The in vivo and in vitro corrosion of high-purity magnesium and magnesium alloys WZ21 and AZ91 Nor Ishida Zainal Abidin a,b , Barbara Rolfe c , Helen Owen d , Julian Malisano a , Darren Martin c , Joelle Hofstetter e , Peter J. Uggowitzer e , Andrej Atrens a,⇑ a The University of Queensland, Division of Materials, St Lucia, Qld 4072, Australia b University of Malaya, Department of Mechanical Engineering, Faculty of Engineering, Kuala Lumpur 50603, Malaysia c The University of Queensland, Australian Institute for Bioengineering and Nanotechnology, Brisbane, Qld 4072, Australia d The University of Queensland, School of Veterinary Science, Gatton, Qld 4343, Australia e ETH Zurich, Department of Materials, CH-8093 Zurich, Switzerland article info Article history: Received 19 March 2012 Accepted 13 June 2013 Available online 26 June 2013 Keywords: A. Magnesium B. Weight loss B. SEM abstract Corrosion was studied in vitro in Nor’s solution (CO 2 – bicarbonate buffered Hank’s solution) at 37 °C, and in vivo implanted in the lower back muscle of rats. Nor’s solution is a good model for HP Mg and WZ21, because (i) the pH is maintained by the same buffer as in blood and (ii) concentrations of corrosive chlo- ride ions, and other inorganic constituents, are similar to those in blood. The higher in vitro corrosion rate of AZ91 was caused by micro-galvanic from second phases. The lower in vivo corrosion rate of AZ91 was tentatively attributed to suppression of micro-galvanic corrosion by tissue encapsulation. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction 1.1. Mg alloy corrosion Mg alloys are gaining interest for biodegradable medical im- plant applications [1–13] due to a good combination of load bear- ing mechanical properties, and controllable corrosion rates. Their poor corrosion resistance, however, has limited their applications where they might be exposed to corrosive solutions containing chloride ions [1,2,14–20]. In addition there are known issues [1,14,21] with the measurement of the Mg corrosion rate using Tafel extrapolation of polarisation curves. The poor corrosion resistance of Mg alloys [1,2,14–17,22–27] results (i) because Mg is an active metal and has a high driving force for corrosion, whereas only weak protection is provided by naturally-occurring surface films and (ii) micro-galvanic corrosion acceleration is caused by second phases. Mg alloys typically cor- rode faster than high-purity (HP) Mg. No alloying element has pro- duced a corrosion rate for a solid–solution Mg alloy lower than that of HP Mg in a technically relevant testing solution like 3% NaCl, un- like Cr in stainless steels, nickel and cobalt base alloys, where Cr alloying produces low corrosion rates [28–36] above a critical Cr concentration by means of a more stable passive film. HP Mg means that the impurity elements (Fe, Ni, Cu and Co) are each below their (alloy dependent) tolerance limit [1,14,15,26]. Above the tolerance limit, the corrosion rate is high. Some studies [26,37,38] have been dominated by the Fe impurity element, even though they aimed to study completely different phenomena. Fishing-line specimens, and plug-in specimen, were developed [17] in response to the known issues [1,14,21] with the measure- ment of Mg corrosion using Tafel extrapolation of polarisation curves. Fishing-line specimens were designed as the most-mini- malist-possible specimen mount, and identified the issue of crevice corrosion for Mg during immersion tests [17]. Plug-in specimens were subsequently designed [17] to have no crevice, and to allow reliable polarisation curves to be mea- sured. Three reasons were identified [17] why Tafel extrapola- tion had previously not yielded corrosion rate measurements in agreement with independent measurements. (1) Crevice corro- sion can occur in the specimen mount when the specimen is mounted in a metallurgical mount, or similar, as widely used in Mg corrosion studies. (2) The evolving hydrogen and corrosion products can cause decoupling of the corrosion and electrochem- ical measurements. (3) Often the polarisation curves, used for Tafel extrapolation, have been measured soon after specimen immersion. The initial corrosion behaviour may not correlate with steady state corrosion behaviour, either in vivo or in vitro, particularly as Mg alloy corrosion can have a substantial incubation period. 0010-938X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.corsci.2013.06.019 ⇑ Corresponding author. Tel.: +61 7 3365 3748. E-mail address: andrejs.atrens@uq.edu.au (A. Atrens). Corrosion Science 75 (2013) 354–366 Contents lists available at SciVerse ScienceDirect Corrosion Science journal homepage: www.elsevier.com/locate/corsci