Effects of freeze-thaw cycles on the tensile properties of geogrids and shear behavior of geogrid-soil interface

The tensile strength of geogrids and the shear strength of the reinforced soil interface are critical technical indicators for designing reinforced soil structures. This study developed a temperature-controlled apparatus to examine the tensile properties of geogrids subjected to freeze-thaw cycles within the soil, and the effect of interface temperature and freeze-thaw cycles on the shear behavior of the geogrid-soil interface. Eight and four groups of direct shear and tensile tests were conducted on the geogrid-soil interface and geogrids, respectively, under varying conditions. The peak tensile strength of geogrids remained similar to that of non-freeze-thaw samples after experiencing freeze–thaw cycles within the soil. tensile properties of geogrids were independent of the number of freeze–thaw cycles. At constant water content (w=6%), the peak shear stress at the geogrid-soil interface decreased by 24% (7 freeze-thaw cycles) compared to the non-freeze-thaw group. The cohesion and friction angle of the geogrid-soil interface decreased with increasing freeze-thaw cycles but tended to stabilize after 4 freeze-thaw cycles. The reinforcement provided by the geogrid enhanced with decrease in the interface temperature. The shear stress was higher at interface temperature under 0 °C, whereas in a non-frozen state, the interface exhibited lower shear stress with a consistent stable value.


Introduction
In recent years, reinforced soil structures have been extensively used in engineering projects, including transportation, municipal, and water conservancy, owing to their excellent seismic performance, rapid construction, and eco-friendly attributes [1,2].However, challenges were encountered when these reinforced soil structures were constructed in cold regions.The freezing and thawing of the soil in such areas could result in frost heaving and settlement, potentially damaging the reinforced soil structures and affecting their conventional functionality.Koerner's comprehensive statistical analysis, encompassing 171 global instances of reinforced soil retaining walls, revealed that approximately 60% of these walls experienced significant lateral displacement, cracks, or even collapse within the panels owing to freeze-thaw [3].This alarming statistic highlights the urgent need for the consideration of the impact of temperature fluctuations and freeze-thaw cycles when designing and constructing reinforced soil structures in cold regions.
The tensile strength of the reinforcement material is a crucial technical parameter in the design of reinforced soil structures.Typically, the tensile strength of reinforcement material is determined by unconfined tensile testing conducted in the air.However, recent research has shed light on the significance of considering confined conditions in assessing the tensile strength of reinforcement material [4,5].The mechanical properties of reinforcement material are significantly affected by confined conditions.Therefore, during the design of reinforced soil structures, the impact of confined conditions on the mechanical properties of the reinforcement material must be considered.Although existing studies have indicated that the mechanical properties of geogrids remain unaltered after experiencing freeze-thaw cycles in both air and water [6,7], the environment wherein these cycles occur does not align with the actual conditions.The mechanical properties of geogrids within reinforced soil structures after undergoing freeze-thaw cycles within soil remain unknown.
The shear behavior of the geogrid-soil interface is attributed to the interaction between soil particle confinement by the geogrid and the friction at the geogrid-soil interface, which affects the safety and stability of the reinforced soil structure.The behavior of the geogrid-soil interface varies with different reinforced soil structures and stress conditions, resulting in variations in research content and testing methods.Wang et al. [8] investigated the geogrid-grit interface's monotonic and cyclic shear behavior through direct shear tests.Shi and Xu [9] studied the impact of normal cyclic loading patterns on the shear property of the geogrid-quartz sand interface through direct shear tests.The test results showed that the normal cyclic load frequency and amplitude dramatically influenced the interface shear stress.In case of the interface characteristics between distinct backfill materials such as gravel or clay and geogrids, Zuo et al. [10] performed pull-out and direct shear tests of geogrids in sand-gravel, cohesive soils, and ballast.They reported a relatively low shear strength for the interface between the geogrid and clay, whereas the interface with sand-gravel and ballast exhibited high shear strength.Kim et al. [11] conducted direct shear tests on three types of coarse-grained soil and geogrids.Their findings indicated that larger particles yielded greater shear strength.
Related studies have indicated the feasibility of employing geogrids as reinforcement materials in permafrost regions [12,13].Zhao [14] studied the protective efficacy of flexible reinforced soil structures by conducting field monitoring and analyzing reinforced soil embankments in frozen soil regions.Chen et al. [15] explored the pullout behavior of geogrids in seasonal frozen clay.They revealed that with increase in the number of freeze-thaw cycles, the cohesion and friction coefficient of the reinforcement-soil interface exhibited minor variations.
Although coarse-grain is a prevalent choice for backfill material in most reinforced soil structures, research on addressing the effects of interface temperature and freeze-thaw cycles on the shear behavior of the interface between geogrids and coarse-grained soil remains limited.Moreover, the tensile properties of geogrids after undergoing freeze-thaw cycles within the soil are yet to b experimentally investigated.Therefore, this study focused on investigating the tensile properties of geogrids subjected to freeze-thaw cycles within the soil and exploring the effects of interface temperature and freeze-thaw cycles on the shear characteristics of the geogrid-soil interface based on the tensile and direct shear tests.

Testing Apparatus
A self-designed large direct shear apparatus with temperature-controlled capabilities was used for the geogrid-soil interface direct shear tests.Figure 1 presents a schematic of the test apparatus, primarily comprising a large-scale multifunctional interface shear equipment and a refrigeration-heating system.The shear equipment comprised various functional modules, including a rigid frame, vertical loading system, horizontal loading system, shear box, control and data acquisition system.Notably, the dimensions of the shear box were 600 mm × 400 mm × 200 mm (length × width × height), with the maximum horizontal and normal forces reaching 200 kN.The refrigeration-heating system comprised freezing pipes, inlet and outlet pipes, insulation cotton, a refrigeration-heating circulator, temperature sensors, and a data acquisition instrument.The main technical parameters of the refrigeration-heating system are presented in table 1.
The tensile testing of the geogrid was conducted using the SANS CMT7504 microcomputercontrolled electronic universal tensile testing machine.It mainly comprised a control system, fixtures, tension sensor, displacement sensor, strain testing system, and data acquisition system.The maximum test tensile force was 50 kN, and the effective width of the fixture was 200 mm.
Table 1.Main technical parameters of refrigeration-heating system.To align with practical engineering requirements, sandy soil with a maximum particle size of 5 mm and a water content of 6% was utilized as the backfill material.The particle gradation curve is illustrated in figure 2. The maximum and minimum dry density of the soil were 1.90 and 1.49 g/cm³ , respectively, with the uniformity coefficient (Cu) of 3.91 and the curvature coefficient (Cc) of 0.91.

Geogrid.
To align with practical engineering requirements, the geogrid was processed before the test with certain transverse ribs were removed to simulate a uniaxial geogrid.Figure 3 presents the processed geogrid sample, which had 14 longitudinal ribs and retained 5 rows of transverse ribs.Table 2 lists the technical specifications for the processed geogrid samples.

Test schedule and procedure.
For the tensile tests, the wide strip tensile method was employed.The geogrid sample included two complete units along the machine direction, with dimensions of 200 mm for both the length and width.
According to the relevant provisions of the Geosynthetics-Plastic Geogrids (GB/T 17689-2008), the tensile rate was set as 50 mm/min.The tests were performed under controlled laboratory conditions, maintaining a temperature of 20 °C and a relative humidity of 64%.
The reinforced soil interface direct shear tests in this study were executed using the strain-controlled method, encompassing a total of eight test groups.the tension of different embedding depths of the geogrid in practical engineering.The shear rate was set at 1 mm/min, and the test was stopped when the stress attained a stable value.After drying the soil, we prepared soil samples with a water content of 6% and filled the soil into the shear box.When the height of the soil reached the middle of the lower shear box, freezing pipes were arranged inside the lower shear box, and the filling process persisted until the top of the lower shear box was realized.Subsequently, the geogrid was positioned between the upper and lower shear boxes, followed by the placement of the upper shear box.This procedure was repeated until the upper shear box was entirely filled with soil.The compaction coefficient of 0.9 was consistently applied in all tests, and the density control method was applied to guarantee that the soil inside the box attained the required compactness.The arrangement of freezing pipes within the shear box is shown in figure 4.
Subsequently, the refrigeration-heating system was activated to initiate the freeze-thaw cycles, the freezing and thawing temperatures were set to -20 and +30 °C , respectively.Figure 5 provides a graphical representation of the temperature variations recorded by T1, T2, and T3 sensors during a single freeze-thaw cycle.When the temperature at sensor T2 reached -10 °C, the freezing process was suspended, thus allowing it to achieve thermal equilibrium for 1 h.Thereafter, the thawing process began, and once the temperature at T2 reached 20 °C, thawing was halted, concluding one freeze-thaw cycle.This process was repeated for the subsequent cycles.Following the completion of the freeze-thaw cycles, the geogrids were extracted for the subsequent tensile tests, or the reinforced soil samples were sent to the shear equipment for the interface direct shear tests.

Tensile test
To reflect objectivity, three geogrid samples were chosen for parallel testing within each group.The resulting average values are listed in table 4. Utilizing the test data from table 4, the relationship curve between the geogrid tensile strength and strain at different freeze-thaw cycles is presented in figure 6.
The strain is defined as the displacement of the clamped end divided by the test length of the reinforcement (L = 250 mm), and is expressed as a percentage.From figure 6, it was evident that the peak tensile strength of geogrids after undergoing freeze-thaw cycles within the soil was close to that of the non-freeze-thaw sample in the same testing environment.Furthermore, increasing the number of freeze-thaw cycles did not result in substantial alterations in the tensile strength of the geogrid.This observation suggested that freeze-thaw cycles exerted negligible influence on the tensile properties of the geogrid.Thus, their impact could be neglected in practical engineering applications.However, for the reinforced soil structures designed and constructed in seasonal frozen soil regions, the geogrid might creep owing to temperature changes during multiple freeze-thaw cycles.Freeze-thaw cycle number Table 4 and figure 6 provide the basis for understanding the relationship between strain at peak tensile strength and the number of freeze-thaw cycles, as shown in figure 7.As depicted in the figure, the strain at the geogrid's peak tensile strength was approximately 7%; however, it changed slightly after experiencing freeze-thaw cycles, and there was no apparent correlation between the geogrid's yield strain and the number of freeze-thaw cycles.This observed behavior can be attributed to the geogrid's limited susceptibility to alterations in its tensile properties caused by the relatively short duration of the simulated freeze-thaw cycles in the test-only 168 h (a value significantly different from the actual duration necessary for geogrid creep to manifest).Consequently, the geogrid exhibited no creeping during the test, and its strain remained unchanged regardless of the freeze-thaw cycle number.

Interface direct shear test
3.2.1.Effect of freeze-thaw cycles on the shear behavior of geogrid-soil interface.Figure 8 illustrates the relationship between shear stress and displacement of the geogrid-soil interface, considering various freeze-thaw cycle counts and diverse normal stresses.Based on figure 8, the variations of the geogridsoil interface shear strength parameters (cohesion, ca, and friction angle, φ) under different freeze-thaw cycle numbers are summarized and presented in figure 9.
Figures 8 and 9 reveal several key findings.First, the peak shear stress at the geogrid-soil interface increased proportionally with the increase in normal stress under the same number of freeze-thaw cycles.
Second, the impact of the freeze-thaw cycles on the peak shear stress of the geogrid-soil interface was considerable.As the tally of freeze-thaw cycles increased, the peak shear stress of the interface experienced a progressive reduction.Compared to the non-freeze-thaw group, the peak shear stress decreased by approximately 24% after 7 freeze-thaw cycles, indicating that freeze-thaw cycles reduced the reinforcement of the geogrid in the sandy soil.However, after experiencing 4 freeze-thaw cycles, the reduction in shear stress becomes relatively minor, indicating that further increasing the freeze-thaw cycle number no longer significantly affected the shear stress of the geogrid-soil interface.Third, the freeze-thaw cycles also influenced the cohesion and friction angle of the interface.As the freeze-thaw cycle count increased, both the cohesion and friction angle exhibited a declining pattern.After 4 and 7 cycles, the difference in cohesion at the interface becomes relatively small.
The freeze-thaw cycles disrupted the structure of the initial soil particles.The voids between coarse particles increased or gradually occupied by finer particles.The embedding effect between the coarse particles and the geogrid was gradually replaced by finer particles or voids, resulting in a reduction in the interface shear strength between the soil and the geogrid.With an increase in the freeze-thaw cycle number, the internal structure of the soil continuously changed, resulting in cumulative effects.The ongoing destruction and recombination of soil particles caused the cohesive force and friction angle at the interface to gradually decrease and finally stabilize.

Effect of interface temperature on the shear behavior of geogrid-soil interface.
In this test, the interface temperature between soil and geogrid was controlled at four specific values: -10, 0, 10, and 20 °C .Subsequently, direct shear tests were conducted under the normal stress of 50 kPa.Figure 11 shows the relationship between the shear stress and displacement at different interface temperatures.
Figure 10 reveals that the shear stress was generally augmented with decrease in the interface temperature.The peak shear stress occurred at an interface temperature of -10 °C , progressively decreasing to its nadir at an interface temperature of +10 °C or higher.Furthermore, once the interface temperature reaches or exceeded 10 °C , the stable shear stress of the geogrid-soil interface remained relatively constant.Thus, the effect of enhanced reinforcement for the geogrid in the soil was substantial within frozen soil regions owing to low-temperature freezing.Additionally, when the interface temperature surpassed a particular value (in this study, 10 °C), additional temperature increases ceased to influence any alterations in the shear strength of the geogrid-soil interface.
When the interface temperature dropped below 0 °C, the water within the soil underwent a phase change, transforming into ice crystals and interlayers.This transformation caused the soil and geogrid to combine into a composite with higher stiffness.During the interface direct shear test, a larger horizontal shear stress was required to shear the geogrid-soil interface.Subsequently, the connections formed by the freezing were broken, thereby reducing the horizontal shear stress and facilitating the appearance of residual strength, which subsequently stabilized.As the interface temperature increased, water ceased to be frozen, and the need for high horizontal shear stress diminished.Therefore, the reinforced soil structure located in permafrost regions would not be subject to adverse effects owing to the freezing of the reinforced soil.However, freezing could actually enhance the reinforcement provided by the geogrid, leading to improved performance at lower temperatures.

Conclusion
In this study, tensile and direct shear tests were conducted on the geogrid experiencing freeze-thaw cycles within soil and reinforced soil interface, respectively.The tensile properties of the geogrid subjected to freeze-thaw cycles within the soil, as well as the effects of interface temperature and freeze-thaw cycles on the shear behavior of the geogrid-soil interface, were studied.Based on the systematical test observations, the specific conclusions of this study are as follows.
(1) Compared to the non-freeze-thaw group, the peak tensile strength of geogrids remained nearly unchanged after experiencing freeze-thaw cycles within sandy soil.Additionally, the increase in number of freeze-thaw cycles had minimal impact on the tensile properties of geogrids.
(2) Freeze-thaw cycles reduced the reinforcement provided by the geogrid within the sandy soil.Compared to the non-frozen-thaw group, the peak shear stress at the geogrid-soil interface decreased by approximately 24% following 7 freeze-thaw cycles.Both the cohesion and friction angle of the geogrid-soil interface exhibited a decline with an increasing number of freeze-thaw cycles but tended to stabilize after 4 freeze-thaw cycles.
(3) The shear stress of the geogrid-soil interface increased as the interface temperature decreased.The shear stress was higher when the interface temperature dropped below 0 °C, whereas in a non-frozen state, the interface exhibited lower shear stress with a consistent stable value.Thus, freezing could enhance the reinforcement provided by the geogrid.

References
[1] Xu C, Luo M, Shen P, Han J and Ren F 2020 Seismic performance of a whole geosynthetic reinforced soil-integrated bridge system (GRS-IBS) in shaking

Figure 1 .
Figure 1.Main technical indices of refrigeration and heating system 2.2.Test Materials 2.2.1.Soil.To align with practical engineering requirements, sandy soil with a maximum particle size of 5 mm and a water content of 6% was utilized as the backfill material.The particle gradation curve is illustrated in figure2.The maximum and minimum dry density of the soil were 1.90 and 1.49 g/cm³ , respectively, with the uniformity coefficient (Cu) of 3.91 and the curvature coefficient (Cc) of 0.91.

Figure 2 .a
Figure 2. Particle gradation curve Figure 3.The geogrid sample after rib removal

Figure 4 . 5 .
Figure 4. Sketch of the freezing pipes arrangement in the shear box Figure 5. Plot for temperature changes inside the shear box during one freeze-thaw cycle

6 Figure 6 .Figure 7 .
Figure 6.Curve of geogrid tensile strengthstrain at different freeze-thaw cycle number Figure 7. Relationship between strain at peak tensile strength and freeze-thaw cycle number

Figure 8 .
Figure 8. Curve of shear stress-shear displacement of the geogrid-soil interface at different freezethaw cycle numbers

Figure 9 .Figure 10 .
Figure 9. Interface cohesion and friction angle versus freeze-thaw cycle number Figure 10.Curve of shear stress-displacement at different interface temperatures

Table 3 .
Arrangement of the tensile test and direct shear test