Basin hopping

Official Resources

• Homepage: Method - Implemented in ASE, Wales Group codes, etc.
• Documentation: https://wiki.fysik.dtu.dk/ase/ase/optimize.html#basinhopping (ASE implementation)
• Source Repository: https://gitlab.com/ase/ase
• License: Varies (ASE is LGPL)

Overview

Basin hopping is a global optimization algorithm used to find the global minimum of a potential energy surface. It transforms the energy landscape into a set of basins of attraction and samples them using Monte Carlo moves followed by local minimization. It is highly effective for cluster structure prediction and finding stable configurations of molecules and defects.

Scientific domain: Global optimization, structure prediction, clusters
Target user community: Computational chemists, materials scientists

Theoretical Methods

• Basin-Hopping (BH) / Monte Carlo Minimization (MCM)
• Canonical Monte Carlo sampling
• Local minimization (L-BFGS, CG, etc.)
• Metropolis criterion
• Temperature-dependent sampling

Capabilities (CRITICAL)

• Global optimization of atomic structures
• Prediction of cluster geometries (Lennard-Jones, metallic clusters)
• Surface reconstruction search
• Adsorbate configuration search
• Implemented in: ASE, GMIN (Wales group), LAMMPS (via Python/fix), etc.

Sources: D.J. Wales and J.P.K. Doye, J. Phys. Chem. A 101, 5111 (1997)

Inputs & Outputs

• Input formats: Initial structure, potential/calculator, temperature
• Output data types: Lowest energy structure, trajectory of minima

Interfaces & Ecosystem

• ASE: Standard Python implementation (ase.optimize.BasinHopping)
• Potentials: Works with any calculator (DFT, classical)

Performance Characteristics

• Stochastic, requires many minimization steps
• Embarrassingly parallel (multiple independent runs)
• Efficient for cluster physics

Application Areas

• Nano-clusters
• Protein folding (simplified models)
• Defect searching
• Molecular docking

Community and Support

• Broad usage in chemical physics
• ASE community
• Wales group (Cambridge)

Verification & Sources

Primary sources:

1. Publication: D.J. Wales and J.P.K. Doye, J. Phys. Chem. A 101, 5111 (1997)
2. ASE Documentation: https://wiki.fysik.dtu.dk/ase/

Confidence: VERIFIED

Verification status: ✅ VERIFIED

• Method: STANDARD
• Documentation: AVAILABLE (ASE)
• Applications: Global optimization, structure prediction, clusters