The Circle of Ancient Iranian Studies
The Causes & Prevention of Stone Decay
at Archaeological Sites in Iran
By S. Z. Lewin
Department of Chemistry,
New York University
During the period 24 Jan. - 31 Jan. 1976, this investigator made a detailed field study of the chemical character, ambient conditions, and physical state of the exposed stone surfaces at the following archaeological sites in Iran; Persepolis, Naqsh-e-Rustam, Pasargadae, Bishapur, Takh Bostan, Bisutun (Kermanshsh), and Ganjnameh (Hamadan).
These several sites which were collected and taken back to N.Y.U., the types of serious decay problems afflicting these monuments can be summarized as follows:
1. Surface Flaking. This is a particularly serious and urgent problem at Persepolis, where important details are being lost from the bas-reliefs on the 'vertical sides of the great eastern stairway of the Apadana.
I. The Problems
These reliefs, which were produced in ca. 480 B.C. during the reign of Xerxes I, had been exposed to weathering only until the year 330 B.C., when the palaces at Persepolis were burned by
As a result of these observations, together with data obtained from laboratory measurements on specimens of stone, both decayed and pristine, from Alexander of Macedonia. The ashes and earth from the mud-brick walls buried the eastern stairway, but not the northern stairway, whose bas-reliefs remained exposed during most of the subsequent time. The eastern stairway was discovered and excavated by Herzfeld in 1933, and because of the fact that they had been covered by earth for the intervening nineteen hundred and sixty-six years, the bas reliefs on the eastern stairway are in excellently preserved condition, whereas those on the northern stairway are very badly mutilated. Consequently, the bas-reliefs of the eastern stairway are among the great treasures of human cultural history.
Since the restoration of the Apadana, starting in the 1930's, the authorities at Persepolis have noticed evidences of decay of the newly exposed surfaces. It was thought that thermal expansions and contractions, due to the heating effect of the sun during the day alternating with the cooling occurring during the night, were the cause of the decay, and a protective canopy was erected over the stairway. However, this was unacceptable aesthetically, and was removed about ten years ago.
Our examinations and analyses show that the cause of the decay is not thermal variations, but is instead the deposition of crystalline materials in the pores and intergranular spaces of the stone, due to cycling between wet and dry conditions. That is, the stone becomes wet in certain zones after every rainfall when water migrates into the stone (a) upwards, by capillary rise, from the adjacent wet earth at the base, and (b) downwards, by diffusion, from the rain which has soaked into the top horizontal surfaces. This water carries dissolved salts in it, and when the stone dries out, the evaporation of the water leaves these salts behind. They crystallize in the pores of the stone at the surface, creating an expansive pressure that, after a sufficient number of wet/dry cycles, eventually causes a thin layer of the surface to lift away from the bulk of the stone, producing a flake or blister. Direct exposure to the sun accelerates this process, by promoting the drying step, but it is the water migrating through the stone and evaporating from the exposed surface that does the damage -- not the thermal variations themselves.
This point is clearly demonstrated by the condition of the bas-reliefs at Bisutun (Behistun), 32 km. from Kermanshah. These are midway up on the exposed sheer cliff face of a tall mountain -- so high that moisture from the ground cannot reach them, yet so far below the horizontal top surfaces that rain soaking into the mountain from above does not diffuse to them. In several isolated spots on the relief there is seepage of water that has found its way through fissures and faults in the rock mass.
These bas reliefs, fashioned around 520 B.C. during the reign of Darius I, have been continuously exposed to the elements for nearly 2500 years, yet are as crisp and sharp as if they were new, except in those isolated places where water seeping through fissures in the rock of the mountain flows out over certain portions of the reliefs. The stone at Bisutun is identical in its nature and properties to that at Persepolis, and during much of the year it is subjected to daily thermal variations of ca. 30-°C (from 20 to 30° during midday to - 10 to O°C at night). Thus, it is clear that without the intervention of migrating water bearing dissolved matter, even 2500 years of exposure to sun and air and rain has not produced the slightest evidence of the type of decay which has occurred at Persepolis in just a few years.
Examination of the surfaces of the bas-reliefs of the eastern stairway of the Apadana at Persepolis discloses that the type of flaking and surface losses that have been occurring during the past 43 years (i.e., since the exposure of these stones in 1933) is the same as that which was present when these surfaces had just been excavated and photographed. That is, during the first 150 years, from 480 to 330 B. C., this stone decay process had already been in progress, and it was arrested during the next 1966 years of burial, only to resume again in the 1930's. In fact, the relative extents of old and new surface losses indicate that the rate of decay in Achaemenian times was about the same as it is at present.
Thus, it is an interesting paradox of history that it was the very destruction of Persepolis by Alexander two thousand years ago that has preserved for us some of the most important cultural records of the founders of the Persian Empire. For if these stones had not been buried, an additional century or two of exposure would have completely obliterated their message.
2. Surface Erosion. Liquid water flowing over the surface of a calcite stone, such as marble or limestone, can produce very extensive losses of the stone. The solubility of calcium carbonate in pure, carbon dioxide-free water is very slight, but it is significantly greater in ordinary ground water and rainwater, both of which contain appreciable amounts of solubility-enhancing substances, such as carbon dioxide and various salts.
The effect of water seeping out of a rock mass and flowing over the surface of a bas relief is to wear away, by direct dissolution, the sculptured detail. This process is occurring in a very serious way in several of the sites studied. The most extreme case, and the one requiring intervention most urgently, is that of the bas-reliefs at Bishāpur, 130 km southwest of Shiraz. Here, water in an irrigation canal that was in continuous, flowing contact with a section of the bas reliefs on the west side of the Shapur river for 150 years had eaten into the stone wall to a depth of 30 to 40 centimetres! Several years ago, this water was re-routed away from the bas-reliefs by Dr. Sarfarāz, thus eliminating this source of erosion. However, the rock of the Koumareh mountain is highly fractured and faulted, and there is a considerable volume of seepage water that flows out of cracks and holes above and in the bas reliefs. This liquid, dripping and flowing continuously over the exposed stone during much of the year has eroded an appreciable part of the surface detail, and if allowed to continue, will in time completely obliterate these important sculptures.
The continually wet stone surface also serves as an ideal substrate for the growth of algae and lichens, and these vegetations make an additional contribution to the decay and disfigurement of the bas-reliefs.
3. Surface Stains. There are isolated instances where foreign matter has impregnated the surface of certain bas-reliefs and other sculptures particularly at Persepolis, producing unsightly stains. The staining material may be an organic grease of oil in some cases, and an inorganic deposit (e.g., iron rust) in others.
These are generally not sources of decay of the stone, but are disfiguring, and should be removed. It is obvious that the removal of the stain must be accomplished by a technique that does not damage the stone, and in the case of a chemically reactive stone, such as the calcium carbonate stones of Persepolis and most other ancient Persian sites, this will require chemical treatments specifically designed for the type of stain present.
4. Vertical Splitting. The dark grey to-blue stone that has been employed in a number of the stone columns and other structural elements at Persepolis and Pasargadae, although it is a true marble or limestone composed of calcite grains, shows as a result of weathering an extreme degree of splitting and cleavage along vertical planes. Externally, it has exactly the appearance of a layered silicate rock, such as a mica, or slate, or schist. It is, because of this splitting, very weak, and subject to accelerated decay due to water invading the cracks and, upon freezing in winter weather, enlarging and extending the interlayer cleavages. This type of weathering deterioration of a calcite marble is extremely unusual, and appears to be unique to the effect of local conditions on the stone in this particular geographic area.
II. Interventions Required
The first two types of stone decay outlined above are sufficiently clear that the kind of intervention needed to preserve the stone is fully understood. The third and fourth types of problems cited above will require further study and experimentation before a technique of conservation can be presented with complete confidence.
1. Persepolis and Pasargadae. To arrest the surface flaking that is destroying some of the detail of the bas reliefs at the Apadana, and that is wearing away some of the stone surface of the tomb of Cyrus, it is necessary to prevent water from migrating (a) from the soil at the base of the monument upward into the stone, and (b) from the top surfaces downward.
This can be accomplished simply by the following two interventions:
The water-repellent chemical must be carefully selected, for most of those currently offered by commercial firms are not effective on an alkaline stone such as these calcite marbles or limestone, and even may produce an undesirable surface discoloration of the stone.
2. Bishapur. To prevent further erosion of the bas-reliefs by water seeping out of the mountain rock mass it is necessary to divert this water around the sculptures. In view of the size and shape of the mountain, this can only be done by making a fairly deep horizontal cut into the stone paralleled to the top line of the bas-relief, and about 2 meters above it. The stone should be removed along this line to a depth into the mountain of a meter or more; how deep it will be necessary to go will depend upon the kind of fissures and rock fractures present, and will be determinable as the work proceeds. This horizontal trench, which should tip downward toward the interior of the mountain, would be filled with concrete, and provided with drains to direct any water collecting at its rear away from the bas-reliefs area.
3. Other Interventions. In those places where large vertical cracks and openings have developed, due to the peculiar weathering pattern of this dark-coloured marble, and where the structural integrity of the part is important (as in columns, door frames, w-all elements), the innermost parts of the cracks should be filled, either with a mastic or a dense concrete, to prevent water held by surface tension in the narrow cavities from freezing in cold weather and exacerbating the deterioration.
The understanding of the manner in which this type of weathering is occurring, and the devising of appropriate chemical treatments to strengthen and consolidate this stone will require a program of laboratory investigations and testing.
Copyright © 1998-2015 The Circle of Ancient Iranian Studies (CAIS)