Iron(II) and Iron(III) Spin Crossover: Toward an Optimal Density Functional

Area of Science:

• Computational Chemistry and Materials Science
• Quantum Chemistry
• Solid-State Chemistry

Background:

• Spin crossover (SCO) phenomena are vital in biochemistry, catalysis, and materials science, with accurate prediction being a high priority.
• Density Functional Theory (DFT) is the primary tool for SCO system prediction but exhibits high sensitivity to methodological choices.
• Existing DFT methods often show imbalanced accuracy for Fe(II) and Fe(III) SCO systems, hindering comprehensive studies.

Purpose of the Study:

• To investigate the performance of various DFT functionals for well-established Fe(II) and Fe(III) SCO complexes.
• To identify DFT functionals that provide balanced accuracy for both Fe(II) and Fe(III) SCO systems.
• To explore the limits of DFT accuracy and precision for iron-based SCO materials.

Main Methods:

• Evaluation of diverse DFT functionals, including GGA, hybrid, meta-GGA, meta-hybrid, double-hybrid, and long-range-corrected hybrid types.
• Systematic testing of customized functional versions with adjusted Hartree-Fock exchange percentages.
• Analysis of accuracy (Mean Absolute Error) and precision (standard deviation of errors) for Fe(II)/Fe(III) SCO systems.

Main Results:

• Adjusting Hartree-Fock exchange to 10-17% generally improves accuracy but often introduces Fe(II)-Fe(III) bias.
• CAMB3LYP-17, B3LYP*, and B97-15 (15-17% HF exchange) show the highest accuracy (<10 kJ/mol MAE).
• The customized PBE0-12 functional offers the best balance (<5 kJ/mol bias, 15 kJ/mol MAE); otherwise, a trade-off between balance and accuracy is necessary.
• DFT's precision limit for iron SCO is around 5 kJ/mol, with various functional behaviors observed (accurate/precise, accurate/imprecise, etc.).

Conclusions:

• Balanced accuracy is crucial for DFT studies of SCO systems, especially where iron oxidation states vary.
• Specific functionals like PBE0-12 provide a good compromise, while others require careful selection based on whether Fe(II)-Fe(III) balance is prioritized.
• Current DFT methods have limitations in achieving the target accuracy and precision for iron SCO, suggesting ongoing methodological development is needed.