Dresden 2014 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

MM: Fachverband Metall- und Materialphysik

MM 4: Computational Materials Modelling I - High throughput/Material discovery

MM 4.1: Talk

Monday, March 31, 2014, 10:15–10:30, IFW D

An automated first-principles search for solid-state lithium electrolytes — •Riccardo Sabatini and Nicola Marzari — Theory and Simulation of Materials, EPFL (CH)

In recent years much attention has been given to high-throughput techniques for materials' discovery and characterization, a complex challenge where state-of-the-art theoretical advancements meet data analytics and automated job management. We present here one of the first high-throughput projects done with our AiiDA framework (Automated Interactive Infrastructure and Database for Atomistic simulations), an open-source infrastructure built for largely automatic materials' design and discovery. Starting from the ICSD database, we build an automatic workflow to compute lithium-ion diffusion with Car-Parrinello molecular dynamics. The database subset of target lithium-containing materials is also ranked by citations number; in descending order, each structure is first optimized with a ground-state variable-cell relaxation, followed by a microcanonical simulation. Several techniques to ensure the quality of the workflow have also been developed and are discussed.