Comment on "Laboratory-Scale Identification of Corrosion Mechanisms by a Novel Pattern Recognition System Based on Electrochemical Noise Measurements" [J. Electrochem. Soc., 166, C284 (2019)]

Papers dealing with electrochemical noise measurements (ENM) published in scientific journals often propose advanced signal processing techniques, in particular to determine the corrosion mechanisms. Most of the time, this is done without first validating their measurements. However, validation is not possible by simply visualizing the time records; rather, it is necessary to measure the EN at different sampling frequencies and check the overlap of the power spectral densities (PSDs) in the common frequency ranges. Such validation is mandatory because many commercial potentiostats do not have anti-aliasing filters before the analog-to-digital converters in the measurement channels. The aliasing of high-frequency signal components to frequencies below half of the sampling frequency corrupts the EN signal in both time and frequency domains. This explains the dramatic scatter of the PSDs of the thermal noise generated by dummy cells, as shown in the round-robin tests organized by the European Cooperative Group on Corrosion Monitoring of Nuclear Materials (ECG-COMON).