Читать книгу Monument Future - Siegfried Siegesmund - Страница 285

The thermal expansion was analysed in two different directions. One according to the preferred c-axis orientation (z) and one perpendicular to z (y).

188In Figure 4, the irreversible length change (residual strain) is given in mm/m after each heating-cooling cycle (20°–90°–20 °C). After three dry cycles (red background) the samples were analysed under wet conditions (blue background). All four marble varieties show residual strain after the first dry cycle, with no or little increases in the following dry cycles. As the residual strain slows down, the adding of water again causes an increase of the residual strain. While the Lasa and Gioia marble shows no significant anisotropy for the residual strain, Großkunzendorfer and Blanco Macael show dependencies according to the analysed sample direction.

Figure 4: Residual strain [mm/m], after heating – cooling cycles under dry and wet conditions for a) Lasa, b) Großkunzendorfer, c) Gioia and d) Blanco Macael.

For these four marble varieties, the acoustic emission was analysed. Figure 5 shows the acoustic emission activity, namely events per time, detected during the first two dry cycles and during the first wet cycle on a Blanco Macael marble.

During the first heating a significant AE activity was detected from a temperature of approximatley 35 °C up to the maximum temperature of 90 °C. A significant part of these events with a cumulated energy of 78 × 10⁵ energy Units (eU) can be related to cracking and crack growth. The onset temperature of the acoustic emission activity of 35 °C indicates that the materials have not been heated up to higher temperatures before. The effect that acoustic emissions do not occur on loadings lower than before was first described by Kaiser 1950 for different mechanically loaded materials. For thermally loaded marbles this so-called Kaiser effect was verified in our own preliminary studies. In the following dwell time, when the thermal equilibrium is reached, the AE activity decreases to nearly 189zero. During the first cooling back to 20 °C, the number of events is comparable to the number of events during the first heating. Also, the cumulated acoustic energy of 56 × 10⁵ eU has the same order of magnitude like during heating. These events may be related to additional cracking, probably due to crack misfit.

Figure 5: Acoustic emission activity, the red graph shows the time-temperature course, while the blue peaks represent the number of events.

In the next cycles a noticeable AE activity was detected during cooling only. The acoustic emission energy in these cycles is significantly lower, nearly zero. This indicates that the main damage occurs during the first heating-cooling cycle. This is verified by the ultrasonic velocity as an indicator for the integrity of the marble (Köhler 1991). For the Blanco Macael marble the ultrasonic velocity measured in the y-direction decreases from 5,530 m/s to 4,450 m/s (Figure 6). In the following dry cycles no significant change of the ultrasonic velocity was detected. These results are in good correlation with the measurements of the thermal strains. No significant further increase of the residual strain was measured after the first cycle under dry conditions.

No AE activity was detected during heating or cooling in the wet cycles. A possible explanation is that no additional cracking occurs, and the water lubricates the grain boundaries and therefore no emissions from friction can be detected. A noticeable number of events was detected only in the dwell time at 90 °C, but the cumulative energy of these events is low. A part of the water evaporated during the dwell time and the upper part of the specimen was exposed to the ambient temperature of 90 °C, whereas the water temperature reached only 80 °C. The source of the events can be localized in the zone of the falling water surface. Therefore, the cause for this AE activity might be thermal or hygric strain in the water surface zone. The ultrasonic velocity of 3,940 m/s measured after drying indicates only a moderate additional degradation of the marble. This contrasts with the significant increase of residual strain in the wet cycles. Accepting that no further cracking occurs, the only possible explanation is an increased residual crack width. As shown in Table 1, the maximum expansions of Blanco Macael in wet cycles are equal or slightly lower than in the dry cycles. Thus, water remains in the pores and keeps them open after cooling. This hypothesis needs to be verified in further studies, e. g. by drying tests and microscopy.

The other tested marbles behave similarly. The highest AE activities and the main loss of ultrasonic velocity were measured during the first heating-cooling cycle.

As shown in Table 2, AE energy and change in ultrasonic velocity are in correlation with the residual strains in the first cycle. The seriate interlobate grain structure of the Grosskunzendorfer in combination with interlocking grain boundaries causes a high AE activity with high energies, indicating the formation of a large number of microcracks.

Figure 6: Acoustic emission activity during heating and cooling and ultrasonic velocity of Blanco Macael (BM), Gioia (GI), Lasa (LA), and Großkunzendorfer (GK) marble. The wet cycles are marked blue.

190Table 1: Maximum expansion [mm/m] for each thermal expansion cycle under dry and wet conditions.

	1.	2.	3.	4.	5.
GK dry wet	0.71 0.56	0.73 0.62	0.75 0.65	0.67	0.69
BM dry wet	0.31 0.28	0.37 0.36	0.39 0.37	0.39
LA dry wet	0.46 0.46	0.48 0.52	0.49 0.55	0.56	0.57
GI dry wet	0.7 0.99	0.75 1.04	0.76 1.05	1.07	1.09

Table 2: Residual strains, corresponding change of ultrasonic velocity (in the y-direction) and acoustic emission ernergy in the first dry and first wet cycle.

	First dry cycle	First wet cycle
	∆VUS [m/s]	E*105 [eU]	Strain [mm/m]	∆VUS [m/s]	Strain [mm/m]
GK	850	515	0.20 ±0.10	NaN	0.40 ±0.20
BM	1080	78	0.20 ±0.05	480	0.40 ±0.20
LA	780	6	0.15 ±0.05	230	0.30 ±0.05
GI	630	3	0.20 ±0.00	90	0.30 ±0.00

Blanco Macael, Lasa and Gioia have a comparable grain fabric and grain boundary geometry. According to their texture, the thermal anisotropy of Blanco Macael with a pronounced preferred orientation of the calcite crystals is more distinctive than the anisotropy of Lasa. This causes more intensive cracking and therefore a higher AE activity. The Gioia marble has a history of prior outdoor exposure. The low ultrasonic velocity before the first heating and the low AE activity can be explained by preexisting damage.