THE USE OF STANDARD REFERENCE MATERIALS FOR QUALITY ASSURANCE IN INDUCTIVELY COUPLED PLASMA OPTICAL-EMISSION AND ATOMIC-ABSORPTION SPECTROMETRY

To ensure that a chemical measurement process yields accurate results, well characterized control materials such as Standard Reference Materials (SRMs) should be analysed. For inductively coupled plasma optical emission spectrometry (ICP-OES) and atomic absorption spectrometry (AAS) methods, the SRM should be similar in chemical matrix to the unknown test material of interest so that the SRM provides similar sources of potential systematic error. Such sources of error may include incomplete dissolution, loss of volatile analyte, and chemical matrix interferences in the plasma, flame or electrothermal atomizer. In addition, the SRM should be chosen so that the analyte levels match the corresponding estimated levels in the unknown and so that the uncertainties in the certified values are small enough to provide a useful test for accuracy. The measurement process itself must be designed so that sources of random error can be evaluated. Each potential source of random error should be replicated to provide for a test of its significance. The resulting data will yield estimates for these components of variance, and the overall standard error can then be calculated. Examples of experimental designs and typical results from an ICP analysis of an SRM are described. Procedures using commonly available software for evaluating components of variance and calculating the mean, standard error, and the effective number of degrees of complex variance component designs are also presented. The results of such an approach are then applied to ascertaining whether or not bias in the analytical results is detected.