Flow and stress coupled characteristics of bimrock
The effect of soil matrix on flow characteristics for bimrocks
In some literature studies, many researchers have studied the flow law of bimrocks (or SRM) in laboratory experiments (Dunn and Mehuys, 1984; Shakoor and cook, 1990; Zhou et al., 2006a, b; Indrawan et ah, 2006; Shafiee, 2008; Wang et ah, 2016a, b, c, d), in-situ experiments (Gao et ah, 2009; Chen et ah, 2012, 2014), and numerical simulations (Liao, 2004; Xu et ah, 2010). Furthermore, some SRM flow laws have been discussed by the previous scholars: Zhou et ah (2006a, b) used the constant head per- meameter, measured the permeability coefficients of the SRM with different gravel contents, and concluded that the flow rule of the SRM follows Darcy’s law. Indrawan et ah (2006) reported that block proportion has an obvious influence on the permeability coefficient of the SRM, and the flow rule of the SRM is close to the rock block percentage. Shafiee (2008) investigated the permeability of a compacted SRM at different confining stresses and proved that the rock block percentage, rock block size, confining stress condition, and the matrix properties have an obvious influence on the permeability coefficient of the SRM, but it almost always obeys Darcy’s law. Gutierrez and Vallejo (2013) performed laboratory experiments to study the hydraulic conductivity of an SRM with a sand matrix, and they concluded that rock blocks change the discharge path and decrease the energy available to water to overcome the shear resistance (drag). Wang et al. (2016) and Dan et al., (2016) conducted laboratory flow tests and found that permeability coefficient of an SRM with clay matrix increases with increasing hydraulic gradient. Liao (2004) and Xu et al. (2010) used the numerical simulation method to study the seepage field distribution and mechanism of seepage failure with different rock block percentages. They both found that the flow of the SRM follows Darcy’s law. Gao et al. (2009) studied the permeability of an SRM with a loess matrix by systematic laboratory tests, in-situ experiments, and numerical simulations. Chen et al. (2012) used the double-ring permeameter and carried out in-situ permeability test of an SRM. Chen et al. (2014) studied the effect of granite gravel content on an improved SRM, and they reported that the permeability coefficient increases significantly with the increment of granite gravel content, especially in the range of 60%-70%. By in-situ experiments, they all concluded that the flow rule of an SRM obeys Darcy’s law.
After the above literature review, it is clear that due to the different properties of the soil matrix and the compaction state in SRMs, some researchers concluded that the flow rule of SRMs obeys Darcy’s law; however, others think not. Flow law is an important hydraulic rule that is crucial to the prediction of the seepage field, pore pressure distribution, seepage force field, and the stability of the SRM structure. Therefore, the basic motivation of the present work is to investigate the relationship of seepage velocity and hydraulic gradient for the SRM with a different soil matrix and to study the seepage characteristics and disclose the permeability mechanism of the SRM by large-scale permeability testing. A self-developed servo-controlled permeability apparatus was used to perform the seepage test in this work.
Materials and specimen preparation
Remolded SRM specimens were large-scale cylinder-shaped with a diameter of 300 mm and a height of 700mm. According to the geotechnical test technical manuals (GB/T 50123-1999) and soil specimen preparation standard BS1377-1 (1990), the diameter of the blocks should not be greater than 30 mm, and the threshold value for the selected rock and soil is 5 mm. Therefore, when grain size is greater than 5 mm, it is defined as a rock block and when less than 5 mm, it is defined as soil. Sieve analysis is used to measure the particle size distribution when the size is larger than 0.075 mm, and the water sedimentation method was used to conduct particle size analysis for clay and mucky soil, which has particles smaller than 0.075 mm. Lithology of rock blocks was marble, and the size range was between 5 and 30mm. The soil was obtained from the pit in the Chinese Academy of Sciences Institute of Atmospheric Physics at a depth of 10 meters. In this work, three kinds of soil matrix were used to perform the flow test; they are clay soil, mucky soil, and fine sand. The clay soil contained lots of clay minerals with strong hydrophilic properties. The liquid limit of hard clay can reach 64.32%, while the plastic limit can reach 36.32%; the plasticity index was about 28, and the liquidity index was about 0.05-0.127 for this soil. These indices indicated that this soil belonged to the typical hard plastic and high plastic clay. From the results, the mucky soil is a high liquid limit clay with high saturation, plastic limit, and plastic index. The saturation is about 90%-93.5%, and the plastic limit is about 22%-24%, while the plastic index is about 25%-36%, and the natural water content is 23%. To identify the mineral composition and mineral content, we both conducted both scanning electron microscope (SEM) and X-ray diffraction (XRD) tests on the clay soil and mucky soil. According to the results of the SEM tests, rod-like and irregular quartz grains with a grain size of about 0.01-0.03 mm can be seen that are probably surrounded by clay minerals. The XRD tests revealed that the clay soil has a higher percentage of clay minerals, such as kaolinite (26.73%), montmorillonite (61.52%), and illite (6.25%). For the mucky soil, XRD tests showed that the contents of kaolinite, montmorillonite, and illite are 28.23%, 68.21%, and 3.34%, respectively. The fine sand has a natural density of 1.73 g/cm3, and the water content is about 13%. The rock block used in the study is marble stone with a size between 5 and 30mm, with a mass ratio of 1:3:2:1 mixed. The natural density of rock blocks is 2.67 g/cm3, and the compressive strength is about 94.5 MPa.
Many researchers (e.g., Donaghe and Torrey 1994; Wang et al., 2015a) have adopted the hand mixing method for mixing rock blocks uniformly with the soil matrix. In order to ensure homogeneity of the specimens, the rock blocks were mixed by hand into the soil several minutes. Dynamic compaction was used for preparing the specimens, owing to the high difference of elastic modulus between the soil matrix and rock block; compactness of the SRM is the actual the compactness of the soil matrix. Soil density is a very important factor affecting the permeability of the SRM. For the same kind of soil matrix with a different rock block percentage, we compact the specimens with different hammer counts. For the specimen with clay and mucky matrix, the density is about 2.12 and 2.08 g/cm3, respectively. For the specimen with sand matrix, the density is about 2.14 g/cm3.