The application of standard two-way curve resolution methods is reported for analysis of three-way Raman image data. Two current curve resolution methods are described: principal factor multivariate curve resolution (PF-MCR), which uses principal factor analysis (PFA) combined with varimax rotation and alternating least-squares optimization (ALS), and orthogonal projection multivariate curve resolution (OP-MCR), which uses a Gram–Schmidt modified orthogonal projection approach (OPA) followed by ALS. The OP-MCR technique is shown to be an extremely rapid method of analysis producing results equivalent to those of PF-MCR in one-third to one-fourth the time. The results from MCR analysis using either method provide the number of chemical species present in the sample, the spectrum of each species for identification, and the concentration image for each species. The additional benefit of image noise reduction also results from the MCR techniques. A brief description of the theory is presented followed by analysis and comparison of results for two real Raman image data. A discussion is given addressing the rapid analysis aspects of OP-MCR and the relative merits and drawbacks of the technique in comparison to PF-MCR. The use of data subsampling is also discussed as a way of decreasing analysis time without loss in accuracy or performance.

The use of Raman microscopy in imaging two emulsion systems is described. Registered optical microscopy and Raman images are collected, the latter describing the chemical basis of the heterogeneity observed in the former. These examples act as a powerful demonstration of the application of the Raman microscopy technique to the analysis and understanding of microstructure in commercial products. The results indicate how the principles of Raman imaging can be applied to complex, multicomponent, multiphase systems of inherently low contrast. Such systems are of importance because they represent a wide variety of commercial product systems, ranging from pharmaceutical creams through skin creams and toothpastes. The use of a software environment for the organization, storage, management, interrogation, and manipulation of multidimensional spectral imaging data is also described. The important factors to be considered in determining the full information content of such data sets are established, and suggestions as to how such data sets can be optimally interrogated are made.

Step-scan Fourier Transform Inflared (FT-IR) Photoacoustic Spectroscopy (PAS) has been used to study the penetration of substances through abdominal pig skin. The experiments were performed in vitro and dimethylsulfoxide(DMS0) was applied either onto the surface or the bottom of the sample. The results demonstrated that DMSO propagated fast through the skin from the bottom sur$ace. The ease of identijication of this substance in this spectral range showed the potential of the technique, suggesting that penetration and distribution of other chemicals through the skin can be investigated by FT-IR PAS.Penetration and interaction of substances through skin have been investigated both for medical and cosmetic purposes, but the mechanism involved is not well understood [l,2]. Based on its ability to perform depth profile analysis, PAS has already been shown to be suitable for the study of penetration of topically applied substances into skin[3-51. Giese et al, for example, have studied the penetration of sunscreen into human skin, in the visible spectral range [S].The absorption bands from specific group components of substances are easily identified in the infrared spectral range, and a technique such as photoacoustic spectroscopy, which provides depth profile analysis in this range, may be convenient to study the distribution of substances through the skin. .

No abstract