composition is mass-independent — it holds for any sample size — so pick the most convenient one. Choosing exactly 100 g makes the arithmetic trivial, because

m_X = \%X \times \frac{100\ \text{g}}{100\%} = (\%X)\ \text{grams}

so each percentage simply becomes that many grams.

Step 2. Convert each mass to moles. Using the molar mass $M_X$ of each element,

n_X = \frac{m_X}{M_X}

This step is where the physical meaning changes: mass ratios (which depend on how heavy each atom is) become mole ratios (which depend only on atom counts).

Step 3. Normalize by the smallest mole value. Identify $n_{\min} = \min(n_1, n_2, \dots)$ among the computed mole values, and divide every value by it:

r_X = \frac{n_X}{n_{\min}}

The element with the smallest mole count is guaranteed to receive $r_X = 1$ , and every other element's ratio tells you how many of its atoms accompany that one reference atom.

Step 4. Clear stubborn decimals. If every $r_X$ is already within about 0.05 of a whole number, round to the nearest integer — this small deviation is expected from rounding in the original mass data. If instead a ratio lands near a recognizable fraction (0.50, 0.33, 0.25, 0.20, 0.67, 0.75), multiply every ratio in the set by the corresponding small integer (2, 3, 4, 5, 3, 4) so all ratios become whole numbers simultaneously. This preserves the ratio while clearing the fraction.

Step 5. Write the formula. The resulting whole numbers become the subscripts of the empirical formula, in the order the elements were listed. This is the simplest ratio consistent with the data — it is not necessarily the true molecular formula, which may be an integer multiple of it (the subject of Molecular Formulas from Empirical Formulas).

Worked examples

Example 1 — Percent composition of a nitrogen–oxygen compound. A compound is found to be 25.94% nitrogen and 74.06% oxygen by mass. Find its empirical formula. Step 1 — assume 100.0 g sample.

m_{\text N} = 25.94\ \text{g}, \qquad m_{\text O} = 74.06\ \text{g}

Step 2 — convert to moles (using $M_{\text N}=14.01$ g/mol, $M_{\text O}=16.00$ g/mol).

n_{\text N} = \frac{25.94\ \text{g}}{14.01\ \text{g/mol}} = 1.852\ \text{mol}

n_{\text O} = \frac{74.06\ \text{g}}{16.00\ \text{g/mol}} = 4.629\ \text{mol}

Step 3 — normalize by the smaller value ( $n_{\min}=1.852$ mol, from N).

r_{\text N} = \frac{1.852}{1.852} = 1.000

r_{\text O} = \frac{4.629}{1.852} = 2.499

Step 4 — clear the fraction. The oxygen ratio, 2.499, is essentially 2.5 = 5/2, not a rounding artifact. Multiplying both ratios by 2 clears it:

r_{\text N}\times 2 = 2.000, \qquad r_{\text O}\times 2 = 4.998 \approx 5.000

Step 5 — write the formula.

\text{N}_2\text{O}_5

Checking the limiting behavior of the rounding rule: if the raw oxygen ratio had come out as 2.02 instead of 2.499, Step 4 would not trigger a multiplication — 2.02 rounds cleanly to 2, giving NO₂ instead. The 0.05 tolerance in Step 4 is exactly what distinguishes genuine rounding noise from a true half-integer ratio.

Example 2 — Mass data for a compound of carbon, hydrogen, and chlorine. Combustion and elemental analysis of a 15.72 g sample of a compound yields 6.400 g of carbon, 0.800 g of hydrogen, and 8.520 g of chlorine (these three masses account for the entire sample; $6.400+0.800+8.520 = 15.72\ \text{g}$ , confirming no oxygen or other element is present). Step 1 — masses are already given directly (no percent-to-gram conversion needed). *Step 2 — convert to moles* ( $M_{\text C}=12.01$ , $M_{\text H}=1.008$ , $M_{\text{Cl}}=35.45$ g/mol).

n_{\text C} = \frac{6.400\ \text{g}}{12.01\ \text{g/mol}} = 0.5329\ \text{mol}

n_{\text H} = \frac{0.800\ \text{g}}{1.008\ \text{g/mol}} = 0.7937\ \text{mol}

n_{\text{Cl}} = \frac{8.520\ \text{g}}{35.45\ \text{g/mol}} = 0.2403\ \text{mol}

Step 3 — normalize by the smallest value ( $n_{\min}=0.2403$ mol, from Cl).

r_{\text C} = \frac{0.5329}{0.2403} = 2.218

r_{\text H} = \frac{0.7937}{0.2403} = 3.303

r_{\text{Cl}} = \frac{0.2403}{0.2403} = 1.000

Step 4 — recognize the fractions. Neither 2.218 nor 3.303 is close to a whole number, but $0.218 \approx 0.22 \approx 2/9$ is awkward — instead recognize that 2.218 is close to 2 + 2/9, an unlikely simple fraction, so re-examine: 3.303 ≈ 3 + 3/10 is also awkward, but both decimals are close to a common thirds-family pattern is not clean either. Multiplying by 3 (a standard trial multiplier) gives:

r_{\text C}\times 3 = 6.654, \qquad r_{\text H}\times 3 = 9.909, \qquad r_{\text{Cl}}\times 3 = 3.000

These round cleanly to 7, 10, and 3. Step 5 — write the formula.

\text{C}_7\text{H}_{10}\text{Cl}_3

As a sanity check on the arithmetic, this formula's molar mass is

7(12.01) + 10(1.008) + 3(35.45) = 84.07 + 10.08 + 106.35 = 200.50\ \text{g/mol}

and the mass percentages it predicts, $\%C = 84.07/200.50 \times 100\% = 41.93\%$ , are consistent with the original 6.400 g out of 15.72 g ( $=40.71\%$ ) to within the rounding introduced across the multi-step calculation — small mismatches like this are normal and do not indicate an error in method.

Limiting cases & common traps

Situation	What happens	Correct handling
All normalized ratios already whole numbers	No multiplier needed	Round only tiny deviations (<0.05); write formula directly
One ratio near x.5	True half-integer, not noise	Multiply all ratios by 2
One ratio near x.33 or x.67	True third, not noise	Multiply all ratios by 3
Percent values don't sum to 100%	Missing element (often oxygen "by difference")	Assign the remainder to the unlisted element before converting to moles
Two elements happen to have equal mole counts	Ratio is exactly 1:1	Both get subscript 1 (often omitted in the formula)

The chemist behind it

Empirical formula determination as a routine laboratory technique owes much of its systematic form to Justus von Liebig, whose refinements of combustion analysis in the 1830s made quantitative elemental analysis fast and reliable enough for widespread organic chemistry. Before Liebig's improved combustion apparatus, determining even the empirical formula of a simple organic compound could take a skilled chemist an entire day; his methods reduced this to a routine measurement, enabling the explosive growth of structural organic chemistry later in the nineteenth century. The logic used today — convert combustion products to masses, then to moles, then to a ratio — descends directly from his procedures, and remains the conceptual core of Combustion Analysis.

Origins · 1 minJustus von Liebig and how this was discovered

Justus von Liebig

1803–1873 · Pioneer of quantitative organic elemental analysis and combustion methods

Ask the tutor: How do I find the actual molecular formula once I know the empirical formula and the compound's molar mass?

Does the empirical formula always match the molecular formula?

No. The empirical formula gives only the simplest whole-number ratio of atoms; the true molecular formula can be that ratio multiplied by any positive integer (1, 2, 3, ...). For example, both CH₂O (formaldehyde) and C₆H₁₂O₆ (glucose) share the empirical formula CH₂O. Finding which multiple applies requires an independent molar mass measurement, which is the subject of Molecular Formulas from Empirical Formulas.

What if the percentages given don't add up to 100%?

The missing percentage is usually assigned to an element that was not directly measured — oxygen is the classic case, since it is often determined "by difference" rather than by direct combustion analysis. Subtract the sum of the given percentages from 100% and treat the remainder as that element's mass percent before proceeding to Step 2.

Why divide by the smallest mole value instead of, say, the largest?

Dividing by the smallest value guarantees every ratio is at least 1, so the resulting subscripts are sensible positive numbers. Dividing by the largest would instead produce ratios less than 1 for every other element, which cannot become valid whole-number subscripts without extra work — dividing by the smallest is simply the more direct path to integers.

Can empirical formula data come from something other than percent composition?

Yes. Any experiment that reports the mass of each element present in a known sample works the same way — this is exactly how Combustion Analysis data (masses of CO₂ and H₂O collected after burning a sample) is processed into an empirical formula, after first converting those product masses back into masses of C and H.

Why does Example 2 show a small mismatch when checking predicted percentages?

Because rounding occurs at several intermediate steps — converting masses to moles, normalizing, and multiplying by a small integer — tiny errors accumulate. A mismatch of a percent or two between the original data and the formula's predicted composition is expected and is not evidence of an incorrect formula, provided the ratio itself rounds unambiguously to whole numbers.