Shear Strength Enhancement of Cemented Reinforced Sand: Role of Cement Content on the Macro-Mechanical Behavior

A series of direct shear tests were carried out on the sandy soil of the Chlef area, from the banks of the Chlef Valley. The soil was initially air dried and sieved through a 2 mm sieve, Fig. 3(a). Chlef sand particles are with a reduced grain shape and quartz with some carbonates, and they are usually isometric particles. Figure 1 shows a microscopic view of an electronic scanning of the tested sand. The basic properties of sand are given in Table 1, the grain size distribution curves of this soil with different cement content are shown in Figure 2. Many researches were performed to study the Chlef sand behavior (Arab et al., 2008, 2009, 2014; Djafar et al., 2011, Belkhatir et al., 2013; Benessalah et al. 2019 ; 2018 ; 2017 ; 2016 ; Della et al. 2011; Aouali et al., 2018; Merabet et al. 2019).

Figure 1

Figure 2

Figure 3

The cement used in this study is the cement of Chlef CEM II/A type of 42.5 classes, density 3.1 and a surface area of 3298 cm²/g. The silica fume used in cement is 8%, according to the standard EN 13263-1, having a density of 2.4 and a surface area of 220 000 cm²/g (Figure 3(b)). The chemical and mineralogical characteristics of the cement smoke are shown in Table 2. The mixture of soil and cement is introduced into a kneader for 2 minutes while adding appropriate water content.

A series of 117 tests was carried out in a rectangular shear box to study the effect of adding cement on the shear strength of the cemented sand. The shear box had a dimension of 60 x 60 mm horizontal cross section; the height of the test piece was 25 mm. All tests were performed according to the standard ASTM D-3080 (2005) with a constant horizontal shear rate of 1.00 mm/min.

Shear stress was recorded as a function of horizontal displacement to an average shear strain of 13%.

The first series in this laboratory study concerned shear tests on unreinforced Chlef sand with different water contents ω = 3, 7 and 10%, the second series carried out involved 108 tests on a mixture of sand-cement with a cement content Cc = 2.5, 5, 7.5 and 10%. The dense relative density (Dr) used was 80%. Samples were sheared under normal stresses sn = 100, 200 and 400 kPa. The sample preparation time was approximately one hour for setting the cement; then they are placed in a room at a temperature of 26° ± 2 for a duration of 7, 14 and 28 days with a humidity of 90%. The average cement concentration is defined as a percentage of the dry weight of sand, according to the following equation:

With: Wc the cement weight and Ws the sand weight.

Each sample was prepared in three layers (Figure 3(c)). The mass of each layer was determined according to the value of the initial relative density (Dr). Each layer was poured into the mold and then compacted manually; the number of strokes was 25 until obtaining the desired height for the dense tests. Table 3 shows all the parameters considered in the test program.

Figure 4(a) illustrates the results of the shear box tests on the unreinforced sand samples with water contents of 3,

Figure 4

7 and 10%, relative density Dr = 80% and sheared under a normal stress of 100 kPa. The results clearly show that an increase in water content has a significant impact on the soil shear strength. Soil resistance decreases with increasing moisture content. Figure 4(b) illustrates the variation of the vertical displacement as a function of the horizontal displacement. It is noted here that an increase in the water content leads to an increase in the vertical displacement, and therefore, the contractiveness of the sample increases, these results are consistent with those found by Benessalah et al. (2015).

Several researchers (Eathcote et al. 1994; Walker et al. 1995; Huang et al, 1998; Schnaid et al, 2001) showed the contribution of cement content in the evolution of soil resistance. Figure 5 illustrates the results of the direct shear

Figure 5

tests on cemented sand samples with a cement content Cc = 2.5, 5, 7.5 and 10% with a water content ω = 3, 7 and 10% emphasizing with comparison between them and the results of the unreinforced sand. These tests were sheared under a normal stress of 100 kPa. The results show that the shear strength of the unreinforced samples increases with increasing horizontal displacement without a peak appearance, while for samples reinforced with cement, it is found that the resistance increases significantly with the increase of the cement content up to a resistance peak of 80, 90, 100 and 110 kPa for Cc equal to 2.5 , 5, 7.5 and 10%, respectively; then the resistance drops followed by a residual resistance to the shear end for samples with water content of 3% (Figure 5(a)). The results of the tests of the samples made with a water content ω of 7 and 10% are illustrated in Figures 5(b) and 5(c); a very significant increase in the strength of the mixture is observed with the increase in the cement content where the shear strength reaches values of 106, 181 and 189 for water contents of 3, 7 and 10% for the samples mixed with 10% cement. Figures 6(a), 6(b) and 6(c) illustrate the variations of the vertical deformation as a function of the horizontal deformation; the results of these tests show a contractiveness phase for the small horizontal displacements followed by a phase of dilatancy, which reduces the deformability of the grounds against the external forces; on the other hand, the unreinforced sand showed only a contractiveness phase of deformation, which represents the vulnerability of unreinforced sand that will be represented by cracking during external loading.

Figure 6

Figure 7 shows the evolution of the maximum shear strength as a function of the cement content. Note that increasing the normal stress increases the shear strength considerably. It can also be seen that the shear strength of the sand-cement mixture increases with an increase of the cement content according to a quasi-linear growth

Figure 7

(coefficient of determination R²= 0.92, 0.98 and 0.98 for the three normal stresses σn = 100, 200 and 400 kPa, respectively).

Although a curved rupture envelope is expected for the sand-cement mixtures under study, by a linear regression between the shear stresses and the normal stresses applied to the samples at a given shear displacement; to determine the shear strength parameters including friction angle and the interlocking between the particles or the intercept cohesion. Variations in internal friction angle and intercept cohesion versus cement content are shown in Figure 8. Here, the interception of cohesion and the friction angle increase as the cement content increases. In particular; Figure 8(a) illustrates the evolution of the interlocking that has been presented in the figure as an intercept cohesion

Figure 8

of the mixture versus the cement content for water contents ω = 3 and 10%. The results clearly show that the intercept cohesion (C) of the soil increases linearly with the increase of the cement content and also with the water content (ω = 10% has greater values of intercept cohesions of all the samples under study, approximately the double of the values obtained with ω = 3%). Figure 8(b) illustrates the evolution of the friction angle with the increase in the cement content (Cc); here, we find that the friction angle (φ) increases significantly and with a linear manner with an increase in the cement content (Cc) for all the mixtures under study. Water content decreases immediately due to cement hydration and long-term pozzolanic reactions as water is absorbed and converted to cement-based products (Bergado et al., 1996, Bennert et al., 2000; Chew et al., 2004). So, the increase in curing time increases the shear strength, and consequently, also increases on the cohesion and the angle of friction because of continuous cementing reactions. Our results are in good agreement with those found by (Consoli et al., 2009; Amini et al., 2013; Asghari et al., 2003). These results show in addition the positive role of the water content in the improvement of the mechanical behavior of cemented reinforced sand and also indicate that the optimum water content is not yet reached taking into account the two values used in this study. Consoli et al. (1998) used the brittleness index as an indicator of the brittle behavior of soil as follows:

where: I_B is the brittleness index, and q_max and q_res are the peak and residual deviatoric stresses, respectively. They indicated that the addition of cement in the composite material leads to an increase in stiffness, brittleness and peak resistance. However, they noted that the inclusion of fibers increases both the peak and residual triaxial strength and reduces stiffness, which makes it possible to change the brittle behavior to a more ductile one.

It is known that cure temperature plays a very important role in the evolution of resistance; a high temperature of the environment generates a faster development of the resistance against a low temperature will slow down the setting process. Several studies have been developed in this direction, such as Hashimoto et al. (2009), Ahnberg et al. (2003), and Umesha and Val (2009). For temperatures below zero, the process will not be triggered by the freezing of water molecules. Figure 9 illustrates the results of stress-strain shear tests for cement-treated samples that were cured at 7, 14 and 28 days. It can be seen from the results that the shear strength increases significantly with increasing curing time, for example, the sample with 10% cement for a curing time of 7 days, has a resistance of 190 kPa; after 14 days, the resistance increases to 270 kPa and continues to increase to reach the value of 290 kPa after 28 days. This increase in resistance is due to curing time

Figure 9

on increasing the rigidity of the soil. Figure 10 illustrates the evolution of vertical deformations as the functions of horizontal deformations. It is noted that the increase in the cement content and the cure time considerably reduces the deformability of the mixtures that is represented by reducing the contractiveness character, and having a dilatant character after around 2 mm of horizontal displacement. These results are in perfect agreement with those found by Zillur Rabbi et al. (2011).

Figure 10

Figure 11 shows the evolution of maximum shear strength as a function of cement content and curing time. The results of the samples show that maximum shear strength increases almost linearly with increasing cement content and also with increasing cure time. These results are attributed to the interaction and also to the cement bonding of the sand particles, and consequently, the resistance and the improvement of the mechanical behavior increases. These results are close concordances with those found by Consoli et al. (1998) and Schnaid et al. (2001).

Figure 11

The intercept cohesion values obtained with deferent cement content for the three curing times namely 7, 14 and 28 days for the samples of high relative density (Dr = 80%), are presented in Figure 12a. It is noted that the intercept cohesion (C) increases very significantly with the increase in cement content according to almost linear growth (R²= 0.952, R²= 0.918, R²= 0.882, for 7, 14 and 28 days, respectively). It is also noted that shear strength increases with an increasing curing time. Figure 12(b) illustrates the evolution of the angle of friction as a function of the cement content. The results clearly show that the angle of internal friction also increases linearly with an increase in the cement content and also with an increase in curing time (R²= 0.986, R²= 0.934, R²= 0.874, for 7, 14 and 28 days).

Figure 12

Table 4 illustrates a summary of our results with those found in the literature, particularly those of Amini et al. (2013), Asghari et al. (2003) and Consoli et al. (2009). Comparing our results with those of Amini et al., (2013), we find that the addition of cement in the sand is

characterized by a certain efficiency because the cohesion value reaches 94 kPa for a cement content of 2% (Amini et al 2013), while Chlef sand needs 10% cement to reach this value for the same relative density (80% and 85%) and with the same cure time of 7 days. According to the results of Asghari et al, (2003); cohesion reaches a value of 375 kPa for a cement content of 4.5%.