3 CONSERVATION OF JUNKERNBODEN

THE SECOND PHASE of this project was to devise a means of removing the salt from the paper and returning the print to the healthiest possible state. Rinsing the sheet in water, an obvious way to remove sodium chloride, was precluded since the weakened paper could not sustain immersion: a small sample placed in deionized water fragmented immediately upon wetting. Therefore, it was necessary to immobolize the fibers so that they could be penetrated by water to dissolve the salt without themselves disintegrating. Several ways of doing this were considered: 1) facing the print with a porous material, 2) placing the print on a suction table, 3) consolidating it with a synthetic resin, and 4) sandwiching it between sheets of a porous material which would support the paper without adhering to the fibers. The idea of facing the print was abandoned immediately because tests proved that it would be impossible to remove the facing without displacing the surface of the print. For several reasons the use of a suction table was eliminated. At the time a suction table was not available. More importantly, the pulpy nature of the paper when wet made it certain that any irregularities or surface texture from the table would be firmly impressed in the paper.

Consolidation materials were considered at length, and a test was performed using soluble nylon on a piece of the margin. This consolidant was selected as the only material customarily used in conservation that is easily permeable to water and is not also water soluble.18 Also, the matte surface produced by this resin would not be uncharacteristic of the presumed original surface of the wood-block printing ink and paper. A 2% solution of soluble nylon in ethanol was applied to the sample with a brush.19 The paper was so absorbent that it was only necessary to touch the sample in one place to saturate it.

It was hoped that a consolidant would impart enough strength to the sheet to allow it to be bathed, and the soluble nylon did seem to strengthen the paper, though no quantitative tests were made to determine the degree of strengthening. However, the consolidated sample was not able to withstand the immersion in water which removal of the sodium chloride would require. The use of soluble nylon as the sole means of immobilizing the fibers during a water treatment was excluded, though the strength it lent to the sample gave promise for its use as an auxilliary treatment. Further support for the sheet, or else an alternative to immersion in water, was deemed necessary.

At this point, it seemed that sandwiching the print between some porous material was the only remaining possibility for support. The sandwich arrangement would provide a method for indirect handling of the sheet but would not immobilize the fibers as, in theory, a consolidant could. Therefore, it would be necessary to remove the salt without immersing the print in water.

A method employed by conservators to remove salt from pottery and stone was studied as an alternative. This procedure makes use of wet paper-pulp poultices or blotters to leach the salt from an object.20 In this case a porous interlayer surrounding the sheet would prevent loose paper fibers from adhering to the poultice or blotter. The material chosen for this purpose was a monofilament nylon woven cloth available in a variety of meshes.21 Since it is woven from single fibers (instead of multiple fibers twisted together to form a thread), it is not fuzzy and does not catch on materials with which it comes into contact. This, together with its inertness to the materials used in the project, made it suitable for use as an interlayer in treatment with poultices. For ease in description in the following paragraphs, this nylon cloth will be referred to by its trade name, Nitex.

Impregnation of the damaged paper with soluble nylon was still considered a possibility for imparting strength to the print. Therefore, tests were carried out on two samples from the margin of the print to determine the advantages of the procedures under consideration. One sample was impregnated with a 2% solution of soluble nylon, dropped on the paper surface with a pipette; the other was not consolidated. When the impregnated sample had dried, both samples were sandwiched between Nitex of 195 mesh, and wet blotters were placed on top. Paper-pulp poultices were not used because they were difficult to apply uniformly and because the flat print did not require material which would mold to irregular contours. It was expected that the dissolved sodium chloride would be pulled into the blotters by capillary action as the blotters dried, and testing the blotters for chlorides after they were removed confirmed this.22

When the blotters were dry, they were removed from the samples and replaced with wet ones. Blotter changes were repeated until, after the eighth change, the chloride test yielded virtually no reaction. The Nitex was easily peeled from the surfaces of the samples, and the papers were examined. In the consolidated sample the soluble nylon had not inhibited the removal of the sodium chloride, but it had not appreciably added to the security of the paper. Since treatment with blotters involved no manipulation of the print itself, the soluble nylon did not seem necessary, and it was decided not to apply any consolidant to the print.

The procedure finally chosen for the treatment of Junkernboden was to sandwich the print between Nitex sheets of 195 mesh and to remove the sodium chloride with wet blotters. Several factors other than those already mentioned were considered important. It was decided that the print should never be allowed to dry until as much sodium chloride as possible had been removed. Repeated recrystallization of the salt was presumed to have been responsible for at least some of the damage to the paper, so the formation of new crystals in the process of treatment was to be avoided. The use of three blotters at a time assured that the print would remain wet while the top blotter dried.

A serious concern was the possibility of mold growth during the lengthy period treatment would require. To prevent this, the blotters were painted with a 7% solution of thymol in ethanol, and the ethanol was allowed to evaporate.23 The printing ink was tested for solubility in water, ethanol, and thymol in ethanol, all with no reaction.

From the condition of the test samples it was known that the paper would be very weak after the sodium chloride was removed. It would be necessary to provide the sheet with additional support to prevent further damage. The traditional lining method using Japanese paper and starch paste was chosen because this method of supporting weakened paper most allows the sheet to retain its integrity. Overall mounting on a solid support would make the paper look unnaturally flat and would also prevent the sheet from expanding and contracting at its own rate. The plan was to line the print as soon as the sodium chloride was removed and before the paper had dried.

In the arrangement of the print during treatment, plexiglas was used as the bottom support to allow examination of the sheet during treatment. The print, which was sandwiched between Nitex sheets, was laid face down on the plexiglas, so that after the salt had been removed the print could be lined without manipulation. The underlying Nitex interposed a “cloudy” layer but did not obscure the image.

The procedure was exactly that of the test samples. After each change of blotters, which occurred twice a day, the used set was tested for chlorides. For consistency, it was decided that 1 sq. in. of each blotter would be soaked in 20 ml of deionized water for 15 min. The chloride test was performed with 10ml of the resulting solution. The extremely strong initial reaction changed to a moderate reaction by the sixth change of blotters, and to a weak reaction by the tenth. The procedure was continued through thirteen changes of blotters, by which time there was virtually no reaction at all. Ten days had elapsed since the treatment had been begun. It is interesting to note which of the three blotters absorbed the most sodium chloride. After the second change of blotters each of the blotters was tested separately; the middle one produced a considerably weaker reaction than the upper and lower ones.

At the very outset of treatment an unforeseen phenomenon occurred. Within 1� hour black stains appeared in the design areas. These stains initially seemed to be bleeding printing ink. Several were streaks and others were tidal rings. All were confined to the upper half of the image, and all were associated with inked areas. None changed appreciably during the remainder of the treatment. They were not reactive to tests with thymol crystals or salt water, so their cause remains unknown. On examination after the treatment was completed and the print was dried, it was decided that they are not of the same black hue nor the same texture as the printing ink.

When the salt removal was completed the Nitex was peeled from the verso. The print was lined first with a thin Japanese paper and then with a heavy one.24. The light-weight paper was used to form a good bond with the reverse of the damaged paper, while the heavier paper provided a strong support. The wheat starch paste was brushed on the lining papers and the linings were applied with hand pressure.25

The print was then turned over for the first time since treatment began and the Nitex was removed from the front. Again, an unforeseen development occurred. A pale gray counterproof of the ink image had been transferred from most areas of the design onto the Nitex. More critically, the brown stain in the lower left quadrant of the print, whose visual effect had diminished during the salt extraction process, now exhibited adhesive properties. Small portions of the surface of the paper in this area stuck to the Nitex no matter how carefully it was removed, resulting in some skinning of the sheet in the design area. When the Nitex was fully removed and dried, however, bits of detached paper and ink could be peeled from the Nitex and replaced on the print, which was still moist, with dilute wheat starch paste left over from the lining process. Virtually all of the inked areas were reattached in their original locations in this way, using a before-treatment photograph as a guide. A minimal amount of the surface ultimately was lost. Later, when the print had dried, the remaining brown stain seemed more like an adhesive layer than it had prior to treatment: it had become harder, and small areas were cracked and cupped.

The lined print was allowed to air-dry completely. It buckled severely, but the drying procedure did allow the fibers to re-mesh themselves without artificial constraint. To flatten the sheet it was sprayed lightly with deionized water on the reverse until it became limp and slightly expanded. The margins of the lining were then folded over the edge of a sheet of eight-ply ragboard and glued in place with PVA emulsion.26 The ragboard mount had been counterlined with Japanese paper on the verso, which was still wet and in an expanded condition when the lined print was mounted. The ragboard bowed severely away from the print as the counterlining dried. The adhered margins of the print lining were peeled off the mount, and the procedure was repeated on another piece of eight-ply ragboard which had not been counterlined. Planar irregularities were eliminated as the print dried, but at no time was the print taut across the board.

Finally, the print was matted so that it could be safely stored in the Fogg Print Department with no need for special handling. The window mat and backboard were cut from four-ply ragboard. The print on its rigid support was adhered to the backboard with acid-free photocorners and white cloth tape.27 A protective mat which would act as a separator for the window mat was cut from paper-faced expanded polystyrene board, the edges of which were sealed with pressure-sensitive tape.28 The polystyrene board was not in contact with the print itself. The rag window mat was adhered to the protective mat, which was hinged to the rag backboard. This package has the appearance of a regular matted print, but supplies the fragile work of art with greater support. It is now stored with the other Kirchner prints in the Print Department of the Fogg Art Museum.