Application of cause-and-effect diagrams to the interpretation of UV-Vis spectroscopic data.

Abstract

The application of cause-and-effect diagrams to the evaluation of thermodynamic data from UV-Vis absorption spectroscopic analysis is demonstrated. The contributions of measurement uncertainty identified from a cause-and-effect diagram are implemented into a Monte Carlo procedure based on the threshold bootstrap computer-assisted target factor analysis (TB CAT). This algorithm aims at an improvement of data comparability and accounts for non-normality, spectral, residual and parameter correlation as well as random noise in target factor analysis. The ISO Type-B measurement uncertainties are included into the process by normally distributed random numbers with specified mean values and dispersions. The TB CAT procedure is illustrated by a flow diagram and a case study of Nd(III) complexation by picolinic acid N-oxide (pic NO) in aqueous solution. Using 12 experimental spectra as input data, the single component spectra and the formation constant 1g betaML of the Nd(pic NO)2+ species are obtained together with the respective probability density distributions. The role of the cause-and-effects approach on the further development of chemical thermodynamics is discussed.