Methods Development: Choosing Calibration Standards

The concentrations of the calibration standards should span the range of expected analyte concentrations in the samples. A wide range of concentrations gives the best precision, but there is one caveat: the common assumption of homogeneous noise may be more justifiable over a more narrow range of concentrations. Ideally, the calibration standards are made so that the typical analyte concentration in the sample falls near the mean concentration of the standards. If time allows, enough standards should be made to verify linearity of analyte response. Nonlinear calibration curves always need more calibration standards for regression model selection, and to achieve desirable precision.

Calibration standards are usually made by diluting a stock analyte solution in volumetric flasks. The calculations are fairly easy once you've done them awhile, but they may take some getting used to at first.

Example

You are to analyze samples for their copper content based on flame atomic absorption spectroscopy (FAAS). You have a stock solution of 850.0 ppm copper (certified by NIST). Describe how you would prepare standards for instrument calibration. The dynamic range of the flame AA instrument is between 5 ppm and 50 ppm Cu.

Answer

In the above example, the linear dynamic range only covers one order of magnitude, so that it is appropriate for the calibration standards to span the entire range. If necessary, samples must be diluted to fall within the calibration range.