Achieving DFT accuracy with a machine-learning interatomic potential: thermomechanics and defects in bcc ferromagnetic iron

By D. Dragoni, T. Daff, Gábor Csányi, Nicola Marzari. 2017

224 citations recorded by the source corpus.

We show that the Gaussian Approximation Potential machine learning framework can describe complex magnetic potential energy surfaces, taking ferromagnetic iron as a paradigmatic challenging case. The training database includes total energies, forces, and stresses obtained from density-functional theory in the generalized-gradient approximation, and comprises approximately 150,000 local atomic environments, ranging from pristine and defected bulk configurations to surfaces and generalized stacking faults with different crystallographic orientations. We find the structural, vibrational and thermodynamic properties of the GAP model to be in excellent agreement with those obtained directly from first-principles electronic-structure calculations. There is good transferability to quantities, such as Peierls energy barriers, which are determined to a large extent by atomic configurations that were not part of the training set. We observe the benefit and the need of using highly converged electronic-structure calculations to sample a target potential energy surface. The end result is a systematically improvable potential that can achieve the same accuracy of density-functional theory calculations, but at a fraction of the computational cost.

Subfields: Materials Chemistry, Metals and Alloys

View Paper: https://www.semanticscholar.org/paper/5dfd89e4e11ffadc19f5c97003e08f4f032d1d8f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Materials Chemistry

Citations in the corpus

Machine Learning for High-entropy Alloys: Progress, Challenges and Opportunities (2022, background): “calculated the GSFEs in BCC iron with a GAP model trained with 150,000 local atomic environments, and the results are in excellent agreement with that of DFT [178].”
The Abinit project: Impact, environment and recent developments (2020, background): “In order to overcome this limitation, many efforts have been devoted to the development of so-called “second-principles” methods, which use first-principles data to build effective potentials [135, 136, 137, 138, 139, 140, 141, 142].”
Machine Learning a General-Purpose Interatomic Potential for Silicon (2018, methodology): “This representation was later simplified to the SOAP representation and kernel [7] and applied to tungsten [67], amorphous carbon [11,68], iron [69], graphene [12], and boron [70].”