Paintings for Posterity

Northwestern faculty team up with the Art Institute of Chicago to investigate classic works

Amanda Gordon

Image by Leta Dickinson / North by Northwestern

Georgia O’Keeffe once described Cerro Pedernal, a volcanic mesa nestled in the Jemez Mountains of northern New Mexico, as her private possession. Looking at the cool turquoise and warm pastel hues in her 1941 painting Pedernal, you sense her intimacy with the desert landscape. If you look a little closer, the colorful oil painting reveals its age. Bits of discoloration and small protrusions puncture the canvas's surface like acne on an otherwise smooth visage.

These deformations are not unique to O’Keeffe. If you visit Northwestern’s Block Museum of Art, you can see them on the surface of Roman-Egyptian funerary portraits. They form when heavy metal ions in pigments react with fatty acids in the oil binding of the paint, producing pus-like bumps on the painting’s surface.

Marc Walton, an associate professor of materials science and engineering, saw these deformations when he visited the Georgia O’Keeffe Museum in Santa Fe back in 2015. The museum’s head of conservation, Dale Kronkright, contacted Walton and his team at NU-ACCESS, the Northwestern University and Art Institute of Chicago Center for Scientific Studies in the Arts, to evaluate the O’Keeffe collection as part of an ongoing research project.

“We’re trying to understand the basic chemistry of how soaps form, why they form and whether we can model them,” Walton says.

In December, Walton received a grant from the National Endowment for the Humanities to conduct research within the emerging fields of technical art history and conservation science. The grant allows Walton and other scientists at NU-ACCESS to investigate soap bubble deformations and develop digital tools to evaluate paintings and diagnose their condition.

Oliver Cossairt, who teaches electrical engineering and computer science at Northwestern, is working with Walton to develop a technique called photometric stereo. This technology will allow museums to monitor individual works of art by capturing a series of photographs with a controlled light source.

In his office overlooking the Shakespeare Garden, Cossairt scoots away from his desk to explain the technique.

“Imagine a basketball in front of you and your eye is the camera,” he says. “Fix the point on the basketball that you’re looking at.” He moves his hand in front of his face. “When I move the light over here, that gets dimmer.” If conservators know where a light source is when taking a photograph, they can examine brightness values for individual pixels in the image. This can be used to determine the underlying surface of the painting at the pixel location.

The technique allows conservators and scientists to decouple the two-fold problem of soap formations: surface protrusions and discoloration, which were previously impossible to distinguish.

Cossairt and the cohort at NU-ACCESS plan to run a series of controlled experiments that accelerate the deterioration process of a sample in different conditions. This will allow computer scientists to create a digital reference and set of statistics modeling the decay. Conservators can then track the pace of soap bubble development for individual pieces of art and determine whether a painting’s surface is stable enough to transport.

The team hopes to develop a software that enables conservators to upload photos they have taken using the photometric stereo technique. The software would then generate statistics and predictions based on both the photos and the controlled experiments conducted at NU-ACCESS.

With access to this data, Cossairt says, conservators can be more confident in their ability to share art with cultural heritage institutions around the world.

Fifteen years ago, when Francesca Casadio came to the U.S. from Italy to work as a conservation scientist at the Art Institute, it became apparent that Chicago's largest museum needed more scientific resources to monitor its sprawling collection. Northwestern emerged as the solution, recruiting students and faculty from its material science, engineering, computer science and art history departments to collaborate with conservators at the Art Institute.

In 2012, the collaboration grew into a formalized research center through the support of the Andrew W. Mellon Foundation. The following year, Casadio hired Marc Walton as the lead scientist for NU-ACCESS.

“The beauty of doing this at NU is that you touch a synapse and then the surge of the signal goes in directions that you’re not able to anticipate,” she says. This academic ethos and new technology enables NU-ACCESS to be an invaluable resource for museums to monitor and assess the state of their collection.

“We used to need this much material to get answers,” Casadio says, picking up my coffee cup. “Now it’s this much material,” she says, pointing to a small crumb on the table. With the expertise and resources of NU-ACCESS, paintings like O’Keeffe’s Pedernal can retain their color, and poignancy, for generations to come.

“Paintings, like us, are going to age,” Casadio says. “But you can do something to slow down the time.”