Concerning trend removal in electrochemical noise measurements

A theoretical and experimental analysis of the effects of trends in the pot ential and current noise fluctuations observed during measurement of electr ochemical noise (ECN) data on the slopes of power spectral density and spec tral noise plots as well as Various other parameters obtained by statistica l analysis of the ECN data in the time domain has been carried out. Drifts in the ECN data can affect the calculated values of the standard deviations of the potential (sigma (v)) and current (sigma (1)) fluctuations and the noise resistance R-n. Correction formulae developed based on previous liter ature data have been used to correct the experimental data. A linear method has been applied to remove the trend from the experimental data for the mi ld steel/NaCl system and good agreement has been obtained between sigma (v) , sigma (1) and R-n calculated from the data after trend removal and from t he correction formulae. The localization index (LI) was 1.00 after trend re moval in all cases demonstrating that LI is not related to a particular cor rosion mechanism. A good agreement has been obtained between the spectral n oise plots after trend removal and the impedance spectra determined at the same exposure time. A moving average removal method discussed in the litera ture was found to produce erroneous data. (C) 2001 Published by Elsevier Sc ience Ltd.