Minimal Basis Iterative Stockholder Decomposition with Ellipsoidal Atoms

Area of Science:

• Quantum Chemistry
• Computational Chemistry
• Molecular Modeling

Background:

• The minimal basis iterative Stockholder (MBIS) method decomposes molecular electron densities into atomic contributions.
• Current MBIS methods primarily use spherical atomic basins.

Purpose of the Study:

• To extend the MBIS decomposition to ellipsoidal atomic basins.
• To evaluate the impact of ellipsoidal basins on the accuracy of atomic multipole moments and electrostatic potentials.
• To explore the potential of ellipsoidal decomposition for deriving anisotropic atomic parameters.

Main Methods:

• Extension of the MBIS decomposition algorithm to accommodate ellipsoidal atomic basins.
• Calculation and comparison of atomic multipole moments and electrostatic potentials derived from spherical versus ellipsoidal MBIS decompositions.
• Constraining the ellipsoidal decomposition to exactly reproduce molecular multipole moments up to the hexadecapole level.

Main Results:

• Ellipsoidal atomic basins did not systematically improve the reproduction of molecular multipole moments and electrostatic potentials compared to spherical basins.
• Constraining the decomposition to reproduce molecular multipole moments slightly enhanced electrostatic potential reproduction.
• The ellipsoidal decomposition yielded atomic parameters describing the anisotropic decay of electron density.

Conclusions:

• While not consistently improving accuracy, ellipsoidal MBIS basins offer a more flexible decomposition.
• The derived anisotropic atomic parameters show promise for applications in force fields and quantum crystallography.
• Further research may refine ellipsoidal basin definitions for enhanced accuracy in electron density decomposition.