Elsevier

Corrosion Science

Volume 176, November 2020, 109009
Corrosion Science

Impact of fluoride concentration on general corrosion of Mg-Zr alloy in a Na-geopolymer and alkaline solutions

https://doi.org/10.1016/j.corsci.2020.109009Get rights and content

Highlights

  • Impact of fluoride concentration on the corrosion rate of magnesium in alkaline solutions and in Na-geopolymer.

  • NaF concentration ranges required to reduce corrosion in alkaline solutions and in Na-geopolymer.

  • Morphology and chemical composition of corrosion products formed on magnesium surface.

  • Proposition of corrosion mechanisms in presence of fluoride in Na-geopolymer and in alkaline media.

Abstract

Corrosion of a metal in cementitious media can be limited when the pore solution of the binder contains corrosion inhibitors, as in the case of immobilization of magnesium nuclear wastes in France. This works aims to elucidate the impact of NaF on the magnesium corrosion, when it is incorporated in the Na-geopolymer formulation. The presence of NaF, even in small concentrations, considerably reduces the amount of corrosion products found on the magnesium/Na-geopolymer interface. Some concentration ranges were considered optimal for reducing corrosion and are related to the type of corrosion film formed on the metal surface, which provides greater protection.

Introduction

Magnesium alloys are known for their reactivity and their corrosion is commonly associated with hydrogen releases. Reducing the corrosion rate of these alloys and consequently the hydrogen evolution is essential in some situations, such as the storage of metallic wastes from the UNGG (Uranium Natural Graphite Gas) nuclear reactors. Preliminary studies pointed out Na-geopolymers containing NaF as the most adapted materials for the immobilization of magnesium wastes due to their low corrosion rate and low hydrogen evolution [1,2].

The geopolymer provides physical and chemical protection for magnesium. It acts as a physical barrier that minimizes contact between the metal and the external environment. Nevertheless, corrosion of magnesium alloys could occur in the geopolymer due to electrochemical reactions with the pore solution of the cementitious matrix, and then reduce the protective effect. There are indications that the uniform corrosion process in Na-geopolymer/NaF is limited by the presence of fluoride ions in its pore solution [2], due to their ability to inhibit the magnesium corrosion [[3], [4], [5], [6], [7]]. However, divergences are listed for the optimal concentration required to obtain the lowest corrosion rate [3,5,[8], [9], [10]]. This phenomenon can be correlated to the structure of the corrosion products formed on the metal/matrix interface, which is directly proportional to the fluoride concentration in the pore solution.

Predicting variations of the fluoride concentration in geopolymer pore solutions is relevant since they depend on the interactions between the liquid and solid phases, as well as their ability to diffuse through the hydraulic binder to reach the metal surface during the corrosion/inhibition process. As the concentration of fluoride may vary, corrosion is subject to modification and this mechanism must be further investigated.

A common method used to study corrosion in a cementitious matrix consists in performing tests directly in the pore solution extracted from the cement paste, or in solutions representative of the interstitial solution [[11], [12], [13]]. The preference for the study in solution is due to its simple setup and interpretation, in contrast to studies conducted in hydraulic binders. Nevertheless, a part of this study was conducted in geopolymer mortars, to compare the responses with the ones obtained in simulated pore solution.

This work aims to understand the general corrosion of magnesium in the presence of fluoride and how its concentration can affect the corrosion inhibition mechanism. In order to better interpret this process, we propose a magnesium corrosion evaluation in geopolymers and in model alkaline solutions, which simulate the pore solution of these materials after the hydration process.

Section snippets

Material and methods

This work is divided in two parts focusing on the study of magnesium corrosion in the presence of sodium fluoride. The first part consists in verifying Mg-0.5 %Zr corrosion when immobilized in Na-geopolymers containing different concentrations of NaF. The second part was performed in model alkaline solutions with concentrations of fluoride equivalent to those found in pore solutions of geopolymers after the hydration process. In both cases, the purpose is to verify the impact of the fluoride

Electrodes and electrochemical system

The same system composed of three electrodes was used in the study of magnesium corrosion in the geopolymer matrices and in the alkaline solutions. The measurements were performed twice for each condition to test their repeatability. The Mg-0.5 %Zr alloy samples were provided by Neyco®, whose composition is Mg =99.95 wt.%, Zr =0.5 wt.% (impurities (ppm): Al < 10, As < 20, Co < 10, Cr = 22, Cu = 2, Fe = 9, Mn = 11, Sb < 10, Si < 10, Zn = 31, Cl = 10). They have a cylindrical shape with a

Electrochemical measurements

An open circuit potential measurement was initially performed during 24 h for embedded samples or 72 h for non-embedded electrodes, followed by polarization measurements. A potentiodynamic polarization scan was performed for the non-embedded electrodes starting from -0.04 V to 2 V vs. EOCP with a scan rate of 0.1 mV.s−1.

Polarization curves were used to characterize the electrochemical behaviour of the electrode/electrolyte system, which required the system to achieve a permanent regime. In the

Characterization of the corrosion products

The electrodes were characterized with a Scanning Electron Microscopy (SEM), a ZEISS EVO MA15 with a Bruker Quantax Energy Dispersion Spectrometer (EDS). For the embedded electrodes, the cross section Mg-Zr/geopolymer was characterized after 28 days of immobilization in each formulation of geopolymer. The protocol before the analysis involved polishing them under ethanol to #1200 grit (SiC paper) using an automatic Struers Tegramin-30 polisher and finished with 9 μm and 3 μm diamond suspensions

Corrosion in geopolymers

Before the potentiostatic polarization tests, the magnesium samples were immobilized for 28 days in the geopolymers, undergoing an initial corrosion process. The Fig. 2 shows the interface of the Mg-Zr/geopolymer for the five mortar mixes (cracks observed in the geopolymer matrix are a consequence of the sample preparation). The sample without NaF was significantly more corroded, with corrosion products thicknesses reaching up to 150 μm.

In the absence of fluorides in alkaline pH levels,

Conclusion

In all the cases presented here, it was clear that the fluoride present in pore solutions of geopolymers, even in small concentrations, was efficient in reducing the corrosion at the Mg-Zr/geopolymer interface. The formulations with 0.05, 0.36 and 1.25 mol.L−1 of NaF in the activation solution of the geopolymers, were however the most effective in inhibiting corrosion. There is a probable formation of MgF2 in all cases and this film can evolve into NaMgF3 for the highest fluoride concentration.

Data availability

The raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.

CRediT authorship contribution statement

C.F. Barros: Conceptualization, Methodology, Writing - original draft, Writing - review & editing. B. Muzeau: Supervision, Writing - review & editing. V. L’Hostis: Supervision. R. François: Supervision.

Declaration of Competing Interest

The authors reported no declarations of interest.

Acknowledgments

Financial support from the CEA (Commissariat à l’Energie Atomique et aux Energies Alternatives) and from ORANO.

References (28)

Cited by (11)

  • Effect of magnesium-to-phosphate ratio on the corrosion resistance of magnesium alloy embedded in magnesium potassium phosphate cement

    2023, Cement and Concrete Composites
    Citation Excerpt :

    Nevertheless, the OCP values in the MKPC pastes are more positive than those in the MKPC pore solutions. This could be associated with the lower water content in the MKPC pastes compared to the pure MKPC pore solutions, which possibly slows down the dissolution of magnesium and the hydrogen reduction reaction [34]. Additionally, the OCP values in both MKPC pastes and pore solutions decreased with an increase in M/P ratio.

  • Insight into synergistic corrosion inhibition of 3-amino-1,2,4-triazole-5-thiol (ATT) and NaF on magnesium alloy: Experimental and theoretical approaches

    2022, Corrosion Science
    Citation Excerpt :

    This is mostly ascribed to the much more unstable nature and less compact microstructure of the oxide films formed on the surface, which can barely support the adsorption of halogen anions on the surface [7,15]. Compared to the reported halide elements, fluoride (F) itself is widely acknowledged as an effective inhibitor for Mg alloys, since it can form insoluble precipitates (e.g. MgF2 and/or NaMgF3) on their surface [16,17]. However, the corrosion inhibitive ability of F- is significantly concentration-dependent.

  • The effect of fluoride on the formation of an electroless Ni-P plating film on MAO-coated AZ31B magnesium alloy

    2022, Journal of Materials Research and Technology
    Citation Excerpt :

    An EIS test was performed after immersion in OCP for 10 min at frequencies of 100 kHz to 10 mHz and using a sinusoidal AC perturbation of 10 mV amplitude. A salt spray test was conducted according to ASTM B-117 [54–58]. The specimens were placed in a 5 wt.

View all citing articles on Scopus
View full text