Methodological Choices for the Spin-Crossover Energy: Mn(taa) as an Example*

Systematic prediction of the spin-crossover (SCO) energy for both an isolated molecule and its condensed aggregates poses a challenge to current density functional approximations, DFAs [1]. With Mn(taa) as the example, we discuss the effects of DFA choice, dispersion corrections, whether and how to use Hubbard U corrections, and technical and procedural sensitivities and subtleties. We show that, with one exception (r²SCAN [2]), no comparatively simple, non-empirical DFA provides predictive accuracy. Systematic procedures to calculate U fail with popular DFAs. User-driven tuning of U becomes a diagnostic of DFA appropriateness. It confirms the suitability of r²SCAN [2]. In conjunction with structural calculation using r²SCAN-L [3], a rapid procedure for assessing SCO becomes possible.

[1] B. J. Houghton, and R. J. Deeth, Eur. J. Inorg. Chem. 2014, 4573 (2014).
[2] J. W. Furness, A. D. Kaplan, J. Ning, J. P. Perdew, and J. Sun, J. Phys. Chem. Lett. 11, 8208 (2020).
[3] D. Mejia-Rodriguez, and S.B. Trickey, Phys. Rev. B 102, 121109(R) (2020).