🔬 materials science

Designing sustainable high-entropy alloys directly from mixed oxides: A thermodynamic perspective

This paper presents a CALPHAD-based thermodynamic framework that defines the specific composition, temperature, and H $_2$ /H $_2$ O ratio conditions required for the complete metallization of CoCrFeMnNi-based mixed oxides, thereby enabling the sustainable direct production of high-entropy alloys via hydrogen-based reduction.

Original authors: Albert Linda, Shiv Shankar, Dierk Raabe

Published 2026-10-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Albert Linda, Shiv Shankar, Dierk Raabe

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Making metal has always been a dirty business. For centuries, the industry has relied on burning fossil fuels to strip oxygen away from rocks, a process that leaves behind a massive cloud of carbon dioxide. As the world looks for ways to build a cleaner future, scientists are turning to hydrogen as a replacement. When hydrogen is used to reduce metal oxides, the only byproduct is water, not carbon. This method, known as hydrogen-based direct reduction, promises a way to create sustainable alloys without the heavy carbon footprint of traditional smelting. However, turning this promise into reality is not as simple as swapping one gas for another. The challenge lies in the complex chemistry of the rocks themselves. Most useful metals are not found as single, pure elements but are locked together in mixed oxides containing several different metals at once. When these mixtures are heated, the metals interact in ways that change how easily they can be freed from their oxygen bonds. If the conditions are not perfect, the process stalls, leaving behind a stubborn, unreduced residue that ruins the final product.

A team of researchers at the Max Planck Institute for Sustainable Materials has developed a new way to predict exactly when and how these mixed oxides will break down. Instead of trying to test every possible combination of metals in a furnace, which would take an impossible amount of time, they built a detailed thermodynamic map using advanced computer modeling. This map acts as a guide for engineers, showing the precise balance of hydrogen and water vapor needed to turn a specific mixture of metal oxides into a pure alloy. The researchers focused on a group of five metals—cobalt, chromium, iron, manganese, and nickel—which are the building blocks for many modern high-performance alloys, including the famous high-entropy alloys used in everything from jet engines to medical devices. By simulating the behavior of these metals across a wide range of temperatures and gas mixtures, they uncovered the hidden rules that govern whether a mixture will fully reduce or get stuck halfway.

The study reveals that the difficulty of reducing a metal mixture depends entirely on which metals are present and in what proportions. The researchers found that adding certain elements can make the reduction process significantly harder, while others make it easier. Specifically, manganese and chromium tend to lock the oxygen in place, creating a very stable oxide that is difficult to break apart. In contrast, cobalt, iron, and nickel generally make the oxides less stable, allowing them to be reduced with less energy and a less pure hydrogen supply. The most surprising discovery, however, is that the effect of adding a metal is not fixed. Depending on the starting recipe of the mixture, adding the same amount of manganese could either help or hinder the process. This happens because the metals rearrange themselves as they heat up, forming different crystal structures that change the chemical activity of the remaining oxygen.

One of the key insights from the work is that the number of steps a mixture takes to turn into metal does not tell you how difficult the job is. A mixture might pass through many intermediate stages, changing from one type of oxide to another several times, yet still be easy to reduce completely if the final step is favorable. Conversely, a mixture that seems to change very little might get stuck at the very last step, leaving a thin layer of stubborn oxide that refuses to disappear. The researchers identified that for mixtures rich in manganese, the final barrier is often a rock-salt type structure that holds onto oxygen tightly. For chromium-rich mixtures, the barrier is a different, very stable structure that requires extremely dry hydrogen to break. In contrast, mixtures of cobalt, iron, and nickel are much more forgiving, often reducing completely even with a moderate amount of water vapor in the gas stream.

The team also mapped out the specific "window" of conditions required for success. They calculated that for some difficult mixtures, the hydrogen gas must be incredibly dry, with water vapor levels as low as a few parts per million, to achieve full reduction. For easier mixtures, the gas can be much wetter. This finding is crucial for industrial design because it means engineers cannot use a single, standard recipe for all alloys. Instead, they must tailor the gas mixture to the specific metal composition they are trying to produce. The study also showed that simply heating the mixture higher does not always solve the problem. While higher temperatures generally help, they cannot overcome the fundamental stability of the most stubborn oxides if the gas mixture is not dry enough.

By linking the stability of the final oxide residue directly to the ratio of hydrogen to water vapor, this framework provides a clear set of guidelines for designing sustainable alloys. It allows scientists to choose a precursor mixture that is naturally easier to reduce, or to adjust the gas conditions to match a difficult mixture. The work moves beyond simple trial and error, offering a predictive tool that can help the metal industry transition to green production methods. The researchers emphasize that while their model predicts the thermodynamic limits—the absolute best-case scenario for what is possible—it does not account for the speed of the reaction or the physical movement of gases inside a furnace. Those are kinetic factors that will need to be addressed in future work. However, by establishing the fundamental boundaries of what is chemically possible, this study provides the essential foundation for building a new generation of metals that are both high-performing and environmentally friendly.

Technical Summary: Designing Sustainable High-Entropy Alloys Directly from Mixed Oxides

Problem Statement
The transition to sustainable metal production requires moving away from fossil-based reductants that generate significant carbon emissions. Hydrogen-based direct reduction (HyDR) of mixed oxide precursors offers a promising alternative, capable of combining reduction and alloying into a single step. However, a critical gap exists in understanding the thermodynamic limits of reducing multicomponent oxide mixtures. Unlike single-oxide reduction, the reduction of mixed oxides involves coupled phase transformations where the formation of metallic and oxide solid solutions alters the chemical activity of both reduced and unreduced phases. Consequently, the thermodynamic stability of the final residual oxide—and thus the required $H_2/H_2O$ ratio for complete metallization—cannot be deduced from the stability of pure individual oxides. Current knowledge lacks a predictive framework to establish the composition–temperature–gas ratio window necessary for designing sustainable multicomponent alloys, such as CoCrFeMnNi high-entropy alloys, directly from mixed oxide precursors.

Methodology
The authors developed a thermodynamic framework based on the CALPHAD (CALculation of PHAse Diagrams) approach to map the multidimensional design space of Co–Cr–Fe–Mn–Ni oxide mixtures.

Computational Setup: Calculations were performed using Thermo-Calc 2025a with the Metal Oxide Solutions (TCOX10) database. The study assumed a single homogeneous FCC solid solution for the metallic phase, consistent with the equilibrium structure of the equiatomic Cantor alloy above ~700 °C.
Sampling: A non-uniform grid sampled 3,526 unique compositions across the quinary system at 21 temperatures (500–1500 °C).
Boundary Extraction: An oxygen chemical potential ( $\mu(O)$ ) scanning strategy was employed to identify phase boundaries. The "onset" boundary (index 1) was defined as the first metal-to-oxide transition encountered when increasing $\mu(O)$ from the metallic side. This boundary represents the final oxide that must be reduced to achieve complete metallization.
Thermodynamic Metrics: The study calculated the onset Gibbs free energy ( $\Delta G_{onset}$ ) and converted this into the minimum required $H_2/H_2O$ ratio ( $r_{min}$ ) for reduction. The reduction driving force ( $\Delta G_{rf}$ ) was defined as the difference between the oxide stability line and the $H_2/H_2O$ equilibrium line.
Statistical Analysis: To quantify composition sensitivity, the authors employed standardized ordinary least-squares (OLS) regression and histogram gradient boosting models. These statistical tools analyzed how adding specific elements (Co, Cr, Fe, Mn, Ni) to binary hosts affects oxide stability, accounting for the constraint that metal fractions sum to unity.

Key Results

Composition-Dependent Reduction Windows: The study establishes that the $H_2/H_2O$ ratio required for complete metallization varies drastically with composition. Co–Fe–Ni mixtures are readily reducible at moderate ratios (e.g., $r=3$ ), whereas Mn-rich and Cr-rich regions require significantly drier gas (higher $r$ ). For instance, at 1100 °C, the median $r_{min}$ for Co–Fe–Ni is ~0.16, while for Cr–Fe–Mn it is ~2,529.
Elemental Effects on Stability:
- Mn and Cr: Generally increase oxide stability, raising the $H_2/H_2O$ requirement. Mn stabilizes HALITE-type (rock-salt) structures, while Cr stabilizes CORUNDUM-type structures.
- Co, Fe, and Ni: Generally decrease oxide stability, facilitating reduction.
- Non-Universality: The effect of an element is not fixed; it depends on the host composition. For example, Cr can stabilize or destabilize an oxide mixture depending on the Co:Mn ratio, and Co can be destabilizing in some hosts but neutral in others.
Phase Selection and Bottlenecks: The dominant "bottleneck" oxide (the last to reduce) varies by system. Mn-bearing systems are typically controlled by HALITE (MnO-rich), Cr-bearing systems by CORUNDUM (Cr $_2$ O $_3$ -rich), and intermediate systems by SPINEL. The study found that Mn-rich HALITE and Cr-rich CORUNDUM impose persistent residual-oxide boundaries that are difficult to overcome even at high temperatures.
Complexity vs. Difficulty: A key finding is that the complexity of the reduction pathway (number of intermediate phase transitions) does not correlate with the thermodynamic difficulty of complete metallization. A system with few intermediate steps can still be thermodynamically blocked by a highly stable final oxide, while a system with many steps may be fully reducible if the final oxide is unstable.
Temperature Dependence: Increasing temperature generally widens the reduction margin (increases driving force), but it does not eliminate the fundamental stability differences imposed by Mn and Cr.

Significance and Claims
The paper claims to provide the first unified thermodynamic framework that links precursor composition, reduction conditions, and the final equilibrium state for multicomponent oxide mixtures. Its primary contribution is establishing that residual-oxide stability is the central equilibrium criterion for designing alloys via HyDR.

The authors assert that their framework:

Identifies the thermodynamic limits and specific $H_2/H_2O$ windows required for complete metallization of CoCrFeMnNi-based systems.
Demonstrates that precursor composition and reducing environment must be co-designed, as the difficulty of reduction is a coupled function of both variables.
Provides guidelines for selecting compositions that are easier to reduce (e.g., avoiding Mn/Cr-rich bottlenecks or adjusting host ratios) to enable low-temperature alloy formation.
Clarifies that while intermediate phase complexity is a feature of the reduction landscape, it is distinct from the thermodynamic barrier to complete metallization.

The study explicitly notes that its predictions are based on equilibrium thermodynamics and do not account for kinetic factors (e.g., diffusion rates, nucleation, porosity, or precursor history) which influence the actual rate and degree of reduction in practice. The framework serves as a thermodynamic baseline for designing sustainable multicomponent alloys directly from mixed oxide precursors, moving beyond simple superposition of single-oxide reduction reactions.

More like this