It has been shown that circular dichroism (CD) and two-photon absorption (TPA) spectroscopies are capable of throwing new light on fⁿ ⇒ fⁿ optical transitions. High-resolution circular dichroism of single crystals can provide information on the chiral environment as well as on the magnetic and electric quadrupolar transition moments of lanthanide ions, and thus can be exploited for the purpose of assignment of bands. TPA, on the other hand, although capable of resolving crystal-field components better than one-photon absorption, exhibits puzzling intensity features, which need to be sorted out before it can be generally used as a diagnostic tool.

The complementarity of one- and two-photon spectroscopy has been utilised for throwing light on the following problems of chemical interest: (1) Weak interaction between identical chromophores separated by insulating bridges gives rise to split states of different symmetries. Two-photon spectroscopy (TPA), in conjunction with one-photon absorption (OPA), has been used to identify the states and hence to estimate the magnitude of interaction in bimolecules and trimolecules. From the shifts between the one- and the two-photon spectra, the splittings have been estimated. Calculations confirm that the dominant interaction is the through-bond one. (2) The second type of problem is the identification ofg andu vibrations in molecules. We have initiated studies on three molecules in jet-cooled conditions: 9,10-dihydro-anthracene (DHA). 9,10-dihydro-phenanthrene (DHP) and octa-fluoronaph-thalene (OFN). Only the one-photon fluorescence excitation spectra have so far been obtained by us and the TPA spectra are under investigation. (3) The third class of molecules discussed here are the Ln³⁺ complexes wherefⁿ⇒ fⁿ transitions are intrinsically two-photon allowed. We have studied two GD³⁺ single crystals. The CF-splittings, observed clearly in TPA, have been fitted with a parametric model. Some of our observations on the variations of TPA intensity patterns from crystal to crystal, such as circular:linear polarisation ratios, relative intensities of transitions to differentJ-states, do not quite fit in with the Axe-Judd-Downer model. The discrepancies call for a reappraisal of the role of ligand in the TPA process.