The National Gallery in London houses some 2300 paintings dating from the 1250s to the early 20th century. Within its halls, art devotees can drink their fill of a late medieval gilded altarpiece belonging to Richard II and masterpieces of Impressionist and Renaissance art by the likes of Michelangelo, Cezanne, Leonardo da Vinci, and van Gogh. What they won't see is the laboratory on the top floor of the Gallery's Northern Extension, where Director of Science Ashok Roy and his six-member team subject these objects to a battery of 21st-century analytical methods.
Their charge is to characterize the history, provenance, and makeup of the artworks in the museum's care; to establish the material characteristics of paintings for their technical history; and to understand how they degrade in order to aid preservation and restoration efforts.
“The museum scientist is not directly involved with conservation treatment of actual artifacts,” explains Richard Newman, Head of Scientific Research at the Museum of Fine Arts (MFA) in Boston, which has two scientists on staff. “However, the museum scientist does provide scientific information on works of art in a museum setting, and this information may help diagnose problems for conservators.”
Employing techniques such as scanning electron microscopy (SEM), gas chromatography coupled to mass spectrometry (GC/MS), Fourier transform infrared (FTIR) spectroscopy, and X-ray diffraction (XRD), the work is undeniably stimulating. But it certainly isn't easy. Budget constraints keep teams small, often forcing researchers to be scientific ‘jacks-of-all-trades’. Plus, the samples are complex (mixtures of inorganic pigments, organic binders, and varnishes, for instance), tiny, and hard to come by.
Such is life in the small but tightly knit world of museum science, where fine art meets high-tech and dedicated teams of scientists ply their trade on objects most of us will only ever glimpse from behind glass.
First Do No HarmBecause the subjects of their analyses are invariably precious, the guiding principle throughout the museum science world echoes Hippocrates: First, do no harm.
“We don't always have the luxury of taking samples from artwork,” says Ken Sutherland, one of two full-time scientists at the Philadelphia Museum of Art, who recently coauthored a paper discussing GC/MS-based analysis of picture varnishes in 19th- and 20th-century American paintings (1). “These days, the instruments are so sensitive, the amount of sample needed is microscopic…but there are cases where you cannot take anything.”
Karen Trentelman, who heads the five-member Collections Research Laboratory at the Getty Conservation Institute (GCI) in Los Angeles, California, adopts a typical strategy. “We always start with the most non-invasive techniques we can, and also the broadest techniques,” she says. Generally, that means imaging (under ultraviolet, infrared, and X-ray radiation) and other nondestructive approaches such as X-ray fluorescence and Raman spectroscopy (a non-destructive technique that infers molecular structure from a material's vibrational properties). If a physical sample must be taken for more detailed analysis, the team collects a tiny sliver of paint, encases it in resin, and then cuts cross sections to reveal the paint strata within. At that point, says Trentelman, “the number of analytic techniques is broader.”
One popular go-to choice: SEM. According to Vern Robertson, technical sales manager at JEOL USA in Peabody, Massachussets, SEM instruments are effectively “mini-labs” in and of themselves. “With a wide variety of detectors and spectrometers, both morphological and chemical information can be collected.”
Backscattered electrons—created by an SEM instrument's electron beam bouncing off atoms in the material—can reveal the multiple layers and thicknesses of paint, since different atomic nuclei have different densities. Such information can provide dating information, identify restorations, or even suggest a forgery, he says.
Each atom also has a unique X-ray signature, identified through the use of energy dispersive spectrometry. X-rays generated as the electron beam interacts with the material can reveal atomic composition, whereas diffraction patterns generated by backscattered electrons can reveal the orientation of microscopic crystals within a material.
