vL27: Benchmark Binding Energies for 27 Nanoscale Noncovalent Complexes

Ka Un Lao published a benchmark dataset of 27 large noncovalent molecular complexes on 2026-05-23. The dataset provides reference binding energies computed with high-accuracy local coupled cluster [CCSD(T)] methods, extrapolated to the complete basis set limit. It is designed for evaluating electronic structure methods, semiempirical approaches, and machine learning potentials for nanoscale systems.

Use Cases

• Benchmarking the accuracy of density functional theory (DFT) methods based on high-fidelity CCSD(T) reference energies.
• Evaluating the performance of machine learning potentials for predicting nanoscale interaction energies.
• Analyzing the role of dispersion and many-body effects in stabilizing large molecular assemblies based on symmetry-adapted perturbation theory (SAPT) analysis mentioned in the description.

Strengths

• Reference energies are computed with a high-accuracy local CCSD(T)/CBS protocol, with validation against canonical CCSD(T).
• Dataset contains 27 complexes with sizes up to 205 atoms, probing nanoscale interaction effects.
• Includes symmetry-adapted perturbation theory (SAPT) analysis revealing interaction character for each system.

Limitations

• Row count and column-level documentation are absent; field semantics must be inferred after download.
• The dataset is small in size (78.7 KB), indicating a limited scope of 27 data points.
• Last updated 2026-05-23 13:34:27; freshness should be verified for the intended application.

Provenance

Source

figshare

Collection Method

Reference binding energies were computed using local coupled cluster [CCSD(T)] extrapolated to the complete basis set limit.

Freshness

Last updated 2026-05-23 13:34:27.