Two bichromophoric systems are presented that contain an N-alkylnaphthalimide electron acceptor and a 4-methoxyaniline (3a) or an aniline (3b) electron donor, respectively. Upon photoexcitation of 3a in cyclohexane electron transfer occurs in the singlet manifold to afford the short-lived (τ f) 0.75 ns) 1 (D +-A-) state in ca. 70% yield. An important decay pathway of this D +-A-state consists of intersystem crossing (ISC) to yield a triplet state localized on the naphthalimide moiety (D-3 A). In a slightly more polar solvent like din -butyl ether, an equilibrium between D-3 A and 3 (D +-A-) is observed by means of transient absorption spectroscopy. Both species decay with an overall decay time of ca. 1 µs. Thus, upon changing the spin multiplicity of the D +-A-state from singlet to triplet, an increase of its lifetime by three orders of magnitude is observed. In more polar solvents like dioxane, THF, and acetonitrile the 3 (D +-A-) state is the only species observed in the transient absorption spectrum, with decay times of ca. 1, 0.5, and 0.1 µs, respectively. The D-3 A state is the precursor state for the 3 (D +-A-) state in these solvents. It is proposed that, upon increasing solvent polarity, the singlet charge-separation process is retarded as a result of the large driving force (-∆G S°> 1 eV), which allows the triplet pathway (D-1 A f D-3 A f 3 (D +-A-)) to compete effectively. Compound 3b possesses a somewhat weaker donor chromophore than 3a resulting in a smaller driving force. The decay of the locally excited singlet state of 3b occurs mainly Via charge separation in the singlet manifold (D-1 A f 1 (D +-A-)). Only in the very polar solvent acetonitrile does the triplet pathway become competitive, and evidence is found for the formation of 3 (D +-A-).

Solvent-dependent switching between two dipolar excited states in a rigidly extended trichromophoric system van Dijk, S.I.; Wiering, P.G.; Groen, C.P.; Brouwer, A.M.; Verhoeven, J.W.; Schuddeboom, W.; Warman, J.M.

The structure, bonding and vibrational properties of the mixed LiLnX4 (Ln = La, Dy; X = F, Cl, Br, I) rare earth/alkali halide complexes were studied using various quantum chemical methods (HF, MP2 and the Becke3-Lee-Yang-Parr exchange-correlation density functional) in conjunction with polarized triple-zeta valence basis sets and quasi-relativistic effective core potentials for the heavy atoms. Our comparative study indicated the superiority of MP2 theory while the HF and B3-LYP methods as well as less sophisticated basis sets failed for the correct energetic relations. In particular, f polarization functions on Li and X proved to be important for the Li...X interaction in the complexes. From the three characteristic structures of such complexes, possessing 1-(C3v), 2-(C2v), or 3-fold coordination (C3v) between the alkali metal and the bridging halide atoms, the bi- and tridentate forms are located considerably lower on the potential energy surface then the monodentate isomer. Therefore only the bi- and tridentate isomers have chemical relevance. The monodentate isomer is only a high-lying local minimum in the case of X = F. For X = Cl, Br, and I this structure is found to be a second-order saddle point. The bidentate structure was found to be the global minimum for the systems with X = F, Cl, and Br. However, the relative stability with respect to the tridentate structure is very small (1-5 kJ/mol) for the heavier halide derivatives and the relative order is reversed in the case of the iodides. The energy difference between the three structures and the dissociation energy decrease in the row F to I. The ionic bonding in the complexes was characterized by natural charges and a topological analysis of the electron density distribution according to Bader's theorem. Variation of the geometrical and bonding characteristics between the lanthanum and dysprosium complexes reflects the effect of "lanthanide contraction". The calculated vibrational data indicate that infrared spectroscopy may be an effective tool for experimental investigation and characterization of LiLnX4 molecules.

The molecular geometry and vibrational frequencies of monomeric and dimeric dysprosium tribromide, DyBr(3) and Dy(2)Br(6), together with the electronic structure of their ground and first few excited-state molecules were determined by high-level computations, electron diffraction, gas-phase infrared, and matrix isolation infrared and Raman spectroscopy. The effect of partially filled 4f orbitals and spin-orbit coupling on their structure was studied by computations. While the geometry of the monomer does not depend on the 4f orbital occupation, the bond angles of the dimer are noticeably influenced by it. The monomer is found to be planar from all methods; the suggested equilibrium bond length of the molecule (r(e)) is 2.591(8) A, while the thermal average distance (r(g)) is 2.606(8) A. Although the gas-phase DyBr(3) molecule is planar, it forms a complex with the matrix molecules in the matrix-isolation spectroscopic experiments, leading to the pyramidalization of the DyBr(3) unit. Our model calculations in this regard also explain the often conflicting results of computations and different experiments about the shape of lanthanide trihalides.