Linear-Scaling Open-Shell MP2 Approach: Algorithm, Benchmarks, and Large-Scale Applications.

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Theoretical Chemistry

Background:

• Accurate electronic structure calculations are crucial for understanding molecular properties and reactions.
• High-spin open-shell molecules present significant computational challenges for traditional methods.
• Existing methods struggle with the scale and complexity of large open-shell systems.

Purpose of the Study:

• To develop a linear-scaling, local second-order Møller-Plesset (MP2) method for high-spin open-shell molecules.
• To enable accurate and efficient quantum chemical calculations on large and complex open-shell systems.
• To extend the accessibility of high-level electronic structure methods to unprecedented system sizes.

Main Methods:

• Implementation of a linear-scaling local second-order Møller-Plesset (LMP2) method based on restricted open-shell (RO) reference functions.
• Utilizing restricted local molecular orbitals for integral transformation and a novel long-range spin-polarization approximation.
• Leveraging iteration- and redundancy-free, integral-direct, and OpenMP-parallel algorithms for computational efficiency.

Main Results:

• RO-LMP2 method achieves computational cost comparable to closed-shell LMP2.
• Extensive benchmarks show local errors below 0.1 kcal/mol for systems up to 175 atoms.
• Successful application to large protein models (up to 601 atoms) with complex electronic structures.

Conclusions:

• RO-LMP2 provides a computationally efficient and accurate method for high-spin open-shell molecules.
• This method significantly expands the scope of accessible systems for MP2-level calculations.
• Enables widespread application of advanced computational methods to large, complex open-shell systems in chemistry and biology.