PSI - Issue 41

G. Agalianos et al. / Procedia Structural Integrity 41 (2022) 452–460 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

455

4

For the detection of the PSCs, a pair of gold-plated electrodes was used. The electrodes were attached on the lower surface of the specimen on both sides of its central cross section at a distance of 50 mm (Figs.1, 2a). The specific arrangement was chosen since, according Triantis et al. (2017), it ensures strong PSC signals and minimization of the electric noise. The electrodes were electrically isolated using highly resistive Teflon plates and they were supported by pieces of flexible material in order to ensure good coupling with the specimen. The PSC values were recorded with the aid of a highly sensitive Keithley 6517A electrometer (capable of recording electric signals from 0.1 fA) using low noise cables to avoid any electromagnetic interference. It should be mentioned that the AE and the PSC techniques started recording at the same time instant. For most accurate synchronization, the outcome of the applied load channel was continuously recorded not only by the internal system of the loading frame but also by both the AE and PSC computers. 3. Results and Discussion As it was expected, all specimens were fractured at their mid-span (Fig.2b). Concerning the fracture load, Class A specimens failed at a load level of approximately 12 kN while Class B specimens failed at approximately 18 kN. Typical load-time curves are shown in Figs.3a,4a for a specimen of Class A and Class B, respectively. In general, the number of the acoustic hits recorded was significantly smaller in Class A specimens compared to the respective number of Class B specimens. In addition, Class A specimens produced lower PSC values as it will be discussed analytically.

(a) (b) Fig. 2. (a) The experimental setup. The pair of gold-plated PSC electrodes can be seen; (b) Photo of a fractured specimen. All six acoustic sensors can be seen in both photos. Fig.3 shows the experimental data for a typical specimen (porosity: 22.86%) of Class A. More specifically, Fig.3a presents the temporal variation of the recorded PSC signal, in juxtaposition to the applied mechanical load. It is seen that the load-time curve, after the inevitable bedding error (approximately for the first 200 s), is almost perfectly linear until the maximum load is attained (equal to about 13.7 kN). Based on the same figure, it can be said that the PSC shows a gradual increase until the time instant of about 800 s (corresponding to a load value about 10 kN or in other words until about 70% of the maximum load), where its maximum value is reached (equal to 4.9 pA). It is also noted that during the time interval from 570 s to about 830 s four consecutive short-lived fluctuations are observed. A qualitatively similar behaviour has been reported in three-point bending tests of marble (Pasiou et al., 2015) and cement based specimens (Stergiopoulos et al., 2015). After attaining its maximum value, the PSC decreases until the time instant just before the catastrophic fracture of the specimen. Then, the recorded PSC signal exhibits an abrupt increase attaining a value equal to approximately 7 pA. The temporal variation of the recorded PSC signal in juxtaposition to the respective amplitudes of the recorded AE hits is presented in Fig.3b. Only 59 hits were recorded during the specific experiment. This could be attributed to the high porosity of the specimen which blocks the acoustic waves from traveling and leads to many reflections of the waves on the specimen’s capillaries. However, when severe damages occur in the specimen, the acoustic wa ves succeed to reach the acoustic sensors. This is why in the beginning of the experiment (until the time instant ~500 s), only a few hits of low amplitude were recorded and a smooth increase of the PSC was observed. On the contrary, the majority of the acoustic hits as well as the most intense electrical activity is observed during the time interval between 600 s and 800 s, where the PSC fluctuations were recorded, indicating intensive microcracking activity. A period of