Elinvar effect in β-Ti simulated by on-the-fly trained moment tensor potential

Abstract

A combination of quantum mechanics calculations with machine learning techniques can lead to a paradigm shift in our ability to predict materials properties from first principles. Here we show that on-the-fly training of an interatomic potential described through moment tensors provides the same accuracy as state-of-the-art ab initio molecular dynamics in predicting high-temperature elastic properties of materials with two orders of magnitude less computational effort. Using the technique, we investigate high-temperature bcc phase of titanium and predict very weak, Elinvar, temperature dependence of its elastic moduli, similar to the behavior of the so-called GUM Ti-based alloys (Sato et al 2003 Science 300 464). Given the fact that GUM alloys have complex chemical compositions and operate at room temperature, Elinvar properties of elemental bcc-Ti observed in the wide temperature interval 1100–1700 K is unique.

Finite-temperature interplay of structural stability, chemical complexity, and elastic properties of bcc multicomponent alloys from ab initio trained machine-learning potentials

“…9(a). Similar effects have been recently reported for bcc Ti [20]. With our ML-based scheme multicomponent alloys can be effectively screened to achieve desired temperature-dependent behavior of elastic properties.…”

“…A temperature-invariant elastic behavior of materials, referred to as elinvar effect, is, e.g., known for multicomponent Nibased alloys [17][18][19]. Recently, the elinvar effect has also been reported for elemental body-centered cubic (bcc) Ti [20]. The inclusion of finite-temperature effects is thus crucial for a reliable prediction of the mechanical properties of materials at elevated temperatures.…”

“…The advent of machine-learning interatomic potentials (MLIPs) [29,30] has made it possible to significantly reduce the computational costs of pure DFT calculations and to access materials properties with near DFT accuracy while preserving the computational efficiency [31] of classical interatomic potentials [20,32,33]. MLIPs have rapidly advanced in the past decade, resulting in a number of approaches [34][35][36][37][38].…”

Machine‐learning‐based interatomic potentials for advanced manufacturing

“…76 The moment tensor potentials (MTPs) are used to predict the high-temperature elastic properties of GUM Ti-based alloys. 77 Some bulk, interface, and defect characteristics are calculated in the range from low temperatures to those close to the melting point of Ti-based alloys. 78 The ML framework is also applicable for high-entropy alloy systems.…”

Accurate prediction of high-temperature elastic constants of Ti0.5Al0.5N random alloy

“…Elastic constants, in general, are used to get insight into the elasto-mechanical behavior of materials [3,4]. However, predicting the elastic constants of materials at elevated temperature corresponding to the operational conditions of the cutting tools, is a challenging computational task, as it requires 1 meV/atom accuracy in energy differences to achieve 1 GPa accuracy for the constants [5]. Therefore, first-principles calculations of the materials elastic constants have recently become a competitive research field.…”