Comparison of density functionals for energy and structural differences between the high- † 5 T 2g : "t 2g … 2ϩ . Since very little experimental results are available ͑except for crystal structures involving the cation in its high-spin state͒, the primary comparison is with our own complete active-space self-consistent field ͑CASSCF͒, second-order perturbation theory-corrected complete active-space self-consistent field ͑CASPT2͒, and spectroscopy-oriented configuration interaction ͑SORCI͒ calculations. We find that generalized gradient approximations ͑GGAs͒ and the B3LYP hybrid functional provide geometries in good agreement with experiment and with our CASSCF calculations provided sufficiently extended basis sets are used ͑i.e., polarization functions on the iron and polarization and diffuse functions on the water molecules͒. In contrast, CASPT2 calculations of the low-spin-high-spin energy difference ⌬E LH ϭE LS ϪE HS appear to be significantly overestimated due to basis set limitations in the sense that the energy difference of the atomic asymptotes ( 5 D→ 1 I excitation of Fe 2ϩ ) are overestimated by about 3000 cm Ϫ1 . An empirical shift of the molecular ⌬E LH based upon atomic calculations provides a best estimate of 12 000-13 000 cm Ϫ1 . Our unshifted SORCI result is 13 300 cm Ϫ1, consistent with previous comparisons between SORCI and experimental excitation energies which suggest that no such empirical shift is needed in conjunction with this method. In contrast, after estimation of incomplete basis set effects, GGAs with one exception underestimate this value by 3000-4000 cm Ϫ1 while the B3LYP functional underestimates it by only about 1000 cm Ϫ1 . The exception is the GGA functional RPBE which appears to perform as well as or better than the B3LYP functional for the properties studied here. In order to obtain a best estimate of the molecular ⌬E LH within the context of density functional theory ͑DFT͒ calculations we have also performed atomic excitation energy calculations using the multiplet sum method. These atomic DFT calculations suggest that no empirical correction is needed for the DFT calculations. © 2004 American Institute of Physics. ͓DOI: 10.1063/1.1710046͔ I. INTRODUCTIONA well-known feature of d 6 Tanabe-Sugano ligand field theory ͑LFT͒ diagrams for octahedral complexes is the reversal of the ordering of the low-spin 1 A and high-spin 5 T in the spin-crossover region of ligand field strength.1 For compounds in this region, spin crossover may be either thermally or optically induced, 2 leading to possible applications in storage and display devices. [3][4][5] We are particularly interested in the phenomenon of light-induced excited spin-state trapping ͑LIESST͒ in octahedral iron II compounds, which involves the optical interconversion of the high-spin ͑HS͒ 5 T 2g and low-spin ͑LS͒ 1 A g electronic states. While this can be understood at a qualitative level using LFT, 1,2 it is also known that the e g orbitals, populated in going from the LS to the HS state, are antibonding, so that b...