Investigation on the corrosion mechanisms of pure magnesium and the effect of friction stir welding (FSW) on the corrosion resistance of aluminum alloy 2524-T3

Abstract

Friction stir welding (FSW) is a process that has proven to be quite efficient when it comes to joining high-strength aluminum alloys, for instance AA2524-T3. This can be justified by the fact that welding aluminum alloys by FSW technique allows (i) reduction of aircraft weight by eliminating the rivets commonly used and (ii) the use of different aluminum alloys that have low specific density and high mechanical strength. However, even though FSW allows the joining of metallic parts without their effective fusion, which theoretically would result in a defect-free weld bead, the heat resulting from the friction of the welding tool causes significant microstructural changes. In consequence, it results in variations of mechanical properties and corrosion resistance in the welded region. It was evaluated the FSW welding process influence on the corrosion resistance of the joined 2524-T3 aluminum alloy. Different tools have been used, such as: optical and scanning electron microscopy; open circuit potential and electrochemical impedance spectroscopy measurements; and corrosion tests: agar-agar test, intergranular corrosion test, and exfoliation corrosion test. It was proven by different techniques that the thermomechanically/thermal affect zone interface on the retreating side was the most susceptible to corrosion among all investigated zones. A parallel study was developed regarding the corrosion mechanism of pure magnesium. It has been the subject of a considerable amount of work, and despite its ubiquity and history, it remains controversial. This is mainly due to the presence of the negative difference effect (NDE), which increases hydrogen formation when the magnesium is biased on the anodic domain. We was performed a detailed analysis of the electrochemical impedance spectra obtained for the Mg electrode during immersion in a sodium sulfate solution. A model was proposed which took into account the presence of: (i) a thin oxide film (MgO) which progressively covered the Mg electrode surface, (ii) film-free areas where the Mg dissolution occurs in two consecutive steps, (iii) a thick layer of corrosion products (Mg(OH)2), (iv) an adsorbed intermediate Mg+ads which is responsible for the chemical reaction allowing the NDE to be explained. From the impedance data analyses, various parameters were extracted such as the thin oxide film thickness, the resistivity at the metal/oxide film interface and at the oxide film/electrolyte interface, the active surface area as a function of the exposure time to the electrolyte, the thickness of the thick Mg(OH)2 layer and the kinetic constants of the electrochemical reactions.