Preparation steps in environmental trace element analysis - facts and traps

The laboratory of CERVA is for several decades involved in the Belgian envi ronmental research. The activity was associated to national monitoring prog rams dealing with trace metal pollution of all compartments of the environm ent (sea and river waters, sediments and organisms but also soils, plants, animal and food samples). Such a monitoring dealing with the total analyte contents in samples needed a comprehensive development of the whole methodo logy associated to the analyses using atomic absorption and emission spectr oscopy techniques. This includes measurement but also preparation steps. Th e latter is the subject of this work. Long-term experience has shown that p recisely sample preparation is the most critical part of the analysis becau se it is responsible for the largest and often hidden sources of errors. Er rors due to contaminations may be usually overcome if necessary precautions are taken concerning reagents, tools and the manner of working. The proble m is different for analyte losses: in this case, the responsible factor is an inappropriate methodology. This is particularly true for preparation of solid samples that have to be brought in a solution in order to satisfy nee ds of introduction systems of most spectroscopic techniques utilized in rou tine laboratories. For some types of samples (e.g. animal tissues), the dis solution is not a problem: it may be readily achieved by several procedures . This is not the case for samples that contain silicates in their matrix ( e.g. soils, sediments, plants) because their complete dissolution cannot be ensured by a simple procedure. This review describes the present knowledge regarding possibilities and errors that concern preparation steps. Tn addi tion, possible effects of the preparation procedure on the quality of measu rement are also systematically discussed. (C) 2001 Elsevier Science B.V. Al l rights reserved.