PSI - Issue 55

Dimos Triantis et al. / Procedia Structural Integrity 55 (2024) 185–192 Triantis et al. / Structural Integrity Procedia 00 (2023) 000 – 000

188

4

The second protocol considered here included uniaxial compression tests of small intact prismatic specimens made again of Dionysos marble. As previously, the load was applied quasi-statically and monotonically up to the fracture of the specimens. The ultimate compressive strength (UCS) for the typical specimen that will be analyzed here was equal to 54 MPa. Additional details about the specific protocol are provided by Stavrakas et al. (2019). A total of N = 122 acoustic events were recorded during the specific experiment. Focusing again on the last loading stages, the temporal evolution of the load applied during the last 300 seconds of the experiment is plotted in Fig.2a together with the energy of each one of the acoustic events recorded.

55

1E+07

1.00

0.75

45

1E+05

0.50

35

1E+03

0.25

Axial stress [MPa]

25

1E+01

0.00

0.01

0.1

1

10

100 1000

0.01

0.1

1

10

100 1000

t [s]

(a) (b) Fig. 2. Uniaxial compression of intact prismatic marble specimens: (a) The temporal variation of the load applied during the last loading steps of a typical test of the second protocol, together with the energy of each one of the acoustic events recorded; (b) The temporal variation of the normalized load applied in juxtaposition to the respective one of the normalized average energy of the acoustic events. The temporal variation of the quantity of interest, i.e., of the normalized average energy of the acoustic events is plotted in Fig.2b, in juxtaposition to the respective evolution of the normalized load applied (L*). The similarity of the plots of this figure with the respective ones of Fig.1b is at least striking. Several seconds before the fracture of the specimen, and as the load approaches its maximum value (i.e., after the time instant t f - t ≈26 s, namely 26 seconds before the fracture of the specific specimen) the normalized average energy of the acoustic events exhibits again an abrupt increase (from about 0.20 to about 0.60) and afterwards it is trapped within a very narrow “window” ranging now in the interval 0.60< * i E <0.76. A clearly distinguishable plateau is again formed, which is now terminated at the time instant t f - t ≈0.3 s, namely three tenths of a second before the specimen’s abrupt disintegration. The response of the remaining specimens of this protocol is quite compatible to the one just described. Comparative consideration of the two protocols indicates interesting similarities between the temporal evolution of the average energy of the acoustic events. At the early loading steps, it is relatively low exhibiting either a slightly increasing trend or a stabilization tendency. Then, several seconds before the load attains its maximum value the average energy of the acoustic events exhibits a sudden jump to a quite higher level (in fact its value becomes almost three times higher) and it tends to be stabilized with some minor fluctuations. This stabilization tendency is terminated just a few tenths of a second before the fracture of the specimens. 4. Discussion: The case of structural tests The applicability of the conclusions drawn in previous section (from the analysis of data recorded from experiments with small scale specimens) in case of large structural elements will be studied in this section. In this direction, the acoustic activity developed in an accurate copy (scale 1:3) of a restored epistyle of the Parthenon Temple is analyzed, in terms of the average energy of the acoustic events. The epistyle tested consists of two asymmetric marble fragments, joined together with the aid of three pairs of threaded titanium bars (shown schematically in Fig.3) and suitable cement paste. The epistyle was prepared by experienced technicians of the Acropolis worksite. It was submitted to monotonic