Sarah Spafford-Ricci and Fiona Graham
Abstract
On the evening of February 16th, 1990, a fire was ignited in a new Native Indian gallery under construction within the Royal Saskatchewan Museum (provincial museum of Saskatchewan, Canada). The fire burned for an hour before it was extinguished. An unusually large amount of soot was unleashed onto and into every part of the three-story museum and its contents. Recovery from this ‘sooty’ fire presented an ideal opportunity for conservators to characterize soot and determine the best methods for its removal from cultural material.
A commercial fire restoration company was contracted to clean the building, furnishings and administrative contents. This process and the on-site cleaning of museum objects by conservators were undertaken concurrently. Careful organization and documentation proved essential to this cooperative effort. Conservators realized that a notation of the priorities, techniques and materials used by such cleaners should be laid out in a disaster plan, detailed at the point of bid for museum approval and then documented during the recovery.
Samples of soot from the interior surfaces of the museum were analyzed by the Canadian Conservation Institute Soot removal was tested by museum conservators prior to the full recovery stage. The materials and techniques required for soot removal were found to be remarkably different from those commonly used by conservators to remove dirt and accretions. Over eight months of cleaning myriad materials, museum conservators developed a strict procedure of progressive cleaning for soot removal that involved the use of direct vacuum suction, followed by dry cleaning methods and then, when appropriate, wet cleaning. In particular, the use of vacuum and dry cleaning appeared to minimize the break-up of soot agglomerations thereby minimizing dispersion and penetration of soot into an object.
Museum conservators found that, regardless of the surface being cleaned, the highest degree of soot removal was possible only if the soot layer was left undisturbed prior to application of vacuum suction. Pressure exerted on a soot layer through handling, temporary storage or non-progressive cleaning would cause a break-up of soot agglomerations and penetration of soot into the surface of an object. As well, soot was more difficult to remove as time went on. As could be anticipated, the degree of soot removal varied by the type of surface being cleaned.
Dry cleaning materials that proved to be most appropriate at lifting and holding soot particles were: Chem-Sponges, Groom/stick, eraser powder and absorbent disposable Webril wipes. Receiving some use were: soft goat-hair (hake) brushes used in a lifting motion, block erasers such as art-gum, kneaded erasers and tac cloths. Wet cleaning was most successful when minute soot particles were ‘lifted’ rather than solubilized. Aqueous surfactant/detergent solutions (e.g. Orvus/water and Aerosol OT/water) and alkaline solutions (e.g. ammoniated water) tended to be more successful than the use of organic solvents. A notable exception was the removal of heavy soot layers from bird and animal mounts where organic solvents were required. Cleaning of soot-covered bird mounts was done with a specially designed vacuum wand, application of a Webril wipe pad and Groom/stick molecular trap, followed by a swabbing of 1% Vulpex in trichloroethylene/rinse and cleaning of mammals was accomplished with the use of a vacuum crevice tool followed by swabbing with ethanol or trichloroethylene.