Cold War Nuclear Bomb Tests Are Helping Researchers Identify Art Forgeries

Traces of carbon-14 isotopes released by nuclear testing enable scientists to date paintings created post-World War II

detecting art forgeries
Researchers extracted paint and canvas fiber samples from a known forgery supposedly dating to 1886 but actually created during the 1980s. James Hamm, Garman Art Conservation Department, Buffalo State College, The State University of New York

A new method of detecting forgeries uses minuscule canvas fibers and paint samples to expose purportedly historic works of art as modern creations.

The technique, catalogued in the Proceedings of the National Academy of Sciences, dates would-be masterpieces by measuring traces of carbon-14 isotopes released into the atmosphere by mid-20th century nuclear bomb testing. Objects made after 1963 hold significantly higher levels of the unstable isotope, allowing researchers to differentiate between pre- and post-World War II paintings.

This isn’t the first time scientists have turned to radiocarbon dating in an attempt to thwart forgers. As Niraj Chokshi explains for The New York Times, the idea of dating art by assessing the organic matter used to bind paint pigments was first floated as early as 1972; previous case studies include a 2015 investigation that debunked the provenance of a supposed Fernand Léger canvas owned by American art collector Peggy Guggenheim.

Still, the approach has its drawbacks. According to artnet News’ Taylor Dafoe, savvy forgers recycle antique canvases and even paint, making it difficult to determine if a painting is original or simply artfully doctored. At the same time, the Economist notes, radiocarbon testing is so destructive that a sample can rarely be analyzed twice. Typically, the process also requires a “sufficient[ly]” sized sample: Given the possibility that a suspected forgery may actually be a bonafide masterpiece, investigators are often reluctant to remove significant amounts of paint. Fragments of a wooden frame or pieces trimmed from the edge of a canvas, on the other hand, “might be [considered] an acceptable loss.”

The new research, led by Laura Hendriks of Switzerland’s ETH Zurich, draws on the latest technological advances to reduce the size of samples needed for testing. Working with a known forgery dating to the 1980s, the team extracted hairlike strands of canvas fiber measuring just a few millimeters long and a paint particle weighing less than 200 micrograms.

Although the painting—titled Village Scene with Horse and Honn & Company Factory—mimics the American primitive folk art style and is signed “Sarah Honn May 5, 1866 A.D,” it’s actually the work of convicted forger Robert Trotter. As Treasures on Trial, an online portal run by Delaware’s Winterthur Museum, Garden & Library, notes, Trotter stripped, repainted and artificially aged worthless old paintings, as well as created lengthy provenance reports that seemingly testified to the works’ authenticity. Ultimately, he admitted to selling 52 falsified paintings and served 10 months in prison. Following Trotter’s conviction, Buffalo State College’s Art Conservation Department acquired the “Sarah Honn” canvas, which has been studied to better understand forgery methods.

According to Chemistry World’s Emma Stoye, Hendriks and her colleagues relied on “standard non-destructive techniques” to identify a suitable paint particle within an existing crack on the painting. Once the researchers had extracted the paint and canvas fiber samples, they used an elemental analyzer to burn the materials into carbon dioxide. This gas was then fed into an accelerated mass spectrometer capable of measuring the ratio of carbon-14 to carbon-12 isotopes present.

Based on carbon isotope ratios, the team found that the canvas could have been crafted at any point between the late 1600s and mid-1900s, suggesting it was likely a recycled, age-appropriate specimen. The binder found in the paint, however, had enough carbon-14 to definitively date it to the post-war period. Additionally, Chokshi writes for The New York Times, the oil used to bind the paint was shown to be derived from seeds harvested between 1958 and 1961 or 1983 and 1989.

Speaking with Stoye, Jilleen Nadolny, a principal investigator at Art Analysis & Research who was not involved in the study, says there are still limitations to the revamped technique. “You have to be very aware when sampling to avoid contamination,” she explains, “and there are huge chunks of time where you don’t get anything specific.”

Greg Hodgins, a physicist who leads a radiocarbon dating lab at the University of Arizona and was also not involved in the new research, echoes this sentiment, telling Chokshi that while the method is “an important advance, … it’s not a silver bullet.”

Crucially, Chokshi notes, carbon-14 isotopes, spurred by ocean absorption and dilution by fossil fuel emissions, are on track to return to pre-war levels. This could lead to inconclusive results further down the road, making it essential to use radiocarbon dating in conjunction with other techniques.

“It can still be useful but it’s going to be more and more difficult,” Hendriks concludes to Chokshi. “It’s kind of like a puzzle coming together.”

Get the latest stories in your inbox every weekday.