Accurate, Affordable and Unsupervised: Analytical F12 Gradients Driven by Generalized Internal Coordinates

Area of Science:

• Quantum Chemistry
• Molecular Spectroscopy
• Computational Chemistry

Background:

• Accurate quantum-chemical predictions are crucial for molecular sciences, including spectroscopy.
• Current benchmark-level methods are computationally expensive for realistic molecule sizes.
• There is a need for affordable, accurate computational methods.

Purpose of the Study:

• To present a modular implementation of the Pisa Composite Schemes (PCS).
• To enable accurate and affordable quantum-chemical predictions for medium-sized and complex molecules.
• To facilitate routine spectroscopic simulations with quantified uncertainty.

Main Methods:

• Modular implementation of Pisa Composite Schemes (PCS) with analytical gradients.
• Hierarchical combination of PCS variants for efficient geometry optimizations.
• Multilayer ONIOM (Our Own N-layered Integrated molecular Orbital and molecular Mechanics) approach combining high-accuracy and low-cost DFT (Density Functional Theory) methods.

Main Results:

• Achieved near-spectroscopic accuracy for medium-sized molecules.
• Successfully simulated challenging systems like CN-substituted PAHs and transition states.
• Demonstrated robust optimizations and efficient frequency calculations.

Conclusions:

• The PCS platform offers a user-friendly and freely available solution for accurate molecular simulations.
• This strategy significantly reduces computational cost while maintaining high accuracy.
• Paves the way for routine, spectroscopically accurate simulations of complex systems.