Shear-induced Dilation Research Articles

Shear-induced fluid localization, episodic fluid release and porosity wave in deformable low-permeability rock salt

Understanding fluid distribution and migration in deformable low-permeability rock salt is critical for geologic disposal of nuclear waste. Field observations indicate that fluids in a salt formation are likely compartmentalized into relatively isolated patches and fluid release from such a formation is generally episodic. The underlying mechanism for these phenomena remains poorly understood. In this paper, a hydrological-mechanical model is formulated for fluid percolation in a rock salt formation under a deviatoric stress. Using a linear stability analysis, we show that a porosity wave (a train of alternating high and low porosity pockets) can emerge from positive feedbacks among intergranular wetting, grain boundary weakening and shear-induced material dilatancy. Fluid localization or episodic release can be viewed as a stationary or propagating porosity wave respectively. Fluid pockets transported via a porosity wave remain relatively isolated with minimal mixing between neighboring pockets. We further show that the velocity of fluid flow can be significantly enhanced by the emergence of a porosity wave. The concept and the related model presented in this paper provide a unified consistent explanation for the key features observed in fluid flow in rock salt. The similar process is expected to occur in other deformable low-permeability media such as shale and partially molten rocks under a deviatoric stress. Thus, the result presented here has an important implication to hydrocarbon expulsion from shale source rocks, radioactive waste isolation in a tight rock repository, and caprock integrity of a subsurface gas (CO2, H2 or CH4) storage system. It may also help develop a new engineering approach to fluid injection into or extraction from unconventional reservoirs.

Reliquefaction behavior of sand and response of pile group subjected to repeated shaking sequence using centrifuge model experiments

A series of centrifuge model experiments were performed in this study to investigate the sand reliquefaction behavior in free field and pile group models subjected to repeated shakings. Toyoura sand was used to prepare the model ground with 50 % relative density and experimented at 50g centrifugal acceleration. A 2 × 2 pile group model was used to examine the response of piles and its effect on sand reliquefaction during repeated shakings. A seismic sequence comprising foreshocks–mainshock–aftershocks–mainshock pattern was imparted to both free field and pile group models. In addition, an independent mainshock was imparted to a free field model and compared with the seismic sequence to examine the effect of foreshocks in triggering future liquefaction. Acceleration time response, excess pore pressure (EPP), ground subsidence, bending moment and lateral displacement of the pile group were measured. Significant de-amplification and unsymmetrical response with presence of shear-induced dilatancy spikes were observed in both free field and pile group models. Results from this study indicate that foreshocks and aftershocks are not sufficient to induce liquefaction. However, liquefaction and subsequent reliquefaction were recorded in both models during mainshocks. Pile group recorded smaller EPP response and subsidence compared to free field model. Higher magnitude of reliquefaction was induced even at moderate depths during the second mainshock at a quicker time. This was primarily associated with the increase in magnitude of anisotropy due to repeated generation and dissipation of pore water which results in decrease in resistance to further reliquefaction. Hardening induced around the vicinity of the pile group due to sand densification has resulted in smaller bending moment values during second mainshock compared to the first.

Temporal evolution of shear-induced dilatancy of rock fractures: controls from surface roughness and normal stress

SUMMARY Understanding the shear-induced dilatancy of rock fractures is important for assessing the permeability evolution and seismic hazard in shale and geothermal reservoirs. The displacement dependence of fracture dilation has been well studied, while the influence of slip velocity is poorly constrained. In this study, we combined displacement- and velocity-dependent aperture models to reproduce the transient shear-induced dilatancy of fractures in sandstone in 16 normal stress unloading tests. Our results show that the combined aperture model can describe the transient fracture aperture evolution during accelerating slip induced by normal stress unloading better than the model dependent only on slip displacement. Slip velocity could enhance the aperture increase on smoother fractures at lower normal stresses and higher slip velocities. Both the dilation factor and characteristic slip distance decrease with increasing normal stress and surface roughness, signifying reduced contribution of slip velocity to transient shear-induced dilatancy at higher normal stresses and surface roughness. The dilation angle increases with the increase of surface roughness, and this increase diminishes at higher normal stresses primarily attributable to more severe asperity wear. These findings highlight the importance of slip velocity in controlling the transient evolution of aperture and permeability of a rock fracture. Our study also provides constraints on the constitutive parameters in the combined aperture model for describing transient shear-induced fracture dilatancy. We suggest that it is crucial to incorporate the velocity-dependent aperture model to simulate the nonlinear evolution of fracture aperture in future analytical and numerical models involving coupled hydromechanical processes in geoenergy systems.

Experimental and modeling study of shear strength of unsaturated fine-grained soils with bi-modal water-retention characteristics

The mechanical behavior and strength characteristics of unsaturated fine-grained soils with dual-porosity are of crucial importance in geotechnical designs. Nanyang fine-grained soils have been selected as typical dual-porosity structure soils to perform experimental tests under a wide range of suction and different initial densities while studying its stress-strain-strength properties constitute the main scope of this study. Axial translation and vapor equilibrium techniques are jointly employed to apply a wide suction range. Our data suggest that soil behavior transits from strain-hardening with shear-induced contraction to strain-softening with shear-induced dilation as suction and density increase. By exploiting a bi-modal soil–water retention curve (SWRC) explicitly separating capillarity and adsorption mechanisms, the shear strength is allowed to be analyzed in the capillary suction stress-shear stress space. The strength envelop exhibits bi-linear characteristics. Building upon these findings, we propose a bi-linear shear strength criterion specifically for dual-porosity fine-grained soils. We utilize the obtained test data to evaluate existing strength criteria based on effective stress and dual stress variables that consider the bi-modal SWRC characteristics. The comparison indicates that the proposed bi-linear shear strength criterion can more reasonably represent the variation of shear strength under a wide range of suction for unsaturated dual-porosity fine-grained soils.

Frictional-dilatancy effect in the description of immersed granular motion

This study develops a dynamic model to better describe the frictional-dilatancy behavior of underwater granular motion. We employ the compressible Navier–Stokes equations as the continuum framework, and introduce μ(J) rheology in treating the constitutive law of the immersed granules. Within the compressible Navier–Stokes framework, the change in granular volume fraction that occurs when the granules undergo shear-induced volumetric dilation (contraction) is considered using the frictional-dilatancy law from soil mechanics. On introducing frictional dilatancy, the constant coefficient of friction at startup in μ(J) rheology, which governs the yielding limit of particles, is replaced by a particle-volume-fraction-dependent evolutionary variable. The proposed model enables an accurate description of the properties of quasi-static deforming granular mass. The validity of the model is verified by classical immersed granular collapse. A comparison with experimental and previous simulation results demonstrated that the introduction of frictional-dilatancy law delays the initiation of submarine granular flow.

Impact of stress-driven crack growth on the emergence of anomalous transport in critically connected natural fracture networks

We study the role of in-situ stresses in fluid flow and mass transport through a two-dimensional fractured rock. The fracture network is based on a real outcrop with a connectivity state around the percolation threshold. We simulate stress-dependent fracture propagation and deformation using a geomechanics model based on the hybrid finite-discrete element method. The hydrodynamic transport through the deformed rock mass is then modeled based on particle tracking simulations. By exploring various scenarios of different stress ratios and orientations, we systematically analyzed how the superimposed multiple geomechanical processes (i.e., compression-induced closure, shear-induced dilation, and stress-induced propagation) control the aperture patterns, flow fields, and transport behaviors of the fracture network. Our results show that an increased stress ratio tends to promote fracture shear dilation and thus enhance flow channelization, resulting in an anomalously early arrival of solute transport. Furthermore, the increased differential stress attempts to drive the growth of wing cracks that bridge pre-existing fractures and form new flow pathways. We observe a qualitative transformation of the flow structure from a disconnected to a connected regime accompanied by a significant alteration of its overall transport properties. This phenomenon is attributed to the superimposed effect of stress-driven crack propagation and shear-induced fracture dilation. By quantitative parameter analyses of global flow and transport properties, we further found that the breakthrough time median of particles is well correlated with the equivalent permeability of the fractured rock when the effect of new cracks is removed. However, this correlation is greatly reduced when fracture propagation is considered; instead, the breakthrough time shows a good correlation with mean tortuosity. Our findings have important implications for the strategy design of geothermal production and waste disposal in critically connected fractured geological media under various stress conditions.

Effects of surface roughness and shear processes on solute transport through 3D crossed rock fractures

The influences of surface roughness and shear processes on fluid flow and solute transport through three-dimensional (3D) crossed rock fractures, a vital element of fracture networks, were systematically investigated. Surfaces of tensile fractures created by splitting granite and sandstone samples along its two orthogonal central axes were optically scanned to generate rough-walled crossed fracture models. Shearing processes on the models were realized by assigning experimentally measured normal and shear displacements to one fracture while fixing the other. Fluid flow and solute transport through the models were numerically simulated taking into account different combinations of inlets and outlets, in which distilled water and solution are injected into the two inlets, respectively. The results show that compared to the parallel-plate model, the rough-walled crossed fracture model exhibits obvious flow channelization and fluid redistribution at the intersection, significantly promoting the mixing. The shear process affects the mixing at the intersection as it induces dilation and geometric change of the intersection. Increasing shear displacement can either enhance or reduce the mixing depending on combinations of the inlets and outlets, and the mixing ratio is controlled by the aperture difference between two outlet branches and the surface roughness. Effects of surface roughness, shear displacement and shear-induced dilation on the mixing ratio are quantified, upscaling of which can be potentially useful for field-scale characterization of solute transport in fractured systems.

The Drucker–Prager criterion-based plasticity theory of amorphous alloys under the complex stress states

The constitutive models of materials are pushed to their limits in a wide range of service scenarios, each with a unique stress state, strain rate, and temperature coupling. This paper successfully captured the stress state sensitivity of amorphous alloys using the free volume theory and the Drucker–Prager yield criterion. The varying yield surface and shear band angle as the dilation coefficient changes are well captured by the FEM simulations. The model can also accurately portray the influence of the stress state, as well as the difference of shear-induced dilation between the tension and compression. This model can facilitate the application of amorphous alloys in complicated local stress states at a large scale and structural level, such as in chiral mechanical metamaterials and gear structures.

Strain assessment of polyethylene pipes in dense sand subjected to axial displacement

Buried polyethylene pipes used in gas distribution systems can experience excessive wall strain when exposed to ground movement that can affect the performance of the pipes in service. This paper presents full-scale laboratory tests performed to investigate the responses of medium-density polyethylene (MDPE) gas-distribution pipes in dense sand when subjected to axial ground movement. Pipes buried in sand in a large test box were pulled at the rates of 0.5 mm/min, 1 mm/min, and 2 mm/min to simulate relative ground movement in the longitudinal direction. The test facility was instrumented to measure pulling force, pipe wall strain, and soil stress. The measured pullout force was significantly higher than predicted using the equations recommended in current design guidelines, which is attributed to the increase of normal stress on the pipe wall by shear-induced dilation of interface soil. The cavity expansion theory was successfully applied to calculate the normal stress increase. The distribution of measured strains was nonlinear along the pipe length. Assuming a parabolic distribution of the strains, simplified equations were developed to calculate pullout resistance and pipe wall strain from the relative ground displacement. The developed method reasonably predicted the pipe strain measured during the tests.

Defects identifier based on coarse grained equivalent stiffness gradient in 2D metallic glasses

Intensive parameters have focused on identification of defects in metallic glasses (MGs), a long-standing stymy due to the lack of long-range order. Rather than diagonalizing full Hessian matrix, here we introduce Coarse Grained Equivalent Stiffness (CGES), which is based on the determinant of local Hessian matrix (LHM), to describe heterogeneities of MGs. Atomic nonaffine displacements (ANDs) are determined by external loadings and internal heterogeneities, so that certain shapes of CGES gradient fields lead to shear-like or dilation-like ANDs, revealing the physical mechanism of local shear, shear-induced dilatancy and dilation. Further molecular simulations indicate that approximate spatial distributions of atoms, which would experience shear or volumetric deformation before yielding, can be estimated by initial CGES gradient fields. These results provide insights into the structural origin of anisotropic defects in MGs.

Strength and ductility enhancement of twin-roll cast Al-Zn-Mg-Cu alloys with high solidification intervals through a synergistic segregation-controlling strategy

Center segregation and banded intergranular segregation (B-IGS) should be well controlled to improve the mechanical properties of twin-roll cast (TRC) aluminum alloys, especially for alloys with high solidification intervals. In the present work, a synergistic grain refinement strategy was designed using an Al-5Ti-B grain refiner and Ti, Zr, and Sc microalloying elements to simultaneously control center segregation and B-IGS in TRC Al-Zn-Mg-Cu alloys. As the grain size decreased, center segregation defects were eliminated and transformed into dispersed B-IGS defects; simultaneously, the width and length of the B-IGS were also reduced. Moreover, the macro-distributions of the alloying elements along the thickness direction became more homogeneous due to a weak shear-induced dilation effect. The well-controlled multiscale segregation improved the uniformity of the alloy macrostructure, accelerated the redissolution of the crystalline phase, dispersed the aggregated residual phase, and refined grains under the T6 state. Hence, the strength and ductility of the alloys under the T6 state were simultaneously improved, and the hardness distribution along the thickness direction became more homogenous. Furthermore, the underlying mechanisms of segregation evolution and strength and ductility enhancements were elucidated. This work provides a novel strategy to effectively control multiscale segregation and produce high-performance aluminum alloys with high solidification intervals by TRC.

Shaking table tests on the liquefaction-induced uplift displacement of circular tunnel structure

Underground structures are susceptible to float and move upward during earthquakes when located in a liquefiable soil deposit. There are examples of this phenomenon in past major earthquake events. In this study, the uplift of circular tunnels in a liquefiable sand layer was investigated with a series of shaking table tests. The research has focused on the buried depth of the tunnel, tunnel diameter, tunnel weight, liquefaction extent, uplift mechanism, and factor of safety against liquefaction-induced uplift. According to the test results, the shallow buried depth, larger diameter, and lower weight can intensify the tunnel uplift, so the displacement in post-liquefaction time continues at the same rate as during the shaking time. Due to the shear-induced dilation, pore water pressure generation around the tunnel was reduced compared with that of the free field. The excess pore water pressure dissipation in the soil overlying the uplifted tunnel was significant, which leads to suction in the soil deposit. Furthermore, the acceleration response of overlying soil with the uplifted tunnel was similar to that of the free field. However, the soil acceleration response around the tunnel without uplift was similar to the base motion.

Granular crystals as strong and fully dense architectured materials

Dense topologically interlocked panels are made of well-ordered, stiff building blocks interacting mainly by frictional contact. Under mechanical loads, the deformation of the individual blocks is small, but they can slide and rotate collectively, generating high strength, toughness, impact resistance, and damage tolerance. Here, we expand this construction strategy to fully dense, 3D architectured materials made of space filling building blocks or "grains." We used mechanical vibrations to assemble 3D printed rhombic dodecahedral and truncated octahedral grains into fully dense face-centered cubicand body-centered cubic"granular crystals." Triaxial compression tests revealed that these granular crystals are up to 25 times stronger than randomly packed spheres and that after testing, the grains can be recycled into new samples with no loss of strength. They also displayed a rich set of mechanisms: nonlinear deformations, crystal plasticity reminiscent of atomistic mechanisms, geometrical hardening, cross-slip, shear-induced dilatancy, and microbuckling. A most intriguing mechanism involved a pressure-dependent "granular crystal plasticity" with interlocked slip planes that completely forbid slip along certain loading directions. We captured these phenomena using a three-length scale theoretical model which agreed well with the experiments. Once fully understood and harnessed, we envision that these mechanisms will lead to 3D architectured materials with unusual and attractive combinations of mechanical performances as well as capabilities for repair, reshaping, on-site alterations, and recycling of the building blocks. In addition, these granular crystals could serve as "model materials" to explore unusual atomic scale deformation mechanisms, for example, non-Schmid plasticity.

Geometry of sliding lamellae dictates the constitutive properties of nacre-like hierarchical materials

The mechanical properties of nacre (mother-of-pearl), which are superior to those of its constituent materials, derive from a micro-architecture formed by the arrangement into parallel laminae of individual aragonite tiles with surface asperities, bonded by organic interlayers. A mechanical model is developed to determine how the contact profile that defines the geometry of the tile surface affects the constitutive response of hierarchical metamaterials of this type when they are strained in the lamellar plane. We consider the structured character of the deformation, consisting of elastic distortion of the tiles and the inelastic contribution from the disarrangements produced by relative sliding of the lamellar interfaces, which is formulated within the thermodynamic framework for irreversible processes in generalized materials. The resistance to sliding associated with constrained shear-induced dilatation of the tiles and the bridging of the organic interlayers are represented by a generalized cohesive-frictional law, effective at the level of averaged separation plane, providing a phenomenological approach that by-passes the difficulty in analytically solving the contact problem at asperities. The nominal stress–strain curves of the nacreous structure are shown to be highly dependent, qualitatively and quantitatively, on the geometry of the contact profiles. The model not only produces results that are in excellent agreement with experiments on nacre, but it can also describe complex stress–strain curves with pinched hysteresis loops having branches of various shapes, as well as serrated inelastic regimes. The versatility of this theoretical approach offers promise as a design tool for innovative metamaterials.

Formation of banded intergranular segregation and control via micro-alloying in twin-roll casted Al–Zn–Mg–Cu alloy with high solidification interval

For twin-roll casting (TRC) aluminum alloys, research on segregation defects has mainly focused on center segregation. However, intergranular segregation has been rarely investigated. In this study, we found that the banded intergranular segregation was a distinctive feature in TRC aluminum alloys. The formation and coarsening mechanism of this banded intergranular segregation in TRC Al–Zn–Mg–Cu alloys were revealed for the first time. The results showed that the banded intergranular segregation was contributed to the shear-induced dilation as the cast-rolling force was gradually applied to the crystals in the semi-solid region. Macroscopic inverse segregation along the dendrite channels occurred under the strong shear-induced dilation, which resulted in significant coarsening of the banded intergranular segregation. Furthermore, a new path controlling the center segregation and the banded intergranular segregation simultaneously by micro-alloying was proposed, which provides a strategy to develop TRC aluminum alloys with a high solidification interval.

Impact of injection rate ramp-up on nucleation and arrest of dynamic fault slip

Fluid injection into underground formations reactivates preexisting geological discontinuities such as faults or fractures. In this work, we investigate the impact of injection rate ramp-up present in many standard injection protocols on the nucleation and potential arrest of dynamic slip along a planar pressurized fault. We assume a linear increasing function of injection rate with time, up to a given time t_c after which a maximum value Q_m is achieved. Under the assumption of negligible shear-induced dilatancy and impermeable host medium, we solve numerically the coupled hydro-mechanical model and explore the different slip regimes identified via scaling analysis. We show that in the limit when fluid diffusion time scale t_w is much larger than the ramp-up time scale t_c, slip on an ultimately stable fault is essentially driven by pressurization at constant rate. Vice versa, in the limit when t_c/t_w gg 1, the pressurization rate, quantified by the dimensionless ratio dfrac{Q_m t_w}{t_c Q^*} with Q^* being a characteristic injection rate scale, does impact both nucleation time and arrest distance of dynamic slip. Indeed, for a given initial fault loading condition and frictional weakening property, lower pressurization rates delay the nucleation of a finite-sized dynamic event and increase the corresponding run-out distance approximately proportional to propto left( dfrac{Q_m t_w}{t_c Q^*}right) ^{-0.472}. On critically stressed faults, instead, the ramp-up of injection rate activates quasi-static slip which quickly turn into a run-away dynamic rupture. Its nucleation time decreases non-linearly with increasing value of dfrac{Q_m t_w}{t_c Q^*} and it may precede (or not) the one associated with fault pressurization at constant rate only.

Static and dynamic behaviors of granular soil reinforced by disposable face-mask chips

The stock of disposable face masks has climbed steadily due to COVID-19, which results in an urgent worldwide environmental problem. This investigation aims to assess the potential of using waste mask chips (MC) as reinforcement in granular soil. To evaluate the mechanical properties of the mask chips - granular soil mixture (MSM), a series of monotonic and cyclic triaxial tests with different confining pressures and MC contents are conducted. Then, the shear behaviors of MSMs are quantified from both static (strength, stiffness, dilatancy) and dynamic (energy absorption capacity, resilient modulus, deformation) perspectives. Furthermore, based on the static and dynamic indexes, the effects of adding MC into granular soil are comprehensively analyzed. All experimental results indicate that the MSMs exbibit an increase of shear strength and a reduction of shear-induced volumetric dilation, but a decrease in stiffness. The addition of MC also leads to an increase in energy absorption but will not affect the cumulative strain of MSMs. Thus, MC has great potential to be used in the embankment construction of road and railway, backfill or reclamation construction, and so on. This experimental study provides a new perspective and thought on the recycling of waste face masks in civil engineering.

Performance evaluation of PM4Sand model for simulation of the liquefaction remedial measures for embankment

Non-linear seismic deformation analyses of an earth embankment founded on liquefiable soil were performed using the FLAC 2D and PM4Sand as the constitutive model to simulate the seismic behaviour of loose sand foundation. PM4Sand model was calibrated to the cyclic behaviour of Nevada 120 sand. The untreated embankment-foundation system was subjected to three sinusoidal shaking events. Though the computed deformations are in reasonable agreement with measured results, the excess pore pressures (EPP) in the central foundation region were found to be overpredicted. The temporal variation of sand permeability was considered for the simulation of untreated sand deposit supporting the embankment. The low effective stresses computed to be regained following the dissipation of EPP for the case with variable sand permeability implies that the numerical model is not capable of accounting the shear-induced dilation in the foundation soil underlying the embankment. So, the class-C1 simulations were performed to evaluate the numerical model by adjusting the dilation parameters in the model. Following this, the three different liquefaction mitigation measures such as dense sand columns within the foundation soil, gravel berms along the side slopes of the embankment, and vertical sheet piles below the embankment toes were simulated and comparisons with experimental data are reported.

Linear and nonlinear fluid flow responses of connected fractures subject to shearing under constant normal load and constant normal stiffness boundary conditions

Linear and nonlinear fluid flow behaviors of connected fractures during shearing under constant normal load and constant normal stiffness boundary conditions are numerically characterized solving the Navier-Stokes equations. The model consists of three fractures having an identical length of 200 mm and a width of 100 mm, among which the bridge fracture is sheared. Both smooth and rough-walled fracture surfaces are generated to investigate the effect of surface roughness on flow streamline distributions and the ratio of flow rate to hydraulic gradient. The results show that the permeability monotonically increases during shearing under constant normal load conditions, due to the shear-induced dilation of fractures. Under constant normal stiffness conditions, the permeability either increases or decreases during shearing, controlled by the competing role played by the shearing fracture and nearby fractures. As shear advances, the shearing fracture dilates and increases the permeability, whereas the surrounding fractures are compressed that decrease the permeability. The flow rate of rough-walled models is overestimated by approximately 45 ∼ 80% comparing to the smooth parallel plate model. For smooth models, the critical hydraulic gradient dramatically decreases with increasing the shear strain from 0.015 to 0.025 and then slightly fluctuates with the increment of shear strain from 0.025 to 0.05. The critical hydraulic gradient monotonically decreases during the shearing process for rough-walled models. The range of critical hydraulic gradient is between 0.001 and 0.02.

The highest potential transmissivities of fractures in fault zones: Reference values based on laboratory and in situ hydro-mechanical experimental data

The transmissivity of a fracture can be related to fracture roughness (JRC0), initial aperture (E0), effective normal stress (σ'n), and tensile strength (σt) of the intact rock, based on the Barton–Bandis model and their data, and the transmissivity (or E0) can increase by shear-induced dilation. Previous studies revealed that the transmissivities of fractures in fault zones, detected as flow anomalies (highly transmissive zones) during borehole investigations at six sites, decrease uniformly with an increasing effective mean stress normalized to σt. If this uniform change in transmissivity is explained by σ'n-dependent fracture-normal displacement following the Barton–Bandis model, those transmissivities represent the upper limit of transmissivities of fractures in fault zones that can increase by shear-induced dilation. To verify this possibility, the E0 of fractures was estimated using those transmissivities, σt, and possible JRC0 and σ'n. Then, using this estimated E0, the changes in transmissivity were simulated, varying σ'n. The results reproduced very well the observed uniform change in transmissivity. The estimated values of E0 are tens of micrometers to a few millimeters, which can occur by slight shear displacements (e.g., 0.05–2.00 mm) during shear-induced dilation, easily achievable in fault zones. Thus, the requirements for the highest transmissivities are slight shear displacements and no/limited fracture-sealing rather than large shear displacements. Transmissivities in fault zone fractures that have already reached the highest transmissivities do not change significantly by shear displacement, while the transmissivities of fractures sealed by mineral filling can increase by orders of magnitude, as confirmed by recent fault-stimulation-experiments.