Monument Future

Conclusions

In the present research structure sample were analysed and the problem of salt crystallization in the structure deterioration was approached.

It was found that salts on the structure surface rarely crystalize due to high environmental RH in the area of study. Therefore, it can be inferred that probably salt plays a minor role in the stone deterioration, while the main causes of deterioration should be further studied.

129Quarry and structure stone were physical-chemical compared through X-ray and petrographical analysis. Quarry stones were found compatible for replacement of deteriorated block, although further research is needed to include more relevant physical and aesthetic variables in the analysis of stone replacement such as colour, texture and moisture transport properties.

Acknowledgements

This Project has been funded by the Universidad de Cartagena, Colombia, through the resolution number 00473 of 2016 “Octava convocatoria a proyectos de investigación, para grupos de investigación visibles (categorizados o reconocidos) en la plataforma Scienti del departamento de Ciencia, Tecnología e Innovación-Colciencias y avalados por la Universidad de Cartagena”. Furthermore, this project has been funded with the support of the European Commission (Grant agreement no. 2014-0873/001-001). This publication only reflects the view of the author, and the Commission cannot be held responsible for any use which may be made of the information contained therein - Elarch Program (project reference number: 552129-em-1-2014-1-it-era mundus-ema21). Finally, Authors would like to acknowledge the development of this work to Universidad Nacional de Colombia by funding DIB 34835.

References

Álvarez-Carrascal, J. L. (2018). Retrospectiva de la construcción del cordón amurallado de Cartagena de Indias, su presente y su futuro. Universidad de Granada.

ASTM C1721 – 15 Standard Guide for Petrographic Examination of Dimension Stone, (2015).

Baronio G, Binda L, Tedeschi C, Tiraboschi C (2003) Characterisation of the materials used in the construction of the Noto Cathedral. Construction and Building Materials 17:557-571 doi: 10.1016/j.conbuildmat.2003.08.007.

Binda, L., Baronio, G., Gavarini, C., De Benedictis, R. & Tringali, S., (1999) Investigation on Materials and Structures for the Reconstruction of the Partially Collapsed Cathedral of Noto (Sicily). In 6th Int. Conf. Structural Studies, Repairs and Maintenance of Historical Buildings, STREMAH 99, Dresden, Germany, pp. 323–332.

Binda L, Baronio G, Tedeschi C, Tiraboschi C (2003c) Experimental research for the choice of adequate materials for the reconstruction of the Cathedral of Noto. Construction and Building Materials 17:629-639. doi:10.1016/S0950-0618(03)00059-X.

Bionda, D. (2005). A graphical user interface to the ECOS thermodynamic model for the prediction of the behaviour of salt mixtures under changing climate conditions.

Cabrera, A., Martelo, R., Martinez, A., & Martinez, R. (1995). La Ruta De Los Hornos.

UNI EN 1936:2007 Natural stone test methods – Determination of real density and apparent density, and of total and open porosity, 1 (2007).

DIMAR. (2018). Boletín Meteomarino del caribe Colombiano. No. 68 Agosto 2018. https://issuu.com/dimarcolombia/docs/boletin201808

Dunham, R. J. (1962). Classification of carbonate rocks according to depositional texture. In W. E. Ham (Ed.), Classification of Carbonate Rocks. American Association of Petroleum Geologists Memoir (pp. 108–121).

Ferretti, D., & Bažant, Z. P. (2006). Stability of ancient masonry towers: Moisture diffusion, carbonation and size effect. Cement and Concrete Research, 36(7), 1379–1388. https://doi.org/10.1016/j.cemconres.2006.03.013

Ghafar Ahmad, A., & Abdul Rahman, H. F. (2010). Treatment of Salt Attack and Rising Damp in Heritage Buildings in Penang, Malaysia. Journal of Construction in Developing Countries, 15(1), 93–113.

Kameni, M., & Orosa, J. A. (2016). Building construction materials effect in tropical wet and cold climates : A case study of office buildings in Cameroon. Case Studies in Thermal Engineering, 7, 55–65. https://doi.org/10.1016/j.csite.2016.01.007

Nasraoui, M., Nowik, W., & Lubelli, B. (2009). A comparative study of hygroscopic moisture content, electrical conductivity and ion chromatography 130for salt assessment in plasters of historical buildings. Construction and Building Materials, 23(5), 1731–1735. https://doi.org/10.1016/j.conbuildmat.2008.09.029

Price, C. (2000). Runsalt. An expert chemical model for determining the environmental conditions needed to prevent salt damage in porous materials, European Commission Research Report No 11, (Protection and Conservation of European Cultural Heritage). Archetype Publications.

Saba, M., Hernandez-Romero, N., Quiñones-Bolaños, E., & Lizarazo-Marriaga, J. (2019). Petrographic of Limestone Cultural Heritage as the Basis of a Methodology to Rock Replacement and Masonry Assessment: Cartagena de Indias Case of Study. Case Studies in Construction Materials. https://doi.org/https://doi.org/10.1016/j.cscm.2019.e00281

UNESCO. (1984). Port, Fortresses and Group of Monuments, Cartagena.

UNESCO. (2017). List of World Heritage in Danger.

Zoghlami, K., Lopez-Arce, P., & Zornoza-Indart, A. (2016). Differential Stone Decay of the Spanish Tower Façade in Bizerte, Tunisia. J. of Materials in Civil Engineering 1–14. https://doi.org/10.1061/(ASCE)MT.1943-5533.0001774.

131

COMPARATIVE ANALYSIS OF VOLCANIC TUFFS FROM EUROPE, ASIA AND NORTH-AMERICA

Ákos Török¹, Luigi Germinario¹, Rubén López-Doncel², Christopher Pötzl³, Siegfried Siegesmund³

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.

1 Budapest University of Technology and Economics, Department of Engineering Geology and Geotechnics, Megyetem rkp. 3, H-1111 Budapest, Hungary (torokakos@mail.bme.hu)

2 Geological Institute, Autonomous University of San Luis Potosí, Mexico

3 Geoscience Centre of the University of Göttingen, Goldschmidtstr. 3, 37077 Göttingen, Germany

Abstract

In this study, volcanic tuffs from Europe (Germany and Hungary), Asia (Armenia) and North-America (Mexico) are compared. The physical properties (density, porosity and pore-size distribution, strength) and mineralogical composition of basic, intermediate and acid volcanic tuffs were measured in laboratory conditions aiming to evaluate these parameters in terms of stone durability. The colour of rhyolitic to andesitic tuffs, the amount of lithic clasts and the proportion of groundmass vary. The mineralogical composition, porosity and textural characteristics of the studied tuffs are very different. The comparative analysis suggested that clay content and distribution within the pores, the amount of glassy groundmass, porosity and pore-size distribution, as well as hygric swelling and thermal behaviour of the tuffs, are the main control factors of the durability.

Introduction

Volcanic tuffs are important lithotypes of our cultural heritage. The material characteristics and the worldwide occurrence allow a widespread use from sculptural elements to massive built structures such as castles and fortresses. Despite the extensive use of tuffs, these materials are prone to weathering and show various forms of decay due to their mechanical properties, textural variability and high porosity.

The current paper provides an overview of the petrophysical, mechanical and petrographic properties of tuffs from Europe, Asia, and North-America. It presents the geological setting and utilisation of these tuffs. It outlines the decay features that are linked to external factors and discusses the differences in durability linked to textural/mineralogical characteristics and stress conditions such as freeze-thaw and salt action. The study also provides a background for the material selections for restoration.

Materials

In the course of its geological history, immense amounts of volcaniclastic material were deposited in the territory of Armenia. Armenian tuffs show a wide variety of color, grain size, clast content and chemical composition. The two examples presented in this study, Hoktemberyan and Golden Armenia, are frequently used in the past and recent construction in many parts of the country (Figure 1a shows the utilization of Qasakh tuff).

In Mexico, volcanic tuffs were and are widely used in the construction of pre-Hispanic, colonial and 132modern monuments. The type and variety of these tuffs exceed half a hundred. In the present study, five tuffs of Central Mexico (San Luis Potosí, SLP), are presented as examples (Figure 1b). The volcanic tuffs of SLP belong to the Paleogene Silicic Large Igneous Province, the Sierra Madre Occidental, the largest ignimbritic province in the world.

Figure 1: Examples of the historical use of the studied tuffs and relevant geographic locations: a) Hovhannavank Monastery, Ohanavan (Armenia), 5th c.; b) Templo del Carmen, San Luis Potosí (Mexico), 17th–18th c.; c) Eger Castle, Eger (Hungary), 13th–18th c.; d) Porphyry house, Chemnitz (Germany), 1868.

Miocene volcanic activity produced a large amount of welded and unwelded pumiceous tuffs in Hungary. Besides the prevailing rhyolitic composition, dacitic and andesitic tuffs also formed in Eastern Hungary. Rhyolitic tuffs are particularly common in a 50 × 10 km area known as Bükkalja Volcanic Field in Northern Hungary.

Emblematic monuments such as the castle or the minaret of Eger were constructed from this material (Figure 1c).

Significant volcanic tuff deposits in Germany are mainly found in the eastern and western Central part of the country. Permian volcanism led to the deposition of the Hilbersdorf tuffs in eastern Central Germany, near the city of Chemnitz.

The Quaternary Weibern tuff is located in western Central Germany, in the municipality of Weibern.

Prominent examples of its application as building stone are, e. g. the Kaiser Wilhelm Memorial Church in Berlin, the Castle Church of Chemnitz or the porphyry house of master stonemason Findewirth in Chemnitz (Figure 1d).

Methods

The tuffs were described using polarized-light microscopy. Bulk and material density, porosity (EN 1936:2006), and water absorption at atmospheric pressure (EN 13755:2008) were measured. Pore-size distribution was analyzed by mercury intrusion porosimetry (MIP) (ASTM D4404-84). Ultrasonic 133p-wave velocity was recorded by direct transmission method (EN 14579:2004). Strength parameters such as uniaxial compressive strength (UCS), tensile strength (ASTM D3967-16) were measured, and Young’s modulus of elasticity was also recorded (ASTM D7012-14). Durability was assessed by resistance to freeze-thaw (EN 12371:2001 with modifications) and resistance to salt attack (EN 12370:1999).

Petrology and Mineralogy

The creamy rhyolitic tuffs from Northern Hungary are characterized by a porphyritic texture with plagioclase phenocrysts. The pumice content is high, and the groundmass is mostly glassy (Figure 2). Critical parameters associated with high textural variability are crystal-groundmass-pumice ratio; grain size; welding degree. These features may vary in outcrops a few kilometres away, and even within the same quarry, to a lesser extent (Germinario & Török 2019).

The German Hilbersdorf Regular tuff, appears in irregularly speckled and marbled in pale pink to dark purple and bright beige to greenish colours. A variety of elongated lapilli inclusions are embedded in the groundmass and millimetre to centimetre-large cavities are partly filled with loose, clayey material. It has a porphyritic texture with mono- and polycrystalline quartz, muscovite, hematite, calcite, feldspar relics and lithic clasts in the glassy matrix (Figure 2). The porphyritic Weibern tuff consists of a fine-grained yellowish-brownish matrix in which pumice lapilli of yellow colour and partly elongated clasts of different but mostly grey colour are embedded. The rock fragments are mostly sandstone and shale as well as basalt and volcanic glass fragments. The matrix mainly consists of analcime, muscovite/illite, quartz and calcite (Wedekind et al. 2013).

The Armenian Hoktemberyan tuffs are trachydacites, while the Golden Armenia varieties have rhyolitic composition. The most popular Hoktemberyan tuff is of characteristic brick-red color, but transitions to orange-brownish and blackish also exist, often given different local trade names. In its fine-grained brick-red groundmass (~75 %), small elongated pumice clasts of red and black colour are embedded. Grey to black rock fragments, as well as huge amounts of white, elongated feldspars and glass particles in the millimetre range give a slightly speckled appearance. Thin-section analyses show feldspar and amphibole phenocrysts as well as volcanic lithoclasts and hematite located in a cryptocrystalline matrix (Figure 2). The yellowish-golden groundmass of Golden Armenia tuff embeds beige, grey and slightly reddish clasts. In thin section, volcanic lithoclasts, quarz phenocrysts, feldspar relics and vitric fragments are oberserved in a glass-rich matrix. Considerable amounts of swellable clay minerals (corrensite) are verified in Pötzl et al. (2018b).

All the Mexican volcanic tuffs of SLP have rhyolitic composition. The percentage of crystals and matrix vary of 30 %–70 %. The rocks contain mainly 134quartz, with an average abundance of 45 %, alkali feldspar (mostly sanidine and orthoclase) with 35 %, and plagioclase (oligoclase to anorthite) not exceeding 30 %. The texture of these rhyolitic ignimbritic tuffs varies from porphyritic hypocrystalline to vitrophyric (Figure 2).

Figure 2: Thin-section photomicrographs in plane- and cross-polarized light from a selection of the studied tuffs (see detailed explanations in the text).

Physical properties

The Hungarian tuffs generally have high effective porosity, from 17 % to 30 % approximately, with a well interconnected pore network, accounting for the almost exclusive presence of open pores in the rock volume. Capillary pores (> 0.1 µm) represent distinctively the most abundant size. The low bulk density, 1.5 g/cm³ on average, is directly related to the high porosity, as well as to the abundance of low-density pumice clasts and glass shards. Considering the mechanical properties, the studied tuffs are weak to moderately strong rocks with compressive strength of 7 and 28 MPa. Saturated conditions produce an extreme deterioration of the mechanical properties, with the strength that may decrease even by 90 % in the weakest tuff varieties (Table 1).

The Hilbersdorf and Weibern tuffs of Germany have high effective porosities of 26 % and 37 %, respectively. While Hilbersdorf tuff contains substantial amounts of micropores (43 %), Weibern tuff is characterized by huge portions (> 86 %) of capillary pores. In comparison to the other tuffs of this study, the Hilbersdorf tuff shows considerably high bulk (1.9 g/cm³) and matrix (2.6 g/cm³) densities, as well as moderate to high p-wave velocity (2.6 km/s), tensile (4 MPa) and compressive strength (32 MPa). The Weibern tuff, on the other hand, is characterized by lower densities and p-wave velocity and tensile strength (1.5 MPa) (Table 1). Both tuffs suffer a strong decrease (up to 40 %) in their mechanical properties when tested under saturated conditions (Table 1).

The Armenian tuffs show a broad range of petrophysical properties. The tuffs of this study have a high effective porosity of 21 % to 36 %. However, the Hoktemberyan tuff is characterized by a much lower ratio of micropores (9 %) and bulk density (1.6 g/cm³). Mechanical properties display the Armenian tuffs as moderately strong with maximum tensile strength values of 5 MPa. While Golden Armenia suffers from a distinct strength decrease under saturated conditions, the Hoktemberyan tuff does not seem to be affected by water in its mechanical properties.

The porosity of the Mexican tuffs ranges from 18 % to 36 % and the density values vary from 2.3 g/cm³ to 2.6 g/cm³, respectively. The pore-size distribution of the studied tuffs are unimodal and bimodal, in addition the average pore radius fluctuates from 0.15 µm to 4.02 µm, dominating the capillary pores. SLP volcanic rocks are also very soft to moderately hard, depending on the degree of welding and showing average values of tensile strength in dry conditions of around 8 MPa, falling extremely to values down to 1.33 MPa in water-saturated conditions. Well-welded ignimbritic tuff samples of SLP can have uniaxial compressive strength of 90 MPa (Wedekind et al. 2013).

Water saturation reduces the strength (both uniaxial compressive and tensile strength) of the tuffs with one exception (Hoktemberyan) (Table 1). The strength of the saturated samples can be as low as one-fourth, but in general, the value is half or two-thirds of the dry one. The porosity of the tuffs is in between 18 to 37 vol%, while matrix densities are relatively uniform with 2.3 to 2.6 g/ cm³. The P-wave velocity of the studied tuffs is in between 2.3 and 3.9 km/s. However, these values do not necessarily indicate differences in porosity (Table 1).

Durability

The durability of the Hungarian rhyolitic least porous welded tuffs may reach 90 freeze-thaw cycles and 20 salt crystallization cycles with minor or no decay in structure and technical properties. However, the most porous softest varieties, which have found a larger application in the historical built heritage, can withstand only few weathering cycles, before disintegration, erosion, and cracking. The magnitude of the differences in porosity, and consequently in the absorption of water and salt solutions, controls primarily the diverse durability.

135Secondary factors are pore-size distribution especially in the range 0.1–10 µm – critical for ice and salt crystallization damage. Considering the textural features, the most significant discriminating factor is the proportions between crystals and the weaker pumice and groundmass.

Table 1: Physical properties of studied tuffs (data of Hungarian rhyolitic tuffs is from Germinario and Török 2019, Hungarian andesite tuff is from Török 2007, data of the German and Armenian tuffs are from Pötzl et al. 2018a and Pötzl et al. 2018b, data marked by * is from López-Doncel et al. 2016; and marked by + is from Wedekind et al. 2013).

The Hilbersdorf tuffs withstand 11 to 21 cycles before complete disintegration due to its high porosity, bimodal pore radii distribution and high amounts of micropores.. The hydric and thermal expansion of these tuffs is greatly influenced by their clay mineral content and ratio of micropores. Maximum hydric expansion can reach values of 7.5 mm/m (Pötzl et al. 2018a).

Armenian tuffs also show great variation in their durability and weathering behaviour, dependent on their petrophysical properties. The resistance against salt attack varies from 16 to way over 150 cycles. Even considerably soft tuffs, like the Hoktemberyan varieties, are not strongly attacked by the salt crystallization and withstand more than 150 cycles. The Golden Armenia, on the other hand, disintegrates after 34 cycles. The main difference between both tuffs is the clay mineral content as well as the much higher ratio of micropores in Golden Armenia.

Regarding the durability of the Mexican tuffs, the conducted salt bursting tests show that after 14 cycles the most porous sample was completely destroyed and two samples, both with a bimodal pore distribution, have resisted at the end of the test 47 and 48 cycles, respectively. All SLP samples 136were tested for hydric and thermal expansion and all of them did not show any expansion, and one sample showed even hydric contraction with values around 0.03 mm/m.