WALLPAPER ON WALLS: PROBLEMS OF CLIMATE AND SUBSTRATE
1 THE WALL AS INTERFACE BETWEEN ENVIRONMENTS
The fact that papers indisputably benefit from stable conditions of temperature and humidity has often led to recommendations that historic buildings be fitted with environmental control systems for the sake of the wallpapers as much as for the protection of furnishings and artifacts. Unfortunately, there is little expertise within the conservation field in selecting the best levels of temperature and relative humidity for historic buildings containing collections. The temperature and relative humidity levels long thought desirable in the museum field, 70°F and 50–55% respectively, are becoming recognized as too high for most historic buildings during winter. There is still the need, however, to point to ways these high levels can damage historic buildings, since some curators are not fully aware of the problem.
126.96.36.199 Condensation in Winter
It is no use trying to preserve a wallpaper while the wall behind it crumbles away due to the effects of condensed moisture on such materials as plaster and structural woodwork. Modern structures designed to be humidified in winter are fitted with continuous vapor barriers, such as aluminum foil, placed just behind the interior surfaces of exterior walls.
In this location, the vapor barrier prevents water vapor from permeating outward to the cold, outer portion of the wall, where it can condense. The vapor barrier does not in itself form a surface on which condensation can occur, because it is located in the warm portion of the wall, where its temperature is kept above the dew point of the humidified air within the building.
The increasing popularity of humidification in winter has placed new stresses on old buildings constructed without vapor barriers. Increased winter heating temperatures during modern times—only just now being mitigated by rising energy costs—have been a major reason why humidification is needed for human comfort and the care of collections.
The current energy-saving trend toward retrofitting old buildings with insulation poses other severe threats to the stability of large numbers of old structures, since pre-existing structural configurations almost always preclude the installation of the uninterrupted vapor barrier that is standard in new construction. Without a proper vapor barrier, insulation only increases the risk of condensation by lowering the temperature of the outer, cold zone of the wall below what it would be if more heat from the interior could reach it. Thus, water vapor diffusing outward through the wall, or moist air leaking outward through the countless small openings in an old structure, is apt to form liquid water or even ice.
Many types of retrofitted insulation are applied in such a way as to fill the wall cavity of an old wooden structure. This blocks off air movements that have traditionally helped to dry out condensation or water that has leaked in from outside, the insulation itself sometimes holding water like a sponge.
188.8.131.52 Possible Condensation in Summer
The phenomenon of condensation in exterior walls in winter is well-established though often forgotten by curators thinking only of the needs of the collections. Less well recognized is the possibility that condensation might occur in summer, when the interior of a historic building is too effectively cooled by air conditioning. In this case the temperature gradient across the wall is reversed, the warm air being on the outside, and the wall being coolest just at the interior paint or wallpaper. On summer days when the outdoor temperature is, perhaps, 95°F and the relative humidity 95%, and when the interior surface of a wall is cooled to 70°F, it seems likely that condensation might occur just behind the finished interior surface of the wall, especially where this surface is relatively impermeable and blocks the escape of moisture into the air within the structure. Multiple layers of oil paint might constitute such a surface.
This inverted summertime version of winter condensation may be one of the reasons interior plaster sometimes becomes crumbly with age; water cyclically condensing and evaporating could damage plaster by gradual recrystallization of the plaster materials, or perhaps through capillary forces. It might also attack wallpapers and wallpaper pastes, and could encourage fungal growth. More study is needed to assess the seriousness of this risk.
We have been discussing the damaging effects of condensation that occur when an old wall is overstressed in its role as the boundary between differing environments. A general approach in mitigating those effects is to reduce to some sensible extent the differential between indoor and outdoor environments. Forexample, by lowering heating temperatures in winter, a desirable level of relative humidity within a structure can be achieved at a lower absolute moisture content of the air. Since it is the absolute moisture content that determines the dew point, the risk of condensation in walls is reduced. Many old wall systems cannot tolerate the wide climatic differentials between outdoor and indoor air that modern home heating practices, or museum standards, produce. Such wide differentials can be safely tolerated only where specific provision is made, as by installing a continuous vapor barrier, or where the mass of a wall is great. Many art museums have massive walls that buffer climatic extremes, but, at least in New England, early wallpapers are more often found in lesser structures.
184.108.40.206 Pattern Staining
Wallpapers and their substrates can be damaged by mechanisms other than condensation, where an old wall is climatically overstressed. A steep temperature gradient across an exterior wall in winter can lead to accelerated dirt deposition on historic paints and wallpapers, by means of the phenomenon known as pattern staining. This effect is widely observed in historic buildings where the wall surfaces of interior spaces are not frequently repainted or washed.
Dirt or smoke particles suspended in air move about when struck by air or water molecules, which are in constant motion. At higher temperatures, the molecules of air or water vapor move more vigorously than at lower temperatures, and strike dirt particles more energetically. Thus where the temperature of air varies between adjacent areas, dirt particles will tend to move from the warmer area, where they are struck harder, toward the cooler area, where the impulses in the reverse direction are generally not so strong.
In winter, the interior surfaces of exterior walls are usually colder than the air they contain, and there is usually a temperature gradient in the air near the wall. Dirt will thus be propelled from the air onto the wall surface, in amounts proportional to the steepness of the temperature gradient extending from the wall surface into the adjacent air. This is believed to be the mechanism that causes the interior surfaces of exterior walls often to be so much dirtier than the surfaces of partitions between rooms, even where there are no dirty hot air registers or steam radiators near outside walls. It accounts for the way the pattern of lathing in exterior walls and in ceilings gradually becomes visible through the plaster, on the surface of paints and wallpapers. The laths are better insulators than the plaster keys between them, so that the interior wall surface is kept a little warmer in winter where it overlies a lath than where it overlies a deep plaster key between laths. Thus, where there is a lath behind, the wall surface is less strongly “attractive” to dirt and remains cleaner.
The most pronounced pattern staining effect is often the clear silhouetting of the heads of nails that affix interior trim on exterior walls: the highly conductive nails lead indoor heat toward the cold outdoors, cooling the layers of paint and putty over the nail head, and causing a black spot to appear on the top layer of paint. The silhouette is often so precise that the character of the nail head—hand wrought, cut, or made from round wire—can be told.
Several approaches can be used to reduce the deposition of dirt on wallpapers by this mechanism. First, filtration systems and clean heating apparatus can reduce the amount of dirt in the air. Secondly, protective varnishes on wallpaper can make dirt removal easier, although it is still important to minimize dirt deposition. Thirdly, insulation can be used behind wall and ceiling surfaces to keep plaster and wallpaper warmer—near the temperature of the adjacent air. However, as discussed above, the problems of insulating old walls without risk of condensation have not been solved.
Finally, the phenomenon of pattern staining can be reduced by lowering indoor temperatures in winter, lessening the steepness of the thermal gradient that impels dirt toward the walls. Pattern staining does not seem to occur actively in unheated buildings.
220.127.116.11 The Best Compromise
Many elegant early houses used only in summer and not heated at all in winter since perhaps 1900, seem to have benefitted from the effects of this annual refrigeration. This can also apply to all their fine furnishings, including wallpapers, which sometimes seem almost miraculously preserved. The wide swings in temperature between summer and winter do not seem to matter too much, since the relative humidity—the more critical factor—is much more constant throughout the year than in heated, unhumidified houses. In addition, one may presume that cold winter temperatures reduce the rates of oxidation of wood, fabrics, papers, and finishes, since the rates of chemical reactions are, as a rule of thumb, halved for each 10°C (18°F) drop in temperature.
On the other hand, there have been many cases where immediate and severe damage occurred to buildings and their contents in the first winter that the heat is not turned on. We need to be able to predict more accurately which effect will occur: drastic peeling of paint and wallpaper, or “preservative refrigeration.”
There are many ways that condensation can occur in unheated buildings. A well-known example is that of a building that becomes filled with warm, humid air during warm, damp weather, and that retains this air while temperatures drop: the sudden cooling may produce condensation within the building. Conversely, interior building surfaces cooled during long periods of cold weather may gather condensation immediately when the weather changes and warm, humid air penetrates from outdoors.
Some amount of winter heating would presumably guard against these kinds of condensation effects by keeping the building interior steadily above the dew point. Thus a middle course, which is not yet well charted, must be struck between the extremes of leaving an old building unheated in winter and attempting to maintain exactly the same temperature and humidity levels throughout the year. The objective must be to eliminate the sharpest peaks and deepest valleys of the annual cycles in temperature and relative humidity, to the extent that this can be done without overstressing the exterior walls. Moderate heating in winter with extremely cautious humidification, and dehumidification in summer without excessive cooling, would probably be a sensible program for many old buildings. More specific guidelines are needed. No material in an old building will suffer more drastically than the wallpaper from our continued lack of knowledge and experience in climate control.
1.2 THE SUBSTRATE
Plaster, the most common substrate for wallpaper, is subject to various ills, most of which could be remedied fairly easily were the wallpaper not present. Depending on the problem found, the plaster treatment intended, and the possibility of access from behind, wallpaper may or may not have to be removed temporarily for the plaster to be treated.
Lime/sand plaster is the type most commonly found behind wallpapers in old New England buildings. An unsanded skin coat of pure lime or lime and gypsum may also be found, although such “neat” plasters were apt to be reserved for smooth finishing of areas meant to be painted. Plasters containing these materials vary widely in strength on account of differences in formulation, and even the soundest plasters become crumbly when attacked by water from roof or plumbing leaks, or when slowly degraded by condensed moisture. Plasters containing clay binders are by nature friable.
The thinness and open porosity of most plaster renders penetration by consolidants relatively easy to accomplish. The problems are encountered in getting access to the plaster that is covered by paint or wallpaper, and in carrying out a treatment of the plaster without affecting these paints or papers. Each case will differ.
This writer has had experience only with organic materials for plaster consolidation, but the use of inorganic consolidants designed principally for carbonate stones might sometimes be justified. These include the methods of Lewin and Sayre for forming insoluble barium salts within stone to act as supplemental binder. Toxicity and alkalinity are concerns with some of these processes, as is the need to keep the treated substance moist for some time.
Some alkoxysilanes are being successfully used as stone consolidants, and appear to work on small plaster samples of various compositions. These materials provide reinforcement through the formation of a network polymer of silica, often incorporating water-repellant alkyl groups.
Where a plaster is permeable enough to accept a more viscous type of consolidant, solutions of organic polymers are relatively simple to use, since no chemical reactions need be carried out to bring about solidification. A wide range of polymers can be used. This writer has directed the consolidation of several plaster walls and ceilings using solutions of polyisobutyl methacrylate, sometimes rendered more flexible with an addition of poly-n-butyl methacrylate (Rohm and Haas Acryloids¯ B-67 and F-10, respectively). The solubility of these materials in aliphatic hydrocarbon solvents of relatively low toxicity is a major advantage when craftsmen must impregnate larger areas of plaster, working in close quarters. Where conditions allow the use of highly flammable solvents, acetone can be combined with hydrocarbon solvents to lower the viscosity of the polymer solution at any given content of polymer solids.1 A 15% solid solution, applied repeatedly over a period of about an hour, can penetrate deeply into typical plasters, effectively strengthening and bonding together base and skim coats.
Research I am now conducting indicates that where the viscosity of a polymer solution of useful solids content is too high for adequate penetration, plaster can be consolidated using low viscosity solutions of acrylic monomer and solvent. Once the solution has been soaked into plaster, polymer can be formed by a chemical reaction and can be made to precipitate out of the solution, lining pore walls.
These techniques will almost inevitably require the temporary removal of wallpapers. Why, then, should the plaster not simply be replaced? In some cases replacement may be justified. In many cases, however, original plaster must be considered a valuable element of a historic building, meriting preservation. Additionally, its texture and irregular character may favorably affect the appearance of re-installed wallpaper.
18.104.22.168 Reattachment of Plaster
Plaster loosened from its laths or other supporting structures can be reattached using a variety of techniques, many of which may not require removal of wallpapers. Depending on conditions of access, a variety of backing plasters or adhesives can be formed or injected so as to secure the old plaster. One technique long used in England to support falling ceilings is to remove some or all laths from above, form a new supporting metallic or cloth mesh, and form through this a new plaster against the carefully cleaned back of the old.
Another approach is to bond loose plaster to its laths, or other substrate, by injecting adhesives. I have used injectable adhesive foams based on acrylic resin emulsions and containing a material that compensates for drying shrinkage.2 Limecasein mixtures are more traditional but may entail the risk of microbiological deterioration.
22.214.171.124 Infill and Replacement
There are usually no major problems in replicating, at least in appearance, an old plaster for the purpose of patching-in or replacement. The traditional materials used in older plasters—lime, sand, gypsum, clay—are still widely available and as easy to use today as they were in past times. Hair from obsolete breeds of farm animals may no longer be available, but this degree of precision in reproduction is superfluous. Indeed, new plaster should probably be “marked” in some way to indicate on close inspection that it is not original. I have used a sparkling crushed garnet abrasive for this purpose in mortar, as part of the aggregate. Well-known practices used in matching mortar, such as extracting the original sand to serve as a guide in selecting new sand, are applicable to the replication of plasters.
Variations from original formulas can save labor and improve durability while not altering appearance. A common early practice in using lime plasters, for example, was to compress the plaster repeatedly with a trowel while it dried, to prevent the development of shrinkage cracks. By substituting gypsum for a portion of the lime, the need for this can be eliminated, since gypsum sets without shrinkage and, by reacting with the mixing water, can give the plaster some strength while it is still wet.
Quite modern additives may be needed in a new plaster meant to replicate an old one. Where an old plaster has fallen away because of insufficient provision for mechanical keying, resinous admixtures can insure the security of new plaster by giving it a more adhesive character. Acrylic resin emulsions are one of the classes of materials used for this purpose, either being added to the plaster or painted onto the substrate immediately before placement of the plaster. Modern emulsion technology thus provides the conservator the means to blend hydrophobic and adhesive polymers into water-mixed masonry-type materials. It is often convenient for the conservator to formulate and make materials, such as crack-filling compounds, that have more durability than ready-made products.
In some cases, a totally new or drastically modified substrate may be required for wallpaper. There might be a case in which a thoroughly modern, lightweight, perlitefilled plaster with thermal-insulating properties would be the best answer when re-mounting a conserved paper. However, details must be worked out to insure that the historically correct appearance of the repapered wall will not be sacrificed. Metal lath, for example, used to provide improved mechanical keying for new plaster, may have no effect on the appearance of walls or ceilings where the original plaster was thick and smooth. In more primitive buildings, with thin plaster over bumpy wood lath, metal lath can spoil the original effect.
126.96.36.199 Wood: A Question of Acidity
Although the strong alkalinity of fresh plaster can damage paper applied too quickly, well cured plaster might conceivably exert a buffering action on the acidity of some wallpapers. This is mere speculation, but at least plaster is not acidic. The practice of applying wallpapers to woodwork was quite common in simple dwellings in New England, especially as styles changed and older wood sheathing, no longer considered attractive, was papered over. In many cases intervening paint layers protect the paper somewhat from the acidity of the wood, but equally often the paper is found directly on the wood. The overlay of interesting papers and woodwork is often a delightful and informative spectacle for the antiquarian, but papers found directly overlying wood would surely benefit from temporary removal and reinstallation over an acid-free liner paper or perhaps a reversible and chemically inactive paint or varnish.
1.3 WHAT TO DO?
Conserving valuable papers on the walls of historic buildings presents many unsolved technical problems, and many philosophical ones on which persons having different viewpoints will seldom agree. The measures that will best preserve the paper may disrupt architectural evidence, damage architectural fabric, and compromise the authentic appearance of the papered wall as a whole. With more experience and study and better techniques, we may learn to reduce the severity of the sacrifices that have to be made.
Historic buildings, particularly those open to the public as museums, offer an opportunity to enjoy works of art in a context far more natural and real than that of the art museum. This is the special value of museum houses—that objects are set about in the places they always stood, that a genuine landscape or a real building is seen through the windows. Not every historic building museum should be permeated with an atmosphere that feels artificial, unrelated to the season of the year and produced by equipment that makes quietly unnatural sounds. Some house museums should have open windows in summer, through which breezes blow.
And yet, how shall we preserve the wallpapers?