Conservation of a tabletop from the Pushkin State Museum of Fine Arts


The tabletop of the second half of the 17th century (Florence, the Grand Duke's Workshop) until 1945 was in the collection of the Dresden State Art Collections (Green Vaults). In 1946 it entered the Pushkin State Museum of Fine Arts. The tabletop is a 121×77 cm stone base covered with Florentine mosaics (pietre dure). A similar tabletop is the Louvre collection.


An interesting detail about this tabletop is that it had never been restored and its surface had never been renewed, which was a common practice: to remove small scratches and abrasions, the surface of the Florentine mosaic would be ground and polished, which damaged the quality of the mosaic. In this case, the surface of the tabletop has survived to this day in its original form. But the object had other serious damages: all four corners of the tabletop were chipped, three of the four identical elements of the mosaic in the corners were lost. They were lapis lazuli hearts, fringed by yellow Siena marble. Fortunately, one heart has been preserved almost perfectly, separately from the tabletop. The traces of the lost elements were well read by the glue seams. A black oil shale renewal runs along the tabletop. One of the plates, about 25 cm long, broke off but was preserved, and there were visible glue traces at the chipped areas, as well as details of metal fasteners. This allowed us to better understand the technology of how the fasteners were mounted, as well as to check the thickness of the plates of the mosaic and its base. The biggest problem was the numerous peelings of the mosaic elements which remained in their historical places on the base. The adhesive mastic with which the mosaic elements were attached was almost completely destroyed, probably due to freezing. The mastic has lost its plasticity and adhesive ability.

Establishing the problem: danger to the object

Based on the amount of damage and the results of the examination of the object before the restoration, it became obvious that it could not go on existing in that condition. the safety of the mosaic was in danger. It was decided to restore the object. The examination was conducted visually using a binocular microscope. Tapping the tabletop made it clear that more than 85% of the mosaic was coming off. Because of the destruction of the mastic, there were voids under the mosaic. Based on the data obtained, a map of damages was made, which was later transferred at a scale of 1:1 to a sheet of thin transparent plastic.

Conservation methods

Based on the examination, the restoration technique was developed. It consisted of the following:

1) to compensate for the loss of the mosaic pieces;

2) to take measures to glue the identified cavities of the adhesive layer, thus strengthening the mosaic set.

The main task was to preserve the historical surface of the mosaic set.

Conservation process

Originally, the casts were taken from the areas of losses, so that based on them and on the preserved element it was possible to make the lost details of the mosaic: lazuli hearts. An aqueous solution of micro cellulose (MC) was used to protect the historical surface during the moulding process. The moulding itself was carried out using cold-curing sealant KLCE. Its features are shrinkability and, on the one hand, sufficient rigidity, on the other hand - sufficient plasticity, which allows for precise and delicate moulding of the details.

Plaster models of losses based on the preserved part (the heart) were cast into the resulting moulds.

The stones (lapis lazuli and old Siena marble) were selected, and the cut and polished stone plates were chosen for their colour. New elements of the mosaic were made based on the plaster models, but they could not be put in their original places before the main mosaic set was consolidated. The new elements were polished in advance, and their thickness was known to be slightly less than the thickness of the historic set. This was done in order to be able to precisely match the new and old elements in height using a glue base.

The main problem of the tabletop was the almost complete detachment of the mosaic set. Fortunately, all the elements of the mosaic remained in their historical places, and only the glue layer of the 17th century was destroyed. It was decided to pump glue into the formed cavities. But how to prevent the bulging and movement of mosaic elements in the process of injecting glue?

This was done in the following way. First, the map of glue layer loss made on a thin transparent plastic was transferred to the "conductor": a sheet of 5 mm thick plexiglass, exactly the same size as the tabletop. Then holes (3 mm in diameter) were drilled on the "conductor" at the points of supposed glue injection. The "conductor" through a multilayer buffer substrate, consisting of a protective layer of MC, lavsan film and transparent sealant T4, was attached to the tabletop surface. The holes in the "conductor" were arranged so that on the tabletop they coincided with monochrome black oil shale inserts, not with the stone so that the set would be less injured. After curing the T4 sealant under the "conductor", it was possible to safely drill the tabletop surface with the thinnest 1 mm diameter drill bit to the voids of the deconstructed adhesive layer. OK-15M glue was pumped through the holes in the cavities using medical syringes with thin needles. To fill the cavities with glue as much as possible and to get rid of air inside them, the tabletop was placed in a specially manufactured vacuum chamber. Vacuuming was carried out in stages. After full polymerization of the adhesive (36 hours), the "conductor" and all the protective "pie" were removed. The assumption that the large cavities would be filled in this way was confirmed. Now that the bulk of the mosaic set has been fixed, it remains to fill the small cavities. The seams between all elements of the set were thoroughly cleaned and dried and then filled with glue. The top was again placed in a vacuum chamber. During the step-by-step vacuuming process, the glue filled the small cavities. After curing the glue, inspection and tapping of the worktop showed that there were no more voids inside.

Only after the consolidation of the glue base, it was possible to install the newly made parts in the corners of the table. Further work on the restoration of the tabletop was to model and grind the fillings in the places of loss, seal the technological holes used to fill the glue in the voids, and apply protective wax.

Problems and solutions

Thus, during the restoration of the tabletop, specialists faced several problems, which were successfully solved by innovative methods:

- preservation of an intact surface of the mosaic set with no layer between it and the base;

- creation of support areas of the adhesive layer, which served as the base for the filling of small cavities;

- exact matching of the surfaces of the old and new pieces of the set (for perfect colour matching it was necessary to find Siena marble mined over two centuries ago).

Confirmation of technological solutions

New technological solutions for the restoration of Florentine mosaics were proposed by the conservator A. Androkhanov after extensive research of the restoration of the screen of Empress Maria Feodorovna from the collection of the Mining Institute Museum in St. Petersburg (wood, bronze, stained glass, Florentine mosaic). The Florentine mosaic and the lapis lazuli plate of the screen were restored in the way described above. The successful restoration of the tabletop from the collection of the Pushkin State Museum of Fine Arts once again confirmed the correctness of those technological innovations.