This project will put pHabs into practice by establishing a reliable and universally applicable procedure for measuring the acidity of any substance. It will also develop a validated set of calibration standards for a wide variety of media. In striving towards a practical and accurate unified pH scale, this project will support and strengthen EU manufacturing and innovation in industries from pharmaceuticals to petrochemicals and semiconductors.

The guide will be helpful for improvement of understanding of consequences of measurement uncertainty and correct choice of methods for testing chemical composition of multicomponent materials and objects. The opportunity to work with CITAC and prepare a joint guide (as an IUPAC TR), will give the widest possible exposure to our work and justifies, in our opinion, the requested extension.

Risks were estimated for cases where the concentration of components in the analysed sample are independent or correlated. Calculations were performed in R-ambient and using a user-friendly MS-Excel spreadsheets to allow the easy application of the developed tools.

The definition of statistically sound criteria for relative retention times, RRT, and ratios of the abundance of ions, RA, of the mass spectrum of the analyte must be able to assess the not-normal distribution of these ratios. A user-friendly MS-Excel spreadsheet, based on Monte Carlo simulations of ratios of highly correlated variables, was developed to define criteria for the variation of RRT and RA. The selected confidence level of the confidence intervals defines the true positive results rate, TP. Another MS-Excel file was developed to simulate the signal noise of the analytical instrument and quantify the probability of the noise producing false positive results, i.e. the false positive results rate, FP. The identification/examination uncertainty was estimated by the Likelihood ratio, LR (LR=TP/FP), of the Bayes Theorem that estimates how more likely a positive result is true than false. Models of the variation of the LR with the level of analyte were developed to estimate the Limit of Examination; the lowest analyte level that can be identified in a conclusive way – i.e. with a LR not smaller than 10⁶.

These tools were successfully applied to the identification of trace levels of pesticide residues in foodstuffs.

This project involved defining strategies for the selection of representative matrices and analytes to simplify the validation of analytical methods and the quality control of analysis performed using multi-matrix and multi-analyte methods of analysis. This selection was based on the equivalence of the multivariate pattern of nutritional parameters of foodstuffs or physicochemical properties of analytes.

It was also developed tools to estimate and correct relevant analyte losses in the analytical method to allow the comparability of measurements produced from sample treatments with different efficiency; i.e. associated with different analyte recovery.

This innovative project aimed at evaluating the uncertainty of routine analysis results obtained at Controlab, Lda, to report that information to clients and use it in tests quality control.