## Standard Additions or Spiking

Ideally, the standards should be made up in a solution containing the same normally expected levels of matrix elements as occur in the sample solution. It should be borne in mind that even if they exert no chemical interference, they could possibly exert a viscosity effect on a nebulized solution (especially with high concentrations of phosphoric or sulphuric acids). If it is not possible to determine the matrix components or prepare standards in a matrix solution, and unless experiments have shown matrix interference to be insignificant, then the method of standard additions, or 'spiking', should be carried out. This is where known amounts of the analyte are added to the sample or sample solution before determination by, e.g. AAS or colorimetry.

There are several methods of carrying out the standard addition method. They all rely, however, on the standard curve being linear over the range of analyte concentrations in the spiked solutions. The addition of an incremental range of standards, rather than just one, will improve precision. One method (Thomas, 1996) involves pipetting a fixed volume (Vx) of sample solution into a series of volumetric flasks, next adding a range (say, 0, 5, 10, 15 and 20 ml) of a standard analyte solution (concentration Cs) to the respective flasks, followed by the reagents, then making up to the mark. The absorbance (y-axis) is plotted against volume of standard added. If a is the intercept on the y-axis, and p is the gradient, then the concentration of the analyte in the unknown solution (Cx) is given by:

Two other approaches are given by Meier and Zund (1993). The observed signal is plotted against the amount of analyte spiked (e.g. mg ml-1 in the directly analysed solution) into the test sample (two or more points). The regression line is extrapolated to y = 0, and the unknown concentration is given by -1 x intercept on the x-axis (which is a negative value). In certain circumstances, e.g. where interference by a matrix component is present, the line is extrapolated to a level where y > 0. The other approach is to use an interpolation rather than an extrapolation, which improves precision with no additional tests. The method is to subtract the reading of the unspiked sample from each of the spiked sample readings. The difference is plotted as the best straight line passing through the origin (x = 0; y = 0). The concentration value corresponding to the reading for the unspiked sample is read from the standard addition line.

## Post a comment