2 MATERIALS AND METHODS

Samples were taken from the completely defaced Egyptian stela, which now has no remaining surface details. Mineralogical, compositional, and petrographic analyses were performed using optical microscopy, x-ray diffraction (XRD), inductively coupled plasma mass spectrometry (ICPMS), ion chromatography (IC), and scanning-electron microscopy (SEM).

The clay fraction in the stone samples was extracted using the following experimental protocol. Fifty grams of crushed limestone were treated with 0.2 N HCl, and the acid insoluble residue was washed repeatedly with distilled water. The clay fraction was then separated by centrifugation. Oriented clay aggregates (air-dried, ethylene glycol [EG]–solvated, and heated to 550�C) were prepared. XRD analysis of the oriented clay aggregates was performed with a Philips PW-1710 diffractometer equipped with graphite monochromator and automatic slit using CuKα radiation. Elemental silicon was used as the internal standard. The clay content was calculated from the weight of the clay fraction following drying at 90�C for 24 hours.

To evaluate the swelling behavior of the clays within the stone, the following experiments were carried out:

1. Samples (2 � 2 � 10 cm) of Naga el-Deir limestone were partially immersed in a beaker filled with distilled water. The water surface was then covered with paraffin wax to allow capillary migration of the water only through the stone pore system. Because this stone was obviously laminated, the samples were oriented with the bedding planes perpendicular to the bottom. Macroscopic damage was recorded photographically, and efflorescence, when present, was analyzed afterward.
2. To identify the damage due to swelling of the clay and to distinguish it from damage due to salt growth (if any), stone pieces (2 � 2 � 5 cm) of Naga el-Deir limestone were placed in a vacuum desiccator, impregnated with ethylene glycol (well known as a material that swells clays), and placed in an oven at 60�C for 12 hours. Once the pieces were removed from the desiccator, they were dried at 100�C for 12 hours. This process was repeated three times. It was observed that no heat-related damage took place when control limestone samples were heated to 100�C (3 cycles). Damage created by expansion of the clay structure due to ethylene glycol solvation was recorded photographically. Changes in porosity and pore size distribution were measured by mercury intrusion porosimetry (MIP). Stone samples were also submitted to the experimental protocol described above using distilled water to swell the clays (wetting/drying cycles).
3. Damage created by expansion of clays due to relative humidity changes was evaluated by placing limestone samples (5 � 3 � 2 cm) in a climate-controlled chamber (Micro Climate Technology, model 92MCG-TC) at constant temperature (20�C) and cycling RH (40–90%). Each cycle took 8 hours. Eighty-four cycles were performed. At the end of the experiment, the total weight lost (scales that fell off) was determined. Total weight loss was expressed as a percentage of the original weight of the samples.
4. Total expansion of the stone due to water absorption was estimated using thermomechanical analysis (TMA). Limestone samples (5 � 5 � 5 mm) were placed in the TMA sample holder (Mettler TMA, model TA300), with the bedding planes parallel to the bottom, at constant temperature (30�C) and relative humidity (50%). When the system was stabilized (typically after 30 minutes from the start of the experiment), distilled water was added, and the isothermal linear expansion of the samples was recorded (scan rate: 6 measurements/minute) for both the initial wetting and the final drying. When maximum expansion was reached, subsequent drying was achieved, and contraction of the sample was recorded. Two wetting/drying cycles were performed for each sample.
5. High-magnification study of clay morphology and distribution within the limestones was performed using a scanning electron microscope.