most of the wall paintings present in the temple had to be consolidated because the preparatory layers, the plaster between the paint layer and the wall, were detached from the latter. So, in order to fill the void, a mortar made of compatible construction materials had to be injected. Hand drills were used to drill 1 to 2 mm diameter holes in the area to be consolidated. The gap was freed of dust and debris through a rubber siphon. Then a surfactant solution had to be injected through syringes. In that way the solution/mortar would spread in a better way. Two different solutions had to be injected: the first one, usually an acrylic binder, when there was a slight detachment and the second one, an acrylic binder mixed with local clays, when there were deeper detachments.
there was a serious detachment of large sections of wall paintings of several centimeters from the wall. The consolidation of this very large areas was accomplished with the aid of flexible polycarbonate sheets used to prop the pictorial layer. Long screw-drill-bits were inserted to fix the polycarbonate to the wall. Once the large detached areas were secured, they were slowly pushed back as much as possible by screwing the long screw-drill-bits of fraction of millimeters a day. When the gap between the wall and the paint layer was small enough, a mortar made out of local clays mixed with an acrylic binder was injected by syringe: the injections were preceded by the introduction of a surfactant solution. This operation took a very long time since the mortar had to be completely dry after every injection so as to prevent any fall of paint layer due to an overweight caused by excess of mortar.
the first floor was a particularly challenging situation, for the majority of the paint layer was not visible because covered by a thick coating of clay leakage. Thus, the first procedure for its cleaning required the use of orthodontic micro-drills, fiberglass pencils and surgical knives to remove all the clay deposits from the paint layer. The same procedure was later employed for the cleaning of the other two floors. This phase was extremely time-consuming, requiring four years to reach a homogenous level of cleaning.
the removal of the altered varnish yielded yet more difficulties. The varnish was of the same composition as the paint layer’s binder; thus, any kind of chemical which would dissolve the varnish would have dissolved the colors as well. Once the proper chemicals were identified, the cleaning solution was applied with cotton swabs through Japanese paper to absorb the varnish as it dissolved. Each floor’s wall paintings required a unique chemical solution mixed in proportion to the strength required.
a good portion of the wall paintings on the north, west and south walls were under threat of biological attack. Fungi were literally wiping out the pictorial layer. Its binder, made out of polysaccharides, together with the damp conditions of those portions of walls, constituted the perfect environment for the growth of micro flora. Most of the binder of the painting had been consumed by the fungi, meaning that the colours, deprived of their glue, had simply vanished. The micro flora was treated as follows: a solution of benzalconium chloride in water was spread on the surface by hand sprayer several times with monthly intervals for one full work season. The spraying was diffused on a larger area than the one required to prevent spores to scatter from an affected surface to a non-affected one.
the extremely detailed quality of the wall paintings meant that a very difficult and time-consuming procedure of pictorial integration was required. An elaborate aesthetic intervention was required to return homogeneity to the wall paintings. Abrasions, missing parts, lacunae and light spots were disturbing a clear view of the pictorial cycle. A huge amount of time was needed to restore the definition of the damaged details, or to reconstruct them. Following the requests of the monks and the local people it was decided to reconstruct as much as possible the pictorial cycle to allow the people to use again their temple and to prevent the turning of a religious building into a museum for tourists.
all the hues were balanced using watercolors or natural pigments mixed with a 3% solution of Arabic gum. So abrasions were touched up as well as faded colors. The losses and the missing parts were drawn with pencils and colored using pigments as base-colors. Then the hues of those lacunae were matched with the original hues of the murals through watercolors. Some reconstructions took place in the south and east corridor in an attempt to bring homogeneity back to the pictorial cycle. No retouching or reconstructions occurred in the north and west corridor because the murals still need to cleaning.
the retouching was carried out exclusively with the aid of selected series of watercolors by touching up abrasions, faded colors and micro-losses of paint layer. Where dark spots were present and the use of watercolors was ineffective, it was decided to employ watercolor pencils, more effective in lightening the dark areas. Reconstructions took place mostly in the north wall, which was severely damaged by a prolonged fungi's activity even though the decorative scrolls of some mandalas were reconstructed also on the eastern and southern side.
the reconstructions were carried out with watercolors and watercolor pencils. In the areas where the color was missing, the presence of color indicators, still visible as original sketch, was of a great help in guiding us putting the right hue in the right place. The old masters used to leave marks, letters or numbers for their students to use a particular color paired with a particular mark. The severed murals still keep these marks, even though most of the colour has been washed away or partially damaged. Hence, it was possible to find out which color had to be used according to the marks found under the paint layer, most of it still visible through human eye or through an infrared camera. The retouching and the reconstructions were quite challenging because on the southern side the colors were still vivid while on the northern side the hues had heavily faded because of the continuous action of the sun coming through a skylight that was closed during the conservation work to stop the deterioration of the murals. Since the retouching and the reconstructions had to match the colors we found after the cleaning process, one side of the floor is brighter and colorful then the other.
