^{1, 2}

^{1}

^{1}

^{1}

^{1}

^{2}

Zonal disintegration have been discovered in many underground tunnels with the increasing of embedded depth. The formation mechanism of such phenomenon is difficult to explain under the framework of traditional rock mechanics, and the fractured shape and forming conditions are unclear. The numerical simulation was carried out to research the generating condition and forming process of zonal disintegration. Via comparing the results with the geomechanical model test, the zonal disintegration phenomenon was confirmed and its mechanism is revealed. It is found to be the result of circular fracture which develops within surrounding rock mass under the high geostress. The fractured shape of zonal disintegration was determined, and the radii of the fractured zones were found to fulfill the relationship of geometric progression. The numerical results were in accordance with the model test findings. The mechanism of the zonal disintegration was revealed by theoretical analysis based on fracture mechanics. The fractured zones are reportedly circular and concentric to the cavern. Each fracture zone ruptured at the elastic-plastic boundary of the surrounding rocks and then coalesced into the circular form. The geometric progression ratio was found to be related to the mechanical parameters and the ground stress of the surrounding rocks.

Recently, many countries have begun to focus on deep resource exploitation. With an increase in embedded depth, the zonal disintegration phenomenon occurs during tunnel excavation. Zonal disintegration refers to “alternating regions of fractured and relatively intact rock masses appearing around or in front of the working stope during the excavation of tunnels in the deep rock mass” [

Zonal disintegration of Taimyrskii mine [

Relationship between the damage of support and the distribution of zonal disintegration [

It is a character of the deep rock mass and has recently been a subject of focus. Many specialists have implemented several types of methods to explain the phenomenon. Sellers and Klerck [

However, the mechanical behaviour of deep rock mass is nonlinear, very complex, and clearly different from the engineering response observed in shallow embedded tunnel engineering. The zonal structure of fracturing discovered does not fit within the framework of the conventional theoretical models. Indeed, there is no convincing explanation, but some arguments on the forming condition and failure mode have been made. For example, Oparin and Kurlenya [

Simulation on the excavation process of tunnel in laboratory, using numerical method, is the most effective method. Based on this, the forming process and phenomena of zonal disintegration can be simulated and the mechanism can be revealed accordingly.

The two methods of physical and numerical simulation have their own merit and are complementary to each other. During the numerical simulation, virtual model is established in computer to research the actual engineering. It can be used to simulate the anisotropism, anisotropic and discontinuity characters of the medium, and the complex boundary condition in engineering.

Compared to numerical simulation, model test can reflect failure process in rock mass visually and truly. The analogical model test is an effective reduced-scale method for researching special engineering based on similarity theory. The model is constructed in a manner similar to that in which engineering prototypes and thus the deformation laws can be monitored using precision devices. The data from the model can be converted to that of an engineering prototype using similarity theory to reveal the stress distribution. Hence, real-world problems can be solved using this methodology.

The different advantages can be used and complementary to each other and thus the mechanism of the zonal disintegration can be revealed.

The similarity theory required that the following similarity criteria must be satisfied in mechanical modeling (Fumagalli [

To carry out the analogical model test, a suitable analogical material is required. The material determines whether the model test can reflect the mechanical response of the engineering prototype. Barite powder, iron powder, and quartz sand are used for the aggregate, and an alcohol solution of rosin is used as the mucilage glue. Through hundreds of sets of proportioning tests, a material referred to as Barites-Iron-Sand cementation analogical material was developed. The proportion of the aggregates and the concentration of the alcohol solution of rosin determine the mechanical behaviour of the material. The proportion of the aggregates and the concentration of the alcohol solution of rosin decide the mechanical behaviour of the material.

The similarity ratio of volume weight for the similar material is set to 1 : 1, while the similarity ratio of geostress is 1 : 25. Thus, the mechanical parameters of the mudstone and the corresponding similar material are shown in Table

Physical-mechanical parameters of the prototype rock and model material.

Material type | Prototype | Similar material |
---|---|---|

Volume Weight/KN·m^{−3} |
2.62 | 2.62 |

Edef/Mpa | 5250 | 175 |

Cohesion/Mpa | 10 | 0.4 |

Friction angle | 43 | 43 |

Uniaxial compressive strength/Mpa | 29.08 | 1 |

Tensile strength/Mpa | 10.22 | 0.674 |

Passion ratio | 0.272 | 0.272 |

Construction of the model: the model size is limited by the reasonable size of the steel frame and 30 were taken as the optimal similarity coefficient

The model was delaminated. The similar material was placed inside the steel drum in layers and then tamped to create the model. The steel drum was 400 mm high, and the wall was 10 mm thick and had an inner diameter of 450 mm. Each layer was 100 mm high, and thus four layers were needed to complete the model. Then, a circular cavern cylinder was preset inside the steel drum to make the tunnel. The cavern diameter was 160 mm and the cavern axis coincided with the central axes of the steel drum. Then the similar material was put inside the steel drum and compressed uniformly to build the model. The measuring components were set up when the material reached the desired height, to monitor the displacement in the surrounding rocks (Figure

Zonal disintegration of Dingji coal mine of Huainan mine area.

The model was split and the cracks distribution inside the model was then determined (Figure

Loading procedure of model test.

As Figure

Results of the model test.

Cracks distribution

Displacement distribution surrounding the cavern

In recent 5 years, many specialists have tried all kinds of numerical methods to simulate the zonal disintegration. Tang and Zhang [

The elements are needed to subdivide continuously during the crack propagation when using FEM. So it is not proper to the fracture problem in rock engineering. In 1999, Belytschko and Black [

the fracture is not need to be considered. The element can be subdivided in any part, including the fracture;

the grid is not need to subdivide again, which improves the efficiency of computing highly;

the shape function can be changed according to the problem in study, which is very flexible.

Many specialists have paid attention to XFEM and applied it to the rock fracture analysis. XFEM is adapted in this study with the criteria proposed to simulate the zonal disintegration. The procedure can be shown (Figure

Numerical procedure of the extended finite element method (XFEM).

The simulations on the crack development are carried to examine the efficiency of XFEM, which is shown in Figures

The numerical model of crack extension trajectory under uniaxial compression.

Crack propagated of rock sample under compression simulated by XFEM.

The numerical tests are carried out on the rock samples containing hole with XFEM, and zonal disintegration is reproduced (Figure

The maximum circumference tension stress criterion was taken as the initiate criterion.

The critical loading depends on the equation

That is, the fracture begins to develop when the equivalent stress intensity factor

The angle of crack initiates is

The stress intensity factor of circular fracture is [

Substitute (

The fracture angle is

The numerical process of zonal disintegration is shown in Figure

Block diagram of the simulating procedure of zonal disintegration.

Zonal disintegration of circular model obtained with XFEM.

Numerical model

Forming process of zonal disintegration

In order to research the forming mechanism deeply, the research on the damage procedure of zonal disintegration is carried on based on the geomechanical model test.

The diameter of model is 0.45 m, with the circle cavern of 160 mm in diameter, which is the same as the model test. The model is divided into 30000 elements (four-node). In order to simplify the issue, the hydraulic pressure is applied on the model, which is 0.5 MPa.

The numerical model are shown in Figure

As Figures

Zonal disintegration phenomenon in surrounding rock mass simulated with XFEM.

Model

The 21 steps

Computing done

And as Figure

Displacement distribution surrounding the cavern.

As shown in the photos of the fracture shapes of model test and numerical simulation, it can be found as follows.

Circular cracks propagated and coalesced to form a circular pattern. Four fracture zones were observed to surround the cavern, in accordance with the fractured shape obtained from the model test. Thus, the occurrence of zonal disintegration was confirmed in the model.

Comparing the fracture shape in numerical model (Figure

Some specialists believe that

Figure

Comparing Figure

Zonal disintegration phenomenon is simulated by XFEM.

The criteria of crack propagation are got. And it is appropriate for the simulation of zonal disintegration.

The results between the geomechanics model test and numerical simulation are consistent. Both the two methods show that the fracture line is the concentric circles of the cave. There are 4 fractured zones that appear surrounding the cavern. And the radiuses of the fractured zones fulfill the geometric progression.

The authors declare that there is no conflict of interests regarding the publication of this paper.

This work was supported by the National Natural Science Foundations of China (Grant no. 51209074); the China Postdoctoral Science Foundation (Grant no. 2012M511189 and 2013T60494); the Fundamental Research Funds for the Central Universities (Grant no. 2012B02714), supported by the State Key Laboratory For Geomechanics and Deep Underground Engineering, China University of Mining and Technology, under Grant SKLGDUEK1206; the Open Research Fund of State Key Laboratory of Geomechanics and Geotechnical Engineering, the Institute of Rock and Soil Mechanics, Chinese Academy of Sciences, under Grant no. Z012008, funded by CRSRI Open Research Program CKWV2012306/KY; and the Key Laboratory of Coal-based CO_{2} Capture and Geological Storage Open Research Program 2012KF08. The authors are deeply grateful for this support.