Experimental Study on Permeability Characteristics of Compacted Backfill Body after Gangue Grouting and Backfilling in the Mining Space

Abstract

The timely injection of gangue slurry into the mining space formed after coal mining can scale up the disposal of gangue and control surface deformation. However, the waterproof effect of gangue slurry in the mining space remains unclear, necessitating urgent investigation into the permeability characteristics of compacted backfill bodies of gangue slurry under the action of overburden. In this study, a multi-field coupled seepage test system for backfill materials was developed based on Forchheimer’s nonlinear seepage law, and a laboratory preparation method for compacted backfill body (CBB) of gangue slurry after grouting and backfilling in mining space under pseudo-triaxial conditions was proposed. Additionally, the pressure bleeding characteristics of gangue slurry under the action of overburden were studied, the variation law of permeability of the CBB with the axial pressure, a particle size range, and cement dosage was revealed, and the determination method for the permeability level of the CBB and its optimization method were put forward. The research results indicate that there are obvious staged characteristics in the pressure bleeding changes in gangue slurry. Axial pressure, particle size range, and cement dosage all have a significant impact on the permeability of the CBB. The permeability level of the CBB of gangue slurry is within the range of poor permeability and extremely poor permeability. After backfilling into the mining space, gangue slurry exhibits a significant water-blocking effect.

1. Introduction

During the production and processing of coal resources, a large amount of coal gangue is generated [1,2,3]. Among these, only a limited amount of high-calorific-value coal gangue is utilized, while a significant quantity of low-calorific-value coal gangue is compelled to pile up. Consequently, land occupation and soil erosion are caused, and geological disasters such as landslides and mudslides can be triggered, which subsequently lead to the degradation of the ecological environment in mining areas. The large-scale disposal of coal gangue has become a major challenge for the environmental development of coal mines [3,4,5]. After coal seam mining, a substantial underground void is formed, that is, the mining space. The generation of the mining space inevitably induces the fracture of the overburden, damages the crack structure of the underground rock layer and the stability of the aquifer, and causes problems such as water resource loss. Consequently, water-preserved mining has emerged as a challenging issue in the development of coal resources [6,7,8]. The gangue grouting and backfilling technique refers to using filling pumps, filling pipelines, and other equipment to transport gangue slurry made of broken gangue and water to the underground mining space [9,10]. Figure 1 shows the technical principle of gangue grouting and backfilling.

Under the premise of meeting the critical speed of slurry and ensuring the stable and efficient conveying of gangue slurry, the timely filling of gangue slurry into the mining space can provide certain support for the overburden and suppress rock movement and surface deformation [11,12,13]. When the gangue slurry is injected into the mining space, it continuously flows in the voids and cracks of the fractured rock mass. The moisture of the gangue slurry is continuously absorbed by the rock layer, and the aggregates in the gangue slurry are gradually compacted by the gravity of the overburden, rock compression, and local stress concentration. In the separation zone, a whole CBB is formed, which plays a bearing and water-blocking role [14,15,16]. Therefore, the gangue grouting and backfilling technology offers a viable solution to the two major challenges associated with large-scale disposal and water conservation mining of gangue, and studying the permeability characteristics the CBB after gangue grouting and backfilling is of great significance.

Extensive research on the permeability characteristics of loose rocks, cemented backfill materials, and other materials has been conducted through theoretical analysis, experimental testing, and other means. The patterns of change in material permeability and the main factors influencing it are investigated. To study the permeability of loose materials or broken rocks, Li et al. [17] tested the permeability characteristics of backfill materials mixed by coal gangue and fly ash using a self-made permeability testing system and introduced fractal characteristics to study the key factors affecting the complex pore permeability of broken coal gangue. The experimental results showed that adding a certain concentration of fly ash into the broken coal gangue can achieve good water resistance. Janjuhah et al. [18] studied the relationship between the permeability of carbonate rocks, porosity, and sound velocity. Chen et al. [19] characterized the particle size distribution of saturated fractured sandstone under different axial stresses through the fractal dimension and studied the corresponding relationship between the particle size distribution of saturated fractured sandstone and permeability. To study the permeability characteristics of cemented backfill materials with different components, Qiu et al. [20] conducted permeability tests on tailing-cemented backfill materials, tested the porosity of solidified tailing-cemented backfill materials through electron microscopy, and obtained the effect of the cement–sand ratio and gangue dosage on the internal porosity and overall permeability of tailing-cemented backfill materials. He et al. [21] conducted rheological, mechanical, and permeability tests on modified magnesium–coal-based solid waste backfill materials and investigated the influence of different dosages and lengths of polypropylene fibers on the properties of modified magnesium–coal-based solid waste backfill materials. The research results indicate that the transmission pathway between the connecting pores of polypropylene fibers is the main reason for the change in permeability. Zhou et al. [22] used aeolian sand as a filling aggregate, cement and fly ash as bonding materials, and quicklime and water-reducing agents as additives and explored the seepage characteristics of sand-based bonding backfill materials under the interaction of multiple factors, obtaining the significance degree of each factor’s influence. Liu et al. [23] conducted nuclear magnetic resonance and centrifugal tests on tailing-cemented backfill materials, found a high correlation between fractal dimension and the permeability of backfill material based on fractal theory, and analyzed the permeability changes of the backfill materials with the curing age and ash–sand ratio. Zhu et al. [24] summarized the influence of axial and confining pressures on the permeability of sand-based cementitious backfill materials and revealed the permeability change mechanism of these backfill materials through laboratory permeation meters, particle size analyzers, scanning electron microscopes, and X-ray diffractometers. In summary, research on the permeability of backfill materials mostly focuses on solid bulk or cementitious materials, and there is less research on pure gangue slurry. Moreover, the study on the preparation method of compacted backfilling system and permeability characteristics of gangue slurry is rarely involved.

Considering the application prospects of gangue slurry grouting and backfilling technology for water-preserved mining, the pressure bleeding characteristics, permeability changes, and waterproof optimization methods of gangue slurry were analyzed in this study. Specifically, a multi-field coupled seepage test system for backfill materials was developed, and a laboratory preparation method for the CBB of gangue slurry was proposed. The staged pressure bleeding characteristics of gangue slurry were analyzed, and the influence of axial pressure, particle size range, and cement dosage on the permeability of the CBB was obtained; the determination method for the permeability level of the CBB and its optimization method were provided. This research provides a theoretical basis for improving the waterproof performance of gangue slurry and enriching the theoretical framework for water preserve mining.

2. Materials and Methods

2.1. Test Materials

The gangue sample was taken from the washing gangue of a certain mine, with a density of 2.1 g/cm³ and a hardness of 167.5. In this test, two ratios of the gangue slurry were used. The mass concentration of gangue slurry, with particle sizes ranging from 0 to 2 mm, was 60%, and the mass ratio of gangue slurries with particle sizes ranging from 0.15–2 mm to 0–0.15 mm was (0–2): 1. The mass concentration of gangue slurry with particle sizes ranging from 0 to 5 mm was 65%, and the mass ratio of gangue slurries with particle sizes ranging from 2 to 5 mm, 0.15 to 2 mm, and 0 to 0.15 mm was (0–0.3): 2:1. The Chinese standard GB/T 8077-2012 [25] was adopted to conduct the dispersion test [26]. Laboratory tests showed that the dispersion degree of the two proportions of gangue slurry was around 230 mm, with good conveying characteristics, as shown in Figure 2. Ordinary Portland cement with a grade of 42.5 was used in the experiment.

2.2. Multi-Field Coupled Seepage Test System for Backfill Materials

To study the seepage characteristics of the CBB of gangue slurry under different influencing factors, a multi-field coupled seepage test system for backfill materials was developed. It mainly consisted of a material loading module, axial stress loading module, liquid seepage module, seepage outlet measurement module, and data acquisition and processing module. Figure 3 shows the details of the designed test system.

(1)

Material loading module

The backfilling material is placed inside the material loading module, which is a cylindrical cavity with a diameter of Ø 150 × 300 mm. Corresponding dynamic and static load experiments can be conducted through the axial piston installed inside the material loading module. The interior of the chamber is a fully sealed structure, made of high-quality 2205 duplex stainless steel, providing a sealed condition for liquid seepage experiments. There are three pressure measuring points arranged on both sides of the material loading module, which can monitor the real-time pressure of the backfilling material.

(2)

Axial stress loading module

The axial stress loading module can apply different ways and sizes of axial stress to the backfilling material. The TC-500D axial stress loading pump is selected to ensure uniform and stable loading of axial stress, offering a pressure range from 0 to 30 MPa. A displacement sensor is connected to the outlet end of the piston under axial compression, which can record the piston displacement in real time and facilitate the calculation of the compression height of the backfilling material. The maximum range of the axial displacement sensor is 200 mm.

(3)

Liquid seepage module

The liquid seepage module can apply different sizes of seepage pressure to the backfilling material, and liquid properties can be changed. In this experiment, ordinary tap water was used as the seepage liquid. The liquid seepage module is mainly composed of an advection pump and a piston container. The advection pump is used for the constant-speed quantitative injection of seepage liquid, with a flow rate range of 0.01–600 mL/min and a maximum working pressure of 15 MPa.

(4)

Seepage outlet metering module

The seepage outlet metering module mainly includes high-precision electronic balance and sealed water buckets of different volumes. This module is used for the real-time monitoring of liquid seepage conductivity and seepage flow rate.

(5)

Data collection and processing module

The data acquisition and processing module consists of a data collector, a computer, and data acquisition control software, which facilitates real-time monitoring and recording of key parameters (such as axial stress and displacement of backfill materials, liquid seepage flow rate, and water pressure during the loading process). The software is developed by the equipment manufacturer and has various functions, such as data display, recording, storage, and processing.

2.3. Testing Method for Permeability Characteristics of the CBB

2.3.1. Preparation Method for the CBB

After being injected into the mining space, the gangue slurry was subjected to pressure and local stress from the overburden, resulting in a pressure bleeding phenomenon. After the free water of the gangue slurry was completely drained, a CBB was formed. The axial stress on the gangue slurry varied at different depths of the overburden [27,28]. To simulate the formation process of the CBB, a laboratory preparation method for the CBB after gangue grouting and backfilling in mining space under pseudo-triaxial conditions was proposed using a multi-field coupled seepage test system for backfill materials. The corresponding proportion of gangue slurry was loaded into the material loading module, and a graded loading and pressure holding method was used to apply axial stress [29]; the bottom valve of the material loading container was opened during the loading process, and the amount of pressure bleeding in real time was recorded. The graded loading and pressure holding method refers to dividing the target loading stress into four levels for loading. Combining the experience of previous studies and the results of the pre-experiment in this study, the holding time of each layer was determined to be 20 min. When the target loading stress is reached, the pressure is continuously held until the amount of water bleeding reaches a stable value. At this time, the CBB is in a saturated state, and subsequent seepage tests can be conducted directly. Figure 4 shows the shape of the CBB.

2.3.2. Principle of Permeability Tests

The CBB is essentially composed of crushed gangue and bound water. Under the action of seepage pressure gradient, water flows through the pores of the crushed gangue in a non-Darcy flow [30,31]. Previous studies have shown that the Forchheimer equation can be used to fit the quadratic relationship between seepage pressure gradient and seepage velocity in the CBB:

$$G_P=-\frac{\mu }{k}-\rho \beta v^2$$

(1)

where G_p is the pressure gradient at both ends of the material loading container; μ is the viscosity of water; k is the permeability of crushed gangue; v is the seepage velocity; ρ is the density of water; β is a non-Darcy flow factor; and when β = 0, the seepage follows Darcy’s law.

In Equation (1), permeability k is used to describe the permeability characteristics of the CBB. The pressure gradient and seepage velocity are tested and collected by the designed experimental system using the steady-state method, the G_p-v scatter plot is plotted, and the quadratic equation between pressure gradient and seepage velocity is fitted. In this way, the permeability k of the CBB under different seepage pressure gradient conditions is obtained.

2.3.3. Permeability Characteristic Test

To study the influence of axial pressure, particle size range, and cement dosage on the permeability characteristics of the CBB samples, a controlled variable method was used to test the permeability characteristics of the CBB, as shown in

Table 1. Test plan for permeability characteristics of the CBB.

Test Number	Particle Size Range/mm	Cement Dosage/%	Axial Stress/MPa	Pressure Gradient/MPa
S1	0–2	0	2	1, 2, 3, 4
S2			4	1, 2, 3, 4
S3			6	1, 2, 3, 4
S4			8	1, 2, 3, 4
S5	0–5	0	2	1, 2, 3, 4
S6			4	1, 2, 3, 4
S7			6	1, 2, 3, 4
S8			8	1, 2, 3, 4
S9	0–5	2	2	1, 2, 3, 4
S10			4	1, 2, 3, 4
S11			6	1, 2, 3, 4
S12			8	1, 2, 3, 4
S13	0–5	4	2	1, 2, 3, 4
S14			4	1, 2, 3, 4
S15			6	1, 2, 3, 4
S16			8	1, 2, 3, 4
S17	0–5	6	2	1, 2, 3, 4
S18			4	1, 2, 3, 4
S19			6	1, 2, 3, 4
S20			8	1, 2, 3, 4

. The gangue slurry ratio remained consistent with that in Section 2.1, and a total of 20 experiments were conducted. Under each test condition, the water pressure was controlled at 1, 2, 3, and 4 MPa for steady-state seepage tests. Considering the economic cost of cement, the maximum cement dosage in this study was determined to be 6% after pre-experimentation [1]. Among them, the cement dosage refers to the percentage of cement mass to the mass of gangue slurry.

2.3.4. Procedures of Seepage Test

The steps of the seepage test on the CBB are as follows.

(1)

Pipeline inspection and system debugging. The connections of equipment pipelines were confirmed, the valves of the axial stress loading module and liquid seepage module were checked, and the integrity of all accessories was verified. The key module was initiated, and the corresponding testing software was debugged.

(2)

Loading and sealing inspection. Firstly, a sealing inspection was performed; then, the material loading container was filled with gangue slurry, and the surface of the material flat was scraped with a distance of 8 cm from the upper port. Finally, the pipeline was connected, and the airtightness was rechecked.

(3)

Axial stress loading. The axial stress loading module provides axial stress through water pressure. A constant-speed and -pressure pump and related valves were opened to absorb and pressurize pure water, and axial pressure can be increased through multiple repetitions of liquid suction and injection. The preparation of the CBB was carried out by using a graded loading and pressure retention method.

(4)

Seepage pressure gradient loading. An advection pump was used to pressurize the bottom inlet of the material loading container. The top outlet pressure of the material loading container was 0, so the pressure gradient of the material was the inlet pressure. The inlet pressure was gradually increased, and after the water stabilized in the CBB, the permeability test switch was opened. The system automatically recorded the pressure gradient and seepage data, and the permeability of the CBB was calculated.

(5)

Test termination. The test equipment was cleaned, and the test data were saved.

3. Results and Discussion

3.1. Staged Pressure Bleeding Characteristics of Gangue Slurry

The pressure bleeding rate of gangue slurry refers to the mass percentage of the water secreted from the slurry under the compression compared to the total mass of water. In the preparation stage of the CBB, the pressure bleeding rate of gangue slurry under different conditions can be obtained by measuring the mass of bleeding at the lower end of the material loading container. The calculation function is shown in Equation (2).

$$\phi =\frac{m_b}{S\times H_0\times {\rho }_g\times w_g}\times 100\%$$

(2)

where m_b is the bleeding quality of the gangue slurry; S is the cross-sectional area of the material loading container; H₀ is the initial height of the gangue slurry loading; ρ_g is the density of the gangue slurry; w_g is the solid mass concentration of the gangue slurry.

The mass of water bleeding at the lower end of the material loading container was monitored every 10 min, and the time was recorded. After calculation in Equation (2), the relationship curve between the pressure water leakage rate φ of the gangue slurry and the axial loading time t is drawn. It is found that the curve exhibits obvious phased characteristics. Figure 5 shows the changes in the pressure bleeding rate of gangue slurry in S1–S4 groups.

As shown in Figure 5, during the axial stress loading process of the gangue slurry, as the loading time increases, the pressure bleeding increases rapidly; then, its growth rate gradually decreases and, finally, gradually stabilizes. This indicates that the gangue slurry is no longer bleeding. The pressure bleeding changes in gangue slurry exhibit obvious phased characteristics, and the duration of the second stage is significantly longer than the first and third stages.

(1)

Phase 1. The pressure bleeding rate shows a linear increasing trend with loading time. At this stage, the gangue slurry is in a supersaturated state, with large particle spacing and more free water in the slurry. Under pressure, free water flows out uniformly from the seepage channel. However, when the gangue particles are in contact with each other, the first stage of gangue slurry pressure seepage ends immediately, so the time is shorter than the second stage. By comparing the slope of the first stage of the S1–S4 curve, it can be found that the pressure bleeding rate of the gangue slurry increases almost uniformly at the beginning of compression and rarely changes with the axial pressure. However, as shown in Figure 5, the greater the axial pressure on the slurry of the same proportion, the shorter the duration of the first stage and the earlier the transition to the second stage. Figure 6 shows the variation curves of the pressure bleeding rate of gangue slurry in the S4 and S8 groups. Under the same axial pressure, the linear growth rate of gangue slurry with particle sizes ranging from 0 to 5 mm in the first stage is significantly higher than that of the gangue slurry with particle sizes ranging from 0 to 2 mm. This is because there are more small-sized gangues in the gangue slurry with particle sizes ranging from 0 to 2 mm, and some of them follow the bleeding of the free water, thereby affecting the water bleeding rate.

(2)

Phase 2. At this stage, the pressure bleeding rate gradually slows down with the increase in loading time. Under axial stress loading, the distance between gangue particles gradually decreases. Through the processes of sliding, rotation, and fragmentation, gangue particles gradually assume a bearing role. The compression effect exerted on the gangue particles and the container’s side walls can lead to an increased number of smaller-sized gangue. Since the small-sized gangue has a larger specific surface and can adsorb more bound water, the pressure bleeding rate gradually decreases. The second stage is the main stage of gangue slurry bleeding and has the longest duration.

(3)

Phase 3. The pressure bleeding rate gradually stabilizes, and all free water in the gangue slurry has already been secreted. The maximum value of the pressure bleeding rate φ_max can be used to describe the bleeding characteristics of the gangue slurry and the moisture dosage of the CBB. At the beginning of the third stage, there is already less free water in the gangue slurry. Under pressure, it can soon be secreted, so the duration of the third stage is significantly shorter than the second stage. The variation curve of φ_max with respect to the axial pressure of the gangue slurry in S1–S8 groups can be drawn. As shown in Figure 7, as the axial pressure increases, the φ_max of the gangue slurry with the same ratio gradually increases. When the axial pressure increases from 2 MPa to 8 MPa, the φ_max of the gangue slurry increases by 15.16% to 16.63%. The φ_max of the gangue slurry with particle sizes ranging from 0 to 2 mm is significantly greater than that with particle sizes ranging from 0 to 5 mm, which is also related to the water retention capacity of small-sized gangue particles. The static bleeding rate of the gangue slurry is about 15%, and after being subjected to continuous pressure, the bleeding rate can reach about 70%.

3.2. Permeability Variation Law of the CBB

3.2.1. The Influence of Axial Compression on the Permeability of the CBB

Figure 8 shows the permeability change curve of the CBB in the S1–S20 groups.

From Figure 8, it can be seen that:

(1)

The permeability of the CBB shows an exponential decreasing trend with the increasing axial pressure, and the fitting degree of the fitting curve is good, with an R² above 0.99. When 4% and 6% cement is added to the CBB with a 0–5 mm particle size, permeability decreases by one order of magnitude with an increase in axial pressure from 2 MPa to 8 MPa. This indicates that the sensitivity of the permeability of the CBB to axial pressure was improved after adding the cement to the gangue slurry.

(2)

When 0% and 2% cement dosage is added to gangue slurry with particle sizes ranging from 0 to 5 mm, the permeability of the CBB is almost linearly negatively correlated with axial pressure. When 4% and 6% cement dosage is added to gangue slurry with particle sizes ranging from 0 to 5 mm, the permeability of the CBB decreases with the increasing axial pressure. It indicates that as the cement dosage increases, the permeability of the CBB transitions from linear to nonlinear, and a cement dosage of 2% is the critical value. The permeability variation in the CBB with a particle size range of 0–2 mm also exhibits nonlinear characteristics, indicating that the reduction in particle size has the same effect as the addition of cement. As the curve illustrates in Figure 8, when the axial pressure is the same, the permeability of the CBB with a particle size range of 0–2 mm is equivalent to adding 2–4% cement in the CBB with a particle size range of 0–5 mm.

3.2.2. The Influence of Particle Size Range on the Permeability of the CBB

By comparing the permeability changes of the CBB in the S1–S4 and S5–S8 groups, the influence of the particle size range on the permeability of the CBB can be obtained, as shown in Figure 9.

From Figure 9, it can be concluded that:

(1)

The permeability of the CBB with the particle sizes ranging from 0 to 5 mm generally shows a linear decreasing trend. When the axial pressure increases from 2 MPa to 8 MPa, the permeability decreases from 5.12 × 10⁻¹⁵ m² to 1.85 × 10⁻¹⁵ m², with a decrease of 63.9%. The permeability of the CBB with a particle size ranging from 0 to 2 mm decreases nonlinearly with the increase in axial pressure, and the decreasing rate gradually diminishes.

(2)

When the same axial pressure is applied, the permeability of the CBB with a particle size range of 0–5 mm is greater than that of 0–2 mm. This is because there are more coarse gangue particles in the CBB with a particle size range of 0–5 mm, leading to higher porosity. In addition, as the axial pressure increases, the difference in permeability between the two CBBs first increases and then decreases. This is because as the axial pressure increases, the coarse-grained gangue in the CBB with a particle size range of 0–5 mm is compressed and crushed into small particle size gangue by extrusion and friction and fills the void. Consequently, the voidage of the CBB with a particle size range of 0–5 mm decreases, resulting in a greater drop in permeability.

3.2.3. The Influence of Cement Dosage on the Permeability of the CBB

Figure 10 shows the curve of the permeability of the CBB with a particle size range of 0–5 mm under different additions of cement dosages. As the cement dosage increases, the permeability of the CBB with a particle size range of 0–5 mm gradually decreases. As the cement dosage gradually increases, the decreasing amplitude in permeability also becomes greater. This can be explained as follows: (1) gaps in the CBB can be filled by cement particles with a micrometer-scale particle size; (2) bonding products are generated by the cement hydration, resulting in a more tightly interconnected solid material within the CBB. The addition of cement reduces the permeability of the CBB by 1–2 orders of magnitude. For example, t 0–5 mm, the CBB without cement exhibits a permeability of 5.12 × 10⁻¹⁵ m² under an axial pressure of 2 MPa, while the CBB with a particle size range of 0–5 mm and cement dosage of 6% has a permeability of 9 × 10⁻¹⁷ m² at an axial pressure of 8 MPa. This indicates that cement exerts a significant optimization effect on the waterproof performance of the CBB.

3.3. Determination Method for Permeability Level of the CBB

At present, there is no unified regulation for the waterproof level of backfill materials. The evaluation of the waterproof performance of backfill materials can be achieved by referring to the classification of permeability characteristics in coal mine strata. The permeability classification of rock and soil is given based on the permeability coefficient K of rock and soil in the Code of Water Pressure Test in Borehole for Water Resources and Hydropower Engineering (SL31–2003) [32,33]. The CBB of gangue slurry belongs to porous media [34,35], and there is a conversion relationship between its permeability coefficient K and permeability k, as shown in Equation (3).

$$K=k\frac{\rho g}{\mu }=k\frac{\gamma }{\mu }$$

(3)

where K is the permeability coefficient, cm/s; ρ is the density of water, g/cm³; g is the gravity coefficient; g = 9.8 N/kg; γ is the weight of water, 9.8 kN/m³; µ is the viscosity coefficient of water, 2.98 MPa·s.

Equation (3) can be used to convert the permeability coefficient K in the classification of rock and soil permeability into permeability k, and the permeability level of the CBB is obtained, as shown in

Table 2. Permeability level of the CBB.

Permeability Level	Permeability Coefficient K (cm/s)	Conversion Permeability k (10−15 m2)
Extremely poor permeability	K < 10−6	k < 3.04
Poor permeability	10−6 ≤ K < 10−5	3.04 ≤ k < 30.4
Weak permeability	10−5 ≤ K <10−4	30.4 ≤ k < 304
Medium permeability	10−4 ≤ K <10−2	304 ≤ k < 3040
Strong permeability	10−2 ≤ K < 1	3040 ≤ k < 30,400
Extremely strong permeability	K > 1	k > 30,400

.

The permeability classification of the CBB in Table 2 is used, and the permeability results in Figure 8 are reclassified, as shown in Figure 11. It can be seen that the permeability levels of all CBBs are within the range of poor permeability and extremely poor permeability, indicating that the gangue slurry has a good water-blocking effect after backfilling into the mining space. As shown in Figure 11, when the axial pressure exceeds 6 MPa, the permeability levels of CBBs all reach extremely poor permeability. When the cement dosage reaches 6%, the permeability level of the CBB under all axial pressure conditions has extremely poor permeability. Increasing the axial pressure of the gangue slurry, reducing the particle size range of the gangue slurry, and adding cement can all improve the permeability level of the CBB and improve its permeability. In the actual construction process, the permeability level of the CBB should be comprehensively determined based on the injection layer of gangue backfilling and the requirements for mine water prevention and control, so as to determine the composition and ratio of gangue slurry.

4. Conclusions

In this study, a multi-field coupled seepage test system for backfill materials was developed, and the pressure bleeding characteristics, permeability changes, and the determination method for the permeability level of the CBB and its optimization methods were investigated. The main conclusions are as follows:

(1)

The developed multi-field coupled seepage test system for backfill materials mainly consists of five parts: material loading module, axial stress loading module, liquid seepage module, seepage outlet measurement module, and data acquisition and processing module. With a maximum axial pressure of 30 MPa and a maximum seepage pressure of 15 MPa, this system can monitor and collect key parameters, such as stress, displacement seepage flow rate, and water pressure, in real time during the experiment.

(2)

After being injected into the mining space, the gangue slurry is subjected to pressure and local stress from the overburden, resulting in a pressure bleeding phenomenon. After the free water is completely drained, the CBB is formed. A laboratory preparation method for the CBB is proposed, and the waterproof performance of the gangue grouting backfill materials is evaluated by testing the permeability of the CBB.

(3)

The pressure bleeding of the gangue slurry exhibits obvious phased characteristics. In the first stage, it shows a linear growth trend with axial loading time. In the second stage, the increasing amplitude of pressure bleeding gradually weakens. In the third stage, pressure bleeding gradually stabilizes and finally stops. Moreover, the second stage has a significantly longer duration compared to the first and third stages.

(4)

The permeability of the CBB decreases exponentially with the increasing axial pressure. After adding cement to the gangue slurry, the permeability of the CBB exhibits a higher sensitivity to axial stress. As the cement dosage increases, the permeability change in the CBB transfers from linear to nonlinear, and the cement dosage of 2% is the turning point of the linear to nonlinear transition. With the same axial stress, the permeability of the CBB with a particle size range of 0–5 mm is greater than that of 0–2 mm. This is because there is a higher porosity in the CBB with a particle size range of 0–5 mm. As the cement dosage increases, the permeability of the CBB with a particle size range of 0–5 mm gradually decreases, and the addition of cement reduces the permeability of the CBB by 1–2 orders of magnitude.

(5)

The permeability level of the CBB of gangue slurry is within the range of poor permeability and extremely poor permeability. After entering the mining space, the gangue slurry has a good water-blocking effect. Increasing the axial stress on the gangue slurry and adding cement can both improve the permeability level of the CBB and improve its waterproof performance.