These materials belong to a class called high-entropy oxides (HEOs). They’re made from five or more metal elements and are known for their strength, stability, and potential use in electronics, energy storage, and extreme environments. But many promising combinations were thought to be unstable or too difficult to make. By lowering oxygen levels during the synthesis process, researchers were able to “lock in” elements like manganese and iron – elements that normally refuse to cooperate under typical lab conditions.

One of the first materials created, known as J52, used magnesium, cobalt, nickel, manganese, and iron. Normally, manganese and iron would react with oxygen and shift into unstable forms. But with carefully reduced oxygen, the atoms were forced to stay in a stable state – one that formed a durable, functional ceramic.

To scale the idea, researchers turned to machine learning. They screened thousands of possible combinations and identified six more that could work under similar low-oxygen conditions. Each one was successfully synthesized and turned into dense ceramic pellets.

The implications are big. This thermodynamic-based method opens new doors in material design – especially for applications that demand durability, like batteries, sensors, and spacecraft. By understanding and controlling something as fundamental as oxygen, scientists are now building materials we once thought were out of reach.

Read the study: “Thermodynamics-inspired high-entropy oxide synthesis.” Nature Communications, 2025.

Source: Thermodynamics-inspired high-entropy oxide synthesis | Nature Communications