Ten years experience of energy efficient climate control in archives and museum stores
Poul Klenz Larsen, Morten Ryhl-Svendsen, Lars Aasbjerg Jensen, Benny Bøhm
The National Museum of Denmark, Department of Conservation, DK-2800 Kgs. Lyngby. and Tim Padfield Devon, UK.
Summary: Ten years experience of low energy museum storage in Denmark confirms the practicality of building lightweight stores on uninsulated floors, in contact with the ground which acts as a thermal buffer. The high summer RH is best reduced by dehumidification, which can be intermittent and therefore solar powered.
Here is the full article as a pdf file (1.2Mb)
Energy efficient climate control has been used for archives and museum stores in Denmark for more than a decade. The challenge is to design the building to keep a moderate temperature and relative humidity all through the year by passive means. In the early days it was supposed that a very heavy building with high heat capacity would be the best. This idea emerged out of experience from historic buildings, and the design of many new archives was based on this assumption. But recent studies of archives and museum stores in use, combined with computer simulation, shows that this is only partly true. A well insulated, lightweight building with one level and an uninsulated floor laid directly on the ground has thermal stability just as good as a more heavily built structure. It is possible to achieve an annual temperature cycle between 8 °C and 16 °C without any heating or cooling. When combined with solar driven dehumidification, the relative humidity (RH) is kept around 50 % through the year. The use of a humidity buffer as the interior wall cladding will compensate for occasional periods of extreme weather. To achieve climatic stability, the infiltration of outside air must be low. External pollutants will be kept out of the store but internal pollutants must be limited by the use of inert materials and finishes.
Figure 1: The Alcazar of Segovia, Spain, is home to the military archive located in the basement. The natural climate is cool and humid.
It was not entirely by intuition that the first passive archives were designed with heavy outer walls. The experience from archives kept in historic buildings for centuries suggested that thermal stability was a key parameter for long term preservation. The military archive in Segovia Castle in Spain is one such example (fig. 1). The archive is situated in the basement, next to the limestone bedrock and with a 2 m thick wall to the outside . The annual temperature variation in the vault is from 10 to 15 °C, compared to 0 -30 °C outside. The relative humidity is rather constant at 70-80 % due to the immense humidity buffer capacity of the many tons of paper. It is the archive itself that moderates the relative humidity, not the heavy walls.
The performance is helped by a very slow infiltration of outside air. The windows are few and small, and only opened occasionally for a few hours by the archivist. This archive has worked for over a century without any active climate control. It does, however, have a disquietingly high RH, which inevitably moves towards the outside RH, because the average temperature inside is equal to the average temperature outside. Similar climatic conditions are observed in most of northern Europe.
Figure 2: St. Catherine’s Monastery in Sinai, Egypt has been a safe refuge for the famous library for 1500 years. The natural climate of the library is warm and dry.
A contrasting example of an uncontrolled environment inside a heavy building is St. Catherine’s Monastery in Sinai (fig. 2). This desert fortress has been a safe refuge for the famous library for 1500 years. The solid walls were put up for security rather than for climate control. The annual temperature variation in this location is 20 - 30 °C and the RH is 15 -30%. The low RH is characteristic of this high desert environment. These climates are far outside the limits set by standards or recommendations for archives, yet the documents are well preserved.
Figure 3: The City Archive in Cologne (1971) was designed with thick outer walls for thermal inertia. The interior was heated and ventilated to control the RH.
The idea of using very heavy buildings for archives was adopted for the Cologne City Archive, built in 1971 as a pioneer effort at climate control without air conditioning (fig. 3). Behind the granite tile facade was half a metre of brick wall set in a concrete frame. The climate control in the archive was partly by thermal inertia, which flattened out the daily variation of outdoor temperature. This was supplemented by cross-ventilation through the narrow windows and by heating in winter. The RH was so strongly buffered by the archived material that it hardly changed over periods of many weeks.
Figure 4: The Suffolk Record Office, Ipswich, UK (1990), had winter heating to 16 °C. The RH was stabilised by the humidity buffer capacity of the documents.
Sadly, this building does not exist any more, but the ‘Cologne model’ survived in several archives that were built in the following years. The Suffolk Record Office in Ipswich (UK) shows the effectiveness of simple climate control without mechanical ventilation (fig. 4). This two storey building has cavity walls of brick and porous silicate blocks and a ventilated attic. The annual average RH is reduced by winter heating, enabling the archive to cruise through the summer, relying on humidity buffering by the contents .
This concept worked for many years, but the temperature of the archive was not always within the narrow limits set by the standard BS5454:2000, so an air conditioning system was installed a few years ago. Current standards and recommendations are not compatible with passive climate control. The international standard ISO 11799 quotes 14 -18 °C and 35 - 50% RH as the allowable range for temperature and humidity in archives and libraries in use . The British Standard 5454 proposes 16 -19 °C and 45 - 60 % RH as limiting values for the specification, with even narrower limits for permitted variation . It takes a brave conservator or librarian to speak against these strict limits for climatic variation.
Figure 5: The Arnamagnæan collection of medieval books and documents is located on the second floor, behind the windowless section of wall, in an office building at the university campus in Copenhagen (2004).
The director of the Arnamagnæan Institute in Copenhagen chose simple, fail-safe climate control over constant temperature and relative humidity ensured by full air conditioning when a new archive was designed for the collection of medieval documents. The archive was located on the second floor in an ordinary office building, next to heated spaces (fig. 5). A multi-layered structure encapsulated the archive . The load bearing structure was 240 mm concrete walls and floors, which gave some thermal inertia to the archive. Thermal insulation was 200 mm mineral wool facing the heated offices but only 50 mm insulation in the outside wall. The thermal insulation was calculated to allow the temperature inside the archive to stay about half way between the outside temperature and the office temperature. An annual temperature variation between 15 and 23 °C was achieved without any regulation within the archive. The inside of the walls was lined with 50 mm porous lime silicate blocks to moderate variation in RH. A small ventilator takes in outside air on occasions when the water vapour content is right for pushing the interior RH towards the specified ideal. It raises the air exchange rate to 0.5 per hour, which has no significant effect on the temperature. As a consequence, the RH is within 50 - 60 % all year, with very little fluctuation. The climate control has worked for 8 years with impressive stability (fig. ).
Figure 6: The climate record for five years from the Arnamagnæan archive shows the annual variation in temperature between 15 and 23 °C (lower trace) and the stability of RH between 50 - 60 %. The intermittent blips in RH show the climate in the conservation workshop when the data logger was tapped
Figure 7: Interior view of a store for historic musical instruments, located in an industrial building. The temperature is 10-23 °C controlled by heat leaking from the adjacent spaces. This keeps the RH between 45 - 60%.
A similar concept of passive conservation heating was adapted for a new store for historic musical instruments in Copenhagen (fig. 7). The single level store was established in an industrial building. It is surrounded by workshops, which leak heat into the store in winter. The walls separating the store from the adjacent spaces are 190 mm porous lime silicate, offering a considerable humidity buffer to the store. The instruments’ thin wooden panels have little inherent humidity buffer capacity, but are very susceptible to variations in RH. However, there is no reason to worry about false notes. The annual variation has been within 45 - 60 %RH for many years. The store has no mechanical ventilation and depends on a low air exchange rate to work. The infiltration of outside air averages one room volume per day. This is much too low an air exchange for human health and comfort, but appropriate for the wellbeing of museum objects.
Outside air is a main source of pollutants and particles that cause corrosion on metals and chemical degradation of organic materials. Nitrogen oxides, which originate mainly from traffic and combustion, will engage in acid hydrolysis after transforming into nitric acid on indoor surfaces. Ozone, a natural component of outdoor air, deteriorates materials by oxidation. For a museum store with a high air exchange rate it would be necessary to filter the intake of new air for pollutant gases and particles. However, for a storage room with a very low air exchange rate (less than one exchange per day) the infiltration of ambient pollutants is just as efficiently retarded by the building fabric itself as can be achieved by mechanical ventilation and filtration. Internally generated pollutants, such as acetic acid, are best prevented by the choice of inert construction and furniture materials and finishes. For collections containing unstable objects, internal recirculation of the air through carbon filters is an option [6,7].
Figure 8: The solid concrete shelter for fighter airplanes at Værløse, Denmark, now used as a temporary store for a collection of historic furniture. The annual temperature variation is 0-25 °C and the RH is controlled to 50% by dehumidification.
Figure 9: Climate record over five years from the concrete shelter on Værløse airfield. The RH is controlled by dehumidification. Notice the initial rise in RH before the dehumidifiers were turned on.
The climate control in the musical instruments store may be passive, but it does use energy: it depends on the heat flow from the rest of the building to function. The annual temperature variation is much larger in an unheated building like the solid concrete shelter shown in Figure 8. It used to protect fighter airplanes in Værløse airfield (Denmark) against a nuclear attack, but is now in use as a temporary store for historic furniture. There is no daily fluctuation in temperature, but the annual cycle is from 0 °C in winter to 25 °C in summer (fig. 9). The large thermal inertia of the half metre thick roof is not enough to even out the temperature over the year. The inside temperature follows the outside daily average. The concrete roof would have to be 4 m thick to give thermal stability, so only underground spaces can maintain a constant temperature over the year.
The RH in the store is controlled by an absorption dehumidifier. It is needed all year as the infiltrating air maintains a high RH, because the inside air temperature scarcely differs from the daily average outdoors. However, the dehumidification requirement is quite small, because the shelter is almost air tight. The air exchange rate is only 0.03 per hour, so it takes one and a half days to replace the air volume of the shelter with outside air. Though this store is not passively controlled, it takes only 5 kWh per cubic metre of storage space, per year, to keep this store dry (table ).
|Arnamagnæan archive||15 - 23||50 - 60||0.05||15||heat|
|Musical instrument store||10 - 25||40 - 60||0.05||10||heat|
|Værløse hangar||0 - 25||45 - 55||0.03||5||dehumid.|
|Ribe museum store||8 - 16||50 - 55||0.04||2||dehumid.|
Table 1: Specifications for the energy efficient archives and stores. AER = air exchange rate.
If the Værløse building had a roof of 500 mm mineral wool instead of concrete, the temperature variation would be much smaller. Such a building does not exist, but computer modeling predicts that the annual temperature variation inside a lightweight store will be 7 - 13 °C . For temperature stability, a lightweight insulated structure is better than heavy weight without insulation. The stability depends on heat storage in the ground below the building. The floor should not be insulated, to allow the seasonal transfer of heat.
Figure 10: A well insulated museum store for a local museum collection in Ribe (2006). The annual temperature variation is 8 - 16 °C and the RH is controlled to 50% by dehumidification.
Temperature measurements in a recently built museum store in Ribe (Denmark) have confirmed that heat diffuses into the ground in summer and back into the space in winter . Apart from being a store for thousands of museum artifacts, this building has a remarkable ability to store energy. In winter the temperature is just about right for lowering the RH to an acceptable level, but in summer the temperature is lower than that outside so there is a need for dehumidification. The annual energy use for this type of store is about 2 kWh/m3. It is neither passive nor energy neutral, but it is very energy efficient (fig. 10).
Figure 11: Cross section of an off-grid museum store. Summer heat is stored in the ground below and released into the building during the winter. The RH is reduced by solar driven dehumidification in summer and by the raised temperature in winter.
The actual storage building in Ribe has a connection to the power grid, but calculations show that with photovoltaic elements covering 5% of the roof surface the store would have a sufficient power supply, so the building would be entirely independent of external energy supply (fig. 11). Due to the unpredictable nature of solar power, the off-grid store needs a humidity buffer. Conventional building materials like brick and concrete are rather poor humidity buffers. The walls should be lined with unfired clay brick with open perforations to allow water vapour free and fast access . Unfired brick is not a standard product in the local building materials supplier, but can be procured from the brickworks before firing. It therefore has the advantage of a relatively small carbon footprint. When combined with a lightweight structure, the embedded energy will be much less than for a massive building made of fired bricks or concrete. The total input of energy over the lifetime of the building is negligible compared with conventional museum stores and archives.
Figure 12: Climate ranges for the different climate control strategies are superimposed on a diagram showing curves of equal hydrolysis reaction rate, as proposed by Sebera , modified by Padfield .
The question remains, is the emphasis on energy conservation acceptable for the long term preservation of the artifacts? When light and air pollution are excluded, temperature is the main driver for chemical degradation of organic materials. The rate of decay for reactions involving moisture (hydrolysis) depends also on the relative humidity. The combined effect of temperature and RH is described in figure 12 [11,12]. The decay rate is set to 1 for a constant 20 °C and 50% RH. The curved lines connect conditions causing a constant relative reaction rate. Low temperature and low relative humidity is obviously better than higher temperature and RH, and temperature is in general more important than relative humidity, over the acceptable range of RH, which cannot go too low because of damaging dimensional change in organic materials. From this point of view there is really no benefit in keeping a constant temperature. The temperature variation itself may be a hazard to objects that become brittle at low temperature, such as some paints and plastics, but only if the object at the same time is experiencing a mechanical shock e.g. during transport. There is no evidence of damage to objects in storage caused by the moderate temperature variation we propose.
Figure 13: Diagram with lines of equal dewpoint superimposed on the relative reaction rate diagram. Reducing the RH by heating does not alter the dewpoint (red arrow) while direct dehumidification by absorbing water from the air reduces the dew point (green arrow).
The decay rate depends on the choice of climate control strategy. Figure 13 shows the decay rate diagram superimposed with lines of equal dew point. The dew point is a measure of water vapour concentration. The diagram shows the effect on the decay rate of the two types of humidity control used in the examples presented in this article. In the case of dehumidification, water vapour is removed from the air, so the dew point falls. The decay rate decreases due to the lower RH, shown by the green arrow in the diagram. In the case of heating to reduce the RH, the dew point remains unaltered, shown by the red arrow. The RH is reduced to the same target level, but the chemical decay rate is higher for heating than for dehumidification. Not only is dehumidification more energy efficient than heating, it is also better for preservation.
Low energy museum storage can be built with no detriment to the collection. Indeed, a lower winter temperature than current standards permit will enhance durability. There is no evidence for a conflict between energy saving and conservation. Current standards evolved at a time when a constant climate, maintained by the best available technology, was implemented as a precautionary measure without regard for energy economy. There is no evidence that constancy of temperature, typically confined within a 5 degree window, is necessary for the preservation of the collection. Low energy storage requires a modest annual variation in temperature but is capable of maintaining the RH within a 10% window.
1. Padfield, T and Larsen, P.K (2004). `Designing museums with a naturally stable climate'. Studies in Conservation 49, 131-137.
2. Padfield, T., Larsen, P.K., Jensen, L.A., and Ryhl-Svendsen, M. (2007). `The potential and limits for passive air conditioning of museums, stores and archives'. In: Padfield, T. and Borchersen, K. (eds.) Museum Microclimates, National Museum of Denmark, Copenhagen. pp 191-198.
3. ISO 11799:2003. `Information and documentation - Document storage requirements for archive and library materials'. International Organization for Standardization, Geneva. (German version: DIN ISO 11799:2003. Anforderungen af die Aufbewarung von Archiv und Biblioteksgut. Deutches Institut für Normung e.V. , Berlin.)
4. BS 5454:2000. `Recommendations for the storage and exhibition of archival documents'. BSI, London.
5. Padfield, T. and Larsen, P.K. (2005). `Low-Energy Air Conditioning of Archives'. In: Verger (ed.) ICOM Committee for Conservation, 14th Triennial Meeting, The Hague, 12-16 September 2005 pp 677-80. James & James (Science Publishers) Ltd., London. (Also published in Journal of the Society of Archivists 27 (2006), 213-226.)
6. Ryhl-Svendsen, M. (2006). `Indoor air pollution in museums: prediction models and control strategies'. Reviews in Conservation 7, 27-41.
7. Ryhl-Svendsen M., and Clausen, G. (2009). `The effect of ventilation, filtration and passive sorption on indoor air quality in museum storage rooms'. Studies in Conservation 54, 35-48.
8. Boehm, B., and Ryhl-Svendsen, M (in press). `Analysis of the thermal conditions in an unheated museum store in a temperate climate. On the thermal interaction of earth and store'. Energy and Buildings 43 pp. 3337-3342. DOI:10.1016/j.enbuild.2011.08.028.
9. Ryhl-Svendsen, M., Jensen, L.A., Larsen, P.K., Bøhm, B. and Padfield, T., (2011). `Ultra-low-energy museum storage', In: Bridgland (ed) ICOM Committee for Conservation, 16th Triennial Conference, Lisbon 19-23 September 2011, 8 pp. ISBN 978-989-97522-0-7 (CD only)
10. Padfield, T. and Jensen, L.A. (2011). `Humidity buffering by absorbent materials'. In: Proceedings of the Nordic Symposium on Building Physics, Tampere, Finland, 29 May - 2 June 2011, 475-482
11. Sebera, D. (1994). `Isoperms: An Environmental Management Tool'. The Commission on Preservation and Access, 1400 16th Street, NW, Suite 740 Washington, DC 20036-2217.
12. Padfield, T. (2004). `The isoperm and the isoburn'.
Poul Klenz Larsen has a masters degree (1990) and a PhD (1998) in building physics from the Technical University of Denmark. He has worked for many years as a consultant for historic buildings. He is a senior consultant at the National Museum of Denmark, and teaches at the School of Conservation in Copenhagen and at the University of Gotland in Sweden. His main interests are energy efficient heating and climate control, salt and moisture in historic masonry structures, traditional mortars for restoration.
Morten Ryhl-Svendsen has a masters degree (2001) and a PhD (2007) in conservation from the School of Conservation at the Royal Danish Academy of Fine Arts. Since 1997 he has worked at the National Museum of Denmark. He is now a senior researcher in preventive conservation, with a special interest in air quality and pollution.
Lars Aasbjerg Jensen received a MSc in Conservation from the Danish School of Conservation in 2003. He worked as a sculpture conservator at the Ny Carlsberg Glyptotek, Copenhagen. He has been with the preventive conservation group at the National Museum of Denmark since 2004.
Benny Bøhm received the M.Sc (1973), PhD (1977) and D.Sc (1999) degrees from the Technical University of Denmark (DTU). He was Nordic Professor of District Heating Systems 1988-2003, working for the Nordic Council of Ministers at the Department of Energy Engineering, DTU. He joined the National Museum of Denmark in 2009. He has written over 180 technical articles and reports on fire safety science, solar and district heating systems and lately on indoor climate and energy consumption in museum stores.
Tim Padfield has a MA in Chemistry from Oxford University, and a PhD in building physics from the Technical University of Denmark. He started as a conservation Scientist at the Victoria and Albert Museum (UK), then moved to Leeds University, Earth Sciences Department. He later worked at the Smithsonian Institution (USA) and then at the National Museum of Denmark. He is now a freelance consultant in preventive conservation.
Here is the full article as a pdf file (1.2Mb)
This article has been submitted in German to Restauro Journal.
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.