Spin-Orbit Coupling via Four-Component Multireference Methods: Benchmarking on p-Block Elements and Tentative...

Area of Science:

• Computational chemistry
• Quantum chemistry
• Electronic structure theory

Background:

• Fully relativistic four-component (4c) methods rigorously incorporate spin-orbit coupling.
• Spin-orbit splittings serve as crucial benchmarks for evaluating the accuracy of these advanced computational methods.
• Previous studies have not extensively explored spin-orbit splittings for p-block elements using 4c approaches.

Purpose of the Study:

• To compute and assess spin-orbit splittings for p-block elements (Boron to Iodine) using various 4c electronic structure methods.
• To evaluate the performance of 4c-CASSCF, 4c-CASPT2, and 4c-MR-CISD+Q methods against experimental data.
• To investigate the impact of basis set size on the accuracy of computed spin-orbit splittings.

Main Methods:

• Utilized fully relativistic four-component (4c) methods based on the Dirac Hamiltonian.
• Employed 4c-CASSCF, 4c-CASPT2, and 4c-MR-CISD+Q computational chemistry techniques.
• Performed calculations using uncontracted Dunning basis sets and the BAGEL software package.

Main Results:

• Most computed spin-orbit splittings were within 15% of experimental values, indicating good accuracy.
• Accurate splittings for light elements (B-F) require large basis sets, while heavier elements show less basis dependence.
• The 4c-MR-CISD+Q method excelled for light elements, 4c-CASSCF for heavier elements, and 4c-CASPT2 for group 13 atoms.

Conclusions:

• Four-component relativistic methods provide accurate spin-orbit splittings for p-block elements.
• Basis set convergence is critical for light elements but less so for heavier elements.
• Specific 4c methods demonstrate varying strengths for different element groups, offering valuable insights for computational chemists.