Advanced Search

Indexed by SCI、CA、РЖ、PA、CSA、ZR、etc .

Volume 31 Issue 6
Dec.  2020
Turn off MathJax
Article Contents

Kun Fang, Huiming Tang, Xuexue Su, Wentao Shang, Shenglong Jia. Geometry and Maximum Width of a Stable Slope Considering the Arching Effect. Journal of Earth Science, 2020, 31(6): 1087-1096. doi: 10.1007/s12583-020-1052-0
Citation: Kun Fang, Huiming Tang, Xuexue Su, Wentao Shang, Shenglong Jia. Geometry and Maximum Width of a Stable Slope Considering the Arching Effect. Journal of Earth Science, 2020, 31(6): 1087-1096. doi: 10.1007/s12583-020-1052-0

Geometry and Maximum Width of a Stable Slope Considering the Arching Effect

doi: 10.1007/s12583-020-1052-0
More Information
  • The stability of an arching slope in deformable materials above strong rocks strongly depends on the shape and width of the span. Equations for a free surface problem that incorporate these two parameters were derived using a simplified two-dimensional arching slope model,and were validated using physical model tests under 1g and centrifugal conditions. The results are used to estimate the maximum excavation width for a weak claystone slope in a lignite mine,for which we calculate a safety factor of 1.31.
  • 加载中
  • Beddoe, R. A., Take, W. A., 2015. Influence of Slope Inclination on the Triggering and Distal Reach of Hydraulically-Induced Flow Slides. Engineering Geology, 187:170–182. https://doi.org/10.1016/j.enggeo.2015.01.006 doi:  10.1016/j.enggeo.2015.01.006
    Bosscher, P. J., Gray, D. H., 1986. Soil Arching in Sandy Slopes. Journal of Geotechnical Engineering, 112(6):626–645. https://doi.org/10.1061/(asce) 0733-9410(1986)112:6(626) doi:  10.1061/(asce)0733-9410(1986)112:6(626)
    Chen, C. Y., Martin, G. R., 2002. Soil-Structure Interaction for Landslide Stabilizing Piles. Computers and Geotechnics, 29(5):363–386. https://doi.org/10.1016/s0266-352x(01)00035-0 doi:  10.1016/s0266-352x(01)00035-0
    Coulomb, C. A., 1776. Essai sur une Application des rÈ gles de Maximis & Minimis à Quelques Problèmes de Statique, Relatifs à l'Architecture. Mémoires de Mathématique et de Physique de l'Académie, 7:343–382 (in French)
    Coulomb, C. A., 1781. Théorie des Machines Simples, en Ayant É gard au Frottement de Leur Partie, et à la Raideur des Cordages. Recueil des Savants éTrangers de l'Académie Royale des Sciences, 10:161–342 (in French) http://ci.nii.ac.jp/ncid/BB10483025
    Das, B. M., Sivakugan, N., 2016. Fundamentals of Geotechnical Engineering. Cengage Learning, Boston
    Davis, R. O., Selvadurai, A. P., 2005. Plasticity and Geomechanics. Cambridge University Press, Cambridge
    Eab, K. H., Takahashi, A., Likitlersuang, S., 2014. Centrifuge Modelling of Root-Reinforced Soil Slope Subjected to Rainfall Infiltration. Géotechnique Letters, 4(3):211–216. https://doi.org/10.1680/geolett.14.00029 doi:  10.1680/geolett.14.00029
    EGAT (Electricity Generating Authority of Thailand), 1985. Thailand-Australia Lignite Mines Development Project. Geotechnical Report, Mae Moh Mine, vol. 3. Mae Moh Mine Planning & Administration Division, Thailand
    Fang, K., 2019. Progression and Onset of Undercut Slope Failure Observed by Surface Velocity in Physical Models Subjected to Arch Action: [Dissertation]. University of Kyoto, Kyoto
    Fang, K., Minamide, K., Pipatpongsa, T., et al., 2018. Effect of Groundwater Seepage in Undercut Slope Centrifugal Model. 53th Japan Geotechnical Society, Takamatsu
    Guo, P., Zhou, S., 2013. Arch in Granular Materials as a Free Surface Problem. International Journal for Numerical and Analytical Methods in Geomechanics, 37(9):1048–1065. http://doi.org/10.1002/nag.1137 doi:  10.1002/nag.1137
    Higo, Y., Lee, C. W., Doi, T., et al., 2015. Study of Dynamic Stability of Unsaturated Embankments with Different Water Contents by Centrifugal Model Tests. Soils and Foundations, 55(1):112–126. https://doi.org/10.1016/j.sandf.2014.12.009 doi:  10.1016/j.sandf.2014.12.009
    Hill, R., 1954. On the Limits Set by Plastic Yielding to the Intensity of Singularities of Stress. Journal of the Mechanics and Physics of Solids, 2(4):278–285. https://doi.org/10.1016/0022-5096(54)90018-6 doi:  10.1016/0022-5096(54)90018-6
    Huang, Z. Y., Zhang, Y. X., Dong, J., 2013. Determination of the Influence Scope of Opencast Mining in Bedding Slope Based on Arch Effect. Safety and Environmental Engineering, 4:34 (in Chinese) http://en.cnki.com.cn/Article_en/CJFDTOTAL-KTAQ201304034.htm
    Ito, T., Matsui, T., Hong, W. P., 1981. Design Method for Stabilizing Piles Against Landslide—One Row of Piles. Soils and Foundations, 21(1):21–37. https://doi.org/10.3208/sandf1972.21.21 doi:  10.3208/sandf1972.21.21
    Jenike, A. W., 1964. Steady Gravity Flow of Frictional-Cohesive Solids in Converging Channels. Journal of Applied Mechanics, 31(1):5–11. https://doi.org/10.1115/1.3629571 doi:  10.1115/1.3629571
    Khosravi, M. H., Pipatpongsa, T., Takahashi, A., et al., 2011. Arch Action over an Excavated Pit on a Stable Scarp Investigated by Physical Model Tests. Soils and Foundations, 51(4):723–735. https://doi.org/10.3208/sandf.51.723 doi:  10.3208/sandf.51.723
    Khosravi, M., Pipatpongsa, T., Takemura, J., 2013. Interface Shearing Resistance Properties between Moist Silica Sand and Surface of Materials Investigated by Direct Shear Apparatus. The Japanese Geotechnical Society, Geo-Kanto
    Khosravi, M. H., Takemura, J., Pipatpongsa, T., et al., 2016. In-Flight Excavation of Slopes with Potential Failure Planes. Journal of Geotechnical and Geoenvironmental Engineering, 142(5):06016001. https://doi.org/10.1061/(asce)gt.1943-5606.0001439 doi:  10.1061/(asce)gt.1943-5606.0001439
    Li, C. D., Fu, Z. Y., Wang, Y., et al., 2019a. Susceptibility of Reservoir-Induced Landslides and Strategies for Increasing the Slope Stability in the Three Gorges Reservoir Area:Zigui Basin as an Example. Engineering Geology, 261:105279. https://doi.org/10.1016/j.enggeo.2019.105279 doi:  10.1016/j.enggeo.2019.105279
    Li, C. D., Yan, J. F., Wu, J. J., et al., 2019b. Determination of the Embedded Length of Stabilizing Piles in Colluvial Landslides with Upper Hard and Lower Weak Bedrock Based on the Deformation Control Principle. Bulletin of Engineering Geology and the Environment, 78(2):1189–1208. https://doi.org/10.1007/s10064-017-1123-3 doi:  10.1007/s10064-017-1123-3
    Ma, S. Y., Xu, C., 2019. Applicability of Two Newmark Models in the Assessment of Coseismic Landslide Hazard and Estimation of Slope-Failure Probability:An Example of the 2008 Wenchuan Mw 7.9 Earthquake Affected Area. Journal of Earth Science, 30(5):1020–1030. https://doi.org/10.1007/s12583-019-0874-0 doi:  10.1007/s12583-019-0874-0
    McCue, S. W., Hill, J. M., 2005. Free Surface Problems for Static Coulomb-Mohr Granular Solids. Mathematics and Mechanics of Solids, 10(6):651–672. https://doi.org/10.1177/1081286505036423 doi:  10.1177/1081286505036423
    Ohta, H., Pipatpongsa, T., Heng, S., et al., 2010. Significance of Saturated Clays Seams for the Stability of Rainfall-Induced Landslides. Bulletin of Engineering Geology and the Environment, 69(1):71–87. https://doi.org/10.1007/s10064-009-0246-6 doi:  10.1007/s10064-009-0246-6
    Pipatpongsa, T., Khosravi, M. H., Doncommul, P., et al., 2009. Excavation Problems in Mae Moh Lignite Open-Pit Mine of Thailand. Proc., Geo-Kanto, Tochigi. 459–464
    Pipatpongsa, T., Khosravi, M. H., Takemura, J., et al., 2016. Modelling Concepts of Passive Arch Action in Undercut Slopes. Australian Centre for Geomechanics, Perth
    Sakaguchi, H., Ozaki, E., Tanaka, M., 1993. Some Relation between Stability of Arches and Plugging of Granular Materials. The Science Reports of Faculty of Agriculture, Kobe University, 20(2):225–229 http://ci.nii.ac.jp/naid/110000410485
    Sokolovskii, V. V., 1965. Statics of Granular Materials. Pergamon Press, Oxford
    Tang, H. M., Wasowski, J., Juang, C. H., 2019. Geohazards in the Three Gorges Reservoir Area, China—Lessons Learned from Decades of Research. Engineering Geology, 261:105267. https://doi.org/10.1016/j.enggeo.2019.105267 doi:  10.1016/j.enggeo.2019.105267
    Take, W. A., Bolton, M. D., Wong, P. C. P., et al., 2004. Evaluation of Landslide Triggering Mechanisms in Model Fill Slopes. Landslides, 1(3):173–184. https://doi.org/10.1007/s10346-004-0025-1 doi:  10.1007/s10346-004-0025-1
    Terzaghi, K., 1936. Stress Distribution in Dry and in Saturated Sand above a Yielding Trap-Door. Int. Conf. on Soil Mechanics Proc., 1:307–311 http://ci.nii.ac.jp/naid/10008851136
    Terzaghi, K., 1943. Theoretical Soil Mechanics. JohnWiley & Sons, New York. 11–15
    Tian, Y. Y., Xu, C., Ma, S. Y., et al., 2019. Inventory and Spatial Distribution of Landslides Triggered by the 8th August 2017 MW 6.5 Jiuzhaigou Earthquake, China. Journal of Earth Science, 30(1):206–217. https://doi.org/10.1007/s12583-018-0869-2 doi:  10.1007/s12583-018-0869-2
    Wang, J. J., Liang, Y., Zhang, H. P., et al., 2014. A Loess Landslide Induced by Excavation and Rainfall. Landslides, 11(1):141–152. https://doi.org/10.1007/s10346-013-0418-0 doi:  10.1007/s10346-013-0418-0
    Wang, W. L., Yen, B. C., 1974. Soil Arching in Slopes. Journal of Geotechnical Engineering, 61–78. https://doi.org/10.1016/0148-9062(74)90721-9 doi:  10.1016/0148-9062(74)90721-9
    Yao, W. M., Li, C. D., Zuo, Q. J., et al., 2019. Spatiotemporal Deformation Characteristics and Triggering Factors of Baijiabao Landslide in Three Gorges Reservoir Region, China. Geomorphology, 343:34–47. https://doi.org/10.1016/j.geomorph.2019.06.024 doi:  10.1016/j.geomorph.2019.06.024
    Zhang, W. Y., 2015. The Analysis and Application Study of Arch Effect in Bedding Rock Foundation Pit: [Dissertation]. Chongqing Jiaotong University, Chongqing (in Chinese with English Abstract)
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(13)  / Tables(1)

Article Metrics

Article views(181) PDF downloads(15) Cited by()

Related
Proportional views

Geometry and Maximum Width of a Stable Slope Considering the Arching Effect

doi: 10.1007/s12583-020-1052-0

Abstract: The stability of an arching slope in deformable materials above strong rocks strongly depends on the shape and width of the span. Equations for a free surface problem that incorporate these two parameters were derived using a simplified two-dimensional arching slope model,and were validated using physical model tests under 1g and centrifugal conditions. The results are used to estimate the maximum excavation width for a weak claystone slope in a lignite mine,for which we calculate a safety factor of 1.31.

Kun Fang, Huiming Tang, Xuexue Su, Wentao Shang, Shenglong Jia. Geometry and Maximum Width of a Stable Slope Considering the Arching Effect. Journal of Earth Science, 2020, 31(6): 1087-1096. doi: 10.1007/s12583-020-1052-0
Citation: Kun Fang, Huiming Tang, Xuexue Su, Wentao Shang, Shenglong Jia. Geometry and Maximum Width of a Stable Slope Considering the Arching Effect. Journal of Earth Science, 2020, 31(6): 1087-1096. doi: 10.1007/s12583-020-1052-0
  • The arching effect, defined as the stress transformation between deformable soils or rocks and adjacent non-deformable sections (Terzaghi, 1943, 1936), is important to the stability of both natural areas and excavated slopes. Based on the similar yielding part and unyielding part (Terzaghi, 1943), these slopes can be defined as arching slopes. Arching slopes have the following characteristics: (1) there is non-uniform or relative displacement between the yielding and unyielding parts of slope; (2) the unyielding part is supported; (3) the locally yielding part is due to the loss of support in the slope; (4) the slope lies on a bedding plane with a lower friction. The characteristics (1)–(3) of arching slope are similar to the condition of a trapdoor. For the third aspect, the loss of support in the slope can be produced by some cases, such as mining excavation in open-pit mining, foundation excavation in house-building, road construction and so on. For the fourth characteristics, if the base resistance in the slope is much larger than the lateral resistance resulted from arching effect, the analysis of arch action can be ignored. Thus, the low interface friction in the slope is considered. A schematic figure of the arching slope is shown in Fig. 1. A better understanding of arching slopes is vitally required in the work and studies of engineers and scholars.

    Figure 1.  Schematic illustration of an arching slope.

    With regard to arching effect in slopes, past studies have mainly focused on the soil arching among slope stabilization piles (Li et al., 2019b; Chen and Martin, 2002; Ito et al., 1981; Wang and Yen, 1974). Specifically, Ito et al. (1981) derived the equations of lateral stress located on a row of piles and considered six important factors including row position, pile diameter, pile thickness, slope angle, and pile head in stabilizing pile design. Bosscher and Gray (1986) examined the effect of arch action in pile retaining walls and provided some information about conservative pile design. Chen and Martin (2002) applied a fast Lagrangian analysis of continua method to investigate the arching area within pile groups. However, the study of the arching effect on a whole slope is limited.

    A few studies have investigated the arch affecting inside slopes instead of pile-slope interactions. Pipatpongsa et al. (2009) reported that an unstable rock undercut slope had a potential failure after the lignite was removed at the toe of the slope in mining excavation. Zhang (2015) presented an engineering problem of a slope in house-building. The slope was close to the constructed building during the process of foundation excavation. In addition, similar arch-shaped failures can be found in road construction near reservoirs (Li et al., 2019a; Tang et al., 2019; Yao et al., 2019; Wang et al., 2014; Huang et al., 2013).

    The engineering challenges of unstable slopes in Mae Moh mine, as typical arching slopes, are of concern. In the northeast pit of the Mae Moh lignite mine (Pipatpongsa et al., 2009; EGAT, 1985), a shearing zone is located within the overburden layers, which causes a few unstable areas. A cross-section of a rock slope in area 4.1, one of the unstable areas in Mae Moh mine is shown in Fig. 2. A part of lignite (300–500 m in length, 80–120 m in height) located at seams Q and K supports the upper rock mass. The rock mass lies on bedding planes filled with clay seams G1 and G2 (Pipatpongsa et al., 2016). The engineering challenges in area 4.1 appear after the excavation of the lignite. The upper rock mass may slide along the bedding plane, which results in loss of mineral resources, schedule delays, and casualties. Considering arching effect in the slope, after the part of the lignite is excavated, the lateral load transfer mechanism can protect the slope in some ranges. As a result, the maximum excavation width or the spacing between the unexcavated areas are crucial indexes in the construction, particularly in some large open-pit slope projects. Moreover, it is essential to design the slope protection method and to monitor the potential slope failure and the failure shape of the arching slope under the maximum excavation width. Hence, it is necessary to investigate the failure shape and critical width of a stable arch in an arching slope.

    Figure 2.  A cross section figure of the unstable grey claystone slope in area 4.1 (modified from Pipatpongsa et al., 2009).

    Aiming at solving engineering problems in Mae Moh mine, Khosravi et al. (2011) presented a simplified physical model of a slope. The undercut slope consisted of a slope part and a base part that was cut. The stable arch was observed in the 1g physical model until the base part was cut to the maximum width. In the centrifugal-scale model, experimental evidence regarding arch failure was observable in 50g slope models (Khosravi et al., 2016). Therefore, the physical model tests of arching slope in 1g and centrifugal scale were suitable for achieving the occurrence of the stable arch and validating the analytical results.

    Several studies have been done on the stable arch forming a free surface. The parabolic shape could be found in the condition where a chain of spherical particles is applied (Sakaguchi et al., 1993). McCue and Hill (2005) provided a continuum mechanics derivation about a surface problem in Mohr-Coulomb criteria and formulated the shape of the stable arch in a hopper. Guo and Zhou (2013) further extended the theory in trapdoor test with two-dimensional analysis. However, previous research has paid little attention to the relevant three-dimensional arching slope with such free surface problems.

    In this study, the shape of the stable arch and maximum excavation width in an arching slope are investigated based on free surface analysis. The stress distribution in a simplified two-dimensional arching slope model is also examined. Depending on the solid boundaries and resulting acting force, theoretical formulations have been developed for estimating the failure arching shape and the critical width in arching slope. Two physical model tests in 1g and centrifugal conditions are used for the validation of the proposed equations. Moreover, the estimated maximum excavation width is applied for stability analysis of a grey claystone slope at the site.

  • The solution of the shape of a stable arch in two dimensions based on the Mohr-Coulomb criterion was reviewed. More details about hoppers and trapdoors were presented in McCue and Hill (2005) and Guo and Zhou (2013). The maximum excavation width in the arching slope was deduced with the consideration of the base friction. The analysis of the slope was related to a three-dimensional problem. The three-dimensional arching slope was simplified in a two-dimensional problem by considering an acceleration on a friction slope along the slope direction instead of the acceleration of gravity in the trapdoor test along the vertical direction.

  • The geo-mechanics sign convention is applied throughout the text. The values of compression and tension in the convention are defined by positive and negative, respectively. To solve the symmetry of the excavation problem in an arching slope, the rectangular coordinate system is first adopted. In the coordinate system, compressive stress components σxx and σzz as well as shear stress component σxz are expressed as (Davis and Selvadurai, 2005; McCue and Hill, 2005; Hill, 1954)

    where σij denotes the components of the Cauchy stress tensor in the coordinate system, ρ denotes the bulk density of the geomaterial, and a is the acceleration acting along the slope direction (the details on the resultant acceleration and forces are explained in the section on the maximum excavation width).

    In the Mohr circle (Fig. 3), the mean stress p and the deviatoric stress q defined as the centre and radius in the circle, respectively, are expressed with major σ1 and minor σ3 principal stresses

    Figure 3.  State of stress confined within Coulomb's failure criterion in Mohr's stress circle.

    By introducing the angle of the minor principal axis ϕ, the stress components are expressed as follows, which are also shown in the Mohr circle.

    A relation between the shear stress and the normal stress with the angle of the minor principal axis is obtained by removing the stress invariants p and q.

    The Mohr-Coulomb yield criterion as a constitutive law is applied (Coulomb, 1781, 1776). In the geomaterial, the stress components must satisfy Coulomb's law within the form of an inequality condition.

    where ϕ denotes the angle of internal friction of geomaterial and c denotes the cohesion of geomaterial.

    The inequality indicates that the stress field is at the limit or under the yielding level. When the material is at the yield condition, Eq. (9) can be expressed as

    By combining Eqs. (5)–(7) and Eq. (10), the equation of equilibrium in Eq. (1) is changed with regard to stress invariant q and the angle of minor principal stress ϕ.

  • The unit normal of an element surface is defined as n=nxi+nzj. The traction vector on the surface is t(n)=nxi+nzj (McCue and Hill, 2005). Therefore, the states on a free surface are shown as follows.

    We rewrite Eqs. (13)–(14) in the equivalent form on a free surface.

    Combining Eqs. (5)–(7), we have the following equations on every point of the free surface.

    Since t(n)=0 in a stress-free condition, the minor principle stress must be zero from Eqs. (13)–(14). In addition, due to the obedience of the Mohr-Coulomb yield criterion for geomaterial from Eq. (10) and Eq. (17), the terms are expressed in the general case where the friction angle of geomaterial is ϕ and the cohesion is c. This is also shown in the Mohr circle.

  • A free surface OB and a solid boundary OA intersect at point O, as shown in Fig. 4b. In geotechnical engineering, the solid boundary is considered a rigid wall used for supporting soil laterally or a stagnant edge of the geomaterial itself. Thus, the friction angle at the boundary depends on the interface friction angle ϕw or internal friction angle ϕ. Due to the geometry and boundary condition of the arch, the states of stress (σrr, σθθ, σ) expressed in polar coordinates are presented. These components of stress also obey static equilibrium conditions in the coordinate direction with a downward-acting force along the slope direction. With the assumption that stress at point O is not a singularity of stress, the equilibrium equation in the polar system is expressed (McCue and Hill, 2005; Hill, 1954). Here, θ is measured anticlockwise from the x-axis.

    Figure 4.  Geometry and boundary conditions in an arching slope. (a) The selection of the direction of the axis in the arching slope; (b) an intersection between the free surface and the boundary in a half-width symmetric side.

    Which obeyed

    where q0 denotes the stress invariant q at point O.

    Alternatively,

    As mentioned above, the solid boundary with regard to the arching slope is unyielding geo-material. To investigate the geometry of the stable arch, the boundary between the yielding part and the unyielding part needs to be clarified. Based on the Mohr-Coulomb criterion, the angle of solid boundary OA can be expressed as

    The angle α measured from horizontal to the free surface is

    where ω denotes the wedge angle between the free surface and solid boundary in Fig. 4. The value is π/2–ϕ.

  • The geometrical shape of the free-surface can be solved according to the derivation of the governing equations, free surface conditions and boundary conditions. In the arching slope, the maximum excavation width or critical excavation width Bmax is defined. When the excavation width B is less than Bmax, the slope is considered safe. Note that the complicated conditions at sites such as stress concentration are ignored.

    Given the critical condition B=Bmax that geomaterial near the free surface yields, the solution of the shape of a stable arch is solved by Eqs. (11)–(12) with the summarised boundary conditions (McCue and Hill, 2005). The polar coordinate system is applied as shown in Fig. 5.

    Figure 5.  Schematic figure of a stable arch in an arching slope.

    Sokolovskii (1965) proposed stress invariant to be a linear function of radius r in the polar system. Stress states change linearly with depth. The problems regarding the arching effect on the equilibrium in the trapdoor test are also investigated (McCue and Hill, 2005; Jenike, 1964). Therefore, the form of stress in the arching slope is also compatible with an angle of minor principal stress in the polar system.

    where q denotes the stress invariant, θ denotes the angle measured from the x-axis in Fig. 5, χ denotes a nondimensional scaled stress variable that depends only on the angle θ (Sokolovskii, 1965), and ϕ denotes the angle of minor principle stress which is a function of the angle θ alone.

    The differential equations are expressed by combining Eqs. (28)–(29)

    In the limiting equation, the friction angle of the geomaterial is constant throughout the arching slope. By adding the boundary conditions, the two differential equations can be solved. It is clear that shear stress is zero at the centre in the symmetry condition for the one boundary condition as shown in Fig. 5. Another boundary condition is the solid boundary between the yielding and unyielding geo-material in the static equilibrium condition. Here, subscripts c and o indicate the location along the central plane and solid boundary plane, respectively.

    The solutions of the stress system and angle of minor principle stress are obtained from Eqs. (28)–(31). Therefore, according to the information about the free boundary description in Eq. (18), the numerical solution of the shape of the stable arch is determined.

  • The maximum excavation width in the arching slope can be calculated by considering the solved stress variable from the last section. The problem in the arching slope is a three-dimensional question due to base friction between the bedding plane and sliding slope. A simplified two-dimensional arching slope model is proposed by considering a resultant acceleration on a friction slope. The diagram of an arching slope lying on a bedding plane is shown schematically in Fig. 6. The acceleration of the slope on a bedding plane is obtained from the resultant forces and mass. The forces the along slope direction are the driving force Fd and resisting force at the bedding plane fb.

    Figure 6.  The acting stress along the slope direction in an arching slope.

    Therefore, the acceleration of the arching slope on bedding plane a is expressed.

    where W and M denote the weight and mass of the slopes, respectively, A denotes the base area between the slope and bedding plane (A=LB), β denotes the slope angle, and ϕ′ and c′ denote the interface friction angle and apparent adhesion between the slope and bedding plane, respectively.

    The maximum excavation width in the slope is solved with the boundary conditions. Taking the boundary condition q0=r0ρaχ(θ0) in Eqs. (30)–(31), the maximum excavation width is shown as follows

    By adding q0 obtained by Eq. (19) and θ0 calculated by Eq. (26) to Eq. (38), it becomes

    With the estimation of 1/χ(θ0)=2.57(tanh 3ϕ)3.5 in a friction angle range of 20° to 80° (Guo and Zhou, 2013) and Eq. (37), the maximum excavation width in the arching slope is expressed as follows. Note that this maximum excavation width only applies in cohesive geomaterials.

  • In this part, a 1g physical model and a centrifugal model of arching slopes were used to check the geometrical shape of stable arches and the maximum excavation width.

  • A 1g physical model of the arching slope containing a slope part and a base part is shown in Fig. 7a (Fang, 2019). To present a complete arching failure, the length of the slope part (80 cm) and the base part (40 cm) are selected according to a set of preliminary tests. The width and thickness of both parts are 130 and 6 cm, respectively. The sand slope model with a 38° slope angle lying on a Teflon sheet is constructed by using compacted silica sand No. 6. Silica sand No. 6 is a poorly graded sand (Khosravi et al., 2013). The benefits of using silica sand No. 6 include the unified standards from factories, recycling in slope model preparation, and the suitable color of the sand for PIV (particle image velocimetry) analysis. Therefore, silica sand No. 6 is chosen in the 1g model test. The unit weight of silica sand No. 6 with 10% water content is 13.68 kN/m3. The friction angle and the apparent cohesion from the direct shear test are 41.5° and 8.0 kN/m2, respectively (Khosravi et al., 2011). The high apparent cohesion results from the suction in unsaturated soil (Ohta et al., 2010). Typically, the range of the stress level for measuring the geomaterial strength properties and the scope of stress change of this material in the physical model tests need to be set closely. In the conducted direct shear test, the range of stress level is from 0 to 100 kPa. However, in the 1g arching slope physical test, stress changes are only from 0 kPa to a maximum of approximately 10 kPa. Therefore, the apparent cohesion in the 1g test from the direct shear test may not be suitable. In this study, the apparent cohesion from the unconfined compressive strength and friction angle σc=2c(1+sinϕ)/cosϕ are estimated. The unconfined compressive strength measured in 1g condition is 2.1 kN/m2 (Fang, 2019). The calculated apparent cohesion of 0.47 kPa is used to maintain close stress levels. Moreover, the interface friction angle and cohesion between silica sand No. 6 and the Teflon sheet are 18.5° and 0.1 kPa, respectively. The detailed parameters of silica sand No. 6 are shown in Table 1 and Fig. 8.

    Figure 7.  Physical model of the 1g arching slope. (a) Physical model of the arching slope in the 1g condition; and (b) failure of the slope at the maximum excavation width in the 1g test (after Fang, 2019).

    Parameter Silica sand No. 6 Edosaki sand
    Water content (w) 10% 10%
    Specific gravity (Gs) 2.65 2.69
    Bulk unit weight (γ) 13.7 kN/m3 15.3 kN/m3
    Relative density (Dr) 10% 90.4%
    Maximum void ratio (emax) 1.132 1.37
    Minimum void ratio (emin) 0.711 0.85
    Average particle diameter (D50) 0.33 mm 0.20 mm
    Friction angle (ϕ) 41.5° 41.1°
    Cohesion (c) 0.47 kPa 3.0 kPa
    Interface friction angle (ϕ') 18.5° 17.0°
    Interface adhesion (c') 0.1 kPa 0 kPa
    . $ {D_{\rm{r}}}{\rm{ = }}\frac{{{e_{\max }} - e}}{{{e_{\max }} - {e_{\min }}}}.$

    Table 1.  Basic properties of the geomaterial in the tests (after Fang, 2019; Fang et al., 2018)

    Figure 8.  Grain size distributions of silica sand No. 6 and Edosaki sand (after Fang, 2019).

    The instruments including earth pressure cells, a continuous- shooting camera for PIV analysis, and a high-speed camera capturing the velocity during failure are applied. The lens of the digital camera and the high-speed video camera are perpendicular to the slope part. The process of the 1g arching slope test generally includes preparation, excavation, and data recording. First, in construction, the main target is to compact the sand evenly and carefully in every layer and to prevent stress concentration or inhomogeneous stress inside the slope. The compaction of the slope part is prepared every 5 cm along the slope direction, as shown in Fig. 7a. Second, the physical excavation using a long knife starts from the centerline at the base part. After the center of the base is cut in 5 cm, the right remaining base is first cut in 2.5 cm, and then the left base is excavated in 2.5 cm to maintain the symmetrical excavation process. This process continues until slope failure occurs. To avoid vibration during manual cutting, a finely excavated operation is needed. Before the next cutting, it takes one min to wait for potential failures. Finally, all data is recorded in a timely manner by the high-speed camera.

    The result of the arching slope is shown in Fig. 7b. The angles between the stress-free surface are 64° ±2° on the right side and 55° ±2° on the left side, respectively. The right part is very close to the theoretical value of 65.8°. The slope fails at 67.5 cm when the excavation length of the right base is larger than the length at the left base of 2.5 cm. Therefore, unsymmetrical cutting may cause a small stress concentration on the left. The difference in the left part results from the small stress concentration during the excavation process. The stable arch forms when the base part is excavated by 67.5 cm in the experiment. The calculated maximum excavation width is 64.3 cm based on Eq. (40) and the parameters in Table 1. The estimation error of the critical width obtained by the proposed equation is 4.7%, which means that the equation can reasonably estimate the critical width. The calculated value is less than the experimental result, which indicates that the calculated value is a conservative estimation. In practice, using the proposed equation produces a considerable undercut estimation on the safe side. Moreover, the selection of the apparent cohesion of the sand needs more attention.

    To compare the shape of the stable arch with the theoretical analysis in Eq. (18), the points of the shape are tracked in the physical model, as shown in Fig. 9. The configuration from the numerical calculation generally fits with the stable arch in the 1g test. The two sides of the arch in both the model and the analysis are close. However, the larger differences in the curve are located at the top of the arch. In the physical model, the failure line at the top shows a nearly straight line, while a curved edge is obtained from the theoretical analysis. These differences are probably due to the preparation of the physical arching slope model. In the compaction of the slope part, the sand slope is compacted along the slope direction every 5 cm. Although the surface of every layer along the slope orientation is scratched in the compaction process, some discontinuities may remain between the two layers. This causes the straight line near the top area of the arch.

    Figure 9.  Comparison of the shape of the stable arch in the 1g arching slope

  • A centrifugal model test was also conducted and checked by using a centrifugal machine (Fang et al., 2018). The detailed information of the device is described by Higo et al. (2015). On the centrifugal scale, the manual excavation process cannot be achieved due to the flight of the device during the test. Thus, the pre-excavation method is applied. Unlike the slope failure that occurred by cutting the base part in the 1g arching slope test, the collapse of the slope is due to the gradual increase in centrifugal force in the g-up process.

    A cross section of the slope model lying on a Teflon sheet is shown in Fig. 10. A soil chamber consists of three rigid walls and a visible front side made of transparent glass. To enhance the arching effect in the arching slope, two pieces of sandpaper were used between the sand slope and soil chamber walls. The width and thickness of the slope are 20 and 5 cm, respectively. The lengths of the slope part and base part are set as 22.5 and 17.5 cm, respectively, based on the dimensions of the soil chamber and the hardwood bedding plane. The inclination of the slope is 30° according to the widely used slope angle in centrifugal slope model tests (Beddoe and Take, 2015; Take et al., 2004). The soil used in the experiment was selected as Edosaki sand, a well-graded sand was recommended in the centrifugal model (Eab et al., 2014), with 10% water content to prevent water leakage and to maintain unchanged water content during the high centrifugal acceleration. The main parameters are as follows: the unit weight is 15.3 kN/m3; the friction angle and apparent cohesion are 41.1° and 3.0 kPa, respectively; the interface friction angle and cohesion between Edosaki sand and Teflon sheet are 17.0° and 0 kPa, respectively. The basic properties of Edosaki sand are shown in Table 1.

    Figure 10.  Schematic figure of the centrifugal arching slope model (modified from Fang et al., 2018).

    The main procedures and detailed preparation for the slope model are as follows.

    (1) Dry Edosaki sand and water with a 10 : 1 mass ratio are mixed and compacted every 1 cm for all base parts with support plates. Due to the location of arching failure in the slope part, the compaction process needs more care. The total length of the slope part is 22.5 cm. The length is first divided into four sections including the lower three parts (5 cm) and the last part (7.5 cm). In every part mentioned above, the slope is perpendicularly compacted every 1 cm layer five times with supports. A total of 20 compactions are conducted for the slope part. Noting that scratching the surface of each soil layer after compaction is needed to avoid discontinuity between two layers.

    (2) After removing the support plates, the base part is cut as a pre-excavation simulation. Since the vertical cut may cause the collapse at the edge of the base part in the centrifugal model, inclined cutting is used to make an average of 11 cm at the base part (top cutting Bt is 12 cm, and bottom cutting Bb is 10 cm, as shown in Fig. 11a).

    Figure 11.  Centrifugal model of the arching slope. (a) Outline of the model test of the arching slope after preparation; and (b) failure of the slope in the centrifugal test (modified from Fang et al., 2018).

    (3) In the last procedure, the soil chamber with the prepared slope is assembled in the centrifugal machine. Centrifugal acceleration acting on the slope starts to increase until arching failure occurs. The acceleration at the slope failure is recorded. The instrumentation for stress analysis can be found in Fang et al. (2018).

    The result of the arching slope is shown in Fig. 11b. The stable arch forms when the acceleration increases to 28.5g in the test. In the theoretical analysis, Eq. (40) is modified to consider the centrifuge acceleration as follows.

    where ng denotes centrifugal acceleration (n is the multiple of acceleration of gravity).

    The calculated maximum excavation width is 11.9 cm, which is based on Eq. (41) and the parameters in Table 1. The estimation error between the estimated critical width and average excavation width (11 cm) is 8.1%. However, under centrifuge conditions, the estimated value is larger than the experimental results. This is because the inclined cutting process is applied in centrifugal model. The upper layer of the base part with 12 cm causes earlier failure.

    The points of the shape from the physical model are also depicted in Fig. 12. To consider the method of cutting, the shapes of a stable arch at 11 and 10 cm are both added for comparison. The physical model points fit inside two predicted lines. The discontinuities occurring in the 1g test are less obvious on the centrifugal scale due to the better compaction process.

    Figure 12.  Comparison of the shape of the stable arch in the centrifugal arching slope.

    In summary, the theoretical analysis of both the maximum excavation width and the geometry of the stable arch is in acceptable agreement with the experimental results of the 1g and centrifugal arching slope model.

  • An undercut slope in area 4.1 of the Mae Moe mine is chosen for a case investigation. A diagrammatic view of the slope with the unstable rock mass and the bedding plane is also shown in Fig. 13. Similar to the simplified arching slope physical model, the slope in area 4.1 can also be considered as the slope part (potentially unstable rock mass) and the base part (the lignite that will be excavated or is already partly excavated). The unstable slope, which has an inclined angle β=18° and a thickness of T=33 m, is composed of weathered grey claystone (EGAT, 1985). The clay seam G1 between the grey claystone in the bedding plane has an angle of shearing resistance ϕ′=12°. According to some investigations at the site (Pipatpongsa et al., 2016; EGAT, 1985), the dry unit weight and saturated unit weight of the grey claystone are γd=16.47 kN/m3 and γs=20.13 kN/m3, respectively. The unconfined compressive strength of the grey claystone measured from a hydraulic jack is σu=0.75 MPa. The peak strength value (ϕ=45° and c=0.932 MPa) and residual shearing value (ϕ=43° and c=0.071 MPa) have been reported (EGAT, 1985). It can be found that the cohesion values in different tests produce considerable differences. Considering the similar value of the friction angle, the cohesion from the unconfined compressive strength and friction angle are also used. The calculated cohesion of the grey claystone is c=0.159 MPa. The detailed parameters of the undercut slope in area 4.1 are shown in Pipatpongsa et al. (2009). In the report, the width B=130 m of lignite was excavated. According to all the information above, the maximum excavation width in area 4.1 and the factor of safety based on the proposed equations are examined.

    Figure 13.  Sketch of a rock slope in area 4.1 with a potential sliding plane.

    The maximum excavation width of lignite in the dry condition is assessed using Eq. (40). The critical width is 399.0 m, and the safety factor of the slope is 3.1 by the following equation.

    The safety factor is high in dry conditions, which means that the slope is stable. However, the most dangerous condition is the rainy season. The fully saturated condition with seepage due to heavy rainfall may cause the collapse of the slope. Therefore, slope stability analysis with regard to rainy conditions is necessary in area 4.1. By considering the extreme condition of the fully saturated slope part with seepage (Das and Sivakugan, 2016), the equation of maximum excavation width is shown here.

    The critical width is 169.9 m and the safety factor of the slope in the fully saturated condition is 1.31 by Eq. (43). In accordance with the analyzing results, the slope is in a safe condition and approaches failure. Thus, the excavation process needs more consideration. If further excavation is planned, excavation with a suitable protection method can be applied, such as constructing compacted wastelands to protect the excavated part or applying a multiple excavation method. Moreover, the predicted failure arch shape can be extended in the application of the effective slope monitoring area in landslide monitoring and the selection of the method of landslide treatment. Those topics related to the shape of the arch in the arching slope can be further investigated in practice. On the other hand, these equations can also be applied to slope problems in reservoir areas (Li et al., 2019a; Tang et al., 2019) and slope analysis in seismic conditions (Ma and Xu, 2019; Tian et al., 2019). Generally, the proposed equations make it possible to obtain the maximum excavation width and to carry out stability analysis of the arching slope based on free surface analysis.

  • The aim of this study is to investigate the geometrical shape of a stable arch and the maximum excavation width in arching slopes. The three-dimensional question of the slope is simplified as a two-dimensional problem by considering a resultant acceleration along the slope direction. Based on a stress-free surface analysis in a hopper, the theoretical equations for the curve of the stable arch and critical excavation width in the arching slope are derived according to the continuum theory.

    A 1g physical model and a centrifugal model were used to validate the proposed formulas. In the physical model tests, the stable arch forms due to the simulated manual excavation process in the 1g condition. Arch failure in the centrifugal arching slope with a pre-excavation process is achieved by increasing centrifugal acceleration. The analytical solutions have acceptable agreement with the experimental measurements in both 1g and centrifugal conditions.

    Finally, a safety factor (1.31) of the slope in the fully saturated condition was calculated in the stability analysis of a grey claystone slope at the site, and alternative methods for further excavation in area 4.1 were suggested. Moreover, the shape of a stable arch can be applied in the monitoring and treatment of arching slopes.

Reference (38)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return