Weighing up silver objects: evaluating past and future conservation methods

Simon Metcalf
Metalwork Conservator, Metalwork Conservation

Figure 1. Detail of a Sheffield plate candlestick. Worn and degraded lacquer has caused characteristic preferential corrosion. Museum no. M.11-1912. Photography by V

Figure 1. Detail of a Sheffield plate candlestick. Museum no. M.11-1912 (click image for larger version)

Over the past three years Metalwork Conservation has surveyed and conserved more than 1500 silver based objects in preparation for the re-opening of the Victoria & Albert Museum's Silver Gallery. This work has enabled us to build up a good practical knowledge of a silver object's display and storage life. We have identified the shortcomings of the protective lacquer currently used and have begun to research and develop a broader range of approaches to conserve silver, including a field trial of pollutant adsorbents.

The main problems for silver and silvergilt are the formation of corrosion (tarnish), and the surface wear associated with its removal. Corrosion is caused by reaction of the silver with gases in the atmosphere (a useful synopsis of literature can be found in the review undertaken by Inaba 1  ). The main causes of corrosion are thought to be hydrogen sulphide, carbonyl sulphide and, to a lesser extent, chlorides. These gases derive mostly from organic materials and in concentrations as low as a few parts per trillion, can cause corrosion 2 .

Corrosion is removed by cleaning, followed by the application of a cellulose nitrate lacquer 3  FrigileneTM (W Canning Ltd). This forms a barrier between the atmosphere and the surface of the silver. There are many advantages to this barrier approach - not least being the considerable experience in the utilisation of the technique. The treatment has the following limitations:

1. Pieces with a complicated construction of silver and organic/mixed media are sometimes not suitable for lacquering;
2. The successful application of lacquer requires a high level of skill and training and can take up to 50% of the treatment time;
3. Loss of lacquer from handling or uneven lacquer application can cause exposed areas of silver to corrode. The corrosion  leads to damage which shows as disfiguring, matted or etched areas on normally brightly polished surfaces (Figure 1);
4. Over time cellulose nitrate lacquer deteriorates in the presence of UV-light, leading to discolouration and reduced protection;
5. During application, lacquer coatings evolve solvents harmful to health (mainly xylene and n-butyl alcohol in the case of Frigilene). A Conservation Department health and safety review in 1995 highlighted the need to minimise staff exposure to these solvents. All work has been restricted to the three fume extraction cupboards in Metalwork Conservation's studio. This has reduced the number of staff able to work on lacquering and incurred extra transportation and handling  costs for silver from out-station storage.

  So why are we still using cellulose nitrate lacquers? The conclusion drawn from evaluating the collection's condition is that the above disadvantages are outweighed by the benefits of this type of coating:

1. Unlacquered objects have been found to corrode in three to twelve months, compare to between five and ten years for lacquered objects. Surface wear from frequent removal of corrosion is therefore drastically reduced;
2. Frigilene has been used for over 26 years at the V&A, so its properties are well known;
3. Trials comparing Frigilene with other products such as ParaloidTM B72 (Rhom and Haas) and CovolacTM (W Canning Ltd) have not found any major differences. They have either proved harder to apply or have failed materials testing  and still contain solvents harmful to health;
4. Though less stable than acrylic coatings Frigilene is easier to apply, and if done carefully does not interfere with an object's appearance.

Access and handling of objects will always be part of a museum's function. Conservation does not have the resources to monitor and control the affects of access to an object's unprotected  surface or the case environment. The Metalwork, Silver and Jewellery Department collection alone comprises over 12,000 silver objects so that many objects are not attended to for long periods. A protective coating supported by a range of complementary techniques is therefore a necessary overall strategy.

In the future, we hope to further develop and research practical methods to broaden our options. A number of measures, such as informing staff of the vulnerability of lacquer and to use regularly changed clean gloves, can help minimise degradation. Techniques such as separating pieces made completely of silver from silver objects with organic elements can also lower local pollutant emissions, as can the improvement of display and storage materials. A monitoring and maintenance regime, which includes the trial of pollutant adsorbents, will also help to prolong the life of treatments.

 Zinc oxide (ICI Katalco) and charcoal cloth (Charcoal Cloth International Ltd) have been identified in laboratory tests by the British Museum as being useful in trapping pollutants which are present in the case environment  5 . To complement this work we have begun a collaborative experiment with Dr S. Watts of Oxford Brookes University and Dr N. Blades, University of East Anglia (formerly of V&A Science Group). The aim, over one year, is to assess what causes corrosion and to see how adsorbents perform in a practical situation.

Figure 2. N. Blades (formerly V

Figure 2. N Blades (formerly V&A Science Group) placing pollutant monitoring tubes in a metalwork storage case. Photography by Simon Metcalf (click image for larger version)

Six storage cases used to house silver objects have been selected; three old mahogany cases  and three modern stove enamelled steel cases. Each set of three has  a control case, a case testing zinc oxide and a case testing charcoal cloth. Diffusion tubes 6   (Figure 2) supplied by Oxford Brookes University are being used to measure levels of  hydrogen sulphide, organic acids (principally ascetic and formic acids) and ozone. Additionally, levels of sulphur dioxide, nitrogen dioxide, and chlorides are being measured using a multi-purpose tube supplied by Gradco International Ltd. Four accurately weighed silver coupons (J. Blundell & sons), two 99.99% pure, and two 92.5% sterling have been placed in each case. At any point during the experiment, the coupons indicate the amount of corrosion formed. At the end of the experiment, they will be analysed to identify the types of corrosion present. This is essential since we cannot be sure we are monitoring all the pollutants that cause corrosion and, for some (e.g. carbonyl sulphide), there is as yet no diffusion tube method available 7  . 

The use of adsorbents may prove useful in protecting silver objects from pollutants, though at present it is unlikely to replace lacquer coatings in such a large collection. The last three year's work has highlighted the successes and limitations of a long established treatment regime as well as generating new approaches to silver conservation. For the future, Metalwork Conservation Section needs to continue to develop and research an ever broader spectrum of techniques for protecting the V&A's silver collection.

References

1. Inaba, M., Tarnishing of silver: a short review, V&A Conservation Journal 18, 1996, pp9-10
2. Blades, N.W., Investigation of the use of absorbents to prevent the tarnish of silver in display and storage cases, V&A Science Group Internal Report 96/31/NB, 1996
3. Heath, D., The Conservation of the Portuguese Centrepiece, V&A Conservation Journal 17, 1995, pp 4-7
4. Pretzel, B., Corrosiveness test results, V&A Science Group Internal Report, 1991
5. Lee., L., (Senior Scientific Officer, British Museum Conservation Department), Personal Communication.
6. Blades, N.W., Measuring pollution in the museum environment, V&A Conservation Jour nal 14, 1995, pp 9-11.
7. Blades, N.W., (Senior Research Associate, School of Environmental Science, University of  East Anglia) Personal Communication, 1996