Automated Construction of Quantum-Classical Hybrid Models

Area of Science:

• Computational Chemistry
• Quantum Mechanics
• Molecular Modeling

Background:

• Developing accurate and efficient computational models for molecular systems is crucial.
• Existing hybrid quantum mechanics/molecular mechanics (QM/MM) methods often require manual parameterization and fixed QM regions.
• Self-parametrizing system-focused atomistic models (SFAMs) offer a parameter-free approach derived from quantum mechanics.

Purpose of the Study:

• To present a protocol for the fully automated construction of QM-classical hybrid models.
• To extend the SFAM approach for seamless integration into QM/MM frameworks.
• To enable automated determination of the QM region size and composition for accurate force calculations.

Main Methods:

• Extension of the self-parametrizing system-focused atomistic models (SFAM) approach.
• Automated evaluation and construction of QM regions based on first-principles calculations.
• Implementation of a local reparametrization scheme for on-the-fly parameter generation.
• System-focused, quantum mechanically derived parametrization for the classical region.

Main Results:

• Achieved fully automated construction of QM/SFAM hybrid models.
• Demonstrated accurate description of atomic forces within the QM region through automated QM region definition.
• Enabled convenient redefinition of the QM region during molecular dynamics simulations.
• Eliminated dependence on pre-existing parameters for the classical model component.

Conclusions:

• The QM/SFAM approach provides a high-fidelity and fully automated method for QM-classical hybrid modeling.
• The protocol facilitates efficient molecular exploration by allowing dynamic QM region adjustments.
• This method significantly reduces the manual effort and expertise required for setting up complex simulations.