Abstract: The painter, Vincent van Gogh, and some of his contemporaries frequently made use of the pigment chrome yellow that is known to show a tendency toward darkening. This pigment may correspond to various chemical compounds such as PbCrO4 and PbCr1-xSxO4, that may each be present in various crystallographic forms with different tendencies toward degradation. Investigations by X-ray diffraction (XRD), mid-Fourier Transform infrared (FTIR), and Raman instruments (benchtop and portable) and synchrotron radiation-based micro-XRD and X-ray absorption near edge structure spectroscopy performed on oil-paint models, prepared with in-house synthesized PbCrO4 and PbCr1-xSxO4, permitted us to characterize the spectroscopic features of the various forms. On the basis of these results, an extended study has been carried out on historic paint tubes and on embedded paint microsamples taken from yellow-orange/pale yellow areas of 12 Van Gogh paintings, demonstrating that Van Gogh effectively made use of different chrome yellow types. This conclusion was also confirmed by in situ mid-FTIR investigations on Van Goghs Portrait of Gauguin (Van Gogh Museum, Amsterdam).

Abstract: On several paintings by artists of the end of the 19th century and the beginning of the 20th Century a darkening of the original yellow areas, painted with the chrome yellow pigment (PbCrO4, PbCrO4·xPbSO4, or PbCrO4·xPbO) is observed. The most famous of these are the various Sunflowers paintings Vincent van Gogh made during his career. In the first part of this work, we attempt to elucidate the degradation process of chrome yellow by studying artificially aged model samples. In view of the very thin (1−3 μm) alteration layers that are formed, high lateral resolution spectroscopic methods such as microscopic X-ray absorption near edge (μ-XANES), X-ray fluorescence spectrometry (μ-XRF), and electron energy loss spectrometry (EELS) were employed. Some of these use synchrotron radiation (SR). Additionally, microscopic SR X-ray diffraction (SR μ-XRD), μ-Raman, and mid-FTIR spectroscopy were employed to completely characterize the samples. The formation of Cr(III) compounds at the surface of the chrome yellow paint layers is particularly observed in one aged model sample taken from a historic paint tube (ca. 1914). About two-thirds of the chromium that is present at the surface has reduced from the hexavalent to the trivalent state. The EELS and μ-XANES spectra are consistent with the presence of Cr2O3·2H2O (viridian). Moreover, as demonstrated by μ-XANES, the presence of another Cr(III) compound, such as either Cr2(SO4)3·H2O or (CH3CO2)7Cr3(OH)2 [chromium(III) acetate hydroxide], is likely.

Abstract: A novel electrothermal atomic absorption spectrometry (ETAAS) method was developed for the halogenation of refractory sample components (Er, Nd and Nb) of lithium niobate (LiNbO3) and bismuth tellurite (Bi2TeO5) optical single crystals to overcome memory effects and carry-over. For this purpose, the cleaning step of a regular graphite furnace heating program was replaced with a halogenation cycle. In this cycle, after the graphite tube cooled to room temperature, a 20 μL aliquot of liquid carbon tetrachloride (CCl4) was dispensed with a conventional autosampler into the graphite tube. The CCl4 was partially dried at 80 °C under the mini-flow (40 cm3 min−1) condition of the Ar internal furnace gas (IFG), then the residue was decomposed (pyrolyzed) by fast furnace heating at 19002100 °C under interrupted flow of the IFG. This step was followed by a clean-out stage at 2100 °C under the maximum flow of the IFG. The advantage of the present method is that it does not require any alteration to the graphite furnace gas supply system in contrast to most of the formerly introduced halogenation techniques. The effectiveness of the halogenation method was verified with the determination of Er and Nd dopants in the optical crystals. In these analyses, a sensitivity decrease was observed, which was likely due to the enhanced deterioration of the graphite tube surface. Therefore, the application of mathematical correction (resloping) of the calibration was also required. The calibration curves were linear up to 1.5 and 10 μmol L−1 for Er and Nd, respectively. Characteristic masses of 18 and 241 pg and the limit of detection (LOD) values of 0.017 and 0.27 μmol L−1 were found for Er and Nd, respectively. These LOD data correspond to 0.68 μmol mol−1 Er and 11 μmol mol−1 Nd in solid bismuth tellurite samples. The analytical results were compared with those obtained by a conventional ETAAS method and validated with X-ray fluorescence spectrometry analysis.