*3.2. Materials*

3.2.1. Soil

The soil samples used in this study were weathered granite soil samples. It is the most extensively distributed soil type in Korea [40] and is a typical nonplastic cohesive soil that contains fine-grained soil. The soil sample was classified as SW based on the USCS. The particle size distribution and engineering properties of the soil are shown in Figure 2 and Table 1, respectively. The model soil for the pullout tests was compacted with an optimum water content of 14.1%, and a maximum dry unit weight of 18.8 kN/m2 was achieved. The shear strength of the soil by direct shear tests was 8.7 kPa, and the internal friction angle was 35.6◦.

**Figure 2.** Particle size distribution of soil.


**Table 1.** Engineering properties of weathered granite soil.

#### 3.2.2. Geosynthetic Strip

As shown in Figure 3, the geosynthetic strips used in the pullout tests consisted of highstrength polyester and polyethylene. The widths of the geosynthetic strips were 50 and 70 mm, respectively, and both had a manufacturing strength of 25 kN. Geosynthetic strips with widths of 50 and 70 mm were used to investigate the pullout resistance characteristics according to the effective area for the same tensile strength condition.

**Figure 3.** Composition of geosynthetic strip.

The wide-width tensile test on the geosynthetic strips with different widths was conducted five times, and the averaged results are shown in Figure 4.

**Figure 4.** Relationship between tensile force and strain on the reinforcement with widths equal to: (**a**) width 50 mm; (**b**) width 70 mm.

For the geosynthetic strip with a width of 50 mm, the maximum tensile strength was found to be 26.8 kN, and the tensile strain was 10.5%. For the geosynthetic strip with a width of 70 mm, the maximum tensile strength and tensile strain were found to be 33.6 kN and 13.3%, respectively. The tensile strength and tensile strain increased as the width increased, and the two geosynthetic strips exhibited similar tensile deformation behaviors. As the tensile strengths obtained from the wide-width tensile test were higher than the manufacturing strength, there was no problem in applying the geosynthetic strips to pullout tests.

### *3.3. Testing Program*

Large-scale pullout tests were conducted in accordance with the ASTM D 6706-01 test method [39].

As shown in Figure 5, the model soil was divided into upper and lower parts with respect to the reinforcement and each part was compacted in three layers with the use of a rammer (impact number per min: 640–680) with an impact force of 14 kN. The compaction rate of each layer was higher than 95%. Figure 5 shows the reinforcement installed in the soil and the deformation measurement positions. The deformation measurement positions were selected to evaluate the effective length through the tensile strain and pullout force distribution in the reinforcement based on previous studies [23,28,30]. In all cases, the pullout deformation was controlled with a strain rate of 1 mm/min, and a backfill height of approximately 8.0 m was applied for the maximum normal stress. In addition, lubrication was performed using wrap and oil to minimize the influence of the wall friction of the soil box.

**Figure 5.** View of the installed geosynthetic strip: (**a**) setup view; (**b**) measurement locations.

ݒߪ In the testing program, different horizontal spacings (260 and 210 mm) were applied for the installation of the two types (50 and 70 mm widths) of the geosynthetic strips based on considerations of the applied block-type facing wall, as shown in Table 2. In this instance, when the horizontal spacing was 210 mm, pullout tests were conducted only in the normal stress condition (100 kPa). This corresponded to a backfill height of approximately 5 m in consideration of the most extensively applied MSE wall height in Korea for examination of the influence of the horizontal spacing.

**Table 2.** Testing program of geosynthetic strip pullout.


#### **4. Test Results and Analysis**

*4.1. Test Results*

Based on the pullout force–pullout displacement relationship from the pullout test results, the pullout strength characteristics of the reinforcement can be evaluated using the maximum pullout force at each normal stress condition. In this study, the maximum

pullout force for pullout strength evaluation was selected by referring to the displacement criteria suggested in the FHWA design criteria [38]. In other words, when the end displacement of the reinforcement installed in soil was less than 15 mm, it was necessary to determine the development of the maximum pullout force. This means that the pullout force corresponding to an end displacement of 15 mm must be applied as the maximum pullout force if the end displacement of the replacement exceeds 15 mm in the case in which the maximum pullout force was developed. Therefore, as shown in Figure 5b, the measurement position of 110 m from the front sides of the linear variable differential transformers installed in the reinforcement was applied as the end displacement.

In the cases of normal stress conditions of 50, 100, and 150 kPa, the maximum pullout forces were found to be 14.6, 20.4, and 26.8 kN for GS50W, and 19.3, 28.2, and 34.3 kN for GS70W, respectively.

Considering the FHWA design criteria [38], the relationships between the pullout force and front/end displacement according to the width of the geosynthetic strip and the horizontal spacing of reinforcement are shown in Figures 6 and 7. In all experimental cases, the maximum pullout force was observed when the front displacement ranged between 25 and 35 mm (Figure 6). Specifically, the maximum pullout force was approximately 5 to 8 kN higher when the width was 70 mm compared with a width of 50 mm. This is similar to the results of the wide-width tensile test. The maximum pullout force generated at the end displacement also satisfied the FHWA design criteria [38] (Figure 7).

**Figure 6.** Relationship between pullout force and front displacement for widths equal to: (**a**) 50 mm; (**b**) 70 mm.

**Figure 7.** Relationship between pullout force and end displacement for widths equal to: (**a**) 50 mm; (**b**) 70 mm.

From the pullout force–pullout displacement relationship at a normal stress of 100 kPa, the pullout force–front displacement relationship and the pullout force–end displacement relationship exhibited similar behaviors regardless of the horizontal spacing of the reinforcement. In other words, it was confirmed that friction-resistance reinforcement with the

same tensile strength exhibited the same pullout resistance and similar pullout behavior if the installation areas per unit width were identical. Therefore, the analysis used for the prediction of the pullout behavior and effective length of the reinforcement was conducted using only the test results with a larger horizontal spacing (260 mm, GS50W, and GS70W).

#### *4.2. Pullout Behavior of Geosynthetic Strip*

Figure 8a–c or Figure 9a–c show the pullout displacements as a function of the distance from the reinforcement front. The front pullout displacement was applied to a maximum of 60 mm after the maximum pullout force was developed. As the front pullout displacement increased, the displacement at each position showed a tendency to increase regardless of the reinforcement width. The displacement difference increased as the distance from the reinforcement front approached zero. This tendency was more obvious when the reinforcement width was wider (GS70W) and during exposure to higher normal stress conditions. Therefore, it was found that the pullout force transmitted to the reinforcement in soil during the pullout process was maximized at the front and decreased as the distance from the front increased. It was also found that the pullout force induced by the increase in normal stress was concentrated at the front.

**Figure 8.** Displacement as a function of distance from the geosynthetic strip front: (**a**) GS50W-50 kPa; (**b**) GS50W-100 kPa; (**c**) GS50W-150 kPa; (**d**) pullout displacement of reinforcement in maximum pullout force.

Figure 8d or Figure 9d show the pullout displacement according to the distance from the reinforcement front under the maximum pullout force condition. As shown in the figures, the pullout displacement decreased as the normal stress increased. At positions near the front, the reduction rate of the displacement increased at normal stress conditions. However, when the distance from the front exceeded approximately 45% to 50% of the entire reinforcement length, the reduction rates were found to be similar.

Figure 10 shows the results of analyzing the pullout force induced in the geosynthetic strips at the maximum pullout condition using the pullout displacement evaluation results according to the distance from the reinforcement front. The distance from the front was applied as the ratio of the displacement measurement position length (Li) to the total length of the reinforcement (L).

**Figure 9.** Displacement as a function of distance from the geosynthetic strip front: (**a**) GS70W-50 kPa; (**b**) GS70W-100 kPa; (**c**) GS70W-150 kPa; (**d**) pullout displacement of reinforcement in maximum pullout.

**Figure 10.** Relationship between pullout force and Li/L according to reinforcement width: (**a**) GS50W; (**b**) GS70W.

It was found that the pullout force induced in the geosynthetic strips occurred for a length ratio (Li/L) of approximately 0.45. In addition, it was found that 57–75% of the pullout force was induced at the front in the case of GS50W, and 68–83% in the case of GS70W, depending on normal stress conditions. For cases GS50W and GS70W, similar pullout forces were induced subject to the length ratio conditions of 0.5 L and 0.47 L, respectively. Furthermore, there was little change in the induced pullout force after 0.75 L and 0.73 L, at which the tensile strain values of GS50W and GS70W (calculated based on Figures 8 and 9) were approximately 1%.
