Mathieu Thoury

Emerging Approaches in Synchrotron Studies of Materials from Cultural and Natural History Collections

Synchrotrons have provided significant methods and instruments to study ancient materials from cultural and natural heritages. New ways to visualise (surfacic or volumic) morphologies are developed on the basis of elemental, density and refraction contrasts. They now apply to a wide range of materials, from historic artefacts to paleontological specimens. The tunability of synchrotron beams owing to the high flux and high spectral resolution of photon sources is at the origin of the main chemical speciation capabilities of synchrotron-based techniques. Although, until recently, photon-based speciation was mainly applicable to inorganic materials, novel developments based, for instance, on STXM and deep UV photoluminescence bring new opportunities to study speciation in organic and hybrid materials, such as soaps and organometallics, at a submicrometric spatial resolution over large fields of view. Structural methods are also continuously improved and increasingly applied to hierarchically structured materials for which organisation results either from biological or manufacturing processes. High-definition (spectral) imaging appears as the main driving force of the current trend for new synchrotron techniques for research on cultural and natural heritage materials.

Use of imaging spectroscopy, fiber optic reflectance spectroscopy, and X-ray fluorescence to map and identify pigments in illuminated manuscripts

Abstract A paradigm using multispectral visible and near-infrared imaging spectroscopy is presented to semi-automatically create unbiased spectral maps that guide the site selection for in situ analytical methods (e.g. fiber optic reflectance spectroscopy and X-ray fluorescence) in order to identify and map pigments in illuminated manuscripts. This approach uses low spectral resolution imaging spectroscopy to create maps of areas having the same spectral characteristics. This paradigm is demonstrated by analysis of the illuminated manuscript leaf Christ in Majesty with Twelve Apostles (workshop of Pacino di Buonaguida, c. 1320). Using this approach the primary pigments are mapped and identified as azurite, lead-tin yellow, red lead, a red lake (likely insect-derived), a copper-containing green, brown iron oxide, and lead white. Moreover, small amounts of natural ultramarine were found to be used to enhance the blue fields around Christ, and a red lake was used to highlight different colors. These results suggest that the proposed paradigm offers an improved approach to the comprehensive study of illuminated manuscripts by comparison with site-specific analytical methods alone. The choice of broad spectral bands proves successful, given the limited palette in illuminated manuscripts, and permits operation at the low light intensity required for examination of manuscripts.

Near-Infrared Luminescence of Cadmium Pigments: In Situ Identification and Mapping in Paintings

A comprehensive study of the luminescence properties of cadmium pigments was undertaken to determine whether these properties could be used for in situ identification and mapping of the pigments in paintings. Cadmium pigments are semiconductors that show band edge luminescence in the visible range and deep trap luminescence in the red/infrared range. Emission maxima, quantum yields, and excitation spectra from the band edge and deep trap emissions were studied for sixty commercial cadmium pigments that span the color range from yellow to red (reflectance transition 470 to 660 nm). For paints containing cadmium pigments, luminescence from deep traps was more readily observable than that from the band edge, although the yield varied widely from zero to around 4.5%. Optimal excitation for emission is found to be in the visible for both pigments in powder form and mixed with a medium. The maxima of the deep trap emission shift with the band gap energy, providing a potentially useful way to assign pigment type even when used in pigment mixtures. The usefulness of the results of the study on mockups was demonstrated by the mapping of cadmium pigments of different hues with the aid of calibrated luminescence imaging spectroscopy in a painting by Edward Steichen, entitled Study for ‘Le Tournesol’ (1920). Analysis of the luminescence image cube reveals at least six unique spectral components, associated with emission from white pigments, paint binder, and cadmium red and yellow pigments. The results were compared with those from X-ray fluorescence spectrometry (XRF) and fiber-optic reflection spectroscopy (FORS) and the results obtained on paint samples containing cadmium pigments. These results show that, when present, the emission from traps can be used as an analytical tool to identify cadmium pigments, to distinguish among cadmium sulfide, cadmium zinc sulfide, and cadmium sulfoselenide, and to map cadmium pigments, even in mixtures.

Nondestructive varnish identification by ultraviolet fluorescence spectroscopy.

The identification of the chemical nature of varnish is essential for art restorers in order to choose a suitable solvent during its removal. Until today, such identification has been performed using chemical analysis after sampling. An innovative technique is presented here, using ultraviolet (UV) fluorescence spectroscopy. The method is nondestructive, workable in situ, and leads to results in real time. It is based on the comparison between the emission spectrum of an unknown varnish with those of fresh, artificially aged, or old reference resins and varnishes, for different monochromatic excitation wavelengths. The resin and the nature of the varnish as spirit, oil, or mixed can thus be identified. Various examples are presented on home-made samples applied on fluorescent backgrounds and on real works of art.

Excitation emission and time-resolved fluorescence spectroscopy of selected varnishes used in historical musical instruments.

The analysis of various varnishes from different origins, which are commonly found on historical musical instruments was carried out for the first time with both fluorescence excitation emission spectroscopy and laser-induced time-resolved fluorescence spectroscopy. Samples studied include varnishes prepared using shellac, and selected diterpenoid and triterpenoid resins from plants, and mixtures of these materials. Fluorescence excitation emission spectra have been collected from films of naturally aged varnishes. In parallel, time-resolved fluorescence spectroscopy of varnishes provides means for discriminating between short- (less than 2.0 ns) and long-lived (greater than 7.5 ns) fluorescence emissions in each of these complex materials. Results suggest that complementary use of the two non destructive techniques allows a better understanding of the main fluorophores responsible for the emission in shellac, and further provides means for distinguishing the main classes of other varnishes based on differences in fluorescence lifetime behaviour. Spectrofluorimetric data and time resolved spectra presented here may form the basis for the interpretation of results from future in situ fluorescence examination and time resolved fluorescence imaging of varnished musical instruments.

Visible and Infrared Reflectance Imaging Spectroscopy of Paintings: Pigment Mapping and Improved Infrared Reflectography

Reflectance imaging spectroscopy, the collection of images in narrow spectral bands, has been developed for remote sensing of the Earth. In this paper we present findings on the use of imaging spectroscopy to identify and map artist pigments as well as to improve the visualization of preparatory sketches. Two novel hyperspectral cameras, one operating from the visible to near-infrared (VNIR) and the other in the shortwave infrared (SWIR), have been used to collect diffuse reflectance spectral image cubes on a variety of paintings. The resulting image cubes (VNIR 417 to 973 nm, 240 bands, and SWIR 970 to 1650 nm, 85 bands) were calibrated to reflectance and the resulting spectra compared with results from a fiber optics reflectance spectrometer (350 to 2500 nm). The results show good agreement between the spectra acquired with the hyperspectral cameras and those from the fiber reflectance spectrometer. For example, the primary blue pigments and their distribution in Picassos Harlequin Musician (1924) are identified from the reflectance spectra and agree with results from X-ray fluorescence data and dispersed sample analysis. False color infrared reflectograms, obtained from the SWIR hyperspectral images, of extensively reworked paintings such as Picassos The Tragedy (1903) are found to give improved visualization of changes made by the artist. These results show that including the NIR and SWIR spectral regions along with the visible provides for a more robust identification and mapping of artist pigments than using visible imaging spectroscopy alone.

Extension of the Kubelka-Munk theory for fluorescent turbid media to a nonopaque layer on a background

Photoluminescence is one of the processes by which photons are emitted after the absorption of incoming photons at a higher energy. But the yield and spectral band shape of the emission can be altered by the optical properties of the luminophore environment through scattering and absorption. To understand these effects on a photoluminescent turbid layer, the Kubelka–Munk model, which is a two-flux approximation of the radiative transfer equation, can be used. Compared to previous works, this translucent layer can be applied on a colored opaque background. The model takes into account the absorption, scattering, and luminescent properties of the layer and the reflection by the background, for both the light excitation and the light emission. The competition between these different optical interactions is studied; e.g., the model can predict the presence of an emission maximum by increasing the thickness of the luminescent layer on a light background. Moreover, the model is extended to two important cases: the presence of a photoluminescent background and the effect of a refractive index discontinuity.

High spatial dynamics-photoluminescence imaging reveals the metallurgy of the earliest lost-wax cast object

Photoluminescence spectroscopy is a key method to monitor defects in semiconductors from nanophotonics to solar cell systems. Paradoxically, its great sensitivity to small variations of local environment becomes a handicap for heterogeneous systems, such as are encountered in environmental, medical, ancient materials sciences and engineering. Here we demonstrate that a novel full-field photoluminescence imaging approach allows accessing the spatial distribution of crystal defect fluctuations at the crystallite level across centimetre-wide fields of view. This capacity is illustrated in archaeology and material sciences. The coexistence of two hitherto indistinguishable non-stoichiometric cuprous oxide phases is revealed in a 6,000-year-old amulet from Mehrgarh (Baluchistan, Pakistan), identified as the oldest known artefact made by lost-wax casting and providing a better understanding of this fundamental invention. Low-concentration crystal defect fluctuations are readily mapped within ZnO nanowires. High spatial dynamics-photoluminescence imaging holds great promise for the characterization of bulk heterogeneous systems across multiple disciplines.

Experimental study on merits of virtual cleaning of paintings with aged varnish

To assess the accuracy of virtual cleaning of Old Master paintings (i.e. digital removal of discolored varnishes), a physical model was developed and experimentally tested using reflectance imaging spectroscopy on three paintings undergoing conservation treatment. The model predicts the reflectance spectra of the painting without varnish or after application of a new varnish from the reflectances of the painting with the aged varnish, given the absorption of the aged varnish and the scattering terms. The resulting color differences between the painting actually and virtually cleaned can approach the perceivable limit. Residual discrepancies are ascribable to spatial variations in the characteristics of the aged varnish (scattering, optical thickness) and the exposed painting (surface roughness).

Implementation of a neural network for multispectral luminescence imaging of lake pigment paints.

Luminescence multispectral imaging is a developing and promising technique in the fields of conservation science and cultural heritage studies. In this article, we present a new methodology for recording the spatially resolved luminescence properties of objects. This methodology relies on the development of a lab-made multispectral camera setup optimized to collect low-yield luminescence images. In addition to a classic data preprocessing procedure to reduce noise on the data, we present an innovative method, based on a neural network algorithm, that allows us to obtain radiometrically calibrated luminescence spectra with increased spectral resolution from the low-spectral resolution acquisitions. After preliminary corrections, a neural network is trained using the 15-band multispectral luminescence acquisitions and corresponding spot spectroscopy luminescence data. This neural network is then used to retrieve a megapixel multispectral cube between 460 and 710 nm with a 5 nm resolution from a low-spectral-resolution multispectral acquisition. The resulting data are independent from the detection chain of the imaging system (filter transmittance, spectral sensitivity of the lens and optics, etc.). As a result, the image cube provides radiometrically calibrated emission spectra with increased spectral resolution. For each pixel, we can thus retrieve a spectrum comparable to those obtained with conventional luminescence spectroscopy. We apply this method to a panel of lake pigment paints and discuss the pertinence and perspectives of this new approach.