An electroanalytical method to detect adulteration of ethanol fuel by using multivariate analysis

This work proposes a simple and inexpensive electroanalytical method to identify adulteration in hydrated ethyl alcohol fuel (HEAF) samples by using multivariate analysis. The methodology was validated in a case study with HEAF samples divided into two categories: (1) unadulterated and (2) adulterated with methanol in different concentration levels (1.0–13% m m−1). Staircase voltammetry was the electrochemical technique employed. Voltammograms were recorded with a copper electrode in an alkaline solution. Four different pattern recognition methods were evaluated for classification of the resulting voltammograms: partial least squares discriminant analysis (PLS-DA); linear discriminant analysis (LDA) with variable selection by the successive projections algorithm (LDA/SPA), genetic algorithm (LDA/SPA), and a stepwise formulation (LDA/SW). The best results were achieved with LDA/GA and LDA/SW models, which all prediction samples were correctly classified.