This communication shows a novel strategy in the field of potentiometric sensors, applied to the determination of nitrate in the presence of chloride interferent. The determination is performed employing the flow-injection analysis technique with four potentiometric sensors featuring cross-term response. The signal processing with a multivariate data treatment, in this case an artificial neural network based on the Bayesian regularization, lets us quantify the concentration of nitrate ion between 0.1 and 100mgl -1NO3- without the need to eliminate chloride interferent. Results obtained with this approach are compared with the direct determination of nitrate using its ion-selective electrode, showing how the new strategy attains a better correlation of obtained versus expected values, especially at the lower concentration levels. The comparison line between these pairs yields an intercept of 0.0 ± 0.3mgl-1NO3- and a slope of 1.01 ± 0.02. © 2004 Elsevier B.V. All rights reserved.
- Electronic tongue
- Ion-selective electrode array
- Neural networks