Elsevier

Engineering Geology

Volume 115, Issues 1–2, 6 September 2010, Pages 96-104
Engineering Geology

Behavior of rounded granular materials in direct shear: Mechanisms and quantification of fluctuations

https://doi.org/10.1016/j.enggeo.2010.06.008Get rights and content

Abstract

Fluctuations of vertical displacement and shear stress are routinely observed during shear tests on granular materials. This study was designed to examine these fluctuations to gain insight into their shear behavior. Three distinct particle size mixtures of glass beads were tested by an intermediate direct shear apparatus at four normal stress levels and three shearing rates. Based on examination of the onset and nature of growth/decay patterns of the fluctuations, four distinct stages were identified during shearing of these model materials. The fluctuations were demonstrated to reflect roughness of shear surfaces and variations in the driving force to accommodate this roughness. The Fast Fourier Transform (FFT) method was used to determine the dominant frequency bands and the amplitudes and wavelengths of the corresponding sine waves. These characteristic wave parameters were then linked to the shear strength parameters through the volumetric dilation rate, enabling determination of the friction coefficient of the system. It was demonstrated that the fluctuations are directly related to the intrinsic properties of soils and test conditions, and the current practice of averaging these fluctuations leads to the loss of valuable information. It is therefore suggested that residual strength of granular materials be presented in a way to recover and process this information.

Research Highlights

►Fluctuations of vertical displacement and shear stress were quantified; the fluctuations’ characteristic amplitude and wavelength were linked to the dilation component of shear stress; the friction coefficients were calculated directly from the raw data, making curve smoothing unnecessary. ►Shearing of rounded granular materials takes place in four stages: contraction, dilation, shear-induced packing and residual state. ►Three main factors influences the fluctuations: uniformity coefficient and normal stress increases whereas shearing rate decreases the amplitude and wavelength.

Introduction

Granular materials are frequently encountered in geotechnical engineering practice and other fields of human activity. Though their shearing mechanisms have been widely investigated over a century experimentally and recently by numerical modeling, there still is no consistent model that can successfully delineate and predict their behavior (Jaeger et al., 1996, Morgan, 1999, Aharonov & Sparks, 2002). One phenomenon, often observed during the laboratory tests and numerical simulations of granular materials, is the fluctuation of shear stress and vertical displacement with shear displacement. Many researchers (Taylor, 1948, Skinner, 1969, Oda & Kunishi, 1974, Shimizu, 1997, Tsai et al., 2003, Fukuoka et al., 2006) have encountered such fluctuations during the laboratory tests conducted by different types of shear apparatuses including direct shear, simple shear, triaxial compression or ring shear. Shimizu (1997) indicated that the amplitude of the fluctuation is so large, especially for the dense sand, that it may cause the shear stress to momentarily drop down even to zero. Fluctuations were also observed in numerical simulations (Bardet & Proubet, 1992, Morgan, 1999, Aharonov & Sparks, 2002, Mair et al., 2002). However, to date, a few studies addressed fluctuations only in descriptive terms (Taylor, 1948, Morgan, 1999, Oda & Kazama, 1998, Mair et al., 2002). In practice, the shear test results are averaged regardless of the magnitude of the fluctuations. Taylor (1948) may be the first to recognize the particle interlocking in the shear zone of granular materials and that the interlocking may cause volume change. He linked the shear stress τ and volumetric dilation rate δy/δx by τ/σ' = (δy/δx) + μ, where σ' and μ are normal stress and frictional resistance of soils and y and x are vertical and horizontal displacements, respectively.

Based on a series of numerical simulations of granular shear zones, Morgan (1999) indicated that the periodical fluctuations in stress with strain correspond to the dilation and contraction of such zones, given by ∆ = (W  W0)/W, where W and W0 are the instantaneous and initial thickness of the zone, respectively. Morgan (1999) also found the magnitude of the fluctuations correlate to the intrinsic properties of soils, such as the particle size distribution of soil grains and interparticle friction. However, these numerical models were configured to simulate the microstructural evolution of fault gouge, represented as a 0.9 cm thick zone composed of a narrow range of granular particles (0.5 to 0.063 mm), which was sheared under high normal stress (70 MPa) by the moving upper boundary. With this type of imposed boundary condition and normal stress level, Morgan's results may not be directly relevant to the cases often encountered in geotechnical engineering practice, where the loading mechanisms and shear zone development is better represented by direct shear and triaxial tests. Mair et al. (2002) observed the fluctuation of shear stress both in laboratory tests and numerical modeling of spherical and angular granular materials. They indicated that the amplitude of the fluctuations decrease with increasing shearing rate.

This paper presents a systematic experimental investigation into the shear behavior of granular materials. The artificial rounded granular materials were tested by means of a direct shear box. The main objective was to explore: 1) mechanisms leading to fluctuations of shear stress and vertical displacement, 2) means of quantifying these fluctuations and 3) the relationships between grain size distribution, test conditions and fluctuation parameters.

Section snippets

Materials

Two types of commercially available granular materials (nonporous soda lime glass sand and glass beads) were chosen as constituent materials to prepare the samples. The particle sizes of glass sand and glass beads had even distributions between 63 μm and 2.0 mm and between 2.0 and 6.0 mm with the mean sizes of 0.32 mm and 3.2 mm, respectively. The glass sand and glass beads had the same specific gravity of 2.48 g/cm3, rounded shape and smooth surface. These materials were essentially free of particle

Results and discussion

The assumption that the normal and shear stresses are distributed uniformly across the shear plane was adopted as the basis for calculation of the effective normal stress (σ') and shear stress (τ) by dividing the measured vertical load and shear force by the corrected area of shear surface. In the following, residual strength τR was used to refer to the shear stress at the residual state. The analysis of the test results was presented in terms of residual stress ratio τR/σ', where σ' is the

Conclusions

The fluctuations of both vertical displacement and shear stress, arising from the direct shear tests of rounded granular materials, were analyzed, quantified and used to draw inferences about the role of soil and test conditions. Based on the foregoing discussion, the following conclusions can be derived:

  • The fluctuations summarized quantitatively by their characteristic parameters (amplitude and wavelength) reflect the degree of irregularity or roughness of the shear surface.

  • Shearing process in

References (15)

  • M. Oda et al.

    Microscopic deformation mechanism of granular material in simple shear

    Jpn. Soc. Soil Mech. Found. Eng.

    (1974)
  • E. Aharonov et al.

    Shear profiles and localization in simulations of granular materials

    Phys. Rev. E

    (2002)
  • ASTM D 3080

    Standard test method for direct shear test of soils under consolidated drained conditions

    (2001)
  • J.P. Bardet et al.

    Numerical simulations of shear bands in idealized granular materials

    Solid State Phenom.

    (1992)
  • H. Fukuoka et al.

    Observation of shear zone development in ring-shear apparatus with a transparent shear box

    Landslides

    (2006)
  • H.M. Jaeger et al.

    Granular solids, liquids and gases

    Rev. Mod. Phys.

    (1996)
  • J.F. Lupini et al.

    The drained residual strength of cohesive soils

    Geotechnique

    (1981)
There are more references available in the full text version of this article.

Cited by (45)

  • Mechanical stabilization of aeolian sand with ceramic brick waste aggregates

    2023, Construction and Building Materials
    Citation Excerpt :

    The roughness and edges of the added ceramic brick particles aggregates, identified by SEM (Fig. 2) also promote this behavior. Recent reported researches, supported on different advanced experimental techniques, have also highlighted similar patterns about the influence of the superficial characteristics on the improvement on the behavior of soils [42–47]. The long-term performance of the waste aggregates employed has been analysed to assess the degradation of this stabilization proposal in practical constructions.

  • Experimental study on the effect of particle size on the shear characteristics of large-displacement soil exposed to heat treatment: Shear fluctuation and heat degradation

    2022, Engineering Geology
    Citation Excerpt :

    They showed that the friction strength and stability of the particle shear zone are affected by the particle shape, PSD, and shear zone evolution. Li and Aydin (2010) carried out ring-shear tests on soil samples from three slopes of the Three Gorges area using a large ring-shear apparatus, and proposed that the shear stress fluctuations are related to the Atterberg limit, particle shape, and shear rate of the material. Higher Atterberg limits and lower particle elongation were shown to be associated with smaller fluctuation amplitudes, whereas higher shear rates lead to smaller amplitudes, in which the shear stress fluctuation amplitude reflects the sample roughness.

View all citing articles on Scopus
View full text