Premium
Studies on paint cross sections of a glass painting by using FT‐IR and Raman microspectroscopy supported by univariate and hierarchical cluster analyses
Author(s) -
Staniszewska Emilia,
Malek Kamilla,
Kaszowska Zofia
Publication year - 2013
Publication title -
journal of raman spectroscopy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.748
H-Index - 110
eISSN - 1097-4555
pISSN - 0377-0486
DOI - 10.1002/jrs.4332
Subject(s) - raman spectroscopy , chemical imaging , univariate , raman microspectroscopy , optics , materials science , infrared , image resolution , raman scattering , analytical chemistry (journal) , chemistry , computer science , artificial intelligence , physics , multivariate statistics , hyperspectral imaging , chromatography , machine learning
Fourier transform infrared (FT‐IR) and Raman spectroscopy is used for the non‐destructive analysis of painting materials and ageing compounds in micrometric cross sections of a glass painting. The combination of both techniques in conjunction with imaging/mapping function provides the spatial distribution of chemical components identified in vibrational spectra. The aim of our work is to show the applicability of the FT‐Raman mapping technique in the detection of painting materials. We also compare Raman information gained by using two laser excitations at 532 and 1064 nm implemented in microspectrometers with different confocality and spatial resolution. In turn among FT‐IR imaging techniques, we compare chemical images recorded in external reflection and attenuated total reflection modes that give chemical images of different size and spatial resolution. Our FT‐IR and Raman imaging characterize a number of painting materials such as pigments, binders, fillers as well as degradation products. Raman maps are constructed by using the univariate analysis. In turn, a profile of IR images requires the use of a more complex methodology. Here, we compare FT‐IR images of the painting cross sections obtained by using the univariate and hierarchical cluster analysis. We clearly show that the multivariate approach is a powerful tool for the credible construction of IR images, providing the relevant chemical information on the multicomponent stratigraphy of the samples. Moreover, the combination of all the methods allows us to demonstrate their degree of utility for the study on the paint cross sections of the works of art. Copyright © 2013 John Wiley & Sons, Ltd.