Predicting and understanding corrosion in molten chloride salts

Molten chloride salts are stable at higher temperatures than many other salts, including nitrate salts, and are thus promising for heat transfer and/or thermal energy storage in concentrating solar power, nuclear power, and other thermal energy storage applications. However, corrosion in molten chloride salts remains a significant problem. While many studies have been devoted to evaluation of corrosion, we find that a comprehensive method for predicting corrosion in molten chloride salts is lacking. Here, we present an evaluation of corrosion in molten chloride salts using Ellingham diagrams and chloride-oxide stability diagrams, which enable prediction of alloy performance in molten chloride salts and allow corrosion results to be interpreted at a fundamental level.


Introduction
A comprehensive body of literature exists on the study of corrosion of alloys in molten chlorides [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19], The focus of corrosion studies has been primarily concerned with Febased and Ni-based alloys, i.e., stainless steels versus nickel super alloys.These studies often experimentally evaluate depletion of individual elements in super alloys.However, Ellingham diagrams, often overlooked in the literature, are a useful tool for predicting behavior of alloying elements.In combination with chloride-oxide stability diagrams, Ellingham diagrams can explain much of the experimentally observed behavior of alloys is chloride salts.Typically, super alloys exhibit high corrosion resistance due to the presence of a passivating oxide layer such as chromium oxide.However, in molten chloride salts Cl − ions challenge this oxide layer, and expose the alloying constituents to further oxidation, creating what has been termed the "chlorine-oxidation cycle" [1][2][3][4][5][6][7][8][9][10], For all alloys corrosion follows the same stages of (1) oxidation of elements in alloy, followed by (2) dissolution or vaporization of oxidized elements.The rate of corrosion has therefore been strongly correlated to the presence of oxidizing impurities in the chloride salt [3, 5-11, 13-16, 18-23].Understanding these studies using Ellingham diagrams and chloride-oxide stability diagrams enables predictive insight into alloy performance.Furthermore, it underscores the importance of limiting corrosive impurities via control of salt exposure to water, oxygen, and other oxidizing agents.

Construction of Ellingham diagrams
The utility of Ellingham diagrams in evaluating molten chloride salts systems can be illustrated in the example case of MgCl 2 .The formation of corrosive impurities primarily stem from the hygroscopic nature of MgCl 2 resulting in the formation of oxide/hydroxide species in the presence of oxygen and moisture via the following reactions: Furthermore, MgOHCl has been found to decompose above 550°C according Kerry Rippy and Liam Witteman have contributed equally to this work.
nb Several studies have found a direct correlation between corrosion rates and concentration of MgOHCl present in the chloride salt [20,23], Furthermore, the formation of HCl (g) and Cl 2(g) via leads to corrosion of alloys in the headspace [7,23,24].The corrosive impurities lead to degradation reactions with alloying constituents M (e.g., Cr, Fe, Ni) according to: This degradation is an electrochemical process that can be explained via half-cell reactions as follows [9,25]: Anodic oxidation of alloying element M: where n is the number of electrons exchanged.Cathodic reduction of corrosive species: where Ox is the oxidizing impurity in this case (e.g.MgOHCl), and Red is the reduced form of the oxidizing impurity.
The complete redox couple reaction: For the electrochemical process to occur spontaneously the change in Gibbs-free energy needs to be negative, which can be calculated via: where ∆G rxn is the change in Gibbs-free energy of reaction (J mol −1 ), F is Faraday's constant (96,485 C mol −1 ), and ∆E rxn is the redox potential of reaction (V ).
Equation 10 allows for the construction of an Ellingham diagram, shown in Figure 1 (constructed for various alloying constituents using the thermodynamic software HSC v8).
According to Figure 1, the most stable chlorides are K, Na, and Mg.Therefore, common alloying constituents (e.g.Fe, Cr, Ni) should theoretically remain stable within the chloride salt.However, the influence of oxidizing impurities such as MgOHCl is not readily captured by simply looking at Equation 10.To assess the effect of oxidizing impurities, ∆E rxn can be further expressed into the anodic and cathodic potentials via the Nernst Equation: (3) where E 0 is the potential under standard conditions (V ), R is the ideal gas law constant (8.314J mol −1 K −1 ), T is temperature (K), and a is the activity, which for a pure solid is unity, i.e., a M = 1.From equation 13, the effect of increasing the concentration of oxidizing impurities (increasing a Ox ) leads to increasing E c -thereby increasing ∆E rxn , and results in a more negative ∆G rxn .So far, the mathematical treatment around the Ellingham diagram is useful in predicting some experimental trends, such as the rate of depletion of certain alloying constituents.The diagram correctly predicts the rate of depletion for Mn > Cr > Fe > Ni [5,11].However, the diagram incorrectly predicts the rate of depletion of Nb, Mo, and W. Several studies have suggested the presence of these alloying constituents to slow down the rate of corrosion [10,26].

Construction of chloride-oxide stability diagrams
The construction of a chloride-oxide stability diagram elucidates more information regarding the thermodynamic behavior of alloying constituents in molten chloride salts [10].Such a diagram was constructed by calculating the ∆G rxn of the oxides versus chlorides of various alloying constituents via HSC v8 (Fig. 2.).
Figure 2 has three primary regions: i. Lower half representing oxide species are more stable than chloride species.
( Several experimental observations can be explained via the chloride-oxide stability diagram.For example, K and Na chlorides are highly resistant to oxidation and can be considered stable in the presence of oxygen and moisture, whereas Mg is not [25].Additionally, upon oxidation of Mn, Cr, Fe, Co, and Ni, the oxide will equilibrate with chloride ions and form chlorides as suggested by the chlorine-oxidation cycle.Lastly, alloys containing W, Mo, Al and Nb have enhanced corrosion resistance due to the formation of a relatively stable oxide that can serve as a passivation layer [10].The utility of the chloride-oxide stability diagram elucidates several experimental observations that the typical Ellingham diagram overlooks.The diagram highlights that even Ni itself falls victim to the chlorine-oxidation cycle, as was observed by Liu et al. [5].Even commercially pure Ni (e.g.Ni-201), which is expected to provide superior corrosion resistance, has been observed to corrode in chloride salt to a point of failure within days [26].

Combined use of Ellingham diagrams and chloride-oxide stability diagrams as predictive tools
Chloride-oxide stability diagrams and Ellingham diagrams are useful tools in evaluating corrosion and interpreting results, even in less well studied systems such as convective molten chloride systems.Though studies under these conditions are limited, notable examples include forced convection studies dating back to 1960s at the Brookhaven National Laboratory [12] and natural convection studies that were conducted in early 2010 between Idaho National Laboratory and the University of Wisconsin Madison [26].We find that these examples are well explained by Chlorideoxide stability diagrams and Ellingham diagrams.
Natural convection corrosion studies elucidated timedependent corrosion mechanisms [26,27].In the initial stages, corrosion is primarily driven by oxidative impurities, which can be understood using chloride-oxide stability analysis.Once the concentration of these impurities diminished, the corrosion was dominated by active dissolution of selective alloying constituents (e.g., Cr) from the hot side, and deposited on the cold site.The studies highlighted the effect of temperature-dependent metal solubilities of species such as chromium chloride.Similar observations were observed in the static corrosion study conducted by Gong et al. [20].Under static conditions, the first 250 hours was primarily impurity-driven corrosion, after which the dominant mechanism became thermal effects resulting in metal solubility differences.Surprisingly, forced convection corrosion studies conducted at Brookhaven National Laboratory observed no appreciable change in corrosion rates compared to static conditions [12].The work highlighted the importance of salt purification and designing a leak tight system with an inert atmosphere.Therefore, an effective purification strategy can minimize corrosion by limiting the impurity-driven corrosion mechanism.

Conclusion
Corrosion remains a significant problem in molten chloride salts systems, but we propose that corrosion behavior of specific alloys can be predicted, and that experimental corrosion evaluation results can be understood, using a combination of Ellingham diagrams and chloride-oxide stability analysis.This method correlates corrosion behavior to fundamental thermodynamics and can be used to identify and explain performance of specific promising alloying elements.Thus, it can be used to identify high-performance alloys and to guide materials development toward appropriate alloys for use in molten chloride salts, which could in turn enable advances in chloride salt based CSP, nuclear, thermal energy storage, and other applications.sarily represent the views of the DOE or the U.S. Government or any agency thereof.Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights.The U.S. Government retains and the publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this work, or allow others to do so, for U.S. Government purposes.
Funding Open access funding provided by National Renewable Energy Laboratory Library.This work was authored in part by the National Renewable Energy Laboratory, operated by Alliance for Sustainable Energy, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308.Funding provided by U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Solar Energy Technologies Office grant CSP #35931 as well as the Colorado School of Mines/NREL Advanced Energy Systems Graduate Program.The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government.The U.S. Government retains and the publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this work, or allow others to do so, for U.S. Government purposes.

Fig. 1
Fig. 1 Gibbs-free energy of reaction of metal to metal-chloride as a function of temperature