Polarizable Atomic Multipole-based Molecular Mechanics for Organic Molecules.

Area of Science:

• * Computational chemistry and molecular modeling.
• * Development of empirical force fields for organic molecules.

Background:

• * Accurate molecular modeling requires robust force fields that capture interatomic interactions.
• * Existing force fields may not sufficiently represent electrostatic interactions, particularly for diverse organic functional groups.

Purpose of the Study:

• * To develop and validate an empirical potential (AMOEBA force field) for small organic molecules.
• * To accurately model permanent and induced atomic dipoles and multipole moments.

Main Methods:

• * Derived permanent atomic multipole moments (up to quadrupole) from ab initio calculations.
• * Fit van der Waals parameters using homodimer QM energies, experimental densities, and heats of vaporization.
• * Validated the force field through gas-phase heterodimer calculations and molecular dynamics simulations.

Main Results:

• * Achieved excellent agreement (within 0.4 kcal/mol) for hydrogen bonding energies of 32 homo- and heterodimers.
• * Hydrogen bond distances showed minimal deviation (<0.06 Å) from QM optimized geometries.
• * Computed liquid properties (self-diffusion, dielectric constants) and solvation free energies matched experimental data.

Conclusions:

• * The AMOEBA force field demonstrates high accuracy for a wide range of organic molecules and phases.
• * The developed electrostatic model and parameterization protocol enable future extensions to more complex systems.