Many thanks to Jean Brown, Programme leader Preventive Conservation at the University of Northumbria, Newcastle, UK, for her generous support and encouragement in general and with regard to the study, which greatly benefited from her comments and suggestions; to the 96 institutions participating in the survey 2009; to the immensely supportive staff of the Kupferstichkabinett, Berlin, in particular to Prof. Irene Brückle and Ulf Schade; to the 15 institutions participating in the preliminary survey in March 2008, via email or by welcoming me into their museums, and to the suppliers who provided me with a wealth of information.
1Collections of oversized art on paper are growing and at the same time have to fit ever scarcer storage space and budgets. As they are extremely sensitive due to size1, material and production (Fiorani et al. 2006), only limited display periods are allowed, and the rate at which they deteriorate greatly depends on the conditions in storage and the quality of storage furniture (Ogden 2001). However, the opinion on what constitutes acceptable/optimal storage furniture is changing rapidly, and theory and practice often drift apart. Preliminary research, carried out at the Kupferstichkabinett, Museum of Prints and Drawings Berlin (Germany), on flat oversize storage currently in use by twenty museums in Europe and the US (Hammer 2008) revealed significant dissatisfaction with various construction features of oversize storage, staff health and safety and size and space required for handling. Many conservators, art handlers and collection managers would be happy to exchange their furniture into something ‘more appropriate’.
- 2 Pers. comm., Debra Evans, Fine Arts Museum of San Francisco, 19/03/2008 and 05/06/2009
2The planning for oversize storage should begin with a collection’s survey to understand how objects are used within an institution. Facini (Facini 2005) conducted such a survey. According to Debra Evans2, who was involved in this project, surveying museums with oversize storage is still a much needed task. Indeed, several museum professionals who took part in last year’s preliminary survey encouraged me to pursue this research.
3Safety to collections and user satisfaction are thecommon criteria to assess furniture material and construction features. As preventive conservation of collections is inherently sustainable, the approach of choosing furniture and support materials with ‘long life, loose fit, low energy, (LL, LF, LE)’ already advocated thirty-eight years ago by the RIBA President Alex Gordon (Gordon 1971), should also be of high importance. When making this the guiding principle for material/product selection one can significantly reduce the environmental impact related to the product, i.e. act environmentally sustainable. Overall sustainability also includes an economic and social component.
4Unfortunately, most manufacturers provide no evidence on the sustainability of their products. Thus, the furniture on offer should ideally be subjected to an independent evaluation. An understanding of the basic guidelines for material specification on the part of the purchaser would also be helpful3.
5Life cycle analysis (LCA) offers a number of tools to assess the environmental impact of a product with varying degrees of accuracy (SEI 2006). Currently, the most suitable method is the eco-indicator method, where an index value, called the eco-point, is assigned to a material or process that summarises the total environmental impact, including damages to ecosystems, human health and resources (Mark Goedkoop of PRe consultants, www.pre.nl). The various environmental impacts are measured individually, allowing the identification of the stage of the life cycle with the greatest impact. The stages extend from raw-material extraction through processing & manufacture, transport, scarcity and toxicity of the material; to product use, maintenance, repair and replacement; and to recyclability and end-of-life considerations, i.e. disposal (Abeyasekera and Matthews 2006).
6This study is a comprehensive approach combining a literature review, qualitative and quantitative analysis of an extensive user survey, supplier information and sustainability analysis to present a holistic view on the advantages and disadvantages of oversize furniture. It aims to provide guidance, in form of a checklist, to museum professionals regarding its selection.
7In the present study, the LCA/eco-indicator analysis is applied to furniture materials generally recommended for the highest protection of works on paper, i.e. to powder coated as well as stainless steel and aluminium (AIC 1994, Ogden 2001). Calculations are also provided for wood and its lacquers, coatings and linings plus handling supports – including their respective manufacturing processes. The approach chosen here comprises the following steps: First, a list of a product’s components is compiled, materials and processes are identified, and a weight (in kg) is assigned to each component for which information was gathered from suppliers and the literature. Each element is then added to the correct section on a table, i.e. the ‘life cycle inventory’, quantified in relevant units (raw materials in kg, electricity in kWh, shipping in tonkm etc) and complemented by its specific eco-indicator (EI) value, as assembled on www.pre.nl. Then the weight of each of the product elements is multiplied by its eco-indicator value to give its final score, the eco-point. The higher the point, the worse the environmental impact of that element. The total number of eco-points can then be calculated for each life-cycle stage. Finally, one can consider which materials cause the worst impact and should possibly be excluded as a choice. Calculations and recommendations are also made on sustainable furniture construction features and environmentally efficient forms of distribution.
- 4 Aluminium recycling would save up to 90% of the energy required for primary production, reducing t (...)
8The results of the EI calculations for various potential storage furniture and support materials are summarised in ill. 1. For steel and aluminium, the materials recommended for the highest protection of works of art on paper, it shows that the EI of steel is 3 times lower than that for aluminium. The energy required for the production of steel amounts to 23 MJ/kg, about 8 times less than the energy used in the production of aluminium. Moreover in comparison to aluminium no raw material is wasted in the manufacture of steel - recycling is an integral part of the process, allowing the alloying elements to be easily recovered and returned to the production process.4 In order to reduce the energy demand (and consequently the eco-indicator) manufacturers of steel furniture may purchase (domestic) carbon steel produced with up to 90% recycled material (e.g. designtograce.co.uk using 60% recycled steel).
Fig. 1 Eco indicator 99 for various potential storage furniture and support materials
9Besides its advantages on conservation grounds, powder coating of steel should be favoured over chroming (stainless steel) as its environmental load is much lower (ill.1, SEI 2006). For powder coating 97% of the paint ends up on the product, which minimises waste. No primer is required and metallic ‘burrs’ are smoothed, eliminating thus the processes (and energy costs) of grinding or filing. On top the coat is easily removed for steel recycling (for more sustainable features of powder coating and innovative coatings see Hammer 2009 and PPCJ 2008-1)
10Ill. 1 shows that the environmental impact of most wooden products is small (all calculated for European wood from sustainable forestry, i.e. Forest Stewardship Council (FSC) certified), which is due to the easy acquisition and procession of raw material. Nevertheless, 20% recycled steel (355 mPt) scores as well as particle board with low formaldehyde content (own data backed up with recent literature data, Bovea and Vidal 2004), 100% recycled aluminium and steel perform even better.
- 5 Some recently developed lacquers/sealants combine low environmental impact and highest protection (...)
11As far as actual usage is concerned, most wooden furniture is emitting formaldehyde and other VOCs, some of them still unknown (Oikawa et al. 2006). Furthermore, the energy demands and monetary costs related to wooden products in museum use will typically accumulate over the product’s life cycle due to maintenance measures. Coatings and recoatings with liquid paint are necessary5, and additionally the application of absorbents to ensure the near-zero-VOC concentrations required for the highest protection of art works (see below, Ogden 2001; Abeyasekera and Matthews 2006 for UK manufacturers); the same applies to lining and relining. The need for regular emission sampling is associated with relatively high personnel expenses; and in the worst case, the furniture may need to be replaced. All of this is likely to turn wooden furniture into an expensive and unsustainable piece (Ulmann 2004).
12As a means of lining wooden furniture archival board should be favoured over any of the plastic products (ill.1). The same applies to handling support sheets. This notwithstanding, in comparison to other support boards the EI is acceptable for e.g. Fome-core or Gatorfoam (Polystyrene); and due to their light weight, the eco points of these boards are reduced even further. Its paper honeycomb cells make the Tycore Honeycomb Mounting Panel another board with acceptable EI values (for details see Hammer 2009). Hexalite Lightweight boards, with their aluminium honeycomb cores and wide mix of materials, are less sustainable and recommended only if very high stiffness is essential.
- 6 If reuse is not possible, remanufacturing, recycling or incinerating should be performed in this o (...)
13Sustainability is not only about material, but also about construction features. To increase overall sustainability products should be favoured which are (1) durable and dimensionally stable, i.e. withstanding both intended use and possible misuse = optimised technical life cycle; (2) attractive not only for the short term = optimised aesthetic life cycle (which should ideally coincide with technical LC); (3) designed for economically viable service and repair; (4) flexible, adaptable, upgradeable, allowing refurbishment and reuse6; (5) easy to (dis)assemble, i.e. having as low as possible a number of components, allowing the use of simple tools, coming as a series of easily accessible modules, and being easy to transport; (6) independent, i.e. they should possess own lateral stability and not depend on other construction elements such as internal walls, in order to allow repositioning and/or refurbishment and; (7) compact rather than fixed storage; (8) low in material input, provided functionality is not compromised.
14Moreover, a product should give as much satisfaction as possible to the user. If not, it will be unsuccessful on the market and (economically, environmentally) unsustainable (Ljungberg 2008).
15Transport considerations are crucial for sustainability. The (environmental) efficiency of transport includes packaging, logistics and again structural considerations. Ideally the product should come in modules, allowing to fit a number of different products inside one truck for an optimisation of delivery schedules. Product weight should be reduced, especially if the transport is over long distances. This saves on fuel and lowers the EI. As a further requirement, packaging should be minimised and, wherever possible, draw on reusable material.
16The eco-indicator can be employed as a measure to estimate the impact of transport on the environment. It is then calculated in tonkm, i.e. the payload in tons multiplied with the distance that it is carried in km. The eco-indicators for a range of transport options are reported in the Design Manual (Goedkoop and Spriensma 1999), and the following examples are calculated from it:
17If a powder coated steel plan chest (100 kg, 20% recycled) is transported from Shanghai to Berlin (8.433 km) by intercontinental airfreight it attains a total LC eco-point of 78. Its environmental impact is then dominated by transport (86% of the sum of production, transport and disposal/recycling, ill.2). Transporting the same plan chest from a local supplier by rail gives an EI of 11 (by delivery van (< 3.5t) 17, by truck (28t) 12), which on average is 7 times less (ill.2). Here the steel production dominates the LC with 74%.
Fig. 2 Eco points for a steel plan chest – Local (500 km by rail) vs. Intercontinental transport (8433 km by intercontinental flight, Shanghai-Berlin)
18This should give an idea of how much the life cycle costs are increased / dominated by air transport over large distances. In principle, other modes of long-distance transport, e.g. by sea, are more environmentally sound. Nevertheless, the example illustrates the importance that should be given to the mode of delivery when it comes to sustainability.
19To summarise the sustainability analysis, the eco-indicator results show that in terms of their production side impact, FSC wooden products are preferable over primary metals but fare equally to recycled metal. However, as wooden furniture requires further treatment and attention the life-cycle impact will be considerably higher. If one adds to this the potential for wood to do damage to collections, (recycled) metal should be clearly favoured. Amongst the different metals, powder coated steel ranks over stainless steel, which in turn ranks over aluminium. If furniture from recycled metals is of archival quality then recycled powder coated steel is the preferred choice. Finally, the analysis showed that the mode and distance of delivery are as important for sustainability as the choice of furniture material.
20Several of the sustainable construction features correspond to the functional properties generally recommended for the highest protection of works on paper/safe handling (e.g. Ogden 1999). Again, this illustrates the sustainability concept underlying preventive conservation. For example, it is one key strategy towards sustainability to choose products with optimised functionality. However, this requires a thorough analysis of the user’s needs, which is precisely one main intention of the current study (see below).
21The survey “Archival quality furniture for oversized works of art on paper” was prepared and posted at www.umfrageonline.com. Questions were set up in the styles: (1) yes/no, (2) open, i.e. text box, (3) closed, i.e. single selection with optional text box or multiple selection, (4) table/matrix, i.e. evaluation table and (5) polarisation profile. The questionnaire link was sent via email to specialist staff of 162 institutions internationally, of which 96 institutions took part in the survey (= 60% response rate). Survey participants were requested to answer 31 questions about (1) their collection and conditions in storage; (2) the (main) method of oversize storage and, if plan chest usage, on construction features, e.g. drawer depth, types of drawer extractions, safety brakes, microclimate and dust prevention in drawers, usage of folders etc. in drawers, dimensions, manufacturer; and (3) furniture materials. Satisfaction was evaluated using a polarisation profile (scores from 1 to 5) on general satisfaction with the method currently in use, protection of objects, cost/performance, frequency of repair, and manufacturer’s service. Regarding plan chests, questions were on satisfaction with design/appearance, easy handling of objects, vibration free opening of drawers, quiet opening of drawers, firm drawer base.
22After a period of one month the survey was closed and the results were subjected to an in-depth statistical analysis. Percentages of distributions were calculated and correlations/ranking established by estimating the Spearman r (for nonparametric samples) or Pearson r (for parametric samples). Significance levels (P) were provided for all results. Variances were calculated and tested across more than two groups (ANOVA with significance analysis, Kruskal Wallis for nonparametric samples) and across two groups (t-test for nonparametric samples = Mann-Whitney test). User satisfaction scorings were developed for the specific parameters mentioned above). The results of the survey analysis were interpreted to derive guidance for oversize purchase.
23The characteristics of the oversize furniture in use in survey institutions correspond in many aspects to the ones recommended in the literature (e.g. Ogden 1999). These are the use of plan chests for best object protection (in 59% institutions), purchased from local/national suppliers (89%); the use of powder coated steel as furniture material (60%); availability of the top as work surface (72%); no sharp edges (76%); >80% drawer extraction (68%); safety brakes (71%); and storage of artworks in folders (53%).
24However, material emission tests are available for only 21% of the furniture, almost exclusively metal made. Test results are rarely available for wood, for which they are essential to ensure object protection. The drawer depth corresponds to the recommended 5cm for only 19% of institutions and drawers seemed to be filled too high. The space in front of the drawer rarely amounts to the recommended space of twice the drawer depth, making handling difficult. Telescopic guiding, scoring best in user satisfaction, is installed in only 19% of all plan chests.
25Generally, the interplay of object protection, user satisfaction and sustainability is well documented in the survey. Thus, for now the traditional flat storage of objects within folders in shallow drawers in powder coated steel plan chests scored best regarding all three points (incl. economic sustainability). The environmental sustainability is as good as that of low VOC particle board, for which however further maintenance costs accumulate due to (re-)coatings, sealants and lining, regular emission tests and the use of absorbents. According to many participants’ comments the maximum chest size offering best object protection and user satisfaction is approx. 2 x 2.5m (ill.3).
Fig. 3 A plan chest with high user satisfaction
- 7 Karl Buchberg, Pers. Comm. 14/03/08
Dimensions 2.4 x 1.5m, formats in excess of 2.4m are rolled7. Prints and Drawings Department MOMA New York
Credits : Karl Buchberg
26With these characteristics steel plan chests soon outcompeted recently introduced aluminium systems for extra-oversized items. The latter gave rise to concerns about the frequent need of repair due to the opening of several drawers at once, about dangerously sharp edges carrying the risk of injuries to staff and damages to objects (ill.4), sagging and sliding problems (documented in ill.5 and Ziwes 2007), and about the high costs. Moreover if produced from primary aluminium they are highly environmentally unsustainable.
Fig. 4 Sharp edges of a drawer
Credits : Astrid Hammer.
Fig. 5 Aluminium plan chests of 4 x 2.5m
Art Gallery of Ontario, Toronto
Credits : Melissa Potter
- 8 In the case study undertaken in 2008 at the Kupferstichkabinett Berlin (Hammer 2008), several supp (...)
27“Made-to-measure” plan chests scored better regarding user satisfaction and at the same time seem more sustainable than off-the-shelf chests. The higher initial costs8 are likely to be paid off as custom made plan chests receive better results regarding performance, repair frequency, service, design, handling and quietness of drawers (*, p<0.05). This confirms insights from the literature (Facini 2005) according to which customised furniture clearly provides a high(er) level of protection for problematic objects, because housings are produced specifically for individual preservation needs. They do not loose their attractiveness/user satisfaction with age as compared to off-the-shelf ones, i.e. the aesthetical life cycle is also optimised. Moreover, their owners seemed more involved in the process of ordering and better informed, which provides the scope and incentive to optimally use and care for the furniture leading to a higher (technical) life time. As one would expect, the dimensions of customised furniture generate greater satisfaction, especially width and (work top) height, which eases handling.
28The disassembly of made-to-measure chests is possible in most cases, an advantage that was initially only assumed for off-the-shelf plan chests. Hence, flexibility is still maintained and reusing parts or the whole product by the same institution or by others is possible. Of course this is not true for all custom-built furniture; some furniture is built to fit in a specific place, reordering of parts might take longer and relocation may turn out to be difficult. According to the LCA, this would be unsustainable, in particular as nowadays reorganisation and relocation have become much more likely events. However, from this survey we may conclude that custom built furniture is increasingly disassemblable so that the sustainability aspect must no longer be traded off against its advantages.
- 9 As the LCA showed, the mode and distance of delivery are as important for sustainability as the fu (...)
29Further, user satisfaction was highest if the manufacturer is local or national9, if emission tests are available for the furniture material, if plan chests are disassemblable, with worktop height, 200cm width and 125cm depth, and with a space in front slightly wider than the drawers’ width and almost double the depth of the plan chest. High scores were also awarded for a drawer depth ≤ 6cm and a (telescopic) drawer extraction of at least 80%; and for drawers without dust brushes and punched bottoms, with safety brakes and no sharp edges. Availability of supports such as tables or cardboards was also appreciated.
30From these results on satisfaction one may conclude that in the process of purchasing new furniture staff should discuss issues with everyone who will use the furniture and make themselves familiar with the different options. This improves the selection and increases the life time and overall sustainability of the purchased furniture.
31Suppliers should be contacted for emission test results, or better, a third party (e.g. a research institute) to perform a test. To reduce overloading of drawers one may install cost efficient shelves for the storage of less fragile works in archival quality folders and boxes. In the long term this increases the life time both of the plan chests and, more importantly, of the objects. Initial shelving costs are therefore soon paid off. As a space saving measure, plan chests should be acquired/fitted with rolls if possible; sturdy archival quality supports may to some degree compensate for the lack of space (Facini and Lussier 2003). Alternatively objects are stored hanging in frames10 or rolled11, if extra large. For more efficient space usage both are sometimes integrated in a rolling rack system made to measure12.
32Dust brushes and punched drawer bottoms are widely used yet seem unnecessary in the light of the survey results. Drawers without dust brushes (and no big gaps between drawers) might be sufficient or even preferred, as the likelihood of microclimate build up is reduced. Drawers without punched bottoms help to reduce potential object damage.
- 13 Greater detail is contained in the thesis (Hammer 2009)
33On the basis of the survey results presented above13, a simple checklist was developed, containing the most important criteria to guide the purchase of oversize furniture (ill.6). It was already applied to the case study ‘Re-housing oversize works on paper in the Kupferstichkabinett Berlin’ (Hammer 2008). As it has emerged from this case study, the selection of an appropriate piece of (oversize) furniture is a time consuming process for any institution. First, the desired criteria need to be established (and possibly ranked in their importance). Second, suppliers need to be identified and relevant information collected. Third, suppliers need to be compared in regard to the criteria, a process which usually involves a lot of trade-offs. Evidently, it is difficult for museum staff to reconcile such an intensive selection process with their day-to-day commitments. Here, any tool that helps to structure and streamline the selection process would be of high value.
Fig. 6 A simple checklist to ease oversize purchase
34In conclusion, oversize furniture should be selected with a holistic perspective, trying in the best possible way to reconcile object protection in terms of construction details and material and the product’s environmental, economic and financial sustainability. As a holistic perspective takes (or should take) due account of financial and other constraints, the solution is likely to be second-best from a purely conservatory point of view. Nevertheless, preventive conservation and sustainability principles should inform even the second-best choices, eventually leading to a process of gradual improvements.
35A convincing system for extra oversized flat storage, being light-weight and durable, remains to be developed. If this should be aluminium furniture, then it would need to be integrated as much as possible into the recycling loop in order to be sustainable. For the moment powder coated steel is the preferred storage furniture material in all aspects, including sustainability (even with only 20% recycling) and, to name just the most important features, built made to measure by local companies, being 200cm width and 125cm depth, with worktop height, and drawers ≤ 6cm deep.