Contamination as a limiting parameter in trace analysis.

The relative importance of the main parameters determining systematic errors due to contamination in trace analysis is evaluated, with respect to the laboratory environment, tools, containers and reagents. Variability in the composition and quality of materials as well as of the relevant analytical data makes the treatment of the subject only semi-quantitative but the evidence suggests that control over contamination primarily depends on good laboratory practice. Among the appproaches applicable to trace concentrations, activation analysis gives results that are least affected by contamination. In other techniques, varying blanks and the need for their control make accurate analysis at the trace level demanding in terms of time and engagement; reliable data in routine determinations at levels much below 1 mg/kg can only be expected from specialized and experienced laboratories.