Hybrid DFT Geometries and Properties for 17k Lanthanoid Complexes─The LnQM Data Set

Area of Science:

• Inorganic Chemistry
• Computational Chemistry
• Materials Science

Background:

• Lanthanoids possess unique properties driving research and industry.
• Existing molecule datasets lack comprehensive comparative data for lanthanoids.
• A gap exists for datasets supporting both quantum chemistry and machine learning.

Purpose of the Study:

• To create a comprehensive, curated dataset of monolanthanoid complexes.
• To facilitate comparative studies on lanthanoid properties, independent of ligand effects.
• To support advancements in quantum chemistry calculations and data science applications.

Main Methods:

• Generated 17,269 monolanthanoid complexes across all 15 lanthanoids in +3 oxidation state.
• Systematically permuted central lanthanoid atoms in ligand structures for comparability.
• Included diverse computational features: optimized geometries, energies, electronic properties, and charge analyses.

Main Results:

• The dataset covers all 15 lanthanoids, various molecular charges, and high spin multiplicities.
• Provided PBE0-D4/def2-SVP optimized geometries and ωB97M-V/def2-SVPD energies.
• Included extensive electronic structure data (HOMO-LUMO, population analyses) and physical properties (dipole moments, polarizabilities).

Conclusions:

• The presented dataset bridges the gap for comparative lanthanoid research.
• It enables independent study of lanthanoid influence on complex properties.
• The openly accessible data serves as a foundation for future lanthanoid investigations.