Xinglin Lei

Detailed analysis of acoustic emission activity during catastrophic fracture of faults in rock

Abstract The detailed time-space distribution of acoustic emission (AE) events during the catastrophic fracture of rock samples containing a pre-existing joint or potential fracture plane is obtained under triaxial compression using a high-speed 32-channel waveform recording system, and the results are discussed with respect to the prediction and characterization of catastrophic fault failure. AE activity is modeled quantitatively in terms of the seismic b -value of the magnitude–frequency relation, the self-excitation strength of the AE time series, and the fractal dimension of AE hypocenters. Consistent with previous studies on rock samples containing a fracture plane with several asperities, the present analyses reveal three long-term phases of AE activity associated with damage creation on heterogeneous faults, each clearly identifiable based on the above parameters. A long-term decreasing trend and short-term fluctuation of the b -value in the phase immediately preceding dynamic fracture are identified as characteristic features of the failure of heterogeneous faults. Based on the experimental results it is suggested that precursory anomalies related to earthquakes and other events associated with rock failure are strongly dependent on the heterogeneity of the fault or rock mass. A homogeneous fault or rock mass appears to fracture unpredictably without a consistent trend in precursory statistics, while inhomogeneous faults fracture with clear precursors related to the nature of the heterogeneity.

The hierarchical rupture process of a fault: an experimental study

Abstract We describe the detailed faulting process of a naturally healed fault containing geometric and mechanical asperities in a granitic porphyry sample, based on data collected with a high-speed acoustic emission (AE) waveform recording system. Asperity failure is examined using the detailed spatio-temporal distribution of AE hypocenters. The initial phase of AE activity is also examined using high dynamic range waveforms. Our experimental results indicate that quasi-static nucleation of the heterogeneous fault is associated with the failure of asperities on the fault plane. The fracturing of an asperity is characterized by a dense spatial clustering of AE events and a changing b -value ( b , hereinafter), which is manifest in three typical stages of failure as follows: (1) foreshocks exhibiting a decrease in b , (2) a period of mainshocks corresponding to a minimum in b , and (3) aftershocks of increasing b . The progressive fracture of several coupled asperities results in short-term precursory fluctuations in both b and AE rate. Furthermore, some AE events possess similar dynamic rupture features to those of earthquakes, having an initial phase associated with the transition from quasi-dynamic to dynamic rupture. We conclude based on these experimental observations that fault rupture has hierarchical characteristics. Quasi-static nucleation of fault rupture represents dynamic fracture of the asperities on the fault plane; likewise, a quasi-static nucleation process characterized by dynamic microfracturing precedes the fracture of an asperity. Since dynamic motions are easier to detect remotely than static deformations, understanding the hierarchical processes underlying fault rupture may thus be helpful for elucidating quasi-static nucleation at larger scales in terms of the dynamic rupture of the asperities at smaller scales. Careful studies of asperity failure in the lab may guide future seismic studies of large asperities on natural faults, potentially making it possible to recognize the final preparation stage before a large earthquake.

Compressive failure of mudstone samples containing quartz veins using rapid AE monitoring: the role of asperities

This paper presents the results of an ongoing experimental investigation of compressive failure of homogeneous and heterogeneous rocks. We used a rapid data acquisition system to monitor the spatio-temporal distribution of acoustic emission (AE) during fault nucleation under conditions of either constant-rate loading or static loading. In order to examine the effect of asperities on faulting, we conducted a series of experiments on mudstone with quartz veins. The bedding planes of the mudstone were oriented at an angle of 30° with the maximum compressive stress and were expected as the fracture plane. The quartz veins are much stronger than the bedding plane of mudstone and thus the samples model faults having strong asperities. Experimental results show that: (1) AE activity initiated close to the peak stress, and almost all AE hypocenters appeared at the intersections of the veins and the fault planes, suggesting that the vein asperities control faulting; (2) the b-value changed with time and shows multiple large and short-term fluctuations; (3) the change of the b-value correlated closely with the spatio-temporal hierarchy of the fracture process; and (4) fault segments along the bedding plane show behavior of slip having large compressive deformation before the peak stress, while the vein asperities show large precursor dilatancy prior to dynamic rupture. These experimental results are helpful for the understanding of seismic precursor phenomena associated with strain localization, dilatancy, change in level of ground water, as well as the b-value.

Effect of grain size on fractal structure of acoustic emission hypocenter distribution in granitic rock

Abstract Triaxial compression tests were carried out on Inada and Oshima granodiorites under constant loading rate conditions at a confining pressure of 50 MPa. Acoustic emissions (AE) were monitored and recorded using a multichannel system, and the spatial distributions of AE in the two samples were examined and compared as the experimental conditions were identical. Approximately 3000 and 1000 events were located in Oshima (fine-grained) and Inada (coarse-grained) granodiorites respectively, and the fractal dimensions of their spatial distributions were estimated by Grassbergers method. The AE hypocenter distribution of the fine-grained granodiorite showed a single fractal structure with a fractal dimension of 2.7, whereas that of the coarse-grained granodiorite showed a band-limited fractal structure. The fractal dimension of the spatial distribution of Inada granodiorite changed from 2.0 to 2.4 when the length scale r became nearly equal to the average grain size (5 mm) of the rock. These results imply that the spatial distribution of newly developed microcracks, which are the sources of AE and the scatterers of elastic wave, can be influenced by the grain of the rock.

Comparison of the microfracture localization in granite between fracturation and slip of a preexisting macroscopic healed joint by acoustic emission measurements

Experiments of fracturation and slip of a preexisting macroscopic healed joint have been performed under triaxial deformation on granite from Mayet de Montagne (France). This granite shows high grain-scale inhomogeneity. Acoustic emissions have been recorded and hypocenters have been determined during the entire experiments. For both rupture experiment and slip experiment, precursory localization of microfractures in the final rupture plane has been observed in the early stage of deformation, well before the dilatancy. It is likely that not only initial closure of favorably oriented cracks but also breaking of partially cemented grains or slipping between grains may occur in the pseudoelastic phase and are already localized on the final rupture plane where the shear stress seems to be concentrate. This behavior is observed in both cases where stress heterogeneity and rupture nucleation are controlled by (1) medium-scale heterogeneity at the grain scale (HS sample) or (2) macroscopic heterogeneity in the form of a preexisting healed joint (JS sample). The sample with the healed joint exhibited ~ 1.6 times more acoustic emission events than the intact sample. The presence of the healed joint significantly weakened the sample.

Effect of fault bend on the rupture propagation process of stick-slip

Abstract An experimental study of stick-slip is performed to examine the effect of a fault bend on the dynamic rupture propagation process. A granite sample used in the experiment has a pre-cut fault that is artificially bent by an angle of 5.6° at the center of the fault along strike, and accordingly the fault consists of two fault segments. The rupture propagation process during stick-slip instability is investigated by analyzing the records of shear strain and relative displacement measured with strain gauge sensors together with the hypocenters of AE (acoustic emission) events detected with piezoelectric transducers. The observed rupture propagation process of typical stick-slip events is as follows. (1) The dynamic rupture started on a fault segment is stopped near the fault bend. (2) The rupture propagation is restarted near the bend on the other fault segment 10.8 ms to 3.5 s after the stop of the first rupture. The delay time of the second rupture decreases with an increase in the slip amount of the first rupture or a decrease in the normal stress acting on the fault segment where the second rupture started. (3) The restarted rupture is not arrested by the presence of a fault bend, and slip occurs over the entire fault. We theoretically analyze the stress concentration near the fault bend to find that the normal stress produced by the preceding slip near the fault bend plays an important part in controlling the rupture propagation. A numerical simulation based on a rate- and state-dependent friction law is performed to interpret physically the retarded rupture in the experiment. The observed time interval of 10.8 ms to 3.5 s between the first rupture and the second is explained by the numerical simulation, suggesting that the rate- and state-dependence of rock friction is a possible mechanism for the retarded rupture on the fault.

Injection-induced fracturing process in a tight sandstone under different saturation conditions

To investigate the influence of local hydraulic condition on the water injection-induced fracture behavior in tight sandstone, injection tests under triaxial compression were conducted on samples collected from the Sichuan Basin, China. By means of acoustic emission (AE) monitoring, the fracturing behaviors of two samples of different water saturation patterns, partly saturated (the middle part was dry) and fully saturated, are investigated. Experiment results indicate that the local hydraulic condition plays a governing role in the mechanical properties, AE productivity, fracture nucleation and the geometry of the shear fracture zone. During axial loading stage, the partly saturated sample demonstrated ~35% higher elastic modulus than the fully saturated sample. During the injection-induced fracturing stage, progressively increasing AE activity and dilatancy, with increasing injected water volume, were observed preceding the dynamic fracture in the partly saturated sample, demonstrating a positive feedback between damage growing and fluid flow. More AEs were located in the dry or partially saturated regions at water front and thus produced dilatancy. However, the fully saturated sample shows very low AE activity (5% of that in partly saturated sample) which was initiated immediately before the dynamic fracture phase. Due to the very low permeability (~0.001 mD), volume of water injected into the fully saturated sample during the loading and creep stages is very limited, indicating the observed large dilatancy is governed by pore pressure increasing due to stress compaction. AE hypocenters demonstrated that an irregular shear fracture zone was created in the partly saturated sample, while a relatively flat shear plane was formed in the fully saturated sample.

The Pathway‐Flow Relative Permeability of CO2: Measurement by Lowered Pressure Drops

We introduce a simple method to measure the relative permeability of supercritical CO2 in low-permeability rocks. The method is built on the assumption of the stability of formed CO2 percolation pathway under lowered pressure drops. Initially, a continuous CO2 flow pathway is created under a relatively high-pressure drop. Then, several subsequent steps of lowered pressure drops are performed while monitoring the associated flow rates. When the pressure drop is lower than a threshold value, the created flow pathway is assumed to be adequately stable and does not vary significantly during successive flows, with the average saturation and flow rate achieving a quasi-steady state. The relative permeability of CO2 is then calculated from the relationship between the pressure drop and flow rate at several lowered pressure drops according to the extended form of Darcys law. We demonstrate this method using both numerical modeling and an experimental test using X-ray CT imaging. The results indicate the validity of the assumption for the stability of flow pathway under lowered pressure drops. A linear relationship between the lowered pressure drops and the corresponding CO2 flow rate is found. Furthermore, the measurement results suggest that the relative permeability of CO2 can still be high in low-permeability rocks if the CO2 saturation is higher than the threshold value required to build a flow pathway. The proposed method is important for measuring the pathway-flow relative permeability of non-wetting fluids in low-permeability rocks.

Effect of tidal stress on fault nucleation and failure of the 2007 M s 6.4 Ning'er earthquake

Based on calculations of the tidal Coulomb failure stress and investigations of the correlation between the Earth tide and the Ning’er earthquake sequence, the processes of fault nucleation and failure were simulated. In these simulations we consider the influence of tidal stresses using the rate- and state-dependent friction laws. Furthermore, the effects on tidal triggering due to the stress amplitude and periodic oscillation properties were investigated, and the triggering effects between the tidal normal and tidal shear stresses were compared. The results showed that the Ning’er earthquake sequence was a physical consequence of tidal effects. A transition period T0 exists between the nucleation and failure processes of a seismic fault. When the period T of stress is equal to or becomes larger than T0, the fault response becomes dependent on the periodic features of the loading stress; however, for T < T0, the response of the fault is nearly independent of the period. Both the tidal normal and tidal shear stresses have similar effect in the nucleation and failure processes; the clock changes generally increase with the maximum amplitudes of the tidal stresses. Tidal normal and tidal shear stresses with positive amplitudes mainly induce earthquake triggering; however, the triggering effects induced by negative tidal stresses are smaller and faults are not sensitive to negative tidal stresses. Our results primarily reveal the physical mechanisms of tidal stress triggering.

Response of Velocity Anisotropy of Shale Under Isotropic and Anisotropic Stress Fields

We investigated the responses of P-wave velocity and associated anisotropy in terms of Thomsen’s parameters to isotropic and anisotropic stress fields on Longmaxi shales cored along different directions. An array of piezoelectric ceramic transducers allows us to measure P-wave velocities along numerous different propagation directions. Anisotropic parameters, including the P-wave velocity α along a symmetry axis, Thomsen’s parameters ε and δ, and the orientation of the symmetry axis, could then be extracted by fitting Thomsen’s weak anisotropy model to the experimental data. The results indicate that Longmaxi shale displays weakly intrinsic velocity anisotropy with Thomsen’s parameters ε and δ being approximately 0.05 and 0.15, respectively. The isotropic stress field has only a slight effect on velocity and associated anisotropy in terms of Thomsen’s parameters. In contrast, both the magnitude and orientation of the anisotropic stress field with respect to the shale fabric are important in controlling the evolution of velocity and associated anisotropy in a changing stress field. For shale with bedding-parallel loading, velocity anisotropy is enhanced because velocities with smaller angles relative to the maximum stress increase significantly during the entire loading process, whereas those with larger angles increase slightly before the yield stress and afterwards decrease with the increasing differential stress. For shale with bedding-normal loading, anisotropy reversal is observed, and the anisotropy is progressively modified by the applied differential stress. Before reaching the yield stress, velocities with smaller angles relative to the maximum stress increase more significantly and even exceed the level of those with larger angles. After reaching the yield stress, velocities with larger angles decrease more significantly. Microstructural features such as the closure and generation of microcracks can explain the modification of the velocity anisotropy due to the applied stress anisotropy.