d-SEAMS: Deferred Structural Elucidation Analysis for Molecular Simulations

Area of Science:

• Computational physics and chemistry
• Materials science
• Chemical engineering

Background:

• Accurate structural analysis is crucial for computational studies of water nucleation and supercooled states.
• Understanding molecular mechanisms requires analyzing evolving ice-like structures, which is computationally intensive.

Purpose of the Study:

• To introduce d-SEAMS, a free, open-source postprocessing engine for analyzing molecular dynamics trajectories.
• To enable qualitative classification of ice structures in confined and bulk systems.
• To provide user-friendly and extensible tools for advanced ice structure analysis.

Main Methods:

• Implementation of recent algorithms for confined ice structure determination.
• Application of topological network criteria for bulk ice structure analysis.
• Proposal and validation of a novel order parameter for quasi-one-dimensional ice building blocks.
• Development of a customizable code architecture using nix and a YAML-Lua scripting pipeline.

Main Results:

• d-SEAMS successfully classifies ice structures in various systems, including confined and bulk water.
• The software integrates novel algorithms and a new order parameter for enhanced structural identification.
• Demonstrated analysis of heterogeneous ice nucleation on a β-AgI surface, homogeneous nucleation, and ice nanotube freezing.

Conclusions:

• d-SEAMS offers a powerful, flexible, and user-friendly solution for analyzing complex ice nucleation and structure formation in molecular dynamics simulations.
• The tool facilitates immediate visual insights and quantitative metrics, aiding researchers in understanding water's phase transitions.
• d-SEAMS advances the field by providing accessible and state-of-the-art methods for ice structure determination.