Transition-element hexafluoride systems in ionic lattices. A Suhf molecular orbital study

An unrestricted Hartree-Fock SCFMO method, based on the MCZDO method of Brown and Roby, suitable for computing spin densities in transition-element compounds, is described. The method is used to study spin densities on fluorine in Cs₂MnF₆, K₂NaCrF₆, K₂MnF₆, K₂NaFeF₆, KMnF₃, RbMnF₃ and KNiF₃ using a “cluster” approximation in which a MF ₆ ^n? unit is explicitly considered. Excellent agreement is obtained between calculated and experimental spin parameters. The effect of the lattice is incorporated by using the electrostatic approximation of Brown, O'Dwyer and Roby. The lattice potential for these highly symmetric systems is found to have little effect on spin densities and charge distributions, but it effects substantial stabilization of the anion molecular orbitals. A general feature of the results is that central-atom 4p orbitals are scarcely involved in bonding, this being confined to the 3d and to some extent the 4s orbitals. Comments are offered on the lack of spin symmetry in the unrestricted Hartree-Fock wavefunctions of these systems, and the need to evaluate the core hamiltonian as accurately as possible.     

Vibrational quenching of excitonic splittings in H-bonded molecular dimers: the electronic Davydov splittings cannot match experiment

The S(1)/S(2) state exciton splittings of symmetric doubly hydrogen-bonded gas-phase dimers provide spectroscopic benchmarks for the excited-state electronic couplings between UV chromophores. These have important implications for electronic energy transfer in multichromophoric systems ranging from photosynthetic light-harvesting antennae to photosynthetic reaction centers, conjugated polymers, molecular crystals, and nucleic acids. We provide laser spectroscopic data on the S(1)/S(2) excitonic splitting Δ(exp) of the doubly H-bonded o-cyanophenol (oCP) dimer and compare to the splittings of the dimers of (2-aminopyridine)(2), [(2AP)(2)], (2-pyridone)(2), [(2PY)(2)], (benzoic acid)(2), [(BZA)(2)], and (benzonitrile)(2), [(BN)(2)]. The experimental S(1)/S(2) excitonic splittings are Δ(exp) = 16.4 cm(-1) for (oCP)(2), 11.5 cm(-1) for (2AP)(2), 43.5 cm(-1) for (2PY)(2), and <1 cm(-1) for (BZA)(2). In contrast, the vertical S(1)/S(2) energy gaps Δ(calc) calculated by the approximate second-order coupled cluster (CC2) method for the same dimers are 10-40 times larger than the Δ(exp) values. The qualitative failure of this and other ab initio methods to reproduce the exciton splitting Δ(exp) arises from the Born-Oppenheimer (BO) approximation, which implicitly assumes the strong-coupling case and cannot be employed to evaluate excitonic splittings of systems that are in the weak-coupling limit. Given typical H-bond distances and oscillator strengths, the majority of H-bonded dimers lie in the weak-coupling limit. In this case, the monomer electronic-vibrational coupling upon electronic excitation must be accounted for; the excitonic splittings arise between the vibronic (and not the electronic) transitions. The discrepancy between the BO-based splittings Δ(calc) and the much smaller experimental Δ(exp) values is resolved by taking into account the quenching of the BO splitting by the intramolecular vibronic coupling in the monomer S(1) ← S(0) excitation. The vibrational quenching factors Γ for the five dimers (oCP)(2), (2AP)(2), (2AP)(2), (BN)(2), and (BZA)(2) lie in the range Γ = 0.03-0.2. The quenched excitonic splittings Γ[middle dot]Δ(calc) are found to be in very good agreement with the observed splittings Δ(exp). The vibrational quenching approach predicts reliable Δ(exp) values for the investigated dimers, confirms the importance of vibrational quenching of the electronic Davydov splittings, and provides a sound basis for predicting realistic exciton splittings in multichromophoric systems.