Orbital dependent complications for close vs well-separated electrons in diradicals

Area of Science:

• Quantum Chemistry
• Computational Materials Science

Background:

• Open-shell diradical systems present challenges for standard electronic structure calculations.
• Density-functional approximations (DFAs) often fail to predict open-shell character in systems like O3.

Purpose of the Study:

• Investigate the accuracy of DFAs for diradicals with varying orbital proximity.
• Identify the source of errors in DFA predictions for O3.
• Develop accurate methods for determining electronic configurations and singlet-triplet splitting.

Main Methods:

• Fermi-Löwdin-orbital-self-interaction-corrected (FLOSIC) formalism.
• Analysis of many-electron overlap matrices.
• Energy calculations using corrected generalized gradient, local density, and Hartree-Fock approximations.

Main Results:

• Standard DFAs incorrectly predict closed-shell character for O3.
• Self-interaction error is identified as the cause of inaccuracies in O3.
• FLOSIC method accurately determines electronic configurations and singlet-triplet splitting for O3.
• Broken-spin-symmetry configurations are shown to be irrelevant for Ms=0 triplet states in O3.

Conclusions:

• FLOSIC corrections are essential for accurate electronic structure of O3.
• Well-separated diradicals like (C21H13)2 are less sensitive to these corrections.
• Accurate results for O3 provide a basis for developing Heisenberg-like Hamiltonians for molecular magnets.