Grain boundaries amplify local chemical ordering in complex concentrated alloys

Local chemical ordering strongly influences the behavior of complex concentrated alloys, yet its characterization remains challenging due to the nanoscale dimensions and scattered spatial distribution of the ordered domains. Here, we study chemical ordering near grain boundaries, demonstrating they can act as microstructural anchor points that amplify chemical order and drive the formation of compositional nanopatterns. Atomistic simulations reveal the development of composition waves with ordering vectors normal to the boundary plane in two distinct material systems, CrCoNi and NbMoTaW. These waves manifest as periodic enrichment-depletion patterns that reflect the underlying chemical ordering tendencies of each system, but with amplified contrast that extends several nanometers into the grain interior before gradually decaying. By examining multiple grain boundary orientations and alloys, we show that both the interfacial segregation profile and the crystallographic terminating plane govern the extent and character of this amplification. This interplay between boundary-dictated directional ordering and the diffuse, untemplated chemical domain evolution within the grain advances our understanding of interface-mediated ordering phenomena and suggests new opportunities for experimentally detecting local chemical order in complex concentrated alloys.

💡 Research Summary

Complex concentrated alloys (CCAs) exhibit local chemical ordering (LCO) that strongly influences their mechanical and functional properties, yet the nanoscale size and random distribution of ordered domains make experimental detection difficult. In this work, the authors use large‑scale atomistic simulations to demonstrate that grain boundaries act as “anchor points” that amplify LCO and generate compositional nanopatterns extending several nanometers into the grain interior. Two representative alloy systems are examined: the face‑centered cubic (FCC) CrCoNi alloy and the body‑centered cubic (BCC) NbMoTaW alloy.

For CrCoNi, a Σ11 <110> symmetric tilt boundary is modeled. Hybrid Monte‑Carlo/molecular‑dynamics (MC/MD) simulations in a variance‑constrained semi‑grand canonical ensemble reveal rapid segregation of Ni to the boundary, followed by the development of Cr–Co short‑range order in the bulk. As temperature decreases, Ni enrichment initially reaches ~80 at.% at the interface, then partially redistributes as LCO strengthens. Importantly, a composition wave with an ordering vector normal to the boundary plane emerges: Ni‑rich and Ni‑depleted layers alternate with a wavelength of a few atomic spacings, persisting up to ~6 nm from the boundary before decaying. The wave’s amplitude and decay length are governed by the balance between strain‑relief‑driven segregation (Ni’s larger atomic volume) and chemistry‑driven ordering (Cr–Co bonding).

For NbMoTaW, a series of Σ3 tilt boundaries with varying inclination angles (φ = 19.47°, 27.94°, 35.26°) are studied. Nb segregates to the boundary, while Mo–Ta B2‑type ordering appears adjacent to the interface. The most pronounced ordering occurs at the Σ3 (221)/(001) boundary; rotating the boundary weakens the order, indicating that the terminating crystallographic plane controls the templating effect. As in the FCC case, a normal composition wave forms, with Mo‑Ta enrichment alternating with Nb‑depleted regions, extending several nanometers into the grain.

Both alloy systems show that (i) grain‑boundary segregation creates a chemically distinct interfacial layer, and (ii) this layer serves as a directional template that enhances pairwise bonding preferences in the neighboring lattice. The resulting “chemical waves” are periodic, have higher contrast than bulk LCO, and decay gradually with distance from the interface. The authors also compare results from an embedded‑atom method (EAM) potential with a neural‑network potential, finding consistent qualitative behavior despite quantitative differences.

The study links the observed amplification to two key factors: the crystallographic nature of the terminating plane (which determines the local atomic arrangement and thus the preferred bonding geometry) and the segregation profile (which supplies the necessary excess of a particular element to seed ordering). These insights suggest practical routes for experimental detection: high‑angle annular dark‑field STEM can exploit the Z‑contrast of the enriched layers; selected‑area electron diffraction or nano‑beam diffraction may reveal extra diffuse spots or superlattice reflections aligned with the wave vector; and atom‑probe tomography could map the enrichment‑depletion oscillations near special boundaries.

Beyond detection, the findings open a design pathway: by engineering grain‑boundary character distributions (e.g., increasing the fraction of low‑energy Σ boundaries that favor strong segregation) and by controlling heat‑treatment schedules to lock in the segregation profile, one can deliberately induce and stabilize amplified LCO. This could be leveraged to tailor mechanical strength, ductility, or high‑temperature stability in CCAs, where ordered clusters impede dislocation motion or suppress grain growth.

In summary, the paper establishes grain boundaries as a universal catalyst for local chemical ordering in both FCC and BCC complex concentrated alloys, elucidates the mechanistic interplay between segregation, crystallography, and ordering, and proposes concrete experimental and alloy‑design strategies to harness this interface‑mediated phenomenon.