Ap spectral shift. cm-' Ap dimer 349 0.020 hexamer 771 0.040 trimer 544 0.029 heptamer 798 0.041 tetramer 660 0.035 octamer 814 0.042 pentamer 729 0.038Results are from Shipman-PCM-based exciton calculations which include monomer Soret band states. All shifts given are Fed shifts relative to the monomer Q, band. bGiven as Af = (1/M) floligomer) -f(monomer), where M is the number of monomers in the oligomer. flmonomer) = 0.65.aggregates studied here represent a somewhat limited model of a crystal, exploratory calculations on more extensively aggregated systems indicate that one-dimensional stacking is the major factor responsible for the observed spectral shifts, and only perhaps a 10% increase in shifts would be expected to arise from other modes of aggregation.Changes in oscillator strength are seen to increase steadily with aggregation, the oscillator strength of the octamer being approximately 6% larger than that of the monomer. As discussed previously, it is expected that these changes are underestimated by the present "limited" exciton calculations, and thus significant hyperchromism would be expected of the larger Me-BPheo-a aggregates. ConclusionsThe present results as well as those obtained in the previous study of related dimer systemsz9 reveal severe problems associated with application of the point-dipole approximation in exciton studies of aggregates of spatially-extended monomers. Unacceptable errors in both spectral shifts and oscillator strengths are produced by this method, regardless of whether experimentally or theoretically determined dipoles are employed. The method appears totally unreliable and should be avoided.The present results also demonstrate the importance of including non-nearest-neighbor interactions in exciton calculations of aggregated systems. Computations in which only nearest-neighbor interactions were considered seriously underestimated the magnitude of spectral shifts, although oscillator strengths were largely unaffected. Neglect of interactions between molecules with center-to-center distances >20 8, appear to produce acceptable errors in spectral shifts.There are insufficient experimental data to determine the accuracy with which Shipman-PCM-based calculations can predict spectral shifts, but it is clear that red-shifts are underestimated by this method. It is possible that larger spectral shifts would be obtained from exciton calculations in which higher quality monomer wave functions were employed. The present FSGObased wave functions contain basis functions which do not extend into space as much as do larger atomic basis sets; consequently, the present exciton calculations might be expected to underestimate the intermolecular interactions responsible for spectral shifts. Beyond this, it has been suggested" that charge-transfer effects may be significant, and consideration of this as well as monomeric charge redistribution may be necessary in order to obtain accurate spectral shift values.Exciton calculations on the present aggregates using the Shipman-PCM appear to be u...

With ab initio molecular orbital calculations, the structures of the cation clusters Mg+(H20)" and their hydrogen-eliminated products (Mg0H)+(H20)"-i are optimized. In Mg+(H20)", the hydration number of the most stable isomer is 3. In (Mg0H)+(H20)"-i, all water molecules are directly bonded to Mg+ for n < 6. The hydration energy of (MgOH)+ is larger than that of Mg+ because of the strongly polarized (MgOH)+ molecular ion; Mg is oxidized halfway to Mg(II). The internal energy change of the hydrogen elimination of Mg+(H20)" is positive for n = 1-5, but becomes negative for n = 6, which is in good agreement with the product switching in the TOF spectrum reported in the preceding paper by Sanekata et al. The effects of isotope substitution and equilibrium constants of the hydrogen (deuterium) elimination reaction observed in their experiment can be explained qualitatively.

No abstract

No abstract