Using electrochemical noise to detect corrosion: Evaluation of a round-robin experiment

Successful measurement of electrochemical noise (EN) signals requires that a number of preconditions be fulfilled. In particular, fundamental understa nding of the dynamics of the corrosion system concerned must be combined wi th an appreciation of the way in which the measurements are made. Broader a cceptance of the technique requires a joint effort by the various corrosion investigators and laboratories concerned. This, in turn, requires that a c ommon basis for discussion be established With this in mind, a round-robin experiment was organized on EN measurements. The participation of 17 organi zations (universities, commercial firms, and research and development labor atories) from five countries was obtained The experiment involved the inves tigation of EN signals from aluminum and stainless steel (SS) exposed under conditions in which pitting corrosion can occur. Data provided by the part icipants were analyzed according to various criteria Evaluation methods inc luded visual inspection of noise in the curves of potential or current vs t ime, calculation of the standard deviation, counting of transients, and tra nsformation of measurements into the frequency domain. Almost all of the pa rticipants were able to detect differences in EN behavior during different stages of the experiment qualitatively. However quantitative evaluation rev ealed a more complex situation. At times, differences in measured EN signal intensity of 2 orders of magnitude were obtained as a result of the differ ent apparatus used (sensitivity, frequency range, etc.). Nonetheless, the f act that more than half of the participants reported data within a relative ly narrow scatterband was a positive result. Additional analysis showed uni form, numerical treatment of the measured data was of decisive Importance t o obtain comparable results. It was clear that the complex interactions bet ween the way in which EN data were recorded (sampling rate, filtering, onli ne data manipulation, and base-line noise level) and the subsequent data ev aluation should receive increased attention. Critical evaluation of results from the round-robin experiment showed comparable EN measurements can be m ade if suitable equipment is used and correct procedures are followed. The experience gained provided a sound basis for further development of EN as a tool for fundamental understanding of various aspects of corrosion and for the solution of practical problems of corrosion protection, particularly i n the area of corrosion monitoring.