HomeCasesComputational prediction of ternary phase diagram

Computational prediction of ternary phase diagram

Introduction

While there are a vast number of possible configurations for crystal structures, only energetically stable ones can exist in the real world.
Therefore, it is crucial to accurately evaluate the relative energy order of various crystal structures for computational materials discovery. Electronic structure calculations have been used in the past, but the computational cost makes it difficult to apply to large-scale screenings.
In this case study, to evaluate the accuracy of PFP, we drew phase diagrams with respect to composition ratios at 0 K using both PFP and Density Functional Theory (DFT).

Calculation models and methods

Phase diagrams can be obtained by calculating the formation energies of all candidate crystal structures and plotting them by composition ratio v.s. energy space.

The phase stability can be evaluated by constructing a convex hull connecting the lowest energy points. The points that make up the convex hull represent stable structures, while points located above the convex hull represent metastable structures. (Metastable structures are not energetically stable, and are decomposed to the stable structures on the convex hull.)

For ternary materials, the composition ratio has two degrees of freedom, which corresponds to a three-dimensional space. For more than three elements, the space becomes even higher-dimensional. In either case, stable and metastable structures can be distinguished by constructing the convex hull.

Results and Discussions

The upper figures show phase diagrams of the ternary Ni-Ga-Nd system calculated using PFP and DFT, respectively. The black points correspond to structures on the convex hull. It clearly shows that PFP well reproduce the  DFT phase diagrams.

The error of PFP in the calculations of phase diagrams are evaluated for all possible combinations of elements up to ternary systems. The lower figure show the energy difference between the metastable structures and the convex hull (energy above hull) (The figure is a log-log plot). The mean absolute error of PPF with respect to DFT was 0.0284 eV/atom.

Calculation Condition

References

[1] Materials Project. https://next-gen.materialsproject.org/
Contact Us
Go to List of Cases
Features
Features

Matlantis: 3 Key Features

Matlantis supports companies exploring innovative materials for a sustainable future.

Versatility

汎用性 / Versatile イメージ
Supports a wide range of elements and structures

Matlantis can simulate properties of molecules and crystal systems, including unknown materials. Any combinations of 72 elements are currently supported, and more elements will be added.

Speed

高速 / High Speed イメージ
Over 10,000x faster than conventional methods

The atomistic simulation tasks that take hours to months using density functional theory (DFT) on a high-performance computer can be finished in only a few seconds using Matlantis.

User-Friendliness

使いやすさ / Easy to Use イメージ
Just open your browser to run simulations

Thanks to the pre-trained deep learning model, physical property calculation library, and high-performance computing environment, no hardware or software installations are required for performing simulations. Unlike conventional machine learning potentials, Matlantis requires no data collection or training by users.

Go to Product Detail