Determination of detection limits for SEM-EDS and m-FTIR analysis of artwork

Scanning electron microscopy and energy-dispersive X-ray spectroscopy (SEM-EDS) and micro-Fourier transform infrared (m-FTIR) spectroscopy are two of the primary analytical techniques used for the characterisation of artwork (Casadio and Toniolo, J. Cult. Herit. 2:71-78, 2001; Bruni et al., Vibr. Spectrosc. 20:15-25, 1999; Bouchard et al., e-Preserv. Sci. 6:27-37, 2009). Despite wide application of these techniques, no studies have been performed to evaluate their detection limits for this type of analysis. The characterisation of minor components used in a piece of art is important because these components may provide key information about the process of creation and answer questions regarding conservation and restoration (Casadio and Toniolo, J. Cult. Herit. 2:71-78, 2001; Bicchieri et al., Spectrochim. Acta B At. Spectrosc. 56:915-922, 2001). This study focused on easel paintings. Several mock-ups were prepared, and the painting layers were created with binary mixtures of three different blues (Prussian blue, phthalocyanine blue and ultramarine blue). Blue pigments have been used extensively in many art pieces and several studies have described problems related to their determination in low concentrations (Bouchard et al., e-Preserv. Sci. 6:27-37, 2009; Bicchieri et al., Spectrochim. Acta B At. Spectrosc. 56:915-922, 2001; Osticiolia et al., Spectrochim. Acta A Mol. Biomol. Spectrosc. 73:525-531, 2009; Giaccai, Mater. Res. Soc. Symp. Proc. 1047:233-242, 2008; Lutzenberger and Stege, e-Preserv. Sci. 6:89-100, 2009). This study indicated that the minimum concentrations at which the presence of the minor pigment is detected in at least half of the determinations performed are between 0.9% and 1.5% for phthalocyanine blue, between 0.8% and 1.3% for Prussian blue and between 0.3% and 1% for ultramarine blue with SEM-EDS; with m-FTIR spectroscopy, the minimum concentrations are between 0.9% and 1.5% for phthalocyanine blue, 2.4% or more for Prussian blue and 19% or more for ultramarine blue. The concentrations measured for some pigments were high and may result in errors in the description of the composition of a piece of art. The use of complementary techniques may address this limitation. On the basis of these results, the application of analytical procedures that include replicate determinations and analyses of areas of 100 × 100 μm(2) or greater and exclude single determinations are recommended to more accurately describe the material composition of a work of art.