Experimental Study on the Mud-Water Inrush Characteristics through Rock Fractures

Hunan Provincial Key Laboratory of Safe Mining Techniques of Coal Mines, Work Safety Key Lab on Prevention and Control of Gas and Roof Disasters for Southern Coal Mines, Hunan University of Science and Technology, Xiangtan, Hunan 411201, China State Key Laboratory of Coal Resources and Safety Mining, China University of Mining and Technology, Xuzhou, Jiangsu 221008, China Department of Civil Engineering and Engineering Mechanics, University of Arizona, Tucson, AZ 85721, USA


Introduction
Plenty of underground engineering constructions such as railway tunnel and coal extractions are threatened by various kinds of water inrush and mud gushing in China [1][2][3][4].Research methods of theoretical analysis, laboratory experiments, field investigations, and numerical simulations have always been performed by many scholars and experts to discover the mechanisms and influential factors of water inrush and mud burst in fractured rock masses [5][6][7][8].Accordingly, numerous prevention and treatment technologies were proposed.Physical tests and site investigations of mudwater inrush show that the main factors, related to watermud burst, are the supply of water, pressure of water, geological structure, engineering excavation, and so on [9][10][11].On one hand, the theoretical analysis and numerical simulations also come up with some effective and practical ways to prevent and treat mud-water burst disasters [12][13][14].
On the other hand, for setting risk countermeasure, eliminating and reducing the adverse influence of risk, many ways of risk evaluation and prediction about mud-water burst in fractured rock masses were raised to disclose potential risk factors [15][16][17][18].
In recent years, there have been more than 100 cases of mud-water burst observed in the southwest mountain, especially in Karst areas, such as Yunnan, Guizhou, and Sichuan, causing serious losses of human life and property and deterioration of construction conditions [19][20][21][22].e fractured rock masses in Karst zones as a complex and widespread geological conditions in China, it is extremely easy to induce geologic hazard especially mud-water gushing when the underground engineering goes through it [23][24][25].At present, the research on mud-water outburst mechanism is mainly concentrated on reaction of water-rich area, damage of floor strata, and prevention and treatment [26,27] in addition to some technical solutions for special conditions that depended on different engineering backgrounds.Although many studies on fluid-solid coupling response in coal mass and fractured rock mass have been performed, indicating that the permeability characteristics of coal rock mass are closely related with their mechanical behavior [28][29][30] rough laboratory experiments, the process of mud-water outburst in Karst areas was performed many times through which we can deeply understand the comprehension of mud gushing in Karst areas and understand the improving and enriching mechanisms of the mud burst in Karst areas.e major chemical components of red clay are SiO 2 , Al 2 O 3 , Fe 2 O 3 , CaO, MgO, and so on, and natural moisture content is 6-10%.Firstly, the soil sample is dispersed and exposed to sun for 7 days (Figure 1); during the period of exposure, a wooden stick was used to crush the clay particles successively and then dried under the temperature of 105 °C-110 °C for 24 h, and the standard soil sieve is used to select three different particle sizes which are 5-10 mm, 2.5∼5 mm, and below 2.5 mm.According to the method of earthwork test GB/T50123-1999, using the digital display measurement instrument of soil liquid-plastic limit to measure soil moisture content under liquid and plastic limit stages, the measurement results are obtained, and these results show that the moisture content of the plastic limit is 19.8% and liquid limit is 30.2%.

Tests of Mud Viscosity
e measurement instrument of mud viscosity is the NDJ-8S rotary viscometer as shown in Figure 2.

Tests' Design and Results Discussion.
ree different kinds of particle size clay are divided into four groups: the particle size of 5-10 mm is short for coarse particle clay, 2.5∼5 mm is short for medium particle clay, below 2.5 mm is short for fine particle clay and mixed particle size clay.After three mud viscosity experiments on each group were conducted, the mean was calculated.Moreover, the beakers and clay samples of each group were labeled as follows: #1, #2, and #3 for coarse clay; #4, #5, and #6 for medium clay; #7, #8, and #9 for fine clay; and #10, #11, and #12 for mixed clay.According to the national standard GB/T20973-2007, mud viscosity tests were performed.Table 1 shows the experimental parameter and design of the soil sample.
Based on the tests' results (Figure 3), the changes of mud viscosity can be divided into three stages: that is, the mud viscosity declines rapidly at the stage of ab, the decline is slow at the stage of bc, and the decline is kept unchanged at the stage after the point c.
As is shown in Figure 3, when the moisture content is constant, mud viscosity does not vary with particle size, and it means that particle size has little ever no influence on mud viscosity which depends on moisture content completely.
At the stage of ab, the water in the mud is no longer in the form of combined water, but it is in the form of free and   2 Advances in Civil Engineering gravity water which can dissolve and erode some soluble bonding material between particles; furthermore, it also has some physical-chemical e ects on clay particles, such as lubrication and softening.ese poor e ects will make the spacing increased between particles, and the chance of interaction reduces, bonding force decreases, and the soil skeleton becomes loose.Meanwhile, at the stage of ab, the viscosity is extremely sensitive to moisture content.At the stage of bc, mud was in a state of ux completely; moreover, the direction and speed of motion of particles were controlled by water.During the process of mud ow, there is no drag force between particles, but for those which cannot dissolve, during the process of owing, the sorting phenomenon occurred, which leads to rearrangement and sedimentation of particles and changes mesostructure and mechanical properties of particles.It is the sorting phenomenon caused by water that has large e ects on the soil skeleton.But at this stage of bc, mud viscosity is less sensitive to moisture content than the stage of ab.
At the stage after the point c, the viscosity of the mud is tending to be the viscosity of pure water 5.46 MPa•s, and the bonding force and the interaction between the particles in the mud can even be ignored.It is evident that when the moisture content remains the same, the change of viscosity does not vary with the particle size.
e foregoing analysis shows that the mud viscosity does not depend on the particle size but depends on moisture content.Meanwhile, the varying tendencies of each viscosity curve are consistent, no matter how complex the process of the interactions about water and soil is.For further investigation of the relationship between the viscosity and moisture content, the mean method was applied to process each mud viscosity curve of di erent particle sizes, and then, MATLAB software was used to perform data tting (Figure 4).
As is shown in Figure 4, at the stage of ab, when the moisture content is below 45%, viscosity will decline rapidly from 10462 MPa•s to 796 MPa•s which is caused by the increase of moisture content in mud; at the stage of bc, when the moisture content is 45%-56%, due to the action of the water, the bonding strength of the mud particle was deteriorative, aerosols were increased, and mud uidity was enhanced, but with the increase of moisture content in mud, the rate of decline of viscosity slowed signi cantly; when the moisture content was above 56%, viscosity slowed down extremely as the moisture content of mud increased; nally, change in viscosity remained unchanged (5.46 MPa•s).Mathematical models of the relationship between moisture content and viscosity were obtained by tting experimental data: v 8.19 × 10 5 exp (−(c/6.96))+ 5.52, where c is the moisture content and v is the viscosity.

Equipment and Method.
A test device of mud-water inrush through rock fractures was designed to study mudwater inrush characteristics though rock fractures.e skeleton of mud-water inrush through rock fractures is shown in Figure 5, and the test device of mud-water inrush through rock fractures is shown in Figure 6.As shown in Figure 5, the box system in the test device can be divided into three parts: water tank, mud tank, and upper tank, and the size of each part, in turn, is 300 mm × 300 mm × 100 mm, 300 mm × 300 mm × 400 mm, and 300 mm × 300 mm × 800 mm, respectively.e inside of the upper tank was stacked with cement bricks with the size of 70 mm × 70 mm × 70 mm as shown in Figure 7. Mud with a moisture content of 19.8%-30.2%was lled between bricks which were piled up in the upper tank.After mud consolidation, mud and bricks are formed as a whole.e space among the bricks can be considered as rock fractures.e mud tank was used to store mud.In this laboratory mud-water outburst tests, mud with a moisture content of 52% was chosen as mud-water outburst tests' material, and the mud-water Advances in Civil Engineering pressure values were set to 0.5, 0.8, 1.0, 1.2, and 1.5 MPa, respectively.At the same time, the mud tank and upper tank are connected through the porous steel plate.On one hand, the porous steel plate can bear the load from all cement bricks which were accumulated in the upper tank; on the other hand, holes in the porous steel plate can allow mud stored in the mud tank to pass through and enter into the upper tank under the water pressure from the water tank.ere is a pressure conversion piston between the mud tank and the water tank.e pressure regulator system is made up of an air compressor, a servo pump, and a stabilizer system.During the process of tests, the water pressure in the stabilizer system was monitored in real time by the servo pump; in addition, another important function of the servo pump is to ensure the presence of adequate water in the stabilizer system.If the water pressure in the stabilizer system is below the preset value

Advances in Civil Engineering
Experimental principles can be summarized as follows: the space among cement bricks (70 mm × 70 mm × 70 mm) to stack each other (11 rows and 4 columns) was regarded as the rock fractures.Meanwhile, the space among bricks was filled with hard plastic clay; cement bricks, which were stacked upon each other, were used to simulate the rock skeleton.After the hard plastic mud was consolidated, the fractured rock mass which was composed of bricks and clay was formed in the upper tank.e fracture widths in the horizontal and vertical directions are about 7 mm and 3 mm, respectively.e water tank is connected to the external stabilizer system; moreover, the pressure and quantity of water in the external stabilizer system were provided by the air compressor and servo pump which were controlled by the computer.Slurry with a moisture content of 52% in the mud tank could pass through the porous steel plate into the top box under water pressure; meanwhile, the physical state of consolidated clay would change from solid to mobile liquid which can flow through the fracture.

Results and Discussion
. Based on the observation and calculation from the physical modeling, the curves of the quality of mud-water mixture and time under different water pressures are obtained (Figure 8 and Table 2).
Based on the test results that were described above (Figure 8), the research results indicate that the changes in the mud-water mixture inrush can be divided into four stages: (1) e preparation stage of mud-water mixture burst (OA curve segment): at this stage, seepage channels of mud-water mixture were not formed, and thus, there was no mud-water mixture collected at the outlet of the test apparatus.Meanwhile, hard plastic clay, which was filled in the fissure, developed into soft-plastic and fluid-plastic states under the effect of washout and erosion of mudwater mixture gradually.is kind of the evolution stage of hard plastic clay from immobility to mobility was named as the preparation stage of mud-water mixture gushing, and the duration time of this phase is T OA .e experiment also revealed that the greater the augment in mud-water pressure, the shorter the duration time of the incubation period.When the pressure increased from 0.5 MPa to 1.2 MPa, the duration of this stage would decrease from 98s to 45 s.But when the mud-water pressure increases to 1.5 MPa, the incubation stage of mud-water mixture outburst fails to be observed, which showed that the hazard of mud-water inrush with the higher pressure has some characteristics of suddenness and violence.(2) e seepage period of mud-water mixture inrush (AB curve segment): mud-water inrush moves in fracture, and the seepage channels of mud-water forms gradually.e seepage velocity is slow, and the duration time of this phase is T AB ; moreover, with the increase of mud-water pressure, the duration time of this phase shortens.When the pressure is increased from 0.5 MPa to 1.5 MPa, the duration time of this seepage stage would decrease from 23 s to 5 s.(3) e nonstable period of mud-water mixture outburst (BC curve segment): at this phase, mud-water mixture is continuously scouring channels and gushing forward to external.e velocity of mud-water mixture outburst is constant; in addition, with the increase of mud-water pressure, the energy stored in the mud tank and the maximum velocity of mud-water mixture outburst would be increased.e experiment showed that when the pressure is increased from 0.5 MPa to 1.5 MPa, the maximum velocity of mud-water mixture inrush of this stage would increase from 1.155 e stable period of mud-water mixture outburst (CD curve segment): at this stage, the passages of mud-water inrush form, keeping a near constant velocity of mud-water mixture inrush.is stage lasts for a relatively longer time.

Conclusions
is paper conducted a series of mud viscosity experiments and chose mud-water with a moisture content of 52% as the tests' material.
e process of mud-water in rock fractures was simulated.e research results can be drawn as follows: (1) e mud viscosity does not depend on the particle size but depends on moisture content.Moreover, the relationship between moisture content and viscosity can be represented by the exponential function.
(2) e process of mud-water inrush can be divided into four stages: that is, the preparation stage of mudwater mixture burst, the seepage period of mud-water mixture inrush, the nonstable period of mud-water mixture outburst, and the stable period of mudwater mixture outburst.
, the previous studies did not research mud viscosity, pressure of mud-water outburst, and flow laws of mud-water on rock fractures.Essentially, solid clay cannot flow, only when mixed with water, it can be converted from solid state into fluid state with different viscosity; at the same time, under the pressure of mud-water which exceeds ultimate bearing capacity of rock pillars between Karst zones and tunnel excavation face, mud inrush and water outburst accidents are easily observed.So in this paper, the change laws of mud viscosity under different moisture content have been studied.e study establishes a similar material simulation model called the test device of mud-water burst in confined Karst to experimentally research the flow characteristics of mud on channels under different mud-water pressures.

Figure 1 :
Figure 1: Clay soil used in the test.

Figure 3 :
Figure 4: e curve of mud viscosity under di erent moisture content.

Figure 5 :
Figure 5: e skeleton of mud-water inrush through rock fractures.

Figure 6 :
Figure 6: Test device of mud-water inrush through rock fractures.

( 3 )
e duration time of the preparation stage and seepage period is shorter, and the maximum velocity of mud-water mixture inrush is larger with the increase in mud-water pressure.is research is supported by the National Natural Science Foundation of China (nos.51774131, 51274097, and 51434006), the CRSRI Open Research Program (CKWV2017508/KY), and the Open Projects of State Key Laboratory of Coal Resources and Safe Mining, CUMT (no.SKLCRSM16KF12).

Table 1 :
Experimental parameters and design of the soil sample.

Table 2 :
Experimental data of mud-water mixture inrush.