As The Triumph of David conservation project gains momentum, we, the chemistry team, have been studying the materials used in the painting. Conservation science is considered science in the service of cultural heritage and is a field of study that most science majors don’t even know exists. Conservation scientists are interested in the chemistry of artistic materials and techniques and how they age in an artwork’s environment. The proliferation of high-powered analytical instrumentation has made it possible to analyze historical objects with in-depth, non-destructive techniques. Our role in the conservation team is to provide data to conservators that can identify the materials used by the original artists as well as the materials used during subsequent restoration campaigns (a campaign refers to restoration efforts separated by time). We are also hoping to provide the art historians with technical information that may help with dating and understanding how Pietro da Cortona and his workshop painted this rare, large format, oil-on-canvas painting. Ultimately, we would like to provide a scientific answer as to why the varnish and the areas of obvious overpaint have darkened and deteriorated so quickly since the most recent restoration campaign in 1956. At this moment, we are busy trying to identify the pigments present in the painting as well as which paint layers are original and which are restoration.
Inorganic pigments, like verdigris or vermillion, can often be identified utilizing x-ray fluorescence (XRF) and scanning electron microscopy paired with energy dispersive x-ray analysis (SEM-EDX). During the past few weeks, we have been focusing our work on XRF as well as optical microscopy of cross-sections that have been taken, but today we are going to focus on some of our XRF results.
XRF is incredibly useful in conservation science not only because it’s nondestructive but also because portable, hand-held instruments are available, meaning that the instrument can be brought to the painting instead of bringing the painting to the lab where the instrument resides. XRF utilizes high energy X-rays to eject inner shell electrons from an element. The outer shell electrons will then fill the electron hole, releasing an x-ray photon in the process. This photon release is known as fluorescence and the particular energy of the released photon is specific to each element. Because of this specificity, we can use XRF to help us identify inorganic pigments in areas of interest. (For a more in depth look at the XRF mechanism of action, see here)
In the case of the soldier’s red robe (as seen in the photo at the left) we found a signal for mercury (Hg), indicating that the red color is probably due to a vermillion (mercury sulfide) pigment. There was also a large signal for lead (Pb) but this signal is found throughout the painting as it is caused by a lead white containing ground layer. His yellow shirt, on the other hand, showed signs of chromium(Cr) and barium (Ba) which would indicate a more modern chrome yellow (post 1800) or lemon yellow (post 1830) paint, thereby indicating that one of the paint layers originates from a more modern restoration campaign.
As mentioned previously in the Metigo map blog post, there are a lot of areas with heavy overpaint present as well as several restoration campaigns. By using XRF to monitor the presence of zinc (Zn), we can assist the conservators in tracking the removal of the restoration materials. Currently, zinc is suspected to be a component of the varnish or in a slightly pigmented glaze from a previous restoration. We suspect this because we see zinc disappear significantly with varnish removal, as seen in the image below, comparing the uncleaned overpaint (red), cleaned but still overpaint (green), and cleaned without overpaint (pink) all on a soldier’s leg in the painting. The spectrum corresponding to the uncleaned overpaint shows a moderate zinc signal, whereas zinc is not detected in the spectrum corresponding to an area of overpaint with just the varnish removed and the spectrum corresponding to a completely cleaned area.
Because XRF simultaneously examines all the layers of the painting from the ground, paint layers, over-painted layers, and varnish, we can’t tell exactly in which layer the pigment is present by XRF alone. However, this XRF data is useful when determining where to micro-sample the painting to get layer-specific pigment information through techniques such as SEM-EDX and fluorescence microscopy, which you can read about in later posts. We are just beginning to unravel some of the mysteries surrounding this painting and our team is learning more every day.
Until next time, Ciao!
Anthony, Amanda, and Kristen.