Ion beam characterisation of paint layers made according to late 18th century techniques J.-O. Lill a,b, * , M. Str€ om b , M. Brenner c , A. Lindroos d a Accelerator Laboratory, Turku PET Centre, Porthansg. 3, FIN-20500 Turku, Finland b Polytechnic Sydv€ ast, Formansall en 2, FIN-10600 Eken€ as, Finland c Department of Physics,  Abo Akademi University, Porthansg. 3, FIN-20500 Turku, Finland d Department of Geology and Mineralogy,  Abo Akademi University, Domkyrkotorget 1, FIN-20500 Turku, Finland Abstract External beam particle-induced X-ray emission (PIXE) and PIGE were applied in a study of layered paint structures. A series of layered paint samples were prepared according to known late 18th century techniques. Special attention was paid to two-layer systems with lead white paint as the top layer. The effects of the opaque layer of lead white on the PIXE results were studied and a method for calculating the thickness of the lead white layer is suggested. The inter- pretation of PIXE results from two-layer systems is discussed in terms of detection limits. Using a 3 MeV proton beam, elements lighter than iron cannot be detected beneath an opaque layer of flake white paint. In the case of calcium the nuclear reaction 48 Ca(p,nc) 48 Sc can be used as an indicator for a chalk ground beneath the lead white paint. However, a higher proton energy (4.2 MeV) is needed in this case. Ó 2002 Elsevier Science B.V. All rights reserved. PACS: 81.70.Jb; 81.70.-q; 82.80.Ej; 82.80.Yc Keywords: PIXE; PIGE; Paint-layer structure; Lead white; Chalk ground 1. Introduction Ion beam analysis has been frequently applied in art and archaeology [1]. In Europe these appli- cations have been organised within the framework of a research project (COST G1 and now COST G8) [2]. Particle-induced X-ray emission (PIXE) analysis of paint has been used to determine the provenance of paintings. The interpretation of the results obtained with PIXE is problematic in the case of layered structures e.g. in easel painting. Neelmeijer et al. have demonstrated a method for depth resolved characterisation of paint layers on test structures by using different particle energies [3]. However, no information about the chalk ground was achieved in the case of a layer system of lead white on chalk [3,4]. This was due to the strong attenuation of calcium K X-rays in the lead white. The background to this work can be found in a study of paint layers on late 18th century neo- classical furniture in Finland/Sweden. One issue was to determine whether the popular ‘‘pearl grey’’ Nuclear Instruments and Methods in Physics Research B 189 (2002) 303–307 www.elsevier.com/locate/nimb * Corresponding author. Address: Accelerator Laboratory,  Abo Akademi University, Porthansg. 3, FIN-20500 Turku, Finland. Tel.: +358-2-2154924; fax: +358-2-2154912. E-mail address: jlill@abo.fi (J.-O. Lill). 0168-583X/02/$ - see front matter Ó 2002 Elsevier Science B.V. All rights reserved. PII:S0168-583X(01)01075-8