Machine learning potential as a guide for eutectic in ultra-refractory multicomponent ceramics

TL;DR

This paper introduces a machine learning-based method to predict eutectic points in ultra-refractory alloys, overcoming experimental challenges due to high melting points.

Contribution

It presents a novel AI-driven criterion for eutectic determination applicable to ultra-refractory systems, verified on the Ti-B-C system.

Findings

Machine learning potential achieves ab initio accuracy.

Method operates effectively in liquid phase without crystalline data.

Successfully predicts eutectic points in Ti-B-C system.

Abstract

The experimental determination of eutectic points is a long-established and widely used technique, but it is generally only practical for systems with relatively low melting points. Many modern, promising materials, however, are ultra-refractory, with melting points exceeding 3000 K. For these systems, conventional melting experiments become prohibitively expensive and technically challenging. Advanced AI modeling can serve as a powerful precursor to guide successful experimentation in such cases. This work proposes a novel criterion for determining the eutectic point concentration in ultra-refractory alloys. The approach is verified using the Ti-B-C system - the most thoroughly studied three-component refractory system to date. The core of the algorithm is a machine-learning interatomic potential, based on a neural network, which achieves accuracy comparable to ab initio methods. Crucially, the algorithm operates effectively in the liquid phase, eliminating the need for information about the solid alloy's crystalline structure to estimate eutectic points.