Model‐based description of indicator displacement assay sensor arrays for quantitation of mixtures

Abstract

AbstractThis paper propounds an approach for the simultaneous analysis of colorless analytes using optical sensor arrays. On the basis of the equilibrium law, the principles of indicator displacement assay (IDA)–based sensor arrays that can be utilized for the efficient quantification of analytes in the mixture solutions through multivariate hard modeling approach are discussed in detail. According to these principles, two different sensor array systems are designed for the simultaneous analysis of the analytes with nonselective signal in the mixtures. Each sensor element of the designed sensor arrays consists of a single indicator and receptor but has different equilibrium environmental conditions.Among all amino acids, only histidine and cysteine reveal significant responses to the Ni‐murexide IDA probe. So, in the first experiment, the Ni‐murexide IDA sensor arrays are applied for the precise and selective quantification of these two amino acids in the mixture solutions. The root mean square error of predictions (RMSEPs) of the prediction dataset obtained by the radial basis function–artificial neural network (RBF‐ANN) model are 3.41% and 4.72% for histidine and cysteine, respectively, in the unknown artificial mixtures. In the second experiment, two metal cations, namely, zinc and cadmium, are simultaneously quantified by another designed IDA‐based sensor array. The metallochromic indicator 4‐(2‐pyridylazo) resorcinol (PAR) is applied in two different equilibrium conditions for each sensor element. The results of predicting these two metals in the unknown mixtures via the RBF‐ANN model are 7.22% and 8.15% for zinc and cadmium, respectively.