Efficient Generation of Permutationally Invariant Potential Energy Surfaces for Large Molecules

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Molecular Modeling

Background:

• Accurate fitting of potential energy surfaces (PES) is crucial for molecular simulations.
• Existing methods for generating polynomial bases can be computationally intensive and may lack efficiency for large molecules.

Purpose of the Study:

• To develop an efficient method for generating fragmented, permutationally invariant polynomial basis sets for fitting molecular energies and gradients.
• To improve upon previous methods by avoiding polynomial repetition and accelerating gradient evaluations.

Main Methods:

• Fragmentation of large molecules into smaller units.
• Generation of permutationally invariant polynomial basis sets.
• Fitting of electronic energies and gradients using the generated basis.

Main Results:

• The method successfully generates fragmented, permutationally invariant polynomial bases for large molecules.
• The approach avoids repetition of polynomials, leading to a more efficient fitting basis set.
• Gradient evaluations are sped up while maintaining the accuracy of the potential energy surface.

Conclusions:

• The described method offers an efficient and accurate way to generate polynomial basis sets for large molecules.
• This technique is valuable for computational chemistry, particularly in molecular dynamics and reaction pathway studies.
• The method demonstrated on N-methylacetamide and glycine shows its applicability to biologically relevant molecules.