Analytical Gradient Using Cluster-in-Molecule RI-MP2 Method for the Geometry Optimizations of Large Systems.

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Theoretical Chemistry

Background:

• Accurate calculation of molecular properties is crucial in chemistry.
• Second-order Møller-Plesset perturbation (MP2) theory provides a good balance of accuracy and cost.
• Local MP2 methods aim to reduce the computational scaling of MP2 calculations for large systems.

Purpose of the Study:

• To present an efficient analytical energy gradient algorithm for the cluster-in-molecule resolution-of-identity second-order Møller-Plesset perturbation (CIM-RI-MP2) method.
• To demonstrate the computational efficiency and applicability of the CIM-RI-MP2 gradient algorithm for large molecules.

Main Methods:

• Development of an analytical energy gradient algorithm based on the Lagrange multiplier method.
• Implementation of a cluster-in-molecule (CIM) approach within the resolution-of-identity (RI) approximation.
• Avoidance of solving coupled-perturbed Hartree-Fock (CPHF) equations for the entire system.

Main Results:

• The CIM-RI-MP2 gradient algorithm achieves significantly lower computational cost compared to other local MP2 methods.
• Benchmark calculations on molecules up to 312 atoms confirm the general applicability of the algorithm.
• Optimized structure of a 244-atom molecule shows good agreement with experimental crystal structure.
• A large-scale calculation for a 972-atom system with 9612 basis functions was completed in 48 hours on 25 nodes (600 CPU cores).

Conclusions:

• The developed CIM-RI-MP2 gradient algorithm is computationally efficient for large molecular systems.
• This method is suitable for obtaining optimized geometries of systems with hundreds of atoms.
• The approach offers a practical tool for high-accuracy electronic structure calculations on complex molecules.