Coupling molecular density functional theory with converged selected configuration interaction methods to study...

Area of Science:

• Computational Chemistry
• Theoretical Chemistry
• Quantum Mechanics

Background:

• Accurately modeling solvent effects is crucial for understanding chemical reactions and molecular properties.
• Existing methods like macroscopic continuum approaches lose molecular-level detail, while sampling-based molecular mechanics can be computationally intensive.

Purpose of the Study:

• To present the first implementation of a QM/MDFT coupling for modeling solvent effects in quantum mechanical calculations.
• To offer a method that retains molecular-level solvent description while remaining computationally competitive.

Main Methods:

• Coupling advanced wavefunction theories with molecular density functional theory (MDFT).
• Incorporating solvent-generated electrostatic potential into the electronic Hamiltonian.
• Deducing solvent charges from solvent particle density obtained via molecular mechanics (MM) Hamiltonian minimization.
• Utilizing MDFT as the MM solver within a QM/MM framework.

Main Results:

• The QM/MDFT framework was successfully applied to compute excited state properties of water and formaldehyde in water at the selected configuration interaction (SCI) level.
• Excitation energies and dipole moments were calculated and compared with experimental and theoretical data.
• Using Hartree-Fock for the solute provided sufficient accuracy for solvent charge prediction in ground states, enabling efficient excited-state SCI calculations.

Conclusions:

• The developed QM/MDFT method offers a computationally efficient and accurate way to model solvent effects at a molecular level.
• This approach overcomes limitations of continuum models and sampling-based methods.
• The findings pave the way for studying more complex molecular systems with solvent effects.