Certification of multiparameter physicochemical reference materials. Part 1: Homogeneity study

Abstract

Certified reference materials (CRMs) are indispensable tools in calibration, validation, and quality control across various fields, including chemical analysis, environmental monitoring, and industrial production. Their proper certification ensures the metrological traceability and reliability of the measurement results in laboratories around the world. This study presents a new application of an established statistical test (equality of coefficients of variation) to support the homogeneity assessment of multiparameter physicochemical CRMs. The method enables statistically justified indirect inferences about the homogeneity of untested properties based on the behaviour of tested ones, supporting more efficient and statistically defensible certification processes. The approach is demonstrated using CRMs that reproduce the density, refractive index, surface tension, and kinematic viscosity. The procedure is consistent with ISO 17034 and addresses the gaps identified in ISO 33405 regarding the assessment of homogeneity in multiparameter materials. By extending homogeneity assessment beyond conventional univariate methods, this work contributes a practical and statistically reliable tool for CRM producers developing complex, multiparameter materials.