English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We propose a scheme for {\it ab initio} configurational sampling inmulticomponent crystalline solids using Behler-Parinello type neural networkpotentials (NNPs) in an unconventional way: the NNPs are trained to predict theenergies of relaxed structures from the perfect lattice with configurationaldisorder instead of the usual way of training to predict energies as functionsof continuous atom coordinates. An active learning scheme is employed to obtaina training set containing configurations of thermodynamic relevance. Thisenables bypassing of the structural relaxation procedure which is necessarywhen applying conventional NNP approaches to the lattice configuration problem.The idea is demonstrated on the calculation of the temperature dependence ofthe degree of A/B site inversion in three spinel oxides, MgAl$_2$O$_4$,ZnAl$_2$O$_4$, and MgGa$_2$O$_4$. The present scheme may serve as analternative to cluster expansion for `difficult' systems, e.g., complex bulk orinterface systems with many components and sublattices that are relevant tomany technological applications today.

Facilitating {\it ab initio} configurational sampling of multicomponent solids using an on-lattice neural network model and active learning