Art Restoration: The Fine Line Between Art and Science

Matthew Chalkley
By Matthew Chalkley December 1, 2010 22:36

Art Restoration: The Fine Line Between Art and Science

You walk into your organic chemistry class of two hundred students. As you glance around, you sort people into two groups: pre-meds and pre-med wannabes. If you are really inventive, maybe you can imagine a few of your classmates as pre-veterinarian or pre-dental. You have probably forgotten that there are actually some chemistry majors in there. What you likely did not even know is that some of your classmates are prospective art conservators. In fact, art conservators have the same undergraduate chemistry requirements as pre-meds.

This is the first hint that the art conservation field is far more scientific than one might first imagine. According to Ian McClure, chief conservator for the Yale University Art Gallery, “any graduate [art] school [with] chemistry graduate or physics graduate applicants [regards them] with a great deal of interest because they have the capacity to go into the practical side and the research side.” A firm science background allows a conservator to have a more fundamental grasp of the increasingly large number of scientific techniques necessary for conservation work.

Even pure research science is being incorporated into the discipline of art conservation. Already the Metropolitan Museum of Art, the Museum of Modern Art, and the Getty Gallery each have a full time scientist devoted to developing specific techniques for conservation. Yale has already committed money towards the building of a centralized conservation lab on West Campus and has been awarded an Andrew Mellon Foundation challenge grant towards an endowment for a research scientist devoted to researching topics relevant to art conservation. Even salary costs will be funded for three years.

History of Conservation

In the 1800s and 1900s, the practice of conservation was quite commonplace. Even paintings by the Old Masters were redone frequently to match the taste of the time. As described by Mr. McClure, a study of treatments of “The View of Delft” by Johannes Vermeer “looked back to something like 1800 and found that it had been modified about 50 times. [The] more famous the painting, the more often its been redone because people simply cannot keep their hands off it.”

However, art conservation from the olden days was far from scientifically precise and was found to be quite harsh and severe. Old conservation manuals suggested covering the entire painting in wood-ash and then wiping it off with water, a process that caused an extremely alkaline substance harmful to the painting to form. Even as recently as twenty years ago conservators were still using solvents that are now considered toxic by hand .

The Modern Conservator

The goal of the modern conservator is to determine in a non-invasive way the remaining original portions of the painting and to gain an understanding of how the painting had been treated over the years. Initial analysis will likely consist of x-rays in order to gain information about how the painting was composed. Such x-rays allow the conservator to form an outline of the painting based on differing absorptions. More specifically, Mr. McClure described: “Lead white paint, often used for figures and faces, is very absorptive so you can see them and also areas of missing paint because even translucent paints become more absorbent with added layers.”

Next, the conservator would use infrared imaging to see the original drawings and losses of paint. Different materials with different absorptive indices were often used in past restoration to allow for their identification. Recent technological developments in art restoration include cameras with fixed wavelengths that allow conservators to pinpoint, for example, carbon-based drawings with distinctive wavelengths at about 1700 nanometers.

Both of these techniques are part of a larger movement to eliminate the previously destructive techniques employed by art conservators of the past. Traditionally, millimeter square portions of paint would be removed from the painting to identify varnish layers and different pigments used. Not only does this remove portions of the painting, but it also illuminates the overall picture less than a modern scan of the entire painting could. These new and improved methods are revolutionizing how priceless works of art can be safely restored to near original conditions.

The Science of Varnish

Once the art conservator has developed a coherent picture of the original painting and all previous attempts to preserve and renew the art, the next step would then be to find the appropriate solvent mixture to remove discolored varnish layers and, if possible, later restoration paint often composed of natural resins or a synthetic resin in a newer painting.

Raman spectroscopy, a recent innovation for art conservators, has made it easier to determine the exact composition of this varnish. In this method, a monochromatic wavelength impinges on the material surface and is then transmitted or reflected. A small portion is absorbed by the molecules of the material and subsequently is reradiated at a slightly different frequency. The measured frequency change for any given wavelength of light is characteristic of the molecule, thus allowing for characterization of the varnish.

Upon determining the identity of the varnish and removing the outer layers, the varnish can be repaired such that any restoration work done is separate from the remaining original work. That way, as Mr. McClure emphasized, “in the future, anything that removes the varnish will remove all the restoration.” This makes the entire process eminently reversible, allowing for stylistic fluctuations common in art conservation.

Developing new varnishes is actually a key area of conservation research. One possibility is the addition of UV absorbers to natural resins to prevent the natural yellowing process. Another possibility is the creation of more stable and easily removed materials with the same refractive index as natural resins’. A conservator might even work with a curator of the museum to customize the glossiness or matte of the varnish to best suit the lighting of the exhibit.

Another key aspect of restoration is that the conservator “does not want to make the picture look perfect, rather just homogenous.” If one area is cracked and another is worn and abraded, then the restorer might choose to touch up the abraded area. However, essential details inevitably were lost in the areas of paint loss, thus causing a more seemingly perfect but less unified painting after restoration.

Lead White Paint

Accurate original paint identification by the conservator can give valuable information for dating and origin. One essential component of most paintings starting in the Medieval Ages and spanning both Western and Eastern painting traditions is lead white paint, or lead hydroxycarbonate. Its traditional manufacture involved incorporation of carbon dioxide and thus traceable radioactive carbon-14 produced from burning of recent plant sources. Modern lead white is usually derived from fossil fuels that lack carbon-14. To date the lead white, the relative rate of incorporation of carbon-14 is determined through high-resolution mass spectrometry due to the difference in mass of the isotopes.

Traditional lead white can be further differentiated from its modern equivalent by its specific carbon-13:carbon-12 ratio characteristic of particular common lead sources in England. The carbon-13:carbon-12 ratio can be determined either through mass spectrometry or NMR spectrometry because only carbon-13 has a spin that exhibits NMR activity. Lead ores also contain radioactive radium-226 that has a long half-life and decays into lead-210 with a half-life of twenty-two years. While most of these isotopes are lost as slag when the ore is purified, the remains can be compared to the relative concentration of radium-226 to estimate the age. The actual measurement done is of lead-210, the next step in the radioactive cycle, because lead-210 is a beta-emitter and thus more difficult to measure against background cosmic radiation than the other two alpha-emitter isotopes.

Unfortunately, these measures cannot effectively date and identify the lead white paint by themselves. Different methods, such as the use of old lead and vinegar, can lead to an isotopically “old” sample of lead but an incorrect carbon-14 content. In fact, some of Han van Meegeren’s paintings were discovered post-mortem through radioactive isotope analysis, which uncovered yet more of the world’s most famous forger’s work.


The opportunity for science to integrate itself further into modern art conservation is ever increasing. In fact, just this October Mr. McClure attended conference on nanotechnology and how it can apply to the field. Novel connections with other disciplines likely lie in the future of art conservation with ever increasing contributions from scientists.

About the Author: Matthew J. Chalkley is a sophomore in Davenport College. He is a chemistry major and works in Professor Hazari’s organometallics lab.

Acknowledgments: The author would like to thank Mr. Ian McClure, chief conservator for the Yale University Art Gallery, for his willingness to share his expertise in the field.

Further Reading:

Craddock, Paul. Scientific Investigation of Copies, Fakes and Forgeries. St. Louis: Butterworth-Heinemann, 2009.

Found in Haas Family Art Library.

Townsend, Joyce. Conservation Science 2007. New York: Archetype Books, 2009.

Found in Yale Center for British Art.

Kirsch, Andrea and Rustin S. Levenson. Seeing Through Paintings: Physical Examination in Art Historical Studies. New Ed. New Haven: Yale University Press, 2002.

Found in Haas Family Art Library and Yale Center for British Art.

Kollandsrud, K. 2006. New Light on the Virgin from Veldre, the Virgin from Østsinni and the Crucifix from Thirteen. I: Medieval Painting in Northern Europe: Techniques, Analysis, Art History. Studies in commemoration of the 70th Birthday of Unn Plahter. London: Archetype Publications Ltd. ISBN 1904982212. p. 76-90.

Matthew Chalkley
By Matthew Chalkley December 1, 2010 22:36