PREVENTIVE CONSERVATION RESEARCH AND PRACTICE AT THE BRITISH MUSEUM
2 TEMPERATURE AND RELATIVE HUMIDITY
2.1 COPPER ALLOY, WEEPING GLASS, IVORY AND BONE
2.1.1 Copper Alloy
In the 1960s and '70s the emphasis of conservation in the British Museum was on the treatment of objects. Stabilization through treatments, such as soaking bronzes in sodium sesquicarbonate solution or stripping to remove corrosion layers, was seen as the appropriate response to deterioration. In the Museum Research Laboratory, studies into the deterioration of copper alloy artifacts had shown that bronze disease could be prevented by controlling the relative humidity to below 35%, stopping the reaction of nantokite with copper metal and moisture from the air to form paratacamite (Organ 1963).
This research was implemented in the design of six new galleries opened between 1968 and 1975. Dehumidified showcases were incorporated to display Middle and Near Eastern copper alloy objects which were particularly prone to bronze disease. The showcases were controlled with desiccant dehumidifiers supplied by Munters Ltd. (Newey 1987). A dehumidified storage area was constructed for the copper alloy objects not on display. Although it took some time to limit the use of chemical stabilization methods to those objects for which a controlled environment could not be provided, control of relative humidity is now the primary method of preventing bronze disease.
2.1.2 Weeping Glass
Weeping glass was shown to be stable if kept at an RH of 38-42%, and all of the weeping or crizzled glass was moved to a dehumidified showcase where the objects could be viewed without being handled (Organ and Bimson 1957; Werner 1958). This decision was based on the identification of potassium carbonate salts on the surface of the glass. At that time, analysis was restricted to those ions expected to be present, so it is not known whether some of the ions detected today on the surface of glass were present in the 1950s (Bimson 2004). The critical relative humidity for potassium carbonate was determined as 42%, above which it takes up moisture from the air, forming surface droplets which give the appearance of sweating and provide an aqueous medium into which more alkali ions can be leached from the glass.
2.1.3 Ivory and Bone
The collection contains ivory and bone objects from many different cultures of widely differing ages. Among the highlights of the collection are the Lewis chessmen, carved from walrus ivory and dated to AD 1150–1200. Some of the chessmen were found to be cracking and exfoliating and in 1970–71 all the pieces were examined by scientists working in Dental Anatomy at University College London, who identified the cause of the problem ( Boyd 1971). The chess pieces which were deteriorating had been carved in walrus ivory which had been boiled. This had caused denaturing which made the ivory more susceptible to cracking in response to changes in relative humidity. In 1978 the chessmen were exhibited in a showcase controlled by a ducted humidifier to 50–60% RH. Recently this has been redefined to 45–55% as the control achieved was always close to 50% and the chessmen remain stable.
Ivory objects from Nimrud which had been burnt in antiquity are not humidified. They have been stable in an uncontrolled storage area that has naturally maintained 40–45% RH and have been seen to react adversely to higher and lower relative humidity. The most dramatic disruption occurred when an ivory lion head (ANE 132697) fragmented following a move to new storage. It was found to have fractured along old cracks which had been consolidated with polyvinyl acetate resin. Soluble salts, gypsum, and halite were present in the cracks and their crystallization at the low relative humidity sustained during the move was considered a contributory factor in the deterioration of the object (Thickett and Bradley 1998).
2.1.4 Control of Relative Humidity in Showcases and Storage Areas
For copper alloy objects susceptible to bronze disease, weeping glass, and ivory, the control of relative humidity in both showcases and storage areas has been achieved by only dehumidifying to maintain a low relative humidity and only humidifying to maintain a higher relative humidity. There has been respectively no counter humidification or dehumidification, and no temperature control other than winter heating. This has meant that for the dehumidified cases the winter low internal relative humidity has been accepted and for humidified cases the summer high relative humidity has been accepted. The relative humidity inside the controlled showcases has drifted accordingly, although the well-sealed showcases in use since 1988 buffer effectively against the peaks and troughs in relative humidity. The fact that no observable damage has occurred to any of the copper alloy or ivory objects in a controlled relative humidity has justified this approach. There have been instances of glass apparently starting to weep in the dehumidified case. Research into this problem is ongoing with the focus on occasional very low relative humidity compared to the target of of 38–42% RH, and high levels of indoor pollutant gases in the dehumidified case.
2.2 COMPLEX MATERIALS
2.2.1 Lindow Man
When Lindow Man, a freeze-dried bog body dated to the Iron Age, was put on exhibition, avoiding the seasonal extremes of relative humidity was thought necessary. Having no idea how the body would react to changes in relative humidity, Museum scientists and conservators were keen to stabilize the environment as much as possible. The body had been immersed in a solution of polyethylene glycol 400 prior to freeze-drying and after treatment was acclimatized to ambient conditions in the Museum (Omar et al. 1989). Because low molecular weight polyethylene glycol was known to take up moisture from the air and “sweat” at high relative humidity, it was decided that the maximum allowed would be 60% RH. It was also decided to set the minimum to 50% RH to limit the potential for dimensional changes.
To supply these conditions the showcase was constructed with an integral humidifier (Defensor PH5) and a dehumidifier (Munters M120), both controlled by a Sauter HSC hygrostat. The humidifier was modified with a large funnel that covered the air outlet, connected by large diameter tubing to a hole in the showcase base. The selection of the dehumidifier was based on a comparison of the operation of the condensation and desiccant equipment. Desiccant equipment which had already been used successfully in the Museum was selected. The dehumidifier was connected to the case following the manufacturer's instructions.
The showcase was monitored using an electronic RH/temperature probe which showed the relative humidity was tightly controlled in the range 53–57%, but the temperature range was 18–30șC. At 55% RH the moisture envelope surrounding the body was 9–18 g/m3, a much wider range than anticipated. However, physically the body has remained in good condition since it was put on display in 1989. The color post-conservation has apparently lightened and because of this in January 2005 an examination of the body was carried out. This showed the body to be in excellent condition, but some extrusion of deteriorated PEG had occurred. Color monitoring of Lindow Man while on display had indicated that fading occurred during the exhibition period between 1987 and 1997, when the light levels on the body were up to 400 lux. In 1997 Lindow Man was moved to the Late Bronze Age and Celtic Europe Gallery where it is displayed under a maximum light level of 50 lux. Since then the color change has slowed dramatically.
2.2.2 Ethnographic Objects
The Museum's Ethnography collection is now stored and displayed at a range of 45–55% RH based on the large number of organic objects. There is no temperature control in the main storage area, but on exhibition, air cooling has been installed to keep the temperature in the galleries below 25șC. Much of the collection is stable but this range of relative humidity promotes the deterioration of some objects.
When the deterioration of Maori objects made of New Zealand flax was investigated, the black dye used to color them, an iron/plant polyphenol complex, was found to be the cause (Daniels 1999). The dye was shown to become acidic causing hydrolysis of the cellulose fibers, and the iron in the dye caused oxidation. The research showed that storing the objects at a lower relative humidity would reduce the rate of hydrolysis. However this might affect the dimensional stability of the objects which have acclimatized to the 45–55% range since 1972. The decision was made to keep black dyed New Zealand flax objects in the main storage area.
An investigation of the mechanism involved in localized browning of sugar objects from Mexico showed the cause to be Maillard browning, a complex reaction of reducing sugars and protein-containing compounds. The sugar commonly used in cooking and sweetening is sucrose which is not a reducing sugar. However, experiments showed the rate of hydrolysis of sucrose to the reducing sugars, glucose, and fructose, increased with increase in relative humidity. The decoration on the objects was made from icing sugar which contains egg white, the source of the protein. As a result of this research, the objects were moved to a specially created low relative humidity, low temperature storage area (Daniels and Lohneis 1997).
2.2.3 Stone Sculpture
Control of both relative humidity and temperature which required full air conditioning, was justified for sculptures from the Great Stupa of Amaravati, India. The sculptures were carved in a green-tinged, partially metamorphosed limestone, commonly known as “Palnad marble,” and were deteriorating by powdering and flaking, causing considerable loss of surface detail. Between 1960 and 1992, several campaigns of scientific investigation were undertaken (Bradley and Freestone 1992) identifying five potential causes of deterioration. The cleavage planes created by the folia caused the stone to split readily. The clay minerals in the stone were subject to softening and volume change at high relative humidity, causing surface flaking. Disruption of the surface through powdering was caused by a combination of crystallization and dissolution of soluble chloride along with dissolution and re-precipitation of calcite by reaction with carbon dioxide and water and surface sulfation. This diagnosis suggested that the sculptures should be stored and displayed at a controlled temperature and relative humidity of 18–20șC, 30-40% RH with filtration to remove sulfur dioxide. Under these conditions in storage and on exhibition in the Asahi Shimbun Gallery of Amaravati sculptures, the sculptures have been stable. (Bradley 2003a).
2.3 RECENT CHANGES IN CONTROL OF RELATIVE HUMIDITY IN SHOWCASES
The method of control used in showcases has improved in the last eight years. In new gallery developments, a new type of control system is being used instead of humidifiers and dehumidifiers. These controllers, developed by Glausbau Hahn Ltd., utilize a Peltier cell which cools air and hence dehumidifies without the use of refrigerants or drying wheels. It is based on a thermo-electrical effect discovered by Jean Peltier in 1834. The system can control to any range of relative humidity by feeding air into the showcase at the required midpoint at a slow rate.
Units of different capacity have been used to control the relative humidity in individual showcases, and from a central unit to groups of showcases in a gallery. The former approach has been used to control the relative humidity in individual small showcases in several galleries and in very large showcases in the Wellcome Trust Gallery where ethnographic objects are displayed. The latter approach has been used to control the relative humidity in all of the showcases in the North America gallery where ethnographic objects are displayed, and in the gallery where objects from Korea are displayed. Passive methods, such as use of conditioned silica gel (Artsorb or Prosorb), are used to control relative humidity in only three showcases in the Museum because of the difficulty of maintaining the in-case conditions when the temperature in the gallery is not controlled and the labor-intensive nature of the installations. Currently there are 103 controlled showcases in the Museum galleries.