Mechanical Behavior of Reinforced Silty Sand: Focus on Application of Nano-Silica or Kaolin Coated Ceramic Fibers as a Reinforcement Material

Many studies have been done on the stabilization of weak soil using conventional chemical stabilizers such as lime, cement as well as modern materials such as nanoparticles; however, very few studies have examined the effect of coated bers on the strength of stabilized soil. This paper presents the results of a series of direct shear tests on soil specimens treated with ceramic ber, nanosilica, and kaolin. The effects of ceramic bers, ber length, nanosilica, and kaolin on the mechanical characteristics and shear strength of silty sand was investigated. The results show that the addition of ber to silty sand resulted in a signicant increase in the strength of the soil specimens. The dilative behavior of the soil specimen decreased with the addition of ceramic bers. The cohesion of the ber-reinforced specimens increased when the ber surface was coated with nanosilica or kaolin particles. The friction angle of the coated ber-reinforced specimens decreased with the addition of nanosilica particles; however, the friction angle of the coated ber-reinforced specimens was practically independent of the kaolin content.


Introduction
Techniques such as conventional stabilization and natural or synthetic ber reinforcement have been proposed for soil improvement. Studies have shown that conventional stabilizers such as cement and lime and reinforcing materials such as natural or synthetic bers can be used to strengthen and improve the mechanical behavior, hydraulic properties and freeze-thaw durability and reduce the swelling potential of geotechnical materials (Consoli et Taha and Taha (2012) indicated that the addition of nano-Al 2 O 3 to soil decreased both the expansive and shrinkage strains. Cui et al. (2018) studied the shear strength parameters and microstructure of silty sand treated with carbon ber and nanosilica. They reported that the shear strength parameters of specimens signi cantly increased with the addition of carbon ber and nanosilica. The addition of carbon ber to soil increased both the friction angle and cohesion; however, the nanosilica increased only the cohesion. Sarli et al. (2020) indicated that the addition of recycled polyester and nano-SiO 2 to loess soil increased its shear strength.
Previous studies showed that kaolin can be used to partially replace cement in in mortar and concrete, which can reduce energy consumption during cement production and is environmentally friendly (Sabir et  The current study investigated the mechanical characteristics of reinforced silty sand containing ceramic bers coated with nanomaterial or kaolin particles as a new stabilizer. The novelty of the current study includes the coating of the ceramic bers with nanosilica or kaolin particles in order to improve the interfacial interaction of the ber matrix. A series of direct shear tests was performed to determine the effect of the ber content and length, and nanosilica or kaolin content on the mechanical characteristics of the stabilized specimens. The interaction between the soil particles and ber and/or nanosilica-or kaolin-coated bers was examined using scanning electron microscopy (SEM).

Materials
The soil was collected from the Sejzi industrial zone to the east of the city of Isfahan in Iran. Fig. 1 shows the grain-size distribution curve and an image of the soil. Table 1 lists the physical and geotechnical properties of the soil. The modi ed Proctor compaction test was used to assess the maximum dry density (MDD) and optimum moisture content (OMC) according to ASTM D-1557. The soil was classi ed as silty sand (SM) according to the Uni ed Soil Classi cation System. Tables 2 and 3 show the physical properties and chemical composition of the ceramic bers, respectively.
The ceramic bers consisted of melted and blown kaolin melt with a high percentage of a mixture of pure alumina powder and mixed silica. After melting and blowing the kaolin melt in a furnace at 2000°C, the alumina and silica mixture was blown in by compressed air. These bers were white, exible, and had lengths of up to 50 mm and diameters of 2 to 3 µm. Tables 4 and 5 present the physical and chemical characteristics of the hydrophilic nanosilica and kaolin particles, respectively. Kaolin is a subgroup of clay that consists of alternate layers of silica and alumina. These form natural morphologies such as hexagonal platelets, rolled sheets and tubes (Wong et al. 2013). Table 6 lists the physical properties of the adhesive.

Testing program
The mechanical characteristics of silty sand treated with nanosilica-or kaolin-coated ceramic ber was investigated by direct shear testing. The effects of the ceramic ber content and length, and the nanosilica and kaolin contents on the mechanical characteristics of specimens were determined. Table 7 summarizes the details of the direct shear tests. The rst group of the tests evaluated the effects of the addition of ceramic ber and the ber length on the mechanical behavior of the reinforced specimens. The second and third groups examined the effects of the nanosilica and kaolin contents, respectively.
The fourth and fth groups of tests examined the effects of ber content and length, and the nanosilica and kaolin contents on the mechanical characteristics of the specimens treated with ceramic bers coated with nanosilica or kaolin particles, respectively. SEM images were used to examine the microstructure of the composites.

Specimen preparation
When preparing the specimens, the required amount of soil was dried in an oven for at least 24 h at approximately 110°C. Afterward, the amount of ceramic ber required for the ber-reinforced specimens was mixed until a uniform distribution of the ceramic ber was achieved in the mixture. Finally, the water was added to the mixture up to the optimum moisture content (OMC) and mixed until the mixture was uniformly moist. The specimens containing nanosilica or kaolin were prepared using a procedure similar to that used for ceramic ber-reinforced specimens.
Nanosilica-or kaolin-coated ceramic bers were created using an adhesive material. The ber coating increased the resistance of the bers to re and environmental effects. When preparing the specimens containing coated bers, the required amount of ceramic bers was weighed and then the adhesive was sprayed onto their surfaces. During spraying, the ceramic bers were rotated to assure that all surfaces of the bers were coated. The spraying time for each 0.5% of ceramic bers was held constant at 15 s. Every attempt was made to spread the adhesive homogeneously onto the surfaces of the ceramic bers. Next, the required amount of nanosilica or kaolin was sprayed onto the surfaces of the ceramic bers. After the ceramic ber coating was prepared with an adhesive layer, the nanosilica or kaolin particles were mixed with the dry soil. Finally, the required amount of water was added to the composite up to the OMC of the soil. All specimens were prepared at the OMC and MDD of the soil.

Testing apparatus
A direct shear test measures the shear strength properties of cohesion and the friction angle of the soil. In this study, a series of direct shear tests was conducted on the specimens according to ASTM D3080. The required amount of composite was poured into the shear box and then was compacted to achieve the MDD. The shear box had dimensions of 100 × 100 mm at a constant horizontal displacement rate of 0.2 mm/min. Three direct shear tests were conducted for each composite at three normal stress values (0.1, 0.2, and 0.3 MPa). The shear stress-shear strain curve, vertical strain-shear strain curve, cohesion, and friction angle of the composite were obtained from direct shear tests.

Direct shear tests results
The results of the direct shear tests on the natural soil specimens are presented in Fig. 2. As shown, the natural soil exhibited dilative behavior. Fig. 3 shows the response of the ceramic-ber-reinforced specimens at a ber content of 0.5% (by dry weight of soil) and ber lengths of 6, 12, and 18 mm. The ). An increase in the ceramic ber length caused a decrease in the dilative behavior.
In general, the peak shear strength of the specimens increased with an increase in the normal stress. The peak shear strength of the ceramic-ber-reinforced specimens occurred at a higher shear strain level compared to the natural soil specimens. The in uence of the increase of ber length did not show a regular effect on the peak shear strength. In other words, it had a dual in uence on the peak shear strength depending on the normal stress. For the reinforced specimens, there was a clear trend of increase in strain with an increase in the ber length at the peak shear stress. The shorter bers provided better ber orientation and dispersion because there were more of them at a given ber content than for the longer bers. This had a direct in uence on the strain at peak shear stress and resulted in greater adhesion strength between the bers and the matrix.
The effects of nanosilica content and kaolin content on the response of the specimens stabilized with additive contents of 0.1% and 0.5% (dry weight of soil) are shown in Figs. 4 and 5, respectively. As shown, the shear strength increased with the addition of nanosilica or kaolin particles to the soil. The specimens containing kaolin particles had a greater shear strength than the specimens containing nanosilica particles at a given additive content. The nanosilica-and kaolin-stabilized specimens appeared to be less dilative (or more contractive) than the natural soil specimens. This was due to strong and su cient bonding between the soil particles and additive particles.
The addition of water to the mixture caused a viscous gel to be produced by the nanosilica or kaolin particles that bound the soil particles together and lled the voids between the soil particles. This led to an increase in the shear strength. The peak shear strength of the specimens stabilized with nanosilica or kaolin particles occurred at a higher shear strain level compared to the natural soil specimens. However, the shear strength decreased with an increase in the nanosilica content from 0.1% to 0.5%. Fig. 6 shows the cohesion and friction angle of the specimens. It can be seen that both parameters increased with the addition of ceramic ber, nanosilica or kaolin particles to the soil. The specimen containing 0.5% kaolin particles had the highest cohesion value at 1.32 MPa. The specimens containing 0.5% ceramic ber had the lowest value at about 0.24 MPa. These values were almost independent of the kaolin content and ber length.
The friction angle increased sharply to the maximum values for specimens containing ceramic ber or kaolin particles. The friction angle was independent of ber length or kaolin content, however, an increase in the nanosilica content from 0.1% to 0.5% caused a decrease in the friction angle and an increase in cohesion. Increasing the nanosilica content increased the cohesion, but decreased the friction angle, which is in good agreement with the results reported by Cui et al. (2018). The total shear strength decreased signi cantly with an increase in the nanosilica content from 0.1% to 0.5% because the rate of decrease in the friction angle was greater than the rate of increase in cohesion. The increase in cohesion was related to the pore-lling in uence of the viscous gel produced by the nanosilica. The decrease in the friction angle could be related to the presence of small particles producing less friction and the excessive agglomeration of nanosilica, as has been explained by Cui et al. (2018).
The responses of specimens containing ceramic ber and nanosilica particles are presented in Figs. 7 to 9. The results show that the specimens containing 0.1% nanosilica exhibited more shear strength than those of containing 0.5% nanosilica. The specimens containing 0.5% nanosilica exhibited less dilation than those of containing 0.1% nanosilica at any ber length. Figs. 10 to 12 show the responses of specimens containing ceramic ber and kaolin particles and reveal that the specimens containing 0.5% kaolin exhibited more shear strength than those containing 0.1% kaolin at any ber length. The effect of the addition of kaolin on the volumetric behavior of the ceramic-ber-reinforced specimens varied according to the ber length. Fig. 13 presents the cohesion and friction angle versus the nanosilica or kaolin contents for different ber lengths. Coating the bers with nanosilica increased the cohesion, but decreased the friction angle. The total shear strength decreased with an increase in the nanosilica content used in the coating from 0.1% to 0.5%. This occurred because the rate of decrease in the friction angle was greater than the rate of increase in cohesion. The increase in cohesion was related to the viscous gel covering the ber surfaces, however, this viscous gel covering did not improve the interfacial bond properties and decreased the friction between the soil particles and ber surfaces.
The shear strength envelopes for the specimens at peak shear stress are shown in Fig. 14. It can be seen that the cohesion of the ber-reinforced specimens increased with the addition of nanosilica or kaolin particles. The increase in cohesion caused by the ceramic ber coating was more pronounced for the specimens containing kaolin particles. The cohesion increased with an increase in the ber length from 6 to 18 mm at a given nanosilica or kaolin content.
The friction angle of the ber-reinforced specimens decreased with the addition of nanosilica particles. In other words, the friction between the soil particles and ceramic ber surfaces decreased after the bers were coated with nanosilica. In contrast to the uncoated ceramic bers, which have a smooth surface, the coated ceramic bers showed different surface characteristics. These included the type of coating (nanosilica or kaolin), being ber wrapped, or having a deformed ribbed surface that provided a bond with the soil. However, the friction angle of the ber-reinforced specimens was almost independent of the kaolin content. The effect of ber length on the friction angle of the ber-reinforced specimens varied in the specimens containing nanosilica and kaolin particles.

SEM analysis
The ber-soil interaction in reinforced soil is complicated, especially at the microscopic scale. Tang et al. (2007) indicated that the binding material properties, applied stress condition, surface roughness of the bers, and contact between the bers and soil particles are the major parameters which govern the micromechanical characteristics of the ber-soil interface. In this work, after shearing, a number of specimens were analyzed using SEM. The SEM micrographs of the kaolin and nanosilica particles, ceramic bers, and specimens reinforced with ceramic bers coated by nanosilica or kaolin are shown in Figs. 15 and 16. It can be seen that the kaolin particles are more angular than the nanosilica particles.
The SEM micrographs show that the surfaces of the ceramic bers were clean and smooth. When the bers were coated with nanosilica or kaolin, some particles clung to the ceramic ber surface and formed an interlock which improved the interactions between the ceramic ber and the sand particles. The SEM images show that the ceramic ber surface coated with kaolin contributed to the bond strength, but the nanosilica particles were less angular than the kaolin particles. It can be concluded that the bers coated with kaolin were able to provide much greater pull-out resistance than the same bers coated with nanosilica.

Conclusions
In this study, a series of direct shear tests was conducted to investigate the effects of the addition of ceramic bers and the ber length, nanosilica content, and kaolin content on the mechanical behavior of silty sand. The test results produced the following conclusions.
It was observed that both the cohesion and friction angle increased with the addition of ceramic ber, nanosilica, or kaolin particles to the natural soil. The shear strength of the specimens increased and the dilative potential decreased with the addition of ceramic bers. The ceramic ber length had no signi cant effect on the shear strength parameters of the reinforced specimens.
The shear strength increased with the addition of nanosilica or kaolin particles to the soil. The specimens containing kaolin particles had more shear strength than the specimens containing nanosilica particles at a given additive content. The nanosilica or kaolin stabilized specimens appeared to be less dilative (or more contractive) than the natural soil specimens.
The friction angle also was independent of the kaolin content; however, an increase in the nanosilica content from 0.1-0.5% caused a decrease in the friction angle and an increase in cohesion. The cohesion of the ber-reinforced specimens increased when the ber surface was coated with nanosilica or kaolin particles. The increase in cohesion after coating the ceramic bers was more pronounced for specimens containing kaolin particles. The friction angle of the coated ber-reinforced specimens decreased with the addition of nanosilica particles; however, the friction angle of coated-ber-reinforced specimens was almost independent of the kaolin content. The effect of ber length on the friction angle of the coatedber-reinforced specimens varied in the specimens containing nanosilica and kaolin particles.      SEM images of: (a) ceramic bers, (b) ceramic bers with kaolin-particle coating, (c) ceramic bers with nanosilica-particle coating.