Microanalytical method for studying paintings by use of fluorescence spectroscopy combined with principal component analysis
- Authors: Mannino, M.; Orecchio, S.; Gennaro, G.
- Publication year: 2013
- Type: Articolo in rivista (Articolo in rivista)
- OA Link: http://hdl.handle.net/10447/104134
The study of several different naturally aged films of organic painting materials, is described. Materials studied include drying oils (walnut, poppy seed and linseed oil) resin-based paint varnishes (the triterpenoid dammar and the diterpenoids sandarac and turpentine) and protein-based binding media (egg yolk and casein). A not expensive methodology for a rapid identification of binding media and varnishes, that applies principal component analysis (PCA) to Excitation Emission (EE) fluorescence spectra, is proposed. PCA was performed on a data matrix where the rows represent studied materials and columns the variables; the variables include the fluorescence intensities associated to different points (coordinates are λex/λem) in EE spectra. Results suggest that the new method can be used to classify micro samples on the basis of their fluorescence spectra, because the placement of the different materials in the score plot reflects their origin and chemical composition. The method described has been used for the analysis of materials of an oil painting on canvas, representing a mother with her child and on three mural paints. A small area of painting has been sampled with a cotton swab soaked in acetone, and subsequently the same area was sampled with a cotton swab and diethyl ether. The results of the application of the proposed microanalytical method to oil painting show that the author has used the linseed oil as binding medium and subsequently the surface of painting has been treated with a varnish containing the sandarac as resin. Given the great diversity and complexity of the products used for the realization of a painting, Fourier Transform Infrared (FTIR) spectra of acetone and diethyl ether solutions have been performed, confirming the results of PCA applied to EE spectra.