Cluster Expansion (CE) is a method that has been used successfully in the past few decades to parameterize and express the configurational dependence of the scalar physical properties. The most widely parameterized physical property is energy computed using first-principles methods, but CE can also be used to parameterize other quantities such as band gap and density of states. In spite of its success and usefulness, CE remains as a niche tool used in a small subfield within the computational materials science, primarily used by specialists. Here, we present a new and versatile implementation of the CE method which is integrated as a part of Atomic Simulation Environment (ASE) open-source package. The code automates the complex set up and construction procedure of cluster expansion while giving the users the flexibility to tweak the settings and to import their own structures and previous calculation results. Recent advancements in cluster expansion such as regularization techniques from machine learning and statistics are implemented in the developed code. The code allows the users to construct cluster expansion on any bulk lattice structure, which makes it useful for a wide range of applications involving complex materials. Two types of materials, a binary metal alloy and a disordered battery cathode material, are analyzed using cluster expansion method to demonstrate the implementation.
|Journal||American Chemical Society. Abstracts of Papers (at the National Meeting)|
|Number of pages||1|
|Publication status||Published - 2019|
|Event||257th ACS National Meeting and Exposition - Orlando, United States|
Duration: 31 Mar 2019 → 4 Apr 2019
|Conference||257th ACS National Meeting and Exposition|
|Period||31/03/2019 → 04/04/2019|