Effect of Crystallographic Orientation on the Corrosion of Magnesium: Comparison of Film Forming and Bare Crystal Facets using Electrochemical Impedance and Raman Spectroscopy

Abstract

The corrosion rate of Mg indicates a strong crystallographic dependence in chloride-containing, alkaline environments that correlates inversely with oxide film thickness. In contrast, a different crystallographic orientation dependency is observed initially during open circuit corrosion in non-chloride containing, near neutral pH buffered, chelating environments such as Tris(hydroxymethyl)aminomethane (TRIS) and Ethylenediaminetetraacetic (EDTA) which minimize air-formed MgO oxides. The origins of the differences in the rates of the coupled corrosion processes as a function of crystallographic orientation were investigated utilizing electrochemical impedance spectroscopy (EIS) and Raman vibrational spectroscopy. High frequency constant phase elements (CPEs) were exploited to determine oxide thicknesses as a function of crystal orientation. In unbuffered NaCl, open circuit corrosion was faster on basal planes while lower corrosion rates were observed on low index, prismatic and pyramidal planes. This variation in rate with crystallographic orientation in 0.6 M NaCl was interpreted to depend on the MgO and Mg(OH) 2 film thicknesses as a function of orientation. In particular, crystal planes with a lower MgO/Mg(OH) 2 thickness corroded at higher rates as assessed in-situ by EIS and suggested by Raman spectroscopy. The crystallographic orientation dependence of corrosion seen initially upon exposure in the buffered neutral pH environments corresponded with crystal facet surface energy for bare Mg but this trend disappeared with exposure time and Mg(OH) 2 films were then detected.

Corrosion Inhibition Study of Aqueous Vanadate on Mg Alloy AZ31

“…This response is commonly observed with Mg and Mg alloys under these types of exposure conditions. 39 4.0x10 -6 6.0x10 -6 8.0x10 values (R p ) were estimated using a method derived from the KramersKronig transforms. 42,43 According to this method, the polarization resistance can be estimated from the imaginary component of the impedance integrated over the measured frequency domain: integral of each fit over its frequency range was computed (Fig.…”

Corrosion of Mg alloys EV31A, WE43B, and ZE41A in chloride‐ and sulfate‐containing solutions saturated with magnesium hydroxide

“…The quality of the fit is attributed to the fact that there were slow changes in the corrosion rate in the present work as is evident from Figures . The fit was not as good as those by Gomes et al and those of Scully and co‐workers; however, those works used solutions that were much less corrosive, so the corrosion behavior was much more stable, and there was much less change in corrosion behavior during the measurement of EIS spectra. The corrosion rates evaluated from the EIS data, P i,EIS,1 and P i,EIS,2 , were consistently smaller than the corrosion rates evaluated from the evolved hydrogen P H .…”

“…Comparison of the corrosion rates, P H , P i,EIS,Rp1 , P i,EIS,Rp2 , and P i,EIS,CT of the WE43B specimen designated as WE29 immersed at the open‐circuit potential in 0.6 M NaCl solution saturated with Mg(OH) 2 for 7 days, where P H is instantaneous corrosion rate evaluated from the volume of evolved hydrogen, P i,EIS,Rp1 is the corrosion rate calculated from the resistance at the lowest frequency, and P i,EIS,Rp2 is the corrosion rate evaluated from the polarization resistance calculated using Equation , and P i,EIS,CT is the corrosion rates calculated from the charge transfer resistance, R 1 …”

“…Figure shows the MgO film thickness on specimens WE29 and ZE21 over the test period, which was calculated using where C is C 1 or C PE , (10 −6 F/cm 2 ), ε is the dielectric constant for MgO (9.8 dimensionless), ε 0 is the permittivity of vacuum (8.84 × 10 −14 F/cm), A is the specimen surface area (cm 2 ), and l ox is the thickness of the MgO layer (nm). Figure shows that the MgO film thickness on WE29 continued to grow from an initial value of 3 nm over the test period.…”

“…Scully and colleagues and Liu et al proposed that EIS data for Mg corrosion, which includes an inductive loop, can be fitted to the equivalent circuit given in Figure , where R s is the solution resistance, C 1 represents the capacitance of the surface oxide (MgO), C 2 represents the double‐layer capacitance, R 1 represents the charge transfer resistance, R 2 is the oxide resistance, R 3 is the resistance of a fresh anode with adsorbed species, and the inductor L represents a reaction involving adsorbed intermediates. They maintained that (a) the equivalent circuit is physically realistic and contains the minimum number of circuit elements to explain the EIS data, and (b) the corrosion current density, i corr , can be evaluated from the Stern–Geary equation: where are the anodic and cathodic Tafel constants ( has a negative value) and R P is the polarization resistance, evaluated as the difference between R s and the low‐frequency asymptote of the impedance as the frequency tends to zero. The equivalent circuit in Figure indicates that R P can be evaluated from: …”