Introduction

Underground rock mass engineering is generally affected by groundwater, leading to the occurrence of various underground rock mass engineering disasters (Lin et al. 2021). Water is one of the most active factors inducing geological disasters, engineering rock mass undergoes deformation, instability, and destruction processes, often involving water (Tan and Yan 2004; Tang et al. 2005). In the process of coal resource mining, the interaction between water and rock coupling can affect the stability of engineering rock mass, posing a serious threat to coal mine safety production (Gu et al. 2021; Zhang et al. 2019a, b).

In recent years, many scholars at home and abroad have conducted a series of studies on water–rock interaction. Previous studies have shown that the presence of water can weaken the strength characteristics of coal and rock masses and reduce the engineering stability of underground rock masses (Tang et al. 2010; Yao et al. 2021; Zhao et al. 2021; Zhao et al. 2021; Yin et al. 2023a, b). Jin et al. (2023a; b) studied the dynamic strength and deformation characteristics of rocks under the action of high-pressure water, as well as the dynamic energy dissipation and failure characteristics of red sandstone under different water contents. They believed that water pressure plays a role in promoting and then inhibiting the energy absorption of rocks. In 1994, Kang (1994) first applied the theory of damage mechanics to analyze the influence of water on rock strength and deformation and established the evolution equation of rock damage variables closely related to stress state when encountering water. Zhang et al. (2017) studied the characteristics of the influence of water content on the damage and deformation of gas containing coal and rock under loading and unloading. They believe that the higher the water content, the lower the triaxial strength of the coal sample, and the faster it loses its ability to resist external loads. In addition, plastic deformation is enhanced, brittle failure is weakened, and the increase in Watts flow rate is smaller. Yang et al. (2023) studied the comparative analysis of deformation failure and energy properties of raw coal and sandstone under uniaxial compression. Zhang et al. (2019a, b) believed that the structural damage of hydrate containing sediment has a significant impact on the multi-field coupling process of hydrate decomposition, and its impact gradually increases with the increase of decomposition time. Li et al. (2021) derived damage variables characterizing the overall damage of coal and rock under different water contents based on elastic damage mechanics, established a segmented damage constitutive model for coal and rock under water force coupling, and obtained the deformation law of coal and rock with water contents. Lai et al. (2020) studied the softening effect and multi-scale damage evolution characteristics of coal and rock samples under hydraulic coupling. It was found that the microscale softening effect of coal and rock samples with different water content states varies significantly. Water can effectively slow down the damage evolution process of coal samples, ultimately weakening the impact tendency of coal samples. Chen et al. (2023) analyzed the mechanical properties and damage characteristics of coal rock composite under water rock interaction and constructed a composite damage model based on immersion time. As the immersion time increases, the damage degree of the composite increases. Deng et al. (2017) analyzed the degradation and damage characteristics of sandstone under the action of water and rock. The degradation laws of parameters such as compressive strength, elastic modulus, cohesion, and internal friction angle of sandstone are obvious, and established damage evolution equations for each mechanical parameter. Esatyana et al. (2020, 2021) studied the characterization of anisotropic fracture toughness of shale using nanoindentation, the characterization of fracture toughness using machine learning (Alipour et al. 2021), and its application in nanoindentation of shale cuttings and core measurement (Sakhaee-Pour et al. 2019), and analyzed the geological mechanics of shale and carbonate rocks at two scales and the macroscopic Young's modulus of shale based on nanoindentation (Li and Sakhaee-Pour 2016, 2018).

The above studies mostly focused on the deformation, failure, and loss characteristics of harder coal and rock under the influence of water, while similar materials are often used to replace the deformation and failure characteristics of soft rock. At present, some important research results have been achieved on the deformation and failure characteristics of rock-like materials. Su et al. (2022) studied the effect of fracture dip angle on rock mass fracture instability and deformation field, and believed that the peak shear stress of rock-like materials shows an increasing decreasing increasing decreasing trend with the increase of fracture dip angle. Liu et al. (2020) investigated the mechanism by which the water cement ratio affects the physical and mechanical properties of rock-like materials under freeze–thaw conditions. As the water cement ratio increases, the uniaxial compressive strength and elastic modulus of rock like materials first increase and then decrease. Li et al. (2023) studied the influence of the number of non-consecutive joints on the dynamic mechanical properties and failure characteristics of a rock-like material, the increased number of joints will change the stress–strain curve from a single-peak shape to a multi-peak shape and enhance the ductility. Liu et al. (2023) proposed a modified framework of the phase-field model coupled with plasticity, to investigate the effects of loading rates on the fracture mechanism in elastoplastic rock-like materials. Fu et al. (2022) studied the specification of tensile crack growth under punch shear test, and determined and compared the direct and indirect tensile strength of gypsum and concrete (rock-like materials). Gao et al. (2022) studied the dynamic fracture mechanism of deep mudstone and sandstone under high in situ stress, the results show that materials similar to mudstone and sandstone mainly show axial fracture tensile failure and crushing failure. The above analysis shows that, in response to the characteristics of low strength, poor integrity, and difficulty in coring of soft rock, many scholars have used similar materials to prepare rock like samples to study the strength, deformation, and damage characteristics under different conditions, and have achieved a series of important research results. However, there is relatively little research on the strength and deformation characteristics of rock like samples under different immersion conditions. Therefore, this study utilizes the fine river sand as the aggregate, cement and gypsum as the bonding materials, and borax as the retarder to make RLS. Firstly, DIT were conducted on them, and then uniaxial compression tests were conducted on DIT of RLS. Finally, the mechanical and energy characteristics of the RLS under water-stress coupling were analyzed. The research results are expected to provide reference for solving the stability difficulties of soft rock chambers.

Test preparation

Sample preparation

This study utilizes fine river sand as aggregate, cement and gypsum as bonding materials, and borax as a retarder, a RLS with a sand cement ratio of 1:1 and a ratio of 155 was prepared. The mold is a plastic cylindrical shape with a height of 100 mm and a diameter of 50 mm, as shown in Fig. 1.

Fig. 1
figure 1

Sample mold

Full size image

After the necessary supplies were prepared, the production of RLS can begin. The production process was as follows:

  1. (1)

    Preparation of raw materials: Weigh the raw materials according to the dosage of sand, cement, and gypsum in each test number in the table.

  2. (2)

    Assembly of mold: To facilitate mold removal, apply a thin layer of oil brush to the inner wall of the mold.

  3. (3)

    Mixing: Add weighed sand, cement, and gypsum to the container in sequence, mix evenly, then slowly add an appropriate amount of water, continue stirring, and add an appropriate amount of borax as a retarder midway and mix evenly.

  4. (4)

    Loading: Fill the mixed material into the mold in sequence and shake evenly.

  5. (5)

    Demolding: Under room temperature conditions, let it stand for 24 h, demold, and number the test pieces separately.

  6. (6)

    Repeat the above steps to make all required samples, and place the prepared samples in a constant temperature oven (20 °C) for 28 days, as shown in Fig. 2a. Each group of uniaxial compression tests requires three specimens, with a total of 20–30 specimens prepared. The prepared specimens are shown in Fig. 2b.

Fig. 2
figure 2

Constant temperature curing box and standard sample

Full size image

Uniaxial compression test

The WDW-300 microcomputer controlled electronic universal testing machine (see Fig. 3) was used to conduct uniaxial compression tests on RLS under DIT. During the test, displacement control was used to load axial pressure (with a loading rate of 0.1 mm/min), and stress, strain, time, and other data could be collected simultaneously during the entire loading process.

Fig. 3
figure 3

WDW-300 microcomputer controlled electronic universal testing machine

Full size image

Analysis of experimental results

Water absorption process and phenomenon characteristics of RLS

We will place the prepared RLS into a water immersion container and add tap water to raise the water surface above the surface of the specimen. The IT is 0, 4, 8, 12, 24, and 48 h, respectively. After the immersion test was completed, the mass was measured and the MC of RLS was calculated. The basic parameters of immersion time are shown in Table 1.

Table 1 Basic parameters of IT
Full size table

Observe the surface condition of the sample during the immersion process, as shown in Fig. 4. When the sample is first placed in water, it bubbles during immersion, indicating that the voids in the sample expand and gas is expelled. The bubbles on the surface of the sample are dense after being immersed in water for 4 h. After immersion in water for 12 h, a bright film was formed on the surface, and some specimens exhibited glossy ripples. After soaking for 48 h, there were no bubbles on the surface of the sample, and there were local flake shedding phenomena on the surface of the sample. With the increase of time, there were a certain amount of peeling debris at the bottom around the sample.

Fig. 4
figure 4

Immersion samples

Full size image

To understand the variation of RLS WC with time, draw a curve of the relationship between sample WC and IT, as shown in Fig. 5. The MC of the sample shows an upward trend with the increase of IT, and the growth rate of MC can be divided into two stages: the rapid increase stage (the sample absorbs moisture fastest within 12 h, the MC reaches 3.31% after 4 h of immersion, and the MC reaches 8.72% after 12 h of immersion, showing a rapid growth); slow growth stage. (The WC reaches 10.98 and 13.10% respectively after soaking for 24 and 48 h, and the increase rate gradually slows down.) It can be seen that the RLS has not yet reached a saturated state after being immersed in water for 48 h.

Fig. 5
figure 5

Relationship curve between IT and MC

Full size image

Stress–strain curve

Due to the influence of water absorption, the internal pore structure of RLS changes, leading to changes in their bearing capacity. Based on this, analyze the impact of IT on the UCS, EM, and other factors of RLS. Generally speaking, when subjected to external forces, the sample undergoes stages such as microcracks being compacted, damage evolution starting, microcracks expanding and penetrating, and macroscopic cracks generating. Due to different MC, the failure forms of the sample vary. Combining the stress–strain curve of the specimen with the crack closure and development process during compression, the stress–strain curve is divided into compaction stage, linear elastic stage, failure stage, and post peak stage. Figure 6 shows the four stages of the axial stress–strain curve of RLS under DIT.

Fig. 6
figure 6

Stress–strain curves of samples under DIT

Full size image
  1. (1)

    Crack compaction stage: With the increase of axial load, it will be found that the axial stress–strain curve shows an upward concave shape, indicating that the required stress gradually increases and the tangent slope gradually increases when the same amount of strain is generated.

  2. (2)

    Linear elastic stage: The stress–strain curve in this stage follows a linear variation pattern, with the slope of the straight section being the EM of the sample. Under external action, the internal microcracks of the sample close, the pores compact, and enter continuous deformation.

  3. (3)

    Destruction stage: At this point, the specimen undergoes continuous compression deformation, and the curve becomes convex as the stress increases. The reason lies in the sliding and unstable propagation of new cracks in the sample. The sample continuously undergoes micro fractures, resulting in significant inelastic deformation and an increase in sample volume.

  4. (4)

    Post peak stage: The specimen has formed a macroscopic fracture surface. If the load is continuously applied, the load-bearing capacity of the specimen will sharply decrease, even to zero, and at this time, the specimen will undergo significant plastic deformation. The micro fracture continues to develop until the adhesive force is lost.

Table 2 presents the mechanical parameters of the samples at DIT after uniaxial compression testing. The UCS of the samples ranges from 1.53 to 10.68 MPa, and the EM varies between 0.203 and 1.389 GPa, with a certain variation pattern: as the IT increases, the UCS and EM both show a decreasing trend. The AUCS (average uniaxial compressive strength) of the non immersed sample is 9.934 MPa, and the AEM (average elastic modulus) is 1.179 GPa. After soaking the sample for 4 h, when the MC was 3.31%, it was found that the UCS and EM decreased by 2.734 MPa and 0.14 GPa, respectively; when the soaking time reaches 8 h and the MC is 6.11%, the UCS decreases by 6.057 MPa and the EM also decreases by 0.495 GPa; when the MC is 8.72%, the UCS and EM decrease by 6.527 MPa and 0.789 GPa, respectively; when the MC reaches 10.98%, the UCS decreases by 6.941 MPa and the EM decreases by 0.882GPa. When the sample was immersed in water for 48 h and the MC reached 13.10%, the UCS decreased by 8.23 MPa and the EM decreased by 0.945 GPa. Compared to the non immersed sample, the decrease in UCS and EM was 82.85 and 80.15%, respectively. It can be seen that with the increase of MC, the EM and UCS of RLS decrease significantly.

Table 2 Mechanical parameters of each sample at DIT
Full size table

Effect of water on the EM of RLS

Figure 7 shows the variation curve of EM with IT. The average value of the EM under non immersion conditions is 1.179 GPa, and the EM of the sample after 48 h of immersion is 0.234 GPa. The EM of the sample gradually decreases with the increase of IT. When the soaking time is 12 h, the EM of the sample decreases by 0.789 GPa, and then the decrease gradually decreases. The EM is an important indicator of the compressive elastic deformation process of a specimen, and its size reflects the strength of the specimen's resistance to deformation and failure. The main factors affecting the existence of EM are the presence and variation of water. During the process of elastic deformation, the internal structure of the sample changes and the movement of cracks and pores, resulting in a significant decrease in friction force under the action of water. The action of water will cause the sample to undergo significant deformation under relatively small loads, reducing its resistance to deformation and failure, and reducing its EM.

Fig. 7
figure 7

Variation curve of EM with IT

Full size image

Figure 8 shows the variation curve of EM change rate with IT, and the change rate of the EM increases with the increase of duration of water-immersion. When the soaking time is 0 h, the change rate of the EM is 0. When the soaking time is 12 h, the change rate of the EM is relatively large (about 67.65%). Afterward, as the soaking time increases, the change rate of the EM tends to stabilize, with changes of 74.81 and 80.15% at 24 and 48 h, respectively. The binding force of the internal components of the sample decreases with the increase of soaking time, and when the sample reaches a certain degree of saturation, it will no longer continue to change.

Fig. 8
figure 8

Variation curve of EM change rate with IT

Full size image

Relationship between MC, UCS, and EM

To represent the relationship between MC, UCS, and EM, the least squares method was used to fit the experimental data and establish an appropriate relationship, as shown in Fig. 9.

Fig. 9
figure 9

Variation curve of uniaxial compressive strength, EM with MC

Full size image

From Fig. 9, it can be seen that as the MC increases, the UCS and EM gradually decrease. Fitting them shows that the correlation coefficients of the two after fitting are 0.9243 and 0.8435, respectively. This can well illustrate the relationship between MC and UCS and EM. As the MC increases, both show a downward trend.

Softening properties

Generally speaking, water can reduce the strength of RLS, which is the softening effect of water. It is usually represented by the "softening coefficient," which is the ratio of the strength of rock specimens in saturated and dry states. Figure 10 shows the relationship between the SC of the sample and the soaking time. It can be seen that as the IT of the sample increases, the UCS and SC of the sample gradually decrease. During the immersion process of the sample, the interaction between the two is relatively strong, with a SC of 0.725 after 4 h of immersion, and less than 0.4 after 8 h of immersion, demonstrating strong softening. With the continuous IT, the softening effect also gradually decreases. As the IT increases, the MC of the sample gradually increases.

Fig. 10
figure 10

Variation curve of average MC, SC with duration of water-immersion

Full size image

Figure 11 shows the variation curve of SC with MC of RLS. It can be seen that there is a certain relationship between the MC and SC of the sample, and the SC of the sample shows a decreasing trend with the increase of MC. The relationship between WC and strength reduction coefficient was fitted, and the correlation coefficient reached 0.9429, showing a good Logarithmic function relationship.

Fig. 11
figure 11

Variation curve of SC with MC

Full size image

Energy evolution properties

Energy calculation principle

The deformation and destruction of the RLS is a process of energy accumulation, dissipation, and release, assuming that the loading process of the specimen does not exchange heat with the outside world, which is regarded as a closed system. According to the first law of thermodynamics, the U (absorbed total energy) of the test machine is (Xie et al. 2008):

$$ U = U_{{\text{d}}} + U_{{\text{e}}} $$
(1)

where Ud is the dissipated energy; Ue is the releasable elastic energy in the order of MJ/m3.

Figure 12 shows the relationship between Ud and Ue, with the total energy input to the specimen from the testing machine being the area enclosed by the stress–strain curve and the strain axis (OCB):

$$ U = \int_{0}^{{\varepsilon_{1} }} {\sigma_{1} } d_{{\varepsilon_{1} }} $$
(2)
Fig. 12
figure 12

Dissipated energy versus releasable elastic energy for uniaxial compression process

Full size image

The integral in Eq. (2) is summed over the area of the tiny trapezoidal bars using the stress–strain curve:

$$ U = \sum\limits_{j = 0}^{m} {\int {\frac{1}{2}\left( {\sigma_{1}^{j} + \sigma_{1}^{j + 1} } \right)\left( {\varepsilon_{1}^{j} + \varepsilon_{1}^{j + 1} } \right)} } $$
(3)

where m is the number of samples calculated for the axial stress–strain curve at any point during the test; j is the sampling point; σ1, ε1 are the stress and strain at the sampling point, respectively.

The Ue can be released can be calculated using the area of the triangle enclosed by ABC:

$$ U_{{\text{e}}} = \frac{{\sigma_{1}^{2} }}{{2E_{0} }} $$
(4)

For the convenience of calculation, the E0 unloading modulus in Eq. (4) is approximated by the EM. Based on Fig. 6, it can be seen that the specimen exhibits better linear characteristics before the peak. For energy calculation, the unloading modulus is replaced by the EM. The dissipated energy can be calculated by selecting the area enclosed by the stress–strain curve of the loading process and the unloading curve (OCA):

$$ U_{{\text{d}}} = \int_{0}^{{\varepsilon_{1} }} {\sigma_{1} } d_{{\varepsilon_{1} }} - \frac{{\sigma_{1}^{2} }}{{2E_{0} }} $$
(5)

Energy storage, absorption, and dissipation properties of rock-like samples

Based on Eqs. (3)–(5), the experimental data were processed, and the curves of energy versus axial stress during uniaxial compression of each specimen at variable loading rates are given in Fig. 13.

Fig. 13
figure 13

Relationship curve between axial stress, energy, and axial strain of RLS at DIT

Full size image

From Fig. 13, it can be seen that the energy and axial stress relationship curves of various RLS exhibit similar variation characteristics under DIT. In the early stage of loading, the U and the Ue of the sample show an upward concave shape with the increase of axial pressure. Part of the energy absorbed by the sample is converted into the released elastic performance, while the other part is dissipated by the compaction and closure of the primary pores and fractures. At this stage, the energy absorbed and released by the sample is relatively less, while the Ud is relatively more, corresponding to the compaction and closure stage of the primary fractures of specimen; as the axial pressure increases, the U by specimen and the slope of the axial stress curve, as well as the slope of the Ue and the axial stress curve, gradually increase. The testing machine continues to work on the sample, and the U and the Ue gradually increase, while the rate of change in Ud is relatively small. The energy absorbed and released by RLS is relatively high, while the Ud is relatively low, this stage corresponds to the linear elastic stage of the stress–strain curve; as the axial compression continues to load, a large number of newly formed microcracks inside the RLS generate, propagate, evolve, and gradually penetrate, leading to significant plastic deformation. When the axial stress reaches the maximum capacity that the sample can withstand, the sample enters the post peak deformation and failure stage, forming a macroscopic penetration of the crack surface. The sample can still absorb a small amount of energy from the outside when sliding along the crack, but the rate of absorbing the total energy is greatly slowed down. At this time, the stress is released, the releasable elastic performance is rapidly reduced, and the dissipation energy is rapidly increased. Extract the energy values at the peak of RLS, as shown in Table 3.

Table 3 Energy properties at peak values of various RLS under DIT
Full size table

From Table 3, it can be seen that under the soaking time of 0 h, the U before the peak of the sample ranges from 4.726 to 7.336 × 10–2 MJ/m3, with an average of 5.654 × 10–2 MJ/m3. The elastic energy that can be released at the peak is 3.940–4.336 × 10–2 MJ/m3, with an average of 4.178 × 10–2 MJ/m3; when the soaking time is 4 h, the U before the peak of the sample ranges from 2.757 to 5.182 × 10–2 MJ/m3, with an average of 4.365 × 10–2 MJ/m3. The elastic energy that can be released at the peak is 2.131–3.133 × 10–2 MJ/m3, with an average of 2.522 × 10–2 MJ/m3; when the soaking time is 8 h, the U before the peak of the sample ranges from 1.157 to 2.975 × 10–2 MJ/m3, with an average of 1.975 × 10–2 MJ/m3. The elastic energy that can be released at the peak is between 0.808 and 1.713 × 10–2 MJ/m3, with an average of 1.178 × 10–2 MJ/m3; when the soaking time is 12 h, the U before the peak of the sample ranges from 1.122 to 1.887 × 10–2 MJ/m3, with an average of 1.628 × 10–2 MJ/m3. The elastic energy that can be released at the peak is 1.012–1.638 × 10–2 MJ/m3, with an average of 1.421 × 10–2 MJ/m3; When the soaking time is 24 h, the U before the peak of the sample ranges from 1.249 to 3.195 × 10–2 MJ/m3, with an average of 1.999 × 10–2 MJ/m3. The elastic energy that can be released at the peak is 1.113–1.391 × 10–2 MJ/m3, with an average of 1.285 × 10–2 MJ/m3; When the soaking time is 48 h, the total absorbed energy before the peak of the sample ranges from 0.850 to 1.090 × 10–2 MJ/m3, with an average of 1.000 × 10–2 MJ/m3. The elastic energy that can be released at the peak is between 0.482 and 0.743 × 10–2 MJ/m3, with an average of 0.596 × 10–2 MJ/m3. It can be seen that over 60% of the U during the loading process of the sample is accumulated in the form of Ue, while less than 40% of the total energy is dissipated by various defects in the sample during the compaction, sliding, and yield stages of the new microcracks. The Ue can promote the expansion, penetration, and sliding of the new cracks inside the sample after the peak, until the sample completely loses its load-bearing capacity. As the MC increases, the U and the Ue at the peak of the sample decrease exponentially as a whole, as shown in Fig. 14.

Fig. 14
figure 14

Relationship curve between total U, Ue, and MC

Full size image

Energy storage rate and dissipation rate of rock-like samples

Figure 15 shows the relationship curves between energy ratio and stress ratio of RLS under uniaxial compression under variable loading rate.

Fig. 15
figure 15

Relationship curve between Ue/U, Ud/U, and σ1/σmax under DIT

Full size image

From Fig. 15, it can be seen that the Ue/U of RLS shows a rapid decrease, rapid increase, slow increase, and decrease to a rapid decrease with the increase of stress ratio. The change characteristics of the Ud/U of the sample are opposite with the increase of stress ratio. It can be divided into three stages of change, with the first stage being the primary fissure compaction and closure stage (σ1/σmax < 0.25), during this stage, Ue/U rapidly decreases and Ud/U rapidly increases. The second stage is the stage where the sample continuously absorbs energy (0.25 < σ1/σmax < 0.8), mainly stored in the form of elastic strain energy that can be released, during this stage, Ue/U gradually increases, and the slope of increase continuously decreases. Ud/U gradually decreases, and the rate of decrease continuously decreases. During this process, micro cracks slowly form inside the sample. The third stage is the energy dissipation stage (σ1/σmax > 0.8), during this stage, Ue/U slowly decreases and then rapidly decreases, while Ud/U slowly increases and then rapidly increases. During this process, the internal microcracks of the sample further expand and form a penetrating fracture surface.

Damage characteristics based on energy evolution

The deformation damage of RLS is closely related to dissipation energy, and its damage variable can be defined by the ratio of dissipation energy to the maximum dissipation energy at the time of RLS failure:

$$ D = \frac{{U_{d} }}{{U_{d\max } }} $$
(6)

In the Eq. (6), Ud is the dissipated energy per unit volume, Udmax is the maximum dissipated energy during specimen failure.

According to the calculation results of Eq. (6), it can be seen that the damage change characteristics of RLS comply with the variation law of Logistic equation, and the equation is as follows Eq. (7):

$$ D = \frac{a}{{1 + e^{{b - c \cdot \varepsilon_{1} }} }} $$
(7)

In the Eq. (7), a, b, and c are constants.

Based on Eq. (7), fit the relationship curve between the damage variable and axial strain of the sample under DIT, as shown in Fig. 16. The fitting parameter are shown in Table 4.

Fig. 16
figure 16

Relationship curve between damage variables and axial strain of each sample under DIT

Full size image
Table 4 Fitting parameters under DIT of RLS
Full size table

Failure forms

The failure modes of RLS under uniaxial compression with DIT are shown in Fig. 17.

Fig. 17
figure 17

Failure morphology of RLS

Full size image

Figure 17 shows that the failure modes of RLS are mainly tensile failure, shear failure, and tensile shear composite failure. After water enters the interior of RLS, it will reduce the friction coefficient between fracture surfaces and weaken the bonding between particles. After the RLS contains water, the lubrication effect of water will reduce the friction coefficient between fracture surfaces and reduce the relative sliding difficulty of fracture surfaces. Water also weakens the bonding between particles in RLS, changes the connection state between mineral particles, and increases the possibility of shear failure of the samples. As the pores and cracks close, the presence of water will generate a certain pore pressure on the rock-like sample. After the sample contains water, the internal pores and cracks gradually close under the load, and the water in the pores and cracks cannot be discharged in time. This will generate a certain pore water pressure inside the sample, causing tensile stress to be applied at the stress concentration point of the primary crack tip of the coal rock, causing crack expansion. After the RLS enters the plastic stage, plastic deformation occurs, and the stress state is complex. At the same time, it is subjected to tensile and shear actions, thereby increasing the possibility of RLS tensile shear composite failure. Which action has a greater impact on the rock, then the specimen will be prone to corresponding forms of damage after being hydrated, which is related to the structure and properties of the sample itself.

Conclusions

This study utilizes the similar material method, fine river sand was selected as the aggregate, cement and gypsum as the cementation material, and borax as the retarder, cylindrical RLS (rock-like specimens) with 1:1 sand-binder ratio were prepared, and uniaxial compression tests were carried out on the RLS under DIT (different immersion time) (0, 4, 8, 12, 24, and 48 h). The mechanical and energy properties of RLS under the action of water-stress coupling are revealed, and the research results are of great value to the deformation and failure characteristics of soft rocks. Some condensed conclusions are as follows:

  1. (1)

    The longer the immersion time, the higher the WC (water content) of the RLS. As the MC (moisture content) increases, the UCS (uniaxial compressive strength) and EM (elastic modulus) of RLS decrease exponentially, while the SC (softening coefficient) decreases logarithmically. The stress–strain curve of RLS under uniaxial compression can be divided into four stages: compaction stage, linear elastic stage, failure stage, and post peak stage.

  2. (2)

    The energy dissipation principle of RLS under uniaxial compression was analyzed, and the energy evolution characteristics of RLS under water-stress coupling were revealed. The U (absorbed total energy) before peak, the Ue (releasable elastic energy), and the WC of the fitted RLS all decrease exponentially.

  3. (3)

    The relationship curves of Ue/U (released elastic energy ratio) and Ud/U (dissipated energy ratio) of RLS during uniaxial compression with the σ1/σmax (axial stress ratio) can be divided into three stages of change, namely the stage of primary fissure compaction and closure (σ1/σmax < 0.25), continuously absorbing energy stage (0.25 < σ1/σmax < 0.8) and energy dissipation stage (σ1/σmax > 0.8).

  4. (4)

    The D (damage variable) was defined by the ratio of Ud (dissipated energy) to the Udmax (maximum dissipated energy) at failure time of RLS, the fitting of the relationship between the damage variable and axial strain conforms to the Logistic equation.