Monument Future

Laboratory investigations

The porosity of the basaltic building stones range between 17.4–23.5 % (Tab. 1). The rock material shows a very high porosity of 30 % (RC) to even 38.2 % for the RF variety (Tab. 1). The fine variety (RF) contains a microporosity (0.001–0.1 µm) of 10.2 %, whereas the RC variety contains 23.2 % microporosity (Fig. 2 below).

The saturation degree S of all investigated samples is high. The S-value of the RF variety is 0.94 and for the RC variety 0.97. The two basalt varieties show similar values: 0.97 for the basalt sample of the foundation and 0.95 for the sample from the wall.

Directional water uptake of the rock material, however, shows different values. Z means perpendicular to the bedding, X parallel to the bedding and Y parallel to the bedding and parallel to the lamination of the stone material. Water uptake for the clastic rock (RC) sample attains a value of around 11.8 kg/m²/√h in all directions. Much lower values 261were determined for the fine-grained variety (RF), which show values of 4.6 and 4.5 kg/m²/√h for the XY direction and 5.2 kg/m²/√h for the Z direction, thus attaining an anisotropy (A) of 13 %. This tendency could also be shown by the measurements of ultrasonic velocity, expecially in the case of the fine-grained rock variety (RF), (Tab. 2).

Table 1: Porosity and density of the investigated stone samples.

The rock material shows a different hydric dilatation. Both varieties show remarkable swelling, while the clastic material (RC) with up to 1.04 mm/m reaches high and critical values (Fig. 4). The fine-grained rock material (RF) only shows half of that dilatation, but a higher anisotropy of more than 50 % (Fig. 4).

Ultrasound velocity decreases in both rock varieties under water-saturated condition (Tab. 2). This particularly affects the RC variety with an average of 38 % and an average of only 7.5 % for the RF variety (Tab. 2).

Table 2: Ultrasonic velocity of the rock samples.

Conclusions
Petrophysical properties

Based on its ultrasonic velocity (Tab. 2) and due to the low values of surface hardness, the natural stone used for the Geghard Monastery can be classified as a low bound stone material with an expected compressive strengh of around 10–20 N/ mm² (Wedekind et al. 2016). Due to its pore size distribution and the high amount of micropores (Fig. 2 g and h), as well as due to the high S-value, the rock material seems to be sensitive to ice- and salt crystallization (Hirschfeld 1912).

The high hydric dilatation as well as the micoporosity of both rock varieties speak for a certain proportion of swellable clay minerals (Wedekind et al. 2013). Both values correlate with each other (compare Fig. 2 below and Fig. 4). Also, the significant decrease in ultrasound velocity under water-saturated conditions suggests the presence of swellable clay minerals and a softening of the rock structure.

Experimental conservation

Because of the onsite observed weathering forms and the measured values of hydric dilatation, the two rock varieties were treated with a swelling inhibitor and then their hydric dilatation was measured again. After the treatment, the clastic material shows a plain reduction of the hydric dilalation of around 40 %. In the case of the RC variety, a reduction of the anisotropy of the hydric dilatation to nearly zero is remarkable (Fig. 5). Furthermore, the fine-grained material (RF) shows a reduction of more than 50 % (from 0.53 mm/m to 1.8 mm/m) in the Z direction (perpendicular to bedding), whereas the XY-direction shows a lower reduction reaching 0.093 mm/m.

During the consolidation test with the silica sol, gel formation sometimes occurs on the sample surface. This can also be attributed to the pretreatment with the swelling inhibitor. Silica sols are ion-sensitive and can therefore only be used to a very limited extent on saline substrates. And, a swelling inhibitor is actually a saline solution. Consolidation tests with the silica acid esther showed a lower consolidation effect, but also a low darkening of the sample material.

It is striking that there is only a very slight increase in strength, in the case of both consolidants to the two directions parallel to the stratification (XY) (Tab. 2). The increase is around 8 % for the RF by using KSE and 7.5 % using the silica sol. The 262RC variety reaches less than 1 % in the case of KSE and around 3 % for the silica sol. The strengthening effect perpendicular to the bedding (Z) is much higher for both rock types and consolidants (Tab. 2). This attains 16 % for the RF and 30 % for the RC variety by using KSE. By using the silica sol, a consolidation effect can be established in the Z-direction for RF with 19 % and for RC at 30 %.

Figure 4: Hydric dilatation of the two rock varieties.

Figure 5: Reduction of the hydric dilatation after the treatment with the swelling inhibitor.

Conclusions

Treatment with the swelling inhibitor turned out to be a successful conservation strategy, which reduces the hydric dilatation as well as the anisotropic behavior of the rock in the case of hydric swelling. The consolidation was able to stabilize the cohesion of the material and significantly reduce the anisotropic properties in the case of ultrasonic velocity.

Acknowledgements

We would like to thank the Administration of Heritage Preservation of the Geghard Monastery as well as the Armenian Apostolic Church for their friendly cooperation.

This work was generously supported by the Volkswagen Foundation (AZ93919).

References

Hirschwald, J. (1912) Die Prüfung der natürlichen Bausteine auf ihre Wetterbeständigkeit. W. Ernst & Sohn, Berlin.

Karapetian, SG., Jrbashian, RT., Mnatsakanian, AK. (2001) Late collision rhyolitic volcanism in the north-eastern part of the Armenian Highland. J Volcanol Geoth Res 112(1), p. 189–220.

Meliksetian, K., Savov, I., Connor, C., Halama, R., Jrbashyan, R., Navasardyan, G., Ghukasyan, Y., Gevorgyan, H., Manucharyan, D., Ishizuka, O. (2014) Aragats stratovolcano in Armenia-volcano- stratigraphy and petrology. In: Geophysical Research Abstracts, vol 16, EGU2014-567-2

Pötzl, Chr., Siegesmund, S., Dohrmann, R., Koning, JM, Wedekind, W. (2018) Deterioration of volcanic tuff rocks from Armenia: constraints on salt crystallization and hydric expansion, Environmental Earth Sciences 77:660, https://doi.org/10.1007/s12665-018-7777-8

Wedekind, W., Poetzl, C., Doncel-López, R., Siegesmund, S. (2016) Surface hardness testing for the evaluation of consolidation of porous low bound stones. In: Hughes, J. J., Howind, T. (Eds.) Science and Art: A Future for Stone. Proceedings of the 13th International Congress on the Deterioration and Conservation of Stone. University of the West of Scotland, Paisley 6th to 10th 2016, Volume I, p. 491–499.

Wedekind, W. López-Doncel, R., Dohrmann, R., Kocher, M., Siegesmund S. (2013) Weathering of volcanic tuff rocks used as natural building stone caused by moisture expansion. p. 1203–1224, 2013. DOI: 10.1007/s12665-012-2158-1.

263

THE THREE FUNERARY STELAE PROJECT

Wanja Wedekind, Christoph Schmidt

IN: SIEGESMUND, S. & MIDDENDORF, B. (EDS.): MONUMENT FUTURE: DECAY AND CONSERVATION OF STONE.

– PROCEEDINGS OF THE 14TH INTERNATIONAL CONGRESS ON THE DETERIORATION AND CONSERVATION OF STONE –

VOLUME I AND VOLUME II. MITTELDEUTSCHER VERLAG 2020.

Applied Conservation Science (ACS), Leinestrasse 24, 37073 Göttingen, Germany

Abstract

Diagnostic investigations, desalination, consolidation and restoration have been applied and carried out on three historical tombstones at the Bartholomew Cemetery in Goettingen (Germany).

The tombstones were investigated by electrical capacity, electrical conductivity with and without testpads, ultrasonic velocity and surface hardness. The method for desalination applied in this study is desalination by capillary flow combined with poultices. The salt content within the poultices was measured by electrical conductivity on poultice material taken in the grid from the stone surface. Furthermore, the stone material of the three tombstones was investigated by porosity, density and hydric dilatation.

Follow-up examinations made it clear that a proof of in situ consolidation is possible. A clear relation to the weatherability due to salts and the degree of alteration as well as the possible development of clay minerals could be determined.

Introduction

Damages on structures built from natural building stones are in many cases due to the impact of salt crystallization. The fundamental prerequisite is the reduction of the stresses created by salinization.

The transport of salts largely occurs by solution in the pore spaces of natural building stones. This can also be utilized by the measures designed to reduce the presence of salts. By far the most practical method for the preservation of historical tomb monuments is a combined infusion and poultice desalination method (Wedekind et. al 2008).

The Bartholomew cemetery and its tombs

The historical Bartholomew Cemetery is closely connected to the growth of the Georgia-Augusta University of Goettingen. The cemetery is the last resting place for many distinguished German and European personalities involved in the humanities and scientific research.

The number of tombs preserved today comprises a total of 167. The types of graves found at the cemetery consist of simple enclosure graves, tomb slabs, steles, gravestones, stone pillars, gothic pinnacle-pillars, obelisks, cubic-shaped columns and two mausoleums.

Preservation and the causes for damage

The construction material predominately used for the tombs at the Bartholomew Cemetery is the highly porous Buntsandstein. Different types of damage and stress-strain phenomenon is evident 264on the stones (Fig. 1). They range from locally-formed holes to finely sanded surfaces, crack formation, flaking and crusts. The analyses show that in large part the material loss is due to the high salt accumulation resulting in salt crystallization (Kracke et al. 2007).

Figure 1: The east and west sides of the three tombstones at the Bartholomew Cemetery in Goettingen.

The objects

The treated and examined sandstone objects are three comparable classicist tombstones from the 19th century. The tombstones are of local sandstone and were made in the workshop of the sculptor and architect Andreas Rohns (1787–1853).

The grave stelae are approximately two meters high. The left tombstone (I, Fig. 1a), is a little wider and taller than the two on the right (II and III, Fig. 1a and c). All three sandstone objects show the same thickness (Fig. 1).

The middle tombstone (II) had obviously been badly damaged in the past and was restored with cement mortar. The tombstones are oriented to the east and the engraved writing is largely disfigured due to weathering to the point of illegibility.

Tombstone material and weathering

In the Göttingen region, the local Buntsandstein unit was deposited during the Early Triassic. The rock fabric is strongly inhomogeneous, as the layering, changes in composition and partly show a high clay content (Kracke et al. 2008).

By comparing all the grave monuments in the historical cemetery, it became clear that, the east sides of the objects are especially affected by weathering. This orientation-dependent damage is due to the prevailing wind and rain direction from the west. While the west sides are regularly moistened and washed off by driving rain, the drying processes, preferably take place on the east sides where harmful salts accumulate (Fig. 2b). The salts are primarily nitrate and sulphate compounds (Wedekind et al. 2008).

Figure 2: a) The desalination of the three tombstones and b) weathering model.

Methods of investigation: diagnosis and conservation

Hydrostatic weighing on drilling cores was carried out to acquire the particle and bulk density as well 265as the porosity (DIN 52102). Hydric dilatation was measured on drill core samples taken perpenticular to the bedding using a dial gauge under conditions of complete immersion in demineralized water.

The structural properties of the areas near the surface as well as the material cohesion of the tombstones were examined using two different methods: micro-hardness and ultrasonic velocity. Both measurements were carried out before the consolidation and after.

An Equotip 3 device (proceq) was used for the surface hardness measurements. A rebound hardness impact device with hardness level D was used.

For the measurements of the ultrasonic velocity, the tombstones were measured with a 52 kHz compression wave transducer. The pundipLab+ (procec) was used as the pulse generator. The measurements on the sandstone tombstones were carried out in a grid by transmission.

To determine the causes of the observed deterioration, the electrical conductivity of the tombstone was investigated using two different methods.

The electrical conductivity was measured using a portable measuring device (Protimeter Surveymaster, General Electric). This technique allows conclusions to be drawn about hygroscopic salts and moisture.

In addition, the electrical conductivity was measured using a cotton test pad moistened with distilled water (Fig. 3a). This measurement method can also help to detect non-hygroscopic and less soluble salts like gypsum.

The three sandstone tombstones were desalinated using a combined process of directional moisture flow and compresses (Domaslowski 2003, Wedekind 2016a). For this purpose, a permanently moist compress was applied to the west side of the tombstone, which was moistened with distilled water over a period of about three months by a system of drip devices (Fig. 2a). Around 50 l of destillated water was used for each tombstone during the desalination process. Drying could only take place on the damaged east sides of the tombstone, on which a poultice was also placed. The rest of the tombstone body was covered with a plastic film (Fig. 2a).

The poultices were made from fine washed sand and cellolose in a volume fraction of 4 : 1 mixed with destillated water. After total drying the poultices were sampled gridwise and each sample diluted with a controlled amount of destillated water with respect to each sampling area, mixed and measured by electrical conductivity. The dried poultice has a porosity of 57 % (about twice as much as the sandstone of the tombstone) and a pore size distribution ranging in the size classes of 1 to 20 µm as measured by mercury intrusion porosimetry.

After salt reduction the damaged areas of the three sandstone tombstones were consolidated with a silica acid esther (KSE 300, Remmers company) (Fig. 3b).

Results

The porosity of the three tombstones varies between 22 and 24 % (Tab. 1). The hydric dilatation for the sample from tombstone I was 0.7 mm/m, for tombstone II 0.1 mm/m and for tombstone III 0.9 mm/m.

A significantly increased electrical conductivity could be measured in the areas with observable weathering (Fig. 4). The three tombstones show different forms and intensities of weathering. In the lower part of the eastside of tombstone I, a semi-circular weathering of the stone due to sanding has developed, whereas the opposite side (west) shows an alveolar-like pitting (Fig. 1a). On the lower part of tombstone II (eastside), a low alveolar weathering can be observed, the backside does not show any relevant deterioration (Figure 1b). Tombstone III shows intense flaking and sanding parallel to the layer over the entire engraved writing part, while the backside is nearly intact (Fig. 1c).

The values for the porosity of the three different tombstones are also comparable (Tab. 1). However, the electrical conductivity values of the west sides of all tombstones are very low (green).

Testing of salt reduction show different results for all tombstones. The highest electrical conductivity in the poulice samples could be measured in tombstone III followed by tombstone I and finally 266tombstone II (Fig. 5). The electrical conductivity of tombstone I ranges from 0.2 to 5.07 mS/cm, for tomstone II from 0.16 to 3.83 mS/cm and for tombstone III from 0.949 up to 12.0 mS/cm.

Figure 3: a) Electrical conductivity measuraments using test pads and b) consolidation with a squirt bottle.

Figure 4: Values of electrical conductivity for a) tombstone I, b) tombstone II and c) tombstone III. Values of electrical conductivity using test pads for d) tombstone I, e) tombstone II and f) tombstone III.

The ultrasonic velocity of the three sandstone tombstones show different values. At tombstone I, the ultrasonic velocity ranges between 2.011 km/s and 2.309 km/s and reaches an average value of 2.154 km/s. Areas with a comparatively weak cohesion could be detected in the middle area of the tombstone (Fig. 6a). After consolidation, an increase in the ultrasonic velocity could only be found in the lower, heavily weathered zone (Fig. 7a).

Tombstone II showed, overall, higher values. These lie between 1.994 km/s and 2.621 km/s and reach an average of 2.25 km/s (Fig. 6b). After consolidation, hardly any changes in the ultrasonic velocity could be detected (Fig. 7b).

Tombstone III showed the lowest ultrasonic velocity values. They range from 1.34 km/s to 2.234 km/s and are only around 1.94 km/s on average. The lowest values are concentrated in the middle and in the lower, middle area of the stone (Fig. 6c). Due to consolidation, cohesion was strengthened in almost all areas that were severely weathered. However, only a few areas reach values of 2 km/s.

Before consolidation, the surface hardness at tombstone I reached a maximum of 444 HLD and a minimum of 303 HLD. The average is 398 HLD. A slight softening can be seen in the upper center of the tombstone (Fig. 6d). After consolidation, the surface strength could only be slightly increased in some parts (Fig. 7d). However, the values reach around 400 HLD, which speaks for a certain strengthening affect.

Comparably higher values between 316 HLD and 499 HLD are achieved with tombstone II (Fig. 6e). On average, however, a slightly lower value of 392 HLD was measured as compared to tombstone I. This also corresponds to the observations and the overall measurements, which illustrate that in particular whose central area of the tombstone shows strong softening (Fig. 6e) The consolidation showed clear successes here.

Almost all areas with comparably low values could be adjusted to the surface strength of apparently intact areas (Fig. 7e). Some of these areas will return to values above 400 HLD after consolidation.

The most pronounced weathering and the lowest surface hardness values were observed and measured in tombstone III (Fig. 6d). They ranged from 487 HLD to just 272 HLD. Only an average of 319 HLD could be given. Due to the consolidation, the values were partially increased. However, 267they do not achieve a satisfactory result everywhere (Fig. 7d). Values of over 400 HLD could not be achieved in any of the measured areas after consolidation.

Table 1: Porosity and density of the investigated tombstones.

Figure 5: Results of salt reduction testing using the single poultice samples for a) tombstone I, b) tomstone II and c) tombstone III.

Figure 6: Values of ultrasonic velocity for a) tombstone I, b) tombstone II and c) tombstone III and values of surface hardness for d) tombstone I, e) tombstone II and f) tombstone III before consolidation.

Figure 7: Values of ultrasonic velocity for a) tombstone I, b) tombstone II and c) tombstone III and values of surface hardness for d) tombstone I, e) tombstone II and f) tombstone III after conservation.

Restoration of the tombstones were done using a hot lime reaction mortar (Fig. 8). Similar mortars were already sucessfully prepared for tuffs and sandstones (Teipel et al. 2020, Wedekind et al. 2016b).