Monument Future

Methodology for the AE and strain monitoring

To monitor the AE and strain, specimens were formed into a cylindrical shape with a diameter of 50 mm and a length of 100 mm. Before the test, each specimen was washed using water to remove contaminants and dried for 10 days in a vacuum desiccator. The specimen was installed in a temperature-controlled chamber after setting up equipment for AE, strain, and air and rock surface temperature measurement. The AE and strain data were recorded using a laptop computer.

The AE equipment employed during the test consisted of an amplifier and a piezoelectric sensor (Fig. 1). In this study, peak amplitude, which is an important parameter in the test because it determines AE signal detectability, was continuously monitored during the entire test period at 1/100 s. 181The sensor was placed on an axis face of the specimen. Notably, vaseline was smeared in the contact area of the sensor and specimen to ensure their coupling effect; then, a c-clamp was used to fix the sensor on the specimen.

Figure 1:

 A schematic diagram of the AE and strain monitoring system.

Self-temperature-compensated strain gauges (10 mm in length) were installed on the center of the specimen in the axial and lateral directions using a three-wire system to reduce thermally induced apparent strain. A dedicated adhesive was used to glue strain gauges to the specimens. The specimen strain was continuously recorded using a measuring unit.

The surface temperature of the specimens was monitored at a 1-s interval using a thermocouple sheet and logger. Air temperature in the chamber was also recorded by a logger at a 10-s interval.

The chamber was programed with a heating–cooling range of 4–84 °C and an RTC of ±2 °C/min based on field measurements. Namely, Peel (1974) reported a maximum rock surface temperature (dark sandstone) of 79.3 °C in the Tibesti Mountains. This temperature is thought to be the highest rock surface temperature ever recorded. Meanwhile, Waragai (2019) reported the results of field measurements conducted during the dry season at Cambodia. The range of the surface temperature of the sandstone specimen varied from a 1.50 °C/min increasing rate to a −1.88 °C/min decreasing rate. As possible temperatures due to insolation, the heating–cooling range and RTC were therefore set inside the chamber: the specimens were heated from 4 °C to 84 °C over 40 min after cooling from room temperature to 4 °C over 8 h and 8 min. Then, the specimens were maintained at 84 °C for 4 h and then cooled to the initial temperature of 4 °C over 40 min. In the test, the temperature change of 4–84 °C was repeated four times. After that, the P-wave velocity of the specimen was measured using a TICO.

Results and Discussion

AE amplitude and strain

Generally, the thermal expansion behavior of rock is affected by the temperature history. To avoid the influence of such a history, termed the Kaiser effect, the peak amplitude of the AE (mV) within the large temperature change that the specimens were first exposed to is shown in Figure 2. The air and rock surface temperatures shown in Figure 2 are data obtained by thinning out every 10 s from the data recorded at each time interval. Regarding 182the AE signal, the integrated peak amplitude for 10 s, excluding the peak amplitude < 100 mV from the measured data, is shown.

Figure 2:

 The AE amplitude and the rock and air temperatures versus time evolutions of the rock samples. A: granite, B: marble, C: sandstone.

There is a difference in the size of the amplitude and the appearance frequency over time of the AE amplitude depending on the rock types; however, it can be seen that the AE signal occurred in all specimens when the temperature increased and decreased. Following the test, no apparent damage such as cracks was found in the specimens. However, the P-wave velocity decreased by 25 % for the granite (3,476 m/s), 7 % for the marble (4,090 m/s), and 0.1 % for the sandstone (3,014 m/s). Therefore, the AE signal is considered to correspond to stress waves when microcracks form at grain boundaries.

The range of the strain due to the temperature change is the largest for the strains of granite (the range of axial strain = 380) followed by that of the marble (351) and sandstone (262). The occurrence of the AE amplitude corresponds to this amount of strain, and the maximum peak amplitude is greatest in the granite (4,140 mV), followed by that of the marble (1,597 mV) and sandstone (1,000 mV). Excluding the effects of crack opening and closing due to temperature changes and the hysteresis effect, the amount of strain and generation of AE signals are closely related. The porosity is lowest for the granite and highest for the sandstone. In other words, microcracks are more likely to occur with the temperature change at the grain boundaries of the granite where the minerals are in closer contact. Because the sandstone has a higher porosity than that of the granite and marble and is not dense, the microcrack occurrence at grain boundaries is considered to be the smallest. Very heterogeneous textures such as sandstone are thought to be less responsive to thermal changes.

AE amplitude and temperature

The rock surface temperature at the time of the generation of the AE amplitude and its frequency by temperature are shown in Figure 3; in Figure 3, data of an AE amplitude = 0 are excluded.

The maximum AE amplitude of the granite is at 20 °C and 60 °C, but the frequency is the highest at 70–80 °C. However, in the case of the marble and sandstone, the maximum AE is observed at approximately 20 °C, and its frequency is large. The appearance patterns for these rocks are different, as shown in Figure 2. The AE amplitude is observed when the temperature decreases in the marble and mainly when the temperature increases in the sandstone.

Figure 3

: The AE amplitude versus the rock surface temperatures and its frequency by temperature. A: granite, B: marble, C: sandstone.

Thermal stress that causes AE is a result of the anisotropy in the thermal expansion properties of different minerals (Sirdesai et al. 2017) and the amount of certain minerals such as quartz. According to Kinoshita et al. (1995), in the case of granitic rock, even when heated at a slow heating rate that does not cause a temperature gradient inside the rock, due to the mismatch in the thermal expansion coefficient of the mineral particles, 183AE signals occur when the temperature reaches from approximately 60 °C to 70 °C, and its amplitude increases with heating. In other words, the reason why the frequency of the AE in the granite increased toward 70–80 °C is probably due to the difference in the thermal expansion coefficient of the constituent minerals. The sandstone is also composed of aggregates of various mineral grains. Although the porosity of the sandstone is high as previously described, the reason that the AE of the sandstone mainly occurred at the time of the temperature increase is thought to be due to the mismatch of the thermal expansion coefficients of the constituent minerals. The main constituent mineral of the sandstone and granite is quartz. Quartz thermally expands more than other minerals: quartz shows a thermal expansion of 0.14 % (⊥ c) and 0.08 % (||c); however, plagioclase shows an expansion of only 0.09 % (||a) and 0.03 % ( ⊥010), from 20 °C to 100 °C, respectively (Skinner 1966). Therefore, AE is generated at a relatively low temperature in rocks containing quartz. In addition, in such rocks, the increase in the AE with a subsequent temperature increase is remarkable.

Because marble is composed of only a single mineral, microcrack occurrence due to inconsistency in the thermal expansion coefficients of the minerals is difficult to recognize. However, calcite shows the thermal anisotropy of 0.189 % ( ⊥c) and −0.042 % (||c) from 20 °C to 100 °C, respectively (Skinner 1966).

AE amplitude and RTC

Figure 4 shows the relationship between the RTC of the rocks and the AE amplitude as well as frequency of the AE for each RTC. In Figure 4, the rate of the temperature increase is shown as a plus, and the rate of the temperature decrease is shown as a minus.

Figure 4

: The AE amplitude versus the rate of temperature change (RTC) of the rock samples and its frequency by RTC. A: granite, B: marble, C: sandstone.

In the case of the granite, the maximum AE amplitude was recorded when the RTC = 1.83 °C/min and the frequency of the RTC = 1.5–2.0 °C/min was approximately 20 %. In the granite, a relatively large AE amplitude is generated as the RTC increases. However, in the marble, the maximum AE amplitude occurred when the temperature decreased. The frequency of the RTC < −1.5 °C/min accounts for approximately 45 % of the whole. In the case of the sandstone, the maximum amplitude is generated at RTC = 1.5 °C/min, though the frequency of the AE during the temperature decrease is high.

Thus, the AE signal occurred when the temperature increased above RTC = 1.5 °C/min in the case of the granite and sandstone and when the temperature decreased below RTC = −1.5 °C/min in the case of the marble.

There have been many field observations of the RTC, but during recent years, it has been reported that large temperature changes have instantaneously occurred. McKay et al. (2009) measured the surface temperature of basalt using thermocouple sheets in the Atacama Desert and the cold deserts 184of Antarctica. It was found that the RTC of ≥ 2 °C/min appeared approximately 8 % on average, and the RTC of ≥ 8 °C/min appeared 0.02 % on average. In addition, Molaro & McKay (2010) measured the surface temperature of basalt and sandstone samples using a 0.375-s interval in Death Valley (USA) during April 2009. As a result, the RTC at 2 °C/min or higher accounted for 71.6 % of the basalt and 66.3 % of the sandstone, respectively.

These studies suggest that rocks subjected to rapid temperature changes due to solar radiation may form microcracks and fracture via thermal shock. In this study, it was presumed that microcracks occurred in the rock samples at an RTC above ±1.5 °C/ min. For this reason, it is believed that long-term continuous temperature change due to radiation, which is effective for microcrack generation, leads to stone deterioration.

Conclusions

AE amplitude was generated in three rock types resulting from a temperature change of from 4 °C to 84 °C, which is probably observed in the field. The generation of AE signals the formation of microcracks due to thermal stress acting on grain boundaries. A large AE amplitude occurs when the temperature gradient is > 1.5 °C/min. This means that microcracks are generated inside the stone even with normal temperature changes in the field. As these microcracks grow, the bond strength between minerals eventually weakens, leading to stone degradation.

The generation, frequency, and pattern of AE differ depending on the rock type. These depend on the mineral composition, structure, and physical and mechanical properties of the rock itself. Granite containing a large amount of quartz has the highest AE generation, followed by that of marble with a uniform mineral composition; the lowest AE generation occurred in high-porosity sandstone.

For cultural properties composed of these three rock types, maintaining a small temperature change will prevent microcrack occurrence, but it is difficult to do so. The expected temperature increase in the future due to global warming is likely to further increase the potential for thermal weathering of cultural properties.

References

Kinoshita N., Abe T., Okuno T. 1995. Thermal expansion behavior of igneous rock at high temperatures and pressures. Doboku Gakkai Ronbunshuu 511: III-30, 69–78.

Matsuoka N., Waragai T., Wakasa S. A. 2017. Physical Rock Weathering: Linking Laboratory Experiments, Field Observations, and Natural Features. J Geography (Chigaku Zasshi) doi:

10.5026/jgeography.126.369

.

McKay C. P., Molaro J. L., Marinova M. M. 2009. High-frequency rock temperature data from hyperarid desert environments in the Atacama and the Antarctic Dry Valleys and implications for rock weathering. Geomorph. 110: 182–187.

Molaro J. L., McKay C. P. 2010. Processes controlling rapid temperature variations on rock surfaces. Earth Surf Process Landf 35: 501–507.

Peel R. F. 1974. Insolation weathering: Some measurements of diurnal temperature changes in exposed rocks in the Tibesti region, central Sahara. Z Geomorphologie, Suppl 21: 19–28.

Richter D., Simmons G. 1974. Thermal expansion behavior of igneous rocks. Int J Rock Mech Min Sci Geomech Abstr 1: 403–411.

Sirdesai, N. N. Singh, T. N., Gamage, R. J. 2017. Thermal alterations in the poro-mechanical characteristic of an Indian sandstone – A comparative study. Eng Geol 226: 208–220.

Skinner B. J. 1966. Thermal expansions. In: Clark, S. P. (ed) Handbook of Physical Constants. Geol Society of America Memoirs 97: pp 75–96.

Waragai T. 2019. Measurements of thermic damage of Angkor dimension sandstone used for World Heritage temples. IAG Regional Conference on Geomorph. Abstr. Athens.

Yamaguchi U., Miyazaki M. 1970. A study of the strength or failure of rocks heated to high temperatures. J Min Metall Inst Japan 86: 346–351.

185

ACOUSTIC EMISSIONS – INSIGHTS INTO THE DECAY MECHANISMS OF THERMALLY TREATED MARBLES

Stephan Pirskawetz

1

, Johanna Menningen

2

, Siegfried Siegesmund

2

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

 Bundesanstalt für Materialforschung und -prüfung (BAM), Berlin, Germany

2

 Geoscience Centre of the Georg August University Göttingen, Germany

Abstract

Since ancient times, marble has been the preferred material for monuments, sculptures, ornaments and architecture. Though the stone is often a chosen material, long-term exposure of marble results in cumulative deterioration of the rock fabric. The rate and extent of deterioration depends on the rock fabric and the climatic conditions. Besides the thermal vulnerability of marble, a combination of thermal and hygric fluctuation accelerates the deterioration process. The weathering sensitivity of marbles can be characterised by the irreversible length change of samples after heating under thermohygric conditions as residual strain. This residual strain is a non-reversible deterioration and caused by microcracking induced by a pronounced anisotropy of the thermal dilatation coefficient of calcite. In brittle materials like marble, cracking and crack growth or friction on crack surfaces are accompanied by release of acoustic waves. The analysis of these acoustic emissions can give a deeper insight into the deterioration mechanism of marble.

In this study, acoustic emissions of thermohygric treated marble were analysed and correlated with ultrasonic velocities, thermal dilatation and residual strains. Therefore, different types of calcitic marble were cyclically heated from 20 °C to 90 °C and after equilibration of the samples cooled down again to 20 °C. While the first cycles were performed under dry conditions, the following were executed in a humid environment. The analysis of acoustic emissions enables one to determine when cracking occurs during the thermal treatment. It is also possible to differentiate microcracking from internal friction. Furthermore, the evolution of deterioration can be estimated based on ultrasonic velocities. The combination of acoustic methods and strain measurement gives an insight into the disintegration mechanism and supports the development of prevention strategies.

Introduction

Due to the exceptional microstructure of marble, thermal fluctuations initially cause an integral decohesion of the calcite crystal grains. This opens the intrinsically dense and tight structure to the ingress of water. Then, starting from the surface, ongoing weather-related temperature change and especially frost-thaw action accelerate the deterioration process. Conventional test methods as described in European standards are not adequate to evaluate the weathering resistance of the various types of marble. For a reliable assessment of the weathering resistance of marbles new methods are needed, covering the special deterioration mechanisms 186of marble. In this study, acoustic emissions of thermohygric treated marble were analysed and correlated with ultrasonic velocities, thermal dilatation and residual strains. The analysis of acoustic emissions enables one to determine when cracking occurs during the thermal treatment.

Marble weathering

The reason for the high sensitivity of calcitic marbles can be found by the pronounced anisotropy of the thermal expansion coefficient of the calcite single crystals (Fig. 1a). Temperature changes lead to anisotropic volume changes of the crystals, resulting in microstresses and microcracks. Even moderate repetitive temperature changes can cause significant deterioration. The extent of the local stresses and the resulting cracks are controlled by grain size, grain-to-grain misorientation as well as type and contact of grain boundaries (Rüdrich 2003, Shushakova 2013).

Figure 1:

 Schematic sketch of the thermal dilatation of calcite (modified after Rüdrich 2003). a) Coefficients of thermal expansion in different directions.