The corrosion behaviour of various Zn-Mg-Al hot-dip galvanised coatings on steel substrates were studied using a combination of time-lapse photography (TLP), the scanning Kelvin probe (SKP) and non-contact height profiling. In a standard experiment 1uL electrolytic droplets comprising varying concentrations of aqueous NaCl were deposited onto the surface of the Zn-Mg-Al alloy coating held in a humid air atmosphere maintained in isopiestic equilibrium (equilibrium with respect to the vapour pressure of water) with a large remote reservoir of aqueous electrolyte which had the same composition as the original droplet. The progress of atmospheric corrosion was then followed using TLP with photographs taken at 2-minute intervals for a period of up to 60 hours. When required, in-situ SKP was used to produce area maps of electrochemical potential if in the vicinity of the electrolytic droplets over a similar period of time. The SKP probe could also be used as a non-contact height profilometer which allowed the topography of the electrolytic droplet and the surrounding Zn-Mg-Al alloy surface to be mapped. This made it possible to study time-dependent changes in droplet volume and shape as atmospheric corrosion proceeds.

In most cases the phenomenon known as cathodic secondary spreading was observed at the droplet periphery, where a thin film (microns) of electrolyte develops and advances away from the droplet as time proceeds. The presence of this thin electrolyte film changes the reflectivity of the Zn-Mg-Al alloy surface with respect to visible light and also produces a marked reduction (drop) in local electrochemical potential. The kinetics of secondary spreading could therefore be followed using a combination of TLP and in-situ SKP. Results are presented which show the dependence of spreading rate on: time, original NaCl concentration and Zn-Mg-Al alloy composition. The role of aluminium and magnesium in modifying (reducing) rates of secondary spreading is discussed in relation to their relative electrocatalytic activities for cathodic oxygen reduction and their distribution within the microstructure of the various alloy coatings.

It is shown, using in-situ profilometry, that at lower initial concentrations of NaCl (and consequently higher relative humidity under isopiestic conditions) the original electrolyte droplets dry out within approximately 2-3 hours of being deposited on the alloy surface. The mechanism of this drying phenomenon is discussed in relation to the relevant mechanisms of atmospheric corrosion. It is proposed that as corrosion proceeds all the Na⁺ cations originally present in the droplet migrate out into the region of secondary cathodic spreading. Conversely, Cl^- anions migrating towards the anodic sites at the droplet centre become sequestered as insoluble (or poorly soluble) zinc corrosion products such as Simonkolleite, Zn(OH)Cl. (The presence of Simonkollite is confirmed using grazing angle Z-ray diffraction.) As these corrosion products precipitate the activity of water in the droplet (and its associated vapour pressure) increase and water evaporates from the droplet as it tries to maintain isopiestic equilibrium with the experimental atmosphere. Corrosion-driven drying was also seen for higher initial concentrations of NaCl but at considerably slower rates. The drying effect, caused by the secondary spreading phenomena, may be an important self-limiting mechanism of atmospheric corrosion affecting Zn-Mg-Al alloy coatings.