Nested sampling in the canonical ensemble: direct calculation of the partition function from NVT trajectories

Area of Science:

• Computational Chemistry
• Statistical Mechanics
• Molecular Modeling

Background:

• Nested sampling is a powerful technique for computing partition functions in molecular systems.
• Current nested sampling methods utilize non-standard phase space sampling, limiting integration with molecular dynamics (MD) simulations.

Purpose of the Study:

• To develop a method for directly computing partition functions using standard canonical (NVT) ensemble sampling.
• To enable the use of readily available molecular dynamics and Monte Carlo codes for nested sampling.

Main Methods:

• The proposed method, nested NVT sampling, involves sequentially selecting simulation temperatures from high to low using histogram reweighting.
• A harmonic umbrella biasing potential is employed to enhance phase space exploration when necessary.
• The procedure also details the calculation of observable expectation values and free energy profiles.

Main Results:

• Demonstrates the direct computation of partition functions using standard NVT ensemble simulations.
• Successfully integrates histogram reweighting and umbrella biasing for efficient sampling.
• Provides a framework for calculating expectation values and free energy landscapes.

Conclusions:

• Nested NVT sampling offers a practical alternative to traditional nested sampling, leveraging standard simulation techniques.
• The algorithm's compatibility with existing codes facilitates broader adoption of nested sampling in computational chemistry.
• This advancement expands the accessibility of partition function calculations for complex molecular systems.