Use of an Electronic Tongue Based on All‐Solid‐State Potentiometric Sensors for the Quantitation of Alkaline Ions

Abstract

An array of eight nonspecific potentiometric sensors was used in combination with multivariate calibration for the simultaneous determination of NH þ 4 , K þ and Na þ ions. The sensors were of the all-solid-state type and employed a PVC polymer membrane. Signals were processed by using a multilayer artificial neural network (ANN). The ANN configuration used was optimized by using 8 neurons in the input layer, 5 in the hidden layer and 3 in the output layer. Use of the Bayesian Regularization algorithm allowed a quick building of an accurate model, as confirmed by random multi-starting of network weights. The system was used to analyze synthetic and river water, waste water and fertilizer samples. Correct results were obtained for the three ions in synthetic and real water samples; in fertilizers, ammonium ion can be determined, while sodium and potassium show biased results.

Dual‐Genic Hybridization Sensor Employing Electrochemical Impedance Spectroscopy

“…Other fixed characteristic of the net was the algorithm used to train the model, the Bayesian regularization. An advantage of this algorithm is that it does not require an internal validation subset of samples, used to prevent overfitting, as it happens with more classical algorithms, such as gradient descent [36,37].…”

“…Linear and sigmoidal transfer functions were tested for the hidden layer, purelin, tansig and logsig, whilst the functions for the output layer were purelin and tansig. Internal parameters used in the training process, based on our previous experience [37,38], were fixed a priori: the learning rate at a ¼ 0.1 and the momentum at b ¼ 0.4. The transfer function of the neurons at the input layer was linear.…”

State of the Art in the Field of Electronic and Bioelectronic Tongues – Towards the Analysis of Wines

“…However, the responses were influenced by the presence of other interferents occurring in the samples and provided increased results [82]. Another system utilized the combination of the developed potentiometric sensor array (electronic tongue for Na þ , K þ and NH 4 þ [83,84]) and enzyme urease covalently attached to carboxylated PVC [85]. The extended version with creatinine deiminase was applied for the analysis of urea and creatinine in urine without necessity to eliminate the alkaline interferences and compensation of their effects.…”

Electronic Tongues Employing Electrochemical Sensors

“…Conventional potentiometric PVC membranes can be used to form the sensor arrays if their formulation is properly considered. An example studied in our laboratories is the environmental determination of ammonium with nonactin-based ISEs, a case where the interference from other alkaline ions, mainly sodium and potassium can be expected -the approach was then to include the potential interferents in the array, and perform the multidetermination of the species of interest plus its interferents, compensating in each case the presence of its complementaries [28]. The number of sensors used was 8, and the processing tool selected was a backpropagation ANN, having as its outputs the different concentrations of the species sought [24].…”