•The painting ‘Saul and David’ was examined by Macroscopic-XRF and hyperspectral imaging.•The results show where in the painting the Cobalt-containing pigment smalt was used.•The results also show which areas of the painting are not original•Results show that Rembrandt employed two types of smalt, probably during two different periods, a finding consistent with quantitative analyses of paint micro samples.•For the first time, the occurrence of different types of smalt in the same painting was demonstrated by means of non-destructive imaging methods.The painting Saul and David, considered to date from c. 1652 and previously attributed to Rembrandt van Rijn and/or his studio, is a complex work of art that has been recently subjected to intensive investigation and conservation treatment. The goal of the research was to give insight into the painting's physical construction and condition in preparation for conservation treatment. It was also anticipated that analysis would shed light on authenticity questions and Rembrandt's role in the creation of the painting. The painting depicts the Old Testament figures of King Saul and David. At left is Saul, seated, holding a spear and wiping a tear from his eye with a curtain. David kneels before him at the right playing his harp. In the past, the large sections with the life-size figures were cut apart and later reassembled. A third piece of canvas was added to replace a missing piece of canvas above the head of David. As part of the investigation into the authenticity of the curtain area, a number of paint micro samples were examined with LM and SEM–EDX. Given that the earth, smalt and lake pigments used in the painting could not be imaged with traditional imaging techniques, the entire painting was also examined with state of the art non-destructive imaging techniques. Special attention was devoted to the presence of cobalt-containing materials, specifically the blue glass pigment smalt considered characteristic for the late Rembrandt. A combination of quantitative electron microprobe analysis and macroscopic X-ray fluorescence scanning revealed that three types of cobalt-containing materials are present in the painting. The first type is a cobalt drier that was found in the overpaint used to cover up the canvas inset and the joins that were added in the 19th century. The other two Co-containing materials are part of the original paint used by Rembrandt and comprise two varieties of smalt, a K-rich glass pigment that derives its gray–blue color by doping with Co-ions. Smalt paint with a higher Ni content (NiO:CoO ratio of around 1:4) was used to depict the blue stripes in Saul's colorful turban, while smalt with a lower Ni content was employed (NiO:CoO ratio of around 1:5) for the broad expanses of Saul's garments. The presence of two types of smalt not only supports the recent re-attribution of the painting to Rembrandt, but also that the picture was painted in two phases. Saul's dark red garment is painted in a rough, “loose” manner and the now discolored smalt-rich layer was found to have been partially removed during a past restoration treatment/s. In contrast, the blue-green smalt in the turban is much better preserved and provides a colorful accent. While the use of different types of smalt in a Rembrandt painting has been previously identified using quantitative EDX analysis of paint cross-sections, to the best of our knowledge this is the first time such a distinction has been observed in a 17th-century painting using non-destructive imaging techniques. In addition to the XRF-based non-invasive elemental mapping, hyperspectral imaging in the visual to near-infrared (VNIR) region was also carried out.

•Macro XRF scanning is used to examine a 15th c. Italian manuscript fragment•MA-XRF data are compared with other single-point analytical techniques•The interpretation of the elemental maps is discussed with the challenges faced•We identify two unusual pigments: an arsenic sulphide glass and a manganese oxide•The use of manganese oxide helps place the fragment within a specific artistic milieuMacro X-ray fluorescence scanning (MA-XRF) is gradually becoming an established technique for the non-invasive analytical investigation of painted surfaces. This paper discusses some of the benefits and limitations of employing MA-XRF for the study of manuscript illuminations. Art historical research on this type of artefacts that is based on scientific measurements is often limited by the fact that usually no sampling can take place. Hence there is a need for non-invasive analytical tools that make it possible to conduct systematic investigations. As a representative example of this type of objects, a 15th century Italian manuscript fragment from the collection of the Fitzwilliam Museum in Cambridge (UK) is investigated. The aims of the study were to gain insight into the materials and techniques employed by Renaissance illuminators and to help answer specific questions regarding the fragment’s authorship and geographic origin. The complementarity and advantages of MA-XRF mapping versus site-specific analyses are discussed. For this purpose, MA-XRF data are evaluated and compared with the results of other analytical techniques. The interpretation of the elemental maps is discussed along with the challenges faced during the analysis.

•Several historical artists' pigments were considered as semiconductors.•The (in)stability of the pigments was theoretically predicted using a thermodynamic approach.•The approach easily and quickly screens semiconductor photostability.•Theoretical predictions correspond to experimental data.Several artists' pigments are semiconductors. Some of these materials appear stable over time, whereas others already show remarkable signs of degradation after limited time periods. The (in)stability of these pigments can be understood using a thermodynamic approach. For several pigment-related materials, the thermodynamic oxidation and reduction potential (ϕox and ϕred) were determined and evaluated considering the absolute energy positions of the valence and conduction band edges and the water redox potentials. The positions of ϕox and ϕred can be used in a fast screening of the stability of semiconductor pigments towards photoinduced corrosion in an aqueous/humid environment. This theoretical approach corresponds well with experimental data on pigment permanence and degradation phenomena found in literature.

Technical progress in the fields of X-ray sources, optics and detectors is constantly enhancing the pace of data acquisition in XRF imaging. This enlarges the size of the hyperspectral datasets and the number of their sub-parts. This paper describes the challenges in processing large XRF datasets featuring several million pixels/spectra and the strategies developed to overcome them. During the investigation of historical paintings by scanning macro-XRF the main challenges are the correct identification of all spectral features in a dataset and its timely processing. For the identification of spectral features different approaches are discussed, i.e. the use of sum spectra, maximum pixel spectra and of χr2 maps. For the time-efficient, artefact-free evaluation of XRF imaging data, different software packages are evaluated and intercompared (AXIL, PyMCA, GeoPIXE and the in-house written datamuncher). The process of data evaluation is illustrated on a large dataset (3.4 MPixels) acquired during the investigation of a version of Caravaggio's Supper at Emmaus (143 × 199.5 cm2). This 17th century painting is currently the largest object entirely scanned with macroscopic XRF.

In this paper we demonstrate that by means of scanning reflection FTIR spectroscopy, it is possible to record highly specific distribution maps of organic and inorganic compounds from flat, macroscopic objects with cultural heritage value in a non-invasive manner. Our previous work involved the recording of macroscopic distributions of chemical elements or crystal phases from painted works of art based on respectively macroscopic X-ray fluorescence or X-ray powder diffraction analysis. The use of infrared radiation instead of X-rays has the advantage that more specific information about the nature and distribution of the chemical compounds present can be gathered. This higher imaging specificity represents a clear advantage for the characterization of painting and artist materials. It allows the distribution of metallo-organic compounds to be visualized and permits distinguishing between pigmented materials containing the same key metal. The prototype instrument allows the recording of hyperspectral datacubes by scanning the surface of the artefact in a contactless and sequential single-point measuring mode, while recording the spectrum of reflected infrared radiation. After the acquisition, spectral line intensities of individual bands and chemical distribution maps can be extracted from the datacube to identify the compounds present and/or to highlight their spatial distribution. Not only is information gained on the surface of the investigated artefacts, but also images of overpainted paint layers and, if present, the underdrawing may be revealed in this manner. A current major limitation is the long scanning times required to record these maps.

Technical progress in the fields of X-ray sources, optics and detectors is constantly enhancing the pace of data acquisition in XRF imaging. This enlarges the size of the hyperspectral datasets and the number of their sub-parts. This paper describes the challenges in processing large XRF datasets featuring several million pixels/spectra and the strategies developed to overcome them. During the investigation of historical paintings by scanning macro-XRF the main challenges are the correct identification of all spectral features in a dataset and its timely processing. For the identification of spectral features different approaches are discussed, i.e. the use of sum spectra, maximum pixel spectra and of χr2 maps. For the time-efficient, artefact-free evaluation of XRF imaging data, different software packages are evaluated and intercompared (AXIL, PyMCA, GeoPIXE and the in-house written datamuncher). The process of data evaluation is illustrated on a large dataset (3.4 MPixels) acquired during the investigation of a version of Caravaggio's Supper at Emmaus (143 × 199.5 cm2). This 17th century painting is currently the largest object entirely scanned with macroscopic XRF.