Evaluation of the analytical measurement uncertainty by two methods in food vitamin A and E determination

Dias, M. Graça; Vasco, Elsa; Santos, Mariana

http://hdl.handle.net/10400.18/493

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
201109_MGD_9IFDC_vitAE_Pr.pdf		118.31 KB	Adobe PDF	Download

Send Feedback

Authors

Dias, M. Graça

Vasco, Elsa

Santos, Mariana

Abstract(s)

Rationale. ISO 17025 prescribes the estimation of the measurement uncertainty and specifies requirements concerning this estimation and how it should be reported. Besides,for a consistent interpretation of an analytical method result it is necessary to evaluate the confidence that can be placed in it. This can be provided by the quantification of its accuracy (trueness and precision) in the form of a measurement uncertainty estimate. Objective. To compare two approaches to estimate the analytical measurement uncertainty in the HPLC analysis of vitamins A (all-trans-retinol and 13-cis-retinol) and E ( -tocopherol) in infant milk powder. Materials and methods. The normal phase HPLC analytical method was based on EN 12822 e 12823-1. Identification was carried out based on retention time and quantitative determination by external standard method. Analytical measurement uncertainty was laboratory estimated based on method in-house validation data (evaluation of calibration curve parameters, precision, reference material analysis, detection and quantification limits) and laboratory performance results in proficiency tests (BIPEA and FAPAS), for infant milk powder. Results. Combined relative standard uncertainty obtained from laboratory validation data was 0.22 for vitamin A infant milk powder content of 100-500 μg/100 g, and 0.080 for vitamin E infant milk powder content of 15 mg/100 g. Estimation based on proficiency tests was 0.17 and 0.14, respectively for vitamin A and E. Conclusions. Results from the analytical measurement uncertainty estimation by the two methods were in good agreement for vitamin A although for vitamin E the result obtained from the intraboratory data was lower than from the interlaboratory data. According to the obtained results to give a correct impression of accuracy with which these constituent levels can be known in samples similar to the analysed, results should present at maximum two significant figures.