The effect of magnesium complex ions on the corrosion behavior of magnesium

Magnesium, a very useful structural metal, must be used with corrosion inhibitors in most circumstances because of its reactive nature. New inhibitors are needed to meet anticipated environmental regulations. A relatively quick and efficient method for determining the merits of potential corrosion inhibitors is required. Assessment of effects of inhibitors on the electrochemical kinetics of the corrosion reaction is also of importance. About 95 ligands, possible candidates for testing by virtue of the high stability constants of their magnesium complexes (K $\geq$ 10$\sp5$), are listed. From those ligands, fluoride, metavanadate, EDTA, DTPA, and Eriochrome Black T were chosen for testing. For comparison, dichromate, iodate, and molybdate anions were also studied. A one-hour immersion test in a 0.1 M NaCl solution adjusted to a pH of 4.00 was developed to rapidly assess the inhibitive effects of a range of concentrations for each compound. Percentage inhibition was determined from weight loss. An initial corrosion rate was determined by analyzing the solution concentration of magnesium as a function of time during the first 13 minutes of immersion. Corrosion rates determined by the two methods were compared. Weight-loss data for the non-chelating compounds, Na$\sb2$MoO$\sb4$ and glycine, paralleled the initial concentration data, although the actual corrosion rates differed. A rotating disk electrode study was made of magnesium in 0.1 M NaCl solutions with several of the ligands. Open-circuit potential was plotted versus the logarithm of corrosion current density (determined from the rotation speed via the Levitch equation). Corrosion rates determined by the rotating disk electrode study were approximately two orders of magnitude lower than the values determined from weight-loss and about 20 times less than values determined from magnesium solution-concentration data. Possible reasons for these discrepancies include: dissimilar Fe:Mn ratios in the three batches of magnesium; large differences between geometric and actual surface areas for weight-loss samples; and the presence of metallic magnesium in corrosion products of the weight-loss test.