Cut edge corrosion mechanisms in organically coated zinc–aluminium alloy galvanised steels

Abstract

Abstract Organically coated galvanised strip steel materials are increasingly important as construction materials. The effects of magnesium additions to the zinc spelter on the microstructure and cut edge corrosion mechanism of zinc aluminium alloy galvanising coatings on strip steels have been found to be significant. Galvanised specimens were prepared on 0.7 mm steel substrates using a hot dip bath composition of near eutectic 4.2 wt-% aluminium-ca. 95 8 wt-% zinc and trace levels of magnesium. Increasing the magnesium content in the metallic coating from 0.01 to 0.05 wt-% leads to the formation of significant areas of subsurface, dendritic, pro-eutectic zinc rich phases. The metallic coating substrates were over coated with asymmetric thicknesses of a 200 μm PVC based coating on one side and a 15 μm polyester coating on the other producing a range of model organically coated substrates identical in every way except for the magnesium level in the metallic coating. The kinetics and mechanism of cut edge corrosion in 5 vol.-%NaCl of these model systems have been investigated using the scanning vibrating electrode technique (SVET). The SVET data have shown that the increasing coating heterogeneity with increased magnesium content leads to increasing numbers of active and intense anodes on the exposed surface of the metallic coating at the cut edge. At magnesium levels of < 0.03 wt-%, asymmetry in the thickness of the organic coating induced localisation of anodic activity near to the thicker (PVC) organic layer, and of cathodic activity localised primarily on the steel/galvanising layer adjacent to the thinner (polyester) organic coating. However, as the magnesium content of the metallic coating was increased (to 0.04 and 0.05 wt-%) the location of anodic and cathodic activity was found to become independent of the geometry of the organic coating. Whilst the number of active anodes on the 20 mm exposed edge was found to be identical for both of the latter magnesium levels those in the 0 05 wt-%Mg specimen were more persistent. In the case of the 0 04 wt-%Mg specimen, 13 of the 21 active anodes deactivated over the first 12 h of immersion whereas with the highest magnesium level deactivation was only observed for 5 of the 21 anodes. Those that remained active in the latter specimen also had significantly greater levels of anodic activity. This appears to be related to interlinking between particles of the pro-eutectic zinc rich phase in specimen of the highest magnesium level, leading to crevice corrosion. The consequence of the microstructural changes induced by trace Mg additions in that the SVET measured total zinc losses over the 24 h exposure period increasing from 28 μg (with 0.01 wt-%Mg), to 44 μg (0.02 wt-%Mg), 61 μg (0.03 wt-%Mg), 67 μg (0.04 wt-%Mg) and 223 μg (0.05 wt-%Mg).