This paper reviews the development history of permeability tests for measuring the hydraulic constants (hydraulic conductivity and specific storage) of porous materials in a laboratory, and systematizes the theories of the methods. The permeability test originated from sand filtration experiments carried out by Darcy in the 19th century. Techniques for measuring hydraulic constants were then devised and, at the same time or successively, the theory was established. This paper presents the exact analytical solutions of six methods, including the constant head, falling head, flow pump, and transient pulse methods, with the same initial condition and different boundary conditions. Based on the analytical solutions, we demonstrated the variations in the hydraulic head distribution within a specimen and the variations in the sensitivity coefficients of the hydraulic head difference between both ends of a specimen in terms of the hydraulic constants and compressible storage at the upstream and downstream sides of the equipment during tests of the six methods. The results were used to characterize the test methods, in terms of the experiment time, experiment precision and error, and method selection.
Uniaxial tension cyclic loading and creep tests were carried out on Inada granite aiming to clarify effects of stress amplitude and loading rate on fatigue behavior. Relationship between fatigue life, deformation behavior and AE activity were also examined. It was found that time to failure in cyclic loading test became shorter than that in creep test for low applied stress. This indicates stress amplitude has a significant effect on fatigue failure. The amount of fatigue damage per cycle increased with decrease in loading rate when the same maximum stress was applied. This is thought to be cause of decrease in fatigue life with decrease in loading rate. On the other hand, the amount of fatigue damage per cycle and fatigue life did not depend on loading rate when the same maximum stress ratio, which was normalized by the uniaxial tensile strength at each loading rate, was applied. Fatigue life of a specimen showing less strain increment tended to be longer and the relationship between the minimum value in the strain increment and fatigue life was represented by a line regardless of loading rate and maximum stress. AE event rate as well as strain increment was inversely proportion to life expectancy before failure. Ratio of AE event rate in unloading to loading began to increase just before failure. These results show that strain increment and AE event are useful parameter to predict fatigue failure.
In order to investigate the influence of water vapor pressure of surrounding environment on mode I fracture toughness of rock, a series of Semi-Circular Bend (SCB) tests under various water vapor pressures were conducted. The water vapor is the most effective agent which promotes stress corrosion of rock. The range of water vapor pressure used in this paper was from 10-3 to 103 Pa. The rocks used in this paper were African granodiorite (AG), Korean granite (KG), Kimachi sandstone (KS), Kumamoto andesite (KA) and Kunum basalt (KB). Measurement of elastic wave velocity and observation of thin section of these rocks were performed to make clear the micro structures, namely distribution of inherent micro cracks and grains. The results of the SCB test show that the fracture toughness of AG, KG, KA and KB were dependent on the water vapor pressure of the surrounding environment and decreases with increasing the pressure and that the fracture toughness of KS was independent on the pressure. It is considered that this decrease is due to stress corrosion promoted by water vapor. The experimental results also show that the degrees of influence of the water vapor pressure on the fracture toughness for AG was greater than the others. Together with the results and observed micro structures, it was concluded that the degree of the influence is dependent on the distribution and the density of inherent micro cracks.
Understanding time-dependent deformation of frozen rock is significant for the assessment of long-term stability of rock slopes in cold region. In this study, to clarify the time-dependent deformation of frozen rock and influence of water content on the deformation, a series of uniaxial compression tests and creep tests were carried out on Shikotsu welded tuff under the dry and wet conditions at -20℃ (-4°F). The uniaxial compressive strength (UCS) of frozen wet specimens was approximately equal to that of frozen dry specimens at the strain rate below 4.2×10-6/s. However, the former was greater than the latter at higher strain rates. The creep life of frozen wet specimens was longer than that of frozen dry specimens at the stress level above 13.4 MPa. On the other hand, the former was shorter than the latter at lower stress levels. Deformation of frozen wet specimens was larger than that of frozen dry specimens. The frozen dry and wet specimens showed different axial strain rate-axial creep strain curves. These differences could be explained by the presence of pore ice. Therefore, water content of rock should be considered for the assessment of long-term stability of rock slopes in cold region.
In this study, a difference between mechanical properties of water saturated and CO2 injected mudstone were investigated. Triaxial experiments, which simulated pressure and temperature at 1000 m deep underground, were conducted under drained and undrained conditions. Specimen were taken from Otadai mudstone Formation which is thought to be suitable for a caprock of carbon dioxide capture and storage reservoir. Experimental results were analyzed using poroelastic theory. Results are summarized as follows. (1) Under the both drained and undrained condition, Young's modulus and Poisson ratio of CO2-injected specimens were smaller than that of water saturated specimens. Triaxial strength of CO2-injected specimens were slightly higher than that of water saturated specimen. (2) Mineral particles do not seem to be dissolved since no significant differences in mechanical properties were observed between CO2-injected and N2-injected specimens. (3) Peak stress of water saturated specimens became larger by higher pore pressure than that of CO2-injected specimens.
In general, granite has anisotropy due to the preferred orientation of pre-existing micro-cracks. It is possible to obtain the information of the preferred orientation of micro-cracks by the measurement of the P-wave velocity. Therefore, by measuring P-wave velocity in various directions, we can obtain the information of three-dimensional distribution of micro-cracks in granite. Assuming that the three-dimensional distribution of micro-cracks in granite was caused by the stress release during coring, it can be possible to assess the stress state at the location of coring in the underground considering the relationship between three-dimensional distribution of P-wave velocity and microcracks. Differential Strain Curve Analysis (DSCA) is one of the methods to assess the stress state in the underground using rock core samples based on the initiation of micro-cracks due to the stress release. Therefore, if the three-dimensional distribution of P-wave velocity can be related to the orientation of micro-cracks determined by DSCA, the stress state in the underground can be evaluate from the measurement of P-wave velocity. In this study, by conducting both P-wave velocity measurements in various directions with a polyhedral specimen and DSCA using granite samples obtained from the same location, the relationship between the three-dimensional distribution of P-wave velocity and that of the crack parameter determined from DSCA was investigated. In addition, we considered the possibility to assess the stress state in the underground from P-wave velocity measurement. From the above measurements, a negative correlation between the P-wave velocity and the crack parameter from DSCA was observed for granite. Additionally, by using the value of the P-wave velocity which is corresponding to the crack-free granite rock, we could show the possibility to assess the stress state in the underground.
Stress-measurement methods based on the overcoring method assume that the rock mass is linearly elastic, isotropic, continuous, and homogeneous. However, a rock mass is actually anisotropic to some degree. In this study, we carried out theoretical, numerical and experimental studies to apply the Downward Compact Conical-ended Borehole Overcoring technique to an orthotropic rock. Numerical experiments with the use of a 3D-FEM analysis were conducted to confirm the applicability and efficiency of the proposed measurement theory.The results showed that the measurement includes a non-negligible error if we do not consider the anisotropy of rock when such anisotropy is strong.For example, the isotropic assumption method gives an error of more than 20 % in the estimation of stress when the degree of anisotropy in the direction of the maximum applied stress exceeds 20 %. On the other hand, the error is less than 5 % in the orthotropic assumption method. Finally, laboratory experiments for four kinds of orthotropic rocks using a true-triaxial compressive apparatus were carried out to verify the efficiency of the proposed measurement theory. These experiments also verified that the proposed method is suitable for use in orthotropic rock.
To clarify the effects of the viscosity of the hydraulic fluid and the weak planes of the rocks on the fracture creation, the hydraulic fracturing experiments were conducted on 15 cm-cubic Inada granite and Ogino tuff with the fracturing fluids of water and the super critical carbon dioxide. The borehole was bored in the perpendicular direction of the rift or bedding planes. Two horizontal stresses of 5 and 3 MPa and the vertical stress of 1 MPa were applied to the specimens. The vertical single fracture along the borehole was created in the specimen by the injection of water, while multiple complicated fractures along the rift or bedding planes were tended to be created by the injection of the super critical carbon dioxide. The numbers of the fracture branches obtained for the super critical carbon dioxide were greater than those obtained for the water. These results suggest that both the viscosity of the hydraulic fluid and the weak planes of the rocks significantly influence the formation of hydraulic fractures.
The 3-D weakness plane model was developed to clarify the mechanisms of influences of various stress states on the strength of rocks. The model assumes numerous planes of weakness whose directions are uniformly distributed in a rock; each plane slips or opens based on the Coulomb criterion with a tension cut-off; and the rock is regarded as failed when the ratio of the failed plane number to that of all planes reaches a certain value. The equal strength parameters were assigned to all planes and no complicated statistical functions are used. The model was applied to true triaxial compression, uniaxial tension, Brazilian and extension tests. The model was very simple but the effects of a stress state, namely, nonlinear increase of peak stress with small intermediate principal stress and nonlinear decrease with large intermediate principal stress in true triaxial tests, larger tensile strength by Brazilian tests than uniaxial tensile tests and the bilinear nature of peak stress in extension tests, were simulated very well. Distributions of stress and failed planes on the Schmidt net increased the understanding of the mechanism of the stress state effects on rock strength.
The transient pulse method is suitable for measuring hydraulic constants of low-permeability rocks in the laboratory. This method is established and reliable; however, there is room for improvement. In this study, we examined analytical solutions of the transient pulse method and introduced a versatile technique of precise test data analysis with sensitivity analysis and experimental error estimation that is based on the nonlinear least-squares method and is applicable to other permeability tests. We also carried out highly precise transient pulse permeability tests in the laboratory and obtained highly precise experimental data. Based on the results, we validated our proposed data analysis technique based on nonlinear least squares. We conducted transient pulse permeability tests using Inada granite, which is a low-permeability rock, and evaluated its hydraulic constants (hydraulic conductivity and specific storage) using the exact solution of the transient pulse method. We estimated the experimental error between the analytical solution and the experimental data simultaneously. When the approximate solution is used, we can increase precision by considering the compressible storage of the upstream and downstream reservoirs and tubing, which is estimated by a calibration test, instead of considering only water compression. Sensitivity analysis revealed that the sensitivity of the hydraulic head difference between the upstream and downstream reservoirs in terms of hydraulic conductivity is an order of magnitude greater than that in terms of specific storage. This means that the relative error of hydraulic conductivity obtained in a laboratory test is an order of magnitude less than that of specific storage. In precise experiments, control of the room temperature is important because the fluids used in pore and confining lines have thermal expansion characteristics that are sensitive to temperature variation, and the differential pressure transducer is also directly affected by temperature change. The pressure pulse should be as small as possible because if a large change in pressure occurs, the effective stress state of the specimen changes and the permeability changes with it.
The fluid flow mechanism in rock fracture is one of the basic elements in the coupled Thermal-HydraulicMechanical-Chemical (THMC) process, which is an important factor for underground rock engineering, especially for the safety assessment of radioactive waste repository. Fluid flow through a rock fracture is usually simplified as a parallel-plates model, which however, brings discrepancies against experimental and numerical simulation results in some contents, mainly due to the surface roughness of rock fracture and the inertial effect of fluid taking placing at high Reynolds numbers. JRC (Joint Roughness Coefficient) is one of the most extensively adapted parameter to quantify the surface roughness of rock fractures, however, the relation between mechanical aperture and hydraulic aperture of fractures developed based on JRC doesn't fit well with experimental and numerical simulation results. In this study, fluid flow through JRC profiles was simulated by solving Navier-Stokes equations, and the flow mechanism in fractures with different roughness was investigated. Based on the simulation results, an equation linking the relation between mechanical aperture and hydraulic aperture, with parameter Z2 which describes the geometrical characteristics of a rough surface and Reynolds number was established. This empirical equation takes into account both of the surface roughness and inertial effect, which can help estimate the fluid flow behavior of rock fractures in 2-D problems.
In radioactive waste disposal, tunnel passing through the ground with high geothermal region, and hot water temporary storage system, the rock mass around the openings will be influenced by coupled chemo-mechanical behavior induced by high temperatures. Then, rock fractures may be hydraulic weakness on low-permeability rock mass. Therefore, the coupled behavior should be examined to evaluate hydraulic property of rock fractures. In this study, flow-through experiments on a single fracture in granite have been carried out under confining pressure and temperature conditions controlled. The fracture permeability monotonically decreased with time at room temperature, and reached a quasi-steady state. Then, after the temperature was raised to 90 ºC, the permeability decreased again throughout the rest of the experiments. The elemental concentrations in fluid samples taken from the outlet were evaluated by ICP-AES. The elemental concentrations increased with increase of temperature. Thus, this behavior should be attributed to coupled chemo-mechanical effects. Meanwhile, another suite of sustained loading experiments using the same granite has been conducted under controlled temperature, confining pressure and mineral dissolution conditions for the purpose of evaluating the deformation behavior of rock fractures. From the tests results, it is found that the difference in compression behavior of the fractures is very small for all the conditions. However, particularly, it is found that displacement of the fractured rock sample under the wet and high temperature conditions is larger than those under the other conditions. The reason for this may be due to microscopic failure and mineral dissolution at the asperity contact areas.
The practical use of rock cavern storage for liquefied fuel (liquefied natural gas, liquid hydrogen, and DME) requires a heat conduction analysis for predicting the temperature distribution in rock mass near the cavern. This analysis is performed by using values for the thermophysical properties (thermal conductivity, thermal diffusivity, and specific heat) of rock mass containing ice. In our research, we attempted to measure the thermal diffusivity of frozen-state wet rock (water-saturated rock) using an optional heating method, which is an unsteady method for measuring thermal diffusivity, and verified the usefulness of this method. Next we investigated the impact of cooling on rock's thermal diffusivity by measuring the thermal diffusivity of rock samples with different porosities. Our results revealed the following observations. Measurements for Ogino tuff, Kimachi sandstone, and Himekami granite showed that the frozen rock's thermal diffusivity at -20℃was 0.51 mm2/s for the tuff, 0.66 mm2/s for the sandstone, and 1.51 mm2/s for the granite. The optional heating method can be used to measure the thermal diffusivity with an uncertainty of approximately ± 5％ (based on measuring the rock sample three times). Furthermore, a comparison between frozen rock and wet rock revealed that the freezing of pore water caused the thermal diffusivity of sandstone and granite to increase by a factor of 1.4. On the other hand, no variations were observed in the thermal diffusivity of tuff due to the freezing of pore water.
In order to gain further insight into changes in the properties and microstructures of granite due to temperature changes, two series of granite samples with different granularities from Aji, Takamatsu City, Kagawa Prefecture, Japan, were subjected to a one-cycle heating and cooling test with maximum temperatures ranging from 200 to 600℃ at heating and cooling rates of 50℃/h. Properties reflecting the internal microstructures, such as the size, weight, effective porosity, and P-wave velocity, were measured before and after the thermal cycle, and microstructures were visualized and observed using the fluorescent approach. The development of microcracks induced by the thermal test was then investigated with an image analysis technique. No significant differences in the changes in size and weight were identified within samples heated to 500℃, but a marked change was identified in the samples heated to 600℃. Increases in the effective porosity and decreases in the P-wave velocity were clearly identified. The effective porosity changed markedly from 500 to 600℃, while the P-wave velocity changed gradually to 600℃. These suggest that changes in the inner structure occur within samples. Microcracks developed in the tested samples, and this tendency was marked in samples treated at higher temperatures. These microscopic observations agree well with the observed changes in properties. Furthermore, a difference in the crack growth pattern between the two series of granite was observed. In the fine-grained sample, grain boundary cracks strongly developed between 500-600℃, while, in the medium-grained one, this tendency was not identified. It was inferred that the phase transition of quartz at 573℃ markedly influenced crack development. On the other hand, no marked differences due to the granularities were identified from 200 to 500℃.
Water inrush is one of the crucial risks encountered by engineers in mining practices. An effective risk assessment of water inrush in a mining excavation requires thorough investigation and understanding of the hydrogeological properties of the host rock mass, in which, the feature (position, volume, etc.) of water-bearing rock structure is a key issue. Since the temperatures of groundwater and surrounding rock masses are usually different from each other, the temperature difference may serve as an indicator for investigating the properties of water-bearing rock mass, and to assess the risk of water inrush during mining. In this study, the temperature variation behavior of a rock mass containing a water-bearing structure during a mining process was investigated through a laboratory model test. The monitoring system consists of a set of Fiber Brag Grating (FBG) temperature sensors and multi-point displacement meters. Experimental results indicate that the variation of the seepage intensity during the mining excavation can be effectively monitored by Pe (Peclet number), which exhibits close relations with the temperature field of rock mass, as well as the instability of rock mass structure induced by the water inrush.
Grouting is a widely used method for sealing fractured rock masses around underground structures to reduce or stop groundwater inflow. One important aspect is the grout penetrability. So far, many experimental studies have been done to detect the penetrability of grout, which depends on the grout injection pressure, aperture of fracture from a borehole, distribution of cement particle and etc.However, the filtration and penetration mechanism of cement-based grout have not been clarified sufficiently yet due to complicated physical and chemical processes of grout, such as pressure-dehydration, consolidation, bleeding, clogging, absorption, sedimentation and condensation etc. To better understand the penetration and filtration mechanism of cement-based grout through rock fractures, a two-dimensional numerical model of coupled Computational Fluid Dynamics and the Distinct Element Method (CFD-DEM) had been developed. The fluid flow was simulated numerically by solving the Navier-Stokes equations using finite deference method. The calculated flow velocity field was used to the cement particle movement solved by DEM applying Newton's laws of motion. For the interaction between fluid and cement particle, immersed boundary method was used. By using a direct numerical simulation technique, the interaction between fluid and particles can be evaluated.In this study, the laboratory injection test (called“ short slot” experiment) performed in Royal Institute of Technology (KTH) in Sweden was simulated to verify the applicability of the newly developed CFD-DEM code. As a result, the simulation results agree qualitatively well with the actual experimental results, and the clogging process during the injection of cement-based grout was successfully reproduced by the CFD-DEM code. Moreover, based on these results, the influence of the particle size distribution of cement on the filtration and penetration behavior was discussed. The presented numerical model in this paper gives a better understanding for filtration and penetration mechanism of cement-based grout.
Beachrocks are coastal deposits cemented mainly by calcium carbonate cement; these deposits are found in the tidal zone of sandy beaches in tropical and subtropical regions. Manmade beachrocks have the potential to inhibit coastal erosions; considering this important application, we performed field investigations and laboratory tests to understand the formation mechanisms of beachrocks in Okinawa and Ishikawa, Japan. We performed a needle penetration test, determination of viable count, elemental analyses and mineral analysis. Our investigation of the formation mechanisms of the beachrocks showed that in Okinawa, the evaporation of seawater and/or urease activity of the microorganisms may have resulted in the precipitation of high Mg calcite (HMC), leading to the formation of beachrocks. On the other hand, in Ishikawa, beachrocks and sand were present near a spring. Here, the pH value of the spring was in the range 4.7 and it has a higher concentration of Al3+. The mixing of spring water with seawater could have led to the precipitation of the cement containing Al and Si between sand particles and, thus resulting in the formation of the beachrock. Therefore, we have interpreted the formation mechanisms of beachrocks.
Stone heritages, such as stone architectures, stone statues, graveposts and stone towers, are generally located outside and they are exposed to the wind and rain. Therefore, they have been defaced and collapsed due to weathering. The main reason of the weathering is the existence and movement of water in the rock heritages, and the appropriate protection method is expected in order to keep heritages in good conditions. Kyusyu Archaeological Laboratory introduced a new method, Aquo-Siloxane Method, in order to protect the stone heritages. In this method, the protection material is painted on the surface of heritages, and the material penetrates into the rock pores. Then the water in the stone structures is captured by the material and it becomes stable. However, the stability of the captured water has been not evaluated quantitatively yet, and the characteristics of Aquo-Siloxane Method also have not been verified yet. In this study, the tests related to the water migration in rock pores, such as water permeation tests, evaporation tests and capillary tests, were applied to the porous rock samples. By comparing the state between with and without protection, the effect of Aquo-Siloxane method was verified. It was found that the water migrations in the pores were restrained in the rock sample once the water was captured by the protection material. The condition when the protection material is painted was also confirmed, and it was found that the protection effect was not influenced by the initial water content of rock samples.