Adaptation d'un algorithme de deuxième ordre pour l'analyse haute-résolution de courbes électrochimiques
|Advisor:||Grenier, Dominique; Miled, Amine|
|Abstract:||This master's thesis describes a new method for analyzing cyclic voltammetry curves for an efficient peak detection and automatic baseline substraction. This method uses two distinct algorithms for a precise characterization of Gaussian redox peaks which are correlated with molecules' concentration in a solution. First, significant improvements are made to an existing algorithm that uses iterative polynomial approximations to suppress the baseline automatically from the voltammetric curves. With these enhancements, the algorithm extracts redox peaks from cyclic voltammetry measurements automatically and allows a better representation of the variation of peak's amplitude according to concentration. In addition, the approximation errors are reduced compared to the initial algorithm. Then, the development of an algorithm for characterizing Gaussian peaks based on the MUSIC second-order algorithm is presented. This algorithm is adapted to characterize the number, position, width and amplitude of redox peaks with high accuracy. Finally, the performances of this algorithm are compared with those of other similar algorithms using simulated and experimental curves. The suggested algorithm leads to a better characterization of non-overlapping peaks as well as distorted peaks. It also reduces the frequency of false detections and allows the precise measurement of peaks' positions in noisy signals.|
|Document Type:||Mémoire de maîtrise|
|Open Access Date:||3 July 2018|
|Collection:||Thèses et mémoires|
All documents in CorpusUL are protected by Copyright Act of Canada.