PyAR

PyAR performs global optimization of nanoclusters to identify the global minimum and other low-energy minima by combining recursive trial-geometry generation, Tabu-list filtering, and gradient-based local optimization.

Key Features:

• Recursive Algorithm for Geometry Generation: Constructs an n-sized cluster from geometries of (n-1) clusters using an evolutionary growth strategy to produce diverse trial geometries.
• Tabu Search Integration: Maintains a Tabu list of previously used trial geometries to prevent revisiting similar structures during optimization.
• Comprehensive Energy Analysis: Evaluates relative energy, singlet-triplet energy difference, binding energy, second-order energy difference, and mixing energy to characterize cluster stability.
• Gradient-Based Local Optimization: Applies gradient-based optimization methods to refine trial geometries and identify local minima.
• Versatile Application Across Various Clusters: Applied to homometallic clusters (Pd, Pt, Au, Al) and multi-metallic systems such as Ru-Pt, Au-Pt, and Ag-Au-Pt.

Scientific Applications:

• Structural and Energetic Characterization: Identification and comparison of global and low-energy minima and calculation of energetic descriptors for nanocluster stability.
• Catalysis Research: Analysis of binding energies and singlet-triplet energy differences relevant to catalytic activity in metal and multi-metal clusters.
• Materials and Electronic Properties: Exploration of stable geometries and energetic landscapes for applications in electronics and materials science.

Methodology:

Uses an evolutionary growth strategy to generate trial geometries by building n-sized clusters from (n-1) clusters, employs a Tabu list to avoid redundant trials, and applies gradient-based local optimization to refine structures and locate minima.