Conservation of a Louis Comfort Tiffany Stained Glass Window

Recently Diane Roberts Rousseau, a stained glass conservator from Western Massachusetts, came to treat our “Window with Hudson River Landscape” by Louis Comfort Tiffany (76.4.22). One of the panels had an old repair which had significantly yellowed over the years. I had the opportunity to ask her some questions about the treatment.

Astrid van Giffen: Why did the window need to be treated?

Diane Rousseau: The window depicts a summer landscape, and people were seeing a sunset–an oddly yellow, squared-off sunset–where none was meant to be. This was actually a heavy layer of adhesive. At some point during the preparation for the original Corning installation in 1977, a large blue piece of the sky area suffered an impact, which resulted in a large, multiple, V-shaped break. The Museum hired a restorer, Mr. Franz Meyer, to repair this glass and to strengthen the lead perimeters of the panels. He used the best materials and approach available to him at the time, but unfortunately the adhesive did not age well. Since he sealed the repair behind a new piece of clear glass, applied to the exterior of the window, there was no way to access and reverse it without taking the panel apart.

Van Giffen: Why could this treatment not done by the Museum’s conservators?

Rousseau: My treatment was actually a team effort with the chief conservator at CMOG, Steve Koob, who taught me many of the specialized techniques used for epoxy joining and casting at a seminar in 2001. My own specialty is conservation of stained glass, which involves quite a lot of joining and filling of broken glass, but also treatment of the matrix–the lead and copper channels that hold a window together. In this case, the matrix had to be partially disassembled to remove the damaged blue piece, so it made sense for Steve and I to work side by side. I opened the patient for surgery, and he took care of the cosmetic end of the treatment–the Hxtal repair.

Van Giffen: How was it repaired?

Rousseau: The window was removed from its display setting and brought over to the Conservation Lab area, where it was laid out on a specially padded bench. Tiffany windows are built in layers, so my first task was to support the stepped surfaces from below, so I could work on them without any danger of stress or deflection. I took a rubbing of the lead and copper-foiled lines, with black wax crayon on vellum – this is a centuries-old documentation technique which is still used, because it efficiently registers not only the placement of the lines but the texture of the glass surface. Digital photos were also taken of the damage before treatment, and as the work progressed.

I disassembled the window from the perimeter edges, in order to minimize intervention on Tiffany Studios’ original material. The perimeters had already been worked on by Mr. Meyer in 1976, and he had heavily reinforced them–by soldering steel bars directly to the outer lead lines, and then floating more solder into any gaps, and over the entire surface of the lead/steel combination. Solder is a robust material, and Mr. Meyer was very liberal with it. I chose specific points to separate the new perimeter from the Tiffany matrix, and desoldered these with rosin-infused copper braid. Once I had a single joint open, I could progressively work my way down the perimeters, until both sides of the panel were loose and flexible. I extracted the damaged piece, with its clear support plate attached, and turned it over to Steve for Hxtal joining. This took place overnight, and the next afternoon, I was able to begin reassembly.

This slideshow requires JavaScript.

Van Giffen: What was the most difficult part of the treatment?

Rousseau: I’d have to say removing such a large, badly shattered piece—the whole width of the panel—without causing more damage to the broken edges, the surrounding glass, or the matrix itself. Tiffany’s copper-foil technique produces delicate, organic metal channels to hold the glass, but they are quite rigid. They’re fabricated by wrapping a pair of glass edges in copper foil, and floating molten solder over their surfaces to join them as one unit, often—as in this window—with a brass fin in the center for lateral support. Each line you see is fabricated in place, customized to that specific piece of glass. If they are mishandled during treatment, they can be distorted—almost too slightly to see, but the result is they won’t hold the glass properly any longer, and there are gaps in the panel: light leaks. Disassembly and reassembly took extreme concentration.

Van Giffen: Did you learn anything interesting about the window or the old repair during the treatment?

Rousseau: Steve and I had first looked at the old repair—a sheet of clear glass, over a heavy layer of yellowed adhesive—and assumed that there was no space between them.  It appeared that Mr. Meyer had laminated the whole unit together with the original piece, for maximum strength. We expected to begin treatment of the broken piece by breaking the chemical bonds of the adhesive, which can take days; separating it from the clear support plate, and then laboriously picking adhesive residue off the fragments. Imagine our surprise when I first had the piece free of the matrix, Steve touched the clear plate, and…. it moved! Mr. Meyer had saved us hours of work, by installing the clear plate as an independent element. The adhesive, now being analyzed, turned out to have very little chemical bond to the glass; it came up in contiguous sheets with only minor effort. For 1976, these treatment choices display remarkable attention to the need for reversibility. And in fact we achieved it; when the Hxtal join was complete, Steve and I independently decided the piece was strong enough to go back into the matrix without a support plate at all. What we see now, is all Tiffany glass.

Van Giffen: Will the new treatment yellow too?

Rousseau: We hope not. All epoxies yellow to some degree, but Hxtal has consistently placed at the top of aging studies carried out over the last four decades. The projected time for significant yellowing—based on lab tests, done in an accelerated-aging chamber—is approximately a hundred years. It’s been in use in conservation labs since the mid-80’s, (including here at CMOG), and performs marvels if handled correctly. In this Tiffany repair, the join lines were very tight, so the cross-section of adhesive is on the order of microns thick. Even if the adhesive were to amber slightly, on opalescent glass it would be difficult to perceive. This is a vibrant, dynamic window in terms of color, and Hxtal has allowed us to return the sky to its true blue glory. Steve will keep an eye on the repair, but I would be surprised to see any change before either of us is ready to retire.