Laboratory studies of atmospheric corrosion—I. Weight loss and electrochemical measurements

Electrochemical studies have been performed with the atmospheric corrosion monitor (ACM) under thin layers of electrolyte which were drying out at R.H. < 100%. Galvanic couples (Cu/steel, Cu/zinc) and one-metal (steel, zinc) ACMs were used. Measurements were carried out as a function of R.H. and Na2SO4 concentration. In addition, weight loss data were collected under identical conditions in thin layer experiments for steel and zinc in 0.01N solutions of NaCl, Na2SO4, HCl, H2SO4 and distilled H2O in air, air + 1 ppm SO2, argon and argon + 1 ppm SO2. The data obtained in air and air + SO2 were compared to weight loss results in bulk solutions. The electrochemical technique makes it possible to follow the changes of corrosion rates with time. As observed in outdoor exposure, a large increase of corrosion rates occurs when the electrolyte layers become very thin, shortly before the surface dries out. These findings explain the results of the weight loss data which show for most environments a much larger corrosion rate than in the bulk electrolyte. An accelerating effect of SO2 was observed for steel at higher R.H. values, while for zinc, no effect occurred in NaCl, Na2SO4 and H2SO4, but an inhibiting effect was measured in HCl and in distilled H2O. Since weight loss and electrochemical data were recorded under identical conditions, it is possible to determine how accurately the ACM data reflect the true corrosion rate. It was found for Cu/steel ACMs that the electrochemical data follow the same trends as the weight loss data, but account for only about 20% of the corrosion rate. Due to larger scatter in the weight loss data, a similar efficiency factor could not be determined for Cu/zinc. For steel and zinc ACMs, the true Tafel slopes are not known, which makes a calculation of corrosion rates doubtful. The low cell efficiency is considered to be due to local corrosion of single cell plates and to i.r.-drop effects. Despite the fact that exact corrosion rates cannot, at present, be obtained from ACM data, the technique appears very valuable for following the changes of atmospheric corrosion behaviour and for time-of-wetness measurements.