Back to

CSI: Conservation Science Indianapolis – Part Three

Our guest blogger today is Mirza S. Baig, a Senior Research & Development Chemist at Vertellus Specialties, Inc. where he does “top secret” catalyst research. His favorite art moment was sitting for two hours in the Sistine Chapel at the Vatican.

Part Three in a series of blog posts on the ongoing technical examination of a purported 1874 Alfred Sisley Impressionist painting (Part One is here, Part Two is here).

In previous blog posts we have laid out the historical background and curatorial concerns regarding the IMA’s purported Alfred Sisley painting House in a Village.  Numerous inconsistencies in the painting’s provenance and several curiosities discovered through imaging studies have cast doubt on the artwork’s authenticity – but at this point, all of these are only circumstantial evidence.  Can a scientific analysis of the painting reveal any concrete evidence of illegitimacy: the presence of anachronistic materials or the use of techniques not normally practiced by Sisley?  One of the easiest tests to conduct is an identification of the pigments used in the work, a task that furthermore can be accomplished without harming the painting.

Paints have been used since the dawn of humankind. Over the centuries they have changed from plant-based dyes and mineral pigments to include complex colorants made in laboratories. One would expect cave paintings from Lascaux to be drawn with red and yellow ochres, brown umbers, and sooty charcoal, while a modern canvas might contain synthetic phthalocyanine blues and bright quinacridone reds.  The advent of new colorants provides the art analyst with date markers – temporal wayposts by which to measure the age of an artwork.  Industrial patents and historical notebooks from chemists provide us with the discovery dates of new colorants, and so their presence on an artwork sets a terminus post quem, or “date after which,” the painting could theoretically have been painted.  The anachronistic use of colorants by forgers can be the telltale sign that an artwork is not what it purports to be.

X-ray fluorescence (XRF) spectroscopy is an instrumental technique that uses a beam of excitation energy to induce emission of characteristic x-rays from elements contained in a sample.  The sample, a painting in this case, can be analyzed directly without having to remove any material from it, even as it hangs on a gallery wall.  A microfocus XRF was used to characterize the elemental signatures of each of the major color passages in House in a Village, and the elements determined are shown in Figure 1.  Click here for a short video that explains the analysis.

Figure 1. XRF analysis locations with major and minor elements identified. A likely pigment is also suggested.

All pigments and dyes are made of molecules comprising certain elements that contribute to the color we see. We can infer from the detected elements which pigment might be present. For example, an XRF analysis of the red field in the central area of the Sisley work (Figure 2, marked position 1 in Figure 1) indicated that the major elements lead (Pb), mercury (Hg), and iron (Fe) were present.  Mercuric sulfide (HgS), called vermilion, is a common historical artists’ red pigment, and so it is likely that this passage of paint contains that colorant, perhaps mixed with basic lead white (2PbCO3.Pb(OH)2) to give a pinkish hue.  But where in the spectrum are the S, C, and O peaks that should also be present due to those pigments?   The weak C and O x-rays are unfortunately absorbed by the air and cannot be detected, while the x-rays from sulfur have the same energy as the Pb x-rays at about 2.34 keV.  A further complication comes from the highly penetrating nature of the x-ray excitation beam; signals are generated from every layer of the painting.  This can explain the strong Fe signal as perhaps coming from a buried layer of the painting – remembering that our work appears to have been painted over an earlier composition – or there is also the chance that an iron-containing red ochre is present mixed with vermilion.  To answer these questions, a more invasive analysis is required.

Figure 2. XRF spectrum of the central red field (marked position 1) with element peaks identified.

Based solely on the results of our non-destructive XRF analysis, we can take our list of suspected pigments and compare them to the time period that the painting was supposedly created. Vermilion red, organic red lake, lead white, Prussian blue, cobalt blue, chrome green, burnt umber, and bone black were all available to artists in the 1870s, but they are also available to artists today!  How does the palette compare to what is known of Alfred Sisley’s preferred artists’ materials?  Anthea Callen summarizes Sisley’s palette as likely to contain most of these pigments.1  Based on this initial cursory survey of the work’s colorants, the painting could be by Sisley, or a contemporary of his, or by a 20th century forger familiar with the typical Impressionist palette.   A closer look at the use of these colors and the media in which they are applied is a good next step to understanding the history and construction of the painting.


Filed under: Art, Conservation, The Collection

Comments are closed.