Spin crossover (SCO) complexes exhibit changes in spin state in response to external stimuli and have applications in molecular electronics, making them challenging materials for computational design. We curate a data set of 95 Fe(II) SCO complexes (SCO-95) from the Cambridge Structural Database which have available low- and high-temperature crystal structures and, in most cases, confirmed experimental spin transition temperatures (T_1/2). We study these complexes using density functional theory (DFT) with thirty functionals spanning across multiple rungs of “Jacob’s ladder” to understand the effect of exchange-correlation functional on both electronic and Gibbs free energies. We also assess the effect of varying the Hartree-Fock exchange fraction (a_HF) in structures and properties from the B3LYP family of functionals. We identify three best-performing functionals, namely modified B3LYP (a_HF = 0.10), meta-generalized gradient approximation (GGA) M06-L, and meta-GGA hybrid TPSSh, accurately indicate SCO behavior for the majority of the complexes. Contrary to observations from prior studies, double-hybrids with higher a_HF values strongly stabilize high-spin states and exhibit poor performance in predicting SCO behavior. Computationally predicted T_1/2 values are consistent among the three functionals but show limited correlation to experimentally reported T_1/2 values or ranks. We rationalize these differences by the lack of crystal packing effects and counter-anions in the DFT calculations to aid prediction of experimentally observed phenomena like hysteresis and two-step SCO behavior. The SCO-95 set of 95 experimental SCO complexes thus presents opportunities for method development, both in terms of increasing model complexity and method fidelity.

Aditya Nandy

Graduate Student, Massachusetts Institute of Technology

Appearances