Building Museum Building-Sized Paint Restoration
Washington DC: 2008
What is known today as the National Building Museum started out as the Pension Building for the processing of Civil War pensions and benefits to widows and orphans. Completed in 1887, the design and construction was overseen by Army Quartermaster Montgomery Meigs. With an inspiration based in classical Italian design, the building looked very different than surrounding construction, but incorporated many of the latest Victorian concepts of providing fresh air and daylight to all the building's occupants in spite of the massive size of the structure. In keeping with period masonry, Meigs was adamant that the mortar, bricks and terracotta embellishments match exactly, but this was hard to accomplish across the 1300 lineal foot perimeter of this 75' tall building not long after the ravages of war. Maybe this is why the brickwork was painted.
When the conservation team was first asked to look at this building, the assumption was that the white blush on much of the masonry was efflorescence from prior cleaning attempts. Efflorescence sometimes occurs when salts are created by acidic or strongly alkaline masonry treatments reacting with the lime in mortar. With the most aggressive masonry treatments using hydrofluoric (or ammonium bifluoride) solutions, the sand in the bricks and mortar itself may be attacked. The first testing by the conservation team consisted of using chelation agents to unlock what was expected to be calcium salts and airborne dirt. Using nitrilotriacetic acid (NTA) and citrate gels to attach to the calcium compounds which otherwise are difficult to remove. We were surprised to find it was completely ineffective. Looking more closely at the bricks, we began to wonder if we were seeing a coating.
A small fragment of brick was viewed using a fluorescence microscope which indicated there was indeed an oil-bound coating on the surface. A pyrrolidinone paint stripper applied to the surface of a brick removed from the building in no time lifted paint and we were looking at a coarse-aggregate orange brick surface more like what one would expect from the interior of clay tile. As we looked more closely, it became clear this paint was not just on the exterior face, but that all six sides of each of the bricks we selected for removal had been coated. In other words, the bricks had been dipped in paint before being laid in the building. This made it clear that the coating wasn't an after-the-fact treatment, but done intentionally to create uniformity across the entire facade of a million and a half bricks. It also explained the odd orange streakiness we had seen near the cornice where earlier aggressive cleaning treatments had exposed some of this coarse orange surface in vertical stripes. We then realized this meant most of the white discoloration was in fact due to paint degradation not unlike the way the finish on an old car loses it colors as the coating begins to refract or scatter light rather than evenly reflect it.
Further laboratory investigation confirmed the paint was in fact a casein paint with a small percentage of oil added to make it more durable for exterior use. Since casein is made by blending milk and lime and the casein was pigmented with natural iron oxide pigments, the brick surface would test for iron (Fe+) and calcium (Ca-) just like the interior of the brick. Thus the type of testing and percentage of those compounds at the surface all were important to being able to separate brick from coating. In other words, the standard petrographic analysis used in masonry testing would not have been able to differentiate the two. As with all conservation investigations, it was necessary to be observant, inquisitive and use a broad range of testing measures. The durability of this oil-bound casein paint with significant quantities of pigment after 120 years is a testament to the quality of these historic paints that are often considered inferior to anything modern. In fact, these paints are also far healthier to use. In cleaning the building bubbles and soap smell began emanating from the surface suggesting it had previously been scrubbed with detergent to remove years of soiling, but that the crew had likely worked during very hot summer weather and not been able to keep the cleaning materials from soaking in. The existence of this sodium laurel sulfate residue in the brickwork would also have contributed to more of a white haze on the facade.
At this point, the scope of work changed from a masonry cleaning project to paint stabilization. But how to clean and repair an 87,500 sf facade? Recognizing that there was dirt and degraded paint on the surface, we wanted to gently lift this soiling away without any chance the cleaner could react with the facade. Chemist Richard Wolbers began shopping for a biodegradable, non-ionic cleaner and finally chose one used in the Silicon Valley electronics industry: Innovative Organics (a division of Saint Gobain) product Amberclean SC19U. In conjunction with gentle scrubbing using a soft bristled brush, the dilute Amberclean washed away dirt, creating a clean canvas for the paint stabilization.
Now the question was how to re-saturate the pigments, providing binder to connect where it had been lost. Again Richard Wolbers went searching with a specific list of characteristics he wanted: good bonding to metal oxides (re-saturation to improve the color density of the red pigments in the paint) but able to be unzipped (made reversible with selective chelators and a rise in pH to 10 and thus not incompatible with the lime mortar), and limited reduction in water transmissivity (we were not looking for a continuous coating nor did we want to begin trapping water). . Richard chose a polyacrylic acid commonly used for suspensions in the cosmetics industry, choosing Lubrizol's Avalure AC-315. This product has the added benefit of providing good UV protection to the remaining paint and has a life of 50 years before needing to be renewed.
After conducting vapor transmission and weathering tests in the laboratory, we found that a 5-7% solution of Avalure in ethanol provided the necessary saturation with limited loss of vapor transmission at the brick, terra cotta or mortar surface. Additionally the "gating" characteristic of Avalure helps to increase water shedding. When applied to the building with a single pass using an inexpensive Wagner power sprayer, the facade begin to generally glow more "warmly" with the red resaturated in all but the most deteriorated areas.
But what about the most damaged areas? Building upon the paintings conservation approach of "in-painting," Richard again went looking for a clear material that could be applied in a waterborne formula that again bonds well to masonry materials but is easily reversible. He was looking for something that even in dilute formulations (5%) could carry a significant amount of iron oxide pigment. He found what he was looking for in Lubrizol's inventory with a product developed by B.F. Goodrich known as Carboset AE-960. It too can be removed by raising the pH to 9 and adding a chelator of citrate or nitrilotriacetic acid (NTA).
After the cleaning with Amberclean, facade saturation with Avalure, and touchup of selective areas using Carboset, the building looked like a well-cared-for 120-year-old brick building, a far cry from the blotchy and dilapidated appearance when the project started with the inappropriate goal of cleaning the masonry before it was recognized this was a painted building. We have to remember that although we think we know and we have "conclusive" archival accounts as well as modern word of mouth about what has happened to the building, we must always go back to the source itself; the building, to unlock its secrets. Every building is unique, a product of specific people and times and having undergone its own history of treatments and interventions. Only with an open mind can we see the evidence contained within the building materials. We need to be vigilant and ask questions.