DETERMINATION OF PH GRADIENTS AND ACIDITY CONSTANTS IN FLOW-INJECTION ANALYSIS SYSTEMS BY EVOLVING FACTOR-ANALYSIS

A general multivariate method for determining pH gradients in flow-injection analysis (FIA) systems was developed. An FIA system was constructed in which an injected sample is merged with a non-linear pH gradient and the method was demonstrated by the injection of acid-base indicators with pK(a) values spanning the range 3.4 - 8.9. The indicator responses were monitored by a photodiode-array spectrophotometer and by application of the evolving factor analysis multivariate peak resolving technique it was possible to determine the concentration profiles of the acidic and basic forms of the indicators. With this knowledge, the pH gradient can be determined. It is further shown that the pH gradient can be exploited to predict pK(a) values of unknown indicators. The prediction errors obtained are comparable to literature results. As an alternative, a multivariate calibration model (partial least-squares regression) was tested for empirically predicting the pK(a) values by using the EFA concentration profiles as variables. The prediction errors obtained by this multivariate approach were also similar to literature results.