Pooled calibrations and retainment of outliers improved chemical analysis

Jens E.T. Andersen, Haedar S.H. Alfaloje

Research output: Contribution to journalArticlepeer-review

7 Citations (Scopus)

Abstract

Analytical chemistry has a large responsibility in society, and credibility and reliability are important concepts associated with chemical analysis. Metrology and Quality Assurance (QA) are key areas of interest in contemporary research. Quality in measurements is illustrated by a series of experiments with several analytical technologies comprising of ICP-MS, GC-MS and AAS. The scientific methodology relies on the concept of reproducibility that depends on type of analyte and type of apparatus. By applying the principle of pooled calibrations it is shown that the performance of the apparatus in terms of levels of uncertainty can be tested in a single laboratory. The uncertainties are compared to predictions of the Horwitz formula. It is suggested that this method is universally applicable not only to the actual technologies but also to other technologies in other fields of science. The results indicate that the procedures outlined in the Eurachem/CITAC Guide are of tremendous value to analytical sciences because they direct researcher's attention towards the concept of consensus values rather than towards true values. Introduction of certified reference materials (CRM's) in metrology has provided much new information on working habits in professional laboratories and CRM's may be applied to establish the true level of uncertainty for a given type of analytical method. Finally, it is proposed to devise a new procedure of method validation that facilitates QA in general, thus saving many resources at laboratories.

Original languageEnglish
Pages (from-to)691-703
Number of pages13
JournalRevue Roumaine de Chimie
Volume58
Issue number7-8
Publication statusPublished - Jan 1 2013

All Science Journal Classification (ASJC) codes

  • Chemistry(all)

Fingerprint Dive into the research topics of 'Pooled calibrations and retainment of outliers improved chemical analysis'. Together they form a unique fingerprint.

Cite this