Shear behavior of basalt fiber modified compacted red mudstone as subgrade fill material

This study experimentally investigated the shear behavior of basalt fiber-modified compacted red mudstone as a subgrade of a high-speed railway. We examined the effects of different vertical stresses, initial water contents, and fiber contents on brittleness and dilatancy. Several direct shear tests were conducted under the aforementioned conditions. The stress displacement curve tended to vary from softening to hardening with an increase in vertical stress. In the Mohr–Coulomb mode, except for the saturated state, the residual internal friction angle was greater than the peak value, whereas the residual cohesion was lesser than the peak value. The maximum cohesion occurred at the optimal water and fiber content, whereas the friction angle exhibited a downward trend with an increase in the contents discussed above. The brittleness index (I b ) was defined as the ratio of the peak to residual shear strengths. The magnitude of the brittleness index reduced with increase in the vertical stress. Moreover, the maximum value occurred at the optimal water and fiber content. Furthermore, the deformation mechanism was discussed based on the dilatancy angle(ψ). Based on the data, three different stages were proposed: Softening–Dilatancy (S1), the Hardening–Dilatancy (S2), and Hardening–Shrinkage (S3).


Introduction
Red mudstone is notoriously problematic for roads and other infrastructure built on it because of its poor mechanical properties; the particles of red mudstone are easily broken and have low strength.Moreover, they easily disintegrate and soften in water [1][2][3].However, according to China's development strategy, both high-speed and overloaded railways are planned to be built in Southwest China, which would unavoidably encounter the issues of red mudstone [4].One common strategy for addressing this is chemical improvement, which involves modifying the soil by adding other materials such as cement and lime [5].This modifies the mechanical behavior of the soil, and research in this field tends to be saturated.Fibers, as a widely studied modified material, can significantly improve the behavior of soil [6].
Over the last several decades, research has been aimed at characterizing the fundamental shear mechanism and identifying the various factors influencing the shear strength, including construction conditions [7], fiber content [8][9][10][11], and the initial state of the specimen [12] have been extensively investigated.In parallel, the evolution of the microstructure of the shear surface, as well as the distribution of the soil particles and fibers, could also be investigated [13].
To date, it has been well accepted that the elastic modulus and shear strength of various types of fibers, including polyester fiber, polypropylene fiber, and basalt fiber, could be ideal enough to modify the soil, 1332 (2024) 012013 IOP Publishing doi:10.1088/1755-1315/1332/1/012013 2 which could satisfy the requirements of construction [14][15][16].Among these fibers, basalt fiber has a relatively large elastic modulus and exhibits a more stable behavior when exposed to water, heat, extreme acid, and alkaline environments [17][18][19].To the best of our knowledge, there are no comprehensive data on the shear mechanism of basalt fiber-modified compacted red mudstone.This study aimed to fill this gap in the literature.
This study conducted several direct shear tests under different vertical stresses, initial water contents, and fiber contents on basalt fiber-modified compacted red mudstone.A comprehensive study of the shear behavior of the specimens under the given conditions was obtained.The study was then further discussed by the interaction of the dilatancy angle () and the brittleness index (  ).

Materials
The basalt fibers used in this study were obtained from the Zhejiang Jinshi Company.The diameter and length were 16 μm and 12 mm, respectively.Moreover, the initial state of the fiber was bundled and was separated into filamentary fibers.The basic physical characteristics of the basalt fibers used are listed in table 1.
Table 1.Physical and mechanical properties of basalt fiber.was used in this study.The material was collected from the superficial layer (1-3 m) of Tianfu District in Chengdu, which belongs to the Jurassic Penglaizhen Group (J3P).The materials was allowed to disintegrate naturally and then was using a hammer mill.Particles smaller than 1 mm were sieved.
Moreover, based on the light compaction test, the maximum dry density of the red mudstone and the optimal water content were 1.77/ 3 and 16.3%, respectively.In addition, based on the Malvern test, the particle of the fully weathered red mudstone was wide in distribution range and good in gradation.
In the X-Ray diffraction (XRD) test, the main components of the red mudstone (from highest to lowest) were quartz, calcite, albite, muscovite, and montmorillonite.Hereafter, the basalt fiber-modified compacted red mudstone is denoted as FMRM.

Experimental program
This experimental study aimed to characterize the shear mechanisms of basalt FMRM, with a particular focus on the effects of vertical stress, initial water content, and fiber content.According to a previous study [20], maximum cohesion occurs at approximately  = 13%; therefore, the tests focused on the fiber content were operated under this certain magnitude of initial water content.The tests series were organized as follows.

Specimen preparation and apparatus for direct shear test
The uniformity of the distribution of the basalt fibers in the specimens is of vital significance and could directly influence the results of the tests.Based on the existing researches [6,10], the sieved particles were first dried in an oven at 105 ℃.Subsequently, the basalt fiber bundle, at five different fiber contents of 0, 0.1, 0.2, 0.3 and 0.4%, was added into the soil and then malaxated until only filamentary fiber remained.During this process, we regarded the basalt fibers to be evenly distributed.The well-mixed soil was then wetted at four different initial water contents (5, 8.7, 13 %, and 16 %) using the water spray method.The soil was then sealed in a plastic bag for 24 h to allow moisture equilibrium.After that, all the specimens were statically compacted in the ring (61.8 mm in internal diameter and 20 mm in height) to a certain dry density of 1.7/ 3 .Further, the specimens at saturated state were wetted up from 13%.
The localized failure of a shallow slope is generally assumed parallel to the slope surface [21,22].When a gentle-angle slope or subgrade is encountered, the shearing mode is best represented by a direct shear test.A direct-shear apparatus (Geocomp) was used for the direct-shear test under constant vertical stress.The specimens were carefully transferred into a shear box to minimize possible residual lateral stress.Subsequently, six series of constant vertical stresses of 25, 50, 100, 200, 300 and 400 kPa were applied in the consolidation mode for 10 min.Thereafter, the pins fixed in the shear box were pulled out and the shear process could be operated in a rate of 0.16933/.The shear test was finished when the maximum displacement reached 10 mm.

Effect on stress-displacement behavior
Figure 1 presents the general shear response of FMRM specimens with a basalt fiber content of   = 0.2% under 100 .Herein, the behavior tendency could be clearly identified.Consistent with the existing data of the literature [20], in a certain range (0 = 5-13%), the higher the water content, the lower the peak shear strength.However, when the initial water content exceeded the one with the maximum cohesion, that is, 0 = 16%, the shear strength dropped significantly.Moreover, the displacement required to reach the peak strength increased with the increase of the water content.In terms of the influence of vertical stress on the shear response at 0 = 13% (see figure 2), the displacement required to reach the peak shear strength increased with the increase in vertical stress.In addition,, the stress-displacement curves tended to vary from softening to hardening when the vertical stresses increased.When the vertical stress and initial water content remained the same (see figure 3), the residual shear strength could be significantly modified by adding basalt fibers.However, the peak shear strength as well as the residual shear strength decreased when the fiber content exceeded   = 0.2%.Thus, this boundary content is the optimal basalt fiber content.According to the existing studies [10,23], the fibers interlace to form a spatial network structure that enhances the friction between the fibers and soil particles.Nevertheless, excessive fibers directly result in an uneven distribution and cannot separate into filamentary fibers.This decreases the contact surface of soil and soil, as well as soil and fiber, which results in a loss of friction and a decrease in shear strength.

Effect on stress-displacement behavior
To obtain the shear behavior of the FMRM, a method for determining the peak and residual shear strengths should first be proposed (see figure 4).
Figure 5 shows the Mohr-Coulomb strength lines for different water contents.In the previous section, the cohesion reached the largest magnitude when 0 = 13%.As a boundary, the peak shear strength increases with the increase in water content before the boundary.However, the peak shear strength decreases when it exceeds 13%.The reason for the phenomenon is that the water inside the specimen first enhances the cohesion of FMRM when the water content is less than 13%.Nevertheless, when it is beyond the boundary, with the increase in water content, the bounded water film wrapped around the soil particles becomes thicker, which has a lubricating effect on the friction between the soil particles.Moreover, the mechanical interlocking between the particles is weakened.According to a previous study [4], red mudstone is sensitive to water, and easily disintegrates and softens when exposed to water.However, basalt fiber shows stability against water [19], implying that the water stability of the red mudstone could be modified by mixing the appropriate content of basalt fiber.Figure 6 shows the effect of basalt fiber content on the peak shear strength.To clearly visualize the variation trends, only 0, 0.2%, and 0.4% are illustrated.From section 4.1, the optimal fiber content was set as   = 0.2% and the rule was obvious that the maximum shear strength occurred at the condition of optimal fiber content.
The cohesion and internal frictional angle were obtained by linear fitting in the form of Shear stress (kPa) Shear stress (kPa) Shear strength /kPa Shear stress (kPa)  =  +    (1) where  is the cohesion of the soil and  is the internal frictional angle.As shown in figures 7 and 8, the peak cohesion increased as the water contents increased before 0 = 13% and then exhibited a drop tendency.The differences between residual and peak cohesion were reduced in this process as well.Moreover, for the internal frictional angle, except for the saturation state, the residual ones exceeded the peak ones and the residual internal frictional angle reduced with the increase in water content.Figure 10.Effect of fiber content on internal friction angle.Figures 9 and 10 illustrate the effect of fiber content on the parameters mentioned in equation ( 1).Note that in Section 4.1, 0.2% was considered as the optimal fiber content.With the increase in the fiber content (less than   = 0.2%), both peak and residual cohesion tended to increase, while they reduced when   > 0.2%.As for the internal friction angle, a downward tendency was simulated in this process.
Except in the saturated state, the residual cohesion of the specimens was smaller than the peak cohesion, whereas the residual internal friction angles were greater than the peak values.This is because, when the vertical load is small, the stress displacement curves of the specimens are mostly softened.Thus, the residual shear strength can be much smaller than the peak shear strength.With an increase in the vertical stress, the stress-displacement curves exhibited hardening, and the residual shear strength was close to or equal to the peak shear strength.Therefore, the coefficients of the Mohr-Coulomb theory should follow the rule that the residual cohesion is always lower than the peak cohesion and the residual internal friction angles are greater than the peak values.
We proposed to use the brittleness index (noted   , defined by [24]) to interpret the magnitude of shear strength.It is defined as the ratio of peak to residual shear strengths.Moreover, it can indicate the contractiveness and the severity of strain softening of soil.As shown in figures 11 and 12, the shear responses of the specimens changed from softening to hardening with the increase in the vertical stress, when approaching the magnitude of   = 1.Moreover, when the initial water content was relatively small, the difference between the peak as well as the residual shear strength was high, and the softening trend was more obvious.In the same situation, for the comparison between different fiber contents, the addition of the fiber could significantly reduce the brittleness, indicating that the basalt fiber could modify the strength characteristics and transfer the shear response to the hardening mode.

Effect on the dilatancy angle
The dilatancy caused by interlocking friction directly affects the strength of the soil and geometrical characteristics of the shear plane at failure [25].It is well studied that dilatancy is an important part of contributing to the strength of the soil for both saturated and unsaturated soil [26,27].Therefore, a dilatancy angle was proposed to interpret the strength of the soil.
Figure 13 illustrates that, with an increase in water content, the dilatancy angle exhibited a tendency to first increase and then decrease.Moreover, the peak value occurs at 0 = 13%.As for the effect of fiber content, figure 14 show that basalt fiber could reduce the magnitude of dilatancy angle.

Results and discussion
The shear behavior of the basalt fiber-modified red mudstone in the compacted state was characterized with a particular focus on the shear strength and dilatancy behavior.The tests demonstrated that the shear strength, brittleness index, and angle were influenced by the initial water content, basalt fiber content, and vertical stress.The results obtained in this study would provide a new construction strategy for railway subgrades designed for red bed regions.Consequently, the following conclusions were drawn: 1) The stress displacement curve was strongly influenced by the initial water content, basalt fiber content, and vertical stress.Moreover, the curve tends to change from softening to hardening as the vertical stress increases.
2) The brittleness index gradually decreased with an increase in the vertical load until the value reached a magnitude of 1.Therefore, except in the saturated state, the residual cohesion was smaller than the peak cohesion, whereas the residual internal friction angle was greater than the peak value.3) Shear dilatancy gradually transforms from dilatancy to shrinkage with an increase in vertical stress.
Moreover, there is a boundary value for the vertical stress to distinguish dilatancy.
Brittleness index and shrinkage.Additionally, in the semi-logarithmic coordinate system, the dilatancy angle decreased linearly with an increase in vertical stress.4) The change in the brittleness index with the dilatancy angle can be divided into three main stages (S1-S3), which implies that the interaction between the dilatancy and shear strength properties of the specimens can be divided into three types: softening-dilatancy (S1), hardening-dilatancy (S2), and hardening-shrinkage (S3).The addition of basalt fiber could bring S2 forward compared to plain soil.

Figure 3 .
Figure 3.Effect of fiber content on stressdisplacement curves.

Figure 4 .
Figure 4. Schematic illustration of peak stress and residual stress.

Figure 5 .
Figure 5.Effect of water content on peak shear strength.

Figure 6 .
Figure 6.Effect of fiber content on peak shear strength.

Figure 7 .
Figure 7. Effect of water content on cohesion.Figure 8. Effect of water content on internal friction angle.

Figure 11 .Figure 12 .
Figure 11.Effect of water content on brittleness index.

Figure 13 .
Figure 13.Effect of water content on dilatancy angle.
. Direct shear test programs.