Machine learning interatomic potential for the low-modulus Ti-Nb-Zr alloys in the vicinity of dynamical instability

TL;DR

This study develops a machine learning interatomic potential to simulate the elastic properties of multicomponent Ti-Nb-Zr alloys at finite temperatures, revealing invar effects, non-linear concentration dependence, and anisotropic elastic behavior near instability points.

Contribution

The paper introduces a machine learning interatomic potential capable of accurately modeling finite temperature elastic properties of complex Ti-Nb-Zr alloys, capturing non-linear and anisotropic effects near instability.

Findings

Predicts invar effect over a wide temperature range.

Shows non-linear concentration dependence of elastic moduli near instability.

Reveals strong anisotropy in Young's modulus for alloy design.

Abstract

Traditionally, alloying and thermal treatment are considered as the main tools for design of new materials. Application of first-principles simulations can significantly accelerate the process of materials design, however, to account for both, multicomponent chemical disorder and finite temperature effects in theoretical simulations is a challenging task. In this work we have trained machine learning interatomic potential to effectively simulate finite temperature elastic properties of multicomponent \beta-Ti94-xNbxZr6 alloys. Our simulations predict the presence of the elinvar effect for the wide range of temperatures. Importantly, we predict that in a vicinity of dynamical and mechanical instability, the \beta-Ti94-xNbxZr6 alloys demonstrate strongly non-linear concentration-dependence of elastic moduli, which leads to low values of moduli comparable to that of human bone. Moreover, these alloys demonstrate a strong anisotropy of directional Young's modulus which can be helpful for microstructure tailoring and design of materials with desired elastic properties.