The oxidation of pure Mg, Mg-3Al-1Zn (AZ31B), and Mg-1Zn-0.25Zr-<0.5Nd (ZE10A) was studied at 85°C in humid air using sequential exposures with H 2 18 O and D 2 16 O for water vapor. Incorporation of 18 O in the hydroxide/oxide films indicated that oxygen from water vapor participated in the reaction. Penetration of hydrogen into the underlying metal was observed, particularly for the Zr-and Nd-containing ZE10A. Isotopic tracer profiles suggested a complex mixed inward/outward film growth mechanism.Magnesium alloys are of great interest for the manufacture of vehicle parts, biomedical implants, and functional applications involving hydrogen storage, batteries, and fuel cells [1][2][3]. Understanding the high reactivity and film formation mechanism(s) associated with Mg alloys is a key to improving corrosion resistance, as well as tailoring corrosion reaction/stability behaviors for biomedical and functional use. Both MgO and Mg(OH) 2 films are formed when Mg alloys are exposed to water [4,5].The corrosion behavior of Mg alloys in aqueous solutions has been widely studied [e.g. 4, 5]. However, oxidation behavior studies at elevated temperatures have not been as widely pursued for Mg [6][7][8][9][10][11][12], but are relevant to applications such as automotive powertrain components, for example. The present authors recently demonstrated the feasibility of performing isotopic tracer studies of the film growth mechanism for the immersion of Mg in water, using sequential exposures in D 2 16 O and/or H 2 18 O water and secondary ion mass spectroscopy (SIMS) analysis [13]. The goal of the present work was to extend this experimental approach to study the oxidation film growth mechanism of Mg in hot humid air.Ultrahigh purity Mg (UHP Mg) and two Mg alloys: AZ31B to represent the Mg-Al-Zn alloy class and Elektron Ò 717 (ZE10A grade, referred to as E717 for brevity), to represent the Mg-Zr-rare earth alloy class, were selected for study (Table 1). The same material batches and sample preparation methods as described in [13] were used in the present work. Test samples $9 mm in diameter and $1-1.5 mm thick were wet ground to P1200 grit finish, cleaned with acetone and deionized water, and dried with an air stream [13]. Samples were stored in a desiccator for at least 24 h prior to the tracer studies [13].Oxidation exposures were done at 85°C in humid air using a sealed stainless steel reaction vessel approach previously developed for geochemical hydration studies [15]. The reaction vessels ($24 cm 3 ) contained an inner Teflon liner to which 2 ml of isotopic tracer water was added, with the test sample suspended above the water in the vapor phase (humid air) in a Teflon mesh basket. The reaction vessels were then sealed and placed in a covered aluminum block holder inside of a convection oven at 85°C. At this temperature, these vessel and fluid volumes result in a two-phase http://dx.