Improving the Efficiency of Electrostatic Embedding Using the Fast Multipole Method.

Area of Science:

• Computational Chemistry
• Theoretical Chemistry
• Quantum Mechanics

Background:

• Embedded-cluster model (ECM) calculations are crucial for studying complex molecular systems.
• Existing ECM methods face computational challenges, particularly in evaluating electrostatic interactions.
• The ORCA quantum chemistry package is widely used for electronic structure calculations.

Purpose of the Study:

• To enhance the computational efficiency of the embedded-cluster model (ECM) within the ORCA software.
• To implement and optimize the fast multipole method (FMM) for ECM calculations.
• To ensure the accuracy and versatility of the FMM implementation for various molecular systems.

Main Methods:

• Implemented the fast multipole method (FMM) using state-of-the-art algorithms within the ORCA package.
• Developed efficient and accurate methods for handling atomic orbital shell pairs.
• Decomposed near-field and far-field electrostatic interactions effectively.
• Integrated the FMM into the embedded-cluster model (ECM) framework.
• Ensured full parallelization of the implemented code.

Main Results:

• Achieved a significant acceleration in the evaluation of electrostatic potential integrals, ranging from one to two orders of magnitude.
• Maintained excellent accuracy, consistently exceeding chemical accuracy (1 kcal/mol).
• Demonstrated the versatility of the implementation for molecular systems using QM/MM approaches.
• The optimized FMM-based ECM is available in ORCA 6.0.

Conclusions:

• The FMM implementation provides a substantial speedup for ECM calculations in ORCA.
• The method offers a robust and accurate approach for large-scale computational chemistry.
• This advancement facilitates more complex and efficient simulations in quantum chemistry and QM/MM studies.