Adaptive-Partitioning Multilayer Dynamics Simulations: 1. On-the-Fly Switch between Two Quantum Levels of Theory

Area of Science:

• Computational Chemistry
• Quantum Mechanics/Molecular Mechanics (QM/MM)

Background:

• Existing two-layer permuted adaptive-partitioning QM/MM methods reclassify atoms dynamically during simulations.
• Generalizing these methods to multiple levels of theory is crucial for complex molecular simulations.

Purpose of the Study:

• To develop and formulate novel multilayer adaptive-partitioning algorithms.
• To enable simulations utilizing multiple quantum mechanics (QM) levels of theory.
• To achieve smooth energy interpolation between different QM levels (Q1 and Q2).

Main Methods:

• Formulation of two new adaptive-partitioning algorithms: permuted and interpolated.
• The permuted scheme uses a weighted many-body expansion of the potential.
• The interpolated scheme employs alchemical QM calculations with mixed basis sets, Fock matrices, and overlap matrices.

Main Results:

• Successful implementation of two novel multilayer adaptive-partitioning algorithms.
• Demonstrated smooth energy curves when water molecules transition between Q1 and Q2 QM levels in test calculations.
• First-time application of alchemical calculations within QM for adaptive partitioning.

Conclusions:

• The developed multilayer adaptive-partitioning algorithms effectively enable multiple QM levels of theory.
• Both proposed methods provide smooth energy interpolation, enhancing simulation accuracy.
• These advancements offer a more versatile and potentially efficient approach for QM/MM simulations.