Prediction of sublayer depth in turbid media using spatially offset Raman spectroscopy.

We demonstrate experimentally the feasibility of monitoring the depth of optically thick layers within turbid media using spatially offset Raman spectroscopy (SORS) in combination with multivariate analysis. The method uses the deep penetration capability of SORS to characterize significantly thicker (by at least a factor of 2) layers than possible with conventional Raman spectroscopy. Typical relative accuracies were between 5 and 10%. The incorporation of depth information into a SORS experiment as an additional dimension allows pure spectra of each individual layer to be resolved using three-dimensional multivariate techniques (parallel factor analysis, PARAFAC) to accuracies comparable with the results of a two-dimensional analysis.