Reflections on one million compounds in the open quantum materials database (OQMD)

Abstract

Density functional theory has been widely applied in modern materials discovery and many materials databases, including the Open Quantum Materials Database, contain large collections of calculated DFT properties of experimentally known crystal structures and hypothetical predicted compounds. Since the beginning of the OQMD in late 2010, over one million compounds have now been calculated and stored in the database, which is constantly used by worldwide researchers in advancing materials studies. The growth of the OQMD depends on project-based high-throughput DFT calculations, including structure-based projects, property-based projects, and most recently, machine-learning-based projects. Another major goal of the OQMD is to ensure the openness of its materials data to the public and the OQMD developers are constantly working with other materials databases to reach a universal querying protocol in support of the FAIR data principles.

Stability and Equilibrium Structures of Unknown Ternary Metal Oxides Explored by Machine-Learned Potentials

“…Interestingly, we find that AuYO 3 , AuInO 3 , and AuTlO 3 are in the same prototype, as do AuNbO 4 and AuTaO 4 , which are not reported in the AFLOW library (see Figure S5). This means that the majority of the newly identified stable phases cannot be obtained through the data-mining approach. ,, For example, in the composition of Y 2 HgO 4 (see Figure d), the lowest-energy structure in the OQMD has an E hull of −18.7 meV atom –1 . In contrast, SPINNER identifies a lower-energy structure with an E hull of −26.0 meV atom –1 , which does not belong to any known prototype.…”

“…Computational databases such as OQMD, AFLOW, and MP provide millions of theoretical structures, most of which are obtained through ion substitution into known compounds or prototypes. − In specific, the theoretical structures in the MP are primarily generated by data-mined ion substitution method, and structures in the OQMD and AFLOW databases are largely predicted by a high-throughput structure generation method using specific types of prototypes. , Such a data-mining approach is highly efficient, but it risks overlooking materials with unknown prototypes. In contrast, the present direct approach is computationally more demanding but allows exploration of the structure space without any constraints.…”

“…The previous section contrasts two approaches, direct versus data-mining ,, in predicting crystal structures. The direct search outperforms the data-mining approach for a specific composition, indicating that the current library of prototypes is still insufficient.…”

Oxygen Vacancy Formation Energy in Metal Oxides: High-Throughput Computational Studies and Machine-Learning Predictions

“…The authors identified over 250 stable (within 25 meV/atom of the convex hull) compounds, with the rare earth and alkaline earth metals predominantly occupying the A site and transition metals the B site. In a recent work, 28 we updated the data set of stable ABO 3 compounds following the addition of numerous new compounds to the OQMD (now containing over 1 million entries 62 ) and computed the vacancy formation energy of the lowest energy structure at each stable ABO 3 composition, highlighting a strong dependence of ΔE vf on the B site cation. We illustrate in Figure 2 the results of the vacancy formation energy calculations of perovskite oxides, plotted against ΔE stab to highlight the previously mentioned 25 meV/atom stability cutoff.…”

Machine-Learning-Assisted Understanding of Polymer Nanocomposites Composition–Property Relationship: A Case Study of NanoMine Database

“…In addition, the National Institute of Standards and Technology (NIST) maintains the NIST Materials Resource Registry (MRR) that allows materials resource registration, helping to integrate data projects across the materials research community . Currently, most existing data infrastructures host computational data obtained by either first-principles calculations (e.g., AFLOW, , Open Quantum Materials Database, − NOMAD CoE, − Materials Project, , etc.) or atomistic simulations (e.g., Khazana, , Computational Materials Repository, − etc.)…”