Quenching Fails to Lock In the B2 Phase in AlCoCrFeNi High-Entropy Alloys

High-entropy alloys have spent two decades promising a new frontier in metallurgy: five or more principal elements jumbled together in near-equal proportions, defying the classical idea that alloys need one dominant metal to stay stable. Among the most studied of these multiprincipal systems is AlxCoCrFeNi, a family in which changing the aluminum fraction alone transforms the alloy from ductile face-centered cubic metal to a strong body-centered cubic material. But within that body-centered cubic domain lies a subtler and more consequential contest, one between two crystal structures that are nearly indistinguishable to most laboratory instruments yet profoundly different in performance. A new study from researchers at NRC-Negev, Ben-Gurion University of the Negev, and the Thermal Research Institute of Israel, published in the Journal of Materials Science: Metallurgy, has now quantified how much of each structure survives when these alloys are heated to near-melting temperatures and then slammed into cold water.

The two contenders are known as A2 and B2. The A2 phase is a disordered body-centered cubic solid solution, in which iron and chromium atoms occupy crystallographic sites in essentially random fashion. The B2 phase is its ordered sibling, an arrangement in which aluminum and nickel segregate to distinct sublattices, producing an ordered lattice enriched in aluminum, nickel, and cobalt. The distinction matters enormously in practice. The relative amounts of A2 and B2 influence how strongly the alloy strengthens, and the ordered B2 phase has been repeatedly linked to reduced corrosion resistance in these materials. Engineers designing turbines, marine hardware, or any component exposed to aggressive environments therefore need to know exactly how much B2 lurks in the microstructure, and whether that amount can be predicted or controlled.

Here is the problem that has frustrated the field for years: the lattice parameters of A2 and B2 differ by less than about 0.4 percent. X-ray diffraction, the workhorse technique for phase quantification in metals, simply cannot reliably separate the two across most of the composition range. Worse, published thermodynamic phase diagrams of the Al-Co-Cr-Fe-Ni system disagree substantially with one another about where the A2, B2, and mixed A2-plus-B2 fields begin and end. Different thermodynamic databases yield different widths for the two-phase field and even different conclusions about whether a nominally single-B2 region exists at high temperature. That disagreement leaves a deceptively simple question dangling: even if theory predicts a single B2 phase at elevated temperature, can that state actually be trapped at room temperature by practical quenching?

To answer it, the team led by Michael Aizenshtein and Shmuel Hayun prepared a series of AlxCoCrFeNi alloys spanning aluminum contents from one to four atoms per formula unit. Each alloy was arc melted under ultra-high-purity argon with a titanium getter to scavenge oxygen, remelted and flipped five times to homogenize the chemistry, and cast into six-millimeter rods. Samples were then held in a vertical tube furnace at 1593 to 1623 kelvin for thirty minutes under flowing argon before being quenched into room-temperature water. The aluminum-leanest composition received the slightly hotter treatment, a deliberate choice to suppress the formation of unwanted face-centered cubic and brittle sigma phases during cooling. Thirty minutes was a pragmatic compromise: long enough to approach the high-temperature state, short enough to avoid excessive coarsening of the phases.

The analytical strategy had to be as layered as the microstructure itself. High-resolution scanning electron microscopy was tried first, but the A2 precipitates turned out to be nanoscale, far below what scanning electron microscopy can quantify reliably. X-ray diffraction with whole-pattern fitting could resolve the B2 superlattice reflections only at higher aluminum contents, roughly x greater than or equal to 2.75. For everything leaner in aluminum, the burden fell to transmission electron microscopy. Thin foils were extracted by focused ion beam milling and examined with a field-emission TEM equipped for energy-dispersive X-ray spectroscopy. Electron diffraction identified the phases; elemental maps revealed their chemistry. The A2 fraction was estimated from the areal fraction of the chromium- and iron-rich precipitates within the B2 matrix, a reasonable approximation of local volume fraction because the precipitates are nanoscale and approximately equiaxed. Where both methods were feasible, their estimates agreed in trend and were averaged.

The elemental maps tell a vivid story of chemical self-organization. In Al2CoCrFeNi, the B2 matrix glows with aluminum and nickel, while the embedded A2 precipitates concentrate chromium and iron. The same partitioning appears at the extreme aluminum-rich end, where selected-area electron diffraction using the forbidden (100) reflection, a fingerprint of B2 ordering, confirmed the ordered phase exists but does not fill the field of view. Dark-field imaging revealed that even Al4CoCrFeNi, a composition that thermodynamics says should be overwhelmingly B2, retains nanoscale A2 precipitates after quenching. A single-B2 state was not experimentally retained anywhere in the composition range studied. The retained A2 fraction decreases steadily with increasing aluminum content, and a linear extrapolation suggests it would approach zero only near 57 atomic percent aluminum, close to the ideal 60 percent needed to perfectly fill the body-center sites of the B2 lattice. Yet the experimental curve flattens above roughly 40 atomic percent aluminum, hinting at a stubborn residual A2 fraction that refuses to disappear.

Comparison with thermodynamic calculations sharpens the mystery. The team ran CALPHAD calculations in Thermo-Calc with the TCHEA7 database, stepping through compositions and temperatures in fine increments. The calculations reproduce the qualitative trends beautifully: more aluminum stabilizes B2, higher temperature stabilizes B2, and a single-B2 field should emerge above roughly x equals 3 near 1550 kelvin. Quantitatively, however, the measured A2 fractions are consistently higher than the equilibrium predictions. The authors are careful not to read this as a failure of thermodynamics. Instead, the discrepancy points to kinetics: during the quench, atoms have just enough mobility to redistribute over short distances and re-form nanoscale A2 precipitates inside the B2 matrix before diffusion freezes out.

A simple back-of-the-envelope estimate makes that argument compelling. At a cooling rate of about 50 kelvin per second, cooling from 1473 to 1273 kelvin takes roughly four seconds. That window allows a characteristic diffusion length of only several tens of nanometers, which is precisely the scale of the precipitates observed in TEM. In other words, the microstructure that survives the quench is not a frozen snapshot of the high-temperature equilibrium state; it is the product of a brief, frantic burst of short-range diffusion during cooling. Rapid cooling suppresses long-range coarsening but cannot eliminate short-range precipitation. This insight has a sobering corollary for the field: conventional heat treatment followed by water quenching is fundamentally inadequate for validating the equilibrium high-temperature A2/B2 boundaries in this alloy system. Any phase diagram built on quenched microstructures is really a map of quench retention, not equilibrium.

The practical implications ripple outward. For designers of high-temperature components, the work shows that the corrosion-relevant B2 fraction can be tuned by aluminum content and processing temperature, but that a purely B2 microstructure remains out of reach below 57 atomic percent aluminum, well beyond the composition range of structural interest. For computational metallurgists, it provides a valuable benchmark: the trends of CALPHAD are trustworthy, but experiments must account for kinetic retention when validating high-temperature phase fields. The authors suggest the path forward lies in combining direct high-temperature characterization, which observes phases in situ without quenching, with longer equilibration treatments and refined thermodynamic modeling. Until then, the hidden nanoscale architecture of these five-element alloys will keep reminding materials scientists that what you measure at room temperature is not always what existed at 1600 kelvin.

Subject of Research: Phase fraction determination of disordered A2 and ordered B2 body-centered cubic phases in AlxCoCrFeNi high-entropy alloys after elevated-temperature heat treatment and quenching

Article Title: Determination of the A2/B2 BCC phase fraction in AlxCoCrFeNi (1 ≤ x ≤ 4) at elevated temperatures

Article References: Aizenshtein, M., Salhov, S., Strumza, E., Fadel, D., & Hayun, S. (2026). Determination of the A2/B2 BCC phase fraction in AlxCoCrFeNi (1 ≤ x ≤ 4) at elevated temperatures. Journal of Materials Science: Metallurgy, 1(1), Article 22. https://doi.org/10.1007/s44492-026-00023-z

Image Credits: AI Generated

DOI: 10.1007/s44492-026-00023-z

Keywords: high-entropy alloys, AlxCoCrFeNi, A2 phase, B2 phase, BCC, X-ray diffraction, transmission electron microscopy, CALPHAD, quenching, phase boundaries, corrosion resistance, thermodynamics