Effect of inter-layer thickness on dynamic mechanical properties of rock mass combined with hard and soft media Authors: Yuanjian Zhang, Zhehang Huang, Hui Xiao, Huajun Xue, Bin Ren, Weihong Yang, Yanbing Wang Published: 2025 ABSTRACT AbstractIn order to explore the influence of interlayer thickness on the dynamic mechanical properties of the rock mass composed of soft and hard media, taking sandstone and granite as the soft and hard rock matrix, the energy evolution and loss laws of the rock mass composed of soft and hard media under different interlayer thicknesses (0 mm, 5 mm, 10 mm, 15 mm) were studied by using the split Hopkinson pressure bar device, and the energy driving mechanism of its failure was explored. The discrete lattice spring method (DLSM) was used to simulate the crack growth, stress time and damage strain evolution of rock mass with different interlayer thicknesses. The results show that with the increase of interlayer thickness, the resistance of interlayer to the crack growth and development of rock mass is gradually enhanced, and the energy consumption is gradually increased. The energy driving mechanism for the failure of the composite rock mass containing interlayers of soft and hard media is: under the impact load, the elastic energy in soft rock, hard rock and interlayers increases continuously, and the elastic energy stored in the interlayers of the composite rock mass first reaches its bearing capacity limit, and the interlayers are rapidly destroyed. After the failure of the interlayer, the stored energy is released instantaneously and transferred to the soft rock and hard rock. Cracks begin to appear in the soft rock. Under the combined action of the energy released by the interlayer and the energy stored in the rock itself, the soft rock will have cracks and produce a little damage. The failure of interbedded rock mass starts at the interbedded cementation surface. With the increase of the interbedded thickness, only one side of the sandstone near the cementation surface is partially fractured. The growth rate of rock mass stress is inversely proportional to the interlayer thickness. The rate of damage degree increasing with strain of rock mass with low interlayer thickness (0 mm, 5 mm) is relatively constant compared with that of rock mass with other interlayer thickness. The damage degree of rock mass with high interlayer thickness (10 mm, 15 mm) increases rapidly with the strain in the late stage of crack growth.Keywords: Interlayer thickness rock mass, Driving mechanism, Split hopkinson pressure bar (SHPB), Discrete lattice spring method (DLSM)Subject terms: Energy science and technology, Engineering, Materials science, Solid Earth sciences Introduction Under the action of impact loading, the existence of interbedding in soft and hard interbedded rock masses is considered as the main factor leading to discontinuity and inhomogeneity of rock masses, which further imposes a significant influence on the properties of rock masses such as strength, deformation and damage. When stress wave propagates in soft and hard interbedded rock masses, a change in interbedding thickness would hinder the propagation of stress wave, which in turn reduces the stress wave amplitude, intensifies energy dissipation and relieves the damage on the specimen. An attempt to investigate the dynamic mechanical properties of soft and hard interbedded rock masses is expected to provide an important guideline for rock breaking and protection1. Regarding the mechanical behavior of rock masses containing interlayers, numerous scholars have conducted systematic research from different perspectives, showing a progressive deepening trend from material constitutive properties to wave propagation mechanisms, and further to macro-dynamic response, providing a solid theoretical and experimental foundation for this study. In terms of the quasi-static mechanical properties of structural plane media, researchers have primarily used direct shear tests to reveal the fundamental mechanical response of filling materials. Kim et al.2 and Vosniakos et al.3, through experiments and discrete element simulations respectively, analyzed the shear characteristics and microscopic mechanisms of filled rock joints, laying the groundwork for understanding the constitutive behavior of interlayer materials. At the level of dynamic stress wave propagation characteristics, the research focus shifted to the influence of joint surfaces on stress wave propagation. Utilizing modified SHPB devices or ultrasonic techniques, Li et al.4, Liu et al.5, and Yang et al.6 systematically investigated the attenuation patterns of P-wave and S-wave propagation concerning filling thickness and material properties. Jia et al.7 further studied stress wave propagation across filled joints from both theoretical derivation and experimental validation, proposing simplified propagation equations. This series of studies clarified the “filtering” effect of joint thickness on stress wave energy4]– [8. At the level of dynamic mechanical response and failure modes of rock masses, research extended further to macroscopic mechanical behavior. Through similar material model tests, Yang et al.9 and Liu et al.10 revealed the controlling effects of joint thickness and loading rate on dynamic strength and failure patterns. Chai et al.11 and Han et al.12 focused on the effects of cumulative damage and filled joints on the dynamic response of specimens. The SHPB tests by Li et al.13 further confirmed the significant influence of joint thickness on stress wave propagation and failure mode. Furthermore, Song Yanqi et al.14, employing three-point bending impact tests, revealed the influence of filling thickness on dynamic crack propagation paths from a fracture mechanics perspective. Wang et al.15 investigated the effects of differences in media mechanical parameters on the mechanical properties and failure forms of composite samples through uniaxial tests, while Taliercio et al.16, Theodoros et al.17, and Giannantonio et al.18 explored the failure criteria and strength conditions of layered media from a theoretical perspective, providing a theoretical framework for understanding the failure mechanisms of layered rock masses. In summary, existing research has formed a relatively complete system covering static mechanical properties, dynamic wave propagation, dynamic mechanical response, and failure theories. However, most existing studies focus on describing mechanical “phenomena,” such as stress waveforms, peak strength, and final failure morphology, yet few studies have systematically revealed the distribution, transformation, and dissipation mechanisms of the internal driving force—“energy evolution”—in rock masses with different interlayer thicknesses, and quantitatively correlated it with the entire process of dynamic crack propagation. The absence of this key mechanistic understanding limits the in-depth knowledge of the essential dynamic failure of soft-hard interbedded rock masses. The works as referred to above cover the model tests with respect to the mechanical properties of rock masses with different interbedding thicknesses under static load, the stress wave propagation properties under dynamic load, and the effects of dynamic mechanical properties of rock masses, as well as the experimental studies on dynamic response and damage of rock masses, but very few studies relate to the energy evolution and loss of soft and hard interbedded rock masses with different interbedding thicknesses. This study builds upon our earlier work published as a preprint, which focused on preliminary experimental characterization19.This study first conductsr SHPB impact tests on soft and hard interbedded rock masses with four different interlayer thicknesses. Furthermore, it investigates the influence of interlayer thickness on the energy evolution characteristics of the soft–hard interbedded rock masses and the relationship between the duration of each stage of energy evolution and interlayer thickness. On this basis, it also discusses the energy-driven mechanism governing the deformation and failure of the soft and hard composite under different interlayer thicknesses. An emerging continuous-discontinuous numerical method for rock dynamics, i.e., the Discrete Lattice Spring Method (DLSM) developed by Zhao et al.20–22, was employed to reproduce the damage evolution and dynamic fragmentation processes of the combination of hard and soft media under dynamic impact conditions with different interlayer thicknesses. Furthermore, the temporal variation relationships of interfacial stress in the interlayer of composite rock masses were analyzed, which revealed the damage-strain evolution characteristics of rock masses with different interlayer thicknesses. Impact experiment of soft and hard interbedded rock masses with different interbedding thicknesses Preparation of specimens Granite and sandstone, two common types of rocks discovered in Sichuan Zigong Quarry, were used for the experiment. Table 1 gives the physical and mechanical parameters of two types of rocks and interlaid materials. The interlaid material as shown in Fig. 1 is the cement mortar specimen obtained by: mixing cement, sand and water in the weight ratio of 1.5:4.5:0.8, pouring into 5 mm, 10 mm and 15 mm thick molds, vibrating to eliminate air pores, releasing from molds after 24 h of solidifying and hardening, curing in a standard curing chamber for a length of 28d. 3 pieces were prepared for each type of filled specimen with total quantity of 12 pieces. The processed granite, sandstone and interbedding materials were bonded together using epoxy resin according to the design of Fig. 1, but the cementation thickness was neglected. The length-diameter ratio of the combined rock mass is 2:1, which was found slightly larger than the ISRM recommended value23 for SHPB experiments. Table 1Basic physical and mechanical parameters of rock and interlayer materials.Rock typeDensity(g·cm− 3)Wave velocity(m·s− 1)Wave impedance(g·cm− 3· m·s− 1)Uniaxial compressive strength(MPa)Protodyakonovcoefficient(f)Sandstone2.483350830838.53.85Granite2.63541514,24114614.6Sandwich material1.842438448618.72.4 Basic physical and mechanical parameters of rock and interlayer materials. Rock mass design and physical picture of soft and hard media combination with different sandwich thickness. Dynamic loading test system The dynamic compression test of the combined rock mass was completed on the SHPB test system as shown in Fig. 2. As a part of the system components, the bullet is 50 mm in diameter and 400 mm in length, the incident bar is 2,000 mm in length and 50 mm in diameter, and the transmission bar is 1,000 mm in length and 50 mm in diameter. All components are made of alloy steel with a density of 7,900 kg/m3 and a modulus of elasticity of 210 GPa24. SHPB test system diagram. Data processing In this paper, the data obtained were processed using two-wave method25 based on the assumption of one-dimensional stress waves, the equation for calculation of strain rate , strain ε and stress σ of the specimen under impact loading can be thus written as: Where: C0, E, A denotes the propagation velocity of the stress wave in bar, the modulus of elasticity of bar, and the cross-sectional area of bar respectively. A0, L0 denotes the cross-sectional area and length of the specimen respectively. Assume that there is no dispersion in the propagation of the stress wave, then the signal recorded by the strain gauges affixed to the bar is the waveform at both ends of the specimen, so the stress at both ends of the specimen can be obtained from the following equation26: Where: σ1 is the stress at the contact end of the specimen and the incident bar, σ2 is the stress at the contact end of the specimen and the transmission bar. εI, εR, εT is the measured incident, reflected and transmitted waveform signals respectively. Experimental results and analysis Analysis of the energy evolution of the specimen during impact process This experiment satisfies the assumptions of one-dimensional elastic stress waves and homogeneity, but neglects the energy exchange between the specimen and the ambience. The energy carried by the stress wave under impact loading is taken as the total absorbed energy for causing damage on the specimen. In the elastic strain phase, the elastic strain energy can be released and stored in the specimen, ending up with the continuous consumption of the total absorbed energy as a result of crack initiation. The total absorbed energy density, the releasable elastic strain energy density and the dissipated energy density of the specimen can be thus written as20. Wherein, U、Ue, Ud is the total output strain energy density, the releasable elastic strain energy density and the strain dissipation energy density per unit volume of the specimen respectively. σ is the dynamic stress of the specimen, ε is the total strain of the specimen and εe is the dynamic strain of the specimen. Figure 3 shows the waveform signals (incident wave, reflected wave, transmitted wave) measured for three types of composite rock masses during impact testing. Analysis of these three waveform signals reveals the propagation characteristics of stress waves within composite rock masses. As seen in Fig. 3, the incident wave amplitudes are essentially identical, while the reflected and transmitted wave amplitudes exhibit differences. Figure 4 illustrates the dynamic stress-strain relationship of four composite rock masses. Figure 5(a) is the energy-time evolution curve of the specimen in case of the interbedding thickness of 0 mm. In the first stage, the pores and defects in rock mass are extruded densely under the impact loading, the strain continues to rise but the stress doesn’t incur obvious changes, so the total strain energy curve does not increase. At 200µs, the curve proceeds to the second stage, that is the elastic deformation stage. At this point, the rock mass is densely compressed, the stress transmission is gradually uniform, there is almost no plastic deformation, and the total absorbed energy in the initial stage is stored in the specimen as elastic strain energy. At 326µs, the curve proceeds to the third stage, that is the specimen failure stage. the elastic strain energy is rapidly released and continues to decrease with the generation of sandstone cracks in rock mass. Finally at 482.8µs, the total input energy density and the dissipated energy density reach the peak value of 0.64MJ/m3, signifying the completion closure of the failure process. Figure 5(b) is the energy curve of the specimen in case of the interbedding thickness of 5 mm. In the first stage, the pores and defects in rock mass are extruded densely under the impact loading, the strain continues to rise but the stress doesn’t incur obvious changes, so the total strain energy curve does not increase. At 150µs, the curve proceeds to the second stage, that is the elastic deformation stage. At this point, the rock mass is densely compressed, the stress transmission is gradually uniform, there is almost no plastic deformation, and the total absorbed energy in the initial stage is stored in the specimen as elastic strain energy. However, as the action of impact loading persists, the interbedding medium is more fragile than the rock, which renders the specimen prone to premature failure, so the dissipation energy begins to gradually increase, and a part of the energy continues to be stored as elastic strain energy, while the other part is dissipated along with the local stress failure. At around 300µs, the curve proceeds to the third stage, that is the specimen failure stage. At this point, the plastic deformation gradually occurs, the crack initiation of interbedding and the cracks in sandstone causes the elastic strain energy stored in the specimen to gradually consume and the density continues to decrease. Finally at 482.8µs, the total input energy density and dissipation energy density reach the peak value of 0.7 MJ/m3, signifying the complete closure of the entire failure process. As shown in Fig. 6(a) and (b), an increase of the interbedding thickness to 10 mm, 15 mm is accompanied by an increase of its peak dissipation energy density by 9.81% and 14.04% respectively, and a shortening of the duration of specimen failure stage by 31.09% and 38.67% respectively. The dissipation energy of the specimen is correlated with the extent of the specimen failure. As the thickness of the interlayer increases, the interbedding tends to be more robust in impeding the crack propagation and development of the entire specimen, and the rock mass with thicker interbedding consumes more energy if it generates the same cracks and extent of failure as that with thinner interbedding. Three-wave signal diagram of four combined rock masses. Equivalent stress-strain curves of four composite rock masses. Energy-time evolution curves of specimens with different sandwich thicknesses(0 mm and 5 mm). Energy-time evolution curves of specimens with different sandwich thicknesses(10 mm and 15 mm). Correlation between the duration of each stage of energy evolution and the interbedding thickness Referring to the duration of each stage in Figs. 5 and 6, the duration of each energy evolution stage for soft and hard interbedded rock masses with different interbed thicknesses is provided. The duration curves for each energy evolution stage of rock masses with varying interbed thicknesses are plotted in Fig. 7. Duration of energy evolution of rock mass with different interbedded thickness. As is seen from Fig. 7, the duration of the pore compacting stage in case of the interbedding thickness of 5 mm is 200µs. As the interbedding thickness increases to 10 mm and 15 mm, the duration reduces to 176µs and 164µs respectively, 12% and 18% shorter than that of the interbedding thickness of 5 mm. The duration of the pore compacting stage is generally inclined to reduce with the increase of interbedding thickness, in which the duration of rock mass curve in the case of the interbedding thickness of 15 mm is significantly shorter than that of any of the other two specimens. The interbedding material is softer than rock and its inner pores are easier to compact than granite and sandstone, so it would be quicker for the interbedding material to reach the compaction limit, while the internal structure of other combination rock mass with smaller interbedding thickness is more compact. As the interbedding thickness increases, the total volume of rock is smaller, i.e., the rock has fewer pores, so it would be quicker for the specimens to be tightly compacted. Figure 7 is the duration curve of the elastic deformation stage, and the duration of the soft and hard interbedded rock mass in this stage gradually decreases as the interbedding thickness increases. The duration of the elastic deformation phase of the specimen with the interbedding thickness of 5 mm is 123µs, and duration decreases to 114 µs and 95 µs as the interbedding thickness increases to 10 mm and 15 mm respectively, which is 7% and 22% shorter than that of the specimen with the interbedding thickness of 5 mm. The elastic deformation stage is the energy accumulation after the specimen is subject to impact loading, i.e., the energy accumulation comes at the limit of the dynamic compression strength of the specimen. A softer interbedding is more likely to fail earlier under impact loading, then the overall dynamic compressive strength of the specimen is reduced, so the duration of the elastic deformation stage becomes shorter. The specimen with the interbedding thickness of 5 mm descends in a smaller degree, the specimen with a less interbedding thickness tends to have a more compact internal structure and comes at the compaction limit more slowly. The specimen with the interbedding thickness of 5 mm tends to come with a longer compaction stage and better integrity, and its capacity to endure deformation is higher, so the time to reach the elastic deformation limit is prolonged. Figure 5 shows the duration of the failure stage of specimens with different interbedding thicknesses, that is the time from crack initiation to completion fracture. The duration curve of soft and hard interbedded rock masses in this stage tends to decrease as the interbedding thickness increases, in which the duration of the specimen with the interbedding thickness of 0 mm is the longest, and that of the specimen of 5 mm, 10 mm and 15 mm reduces by 11.54%, 39.01% and 44.51% respectively. As the interbedding thickness increases, the effects of the interbedding on crack propagation and hinderance are greater, most of the energy is dissipated in the interbedding, so the energy allocated for damaging sandstone is lower. The energy driven mechanism accounting for failure in rock masses with different interbedding thickness The failure of rock mass driven by energy can be attributed to two factors: first, the elastic energy stored in rock mass comes at the bearing capacity limit; second, the elastic energy stored in rock mass comes at the bearing capacity limit under the action of external force. In this regard, it is prioritized to figure out the energy storage characteristics of soft and hard interbedded rock masses. Figure 8 is the stress-dependent elastic energy density of soft and hard interbedded rock masses with interbedding thickness of 0 mm, 5 mm, 10 mm, and 15 mm respectively. As the stress increases, the elastic energy density of any of the four interbedded rock masses gradually increases, but the rate at which the elastic energy density increases can be ranked in an ascending order: 0 mm<5 mm<10 mm<15 mm, revealing that the rate at which the elastic energy density increases is positively correlated with the interbedding thickness of rock mass. The elastic energy density of the soft and hard interbedded rock mass with interbedding thickness of 5 mm changes in a similar pattern to that of 0 mm. Figure 9 shows the variation of the peak elastic energy density of the four interbedded rock masses. The peak elastic energy density of each of the rock masses with interbedding thickness is lower than that of no interbedding thickness, which further implies that the elastic energy storage capacity of soft and hard interbedded rock masses weakens in the presence of interbedding, and its peak elastic energy density is lower than that no interbedding thickness. The rate at which the elastic energy stores, can be ranked in an ascending order: 0 mm < 5 mm < 10 mm < 15 mm. The capacity at which the elastic energy stores can be ranked in a descending order: 0 mm > 5 mm > 10 mm > 15 mm. In this regard, the energy driven mechanism accounting for failure in soft and hard interbedded rock masses can be summarized: the elastic energy inside soft rock, hard rock and interlayer continues to increase under the action of impact loading, the elastic energy stored in interbedding of weaker strength comes at the energy storage limit and rapidly fails earliest. After the failure of interbedding, the stored energy is instantaneously released and transferred to soft and hard rocks, and cracks start to appear in soft rocks. Under the combined action of the energy released from the interbedding and the energy stored in rock mass, soft rocks will be subject to cracking and slight failure. Elastic energy density of four rock masses varies with stress. Variation characteristics of peak elastic energy density of four intercalated rock masses. The energy evolution curve clearly depicts the accumulation, conversion, and dissipation of energy within the composite rock mass under impact loading. This energy process ultimately manifests through macroscopic fracture and damage in the material. Although high-speed cameras were not employed to record the surface fracture process of specimens in this experiment, the numerical simulations based on the DLSM in Sect. 4.2 fully reproduced the dynamic crack propagation and final failure modes within specimens with different interlayer thicknesses. The simulated results—including crack initiation timing, propagation velocity, and final failure modes—exhibit high temporal and mechanistic consistency with the energy evolution stages analyzed in this section. Specifically, the elastic energy storage phase corresponds to the pre-initiation stage, while the energy dissipation phase aligns with the crack propagation period. This provides robust visual and quantitative support for the energy dissipation theory. DLSM numerical simulation analysis of stress wave transfer and stress, damage changes in rock masses with different interbedding thicknesses DLSM, a novel continuous-discontinuous numerical method for calculating rock dynamics developed by ZHAO et al.20, is used to evaluate shear deformation by the least squares method and found with a higher calculation efficiency. More details of the DLSM model and its applications are provided in relevant references21–27. Modeling of rock masses with different interbedding thicknesses Figure 10 shows the overall dimensions of the interbedded rock mass model are identical to thatof the specimen, with a diameter of 50 mm and a height of 100 mm, and the diameter of the spherical particles is 0.25 mm, This size was determined through mesh sensitivity analysis to ensure the model strikes a balance between computational efficiency and accuracy. Smaller particle diameters facilitate more precise simulation of crack initiation and propagation processes. The spring failure parameter u* t under tensile conditions and the spring failure parameter u* s under shear conditions are 0.005, These parameters represent the critical failure displacements of springs under tensile and shear conditions. Their values were determined by calibrating the DLSM model to match experimentally observed dynamic fracture behavior in rock. The parameters for sandstone, granite, and interbedded materials used in the model—including density, wave velocity, wave impedance, and uniaxial compressive strength—were all derived from experimental measurements in Table 1. This ensures consistency between the numerical model and physical specimens in their macroscopic mechanical behavior. A velocity load was applied at the top of the middle section of the model to simulate the dynamic load in the course of test, defined as a velocity boundary condition, and the specific velocity load parameters are: Model of rock mass with different interbedded thickness. Where vd is the applied dynamic rate load;t0 is the time required for the rate load increased from 0 to a given value. Here, all models determine this time as 50 µs based on the duration required for stress waves to undergo multiple reflections within the specimen. As the load is initially applied, a slow increase of the load permits the specimen to readily reach a state of dynamic equilibrium. This rise time is usually taken as the time taken by the stress wave completing 5 round trips inside the specimen, and loading velocities lead to different loading rates. Analysis of dynamic crack propagation process Figure 11 shows the crack propagation process of the rock mass with interbedding thickness of 0 mm under the action of impact loading. Cracks first appear at the sandstone end. Over time, cracks at the sandstone end gradually propagate toward the cementation surface, while cracks and compressive bulging occur in the sandstone at the cementation surface. Ultimately, the sandstone separates from the granite. Shear failure occurs in the sandstone, while no cracks appear on the granite side. Under impact loading, shear fracture surfaces develop in the sandstone at the end prior to those at the cementation interface. Over time, these shear fracture surfaces in the sandstone progressively expand. Ultimately, the shear fracture surfaces penetrate the entire sandstone mass. Consequently, shear failure dominates in the sandstone, while the granite exhibits no significant damage. Figure 12 shows the crack propagation process of the rock mass with interbedding thickness of 5 mm under the action of impact loading. Stress concentration occurs in the interbedding first, followed by compression and bulging. The weak strength and more pores of the interbedding and the large extent of compressional deformation lead to a greater reflection and transmission of the stress wave in the interbedding and on the cementation face. Under the combined action, local cracks are generated and gradually separated at the contact between the interbedding and the cementation face. As the cracks in the interbedding increase, the interbedding flakes off, new cracks occur in the sandstone of the cementation face and further extend to the ends of the sandstone. In the rock mass failure stage, the energy in the interlayer is released. As a result, a plenty of interbeddings flake off and the interbeddings are failed. The cracks in the sandstone are partially penetrated by the energy released from the failure of interbeddings. Throughout the loading process, the interbedding is first failed, a part of penetrated cracks occur in softer sandstone. The direction along which the rock matrix with interbedding thickness of 5 mm is parallel to that of the maximum principal stress, and the final failure mode is the crushing failure of the interbedding and a part of penetrated cracks in sandstone. Figure 13 depicts the crack propagation process for the rock mass with the interbedding thickness of 10 mm under impact loading. Compression cracking initiates at the middle section of the interbedding first, accompanied with a few flaky fragments. Afterwards, the main crack disposed at a certain angle to the loading direction of the maximum principal stress and the “crush zone” occurs at the interbedding where granite contacts sandstone. The main crack is classified as a compression shear crack under the combined action of compressive stress and shear stress. At around 80µs, a new crack generated at the sandstone of the cementation face further propagates to the end of the sandstone. At this point, the interbedding is further separated from the rock and new flaky fragments are formed subject to the propagated cracks and the “crush zone”. At the failure stage of rock mass, flaky fragments at the interbedding increase quickly, and the cracks and the “crush zone” at the interbedding converge at an accelerated rate. The energy stored in the interbedding and rock is released rapidly and the cracks and the “crush zone” finally converge and penetrate through under the transmission and reflection of stress wave, the interbedding appears as lumpy compression shear failure, and a part of non-penetrated cracks are formed out of the cracks in the sandstone. Figure 14 depicts the crack propagation process for the rock mass with the interbedding thickness of 15 mm under impact loading. At around 96µs, crack initiates at the interbedding of the cementation face, and a split tensile crack disposed at a certain angle with the direction of the maximum principal stress and the “crush zone” is thus formed. Over time, new cracks formed at the sandstone of the cementation face further propagate to the end of the sandstone, and flaky fragments appear in the interbedding. At around 156µs, the cracks at the interbedding propagate wider and converge with the “crush zone” until an obvious local fracture is formed. At the specimen failure stage, the energy converged in the rock and the interbedding is released and under the combined action of stress wave transmission and reflection, shear cracks disposed at a certain angle with the direction of the maximum principal stress appear at the interbedding. The final failure mode of the interbedding is compression shear failure, and non-penetrated cracks appear out of a few cracks in the sandstone. Crack propagation process of rock mass with 0 mm interbedded thickness. Crack propagation process of rock mass with 5 mm interbedded thickness. Crack propagation process of rock mass with 10 mm interbedded thickness. Crack propagation process of rock mass with 15 mm interbedded thickness. Interfacial stress-time monitoring curves Figure 15 depicts the tress-time curves derived at the cementation face in rock masses with four interbedding thicknesses. The curve consists of four stages depending upon the velocity at which the stress increases: in the slow growth stage, the stress starts to act on the specimen and the interbedding is first compacted under stress. At this point, the stress has just been applied to the specimen and the effect of propagation is poor, and the stress at the interbedding develops at a low velocity. As the stress gradually propagates towards the interbedding, the interbedding is completely compacted and the curve proceeds to the rapid stress growth stage. All parts of the rock mass are gradually and completely compacted under the action of the fast-growing stress, the stress propagates at a higher velocity and the stress at the cementation face grows rapidly. As the slope of the stress growth gradually slows down, the curve proceeds to the third stage. Cracks gradually propagate from the interbedding to the sandstone, and the interbedding between two rocks destabilizes quickly and loses the bearing capacity due to crack propagation. At this point, the stress approximates the maximum value and the curve proceeds to the fourth stage. The cracks at the sandstone gradually increase and propagate towards the transmission bar, at which point the specimen as a whole rapidly destabilizes and loses the load-bearing capacity, and the stress rapidly decreases because of the destabilizing specimen. Taking the time when the rock mass is subjected to impact loading as the starting point, the stress wave passes through the interbedding and starts to amass energy in the compact interbedding, and the stress starts to grow rapidly. In the stress rapid growth stage, the curves with respect to four interbedding thicknesses differ in growth rates, in which the curve of the non-interbedded rock mass grows at a higher velocity, i.e. the highest stress growth rate. As the interbedding thickness gradually increases, the velocity at which the stress grows surprisingly reduces. At this point, the stress wave propagates repeatedly in the interbedding. However, an increase of the interbedding thickness results in a longer response distance of stress and provided with a given incident energy, the stress response will slow down as the interbedding thickness increases. The stresses with respect to the four rock masses reach their respective peaks of 64.5 MPa, 46.33 MPa, 38.63 MPa and 34.82 MPa as the stress gradually increases. The peak stress for the rock masses with interbedding thickness of 5 mm, 10 mm and 15 mm reduces by 28.17%, 40.10% and 46.02% as relative to the non-interbedded rock mass respectively. The presence of the interbedding reduces the peak stress in the rock mass, in which the peak stress for the rock mass with interbedding thickness of 10–15 mm reduces at a higher level. Subsequently as the crack propagation in the sandstone intensifies, a large number of micro cracks and damages are formed in the specimen, the bearing capacity reaches its ultimate limit and the stress begins to drop rapidly. As the interbedding thickness increases, the faster the rock mass is crushed to fragments under impact loading, the more likely the specimen destabilizes, the faster the overall strength of the specimen decreases, and the lower the ultimate bearing capacity. Stress-time curves at the bonding surface. Damage-strain evolution characteristics of rock masses with different interbedding thicknesses The deformation damage of rocks during fracture in the DLSM numerical simulation software is closely correlated with the strain dissipation energy, and the extent of such deformation damage D can be defined by the ratio of the dissipation energy to the maximum dissipation energy in the course of rock failure: In Eq. (11), Ud is the strain dissipation energy per unit volume, and Ud max is the maximum dissipation energy in the course of rock failure. Figure 16 is the damage-strain evolution curve for rock masses with different interbedding thicknesses, from which the damage extent of rock masses with different interbedding thicknesses continues to increase with strain and eventually stabilizes. The rates at which the damage extent of rock masses with four interbedding thicknesses aggravates with strain in the initial compaction stage are basically the same. As the strain increases, there are cracks appearing in rock masses which then rapidly enter the yielding stage and plastic deformation occurs. The rates at which the curves of rock masses with four interbedding thicknesses begin to obviously vary, in which the damage extent of the non-interbedded rock mass aggravates significantly and its growth rate is greater than that of the interbedded rock masses. The spacing between two rocks in the non-interbedded rock mass is shorter than that in the interbedded rock mass, and stress concentration is more likely to occur at the cementation face, which can easily form penetrated cracks in softer sandstone, thereby accelerating the damage crushing of rock mass. The presence of interbedding thickness takes a longer time for cracks occurred in the sandstone, and the granite end is also less affected by the crushing displacement of the sandstone end, resulting in a slower aggravation in the overall damage extent of the specimen. The strain-dependent rate at which the damage extent of rock masses with different interbedding thicknesses grows would vary. The strain-dependent rate at which the damage extent of the rock mass with a smaller interbedding thickness (0 mm, 5 mm) grows is more constant as relative to rock mass with any other interbedding thickness, and the strain-dependent rate at which the damage extent of the rock mass with a greater interbedding thickness (10 mm, 15 mm) grows gradually in the late period of crack propagation. For rock mass with a lower interbedding thickness, the effects of the interbedding on crack propagation and the velocity at which cracks propagate are slight. For rock mass with a higher interbedding thickness, cracks are already present in the sandstone at the late stage of crack propagation and the interbedding has been completely crushed. The unexpected energy release from the energy previously amassed in the interbedding paces up the crack propagation in the sandstone, resulting in a rapid aggravation in the damage extent. Damage degree - strain evolution curve of rock mass with different interlayer thickness. Discussion This study systematically reveals the influence of interlayer thickness on the dynamic mechanical behavior of alternating hard and soft rock layers through SHPB experiments and DLSM numerical simulations. The preceding results have presented phenomena such as energy evolution and crack propagation. This discussion aims to integrate these independent observations into a coherent physical picture, delving into the core role of interlayers in stress wave propagation and how energy sequentially drives the ultimate failure of the rock mass. Experimental data clearly demonstrate that the presence and thickness of laminations significantly alter the dynamic response of rock masses. As shown in Fig. 15, the peak stress at the cemented interface decreases sharply with increasing interlayer thickness from 64.5 MPa at 0 mm to 34.82 MPa at 15 mm, a 46% reduction, while the stress rise rate also slows considerably. This phenomenon stems from the reflection and transmission of stress waves at interfaces between media with differing wave impedances. According to Table 1, the wave impedances of granite, interlayer material, and sandstone are 14,241, 4486, and 8308 g·cm⁻³·m·s⁻¹ respectively. When stress waves propagate from high-impedance granite into the low-impedance interlayer, strong reflection occurs, confining part of the wave energy in front of the interlayer or causing it to return to the incident rod. The thicker the interlayer, the longer the path along which stress waves undergo multiple reflections and transmissions within it, leading to more complete energy dissipation primarily through plastic deformation and internal friction within the interlayer material. This effectively transforms the interlayer into a wave impedance filter, where its thickness directly determines filtering efficiency: greater thickness reduces the effective stress wave energy transmitted to the underlying sandstone and available to drive its failure. This directly explains the phenomena observed in Figs. 5 and 6: although total dissipated energy increases with laminate thickness—as the laminate itself dissipates more energy—the actual energy share causing macroscopic through-fracturing in the sandstone decreases. This aligns perfectly with the simulation results in Figs. 11, 12, 13 and 14, where the severity of sandstone failure diminishes as laminate thickness increases. Furthermore, this filtering effect triggers an “energy valve” phenomenon. As shown in Fig. 7, as the interlayer thickness increases fre is significantly reduced. This occurs because the thicker interlayer material reaches its strength and energy storage limits earlier under stress wave action, leading to failure. This premature failure of the interlayer causes its stored elastic energy to be released prematurely. This disordered early release disrupts the process of stable energy accumulation within the rock mass that ultimately drives systematic fracturing of the sandstone. Furthermore, this filtering effect triggers an “energy valve” effect. As shown in Fig. 7, as the interlayer thickness increases from 5 mm to 15 mm, the duration of the elastic deformation stage shortens from 123µs to 95µs, and the duration of the failure stage is shortened even more dramatically. This is because the thicker interlayer material reaches its strength and energy storage limit earlier under the stress wave and fails prematurely. This prematurely failed interlayer acts like an energy valve opened too early, causing the elastic energy it stored to be released prematurely. This disordered early release disrupts the process of stable energy accumulation within the rock mass ultimately leading to systematic fracture of the sandstone. Consequently, energy is wasted on the violent fragmentation of the interlayer itself, only capable of inducing localized damage in the sandstone (Figs. 13 and 14). Based on the energy evolution curves (Figs. 5 and 6) and the dynamic crack propagation process (Figs. 11, 12, 13 and 14), an energy-driven rock mass failure model can be constructed: During the initial loading phase, elastic strain energy accumulates synchronously within the granite, sandstone, and interlayer. However, due to differences in strength and wave impedance among the three components, the rates and capacities of energy storage vary. The interlayer, possessing the weakest mechanical properties, first reaches its energy storage limit. During the initial loading phase, elastic strain energy accumulates synchronously within the granite, sandstone, and interlayer. However, due to differences in strength and wave impedance among the three components, the rates and capacities of energy storage vary. The interlayer, possessing the weakest mechanical properties, first reaches its energy storage limit. The failure of the interlayer marks the transition point where the system shifts from energy storage dominance to energy release dominance. At this moment, the elastic energy stored within the interlayer is instantaneously released. This released energy combines with the stress wave energy still being input, forming a powerful energy pulse that acts upon the adjacent sandstone. The failure of the interlayer marks the transition point where the system shifts from energy storage dominance to energy release dominance. At this moment, the elastic energy stored within the interlayer is instantaneously released. This released energy combines with the stress wave energy still being input, forming a powerful energy pulse that acts upon the adjacent sandstone. Cracks invariably initiate at the interlayer or cementation surface before propagating into the sandstone. This confirms that failure is driven from the interior outward and from weaker to stronger zones. In thin interlayers, the concentrated released energy sufficiently drives crack propagation throughout the sandstone. Conversely, in thick interlayers, excessive energy dissipation within the interlayer disperses the released energy, resulting in only localized damage to the sandstone. Cracks invariably initiate at the interlayer or cementation surface before propagating into the sandstone. This confirms that failure is driven from the interior outward and from weaker to stronger zones. In thin interlayers, the concentrated released energy sufficiently drives crack propagation throughout the sandstone. Conversely, in thick interlayers, excessive energy dissipation within the interlayer disperses the released energy, resulting in only localized damage to the sandstone. In summary, under impact loading, the failure of alternating hard and soft rock layers constitutes a sequential energy process of “storage-release-redrive” regulated by the thickness of the interlayers. The core function of these interlayers lies in their regulation of spatial energy distribution through wave impedance mismatch, while their own initial instability controls the timing of energy release. Consequently, interlayer thickness determines both the severity and localized nature of the ultimate failure mode. Conclusion SHPB and DLSM numerical calculations were used to examine the impact dynamics of soft and hard interbedded rock masses with different interbedding thicknesses, with some conclusions drawn as follows: (1) The value of the dissipated energy of the specimen under impact loading depends upon the degree of difficulty in causing specimen failure. As the interbedding thickness increases, the capacity of the interbedding to impede the crack propagation and development of the entire specimen would intensify. More energy is expected to consume if the rock mass with a greater interbedding thickness generates the same cracks and extent of failure as the rock mass with a smaller interbedding thickness. The duration of each energy evolution stage is negatively correlated with the interbedding thickness. As the interbedding thickness increases, the capacity of the interbedding to impede the crack propagation and development of the entire specimen would intensify, most of the energy is dissipated in the interbedding with less energy transferred to the sandstone. (2) The energy driven mechanism accounting for failure in soft and hard interbedded rock masses: under the action of impact load, the elastic energy in soft rock, hard rock and the interbedding continues to increase, in which the elastic energy stored in soft and hard interbedded rock masses first reaches the maximum value and the interbedding rapidly fails. Once failed, the elastic energy stored in the interbedding is instantaneously released and transferred to soft and hard rocks, where some cracks initiate in soft rock. Under the combined action of the energy released from the interbedding and that stored in rock mass, cracks will initiate in soft rock accompanied with slight damage. (3) Based on the test results and DLSM numerical calculations, the failure of the sandstone with interbedding originates at the interbedding cementation face. In case of thinner interbedding (thickness ≤ 5 mm), the sandstone is subjected to failure due to occurrence of penetrated cracks and the extent of failure on the sandstone where it contacts the cementation face tends to be more severe. As the interbedding thickness increases, the sandstone is only partially fractured at the side close to the cementation face. (4) The rate at which the stress grows on the cementation face of the non-interbedded rock mass is faster. As the interbedding thickness gradually increases, an increase of the interbedding thickness results in a longer response distance of stress, and the rate at which the stress grows on the cementation face gradually slows down. The strain-dependent rate at which the damage extent of rock masses with different interbedding thicknesses grows would vary. The strain-dependent rate at which the damage extent of the rock mass with a smaller interbedding thickness (0 mm, 5 mm) grows is more constant as relative to rock mass with any other interbedding thickness. The strain-dependent rate at which the damage extent of the rock mass with a greater interbedding thickness (10 mm, 15 mm) grows gradually in the late period of crack propagation. Acknowledgements This work was financially supported by the National Key Research and Development Program of China (2023YFC3805900); the Key Research and Development Project of Metallurgical Corporation of China Ltd. (YCC2023Kt01); the Special Project of Inspection and Certification Co., Ltd, MCC (2024ZYZX04). Author contributions Conceptualization, Yuanjian Zhang.; methodology, Yanbing Wang.; software, Huajun Xue.; formal analysis, Weihong Yang, Bin Ren.; data curation, Zhehang Huang.; writing—review and editing, Hui Xiao. All authors have read and agreed to the published version of the manuscript.”. Funding This work was financially supported by the National Key Research and Development Program of China (2023YFC3805900); the Key Research and Development Project of Metallurgical Corporation of China Ltd. (YCC2023Kt01); the Special Project of Inspection and Certification Co., Ltd, MCC (2024ZYZX04). Data availability The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.