Past & Present Work

Project: The conservation of a strappato wall-painting on hessian
Client: Norfolk Museums Service, Castle Museum, Norwich, NR1 3JU.
Date: 2002
Undertaken by Ned Scharer

Description of object
The secco painting was detached from the interior of a house in Wymondham, Norfolk, in the 1960's. It was removed using the strappato technique but was never detached from its hessian, muslin and animal glue facing. Over the subsequent years the glue had cross-linked, causing discoloration, extreme buckling and embrittlement. It was critical to remove the facing and excessive glue from the painting underneath, in order to halt the loss of pigment shards.

The essential difficulty was the age of the glue and the fact that there was no other material within the object to exploit which would facilitate the removal of the facing. The standard procedure for the removal of strappato facing relies on the strappato medium either being different from that of the pigment binder, or on a separation layer having been painted over the wall-painting prior to treatment. Such paintings are usually re-backed soon after stripping while the adhesive is fresh. After forty years, another major complication was the shape of the object; it had distorted so badly that treatment would be impossible without first flattening it.

Extensive research and analysis gave a good understanding of the nature and composition of the object. To leave the piece in its existing state would have been irresponsible and would lead to its inevitable degradation. However, at this stage I was not sure if the client's objectives, re-backing of the painting and removal of the facing, would be achievable.

Overview of the conservation treatment
I developed various techniques that used steam to flatten the painting and remove the glue, hessian and muslin. I exploited the tight time differential between dampening the hessian and subsequently the muslin, before the lower levels of adhesive, incorporating pigment, became too soft. This maximised the hygroscopic nature of the animal glue as well as its low water permeability. Hand made Japanese paper (usimino) and AC33 (acrylic dispersion; methyl methacrylate ethyl acrylate co polymer) were used to form the primary backing.

For rigidity, I mounted the painting onto polyethylene foam, adhered to Aerolam (aluminium honeycomb, sandwiched between two layers of epoxy resin and fibreglass sheets) using paraloid B72 (acrylic resin; methyl acrylate ethyl acrylate co polymer). This innovation adds considerably to the reversibility of the mounting over other methods. The future removal of this backing would not necessitate the use of heat.

Lateral and intuitive steps
The project required a combination of logical and lateral thought processes to solve unique difficulties. Instinct and intuition were as vital as practical experience in finding an appropriate methodology. The results of each process governed subsequent stages of the conservation treatment

Analyses
Visual analyses under microscope
UV light
Polar light microscopy
Solvent spot tests

All investigations, trials and techniques were first tested on control samples and models of the object.

Flattening the painting
The difficulties in flattening the object were exaggerated by its thickness and brittle nature. It was necessary to soften the whole system throughout otherwise the object would crack. Tests showed that the glue was still readily soluble in warm de-ionised water. I decided that the only way to relax the object to a point of malleability was to use high relative humidity. This however had its risks, including potential destabilisation of the thin pigment layer. I built a permeable tray, using aluminium mesh, to the dimensions of the object which I placed face-down, allowing the steam to travel through the layers of facing, reaching the fine paint layer last. This technique allowed me to control the process, waiting for the exact point at which the painting was ready to be manipulated. This was a gentle technique that allowed the object to soften without force. Once the painting was in a flatter state, I kept it under weights for a period of two weeks, so that the glue would dry in its reformed state.

The removal of the glue, hessian and muslin
Before attempting to remove the facing, it was necessary to back the painting. Although the object was flat, it was still creased and folded and a rigid permanent backing at this point would have compromised the final outcome. I decided to use a backing that would facilitate manipulation after the removal of the 1960’s facings; the long-term benefits of this would outweigh the short-term difficulties. I used a fine Japanese tissue and AC33 (acrylic dispersion; methyl methacrylate ethyl acrylate co- polymer) adhesive that would dry transparently, in case the removal of the facing was not possible: a transparent backing would still allow the image to be partially visible from the reverse.

Due to the extent of cross polymerisation of the glue, an energetic process was required to break it down. Warm de-ionised water in practice was not efficient, as I needed a longer contact time. A gel increased the contact time but was found to cool too quickly. Trials showed the best technique for removing the hessian and dissolving the glue was direct localised steam, using a dental steamer with a nozzle. The steam swelled and dissolved the outer layer of the glue temporarily, creating a range of gelatinous consistencies that allowed for a controlled removal. With this simple technique I developed sufficient skill to remove the glue and hessian in a controlled manner.

The removal of the secondary layer of muslin was extremely problematic, separated as it was from the pigment by only a thin glue layer, which I hoped would remain as a barrier. Initially, I mechanically removed the muslin with a Laponite (synthetic colloidal clay) and de-ionised water gel but this was time consuming and indiscriminate. I decided to try the steam again as I felt that here I had developed a technique over which I had the most control. I applied the steam twice, the first time for about five seconds, which softened the glue, the second about a minute later immediately dissolved the glue, the muslin then peeling away easily, without harming the pigment layer.

With the fabric removed, the design of the painting was visible, but it was still obscured by the layer of glue and stubborn remnants of fabric. I felt confident enough to continue with the steam technique and under magnification applied short bursts of steam which removed the dissolved glue and water before it softened the paint layer. It was not possible to remove all the glue, as it was holding the pigment in place, however enough was removed to allow the image and its colours to show.

Backing the painting
The paper backing of long-fibre hand-made Japanese paper (usimino) proved ideal, it allowed the painting to be heated and the remaining creases to be ironed out. The AC33 also remained soluble in acetone and with care could be removed from the painting. It was now stable and ready to be mounted onto a rigid backing so that it could be stored, moved or displayed. Ciba Aerolam board was used for its lightness and high tensile strength. I felt that the secondary backing needed to be easily removable, because the final destination of the painting was not known. In order to remove the object from its secondary support without subjecting the painting to further heat treatment I developed a removal technique that only required small amounts of solvent.

Polyethelne foam coated in paraloid B72 (acrylic resin; methyl arcylate ethyl methacrylate co polymer) was adhered to the Aerolam and acts as a barrier between both the backing board and the lining of the painting, which was coated in AC33 acrylic dispersion. Although AC33 and Paraloid B72 are both acrylics that are soluble in organic solvents, the difference in the solubility times can be exploited to facilitate removal.

In Conclusion
The painting is now flat and the image revealed. The facing is removed leaving only the glue that is dispersed within the pigment layer. The options to the client for display are open as both backing supports are stable enough to remain or be removed without any further risk to the painting, unlike the heat treatment processes described recently in published literature.

Face of painting covered in hessian and glue

Reverse of painting

Completed conservation. click for larger image

Detail