Estimate of the uncertainty in measurement for the determination of mercury in seafood by TDA AAS

• Published in: Journal of environmental science and health. Part B, Pesticides, food contaminants, and agricultural wastes Vol. 50; no. 8; pp. 622 - 631
• Main Authors: Torres, Daiane Placido, Olivares, Igor R.B., Queiroz, Helena Müller
• Format: Journal Article
• Language: English
• Published: England Taylor & Francis 03-08-2015; Taylor & Francis Ltd
• Subjects: accreditation; Animals; Brazil; Calibration; Dispersion; Food Analysis - methods; Food Analysis - statistics & numerical data; Food Contamination - analysis; Food Contamination - statistics & numerical data; Food quality; Least-Squares Analysis; Measurement; Mercury; Mercury - analysis; Reproducibility of Results; Scientific imaging; Seafood; Seafood - analysis; Spectrophotometry, Atomic - methods; Uncertainty; Uncertainty in measurement; Brazil; ASW, Brazil; seafood; accreditation; Uncertainty in measurement; mercury
• ISSN: 0360-1234; 1532-4109
• DOI: 10.1080/03601234.2015.1028861

An approach for the estimate of the uncertainty in measurement considering the individual sources related to the different steps of the method under evaluation as well as the uncertainties estimated from the validation data for the determination of mercury in seafood by using thermal decomposition/amalgamation atomic absorption spectrometry (TDA AAS) is proposed. The considered method has been fully optimized and validated in an official laboratory of the Ministry of Agriculture, Livestock and Food Supply of Brazil, in order to comply with national and international food regulations and quality assurance. The referred method has been accredited under the ISO/IEC 17025 norm since 2010. The approach of the present work in order to reach the aim of estimating of the uncertainty in measurement was based on six sources of uncertainty for mercury determination in seafood by TDA AAS, following the validation process, which were: Linear least square regression, Repeatability, Intermediate precision, Correction factor of the analytical curve, Sample mass, and Standard reference solution. Those that most influenced the uncertainty in measurement were sample weight, repeatability, intermediate precision and calibration curve. The obtained result for the estimate of uncertainty in measurement in the present work reached a value of 13.39%, which complies with the European Regulation EC 836/2011. This figure represents a very realistic estimate of the routine conditions, since it fairly encompasses the dispersion obtained from the value attributed to the sample and the value measured by the laboratory analysts. From this outcome, it is possible to infer that the validation data (based on calibration curve, recovery and precision), together with the variation on sample mass, can offer a proper estimate of uncertainty in measurement.