The sixteenth and seventeenth-century Venetian manuscripts on glassmaking contain a number of recipes on yellow pigment production, later named 'anime', based on Pb-Sn and/or Pb-Sb oxides. Since antiquity, these oxides have been employed intermittently as opacifying and colouring agents to obtain coloured glasses, enamels and pigments.In 1998, different laboratories provided evidence of the existence of a 'non-standardized' yellow pigment (Pb-Sn-Sb triple oxide) very close in composition and structure to both the Pb-Sn type II (silicoestannate) and Pb-Sb (antimonate) oxides. This Pb-Sn-Sb yellow pigment was identified by SEM-EDX in seventeenth-century Italian paintings.In previous studies, the authors of the present work have investigated with Raman spectroscopy the presence of Pb-Sn-Sb oxide (pyrochlore structure) in some Italian paintings of the seventeenth century, using as standard a synthesized pyrochloric triple oxide (Pb 2 SnSbO 6.5 ).In this paper, we provide evidence for the first molecular identification with Raman spectroscopy of said pigment in Italian pottery from the sixteenth century. We have identified this pigment in two pieces of decorated pottery: a 'plate' manufactured in Casteldurante and attributed to Andrea Negroponte style (c 1550), and a Venetian 'albarello' (pharmacy jar) attributed to Domenego (c 1570-1580). We have synthesized the reference compound in accordance with the theoretical formula of the pyrochloric triple oxide (Pb 2 SnSbO 6.5 ). The Raman spectra obtained with this material provided molecular information identical to the yellow pigment measured in the two Italian ceramics. To conclude, the experimental results presented here indicate one simple fact: Pb-Sn-Sb triple oxide yellow pigment has been used during the sixteenth and seventeenth centuries without distinction both in Italian ceramics and paintings.

This investigation is focused on the identification in ceramic artworks of certain nonstandard yellow/orange pigments whose composition is based, fundamentally, on lead, tin, and antimony oxides with or without silica. In this work, a comparative study (temporal and geographical) of the employment of these yellow pigments in different production centers, from Italy (Pesaro and Montelupo) and Spain (Talavera de la Reina), during the Renaissance and Baroque epochs has been proposed. For this purpose, special very ancient yellow pigments were acquired from the Stazione Sperimentale del Vetro, Murano-Venezia (Italy). These reference pigments have been produced following strict and rigorous manufacturing recipes corresponding to Venezian fabrication processes for the 16th and 17th centuries. On the other hand, the portable characteristic of a new optical fiber Raman system has allowed us the access into the Museo de Cerámica de Barcelona in order to investigate the composition of the yellow and orange colors of an important ceramic collection belonging to this museum. The results are in good agreement with the ones obtained by other authors who have investigated the same topic. It is notable, first, the excellent performances that this portable Raman system offers in the direct and non-invasive analysis of ceramic artworks and, second, the coincidences of the molecular results among these yellow pigments. This fact confirms that these pigments were commonly used either in Italian and Spanish ceramic objects during both Renaissance and Baroque epochsPeer ReviewedPostprint (published version

This paper investigates, using only Raman spectroscopy, the spectral differences among the synthetic ultramarine blue (J. B. Guimet) and three natural ultramarine blues (lazurites) corresponding to three lapis lazulis from different geographical sources (Chile, Afghanistan and Siberia). The synthetic ultramarine blue used as reference pigment is the identified in an artwork created around 1917 by the Catalonian modernist painter Ramon Casas i Carbó. This synthetic pigment is compared with the three natural ultramarine blues and, in his turn, the latter are also compared each other. In order to quantify the spectral differences, a rigorous methodology based on the Euclidean distance between homologue spectral segments (or intervals) has been developed and presented here.

Fluorescence, due to the binding media, is the main problem that one encounters in the molecular analysis of artistic pigments using Raman spectroscopy. With the object of minimizing this problem, we propose the use of a semiconductor IR laser in Raman spectral acquisition and the application of local irradiation with a pulsed UV laser on the analysis zone. The Raman analysis with an IR source shows advantages compared to that with visible lasers, such as the reduction of fluorescence and its shot noise, although a price has to be paid in the form of the small intensity of the obtained spectra with the same acquisition time. Also, in this paper we demonstrate that controlled levels of pulsed UV radiation over the analyzed painting (pigment + binding media) can improve even more the Raman spectral quality obtained with the IR laser, which leads us to conclude that the local use of a pulsed UV laser, prior to Raman analysis with the IR source, can bring optimum quality results in pigment identification. The spectral quality of these experimental results for different pigments has been measured by calculating in each case the signal-to-noise ratio (SNR) of the corresponding spectra. From a quantitative point of view, in some practical cases (chromium yellow and ultramarine blue) an SNR improvement of 16 dB is achieved when pulsed UV irradiation and IR laser Raman analysis are used instead of a simple Raman analysis with a visible laser.

The fundamental idea of the present work concerns photon technology performance. On the one hand, in order to improve the Raman spectral quality when pigment analysis is carried out, we use UV pulsed laser radiation (266 nm) and its non-thermal photoablation-induced effect. In this way, the unwanted superficial layers (old varnishes, for example, in easel paintings) can be locally removed and so a reduction of the total fluorescence in the spectrum of the analyzed zone is achieved. On the other hand, for the purpose of analyzing the quality of the process, we employ Raman spectroscopy to obtain molecular information about the previous and later compositions of the surface irradiated with UV pulses. The fluorescence reduction of the Raman spectra and its consequent spectral quality improvement are quantified by a proposed figure of merit (signal-to-noise ratio) which is calculated in a suitable spectral range. The UV pulses are obtained starting an Nd : YAG pulsed laser (1064 nm) in combination with non-linear optical crystals, and the Raman system used in the analytical measurements is a Jobin Yvon Induram with optical fiber technology and an He-Ne continuous laser (632.8 nm). A Haas shutter guarantees the spatial focusing of the optical pulses over a concrete point of the irradiated sample. Based on the experimental results, two conclusions can be established: first, non-thermal photoablation is very effective at removing small areas of the superficial varnished layers; second, Raman spectroscopy and the proposed figure of merit are very useful for measuring the resulting improvement in spectral quality.