reasonably well with a weak feature found in the electron-energy-loss spectrum at 5.85 eV.7 *A feature in the experimental spectrum at 6.04 eV has been previously assigned as the 2*BU state.7 Given the error in our calculated ionization potential, it is unrealistic to attempt to choose between these two features based on the calculated results.One question which remains unanswered by the present study is the cause of the seeming disappearance of the 2'Ag state in the gas-phase fluorescence spectrum. Our results indicate that in relaxed excited-state geometries the 2*Ag state is found to lie significantly below the l'Bu state. Since the calculations are performed on the isolated molecule, the cause of this discrepancy remains unresolved. Clearly further work is required to answer this question. V. ConclusionResults are presented from ab initio Cl calculations for several low-lying excited states of «//-Zranj-octatetraene. In a vertical transition from the ground state the lowest singlet excited state is found to be of 'Bu symmetry. This state is essentially valencelike.The second excited singlet state, is a 2*Ag state at the ground-state geometry, the zeroth-order description of which is multiconfigurational and can be identified with the so-called "doubly excited" state found in long-chain polyenes. However, application of a correction for the 2*Ag state based on the estimated 0-0 transition energy makes the l'Bu and 2'Ag states essentially degenerate at the ground-state geometry. Relaxation of the excited-state geometries leads to the 2*Ag state having the lowest 0-0 transition energy. Reasonable agreement is found with experiment where comparisons can be made, and we predict that the 2'BU state is of 3p* Rydberg character and lies near 5.65 eV.

The solvation of CO(2) in 1-butyl-3-methylimidazolium acetate (Bmim Ac) has been investigated by (1)H, (13)C, and (15)N NMR spectroscopy at low CO(2) molar fraction (mf) (x(CO(2)) ca. 0.27) corresponding to the reactive regime described in part 1 of this study. It is shown that a carboxylation reaction occurs between CO(2) and Bmim Ac, leading to the formation of a non-negligible amount (~16%) of 1-butyl-3-methylimidazolium-2-carboxylate. It is also found that acetic acid molecules are produced during this reaction and tend to form with elapsed time stable cyclic dimers existing in pure acid. A further series of experiments has been dedicated to characterize the influence of water traces on the carboxylation reaction. It is found that water, even at high ratio (0.15 mf), does not hamper the formation of the carboxylate species but lead to the formation of byproduct involving CO(2). The evolution with temperature of the resonance lines associated with the products of the reactions confirms that they have a different origin. The main byproduct has been assigned to bicarbonate. All these results confirm the existence of a reactive regime in the CO(2)-Bmim Ac system but different from that reported in the literature on the formation of a reversible molecular complex possibly accompanied by a minor chemical reaction. Finally, the reactive scheme interpreting the carboxylation reaction and the formation of acetic acid proposed in the literature is discussed. We found that the triggering of the carboxylation reaction is necessarily connected with the introduction of carbon dioxide in the IL. We argue that a more refined scheme is still needed to understand in details the different steps of the chemical reaction in the dense phase.

The interaction of s2d2 and sd3 Ti atoms with the CO2 molecule has been studied using density functional theory at the gradient-corrected level. The ground state Ti inserts with no energy barrier into a CO bond resulting in an OTiCO insertion product. The intrinsic reaction coordinate for the insertion process has been defined and the reaction mechanism has been investigated by analyzing various structures along this path. The singlet and triplet states of the final product are very close in energy. The comparison of the predicted vibrational frequencies and isotopic shifts for both states with those from matrix isolation infrared data reveals that the insertion product formed in low-temperature argon matrix corresponds to singlet state OTiCO. Ti(CO2) complexes in various coordination modes have also been located on the triplet and quintet potential energy surfaces, from which the triplet state (O,O) coordination mode is the most stable one lying, however, about 30 kcal/mol above the OTiCO molecule.

The interaction of the CO2 molecule with nickel atoms was studied by using matrix isolation spectroscopy and density functional theory. In argon dilute matrices, no reaction occurs, even after annealing the deposit. In neat CO2 matrices, it is shown that carbon dioxide forms a 1:1 complex with nickel which is characterized by its UV−visible and FTIR absorptions, including isotopically labeled species. Theory predicts the side-on coordination mode to be the most stable. The binding energy of the side-on Ni(CO2) complex is estimated to be 18 kcal/mol. The calculated OCO angle is 145°, which is quite a large value compared to those encountered in other known CO2 complexes. In dinitrogen matrices, the yield of CO2 complexation is considerably enhanced relative to that in argon dilute and neat CO2 matrices, which is attributed to the formation of unsaturated Ni(N2) n complexes prior to CO2 coordination. The CO2 binding energies calculated for the Ni(CO2)(N2) n (n = 1, 2) complexes (respectively 32 and 4 kcal/mol) suggest that CO2 probably coordinates to the Ni(N2) complex. This is a very interesting result, owing to the fact that CO2 does not react with nickel atoms in dilute argon matrices.