Monument Future

Initial situation and objective

Two columns made of Hilbersdorf porphyry tuff carry the vault on the ground floor of the town hall in Oederan, Germany (Figure 1). The volcanic Hilbersdorf porphyry tuff is a regionally used historic building material with a wide range of qualities (SIEDEL 2006, KREISSL 2010, WEDEKIND et al. 2013). The columns showed considerable surface loosening due to the influence of damaging salts leading to significant reduce in the static cross-section. Primary static assessments based of some assumed material parameters determined a high probability of failure. That’s why as an emergency 216measure a massive reinforced concrete jacket was placed around the shafts of the pillars to support the load transfer.

Figure 1: Town hall Oederan (Germany, Saxony, 16. Century A. C.).

To find a more appropriate solution including the preservation of the columns the knowledge of the real material parameters were necessary. Thus the minimum compressive strength of the stone material and the average weathering depth had to be determined in a low destructive measure. The data is needed for the calculation of the effective residual cross-section.

Approach

There is a basic linear relationship between the uniaxial compressive strength and the drilling resistance of natural stones (DELGADO et al. 2000, WENDLER, E., SATTLER, L. 1996). For any material the regression line must be determined in detail. From original Hilbersdorf porphyry tuff construction material from the Oederan Town Hall 10 test dummy cubes were made (Figure 2). On each cube one drilling resistance measurement was proceeded, using DRMS cordless device (SINT Technology, Italy). Afterwards the compressive strengths was determined. The direction of pressure was in the axial direction to the drill channel based on the assumption that borehole does not affect the result of the compressive strength measurement significantly.

Figure 2: Test cube dummies, a) prior to measurements, b) after measurements, note: fracture path by compression test is not correlated to preceeding hole of drilling resistance.

On site, in the townhall of Oederan, openings were cut into the concrete mantles around the columns (Figure 3). Those windows allowed to carry out three drilling resistance measurements in each column. Measurements aimed to estimate the minimum strength of the natural stone material of the column shafts and, moreover, to gather information of the weathering depth.

217

Figure 3: Investigation windows in the concrete jacket of the two columns.

Results

The result of drilling resistance measurements carried out with DRMS is a force given in Newton referred to increment of the drilling depth section (Figure 4). With the exception of one measurement (#08_oed8.drm), which was excluded from further evaluation, the profiles were quite balanced. The mean value of the drilling resistance was calculated very precisely. This was compared to the compressive strengths measured according to DIN EN 1926 and the reference line was determined by linear regression (Figure 5).

Figure 4: Drilling resistance on test cubes.

Figure 5: Regression function of drilling resistance and compressive strength.

The drilling resistance profiles determined on the columns have shown some more complex profiles (Figure 6).

Figure 6: Drilling resistance measured on the two original columns.

The mechanical properties of the Hilbersdorf porphyry tuff has been changed to different depths. Most probably those changes are due to salt action leading to loosening, compaction of the surface 218with underlying crumble zones, etc.. The zone of unweathered material is detected at different depths within the profile. That’s why the evaluation to an average value and converted into compressive strengths was done in two parts: the drilling resistances from profile depths of 5 mm and 15 mm were combined and plotted/calculated to the regression value (Figure 7).

Figure 7: Compressive strength deduced from drilling resistance.

Conclusions

On the basis of a linear relationship between compressive strength and drilling resistance, it can be estimated that the strengths essentially lie behind an assumed weathering zone of approx. 15 mm depth in the typical spans of unweathered Hilbersdorf porphyry tuff. According to the available results, compressive strengths from approx. 24 N/ mm^² estimated and basis for further calculations. The value correspond to the literature data for Hilbersdorf porphyry tuff (e. g. Siedel 2006, KREISSL 2010, Wedekind et al. 2013).

However, in one case of the measurements, the depth of softening could not be finally detected. With a drilling depth of 15 mm, there is an estimated compressive strength of only approx. 15 N/ mm^² at this point (column 1 middle hole #07_oed_1).

In order to provide proof of the load-bearing capacity of the columns, this minimum value of the compressive strength should be used as a maximum, additionally reduced by appropriate factors. The statically effective cross-section must be reduced at least by the amount of material loss plus 15 mm weathering depth.

Finally, two uncertainties have to be taken into account regarding the statements made above:

1. The two examination windows give solely partional insight to weathering condition of the column surfaces. Both, LANGE (2011) aswell as DU PUITS (2016) examined the columns before their sheathing and report massive surface losses of up to 5 cm depth.

2. The regression line was determined on the basis of dry-dried material. The literature does not provide information on the actual relationship between compressive strength and drilling resistance under the corresponding moisture conditions. KREISSL (2010) determined significant reduction in compressive strength on fully wet Hilbersdorf porphyry tuff material. In the case of the Hilbersdorf porphyry tuff dealt with here, the information about compressive strength as a funstion of relative humidity as possible in the town hall would have to be elaborated through additional investigations.

The case study of the town hall columns in Oederan compehensibly shows a valuation of the compressive strength of inorganic building material. The in situ determination is based on minimal invasive drilling resistance measurements on site correlated to laboratory parameters.

219References

Delgado-Rodriques, J. & Costa, D 2000: A new method for data correction in drilling resistance. Tests for the effect of drill bit wear. – Int. J. Restoration of Buildings and Monuments, 10: –18, Zürich.

DIN EN 1926 Prüfverfahren für Naturstein – Bestimmung der einachsigen Druckfestigkeit; 2006.

Du Puits V. 2016 Vorgezogener Ausschnitt aus dem Memorandum 1 – Chemnitz 03/2016.

Kreißl, S. (2010) Eigenschaften und Schadensphänomene des Hilbersdorfer Tuffs sowie Möglichkeiten der Steinergänzung mittels Mörtel. Diplomarbeit, 156 S.; http://opus.ba-glauchau.de/opus/volltexte/2010/1357/pdf/Diplomarbeit.pdf

Lange, M. 2011 Befunduntersuchung Rathaus Erdgeschoss – Schloss Kaufungen 06/2011.

Siedel, H. (2006) Sächsische „Porphyrtuffe” aus dem Rotliegend als Baugesteine: Vorkommen und Abbau, Anwendung, Eigenschaften und Verwitterung. Institut für Steinkonservierung e. V. Bericht Nr. 22, Mainz.

Wedekind, W., López-Doncel, R., Dohrmann R., Kocher M.& Siegesmund S., 2013. Weathering of volcanic tuffrocks used as natural building stone caused by moisture expansion. Environmental Earth Science. 69:1203-1224. DOI 10.1007/s12665-012-2158-1.

Wendler, E., Sattler, L. (1996): Bohrwiderstandmessung als zerstörungsarmes Prüfverfahren. – In: Wittmann, F. H. Gerdes, A. (Hrg.): Proc. 4. Intern. Koll. Werkstoffwissenschaften und Bauinstandsetzen (MSR IV’96): 145–159, Esslingen.

220

221

HISTORICAL MAN-MADE CAVES IN JAPAN: VULNERABILITY OF ROCKS AND CULTURAL ASSETS IN THE UNDERGROUND ENVIRONMENT

Luigi Germinario, Chiaki T. Oguchi

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.

Department of Civil and Environmental Engineering, Saitama University, 255 Shimo-Okubo, Sakura-ku, Saitama-shi, 338-8570 Saitama-ken, Japan

Abstract

Many social and religious traditions of Japan are deeply rooted in the underground landscape. Over the centuries, sacred hypogea have been frequented for Buddhist and Shinto practices and for burials, caves used as shelters during wars and persecutions, 222tunnels excavated for raw-material exploitation or industrial manufacturing. This study concerns man-made cave sites in central Japan (Kanto region) of diverse age, from the Kofun, about 1,500 years ago, through the Kamakura and Edo periods (13–19th century), until the modern era. They were dug into soft and porous sedimentary rocks, namely volcanic tuffs and tuffaceous mudrocks, which show varied signs of decay, related to salt weathering and water interaction. The first results of the characterization of the textural, mineralogical, petrophysical, and chemical properties of the rock and its weathering products are presented here. The vulnerability of the underground sites is correlated with the relevant environmental conditions, by the monitoring of air temperature and relative humidity. The secondary phases forming crusts and efflorescences on the cave surfaces are mostly sulfates of diverse chemistry. A critical parameter determining their crystallization, stability, or deliquescence is relative humidity, often extremely high, while their composition is controlled most notably by rock mineralogy. The eventual outcomes of this research are expected to support the adoption of countermeasures for preserving and promoting the underground cultural heritage and stone artifacts enshrined therein, and give indications about the influence and safety of visitor traffic.

Keywords: Anthropic cave; Stone decay; Microclimatic monitoring.

Introduction

The landscape of modern Japan is world-renowned for the vertical development and multiform skyline of its megalopolises, shaped by the intense urbanization begun in the post-war era. However, it endured changing fortunes in the course of history, in view of the countless natural catastrophes (earthquakes, tsunami, typhoons, volcanic eruptions) and human-made disasters (wars, fires). Indeed, many social and religious traditions in Japan rather keep a strong cultural bound with the underground world, and with the protection, isolation, and quietness that it offers. Over the centuries, sacred hypogea have been frequented for Buddhist and Shinto practices and for burials, caves used as shelters during wars and persecutions, tunnels excavated for raw-material exploitation or industrial manufacturing. Our interest dwelled on three among the many underground historical sites in the Kanto region, in central Japan (Fig. 1):

— Taya Caves (Yokohama) – excavated and sculpted by Buddhist monks of the Shingon Esoteric sect from the Kamakura until the Edo period (13th to 19th century), and dedicated to ascetic training, rituals, and later pilgrimage. The caves are a maze of halls and galleries extending for about 600 m on three stories, decorated with hundreds of rock-cut high and low reliefs, picturing deities and masters of Buddhism, shrines, real and fantastic animals, vegetal motifs, mandala, zodiac signs, and family crests (Ogata 2019).

— Yoshimi Hundred Caves (Yoshimi) – a composite Kofun, term used for sacred tumulus and megalithic burials of emperors, kings, and aristocrats, widespread in Japan about 1,500 years ago. The Yoshimi Kofun includes 219 hillside-cut tombs dated to the 6–7th century. A part of them were destroyed and tunneled during the Pacific War, making room for a production plant of aircraft parts and munitions, sheltered from the American air raids (Ikegami 2018).

— Oya quarry district (Utsunomiya) – a complex of underground sites of extraction of Oya stone, a popular building and carving material exploited since the Edo period (17th century) up to the present. Many quarries were abandoned, others converted into geoheritage and tourist attractions (e. g., History Museum, Heiwa Kannon monument, Keikan Park) (Seiki et al. 2017).

Figure 1: The underground sites under investigation (photo of Taya Caves by S. Sonoda).

Research outline

This study is aimed at investigating the vulnerability of the underground sites and its relationship with the relevant environmental and microenvironmental conditions, focusing on the properties and deterioration of the rock those caves are excavated into. The eventual outcomes are expected to support the adoption of countermeasures for preserving and promoting those sites and the stone artifacts enshrined therein, and give indications about the influence and safety of visitor traffic.

Here we introduce the first results of the characterization 223of the textural, mineralogical, petrophysical, and chemical properties of the rock and its alteration products, reserving particular attention to salt weathering and rock-water interaction. The principal techniques applied were optical microscopy, scanning electron microscope, X-ray powder diffraction, X-ray fluorescence, and mercury intrusion porosimetry. These were combined with the long-term monitoring of air temperature and relative humidity (RH).

Rock characterization

The rock of Taya Caves is a Pleistocene tuffaceous marine siltstone, grain-supported and well sorted, composed mainly of quartz, plagioclase, and lithoclasts rich in illite and smectite clay minerals, also distributed in the matrix. Two slightly different siltstone varieties are observable, which can be distinguished by their color and the presence (or not) of calcareous bioclasts.

Yoshimi Hundred Caves are excavated into a Miocene tuff with dacitic to andesitic composition, which shows a certain lithological variability, ranging from a coarse-grained pumiceous type – with porphyritic texture, plagioclase phenocrysts, abundant lithoclasts, and hypocrystalline groundmass – to a fine grained, mostly glassy type.

Finally, Oya stone is a Miocene ignimbrite having a rhyolitic to dacitic composition, with fiamme and porphyritic texture, phenocrysts of plagioclase and quartz, and hypocrystalline groundmass. The grain size is highly variable, and the typical clay clusters may reach a size of several centimeters.

The rocks of all the studied sites share, other than the pyroclastic-related sedimentary origin, also the silicate composition (plagioclase, quartz, and clay minerals) (Fig. 2). Moreover, they are all soft rocks with very high porosity. We determined an average porosity of around 45 %, and a pore-size distribution characterized by the prevalence of capillary pores (> 0.1 µm), those most involved in liquid water absorption.

Figure 2: Thin-section photomicrographs in plane- and cross-polarized light of the studied rocks.

Salt weathering and rock-water interaction

Gypsum is the only secondary phase recurring in crusts or efflorescences on the cave surfaces of all the three studied sites (Fig. 3). The best examples of gypsum crusts, thick and compact, were observed in Taya Caves, where they may jeopardize the readability of the carved decorations; they have a composite stratigraphy, characterized by the presence of an intermediate layer of calcite between the surface gypsum and the host rock. As 224for the gypsum efflorescences, the most extensive were observed in the WWII tunnels of Yoshimi Hundred Caves, where salt weathering is actually particularly severe, accounting for the crystallization of other sulfates, of Na, Al, Fe, and Mg – jarosite, alunogen, halotrichite, alum-Na, tamarugite, epsomite, and thenardite (Horiguchi et al. 2000; Oguchi et al. 2010). Oya stone also suffers from sulfate-rich efflorescences, constituted of gypsum, mirabilite, and thenardite, each phase preferentially crystallizing in different microenvironments. Low-crystallinity efflorescences were finally found in Taya Caves, composed of chlorides (sylvite, in particular), phosphates, and sulfates.

Figure 3: Examples of gypsum crusts and efflorescences with the relevant analytical data.

In addition, we conducted a complementary study of the rock-water interaction in Taya Caves, considering the lack of previous researches and the constant presence of percolating rainfall, rising damp, and extremely high humidity in that environment. We measured an extremely rapid water absorption (~25 %), a more contained yet significant adsorption of hygroscopic water (~5 %), and a very fast decay during the jar slake test (Santi 1998). These findings point out a high susceptibility to clay mineral-promoted swelling and slaking deterioration, which may produce decay patterns like erosion, rounding, scaling, peeling and, in the long term, lead to structural decay and collapses.

Lithological and environmental constraints

Gypsum formation in Taya Caves is triggered by the dissolution of the calcareous bioclasts and pyrite crystals in the rock, releasing Ca and S in solution. This is validated by the occurrence of gypsum solely on the fossiliferous rock type, which is also rich in pyrite – diagenetic or secondary, often included inside the shells. Pyrite has been thought to be the precursor also of the sulfate-rich efflorescences in Yoshimi Hundred Caves (Oyama & Chigira 1999), whereas about the gypsum origin 225on Oya stone we need further investigations. The ubiquity of gypsum can be explained in terms of the relevant microenvironmental conditions. The deepest levels of the studied underground sites, the most environmentally isolated, may have a nearly constant RH of about 100 %. This value, under the influence of external airflows, gets significantly lower and more variable closer to the cave entrances, the microclimatic monitoring reveals (Fig. 4). Gypsum has an extremely high deliquescence relative humidity (DRH) – higher than 99 % (Charola et al. 2007) – so that is stable in many environments, even very humid, provided that the substrate is not wet.

Figure 4: Microclimatic monitoring during summer and early fall (at different distances from the entrances of the underground sites and outdoors).

With broader fluctuations and lower values of RH, the crystallization of a number of other phases, typically with lower DRH, can occur. Emblematic are the findings on Oya stone, which disclose the exclusive presence of gypsum in the deep, extremely humid quarry levels, whereas the main component of the efflorescences in the middle levels or semi-underground quarries is mirabilite (DRH = 97 % at 15 °C), then replaced by thenardite (DRH = 86 %) in the dryer external environment (Steiger & Asmussen 2008). The efflorescence formation may vary temporally, other than spatially, with the climatic season: like for the sulfates in Yoshimi Hundred Caves, almost all crystallizing during the dry winter season. The conditions of lower and variable RH are associated with more frequent cycles of crystallization/dissolution, hydration/dehydration, hygroscopic adsorption/desorption, and more damaging mechanical stresses generated on the stone. The rate of stone decay related to the simple interaction with liquid water, instead, upsurges in summer and early fall, when most precipitations are concentrated and typhoons may occur.

The results suggest that the chemistry of salt weathering is often controlled by the contribution of the rock-forming minerals. Another source that is worth mentioning is related to the efflorescences in Taya Caves that, given their composition and the presence of a cultivated land above the cave vault, is possibly represented by the agricultural chemicals used therein, which migrated downwards in solution.