Abstract
Corrosion rates were determined by curve-fitting polarization data using nonlinear regression analysis. Continuous corrosion rate monitoring was implemented thru the use of repetitive scans and real time calculations. Polarization data was fit to the more appropriate of two electrochemical models of the corrosion process: an activation equation when both the anodic and cathodic reactions are under activation control; and an activation/mass-transport equation when the anodic reaction is under activation control and the cathodic reaction is mass-transport limited. Generally, the activation equation offers a better fit to polarization data in flowing water, while the activation/mass-transport equation is more appropriate in stagnant water and/or when a significant scale forms. For laboratory testing in ASTM substitute ocean water, corrosion rates determined by curve-fitting polarization data were compared to those determined by the traditional linear polarization resistance (LPR) method, by an electrical resistance probe, and by weight loss. Various excitation waveforms were galvanostatically applied. Results demonstrated that the excitation waveform has a significant influence on the calculated corrosion rate. Using the appropriate waveform, curve-fitting polarization data can give better agreement with weight loss corrosion rate than the traditional LPR method. However, corrosion rates determined by curve-fitting are more sensitive to errors in the electrochemical data than corrosion rates determined by the LPR method.
