A voltammetric method for the simultaneous determination of oxidizable amino acids (cysteine, tyrosine and tryptophan) is proposed. The method is based on the analysis of multivariate data, which consists of cyclic voltammograms registered in the range from 400 to 1100 mV with a graphite- methacrylate composite electrode, by using multivariate calibration methods. Cyclic voltammograms of pure standards of these oxidizable amino acids show characteristic shapes. Thus, the resolution of mixtures of them by multivariate calibration lies in the differences in their voltammograms. The main difficulties encountered in these analyses are the high degree of overlapping and the lack of bilinearity of voltammetric data. The prediction abilities of principal component regression (PCR) and partial least squares regression (PLS) were assayed to the analysis of oxidizable amino acids in a series of synthetic samples. The quantification of cysteine and tryptophan was especially satisfactory since they are the most characteristic amino acids.
- Cyclic voltammetry
- Graphite-methacrylate composite electrode
- Multivariate calibration
- Oxidizable amino acids
- Partial least squares regression
- Principal component regression