Dual-energy computed tomography for non-destructive characterization of pigments in cultural heritage

The analysis of works of art (artifacts) requires non-destructive techniques that offer both morphological and material characterization. In this study, we show how Dual Energy Computed Tomography (DECT) could be used to characterize natural pigments ( e. g ., lead white, red lead, cinnabar, red ochre, etc.) in the painting layer of an18th-century icon. Image acquisition was performed on a clinical single-energy multi-slice CT scanner using four different X-ray tube potentials at 80, 100, 120, and 140 kV. Energy dispersive X-ray fluorescence (EDXRF) analysis was performed to validate material characterization results obtained from DECT. The optimal pair of X-ray tube potentials was found to be 80 and 120 kV, considering the maximum separation of the four regions of interest in two-dimensional Hounsfield units (HU) histogram space. Using lead white and cinnabar as a decomposition basis, maps of lead-based and mercury-based pigments were created from the dual energy (80 and 120 kV) CT scans using the material decomposition method. The obtained results were successfully validated against EDXRF measurements. The material maps were then used to analyse red pigments, i.e., distinguish red ochre and cinnabar pigments and exclude the presence of red lead from selected regions within the icon. The examination of the icon showed the new potential of the presented technique to provide valuable material composition information for the conservation and preservation study of artifacts while maintaining their integrity. (c) 2025 Elsevier Masson SAS. All rights are reserved, including those for text and data mining, AI training, and similar technologies.