Conservation Journal

Autumn 2002 Issue 42

Continued studies in the deterioration of glass

Sarah Fearn
Research Associate, Department of Materials, Imperial College, London

It has been well documented over many years that certain types of glassware displayed within the glass collection at the V&A are susceptible to deterioration over time. Typically the group of artefacts particularly prone to this process is Venetian glassware of the 16th and 17th centuries. It has been suggested that the instability of glass from this period is due to the purification of the raw products, such as plant ashes, for obtaining ‘a more clear and crystal glass’ ¹ . Refining the raw materials produced a clear glass that could be worked at lower temperatures, but the composition of the glass had been altered. The purification process led to a glass with a low calcium oxide (lime) content and a high sodium oxide (soda) or potassium oxide (potash) content.

Figure 1. An electron image of a severely crizzled surface on a Venetian goblet. Photography by Sarah Fearn (click image for larger version)

In all glasses, sodium and potassium oxides are hygroscopic and therefore attract water. The surface of the glass therefore, absorbs moisture from the air. It is this interaction between the glass surface and atmospheric water that starts the deterioration of the glass. This deterioration manifests itself in a number of ways. In the early stages a glass object may initially acquire a dull foggy appearance. Under humid conditions, droplets of moisture appear on the surface, as the hygroscopic alkali salts delliquesce ² , a condition referred to as ‘weeping’.

As the deterioration progresses a series of fine microcracks start to become visible. This stage is known as ‘crizzling’ and can eventually lead to the formation of flakes and pits on the surface; an example of this is shown in the electron micrograph of Figure 1. With continued leaching of the alkalis to the surface of the glass, it becomes increasingly alkaline. Once the alkalinity of the surface reaches a pH of 9 and above, dissolution of the strong silica network occurs and the glass object will lose much of its mechanical strength and collapse.

In addition to degrading the visual appearance of the glass artefact, the corrosion process has some more subtle effects. As a result of the changes at the surface of the glass due to the corrosion process the surface composition is not representative of the bulk composition of the glass.

Much work has been carried out into the corrosion and deterioration of glass in an effort to identify what environmental conditions should be applied for the storage and display of these vulnerable glasses. In 1959 workers at the British Museum recommended that unstable glasses be kept at relative humidities below a value of 42% ³ . It was reasoned that at relative humidities below 42% potassium is not leached out, and remains in situ. However more extensive studies by Brill, based at the Corning Museum of Glass, New York, suggested that lower humidity values could be just as problematic as the glass surfaces become dehydrated. Brill suggested therefore that a range of 40-60% relative humidity was a safe range for storing glass objects ⁴ . From these two examples alone it can be seen that the conditions for storing glass has not been clarified, and poses a problem for curators and conservators wanting to display vulnerable objects without compromising their longevity.

With the problem of displaying glass artefacts briefly outlined above in mind, the V&A, in collaboration with the Materials Department of Imperial College, instigated a series of research programmes concerning the conservation of glass. Through this collaboration, the V&A is able to access surface-analytical techniques which would not normally be available to the Museum’s Conservation Department. It also has the advantage that studies dedicated to one problem over a period of time can be carried out.

The first such collaboration concluded in 1996 when Ryan completed a three-year PhD study in to the atmospheric deterioration of glass ⁵ . In this study replica materials, of a low lime high soda or potash content, were fabricated and aged under a variety of temperatures and relative humidities. These samples were then analysed using a surface analysis technique called SIMS (secondary ion mass spectrometry). This technique enables the researcher to follow the diffusion of elements through a material. With relation to the corrosion of glass, therefore, it is possible to detect very small changes in composition at the surface of the glass in relation to the bulk material. This enabled a more accurate assessment of the appropriate relative humidity at which to store and display unstable glass and a value of 38% ± 3% was suggested. However, this research was limited to one group of glasses and the humidity value may not be pertinent for other glass compositions.

Further to this work a second collaborative study was carried out between Imperial College and the V&A Conservation Department to look into the possibility of active conservation procedures. In this case, Hogg investigated the possible application of silanes to the surface of the glass ⁶ . By applying a single atomic layer of molecules to the surface it is possible to turn the hygroscopic glass surface hydrophobic. This work was again carried out on replica materials and found to repel water. It was also noted that the surface treatment did not alter the appearance of the glass surface to the human eye, and remained smooth. The application of surface treatments does, however, pose an ethical dilemma in that these treatments are not reversible.

Recently, through gaining funding for a further three year project from the Leverhulme Trust, the research has resumed between the Materials Department at Imperial College and the V&A’s Ceramics and Glass Conservation Section. Since these collaborative projects first started there have been some major advances within the field of surface analysis. These will be applied to investigate more the effects of the environment on unstable materials, and clarify some of the mechanisms at work during the corrosion process. Another aim will also be to investigate further the potential use of surface treatments to halt the deterioration process that are acceptable to the conservator from an ethical and practical point of view. Finally, work will be conducted to identify any effects that cleaning may have on these objects.

If this work is successful it could hopefully extend the lifetime of many unstable historical glass artefacts for years to come.