Are oil paintings slowly eating themselves alive?

Original Article: Revealing the Nature and Distribution of Metal Carboxylates in Jackson Pollock’s Alchemy (1947) by Micro-Attenuated Total Reflection FT-IR Spectroscopic Imaging

It is well-known to conservators and the public alike that external forces—air pollution and acid rain, physical damage, bacterial or fungal growth, exposure to light or moisture—can damage these masterpieces, but what about internal issues? Most of us have never stopped to think that a painting could be eating itself alive!

It’s a big jump from true science to science fiction, and paintings aren’t really alive; ergo, they cannot eat anything, let alone themselves. However, the metaphor is accurate. Parts of the oil binder and the metal pigment colorants in oil paints, such as zinc or lead white, can react to form a serious problem—metal carboxylates, or colloquially, soaps. These soaps transform the previously colored paint into something translucent. They “eat away” the original paint and wreak havoc by popping up either to the surface as protrusions, cracks, or hazy films, or between layers of paint causing separation or delamination. Ironically, metal soaps—especially aluminum stearate—are commonly added during the paint manufacturing process to produce specific properties (e.g., wettability, texture, or consistency) in the final product. Aluminum stearate, in particular, increases the paint’s wettability and produces a final consistency akin to soft butter. Once the presence of these soaps has been noticed in a painting, we are left with a few questions: 1) Are these soaps just a part of the original paint or are they the result of an internal chemical reaction? 2) How are these soaps distributed? 3) Are they dangerous to the paintings, and if so, how dangerous?

Gabrieli et al. have tried to answer these questions with micro-attenuated total reflection Fourier transform infrared spectroscopy (micro-ATR-FTIR). FTIR works on the principles of induced molecular vibrations. Since molecules are very small, it doesn’t take a whole lot to get them to move—a beam of infrared light will do the trick! Certain groups of molecules, called functional groups, move in different ways. These movements produce peaks with specific shapes and locations in a spectrum that can be used like molecular fingerprints to identify compounds.

Figure 1. (a) Alchemy (1947) by Jackson Pollock. (b) Magnified views of ZnO/TiO2 white (I), phthalocyanine blue (II), and cobalt phosphate violet (III). (c) FT-IR spectra of the aforementioned paints which show sharp signals (marked with dotted lines) that match those of the zinc stearate spectrum shown in grey.

Figure 1 shows the FT-IR spectra for three paint samples from Jackson Pollock’s Alchemy. Using micro-ATR–FT-IR, the authors were able to create 2-D “maps” showing the distribution of aluminum stearate (from the original paint) and zinc stearate (produced within the paint by chemical reaction) by integrating the characteristic peaks corresponding to these two compounds for all of the spectra taken across an surface area of paint. A visual map is produced by assigning a color scale to the integral values, in which blue is low/zero and red is high. Figures 2b and 2c show the resulting heat maps for zinc soaps and aluminum soaps, respectively, over the same spatial area. These images help to answer questions 1 and 2. Both zinc and aluminum soaps are present in the paint, and the zinc soaps seem to form around the aggregates of aluminum soaps.

micro-ATR-FT-IR Images from Model Samples
Figure 2: (a) micro-ATR-FTIR spectra for 4 points across a model paint sample. The numerical labels on the spectra match the numbers on (b) and (c) indicating where on the sample they were taken. (b) and (c) are chemical maps of the same sample area produced from integration of the zinc stearate band at 1540 cm-1 (b) and the aluminum stearate band at 1588 cm-1 (c). A color spectrum from blue to red indicates concentration from a zero/low integration to a higher integration, respectively. Scale bar: 10μm.

Question 3 is harder to definitively answer. The zinc soap distributions for six different Alchemy samples are shown in Figure 3. There are two different sizes and arrangements of aggregations: 1) A1, A2, and B2 show smaller, more dispersed aggregates that are visually similar to the 30 day aged model samples made by the authors; 2) A3, B1, and B3 show fewer, but larger, soap aggregates that are attributed here to a more advanced stage of saponification. Logistically, the larger an aggregate grows, the more likely it is to cause damage to the surrounding area. Now that we know this process is happening and is at least partially fueled by the paints’ composition, we must find a way to prevent (or at least minimize) future damage.

5 heat maps for each fo 6 different samples from Alchemy
Figure 3: The first row of images are visible microscopic images of six different paint samples taken from Alchemy, either from the painting (A samples) or from the frame (B samples). Rows 2-5 show micro-ATR-FTIR images integrated for the zinc stearate peak (1540 cm-1). A1, A2, and B2 show similar distributions with many smaller aggregates. A3, B1, and B3 show similar distributions with fewer, but larger aggregates. Scale bar: 10μm

All figures reproduced/adapted with permission from Gabrieli, F.; Rosi, F.; Vichi, A.; Cartechini, L.; Buemi, L. P.; Kazarian, S. G.; Miliani, C. Analytical Chemistry 2016, 89 (2), 1283–1289. Copyright American Chemical Society. Alchemy is copyright 2009 Pollock-Krasner Foundation / Artists Rights Society (ARS), New York.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s