Title: Influence of the approach to calibration on the accuracy and the traceability of certified values in certified reference materials
Authors: ZELENY ReinhardSCHIMMEL Heinz
Citation: TRAC-TRENDS IN ANALYTICAL CHEMISTRY vol. 33 p. 107-116
Publication Year: 2012
JRC N°: JRC67363
ISSN: 0165-9936
URI: http://publications.jrc.ec.europa.eu/repository/handle/JRC67363
DOI: 10.1016/j.trac.2011.10.009
Type: Articles in periodicals and books
Abstract: The use of calibrants properly characterized with respect to identity and purity is an important requirement to achieve accurate, reliable results in chemical measurements. In particular, in generating data that contribute to assigning a property value to a certified reference material (CRM), compromises often made for routine measurements (non-verification of calibrant purity) are unacceptable, and strict adherence to calibrant quality requirements is indispensable. We compared two approaches to calibration within the framework of laboratory intercomparisons for assigning values to two recently released CRMs in the veterinary drug-residue field, namely chloramphenicol in pork (ERM-BB130) and oxytetracycline in milk (ERM-BB492). First, all laboratories used the same calibrant (common calibrant, a commercially available substance, extensively characterized for purity and identity), and second, each laboratory used its in-house calibrant (commercially available substance, purity as indicated by the provider). The laboratories, mostly accredited to ISO/IEC 17025, used validated methods based on gas or liquid chromatography coupled to mass spectrometry or ultraviolet spectrophotometry. Each measurement series included two calibration curves, one with the common calibrant, and one with the in-house calibrant. For both materials, when the common calibrant was used, the mean of accepted laboratory mean values was 7% and 4% lower for BB130 and BB492, respectively. This shows the impact on accuracy and hints at an overestimation of the purity of the in-house calibrants. Moreover, when the common calibrant was used, the variation among the laboratory mean values in the intercomparisons decreased by 9% (BB130) and 39% (BB492), respectively. The smaller uncertainty from the characterization study resulted in a lower expanded uncertainty of the certified values (4% lower for BB130, and 7% lower for BB492, compared to using in-house calibrants). The use of a common calibrant thus significantly affects and improves the accuracy and provides the basis for a proper metrological traceability statement for the certified values in the CRMs.
JRC Directorate:Health, Consumers and Reference Materials

Files in This Item:
There are no files associated with this item.

Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.