Flattery for faience: imitating ancient materials reveals lost manufacturing secrets. (Cover Story).

While students swarmed the halls at the Massachusetts Institute of Technology one Friday late last November, a dozen or so outsiders crowded into a small glass shop in the basement of Building 4. They were recapitulating the first steps of an ancient art. Standing around a workbench, their hands dusty and gray, the group listened to Carolyn Riccardelli, who gingerly patted at a pasty mixture that had the consistency of wet toothpaste.

Riccardelli, a conservator at the J. Paul Getty Museum in Los Angeles, was showing the group of materials scientists, archaeologists, and conservators how to formulate and mold an ancient Egyptian material known as faience. It's a type of ceramic with a quartz core and glazed surface. A well-known example is an 8-inch-long, bright-blue hippopotamus at the Metropolitan Museum of Art in New York.

Leaders of a symposium on the materials-related aspects of art and archaeology at a national meeting of the Materials Research Society (MRS) in Boston had brought attendees to the MIT shop to let them literally get a feel for how the ancient material was made.

Using a spoon, Riccardelli patted the delicate paste into a clay mold. When the amateurs tried their own hands at this technique, they quickly realized that just a little too much or too little water made the material fall apart. It sometimes took several tries to mold a simple shape, such as grapes. "This is why we look at ancient faience and we go `How did they do that?'" Riccardelli says. "It's difficult to work with."

It's also worth the trouble, she says. Scientists and curators can achieve an understanding of ancient materials and cultures that would be impossible without getting their hands dirty, Riccardelli argues. For this reason, many researchers--such as those at the MIT workshop--have become increasingly interested in gaining a craftsman's knowledge of materials processing. With this approach, Riccardelli and other researchers have revealed fine details of faience manufacture and composition that were lost for thousands of years.

Ancient Egyptians began making faience more than 6,000 years ago, and archaeologists have studied it intensely in the past century. Even so, new work using powerful microscopy and spectroscopy techniques continues to uncover long-lost secrets about the faience industry. Researchers including Riccardelli, however, have been showing that another fruitful way of demystifying ancient arts is to replicate them. Her motto could be, To make faience is to understand it.

"The Egyptians didn't leave recipes, so we really don't know how they made things," says curator Rita Freed of the Museum of Fine Arts, Boston. "The only way we can find out is when people try to replicate things.... That way we can understand just how much skill is involved, how much science is involved, and how sophisticated the whole process was."

Ancient Egyptians prized objects such as beads and vessels constructed from faience, probably because the surfaces look like gemstones. Made mostly of silica, from such sources as quartz and sand, faience usually contains several other components: calcium carbonate, a water-soluble alkaline substance such as sodium bicarbonate, and a chemical colorant. Copper oxide, for example, gives faience a distinct blue color reminiscent of lapis lazuli.

One common method of faience preparation--which Riccardelli uses in her research and demonstrated at the MIT workshop--includes a technique called efflorescence glazing. In this process, the faience maker mixes the ingredients together with a little water and then pats, taps, and molds the paste into shape. As the material dries, the colorant and water-soluble salts move to the surface and form a crust. After the material is completely dry, it's heated in a kiln.

This creates a glossy, colorful coating.

A day before the workshop, Riccardelli spoke at the MRS meeting about how ancient Egyptians often inlaid pieces of different colors into one other. "It's just some of the most gorgeous colored stuff in faience," she says.

While studying faience inlay, Riccardelli had wondered how ancient people managed to create certain fine details. She says that the artisans' creation of such exquisite inlaid objects shows that they had clear knowledge of how the materials behave in different situations.

She became particularly curious about two aspects of faience inlay: First, certain objects with inlaid designs--like the petals in a flat rosette--contain a distinct glazed layer under the inlay. Second, many objects have a separation between the inlay and the background piece. She wondered what skills and procedures are necessary to make the glazed layer or separation lines.

Several years ago, she began a systematic study to determine what procedures the ancient Egyptians could have used to produce the distinctive details. Making her own pastes, Riccardelli tested a multitude of techniques in an attempt to replicate ancient rosette designs. For example, she inlaid wet paste petals on a wet rosette background, dry petals on a wet background, and wet petals on a dry background. She also fired dry petals and backgrounds in the kiln and then inlaid wet paste petals on the fired backgrounds. In another approach, she put fired petals into wet backgrounds before the entire rosette was fired.

After each attempt to replicate the ancient inlaid faience, Riccardelli used microscopes to examine her wares' surfaces and cross sections of broken pieces. She then compared the results with observations she made on 60 fragments of ancient faience objects in museums in London, New York, and Boston.

Riccardelli found that two of the inlaying procedures she tried replicated the glazed layers of the originals particularly well. In one case, she allowed the background to dry completely and then added wet paste for the petals. In the other case, she fired the dried background before adding wet paste.

Although confident that she's replicated the glazed layer she's seen in many ancient objects, Riccardelli is continuing to look for a procedure that more exactly produces the separation lines between inlaid and background pieces.

Some researchers had previously suggested that the lines formed spontaneously as two differently colored faience pastes dried, shrank, and pulled away from each other. But faience pastes don't shrink enough to produce the observed pattern, Riccardelli contends. In practice, Riccardelli has found that the lines aren't easy to make, and she surmises that they didn't form accidentally.

Riccardelli is the first person to show how to inlay and glaze objects in a way that replicates certain details seen in many artifacts, says Jennifer Mass of the Winterthur (Del.) Museum, who was an advisor for the project while Riccardelli worked on it as a student at State University College at Buffalo in New York. Riccardelli later took her project to the Getty Museum and has since moved on to the Metropolitan Museum of Art.

"I thought everything about faience was known and done," symposium organizer Pamela Vandiver, a materials scientist at the Smithsonian Center for Materials Research and Education in Suitland, Md., told the audience before Riccardelli presented her results at the MRS meeting. "She's blown a lot of us away."

Patricia Griffin of the Cleveland Museum of Art is another investigator probing still hidden techniques behind ancient Egyptian faience--and using replication experiments to do it.

Griffin wanted to identify the composition of individual faience objects in her museum's Egyptian collection and the processes that were used to make them. Although she sought this information to better understand and catalog the museum's objects--many details of which were published in a 1999 catalog--she was also personally interested in a bigger picture.

Griffin wanted to know how and why similar items from different periods vary in their details. For example, did faience paste composition change ever so slightly, permitting better handling and more precise detailing in the later of two sets of similar objects made 700 years apart?

Signing up for space in night ceramics classes at the Cleveland Institute of Art across the street from her museum, Griffin began to experiment with different compositions for faience paste. She tried varying the chemistry and the coarseness of the paste. She also compared how easily each mixture could be handled, molded, and cut with instruments.

Griffin then analyzed the structure composition of her modern faience objects using advanced instrumental techniques, such as scanning electron microscopy and X-ray fluorescence spectroscopy. With the results of these analyses--performed at the museum and at nearby Case Western Reserve University and the NASA Glenn Research Center, both in Cleveland--Griffin used her handmade objects as standards against which to compare similarly analyzed samples from the Cleveland Museum of Art's collection.

At the MRS meeting, she reported that the composition and manufacturing of the objects in the collection did in fact vary over the centuries. For example, later artifacts, such as a funerary figurine from about 350 B.C., are remarkable both in their detail and craftsmanship. They contain more calcium and their pastes consist of smaller particles than objects from 600 to 700 years earlier, which show less detail.

The ancient artisans probably added the calcium in the form of calcium oxide--burnt lime--rather than calcium carbonate, or limestone, Griffin suspected. This would give the paste the consistency of plaster, she says. And by trying out new recipes, Griffin found that it was easiest to mold and carve objects into their final shapes when the paste is fine and contains higher concentrations of calcium added as lime.

Materials science has long been part and parcel of the study of ancient objects. Indeed, art and archaeology has been part of the agenda of Materials Research Society meetings for many years. Last November's symposium on art and archeology, however, was the first occasion when organizers invited craftspeople to participate.

"It's the first time we've put together craftsmen with scientists with archaeologists and art historians," says Vandiver. Craftspeople, she says, know how to process materials, and they're particularly insightful about how those materials behave and respond under many different conditions.

Consider glassmaker Dudley F. Giberson of Joppa Glassworks in Warner, N.H. At the symposium, he explained how he makes replicas of a small jar called an Egyptian-style core vessel. He carefully coats a vessel-shaped object made of clay, sand, sawdust, and dung with powdered glass, called frit. The frit-coated object is heated in a furnace for a few hours until the frit partially fuses. Then, Giberson holds the object directly over a heat source to add glass handles, rims, and other decorations. After the vessel cools, he digs out the core.

Giberson's glass vessels closely resemble some ancient artifacts, down to tiny bubbles visible only with a microscope, notes Vandiver.

While Riccardelli taught faience inlay techniques in the MIT basement, Giberson simultaneously demonstrated his painstaking craft in the same shop. "One thing I've realized is that the Egyptians had plenty of time," he says.

The MIT workshop was more than just a diversion for conference-weary scientists. Replication is increasingly a part of how archaeologists go about their research, reiterates Freed. "The idea of archaeology is to reconstruct the life of ancient man or really to better understand ourselves in a way," she says. Undertaking the actual physical actions and procedures ancient people used for making real things is one way to do this.

Vandiver adds that many archaeologists abide by the maxim that they need to understand the entire lifetime of an artifact, including its transformation from raw materials.

"When you find an artifact ... you want to get back to when it was not just an artifact dug up, but when it was actually an object working in a culture," she says. "Then, you want to work your way back to the raw materials that actually formed that object."

Sometimes making an object can quickly reveal a long-hidden technique. "You don't really get it until you get that material in your hands," says Riccardelli. "You can come up with all kinds of ideas, but then you should try it. A lot of questions get answered really fast."