Experimental Study on Cyclic Shear Performance of the Four-Way Geogrid Reinforcement–Soil Interface

Abstract

This paper presents the results of horizontal cyclic direct shear tests at the reinforced soil interface of a four-way polypropylene geogrid reinforced sandy soil. The influence of normal stress and shear displacement amplitude on the shear stress, shear stiffness, and damping ratio of the reinforced soil interface are evaluated by varying the normal stress and shear displacement amplitude. Dynamic shear characteristics of reinforced soil interface under normal constant load were investigated by using a large dynamic straight shear apparatus. Peak interface strength increases with increasing amplitude of normal stress and shear displacement amplitude. The larger the normal stress and shear displacement amplitude, the fewer cycles are needed to attain peak interface strength. At low-magnitude normal stress levels, the peak shear stress and shear stiffness tend to stabilize after an initial increase during the cycling process, and the damping ratio decreases and then stabilizes with the increase in the number of cycles; whereas when the normal stress level is high, the peak shear stress and shear stiffness increase and then decrease during the cycling process and eventually stabilize, and the damping ratio decreases and then increases and finally stabilizes with the increase in the number of cycles. Moreover, under the same number of cycles, the corresponding shear stiffness decreases with an increase in shear displacement amplitude, while the damping ratio increases.

1. Introduction

A geogrid is a type of geosynthetic material with high operational efficiency that is extensively implemented in reinforced soil structures [1,2,3,4,5,6]. The interaction mechanism between grids and soil is a crucial topic in geogrid research. Several studies have been conducted by scholars at experimental, numerical simulation, and theoretical analysis levels. These studies have found that the mechanical properties of the grids, soil particle gradation, and particle size are the key factors that affect the geogrid–soil interface. However, most of the studies have been based on static loading conditions [7,8,9,10,11].

During reinforced soil projects, the structure is exposed to dynamic factors such as traffic loads, wave loads, and mechanical loads resulting in cyclic shear at the geogrid–soil interface. Consequently, numerous scholars have begun emphasizing the shear characteristics of the interface under dynamic effects due to cyclic loading. Zhang Ga and Zhang Jian min [12] employed a cyclic loading shear instrument to investigate the fundamental principles and force deformation mechanism of interfacial mechanical properties between coarse-grained soil and structural contact surfaces subjected to repeated loading. The results revealed that two physical change mechanisms, namely, granular crushing and compression of soil near the structure surface, occurred during the shear process. Liu et al. [13] found the enhancement of peak shear stresses by cyclic shear through horizontal cyclic direct shear tests and pointed out that the shear displacement amplitude affects the shear shrinkage. Wang et al. [14] investigated the effect of the particle size of the soil on the cycle shear behavior of the interface and found that the interfacial shrinkage increases as the particle size increases. Ying et al. [15] performed cyclic direct shear tests on geogrid–gravel interfaces, simulating the effects of seismic loading on a reinforced soil structure. They found that the shear stiffness of the geogrid–gravel interface increased with the increase in normal load.

The majority of research on the cyclic shear properties of the geogrid–soil interface focuses on uniaxial or biaxial geogrids. Shi et al. [16] studied the interfacial properties of uniaxial and biaxial plastic geogrids by means of pullout and straight shear tests, and found that for unidirectional grids with reinforcement, the pullout and straight shear curves are usually strain-softened, while for bi-directional grids with reinforcement, the curves are generally strain-hardened. Mahigir et al. [17] investigated the effect of cyclic loading on the displacement of uniaxial polyester geogrids by carrying out multistage pulling tests with a large pulling device. Liang et al. [18] proposed a trilinear shear stress-displacement damage softening model for a unidirectional geogrid–soil interface based on the results of pullout tests. Gao et al. [19] fabricated uniaxial, d biaxial, and triaxial polylactic acid geogrids based on 3D-printing technology, and investigated the mechanical properties of these grids through an indoor pullout test. There is little research literature on four-way geogrids. Si et al. [20] designed a new four-way geogrid and demonstrated the feasibility of the design through mechanical test and numerical simulation. In summary, experimental and theoretical research on the cyclic shear performance of the four-way geogrid–soil interface lags behind its engineering application, and the advantages and disadvantages of four-way geogrid reinforcement remain unclear. This paper investigates the cyclic shear characteristics of four-way geogrid–soil interfaces under different normal stress and shear amplitudes using an indoor large-scale dynamic straight shear test system. The purpose is to uncover the reinforcing mechanism of the interface, improve the selection of reinforced soil engineering, and promote the significant application of new materials.

2. Materials and Methods

The test utilized Fujian standard sand as the filler material, with a particle size range of 0.1–4 mm determined through a sieving test. The sand has an effective particle size of d₁₀ = 0.23 mm, a continuous particle size of d₃₀ = 0.44 mm, and a limited particle size of d₆₀ = 0.97 mm. The sand exhibited an uneven coefficient of Cc = 4.43 and a curvature coefficient of Cu = 0.83. The specific gravity of the soil particles was found to be Gs = 2.71. Figure 1 shows the grading curve of the particles. The maximum dry density was determined to be ρ_dmax = 1.84 g/cm³, while the minimum dry density was ρ_dmin = 1.35 g/cm³. For testing purposes, we used the TGSXG1515PP four-way geogrid model manufactured by Shandong Tai’an Luder Company (Tai’an, China). Compared to the standard two-way grating, this geogrid has increased rhombic diagonal ribs that form two new types of nodes (node Ⅰ and node Ⅱ), as Figure 2 illustrates.

Table 1. Construct parameters of four-way geogrid.

Name of Part	Length (mm)	Width (mm)	Maximum Elongation (%)	Ultimate Tensile Strength (kN/m)
Horizontal rib	60	3	6.3	17.5
Normal rib	60	3	5.7	15.5
Diagonal rib	20	2	6.2	18.2
Node Ⅰ	20	15	-	-
Node Ⅱ	20	8	-	-
Node O	3	3	-	-

presents the grating’s structural parameters while

Table 2. The technical parameters of the dynamic straight shear system.

Name of Project	Value
Maximum output test force in vertical direction	60 kN
Maximum output test force in transverse direction	35 kN
Accuracy of test force	≤±1%
Maximum stroke of horizontal actuator	150 mm
Maximum operating frequency	5 Hz
Internal dimensions of the upper shear box	405 mm × 305 mm × 150 mm
Internal dimensions of the lower shear box	305 mm × 305 mm × 150 mm

shows its mechanical parameters.

The large-scale dynamic straight shear instrument used in this paper, as Figure 3 shows, includes an mainframe instrument system, an oil source system, and a data acquisition system. The shear test system consists of a servo, sensor, and auxiliary equipment. The oil source system mainly consists of an oil tank, motor, oil pump, hydraulic valve, and electric control cabinet to provide power for the servo actuator. The data acquisition system consists of a plug-in controller, software, computer, and printer. The straight shearer controls the vertical and horizontal actuators through a computerized closed loop. The vertical actuator is mounted on a crossbeam which is supported and connected by two supports. The lower end of the piston rod of the vertical actuator is connected to a pressure beam, and the upper-side limit frame can be connected under the pressure beam to complete the vertical loading of the specimen. The vertical and horizontal movements of the shear box are recorded by the data acquisition system. The upper shear box is connected to a vertical hydraulic servo with a vertical movement of 150 mm. The lower shear box is longer than the upper one in order to maintain a constant shear area during the shearing process. Table 2 shows the technical parameters of the dynamic straight shear.

When conducting the four-way geogrid–soil interface shear test using the system, it is essential to ensure that the geogrid remains tightly secured and its relative position with the lower box remains unchanged. Since the geogrid cannot be fixed between the upper and lower boxes upon delivery of the instrument, a modification has been made to prevent any horizontal movement, deflection, or folding of the geogrid during shear testing. A self-designed four-way geogrid fixing device has been implemented for this purpose. As Figure 4 illustrates, this device consists of a fixed steel plate and four hooks for securing the geogrid. The geogrid fixing device effectively secures the geogrid on the horizontal plane of the lower box to maintain its tautness throughout cutting operations. During usage, one side of the geogrid is firmly affixed to the front of the lower shear box using a fixed steel plate, while another side is fastened to it with a grille fixing hook. During the cyclic shear test, the upper shear box is fixed, and the lower shear box starts the cyclic shear from the equilibrium position and finally returns to the equilibrium position.

Citing previous research findings [21,22], it has been determined that there is minimal change in mechanical indexes of the geogrid–soil interface after 1000 cycles of cyclic shear. Therefore, the test is performed for 1000 cycles. To ensure the evenness of the shear surface during the test, the lower shear box was filled with aluminum blocks measuring 405 mm × 305 mm × 100 mm in length, width, and height, respectively. Afterward, 1.6 kg of sand particles was added and compacted in layers until level with the lower box. The mass of sand particles in both the upper and lower boxes was weighed using an electronic scale prior to each filling, guaranteeing consistent sand compaction throughout the test. Four normal stresses (60 kPa, 90 kPa, 120 kPa, 150 kPa) were applied, representing initial static stresses on the soil at depths of 3–8 m below the filling surface. A sine wave can be used to simulate traffic load [23], so this test employs a horizontal displacement loading mode using a sine wave. The frequencies used for the cyclic straight shear test were 0.5 Hz and 1 Hz, which are common frequencies for traffic loads on road pavements.

Table 3. Cyclic shear test conditions.

Normal Stress σ/kPa	Shear Displacement Aw/mm	Shear Frequency f/Hz
60	1.5	1
90	1.5	1
120	1.5	1
150	1.5	1
120	0.5	0.5
120	1.5	0.5
120	2.5	0.5

details the specific test conditions.

3. Results and Discussion

3.1. Shear Stress

3.1.1. Effect of the Normal Stress

The four-way geogrid–soil interface was tested under a f of 1 Hz and A_w of 1.5 mm for 1000 cycles of shear. Figure 5 shows the resulting different cycles of the geogrid–soil interface shear stress and shear displacement curves. Referring to the literature [24], the average of the absolute values of the maximum and minimum shear stresses in a hysteresis loop is defined as the average peak shear stress of the hysteresis loop; the average peak shear stress of the first hysteresis loop is defined as the initial interfacial shear stress; the maximum average peak shear stress that occurs in all the hysteresis loops is defined as the peak interfacial strength, and the average peak shear stress that achieves the stabilized residual stage is defined as the interfacial residual strength. It is evident that the shape of the hysteresis loop at the interface between the geogrid and soil undergoes significant alterations with an increasing number of cycles. During the initial stages of the cyclic shear test (n ≤ 50), the hysteresis curve appears long and elliptical, but as the number of cycles increases (100 ≤ n < 500), the middle segment of the hysteresis loop protrudes outward. The specimen displays a lengthy pike-shaped structure with a broad mid-section flanked by two pointed ends. Subsequently, as n ranges from 500 to less than 1000, the bulging middle exhibits increasing fullness, and the hysteresis loop adopts a parallelogram shape.

Figure 6 displays the graphs for the average peak shear stress variation at the interface of the soil reinforced with a four-way geogrid subjected to various normal stresses and numbers of cyclic shear. The results demonstrate that both the initial shear stress and peak interfacial strength increase as normal stress is raised. Specifically, when σ increases from 60 kPa to 150 kPa, there is a 31.02 kPa increase in initial shear stress and a 41.60 kPa increase in peak interfacial strength. This aligns with the Moore–Coulomb strength criterion, indicating that material shear strength increases with increasing normal stress. When the stress is equal to 60 kPa or 90 kPa, the peak shear stress rises with the growing number of cycles, ultimately stabilizing. Furthermore, the larger the stress, the greater the number of cycles required for stabilization. Moreover, the geogrid reinforcement–soil intersection depicts shear-strengthening properties. When the stress level is 120 kPa and 150 kPa, the peak shear stress increases with the number of cycles, subsequently decreasing and eventually stabilizing. Additionally, the number of cycles needed to attain the interfacial residual strength increases with the growth in stress level. During this phase, the interfacial residual strength does not undergo significant alterations, and the geogrid–soil interface displays shear-softening characteristics. This result is very similar to the experimental results of Wang et al. [25]. The possible cause of this phenomenon is that the value of σ is too high, which exacerbates the deformation and wear of the grid during the shearing process. As a result, the interlocking effect between the soil and the grid weakens. The greater the vertical stress, the smaller the corresponding cycle times when reaching the peak shear stress can be observed simultaneously [26]. With increased vertical stress, soil particles densify at a faster rate, leading to earlier interlock between soil particles and grille [27]. Consequently, the appearance of peak shear stress occurs earlier within the entire cycle.

3.1.2. Effect of the Shear Displacement Amplitudes

Figure 7 provides the shear stress–shear displacement curves for the interface between the geogrid and soil under varying shear displacements at 50, 100, 500, and 1000 cycles. In comparison of the above curves, it is evident that the area of the hysteresis loop grows with the increase in A_w and the number of cycles. Furthermore, when A_w is large, the hysteresis curve will decline with the increase in the number of cycles, resulting in strain accumulation. This degradation is obvious in the first 100 cycles and then gradually tends to be stable. A_w has a significant influence on the development of the peak shear stress. When A_w is 0.5 mm, the peak shear stress shows a trend of increasing with the increase in the number of cycles; when A_w is 1.5 mm, the peak shear stress shows a trend of first increasing and then decreasing with the increase in the number of cycles; when A_w is 2.5 mm, the peak shear stress shows a trend of decreasing with the increase in the number of cycles. When A_w is large, the hysteresis curve will decline with the increase in the number of cycles resulting in the accumulation of strain, and this degradation is obvious in the first 100 cycles and then gradually stabilizes. The peak value of shear stress decreases with the increase in the number of cycles. The slope of the hysteresis loop decreases with the increase in the number of cycles and decreases with the increase in the displacement amplitude.

Figure 8 displays the curves depicting the alternating shear stress at the interface of the geogrid and soil. The curves show the changes in the average peak stress with the number of cycles for different amplitude values of displacement at a normal stress of 120 kPa and f = 0.5 Hz. The increase in A_w leads to an increase in both the initial shear stress and peak interface strength. When A_w increases from 0.5 mm to 2.5 mm, there is a 26.95 kPa increase in initial shear stress and a 32.65 kPa increase in peak interfacial strength. When A_w = 0.5 mm, the average peak shear stress curve increases rapidly at the beginning of the cycle and then stabilizes. However, for A_w values of 1.5 mm and 2.5 mm, the average peak shear stress rapidly reaches the peak strength of the interface at the beginning of the cycle and then decreases rapidly and finally stabilizes. In the shearing process, the sand particles are crushed by collision, and the crushed fine particles fill the pores between the coarse sand particles. The density of the sand particles near the interfacial shear zone increases, leading to an increase in the average peak shear stress. At the same time, the grating will be worn down by the constant collision with sand and soil, causing the peak shearing stress to decrease. When the A_w is small, the degree of grating wear is small and the degree of particle fracture increases with the number of cycles and then tends to stabilize, so there is an overall trend of shear hardening. When the A_w is larger, the particle crushing and the grating breakage become more intense, so that the peak value of the average shear stress shows a trend of hardening and then softening. The phenomenon that grids exhibit different shear properties at different shear displacement amplitudes is the same as the findings of Liu et al. [28]. Moreover, the phenomenon that the interfacial residual strength decreases with increasing A_w can also be explained by the above point of view.

3.2. Shear Stiffness and Damping Ratio

Shear stiffness and damping ratio are important metrics used to quantitatively study the dynamic properties of the geogrid–soil interface [29]. As Figure 9 shows, the shear stiffness K, which defines a reciprocal cycle, is calculated from Equation (1) and the damping ratio D is calculated from Equation (2):

$$K=\frac{K_1+K_2}{2}=\frac{{\tau }_1+{\tau }_2}{2Aw}$$

(1)

$$D=\frac{D_1+D_2}{2}=\frac{1}{2}(\frac{A}{4\pi \frac{1}{2}{\tau }_1{A}_w}+\frac{A}{4\pi \frac{1}{2}{\tau }_2{A}_w})=\frac{A}{4\pi A_w}(\frac{1}{{\tau }_1+{\tau }_2})$$

(2)

In the equation, τ₁ represents the maximum positive shear stress of the hysteresis loop, while τ₂ denotes the maximum negative shear stress of the hysteresis loop. Moreover, A_w represents the amplitude of the shear displacement, and A is the area enclosed by the hysteresis loop.

3.2.1. Effect of the Normal Stress

Figure 10a shows that the shear stiffness of the geogrid–soil interface changes differently under various normal stress levels. When the normal stress level is low, the shear stiffness gradually increases with the number of cycles. The smaller the σ, the more cycles are required for the shear stiffness to reach its maximum value. Under high normal stress conditions, the shear stiffness reaches its maximum value with fewer cycles as stress increases. Additionally, the rate of decline is faster with larger values of σ. This phenomenon may be due to the wear and deformation of the grating under high stress, which causes particle breakage and a decrease in the interface’s ability to resist deformation. The larger the value of σ, the greater the degree of wear and tear on the grating and particles, resulting in a faster degradation of the interface’s strength [30]. The damping ratio decreases with an increase in cycle times under low normal stress and tends towards a stable value. Conversely, under high normal stress, the damping ratio increases with an increase in cycle times and tends toward a stable value. This suggests that the geogrid is not deformed under low normal stress, resulting in relatively low interfacial energy dissipation. However, under high normal stress, the geogrid is deformed, leading to an acceleration in interfacial energy dissipation [31].

3.2.2. Effect of the Shear Displacement Amplitudes

Figure 11a illustrates the variation curves of the shear stiffness of the geogrid–soil interface at different shear displacement amplitudes under a normal stress of 120 kPa. The results indicate that when A_w = 0.5 mm, the shear stiffness exhibits shear hardening with the number of cycles, while for A_w > 0.5 mm, the shear stiffness shows the phenomenon of softening with the number of cycles. The grating is more susceptible to wear and deformation when the shear amplitude is relatively large. This wear and deformation can lead to a degradation of the interface shear strength. As the shear displacement amplitude increases, the corresponding shear stiffness decreases, which is consistent with the conclusion reached by Vieira et al. [32] at the same number of cycles. As the number of cycles increases, the shear stiffness for different shear displacement amplitudes eventually converges to a stable value. Figure 11b displays the variation curves of the damping ratio of the geogrid–soil interface for different shear displacement amplitudes. The damping ratio exhibits cyclic shear hardening as the shear displacement amplitude increases. When the shear displacement is small (A_w = 0.5 mm), the damping ratio curve does not show significant fluctuations with the number of cycles. However, as the shear displacement increases (A_w > 0.5 mm), the damping ratio curve becomes more irregular. The damping ratio increases as the shear displacement amplitude increases with the number of cycles. The larger the damping ratio, the faster the rate of energy dissipation at the geogrid–soil interface. Similarly, the larger the shear displacement amplitude, the faster the rate of energy dissipation at the geogrid–soil interface [33].

4. Conclusions

This paper investigates the effects of normal stress and shear displacement amplitude on the cyclic shear performance of the geogrid–soil interface through cyclic shear tests. The main conclusions can be drawn as follows.

(1)

The results indicate that the amplitude of shear displacement is a key factor affecting the development of hysteresis loops. When the shear displacement is less than 1.5 mm, the hysteresis loops have a long pike shape throughout the cyclic process. However, when the shear displacement is greater than or equal to 1.5 mm, the shape of the hysteresis loops at the geogrid–soil interface changes significantly with an increase in cyclic cycles. At the early stage of the test, the shape is similar to a long ellipse. As the number of cycles increases, the hysteresis loop bulges out in the middle, taking on a shuttle-like shape with two sharp ends and a wide center. As the number of cycles continues to increase, the middle bulging section becomes fuller, resulting in a hysteresis loop that resembles a parallelogram.

(2)

The curves depicting the change in shear stress and shear displacement at the interface between the geogrid and soil can be divided into three stages based on their characteristics: strain hardening, hardening and softening differentiation, and stabilization. It is observed that the number of cycles corresponding to the peak average shear stress at the interface decreases with an increase in the normal stress. Under different rules of change, the peak average shear stress, shear stiffness, and damping ratio were measured in high and low normal stress conditions.

(3)

The area of the hysteresis loop increases as the shear displacement amplitude increases. Under the same number of cycles, a larger shear displacement amplitude corresponds to a smaller shear stiffness and a larger damping ratio.

Due to the limitation of time and experimental conditions, a more in-depth study of the four-way geogrid reinforced soil interface is not carried out in this paper. Future scholars can also study the interaction between large particle sizes and four-way geogrids, the characteristics of the reinforcement–soil interface under vertical cyclic loading, and the influence of the grid node structure on the characteristics of the reinforcement–soil interface.