Simone Cagno

Revealing hidden paint layers in oil paintings by means of scanning macro-XRF: a mock-up study based on Rembrandt's “An old man in military costume”

Over the past several decades the oeuvre of Rembrandt has been the subject of extensive art historical and scientific investigations. One of the most striking features to emerge is his frequent re-use of canvases and panels. The painting An Old Man in Military Costume (78.PB.246), in the collection of the J. Paul Getty Museum, is an example of such a re-used panel. Conventional imaging techniques revealed the presence of a second portrait under the surface portrait, but the details of this hidden portrait have not yet been revealed. Vermilion (HgS) has been identified to have been used nearly exclusively in the flesh tones of the lower painting, suggesting that element-specific XRF imaging might successfully image the hidden portrait. To test this hypothesis, a full-scale mock-up of the painting was created, including a “free impression” of the hidden portrait, reproducing as closely as possible the pigments and paint stratigraphy of the original painting. XRF imaging of the mock-up painting was conducted using three different XRF imaging systems: a mobile X-ray tube based system and two synchrotron-based setups (one equipped with multiple SDDs and one equipped with a Maia detector). The sensitivity, limits of detection and imaging capabilities of each system under the chosen experimental conditions are evaluated and compared. The results indicate that an investigation of the original painting by this method would have an excellent chance of success.

Study on the impregnation of archaeological waterlogged wood with consolidation treatments using synchrotron radiation microtomography

In favourable conditions of low temperature and low oxygen concentration, archaeological waterlogged wooden artefacts, such as shipwrecks, can survive with a good state of preservation. Nevertheless, anaerobic bacteria can considerably degrade waterlogged wooden objects with a significant loss in polysaccharidic components. Due to these decay processes, wood porosity and water content increase under ageing. In such conditions, the conservation treatments of archaeological wooden artefacts often involve the replacement of water with substances which fill the cavities and help to prevent collapse and stress during drying. The treatments are very often expensive and technically difficult, and their effectiveness very much depends on the chemical and physical characteristics of the substances used for impregnation. Also important are the degree of cavity-filling, penetration depth and distribution in the structure of the wood. In this study, the distribution in wood cavities of some mixtures based on polyethylene glycols and colophony, used for the conservation of waterlogged archaeological wood, was investigated using synchrotron radiation X-ray computed microtomography (SR-µCT). This non-destructive imaging technique was useful for the study of the degraded waterlogged wood and enabled us to visualise the morphology of the wood and the distribution of the materials used in the wood treatments. The study has shown how deposition is strictly related to the dimension of the wooden cavities. The work is currently proceeding with the comparison of synchrotron observations with the data of the solutions viscosity and with those of the properties imparted to the wood by the treatments.

Compositional analysis of Tuscan glass samples: in search of raw material fingerprints

SEM-EDX quantitative analyses were performed on 91 samples of vessel glass originating from a number of sites throughout Tuscany. A comparison with Venetian glass data is made in order to evaluate differences in glassmaking materials. The compositions of the samples provide clues about the different raw materials used in relation to recipes that changed slightly during the period covering the late Middle Ages to early Renaissance. Two kinds of ash used as a fluxing agent were identified, one of which is of east Mediterranean origin while the other is likely to be Spanish. Two types of silica sources were also found. One of these was recognized as being quartziferous river pebbles, while the other is proposed to be local sand.

Evaluation of manganese-bodies removal in historical stained glass windows via SR-μ-XANES/XRF and SR-μ-CT

The speed and effectiveness of a conservation treatment used for stained glass windows have been investigated. Dark-coloured Mn-rich stains can be found in the alteration layer of ancient glass artefacts and cause the surface to turn brown/black: this phenomenon is known as Mn-browning or Mn-staining. While in glass manganese is present in the +II or +III oxidation states, in the Mn-rich bodies, manganese is in a higher oxidation state (+IV). In restoration practice, mildly reducing solutions are employed to eliminate the dark colour and restore the clear appearance of the glass. In this paper the effectiveness and side effects of the use of hydroxylamine hydrochloride for this purpose are assessed. Archaeological fragments of stained glass windows, dated to the 14th century and originating from Sidney Sussex College, Cambridge (UK), were examined by means of synchrotron radiation (SR) based microscopic X-ray Absorption Near-Edge Spectroscopy (μ-XANES) and microscopic X-Ray Fluorescence (μ-XRF) and with high resolution computed absorption tomography (μ-CT) before, during and after the treatment. The monitoring of the glass fragments during the treatment allows us to better understand the manner in which the process unfolds and its kinetics. The results obtained reveal that the hydroxylamine hydrochloride treatment is effective, but also that it has a number of unwanted side effects. These findings are useful for optimizing the time and other modalities of the Mn-reducing treatment as well as minimizing its unwanted results.

Iron speciation in soda-lime-silica glass: a comparison of XANES and UV-vis-NIR spectroscopy

Scientific analyses of ancient glasses have been carried out for many years using elemental chemical analysis. However, it is known that the control of the redox conditions in the glass melt has a strong implication on the final hue of glass because it affects Fe2+/ΣFe. Therefore an increasing number of studies on the redox conditions have been published in recent years by means of synchrotron based X-ray absorption spectroscopy. This is a technique which is not easily accessible and requires dedicated facilities. In this paper we describe an alternative approach by means of optical absorption spectroscopy. We synthesised 10 soda-lime-silica glasses with known redox conditions and iron concentration to calibrate the absorption at 1100 nm as a function of Fe2+ concentration. The linear extinction coefficient was also determined. These glasses were also studied by means of X-ray Absorption Near Edge Structure (XANES) spectroscopy. Electron paramagnetic resonance spectroscopy was additionally used as an ancillary method to verify the quality of our data. Furthermore 28 samples from real archaeological samples were analysed by XANES and optical spectroscopy as a case study. The Fe2+/ΣFe values obtained were compared and demonstrated that the two techniques were in good agreement with each other. Optical spectroscopy can be applied in situ with moderate sample preparation to determine the concentration of Fe2+. To investigate the redox conditions, especially as a first screening approach, this methodology is an important tool to take into consideration before applying more sophisticated techniques such as XANES, which is more elaborate and requires high-tech resources.

The identification of chromophores in ancient glass by the use of UV-VIS-NIR spectroscopy

In this publication optical spectroscopy is considered to be a supplementary technique to study ancient colored glass. It results from a systematic study of the UV-VIS-NIR transmission spectra of intentionally colored glass fragments from various archaeological and historical sites and dated from the Roman period to the 21th century AD. The main goal consists of defining optical sensing parameters for this type of material. The considered colorants are iron, cobalt, manganese, copper and chromium. It is proved that many cases exist where optical spectroscopy can be seen as a straightforward, non-destructive, low-cost and in-situ applicable technique in identifying authentic material or to obtain information about the origin of the material. Possible sensing parameters are defined as the absence/presence of absorption bands characteristic for a specific coloring metal oxide and the spectral position of these bands. These parameters could reveal information about the applied furnace conditions and/or to the composition of the glass matrix. It is shown that the cobalt absorption band situated around 535 nm for soda rich glasses (Roman and industrial times) is shifted towards 526 nm for potash rich glasses (medieval and post-medieval times).

A XANES study of chromophores: the case of black glass

We studied the Fe K-edge X-ray absorption near edge (XANES) spectra of several Roman black glass fragments in order to determine the Fe3+/ΣFe ratio of these materials. The selected archaeological glass samples cover the period 1st–5th century AD in nine different sites of the North Western provinces of the Roman Empire. The fragments belong to two different compositional groups demonstrating a diachronic evolution: early Roman HMG (High Magnesia Glass) and Roman Imperial LMG (Low Magnesia Glass). The first group contains natural Fe levels (below 2 wt% as Fe2O3), while the LMG has concentrations above 5 wt%. This difference is also reflected by Fe3+/ΣFe values. Low iron glass was produced under strongly reducing conditions in order to obtain the black colour, with average Fe3+/ΣFe values ≈ 0.17. LMG glass is somewhat more oxidised (Fe3+/ΣFe ≈ 0.4–0.5). While HMG glass required active control of the furnace environment, LMG was made under ambient atmosphere and its higher oxidation degree is mainly determined by the chemistry of the raw glass.

Micro-XANES study on Mn browning: use of quantitative valence state maps

Historical glass, especially non-durable medieval glass, can undergo corrosion. This sometimes results in the formation of dark-coloured manganese-rich inclusions that reduce the transparency of the glass. While unaltered bulk glass contains manganese mainly present in the +II valence state, inside the inclusions Mn is present in higher valence states (+III to +IV). Two different strategies may be considered by conservators when aiming to improve the transparency. One is based on the reduction of highly oxidised black/brown compounds using mildly reducing solutions, while the other focuses more on the extraction of manganese from the inclusions by the application of chelating agents. In this paper, a method for quantitative mapping of the Mn speciation inside partially corroded historical windowpanes based on X-Ray Absorption Near-Edge Structure (XANES) spectroscopy is discussed. The calibration of such Mn valence state maps based on the combo method, a fairly reliable way to determine the oxidation state, is described in more detail. This method is used to evaluate the effect of reducing treatments on historical glass, dated to the 14th century and originating from Sidney Sussex College (Cambridge, UK), suffering from Mn browning. Glasses were examined by means of Synchrotron Radiation (SR) based microscopic X-Ray Absorption Near-Edge Structure (μXANES) spectroscopy and microscopic X-Ray Fluorescence (μXRF). X-Ray elemental distribution maps of glass cross-sections are recorded at different energies, while Mn K-edge spectra are used to convert these into Mn valence state (VS) maps. Such valence state maps will allow evaluation of a reducing treatment.

Composition of Facon de Venise glass from early 17th century London in comparison with luxury glass of the same age

SEM-EDX and LA-ICP-MS analyses were performed on a set of early 17th century London glass fragments. The samples originate from two archaeological sites (Aldgate and Old Broad Street) where glass workshops were active in this period. The great majority of the samples are made of soda glass. Two distinct compositional groups are observed, each typical of one site of provenance. The samples originating from the Old Broad Street excavation feature a silica-soda-lime composition, with a moderate amount of potash. The samples from Aldgate are richer in potassium and feature higher amounts of trace elements such as Rb, Zr and Cu. The distinction between the two groups stems from different flux and silica sources used for glassmaking. A comparison with different European glass compositions of that time reveals no resemblance with genuine Venetian production, yet the composition of the Old Broad Street glass shows a close similarity to that of fragments produced ‘à la façon de Venise’ in Antwerp at the end of the 16th century. This coincides with historical sources attesting the arrival of glassworkers from the Low Countries in England and suggests that a transfer of technology took place near the turn of the century.

Zinc and Iron Concentration as Affected by Nitrogen Fertilization and Their Localization in Wheat Grain

Nearly half of the world cereal production comes from soils low or marginal in plant available zinc, leading to unsustainable and poor quality grain production. Therefore, the effects of nitrogen (N) rate and application time on zinc (Zn) and iron (Fe) concentration in wheat grain were investigated. Wheat (Triticum aestivum var. Krabat) was grown in a growth chamber with 8 and 16 h of day and night periods, respectively. The N rates were 29, 43, and 57 mg N kg-1 soil, equivalent to 80, 120, and 160 kg N ha-1. Zinc and Fe were applied at 10 mg kg-1 growth media. In one of the N treatments, additional Zn and Fe through foliar spray (6 mg of Zn or Fe in 10 ml water/pot) was applied. Micro-analytical localization of Zn and Fe within grain was performed using scanning macro-X-ray fluorescence (MA-XRF) and laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS). The following data were obtained: grain and straw yield pot-1, 1000 grains weight, number of grains pot-1, whole grain protein content, concentration of Zn and Fe in the grains. Grain yield increased from 80 to 120 kg N ha-1 rates only and decreased at 160 kg N ha-1 g. Relatively higher protein content and Zn and Fe concentration in the grain were recorded with the split N application of 160 kg N ha-1. Soil and foliar supply of Zn and Fe (Zn + Fes+f), with a single application of 120 kg N ha-1N at sowing, increased the concentration of Zn by 46% and of Fe by 35%, as compared to their growth media application only. Line scans of freshly cut areas of sliced grains showed co-localization of Zn and Fe within germ, crease and aleurone. We thus conclude that split application of N at 160 kg ha-1 at sowing and stem elongation, in combination with soil and foliar application of Zn and Fe, can be a good agricultural practice to enhance protein content and the Zn and Fe concentration in grain.