Efficiently Capturing Weak Interactions in ab Initio Molecular Dynamics with on-the-Fly Basis Set Extrapolation.

Area of Science:

• Computational Chemistry
• Molecular Dynamics
• Quantum Chemistry

Background:

• Accurate simulation of conformational changes requires capturing weak interactions.
• Large basis set electronic structure calculations are computationally expensive for ab initio molecular dynamics (AIMD).

Purpose of the Study:

• To develop a computationally efficient method for obtaining AIMD trajectories that agree with large basis set calculations.
• To enable accurate studies of weak interactions and conformational changes in large molecular systems.

Main Methods:

• A novel approach based on molecular fragmentation and range-specific interaction repartitioning.
• Utilizes a generalized ONIOM method with a set-theoretic inclusion-exclusion principle for assembling noncovalent interactions.
• Employs simplex decomposition for efficient calculation of many-body interactions.

Main Results:

• AIMD trajectories were generated with accuracy comparable to large basis sets (e.g., 6-311++G(2df,2pd)) at a fraction of the computational cost.
• The method achieved computational effort similar to smaller basis sets (e.g., 6-31+G(d)) for systems exceeding 100 atoms.
• Accurate conformational stabilization energies and AIMD trajectories (Born-Oppenheimer and Car-Parrinello-type) were obtained efficiently.

Conclusions:

• The developed method significantly reduces the computational cost of AIMD simulations while maintaining high accuracy for weak interactions.
• This approach makes large basis set quality calculations feasible for complex molecular systems, including polypeptides.
• Enables efficient and accurate simulations of conformational dynamics and noncovalent interactions.