Experimental evaluation of the surface specificity of sum-frequency vibrational spectroscopy

Summary form only given. Surface vibrational spectroscopy via infrared-visible sum-frequency generation (SFG) has been developed into a powerful tool to study the structures of surfaces and interfaces. It is based on the principle that SFG is forbidden under electric-dipole approximation in a medium with inversion symmetry, but necessarily allowed at a surface or interface where the inversion symmetry is broken. However, there is always the question whether the bulk quadrupole contribution to the reflected surface SFG spectrum is indeed negligible. To answer this question, we need to measure the quadrupole bulk contribution separately. This can be done with SFG in transmission. Because of the longer-coherence length for phase matching in transmission, the quadrupole bulk contribution could dominate over the dipole-allowed surface contribution. Although the transmission measurement cannot determine all the independent tensor elements of the quadrupole nonlinear susceptibility of the bulk, it provides at least an order-of-magnitude estimate of the bulk contribution to surface SFG. We have carried out quantitative measurements and comparison of reflected and transmitted SFG spectra for a number of samples and learned whether their reflected SFG spectra are surface-dominated.