|
|
|
# MC3D: The Materials Cloud FAIR and full-provenance materials database
|
|
|
|
|
|
|
|
|
|
|
|
`Name` <be> Dr. Michail Minotakis <br>
|
|
|
|
`Affiliation`<be> PSI Center for Scientific Computing, Theory and Data, 5232 Villigen PSI, Switzerland
|
|
|
|
|
|
|
|
## Abstract
|
|
|
|
|
|
|
|
Highly curated databases of materials and of materials properties have become an increasingly precious resource for a variety of applications, from property prediction with machine learning methods to material screening through state-of-the-art first-principles simulations. In this talk, I will present our efforts in creating the first fully provenance-based database of (bulk, 3D) materials structures and properties. An important feature is that this database aims first to curate and collect experimentally known materials - given the challenges of predicting reliably thermodynamical stability from first-principles. So, it starts from structures extracted from three major (and mostly, but not only, experimental) databases: the Pauling File (MPDS)1, the Inorganic Crystal Structure Database (ICSD) [Ber87], and the Crystallography Open Database (COD) [Gra09]. Following a curation process to remove non-stoichiometric compounds, duplicates, incomplete entries, theoretically predicted structures, and high-pressure or high-temperature experiments, the resulting 72,609 unique stoichiometric compounds are refined using density-functional theory (DFT) calculations at the PBEsol level, as implemented in Quantum ESPRESSO [Gia17,Gia20] complemented by the open-source SIRIUS library2 for efficient use of acceleration and GPU hardware.
|
|
|
|
|
|
|
|
All stages of the workflow, from curating the database to retrieving and storing the results, are orchestrated by the AiiDA3 materials informatics infrastructure. This ensures that the entire provenance of the simulations is preserved, resulting in a database that maintains complete traceability. The data is then made openly available on the Materials Cloud4. In this presentation, I will showcase our protocols and their validation, highlighting the use of AiiDA's advanced automation and error-handling features to create robust workflows for electronic structure simulations, as well as mention some of the current applications for which the database is already being used.
|
|
|
|
|
|
|
|
## Bibliography
|
|
|
|
- [Ber87] G. Bergerhoff, I. D. Brown, and F. Allen, Crystallographic databases, Int. Union Crystallogr., Chester 360, 77-95 (1987).
|
|
|
|
- [Gra09] S. Gražulis et al., J. Appl. Crystallogr. 42, 4, 726-729 (2009).
|
|
|
|
- [Gia17] P. Giannozzi, et al., J. Phys.: Condens. Matter, 29, 465901 (2017).
|
|
|
|
- [Gia20] P. Giannozzi, et al., J. Chem. Phys., 152, 154105 (2020).
|
|
|
|
1. https://mpds.io/#start
|
|
|
|
2. https://github.com/electronic-structure/SIRIUS
|
|
|
|
3. http://aiida.net
|
|
|
|
4. http://mc3d.materialscloud.org |
|
|
\ No newline at end of file |
