(1. Beijing Urban Construction Group Beijing 100025 2. Institute of Geology and Geophysics, China Academy of Sciences, Beijing 100029)

Soil-rock mixture slope is a heterogeneous slope formed by landslide accumulation, residual slope accumulation, landslide accumulation and flood accumulation. Because of the heterogeneity of its material composition, its deformation characteristics and failure mechanism are significantly different from other slopes. According to the characteristics of its material structure, the soil-rock mixture slope should be treated by mesoscopic treatment technology, which can be treated as homogeneous continuum, equivalent homogeneous continuum, heterogeneous discontinuity and discontinuity controlled by structural plane.

Earth-rock mixture; Meso-treatment technology; side slope

Soil-rock mixture slope is composed of soil mixed macadam or macadam and macadam or macadam mixed soil [1, 2]. Generally developed in Quaternary loose accumulation layer, it is mainly formed by landslide accumulation, residual slope accumulation, landslide accumulation and flood accumulation. Soil-rock mixture slopes are widely distributed in China, such as granite residual soil slopes in Hong Kong, Guangdong and Fujian, ancient landslides and landslides widely distributed in southwest China, the Three Gorges reservoir area of the Yangtze River and the middle and upper reaches of the Yellow River, and moraine slopes distributed in Tibet [3]. Because this kind of slope is usually large in scale, with many influencing factors, sudden instability and complex sliding conditions, it often brings serious harm and huge property losses to the national economic construction, people's normal life and life safety. Therefore, it is of special theoretical and practical value to systematically analyze, study and summarize this kind of slope.

For soil slopes and rock slopes, people have carried out long-term, systematic and fruitful research work and gained rich practical experience. The deformation characteristics of soil-rock mixture slope are significantly different from other slopes because of its different material composition and structure. At present, people's research on it is only in the qualitative analysis stage, or through the correlation analysis between it and some factors to explore the mechanism of landslide [4,5], or through model tests to analyze stability [6,7], lacking the theoretical and technical methods like studying soil slopes and rock slopes. The author thinks that the essence of the problem, that is, the uniqueness of soil-rock mixture slope, should be recognized first, and then these problems should be analyzed and studied from a new angle.

1 concept of mesoscopic machining technology

The study of failure mechanism of rock and soil can be divided into macro, micro and micro categories according to the size of the research object. At present, there is no unified standard for the division of these three categories. According to the viewpoint of Xie Qiang et al. [8], the faults and joints that are widely developed in the field and affect the mechanical properties of engineering rock and soil are classified into macro level; The cracks that develop in the geotechnical structure and directly affect the mechanical properties of geotechnical belong to the meso-level. Dislocations developed in mineral crystals generally have no direct influence on the macro-mechanical properties of rock and soil, and are classified as micro-level. From the perspective of research implementation, macro research is mostly combined with field investigation and field experiment. Due to the observation means and cost, the research done is always limited. Microscopic research is mostly carried out in the form of laboratory mechanical experiments and microscopic observation of rock samples. As far as the feasibility of existing economic and technical means is concerned, the study of rock and soil fracture process at meso level is the main research method at present.

The traditional study of rock mechanics adopts the experimental and analytical methods of macro-mechanics, and has achieved many remarkable results. With the continuous improvement of many important geotechnical engineering requirements and the gradual deepening of research work, people gradually realize that it is obviously not enough to discuss the mechanical mechanism of geotechnical engineering only from the macro-scale phenomenological point of view. In 1998, Sun Jun proposed that rock mechanics should be further discussed from the micro, even micro-scale, and combined with the "three views" to be more comprehensive and in-depth.

The so-called mesoscopic is between micro and macro, which is large enough relative to the micro scale of molecules and atoms, but small enough relative to the macro scale of the objects involved; This is a relative concept of different materials and their different engineering feature sizes. For rocks, the mesoscale may be in the range of centimeters or even meters. People generally have to go through the repeated process of macro-meso-micro-macro to understand the laws of natural phenomena. Higher-scale research is always based on lower-scale research. This means that the occurrence of macroscopic phenomena can be explained by microscopic research, and the occurrence of microscopic phenomena can be explained by microscopic research.

2 the necessity of meso-therapy

Some heterogeneous soils, such as gravel deposits and colluvium, may contain particles of different sizes, including boulders, which bring some special problems to geotechnical engineers. The difficulty in sampling and testing these substances has been well reported [9] and solved to some extent [10]. However, the method of evaluating the slope stability has not attracted enough attention. The slope analysis method of conventional soil is actually based on relatively good gradation, and its properties are assumed to be independent of unit size. However, for the soil-rock mixture slope, due to the heterogeneity of its material composition, it needs to be treated from the perspective of unit size, that is, meso-level.

The mechanical properties of soil-rock mixture are mainly influenced by its particle size composition, that is, by the size, shape, uniformity, roundness and especially porosity of its constituent particles. If clay, silt or organic matter are involved, the nature will be more complicated.

Figure 1 Three failure modes of soil-rock mixture slope

When the soil-rock mixture slope contains a small amount of large particles, there are generally three failure situations [1 1] (Figure 1), and the failure mechanism conforms to the principle of minimum shear strength. What kind of phenomenon occurs depends on the relative strength of particles and surrounding matrix, the relative size of particles and crushing zone, the content of large particles and stress level. For example, passing through a particle may occur when the particle is weak. If there are only a few large particles, they tend to deviate from particle failure, and the geometry of fracture area will change slightly to some extent because of the existence of particles. Hencher & Martin [12] thinks that when the content of large particles exceeds 30%, such as the collapse in Hong Kong, the large particles are generally disturbed when they pass through the fracture surface of the filler. If there are many large particles, a serrated fracture will be formed (Figure 2). If the particles are too large compared with the slope size, or the filler tends to fail along the thin fracture surface, the possibility of expanding the fracture zone is less than the possibility of deviating from the particle type. The failure of deviating particles is more likely to occur in the low stress area where dilatancy may occur, rather than in the high stress area where dilatancy is limited, and the broken area may widen in the high stress area.

Fig. 2 Sawtooth fracture surface of soil-rock mixture slope

In the above discussion, it seems that large particles either increase the stability of the slope or have little effect. However, large particles may reduce the stability of the slope. This happens when the shear strength along the interface between large particles and filler or large particles is less than the strength of filler itself, and it accounts for a considerable proportion. This effect will be more obvious if the position occupied by large particles is less than the ideal position. In addition, it should be noted that the existence of large particles will significantly increase the drainage of slope.

From the above analysis, it can be seen that in order to study the stability of slope more accurately, it should be treated from a microscopic point of view. For major projects, detailed investigation should be carried out as much as possible to determine the heterogeneous geological profile of the soil-rock mixture slope, find out the large gravel blocks that affect the slope stability, and consider these blocks in the stability analysis.

3 meso-processing method

For the specific project to be studied, firstly, a geotechnical engineering model that reflects the actual situation of the project as much as possible should be established on the basis of relatively sufficient survey data. Its central idea is what kind of substance or medium it should be regarded as first, so its content should include the basic geological conditions related to the slope, such as the size of the slope, the granularity and uniformity of the medium contained, the weathering degree of rock mass, the development degree of discontinuities such as joints, and the mutual combination of rock layers. Different engineering conditions should have different treatment methods. The following are the treatment methods for different slope projects according to the mechanical properties of soil-rock mixture materials, that is, according to the engineering geological conditions, the slope materials are regarded as homogeneous or heterogeneous, continuous or discontinuous.

3. 1 uniform continuum

Figure 3 can be regarded as a uniform continuous slope.

The most commonly used method is to regard the soil-rock mixture as a basically uniform continuum, and take the parameters of the most important substance as the parameters of regional representation. This method is suitable for the case that the soil-rock mixture contains a small amount of gravel blocks, and the size of the blocks is too small compared with the size of the slope to affect the overall mechanical properties (Figure 3). At this time, it is effective to regard rock and soil as homogeneous materials, and its engineering parameters have no obvious changes compared with pure soil, so there can be no structural control behavior between block and soil. In this case, it is not necessary to carry out on-site in-situ testing, and only laboratory testing can obtain the required parameters.

3.2 Equivalent Uniform Continuum

When the soil-rock mixture contains more gravel, but the gravel distribution is relatively uniform and the size is relatively small compared with the slope size, it can be regarded as an equivalent homogeneous continuum (Figure 4). From the previous analysis, it can be seen that the mechanical properties of gravel are enhanced, but due to the difficulty of test methods, many engineers still regard the properties of the weakest component as the representative properties of this area, which is generally conservative.

Figure 4 can be equivalent to a uniform continuous slope.

Although it is simplified as a whole, that is, the properties of a certain substance represent the properties of a large area, the sensitivity of the substance to disturbance means that the data obtained from laboratory tests are not credible. Therefore, many designers prefer to rely on in-situ tests and empirical formulas. In addition, it is worth noting that in the soil-rock mixture, the contact surface between gravel and soil becomes discontinuous when it is deformed. In many cases, the appearance of discontinuity can not be ignored, even if there is no obvious mechanism, then discontinuity may occupy the main position and have a great impact on strength.

3.3 Heterogeneity and Discontinuity

When the soil-rock mixture slope contains gravels larger than the slope, even if the content is not high, but the position is important, it must be regarded as heterogeneous discontinuity (Figure 5). At this time, the existence of large gravel will inevitably affect the failure form and mechanism of the slope, and the influence of large gravel must be considered when analyzing the slope stability.

Fig. 5 Heterogeneous discontinuous soil-rock mixture slope

When dealing with heterogeneous rock and soil, it is difficult to determine the parameters with practical significance because of the strong core in weak matrix. At this time, it is impossible to test the sample with the representative size of rock and soil, so the distribution of strength and modulus of rock and soil from core rock to weaker material is either ignored (assuming that it is generally conservative when there is no structural control), or theoretical treatment and simulation are carried out [13]. In this regard, many scholars have conducted different degrees of research. Anon and West et al. [1 1 ~ 14] tried to calculate the additional shear strength caused by the expansion of irregular fracture surface when analyzing the slope with high gravel content, and its geometry was controlled by staggered debris. Irfan and Tang [15] gave the experimental guiding principle of additional strength caused by inclusions with higher strength in accumulation. As long as they have the same characteristics, the same method may be applied to heterogeneous rocks and soils. The work of Brett [16] and Jaros [17] published later reported the difficulty of obtaining actual parameters when predicting the settlement of foundation containing rock core. Jaros attributed the low settlement of the building to the influence of the timely core rock in the residual soil under the building, but Brett thought that the observed settlement could also be attributed to the residual soil of the parent rock with slightly different parameters. Demello [18] used the same method to consider the same problem when predicting the settlement of buildings in areas with different compressibility, which he attributed to the proportion of different substances and different compressibility.

3.4 discontinuity of structural plane control

When the slope contains a large number of crushed stones, such as weakly weathered rock mass and fractured rock mass, the content of soil fillings is small, and the discontinuous surfaces such as joints between blocks play a leading role, which should be regarded as discontinuities controlled by structural planes (Figure 6). Obviously, when the behavior of rock and soil is completely controlled by discontinuities, conventional rock mechanics methods can be used, but the influence of weathering on the shear strength of potential sliding surfaces must be paid attention to.

Fig. 6 Rock-soil mixed slope with controlling structural plane

From the above analysis, it can be seen that the above four methods are mainly distinguished by the content and size of gravel or block stone in soil-rock mixture. However, there is no specific limitation on its content and size, and it can only be determined according to the specific situation on site and engineering experience. The author gives the following suggestions: when the stone content is less than 10%, it can be considered as the first case; When the stone content is 10% ~ 25%, it belongs to the second case; When the stone content is 25% ~ 70%, it belongs to the third situation; When the stone content is more than 70%, especially more than 90%, it can be considered as jointed rock mass. For the first and fourth cases, soil mechanics and rock mechanics can be used to deal with them. In the second case, the strength parameters of soil-rock mixture can be determined according to the research results in the literature, and the corresponding calculation and analysis can be carried out. For the third case, there are two specific methods: first, after detailed investigation, find out the location of large gravel blocks, and use the measured structural model of soil-rock mixture for numerical simulation. Of course, this is an ideal method, which is difficult to realize; Secondly, according to the local statistical data, Monte Carlo method is used to simulate the distribution of gravel, and numerical simulation is carried out on the basis of random structure model, which is a practical and feasible method.

Thank you for the great assistance from many fields and scientific research institutions in the process of writing this article. I would like to express my heartfelt thanks to them, especially Yin Yueping, former deputy chief engineer of China Institute of Geological Environment Monitoring, Zhang Nianxue and Qu Yongxin, researchers of Institute of Geology and Geophysics, Chinese Academy of Sciences!

refer to

Guo Jun, Xia Tianhua. Nonlinear finite element method for stability analysis of over-compacted soil slope. See: Proceedings of the International Symposium on the Application of Computer Methods in Rock Mechanics and Engineering. Wuhan: Wuhan Technical University of Surveying and Mapping Press, 1994.

Hunan Water Conservancy and Hydropower Survey and Design Institute. Engineering geology of slope. Beijing: Water Conservancy and Electric Power Press, 1983.

Sun Guangzhong. Typical landslides in China. Beijing: Science Press, 1988.

Wang fadu. Preliminary study on the characteristics of shallow sedimentary landslide and its relationship with rainfall. Hydrogeology and engineering geology, 1995, (1)

Xie Shouyi, Xu Weiya. Study on mechanism of landslide induced by rainfall. Journal of Wuhan University of Hydraulic and Electric Power,1999,32 (1)

Liu Guanghua. Study on landslide mechanism and prevention countermeasures of loose accumulation layer along the river-a case study of Chongqing city. See: Proceedings of the Second Sino-Japanese International Symposium on Geological Environment in the Three Gorges Reservoir Area. Beijing: Coal Industry Press, 1996.

[7] Singero Komada. Slope instability induced by dam reservoir operation and its mechanism (internal data)

Xie Qiang, etc. Test and analysis of rock micromechanics. Chengdu: Southwest Jiaotong University Press, 1997.

[9] You Xinhua. Stochastic structure model of soil-rock mixture and its application. Doctoral thesis of North Jiaotong University, 200 1

[10] the problem of measuring the characteristics of moraine by Watkins D.J. Master thesis of Imperial College London, University of London, 1969.

[1 1] West L J, Hencher S.R. Evaluation of slope stability in heterogeneous soil. Balkema landslide: Rotterdam, 199 1

[12] Hencher S.R., Martin R.P. Description and classification of weathered rocks for engineering in Hong Kong. Proceedings of the 7th Southeast Asia Conference. Geotechnical engineering conference. Hong kong. 1982， 1

[13] lindquist s, Goodman r e. strength and deformation characteristics of physical model mixture. Proc. 1 St. NARM symptoms. Austin Balkema Rotterdam, 1994

See you later. Effect of coarse particle size on shear strength of colluvium. Report of geotechnical engineering office 23. Hong Kong, 1993

[15] Irfan R.Y. Potential construction problems of bridge foundation in complex marble stratum in Hong Kong. Engineering geology of architectural geology society. Special publication of engineering geology in London, 1995

[16] Architecture in tropical soil. Proceedings of the 2nd Singapore International Conference on Tropical Soil Geomechanics. Barkema, 1988

[17] Jaros m.b. Settlement of two multi-storey buildings built on the residual lava of Venterdorp. Seminar on Soil-Structure Interaction of South African Institution of Engineers. Durban, South Africa, 1978

[18] Demello v. F. B. Thoughts on soil engineering applicable to residual soil. Proceedings of the Third Southeast Asia Soil Engineering Conference 6438+0972