The dynamics of metal-to-ligand charge transfer (MLCT) in a cis-bis(4,4'-dicarboxy-2,2'-bipyridine)-bis(isothiocyanato)ruthenium(II) dye (N3) are compared for the free dye in solution and the dye adsorbed on the surface of the TiO(2) nanoparticles from resonance Raman spectroscopy. The 544-nm MLCT absorption band of N3 adsorbed on TiO(2) is slightly blue-shifted from that of the free N3, indicating a weak electronic coupling between N3 and TiO(2). The resonance Raman spectra of N3 and the N3|TiO(2) complex obtained upon excitation within the lowest-lying MLCT singlet state of the dye are similar except for slight shifts in band positions. Resonance Raman cross sections have been obtained for the vibrational modes of both N3 and N3|TiO(2) with excitation frequencies spanning the 544-nm MLCT band. Self-consistent analysis of the resulting resonance Raman excitation profiles and absorption spectrum using a time-dependent wave packet formalism over two electronic states yields mode-specific vibrational and solvent reorganization energies. Despite the weak electronic coupling between N3 and TiO(2) in N3|TiO(2), adsorption strongly affects the reorganization energies of N3 in the intramolecular MLCT state. Adsorption of N3 onto TiO(2) increases the absolute Raman cross section of each mode by a factor of ca. 1.6 and decreases the vibrational and solvent reorganization energies by factors of 2 and 6, respectively. The excited-state dynamics of N3 adsorbed on the surface of TiO(2) nanoparticles were observed to be independent of the number of N3 molecules adsorbed per TiO(2) nanoparticle. The effect of TiO(2) on the dynamics of the adsorbed N3 is primarily due to both mode-specific vibrational and electronic pure dephasing, with the dominant contribution from the latter process.

Resonance Raman spectra of alizarin-sensitized TiO2 nanoparticles have been obtained at excitation wavelengths throughout the 488-nm charge transfer absorption band. The resonance Raman spectrum of the alizarin-sensitized TiO2 nanoparticle is significantly different from the spectrum of free alizarin, consistent with a chemisorption-type interaction. This interaction is probably chelation of surface titanium ions by the hydroxy groups of alizarin, supported by the observed enhancement of bridging C–O modes at 1326 cm−1. In contrast to resonance Raman intensity analysis of homogeneous electron transfer where vibrations of both the donor and acceptor are observed, self-consistent analysis of the resulting resonance Raman excitation profiles and absorption spectrum using the time-dependent wave packet propagation formalism show mode-specific reorganization along alizarin vibrations exclusively; no resonance-enhanced vibrations attributable to the TiO2 moiety are observed. Therefore, the total resonance Raman-derived reorganization energy is only 0.04 eV, significantly smaller than the observed outer-sphere reorganization energy of 0.2 eV for this system and inner-sphere reorganization energies measured for other molecular systems. The discrepancy is ascribed to a significant environmental component to the outer-sphere reorganization energy arising from rapid dephasing of surface TiO2 units involved in adsorption by strongly coupled interior bath vibrations.

Resonance Raman and resonance hyper-Raman spectra of the "push-pull" conjugated molecule 1-(4'-dihexylaminostyryl)-4-(4"-nitrostyryl)benzene in acetone have been measured at excitation wavelengths from 485 to 356 nm (two-photon wavelengths for the nonlinear spectra), resonant with the first two bands in the linear absorption spectrum. The theory of resonance hyper-Raman scattering intensities is developed and simplified using assumptions appropriate for intramolecular charge-transfer transitions of large molecules in solution. The absorption spectrum and the Raman, hyper-Rayleigh, and hyper-Raman excitation profiles, all in absolute intensity units, are quantitatively simulated to probe the structures and the one- and two-photon transition strengths of the two lowest-energy allowed electronic transitions. All four spectroscopic observables are reasonably well reproduced with a single set of excited-state parameters. The two lowest-energy, one-photon allowed electronic transitions have fairly comparable one-photon and two-photon transition strengths, but the higher-energy transition is largely localized on the nitrophenyl group while the lower-energy transition is more delocalized.

Resonance Raman and resonance hyper-Raman spectra and excitation profiles have been measured for a "push-pull" donor-acceptor substituted conjugated polyene bearing a julolidine donor group and a nitrophenyl acceptor group, in acetone at excitation wavelengths from 485 to 356 nm (two-photon wavelengths for the nonlinear spectra). These wavelengths span the strong visible to near-UV linear absorption spectrum, which appears to involve at least three different electronic transitions. The relative intensities of different vibrational bands vary considerably across the excitation spectrum, with the hyper-Raman spectra showing greater variation than the linear Raman. A previously derived theory of resonance hyper-Raman intensities is modified to include contributions from purely vibrational levels of the ground electronic state as intermediate states in the two-photon absorption process. These contributions are found to have only a slight effect on the hyper-Rayleigh intensities and profiles, but they significantly influence some of the hyper-Raman profiles. The absorption spectrum and the Raman, hyper-Rayleigh, and hyper-Raman excitation profiles are quantitatively simulated under the assumption that three excited electronic states contribute to the one- and two-photon absorption in this region. The transition centered near 400 nm is largely localized on the nitrophenyl group, while the transitions near 475 and 355 nm are more delocalized.

The two-photon-resonant first hyperpolarizabilities associated with hyper-Rayleigh and hyper-Raman scattering are reported for 4-dimethylamino-4-nitrostilbene in 1,4-dioxane, dichloromethane, acetonitrile, and methanol, and for an ionic analog, 4-N,N-bis(6-(N,N,N-trimethylammonium)-hexyl)amino-4-nitrostilbene dibromide in methanol and water. Resonance Raman and hyper-Raman excitation profiles are also measured and modeled. The resonance Raman and hyper-Raman spectra show very similar relative intensities which do not vary much as the excitation frequency is tuned across the lowest-energy strong linear absorption band, suggesting that a single resonant electronic state dominates the one- and two-photon absorptions in this region. The absorption, resonance Raman, and hyper-Raman profiles can be simulated reasonably well with a common set of parameters. The peak resonant (absolute value of beta)2, measured by hyper-Rayleigh scattering, varies by about 50% over the range of solvents examined and shows a weak correlation with the linear absorption maximum, with the redder-absorbing systems exhibiting larger peak hyperpolarizabilities. The experimental hyper-Rayleigh intensities are higher than those calculated, possibly reflecting contributions from nonresonant electronic states.