The established mechanical property and permeability test platform for grouting-based cemented coal bodies are composed of a test device for the sealing material properties, a grouting cementation test device, and an analytical test device for cemented coal bodies. The test device for the sealing material properties is equipped with a ZKS-100 gelation time meter and a consistometer, and the grouting cementation test device consists of a detachable cemented coal body mold that was designed independently and a manual grouting equipment, as shown in Figure 1 and Figure 2. To be specific, the artificial core molds are made from high-quality stainless steel, and their working pressure is no more than 40 MPa. Also, they can be divided into large-sized and small-sized molds. While the large-sized mold is utilized to make cylinders with a height of 100 mm and a diameter of 50 mm from coal samples, the small-sized mold is capable of producing cylinders with a height of 50 mm and a diameter of 25 mm. As its motion system, the press is supported by a high-quality straight sliding bearing. Considering that the press is independent of the running resistance, it can be flexibly operated, ensuring that all the samples are uniformly pressed. In terms of the booster system, it is a fast and stable jack that is easy to operate. At last, the analytical test device for cemented coal bodies includes an RMT-150 rock mechanical test system and a STY-2 gas permeameter. Respectively, the RMT-150 system is under digital control, has a high degree of automation, and can be used to test the mechanical properties of various materials, including rock. The maximum output of the vertical hydraulic cylinder is 1,000 kN, the piston stroke is 50 mm, the deformation rate is 0.0001–1 mm/s, the loading rate is 0.01–100 kN/s, and the displacement sensor accuracy is 1.5‰ F.S. The STY-2 gas permeameter is designed based on the Darcy law, and it was adopted to measure the gas capability to penetrate porous rocks at a certain pressure difference.

PMCC samples with diverse additive ratios were selected as the main material to test the viscosity and gelation time along with the permeability of cemented coal bodies and to perform unconfined compression strength experiments. According to different ratios, coal particles with sizes ranging from 4,000 to 8,000 μm were screened, mixed in a ratio of 1:1, and subsequently compressed into cemented coal bodies (Figure 3). Separately, cemented coal bodies with sizes of 50 × 100mm (Figure 3A) and 25 × 30 mm were used to perform mechanical property experiments and relevant permeability tests, respectively. Before the experiment, the samples were dried in a 105 °C drying box, and then placed in a dryer to wait for the samples to come to room temperature for testing.

The water–cement ratio refers to the mass ratio of water to cement in cement grouts (i.e., mw/mc), and it is an important influencing factor with regard to the effects of grouting. If the water–cement ratio is excessively high, a lot of fine cracks may be generated after the hardening of a cemented body. Consequently, the strength of the cemented body is lowered, but its permeability is increased. On the contrary, colloids fail to be sufficiently formed when the water–cement ratio is too low, causing the grout workability to decline and making it difficult to form compact cemented bodies. Moreover, if the unreacted cement undergoes a secondary hydration reaction after the cemented body is hardened, the hydration products may produce swelling stresses, leading to more cracks. Thus, optimization tests under mw/mc ratios of 0.3:1, 0.5:1, 0.7:1, and 0.9:1 were conducted to test the viscosity and gelation time of cement grouts. Additionally, the cement grouts of the respective water–cement ratios were mixed with coal particles to produce cemented coal bodies through pressing. Furthermore, the cemented coal bodies were subjected to compression tests. For relevant results, please refer to Table 1. It should be noted that the used cement (strength grade: 32.5) in these tests was purchased from Harbin Cement Co., Ltd., Yatai Group.

Based on the aforementioned experiments, it was proven that the grout viscosity is rather high when the water–cement ratio is 0.3; however, it showed a non-flowing character in this condition, failing to meet the relevant testing requirements. In accordance with the experimental data in Table 1, the curves were portrayed to reflect the influencing relations of the water–cement ratio on the viscosity and compressive strength, as presented in Figure 4. Clearly, the grout viscosity gradually decreased with the increase in the water–cement ratio and was also at a moderate level, thus meeting the test requirements. Also, based on the compressive strength curves, the compressive strength of the cemented coal bodies was 3.29, 6.02, and 4.37 Mpa when the water–cement ratio was set to 0.5, 0.7, and 0.9, respectively. With the increase in the water–cement ratio, the compressive strength of the cemented coal body inclined to first go up and then go down. This indicates that the peak compressive strength was formed within this range of the water–cement ratio. Through the binomial fitting of the testing results, it became clear that the peak value appeared when the water–cement ratio ranged between 0.6 and 0.85. Based on the mentioned analysis results, we determined the best three water–cement ratios for composite sealing materials: 0.65, 0.75, and 0.85.

During borehole sealing, both the viscosity and gelation time of the used sealing materials are critical parameters. In detail, viscosity has a direct influence on the flowing properties and thus indirectly affects the grouting results. Also, studying the gelation time may help us obtain a more appropriate diffusion radius and more satisfactory sealing effects of the sealing material. Moreover, the gelation time is constituted by the initial and final setting times. The initial setting time refers to the time partially consumed by the sealing materials in losing their plasticity, and the final setting time refers to the time taken by the sealing materials to reach their inherent properties and completely lose their plasticity. The viscosity of PMCC was obtained based on the time taken by the PMCC grout to completely flow out of the consistometer. The gelation time was calculated using the penetrating resistance strength into PMCC. The water–cement ratio, polymer amount, and another two additives (A and B) were selected as influence factors and orthogonal experiments based on the four factors and three levels were designed. The viscosity and gelation time testing results of PMCC were demonstrated, as shown in Table 2.

The permeability of the cemented coal body was measured according to the flow of the gas penetrating through cemented coal bodies with given sizes under differential pressure (Yu and Page, 1991; Yu and Page, 1995; Rilem, 1999). In this process, the inlet pressure was set at 0.4 MPa and nitrogen was used as a medium. Considering that the gaseous medium falls into the category of compressible fluids and that gas volume changes may affect the corresponding flow velocity, the permeability computing equation should be modified based on Darcy law. That is, the designed formula was modified into the following equation by introducing another parameter of the mean volume flow rate: K = 2 P 0 Q 0 μ L A ⋅ ( P 1 2 − P 2 2 ) .

In the aforementioned equation, K (unit: µm²) denotes a proportional constant for the tested coal body to be permeable, P₀ (unit: MPa) denotes atmospheric pressure, Q₀ (unit: ml/min) denotes the gas flow rate at normal pressure, μ (unit: μPa·s) denotes the viscosity of the gas medium, L (unit: m) denotes the length of the tested coal body, A (unit: m²) denotes the cross-section of the tested coal body, and P₁ (unit: MPa) and P₂ (unit: MPa) denote the inlet and outlet pressures of the coal rock retainer, respectively. Table 2 lists the permeability testing results of the cemented coal body.

To perform unconfined compression strength tests for the cemented coal bodies, a strain gauge was utilized to obtain relevant strain data by touching the coal body. The corresponding testing platform is given in Figure 5. During the experiment, the surface of the cemented coal bodies was wrapped with plastic film, and the strain gauge was tested close to the surface. Clearly, the total number of horizontal and vertical strain collection sheets was 8. For the purpose of acquiring comparatively complete relation curves of the stress and strain, displacement control was conducted to gradually increase the axial loads at a velocity of 0.010 mm/s. Additionally, mechanical property experiments were performed three times in each group, and the corresponding values were averaged. Their average values were used as the final testing results. This way, the experimental results could be ensured to be more accurate. In the testing process, the used software automatically noted down the stress–strain relations, including the axial and radial strains.

As reflected in the experimental results, the cemented coal body failure process conforms to the elastic-plastic deformation curves. The typical relation curves of stress and strain are given in Figure 6. As revealed in this figure, the curves on the left reflect the relations between the stress and radial strains. At the initial stage of applying the axial load, the radial strain kept decreasing for a period of time and then gradually increased until brittle failures occured. The curves on the right represent the stress and axial strain relations. At the initial stage of applying the axial load, the axial strain gradually increased along with the load increase and eventually reached the corresponding yield limit that led to fracturing. The failure process of the cemented coal body is presented in Figure 7. Before Point A, the cemented coal body showed no obvious changes, and the elastic deformation played a crucial role. In this case, the relation curves of stress and strain were approximately straight. After Point A, the microcracks inside the cemented coal body were extended and developed under the action of stress concentration, giving rise to partial plastic deformation. Under such circumstances, the strain increases were accelerated, and the stress–strain curve gradually deviated from the straight line and bends downward. In addition, the development of microcracks led to greater lateral deformations of the cemented coal body. It is worth mentioning that microcracks were stably developed at this stage. After the failure process reached Point B, some interior microcracks began to be interconnected, and an unstable crack development took place. Accordingly, a lateral deformation was built up. Under the long-term actions of stress, cracks were unceasingly developed, eventually leading to failure. At last, the failure process arrived at Point C. The interconnected cracks inside the cemented coal body formed a failure surface. In this case, the strain growth rate obviously increased, and the peak compressive strength was acquired at Point C (Sagar and Sivakumar, 2021).

In orthogonal test analysis, R represents the range of factors. R reflects the maximum variation range of test results for target parameters when factors change. R = MAX { k 1 , k 2 , k 3 } − MIN { k 1 , k 2 , k 3 } .

The test data was analyzed using the variance method to better understand the degree of influence of factor level changes in the test process on the target parameters. First, we calculate the test mean Y ¯ : Y ¯ = 1 9 ∑ i = 1 9 Y i .

Then, we calculate the sum of the squares of deviations of all factors and errors in the experiment Q x : Q x = ∑ j = 1 3 ( k j x − Y ¯ ) 2 ( x = 1,2,3 ) .

We calculate the degree of freedom of each factor in the orthogonal experiment and the degree of freedom of the test error and divide the sum of the squares of deviation of factors and errors by their respective degrees of freedom to get the average sum of squares of deviation. The sum of the average deviations squared of the factors divided by the sum of the average deviations squared of the errors gives the variance ratio F.

The orthogonal experimental results were subjected to both range and variance analyses. The corresponding analytical results were demonstrated, as shown in Table 3. According to these results, the viscosity and gelation time completely had the same influencing rules. Among the four factors affecting the target parameters, factor B had the most significant influence on the viscosity and gelation time, followed by the water–cement ratio, factor A and the polymer amount. Moreover, the viscosity of PMCC was linearly correlated to the corresponding gelation time. The greater its viscosity, the shorter its gelation time, and vice versa. In line with the range analysis results of the compressive strength, the influence of the water–cement ratio on the uniaxial compressive strength of the cemented coal body was most significant. In this aspect, the amount of additive B took second place as far as its impacts on the uniaxial compressive strength are concerned. Furthermore, the sensitivity of the water–cement ratio to the compressive strength was similar to that of the amount of additive B. As for the polymer and the amount of additive B, their sensitivity values were rather low. Also, through the range analysis results of the permeability, it was proven that the amount of additive B had the most significant influence on the permeability of the cemented coal body, while the actions of the polymer amount and water–cement ratio ranked second and third, respectively. Additionally, the sensitivity of the amount of additive A to the target parameters was lowest.

Concerning the variance analysis viscosity results, the variances of factors W/C, polymer, A and B turned out to be 3.163, 1.000, 1.436, and 23.985, respectively. Among them, both B and W/C separately went beyond their critical values (F_α=0.10 for B and F_α=0.25 for W/C). This manifests that the water–cement ratio and additive B amount had obvious influences on the material viscosity. The variances of factors polymer and A were both below the critical values of F_α=0.25, indicating that neither polymer nor the additive A amount had significant influences and that none of them served as an important influencing factor on the experimental results. The gelation time analysis results are the same as those of the viscosity. In other words, factors W/C and B had a significant impact, while no significant influences of the polymer or A were found. Based on the compressive strength analysis results of the cemented coal body, factors W/C and B significantly affected the compressive strength. However, the polymer influence on the compressive strength was moderately significant. In terms of the mechanical properties of the cemented coal body, no significant influences of factor A were proven. With regard to the permeability analysis results, factor B significantly influenced the permeability of the cemented coal body but had no significant impacts on the target parameters.

To identify the influencing relations of the viscosity of the sealing material on the compressive strength/permeability of the cemented coal body, viscosity–compressive strength and viscosity–permeability curves were portrayed according to nine groups of different experimental systems to reflect their influencing relations (see Figure 8).

As shown by the viscosity–compressive strength curves, the viscosity of PMCC had a variation tendency identical to that of the compressive strength of the cemented coal body when the experimental system changed. This reveals that viscosity is positively correlated to compressive strength. In other words, the higher the viscosity, the greater the compressive strength. According to the viscosity–permeability curves, it can be seen that the variation tendency of permeability is contrary to that of viscosity. That is, a higher viscosity of the sealing material corresponds to a lower permeability of the cemented coal body. Therefore, the relations between the compressive strength of the cemented coal body and its permeability could be clarified. That is, the greater the compressive strength, the lower the permeability.

In accordance with the experimental results in Table 2, the cemented coal body adopting a mix ratio of System 3 had the greatest compressive strength, largest elastic modulus and an unsatisfactory permeability. In addition, the sealing materials in this system presented the highest viscosity and also featured the earliest initial setting time and poor groutability. With respect to the mix ratio selected for System 2, it made the compressive strength of the cemented coal body slightly lower than that in the case of System 3. Additionally, the permeability of the cemented coal body subjected to System 2 was lowest among the nine experimental groups. In terms of the sealing materials in System 2, their viscosity was moderate, which contributed to their grouting into the coal body. From the perspective of PMCC mix ratio optimization, System 2 had a comparatively appropriate sealing material mix ratio, thus applying obvious mechanical reinforcement effects and sealing actions on the grouting-based cemented coal bodies.

The viscosity of sealing material and the permeability and mechanical properties of cemented coal bodies are all under the obvious influence of the additive mix ratio. Through analysis, it was deemed that the hydration reaction degree of PMCC and polymer properties determined both the strength and permeability of cemented coal bodies to some extent. In the hardening process of the sealing material that contacted the coal body, a hydration reaction took place between the cement clinker and water, producing hydrates. Then, the hydrates were mutually overlapped and bound by virtue of multiple attractions, forming a cemented body structure and further producing more strength. Moreover, sufficient hydration reactions corresponded to a closer overlapping and binding of hydrates as well as to more stable structures and greater strength. As a result, the number of porous cracks within the corresponding cemented body declined and the permeability of the cemented body was thus lowered. Simultaneously, the polymer has good bonding properties and can form an interspace reticulate microstructure, which can fill the pores of hydration products, improve the density of hydration products, and produce flexible reinforcement. As a result, the strength of the cemented coal body was further improved and the permeability was reduced.