Skip to content
Omnikalc

Calibration curve concentration calculator: concentration of an unknown and its confidence interval

Enter the calibration line, the signal of the unknown and the regression statistics to get the concentration, its uncertainty and a confidence interval.

Updated on
Your inputs

Value out of range

Value out of range

Value out of range

Value out of range

Value out of range

Value out of range

Value out of range

Value out of range

%

Value out of range

Results

Concentration of the unknown sample

4.0000

Standard uncertainty of the concentration
0.0506
Critical t value
2.7764
Lower bound of the confidence interval
3.859
Upper bound of the confidence interval
4.141

A calibration line turns a signal into a concentration, but the answer has an uncertainty. The calibration curve concentration calculator takes the line, the signal of the unknown and the regression statistics, and gives the concentration, its standard uncertainty, the critical t value and the confidence interval.

How the calculation works

  • From your calibration regression, enter the slope m, the intercept b, the number of standards n, the residual standard deviation s y/x, the mean signal of the standards and Sxx, the sum of squared deviations of the standard concentrations.
  • Enter the mean signal y0 of the unknown and the number of replicate readings k. The concentration is x0 = (y0 − b) ÷ m.
  • Its standard uncertainty is s(x0) = (s y/x ÷ m) × √(1 ÷ k + 1 ÷ n + (y0 − ȳ)² ÷ (m² × Sxx)).
  • The confidence interval is x0 ± t × s(x0), with t taken from the Student distribution with n − 2 degrees of freedom.

Example

A calibration line has m = 0.085, b = 0.010, n = 6 standards, s y/x = 0.006, ȳ = 0.435 and Sxx = 70. An unknown read three times averages 0.350. The concentration is (0.350 − 0.010) ÷ 0.085 = 4.0, and its standard uncertainty is 0.0506. With t = 2.776 for 4 degrees of freedom, the 95 % interval is 3.859 to 4.141.

Why it matters

Analytical chemists, environmental labs and food testers must report a concentration with its uncertainty. This is the standard inverse-prediction formula for a straight-line calibration, and it avoids the common mistake of quoting a bare number.

Good to know

The formula assumes a straight line, equal variance across the range and standards measured without error. If the line is curved or the variance grows with concentration, use a weighted or nonlinear fit. The unknown should fall inside the range of the standards.

The linear regression calculator produces the slope, the intercept and the residual spread, and the Beer-Lambert calculator links absorbance to concentration.

Frequently asked questions

How do I find the concentration of an unknown from a calibration curve?

Measure standards of known concentration, fit a straight line of signal against concentration, then read the unknown by inverting the line. The concentration is the measured signal minus the intercept, divided by the slope.

Why does the uncertainty depend on the mean of the standards?

A calibration line is most precise near the center of the data. The farther the unknown's signal is from the mean signal of the standards, the larger the uncertainty, which is why the formula includes the squared distance from the mean.

How can I reduce the uncertainty?

Read the unknown several times, use more standards, spread the standards over a wider range and keep the unknown near the middle of the curve. Improving the fit by reducing the residual standard deviation also helps.

Where do Sxx and the residual standard deviation come from?

Most regression tools report the residual standard error of the fit. Sxx is the sum of the squared differences between each standard concentration and the mean concentration, or equivalently the variance of the concentrations times n − 1.

Was this calculator helpful?

Thanks for your feedback!

Results are estimates provided for information purposes. They do not replace professional advice.