ABSTRACT
A new method has been developed to automatically identify Raman spectra, whether they correspond to single- or multicomponent spectra. The method requires no user input or judgment. There are thus no parameters to be tweaked. Furthermore, it provides a reliability factor on the resulting identification, with the aim of becoming a useful support tool for the analyst in the decision-making process. The method relies on the multivariate techniques of principal component analysis (PCA) and independent component analysis (ICA), and on some metrics. It has been developed for the application of automated spectral analysis, where the analyzed spectrum is provided by a spectrometer that has no previous knowledge of the analyzed sample, meaning that the number of components in the sample is unknown. We describe the details of this method and demonstrate its efficiency by identifying both simulated spectra and real spectra. The method has been applied to artistic pigment identification. The reliable and consistent results that were obtained make the methodology a helpful tool suitable for the identification of pigments in artwork or in paint in general.
ABSTRACT
The interpretation of a Raman spectrum is based on the identification of its characteristic molecular bands. However, the assignment of the vibrational modes is often compromised by the presence in the spectrum of an intense fluorescence background that covers the measured spectra. Several techniques have been employed to minimize the presence of this fluorescence in order to resolve and analyze Raman spectra. In this paper a new automated method for fluorescence subtraction is described, based on morphology operations. This method is compared with the most commonly used polynomial fitting methods. Results indicate that the proposed automated method is efficient in fluorescence subtraction and retains the line shapes and positions of the Raman bands in the spectra.
ABSTRACT
Raman spectroscopy provides useful information to detect and identify pictorial materials in artworks, although some problems are involved when the identification of individual pigments in mixtures is treated. With the hypothesis of the Principle of superposition, the mixture spectrum should be the direct sum of each pondered individual spectrum. In this work, we will show several mixtures where it can be noticed that the mixture spectrum is not qualitatively proportional to the sum of pondered individual spectra. Also there were some cases where the bands of one of the pigments could not be detected in the mixture spectrum. This non-linear behaviour could be attributed to specific proprieties of each material that are revealed when they interact with each other. We conjecture that, for instance, the different reflectances or the wavelength of the laser source could be determinant factors of the obtained results. In this paper an experimental method has been designed in order to characterize the quantitative behaviour of the Raman bands corresponding to each pigment in a mixture. Adequate coefficients are defined and calculated to facilitate the study of the spectral contribution of the different components of a mixture.
