Propagation of Measurement Uncertainty

Introduction

Stoichiometric coefficients are exact within a chosen balanced equation, but masses, volumes and concentrations are measurements with uncertainty. A long calculation can look precise because a calculator displays many digits. A defensible final answer reflects the reliability of its measured inputs and any model assumptions such as purity, gas collection or reaction completion.

Core explanation

For a mass-to-mass reaction calculation, the pathway is measured reactant mass → reactant moles → target moles → target mass. Formula masses use tabulated atomic masses at suitable precision, and coefficient ratios such as 2:1 do not themselves add measurement uncertainty. If 0.243 g Mg is weighed and converted using M(Mg) = 24.3 g mol⁻¹, the amount is 0.0100 mol to three significant figures. Under Mg + 2HCl → MgCl₂ + H₂, ideal H₂ amount is also 0.0100 mol to precision justified by the input, not 0.010000000 mol simply because the ratio is exact.

Significant-figure rules are a useful reporting convention for school calculations. For multiplication and division, a final result normally has no more significant figures than the least precise measured factor, unless a stated uncertainty analysis supports a different report. Keep extra digits during intermediate steps to avoid rounding drift. Exact counting numbers and balanced coefficients do not set the limit. A concentration stated as 0.100 mol L⁻¹ has three significant figures, while a measured volume 20.00 mL has four; their product should generally be reported to three significant figures.

Subtraction deserves special attention. Titrant volume is final burette reading minus initial reading. The appropriate decimal-place precision of that difference depends on the two readings, not simply the count of significant digits in either. If readings are 1.20 mL and 19.60 mL, delivered volume is 18.40 mL. Both readings carry instrument uncertainty, so the delivered volume has uncertainty from both. Similarly, mass loss calculated as initial minus final mass combines uncertainty from two weighings.

For independent uncertainties in a product or quotient, a first approximation combines relative contributions in quadrature. If m and c each have relative standard uncertainties u m/m and u c/c, then a result proportional to m/c has relative uncertainty approximately sqrt[(u m/m)² + (u c/c)²], assuming independence and small errors. At an introductory level, identifying the dominant input uncertainty and reporting sensible digits is often enough. Adding all percentage uncertainties gives a conservative bound in some settings but is not the same as a statistical standard uncertainty.

Model uncertainty can dominate instrument precision. A gas volume measured to 0.1 mL does not make a metal purity estimate accurate to 0.1% if gas leakage is uncontrolled. A clean-looking number from PV = nRT may still rely on ideal-gas approximation or an uncertain water-vapour correction. State relevant assumptions alongside the numerical answer.

An exact coefficient ratio also means scaling an equation does not change uncertainty. Writing H₂ + ½O₂ → H₂O or 2H₂ + O₂ → 2H₂O gives the same amount and precision when the same measured H₂ sample is used. Coefficients select the conversion, while measurement and model quality determine confidence.

Step-by-step reasoning

1. Identify every measured input, its unit and stated precision or uncertainty. 2. Keep balanced coefficients and counting factors separate as exact numbers. 3. Carry extra digits through mole and mass conversions while showing units. 4. For differences, consider the uncertainty of both readings; for products and quotients, compare relative uncertainties. 5. Round the final result sensibly and state assumptions that might exceed instrument uncertainty.

Visual explanation

Draw a chain 0.243 g Mg → 0.0100 mol Mg → 0.0100 mol H₂ → 0.0202 g H₂. Put a small “measured, 3 significant figures” label on 0.243 g and “exact ratio, 1:1” on the middle arrow. The precision label flows forward without being improved by the coefficient.

Real-world analogy

If a ruler measures a board only to the nearest millimeter, cutting it into two equal pieces does not reveal the board's original length to the nearest micrometer. An exact division rule does not create information absent from the measurement. A balanced equation behaves like that exact rule.

Real-world example

A titration may use a precisely standardized solution but a poorly read endpoint. The concentration result is then limited by endpoint and volume uncertainty more than by the concentration label on the titrant bottle. Repeating concordant trials can help estimate random scatter, while calibration addresses systematic error.

Why?

Why avoid early rounding? A chain may multiply several non-integer molar-mass and concentration factors. Rounding each intermediate result can shift the final answer unnecessarily. Retaining guard digits preserves the information in measured inputs until the last reporting step.

Common misconception

“A calculator output with eight decimal places is eight-decimal-place chemistry.” The display is arithmetic precision, not measurement accuracy. Final digits must be supported by input measurements and a credible physical model, including reaction completion and sample identity.

Worked example

Mass of Mg is 0.243 g, M(Mg) = 24.3 g mol⁻¹ and M(H₂) = 2.016 g mol⁻¹. For Mg + 2HCl → MgCl₂ + H₂ with acid excess, n(Mg) = 0.0100 mol, n(H₂) = 0.0100 mol, and m(H₂) = 0.02016 g before rounding. Because the measured Mg mass and supplied Mg molar mass each have three significant figures, report theoretical H₂ mass as 0.0202 g under the simple significant-figure convention. A gas leak would affect an observed mass but not this ideal prediction.

Quick check

1. Does the exact 2:1 coefficient ratio in a reaction make a three-significant-figure mass measurement exact? Answer: No. The ratio is exact within the equation, but the measured mass still limits reported precision.

Exam focus

Write units through the chain and round once at the end. Do not treat equation coefficients as measured values. For a difference of two readings, retain the appropriate decimal place and recognize both readings contribute uncertainty.

Advanced insight

In formal uncertainty analysis, correlated inputs need covariance terms; for example, two masses measured with the same balance calibration can share systematic error. Relative standard uncertainty propagation is an approximation based on a local linear model. Significant-figure rules summarize precision, but a stated uncertainty interval gives richer information.

Summary

Stoichiometric ratios are exact mathematical links in a specified balanced model, while measured masses, volumes and concentrations carry uncertainty. Preserve guard digits during calculation, report a justified final precision, and identify experimental or model limitations that arithmetic rounding cannot repair.

Practice questions

1. How many significant figures are in 0.243 g? Answer: Three. 2. Is the coefficient 2 in Mg + 2HCl measured to one significant figure? Answer: No; it is an exact stoichiometric coefficient in the balanced model. 3. What delivered volume follows from burette readings 1.20 mL and 19.60 mL? Answer: 18.40 mL, before considering the instrument's uncertainty. 4. Why can precise gas-volume readings still give a poor product amount? Answer: Leaks, water-vapour assumptions or incomplete collection may dominate the numerical reading precision.