Percolating Corrosion Pathways of Chemically Ordered NiCr Alloys in Molten Salts

TL;DR

This study addresses how chemical ordering in Ni–Cr alloys modulates corrosion in molten fluoride salts. It combines reactive molecular dynamics with a ReaxFF description and first-principles calculations to show that long-range Ni2Cr ordering forms continuous Cr networks that percolate to the surface, accelerating Cr dissolution and surface degradation, while short-range ordering or random solid solutions exhibit slower, diffusion-limited corrosion. The results link percolation topology to corrosion kinetics and reveal a lower interfacial Cr-dissolution barrier in the ordered state, providing a mechanistic explanation for experimental observations. The findings suggest strategies to mitigate corrosion—such as promoting SRO or alloying with Mo to disrupt Cr connectivity—relevant to the design of Ni–Cr-based materials for molten-salt systems.

Abstract

Recent experiments have shown that chemical ordering in NiCr alloys can significantly accelerate corrosion in molten salt environments. However, the underlying mechanisms remain poorly understood. Using reactive molecular dynamics and first-principles calculations, we show that long-range ordered Ni$_2$Cr in Ni-33at.%Cr alloys corrodes far more rapidly in FLiNaK salt at 800°C than short-range ordered or random solid solutions. This accelerated attack originates from percolating Cr pathways that enhance near-surface diffusion and a lowered energetic barrier for Cr dissolution, as confirmed by first-principles calculations. Contrary to earlier explanations that attributed this behavior to residual stresses, our stress-free simulations demonstrate that ordering alone accelerates the degradation. These results establish percolation as a critical link between chemical ordering and corrosion kinetics, offering a mechanistic basis for experimental observations.

Figures (8)

• Figure 1: Optimized simulation cells for (a) LRO, (b) SRO, and (c) RSS (110)-surface-oriented Ni–Cr alloys in contact with FLiNaK salt at 800 oC and atmospheric pressure.
• Figure 2: Time evolution of Cr and Ni dissolution from the NiCr alloys with LRO, SRO, and RSS configurations, in molten FLiNaK at 800oC. Shaded regions indicate standard deviations.
• Figure 3: Snapshots of NiCr slabs with (a) LRO, (b) SRO, and (c) RSS after 3 ns at 800°C in contact with molten FLiNaK salt. Salt ions are removed for clarity. (d-f) Corresponding surface meshes of the LRO, SRO, and RSS slabs, respectively, color-coded by the local $Z$ coordinates. Each surface morphology is reconstructed by isolating the uppermost 3.5 Å layer of each slab.
• Figure 4: Average atomic number density profiles along the $Z$-direction of NiCr alloys at 800 $\mathrm{^o}$C with (a) LRO, (b) SRO, and (c) RSS configurations at 0 ns (dahsed) and 3 ns (solid). (d) Change in areal density of Ni and Cr across the toplayers after 3 ns.
• Figure 5: Charge distribution along the $Z$-direction of the NiCr alloys after 3 ns at 800 $\mathrm{^o}$C: (a,d) LRO, (b,e) SRO, and (c,f) RSS slabs. (a-c) Cr charges and (d-f) Ni charges. Dashed lines mark the initial layer positions. Atom charges are color-coded based on the original layers in the initial perfect structure.