Atomic Structures and Energies for Water Clusters and a Transition-Metal Oxide

Atomic structures of water clusters, bulk water, and rock-salt Li8Mo2Ni7Ti7O32 are provided in XCrySDen format with total energies as meta information. The data was generated via ab initio molecular dynamics simulations using VASP and Turbomole software, as described in the associated research article. The dataset is intended for constructing and applying artificial neural network potentials via the aenet package.

Use Cases

• Training neural network potentials for water systems based on atomic structures and total energies.
• Benchmarking density functional theory methods for transition-metal oxides using provided atomic configurations.
• Developing machine learning models for molecular dynamics that incorporate atomic forces via Taylor expansion, as suggested by the title.

Strengths

• Structures were generated using established ab initio molecular dynamics (AIMD) simulations with specific density functionals and corrections (e.g., revPBE+D3, SCAN).
• Total energies are included as meta information for each atomic structure.
• The data format is compatible with the aenet package for neural network potential construction.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

University of St Andrews

Collection Method

Generated via ab initio molecular dynamics (AIMD) simulations using Vienna Ab Initio Simulation Package (VASP) and Turbomole software.