PSI - Issue 10

D. Triantis / Procedia Structural Integrity 10 (2018) 11–17 D. Triantis / Structural Integrity Procedia 00 (2018) 000 – 000

12

2

processes, crack growth, dislocations movement, inclusions cracks, etc, constitute the AE sources. Thus the AE technique is developed and further improved in order to be used as a valuable tool for the monitoring and under standing of the mechanisms of dynamic processes, and warning of the upcoming failure (Rao and Lakschmi (2005); Colombo et al. (2003); Cox and Meredith (1993)). Fracturing process involves nucleation, growth and micro-cracks coalescence to the final breaking. The state of stress on rock materials generates a field in the specimen under study. The AE are mostly produced by the growth of microcracks. Generally, the AE are generated, at different spatial and temporal scales. It must be noted that the release of seismic waves related to an earthquake and the AE released from solids during fracture process have similarities. The initial link between AE and seismology was attempted by Ono and Ohtsu (Ohtsu et al. (1991); Ono (1993); Ohtsu (1994)), by transferring earthquake data processing techniques to AE data processing, resulting to the adjustment of several seismological techniques to the field of civil engineering. During the AE, different fracture modes generate different types of AE signals, of various frequency ranges and amplitudes. In general, micro-cracks generate a large number of small amplitude AE, while macro-cracks generate fewer events, but of higher amplitudes. A popular method used for damage quantification of a material under mech anical loading is the b-value analysis. The b-value was originally defined in seismology and then it was expanded and used in the AE signals in engineering materials (Colombo et al. (2003); Kurz et al. (2005)). The method of analysis of b-value has been recently modified using statistical values such as mean and standard deviation of each amplitude, and the most modern method is known as “improved b - value” (I b -value) (Rao and Lakschmi (2005)). In parallel with the study of the I b -value, two additional AE parameters have been used to analyze the failure mode: the average frequency (AF) of the AE signals that is defined as the number of counts divided by the duration, and the parameter RA (Rise Time/Amplitude) (Ohtsu and Tomoda (2008); Aggelis et al. (2011)). It is generally accepted that cracks generating ΑΕ signals of relatively high AF and lower RA are related to tensile cracking mode (Mode I), while shear mode cracking (Mode II) corresponds to lower AF and high RA. In this work, the acoustic emission (AE) study focuses on the change of key ΑΕ parameters, such as the b-value, the RA (Rise Time/Amplitude) and the AF (Average Frequency), when quasi-brittle materials as marble, are subjected to uniaxial compression. Emphasis is placed on the ΑΕ recorded near failure, when different types of AE signals with varying frequency ranges and amplitudes are observed. 2. The material and experimental process A prismatic sample of Dionysos marble, of dimensions 40 mm x 40 mm x 100 mm was used. Detailed description of the mechanical and physical properties of the specific material can be found in previous papers by Kourkoulis et al. (1999, 2010) and Pasiou and Triantis (2017). Initially, the strength of similar samples against uniaxial compression was checked, and failure was observed within a range from 62 MPa to 65 MPa. Preliminary tests conducted on similar specimens, extracted from the same rock volume, show that the transition from the linear to the non-linear region of the mechanical behavior of the material occurs at about 70% to 75% of the fracture stress. Mechanical loading was applied during four distinct stages until failure, as follows: During the first stage (A), the stress increased at a constant rate of 0.44 MPa/s up to 60.5 MPa. A second loading stage (B) of duration exceeding 100 sec followed, during which the stress remained constant at 60.5 MPa. Given that at the end of stage B the AE activity was very limited, a slight increase of the stress by 3 MPa followed (stage C), at the same rate (0.44 MPa/s). During the last stage (D), the stress was kept constant at 63.5 MPa. The sample failed after 60 sec. The continuous temporal recording of the applied stress during all four stages is depicted in Fig.1. An acoustic sensor recording the AE hits was placed in the middle of the side surface of the sample and coupled on the specimen using vacuum grease (see Fig.2a). One preamplifier with 40 dB gain with analogue band-pass filters in the range of 20-400 kHz was also used. The equipment and the software used were by Mistras Group, Inc.

3. Experimental results and discussion

During the whole experimental procedure up to the fracture of the specimen, 3335 ΑΕ hits were recorded with amplitudes equal or higher than 40 dB. After collecting the AE data, filtering work was conducted on the recorded AE hits, aiming at excluding hits probably attributed to noise. In this context, AE hits of less than 10 ms duration and count less than 2 were excluded (Calabrese et al. (2012)).