THE CLEANING OF DAGUERREOTYPES: COMPARISON OF CLEANING METHODS
M. Susan Barger, S.V. Krishnaswamy, & R. Messier
MOST CLEANERS remove part of the original material from the object that is being cleaned and are, therefore, “damaging” to that object. The selectivity of any cleaning method is directly related to the ratio of sulfur-to-silver material removed and is thus inversely proportional to the amount of “damage” resulting from cleaning. Further, the regulation of the rate(s) of material(s) removal gives control over the amount of “damage” a cleaner might do to an object. Implicit in this is that methods offering higher selectivity and control are also more forgiving in that they increase the latitude between cleaning and irreparable damage to an object. For the daguerreian system, the ideal cleaner would remove only sulfur, or barring that, only silver sulfide, leaving the other daguerreotype constituents (Ag, Hg, Au, Cu) intact. It would also proceed at a rate that could be easily controlled.
The solvent cleaners, KCN and thiourea, tend to alter unselectively both the image particles and the daguerreotype surface by dissolving unwanted silver sulfide tarnish, as well as all other daguerreotype constituents. Further, changes in particle shape and distribution resulting from these cleaners suggest that the combination of the solvent and the daguerreotype constituents makes an active chemical system which causes the random removal and redeposition of materials possibly by a mechanism like autocatalytic plating. The action of solvent cleaners tends to be nonuniform, which may be due to uneven wetting of the sample surface by the solvent, and which is apparently reduced by the addition of a surfactant (e.g., Photo Flo) to the cleaner formula.16
In keeping with the general consensus since the adoption of thiourea cleaners, these results show KCN is an unacceptable daguerreotype cleaner. Even though thiourea shows a tendency towards the unacceptable traits of KCN, the results indicate that its use might be warranted in some cases, if used with extreme caution. The work reported by Rempel5 suggested that it is possible to control the rate and selectivity of thiourea cleaners by altering pH and thiourea concentration. Thus, if thiourea cleaners are used, it is recommended that the cleaner formulation include a surfactant and that the pH and thiourea concentration be kept low.
Sputter cleaning gives good retention of both microstructure and the gross optical characteristics of the daguerreotype. The microstructure shows a characteristic roughening of the substrate surface due to sputtering bombardment effects; however, this roughening is not of sufficient size to alter the optical scattering properties of the material and thus is not seen by the viewer. Potentially, it could also be controlled by refinement of the sputtering parameters. Clearly, the four-minute sputtering conditions were too drastic and resulted in a great deal of damage, which could possibly be avoided by an alternation of other sputtering parameters besides a reduction in sputtering time. Further, this technique, which is well understood and highly controllable, is dry and avoids all of the effects associated with the use of solvent solutions.
The results described above show the overall trends in the cleaning effects for the same material, i.e., the same daguerreotype. It should be kept in mind that these results show the effects of the solvent cleaners at the recommended formulations. The sputter cleaning results only look at changes in sputtering time, and there has been no other attempt to optimize this process. However, the sputtering results are very promising and appear to be generally less damaging than the results for the solvent cleaners. It is hoped that further research will provide an understanding of the effects of sputtering on daguerreotypes and will also lead to a workable system for cleaning daguerreotypes. Such a system should be based on (a) daguerreotype evaluation, (b) optimization of sputtering parameters, and (c) improved control over sputtering rate and end point determination.
These results do not consider the effects of repeated cleaning of daguerreotypes by any method, or of cleaning by one method followed by subsequent cleaning by another method. This is an important consideration especially since, over the years, there have been repeated recommendations that daguerreotypes be routinely cleaned whether such treatment was warranted or not.17 It stands to reason that if the image particles on a daguerreotype have been damaged or removed during some prior cleaning, that removal of subsequent tarnish may result in the apparent removal of the image if the new treatment reveals a severely altered microstructure. It is possible that the silver sulfide tarnish layer may act as a partial antireflection coating, depending upon viewing angle and sulfide thickness, and therefore could actually enhance an image that appears faded due to an altered microstructure. Further work is underway to devise methods for evaluating the condition of daguerreotypes in order to determine when treatment of any sort is advisable.