Time-of-flight secondary ion mass spectrometry (ToF–SIMS) has recently been shown to be a valuable tool for cultural heritage studies, especially when used in conjunction with established analytical techniques in the field. The ability of ToF–SIMS to simultaneously image inorganic and organic species within a paint cross section at micrometer-level spatial resolution makes it a uniquely qualified analytical technique to aid in further understanding the processes of pigment and binder alteration, as well as pigment–binder interactions. In this study, ToF–SIMS was used to detect and image both molecular and elemental species related to CdS pigment and binding medium alteration on the painting Le Bonheur de vivre (1905–1906, The Barnes Foundation) by Henri Matisse. Three categories of inorganic and organic components were found throughout Le Bonheur de vivre and co-localized in cross-sectional samples using high spatial resolution ToF–SIMS analysis: (1) species relating to the preparation and photo-induced oxidation of CdS yellow pigments (2) varying amounts of long-chain fatty acids present in both the paint and primary ground layer and (3) specific amino acid fragments, possibly relating to the painting’s complex restoration history. ToF–SIMS’s ability to discern both organic and inorganic species via cross-sectional imaging was used to compare samples collected from Le Bonheur de vivre to artificially aged reference paints in an effort to gather mechanistic information relating to alteration processes that have been previously explored using μXANES, SR-μXRF, SEM–EDX, and SR-FTIR. The relatively high sensitivity offered by ToF–SIMS imaging coupled to the high spatial resolution allowed for the positive identification of degradation products (such as cadmium oxalate) in specific paint regions that have before been unobserved. The imaging of organic materials has provided an insight into the extent of destruction of the original binding medium, as well as identifying unexpected organic materials in specific paint layers.

Copper-based pigments are common in works of art that show signs of decay on green and blue areas and are frequently associated with the degradation of organic substrates and/or media (drying oils, cellulose, etc.). The exact causes of degradation remain unknown. This prompted us to study possible starting and degradation products of one especially reactive copper pigment, verdigris (copper acetate), as well as pigments of the same family (salt and soap greens). Preparation of pigments using historical methods was followed by spectroscopic and crystallographic characterization using Fourier-transform infrared (FTIR) spectroscopy, X-ray diffraction (XRD), and X-ray photoelectron spectroscopy (XPS). Synthesis of verdigris and verdigris-like pigments resulted in a mixture of starting polymorphs of verdigris, including neutral and basic verdigris. With accelerated aging, pigments degraded to a polymorph of basic verdigris when not affected by organic media, whereas pigments on cellulosic substrates showed oxidized copper species. With this study, we are beginning to understand verdigris starting materials and highlight the complex interactions between pigments and substrates that influence pigment degradation pathways.

The interface of art and science provides a broad range of educational and collaborative projects at various learning stages. Therefore, the use of historic artists’ materials for teaching chemistry is receiving more attention. We prepared and used copper acetate (verdigris pigment) for a series of interconnected, lab-based activities, which can be applied to high-school-level chemistry, to undergraduate general chemistry, and further to heritage conservation science research for emerging art conservators. The synthesis and degradation processes of artists’ materials like this pigment allow instructors to illustrate scientific concepts like redox chemistry, while extending the vision of science to arenas beyond the classroom.

Copper-containing materials such as verdigris are commonly found in historic and artistic works of art, often at advanced states of decay. Applied on paper as inks and watercolors, many of which needed a binder such as gum arabic, the intrinsic instability of this pigment was known since the medieval period. The decay of verdigris (a mixture of copper acetates) as a pigment, as watercolor, and as a dye, was studied using a combination of vibrational (Fourier transform infrared) and X-ray photoelectron spectroscopy (XPS) and X-ray diffraction (XRD) instrumental techniques. Changes in the copper oxidative states and the formation of copper oxide were monitored during accelerated degradation as powdered solids and applied on mockup samples (with and without binder). Accelerated aging of both commercially available and synthesized verdigris pigments showed the presence of an intermediate species, Cu(CH3COO)2•3Cu(OH)2•2H2O, which points to the beginning of the decay processes, that culminates in the formation of Cu(II) oxide. However, the presence of gum arabic seems to delay deterioration, by temporarily reducing Cu(II) to Cu(I), even when the final product includes Cu(II). This novel application of XPS and supporting techniques has significant implications in art conservation, as the identified behavior helps explain the better preservation state of some works of art.

Cadmium sulfide (CdS) based yellow paint is fading, flaking, and discoloring with age in billions of dollars worth of Impressionist through Expressionist masterpieces from the late 19 th and early 20 th centuries. Characterization of the morphology, chemistry, and crystal structure of paint particles is critical for understanding CdS pigment degradation, and the role of other cadmium compounds in paint synthesis and aging [1]. Here, we use scanning transmission electron microscopy (STEM) to identify nanoparticle structures in a sample of cadmium yellow paint from the Edvard Munch's The Scream (c. 1910, Munch Museum), taken from a region of flaking yellow paint in the water adjacent to the two background figures on the bridge (Fig. 1a), and prepared for STEM by focused ion beam (FIB) milling.Spectroscopic mapping of the paint sample by electron energy loss spectroscopy (EELS) shows that although most of the particles in the sample are cadmium compounds (Fig. 1b), only a small minority contain sulfur (Figure 1c), suggesting that much of the CdS that would have originally been present at synthesis has degraded. EELS mapping reveals that a majority of paint particles in the sample consist mainly of cadmium carbonate (CdCO3). Compounds such as CdO, CdCl2, and CdCl(OH) may also be present, as observed in XEDS maps of a nearby FIB section, which showed similar carbonate/sulfide distributions, and in µXANES analysis of paint from the same region.Crystallographic datasets with a sub-nm probe were acquired using a high-speed pixel array detector [2]. Large (>10 nm) crystal grains in the sample were mapped using the asymmetry of diffraction patterns from datasets acquired at different sample tilts. Overlaying the crystal grain map and sulfur EELS signal shows that almost all CdS particles are located in regions with no large crystal grains (Fig. 2a). The diffraction patterns from CdS particles show a symmetric polycrystalline ring shape, indexed to cubic Hawleyite (Fig. 2b). From probe size and sample thickness, this suggests that the CdS consists of small (<10 nm) nanocrystalline particles; closer examination by ADF STEM (Fig. 2c) shows CdS nanoparticles in the range of 2-10 nm. CdS is highly photosensitive to oxidation [3], and CdS nanoparticles are particularly reactive due to their high surface area to volume ratio. This may be a key reason for the aging of cadmium yellow paint in the Scream. Given the higher reactivity of smaller nanoparticles, we hypothesize that smaller (~10 nm) and rounded CdCO3 grains are CdS aging products, while larger (~100 nm), well-faceted CdCO3 crystal grains were present from paint synthesis [4].