Taslagyan Karen 201409 PHD

The Shear Strength of Granular Soils under the Influence of Vibration
by
Karen Taslagyan
A thesis submitted in partial fulfillment of the requirements for the degree of
Doctor of Philosophy
in
Geotechnical Engineering
Department of Civil and Environmental Engineering
University of Alberta
© Karen Taslagyan, 2014

Abstract
For some geotechnical design projects where soils are exposed to vibration, it becomes
necessary to evaluate the strength and deformation characteristics of the soil under existing
and/or anticipated vibrations. If large displacements are anticipated, the design of an earth
structure, such as an earth or tailing dam, has to take into account the residual shear strength
characteristics of the soils that underlie or/and constitute the structure. In this case, if the soil is
subjected to vibration, depending on the magnitude of the vibrations, it may undergo some loss
of strength, thus resulting in excessive deformation and compromising the stability of the
structure. Therefore, it is of paramount importance to predict and consider the effects of
vibrations on the residual strength of soils; that is, the effect of vibration on the critical state, a
unique state independent of the initial density (void ratio) of the soil that can be reached upon
sufficiently large shear deformations.
This thesis presents a summary of an experimental program performed to investigate the
influence of vibration on the strength and deformation characteristics of dry granular media with
no cementation bonds. A new vibrating direct shear apparatus is designed and built to evaluate
the mentioned characteristics under a wide range of vibrational accelerations. The apparatus is
used throughout the research project to investigate the behavior of dry granular materials at their
residual strength states under the influence of vibration with different intensities. The impact of
vibration on the overburden (normal) stress and the peak strength of granular materials is
evaluated. Experiments have also been conducted to study the post-vibration behavior of
granular materials, as well as the particle shape effect on the strength loss and deformations of
the granular media during vibration. Finally, the behavior of the shear zone in the granular
ii
materials before, during and after the application of vibration at the pre-peak and residual
strength states is investigated.
About 200 samples of different granular materials have been tested by using the new
vibrating direct shear apparatus with different testing modes, procedures, normal stresses and
vibration intensities. A number of plots are presented that show the behavior of the granular
materials under the impact of vibration in different testing conditions. The Mohr-Coulomb
equation for granular materials has been modified to account for the effect of vibrational
fluidization. A new pattern of the shear zone in granular media is provided, which outlines the
deformation and shear resistance before, during and after the application of vibration.
iii
Acknowledgements
First of all, I would like to thank my supervisor, Professor Dave H. Chan, for his invaluable
support and guidance throughout the whole course of this research. Professor Chan has always
found the time to kindly share his knowledge and inspire new ideas. I also want to extend my
gratitude to my co-supervisor, Professor Norbert R. Morgenstern, for his important comments
and suggestions on the research project. The discussions with Professor Morgenstern have
always been encouraging and stimulating.
Many thanks to Christine Hereygers, Steve Gamble and Chris Krath for their assistance in
building the laboratory apparatus used in this research.
The financial support for this project was provided by the Natural Science and Engineering
Council of Canada (NSERC Postgraduate Scholarship) and the Alberta Innovates Technology
Futures (Graduate Student Scholarship).
Special thanks to my family and friends for their endless help and motivation.
iv
Table of Contents
Abstract ........................................................................................................................................... ii
Acknowledgements ........................................................................................................................ iv
Table of Contents ............................................................................................................................ v
List of Figures .............................................................................................................................. viii
List of Tables ................................................................................................................................ xii
1. Chapter 1: Introduction .............................................................................................................. 1
1.1 Statement of Problem ........................................................................................................... 1
1.2 Research Objectives ............................................................................................................. 3
1.3 Scope of the Thesis .............................................................................................................. 4
1.4 Organization of the Thesis ................................................................................................... 6
1.5 References ............................................................................................................................ 8
2. Chapter 2: Literature Review ..................................................................................................... 9

3. Chapter 3: A Direct Shear Apparatus with Vibrational Loading ............................................. 16
3.1 Abstract .............................................................................................................................. 16
3.2 Introduction ........................................................................................................................ 17
3.3 Strain controlled vibrating direct shear apparatus .............................................................. 20
3.4 Stress controlled vibrating direct shear apparatus .............................................................. 24
3.5 Testing of the Apparatus .................................................................................................... 27
3.5.1 Strain Controlled Testing Mode .................................................................................. 27
3.5.2 Stress Controlled Testing Mode .................................................................................. 32
3.6 Conclusions ........................................................................................................................ 35
3.7 References .......................................................................................................................... 35
4. Chapter 4: Effect of Vibration on the Critical State of Dry Granular Soils ............................. 37
4.1 Abstract .............................................................................................................................. 37
4.2 Introduction ........................................................................................................................ 37
4.3 Previous Studies on Vibrational Loading on Soils ............................................................ 39
v
4.4 Testing Equipment and Procedures .................................................................................... 43
4.5 Tested Materials ................................................................................................................. 46
4.6 Test Results under Vibration .............................................................................................. 51
4.6.1 The Effect of Strength Loss Due to Changes in Normal Stress Caused by Vibration 65
4.6.2 The Effects of Acceleration on Residual Strength Loss .............................................. 68
4.7 Conclusions ........................................................................................................................ 72
4.8 References .......................................................................................................................... 73
5. Chapter 5: Vibrational Fluidization of Granular Media ........................................................... 78

5.1 Abstract .............................................................................................................................. 78
5.2 Introduction ........................................................................................................................ 78
5.3 Testing Equipment and Procedures .................................................................................... 80
5.4 Tested Materials ................................................................................................................. 83
5.5 Test Results ........................................................................................................................ 86
5.6 Conclusions ...................................................................................................................... 113
5.7 References ........................................................................................................................ 115
6. Chapter 6: Effect of Vibration on the Shear Zone in Dry Granular Materials ....................... 120
6.1 Abstract ............................................................................................................................ 120
6.2 Introduction ...................................................................................................................... 120
6.3 Testing Equipment and Procedures .................................................................................. 124
6.4 Tested Materials ............................................................................................................... 127
6.5 Test Results ...................................................................................................................... 129
6.5.1 Pre-Peak and Residual Strength States of Granular Materials under Vibration ........ 129
6.5.2 Post-Vibrational Strength and Volumetric Changes of Granular Materials .............. 137
6.6 Conclusions ...................................................................................................................... 150
6.7 References ........................................................................................................................ 152
7. Chapter 7: Summary, Conclusions and Recommendations ................................................... 155
vi
7.1 Summary .......................................................................................................................... 155
7.2 Conclusions ...................................................................................................................... 156
7.3 Recommendations ............................................................................................................ 159
References ................................................................................................................................... 161

APPENDIX ................................................................................................................................. 172
Part 1 ...................................................................................................................................... 172
Part 2 ...................................................................................................................................... 187
Part 3 ...................................................................................................................................... 268
Part 4 ...................................................................................................................................... 329
vii
List of Figures
Figure 3.1 Schematic of the vibrating direct shear apparatus ....................................................... 19

Figure 3.2 (a) – Conventional strain controlled direct shear apparatus with load cell & (b) – with
proving ring. ................................................................................................................. 21
Figure 3.3 (a) – Strain controlled vibrating direct shear apparatus with load cell and (b) – with
proving ring. ................................................................................................................. 22
Figure 3.4 Prototype of the strain controlling vibrating direct shear apparatus............................ 23
Figure 3.5 (a) – Stress controlling vibrating direct shear apparatus with a load cell and (b) – with
a proving ring. .............................................................................................................. 25
Figure 3.6 Prototype of stress controlling vibrating direct shear apparatus. ................................ 26
Figure 3.7 Photograph of sand taken by using a microscope ....................................................... 28
Figure 3.8 Particle size distribution of the tested sand samples ................................................... 28
Figure 3.9 Static and dynamic shear strength envelops of sand ................................................... 30
Figure 3.10 Shear stress (a), vertical displacement (b), and horizontal (c) and vertical (d)
acceleration responses versus normalized horizontal displacement of sand under a
normal stress of 50 kPa. ............................................................................................... 31
Figure 3.11 Displacement and time response of stress controlled direct shear testing with
vibration at different levels of peak stresses with normal stress of 50 kPa .................. 33
Figure 3.12 Shear stress (a), horizontal (c) and vertical displacement (e), horizontal (d) and
vertical (f) acceleration responses versus time of sand in stress controlled tests at 50
kPa normal stress and shear stress equal to 50% of the static shear strength............... 34
Figure 4.1 (a) – Strain controlled direct shear apparatus and (b) – modified strain controlled
vibrating direct shear apparatus. ................................................................................... 44
Figure 4.2 Photograph of the modified direct shear apparatus ..................................................... 45
Figure 4.3 Particle size distribution of the tested materials .......................................................... 48
Figure 4.4 Pictures of the glass beads and sand under a microscope at magnification of 50x. (a)
0.1 mm glass beads; (b) 0.55 mm glass beads; (c) fine sand; and (d) coarse sand ...... 49
Figure 4.5 Shear strength diagrams of 0.1 mm glass beads (a), 0.55 mm glass beads (b), fine sand
(c) and coarse sand (d).................................................................................................. 51
viii
Figure 4.6 Shear stress, vertical displacement, horizontal and vertical accelerations versus
normalized displacement of fine sand subjected to a normal stress of σ=50 kPa,
vibration frequency of 140 Hz and vibration force of 5.18 N. ..................................... 54
Figure 4.7 Shear stress plots of three fine sand samples tested at normal stress of 118 kPa and
different vibration accelerations ................................................................................... 55
Figure 4.8 Shear strength envelops of 0.1 mm glass beads tested at: (a) vibration force of 1.61 N;
(b) vibration force of 3.22 N; (c) vibration force of 3.71 N; (d) vibration force of 5.18
N and (e) vibration force of 7.14 N. ............................................................................. 60
Figure 4.9 Shear strength diagrams of 0.55 mm glass beads tested at: (a) vibration force of 1.61
N; (b) vibration force of 3.22 N; (c) vibration force of 3.71 N; (d) vibration force of
5.18 N and (e) vibration force of 7.14 N. ..................................................................... 61
Figure 4.10 Shear strength diagrams of fine sand tested at: (a) vibration force of 1.61 N; (b)
vibration force of 3.22 N; (c) vibration force of 3.71 N; (d) vibration force of 5.18 N
and (e) vibration force of 7.14 N. ................................................................................. 62
Figure 4.11 Shear strength diagrams of coarse sand tested at: (a) vibration force of 1.61 N; (b)
vibration force of 3.22 N; (c) vibration force of 3.71 N; (d) vibration force of 5.18 N
and (e) vibration force of 7.14 N. ................................................................................. 63
Figure 4.12 Total strength loss and strength loss due to reduced normal stress for 0.1 mm glass
beads. ............................................................................................................................ 66
beads. ............................................................................................................................ 66
Figure 4.14 Total strength loss and strength loss due to reduced normal stress for fine sand. ..... 67
Figure 4.15 Total strength loss and strength loss due to reduced normal stress for coarse sand. . 67
Figure 4.16 Total strength loss vs. horizontal acceleration of 0.1 mm glass beads ...................... 70
Figure 4.17 Total strength loss vs. horizontal acceleration of 0.55 mm glass beads .................... 70
Figure 4.18 Total strength loss vs. horizontal acceleration of fine sand....................................... 71
Figure 4.19 Total strength loss vs. horizontal acceleration of coarse sand ................................... 71
Figure 5.1 (a) – Strain controlled direct shear apparatus (LVDTs that measured the vertical and
shear displacements are not shown for simplicity) and (b) – modified strain controlled
vibrating direct shear apparatus. ................................................................................... 81
Figure 5.2 Modified direct shear apparatus. ................................................................................. 82
ix
Figure 5.3 Particle size distribution of the tested materials. ......................................................... 84
Figure 5.4 Photographs of the glass beads and sand samples under a microscope at magnification
of 50x. (a) - 0.55 mm glass beads; (b) - fine sand; and (c) - coarse sand ..................... 85
Figure 5.5 An example of test results of a fine sand sample tested at σ=200 kPa, =60 Hz and
h=2.3. .......................................................................................................................... 88
Figure 5.6 Shear strength diagrams of glass beads tested at (a) - h0.44, (b) - h1.38, (c) -
h2.12 and (d) - h3.4. ............................................................................................. 94
Figure 5.7 Shear strength diagrams of fine sand tested at (a) - h0.42, (b) - h2.42 and (c) -
h4.25. ........................................................................................................................ 96
Figure 5.8 Shear strength diagrams of coarse sand tested at (a) - h0.44, (b) - h2.5 and (c) -
h4.28. ........................................................................................................................ 97
Figure 5.9 Friction angles of the tested materials at different vibration intensities. ..................... 99
Figure 5.10 σf values at different vibration intensities. (Note that σf = 0 at h = 0. These points
represent a static condition). ....................................................................................... 100
Figure 5.11 Effect of vibration on the strength loss and compressive deformation of glass bead
samples tested at h = 0.4. .......................................................................................... 103
samples tested at h = 2. ............................................................................................. 105
Figure 5.15 Effect of normal stress on post-vibrational strength and compressive deformation of
glass beads tested at σ = 8 kPa. .................................................................................. 107
glass beads tested at σ = 23 kPa. ................................................................................ 108
x
glass beads tested at σ = 118 kPa. .............................................................................. 111
vibrating direct shear apparatus. ................................................................................. 125
Figure 6.2 Modified direct shear apparatus. ............................................................................... 126
Figure 6.3 Particle size distribution of the tested materials. ....................................................... 128
Figure 6.4 Photographs of the glass beads and sand samples under a microscope at
magnification of 50x. (a) 0.55 mm glass beads and (b) fine sand. ............................ 129
Figure 6.5 Shear strength diagrams of 0.55 mm glass beads (a) and fine sand (b). ................... 132
Figure 6.6 Vibration of fine sand at pre-peak and residual strength states. ................................ 134
Figure 6.7 Vibration of 0.55 mm glass beads at pre-peak and residual strength states. ............. 135
Figure 6.8 Schematic cross-section of the shear zone in a granular material. ............................ 137
Figure 6.9 Test results of dense 0.55 mm glass beads tested at σ = 23 kPa. .............................. 140
Figure 6.10 Test results of loose 0.55 mm glass beads tested at σ = 23 kPa. ............................. 141
Figure 6.11 Test results of dense fine sand tested at σ = 23 kPa. ............................................... 142
Figure 6.12 Test results of loose fine sand tested at σ = 23 kPa. ................................................ 143
Figure 6.13 Shear strength diagrams of 0.55 mm glass beads (a) and fine sand (b). ................. 146
Figure 6.14 An example of test results of fine sand sheared over smooth glass surface. ........... 148
Figure 6.15 Example of test results of 0.55 mm glass beads sheared over smooth glass surface.
.................................................................................................................................... 149
xi
List of Tables
Table 3.1 Measured peak, residual and vibro-residual strengths of sand samples ....................... 29
Table 4.1 Physical characteristics of the glass beads and sands ................................................... 47
Table 4.2 Test results of 0.1 mm glass beads ............................................................................... 56
Table 4.3 Test results of 0.55 mm glass beads ............................................................................. 57
Table 4.4 Test results of fine sand ................................................................................................ 58
Table 4.5 Test results of coarse sand ............................................................................................ 59
Table 4.6 Friction angles of 0.1 mm glass beads. ......................................................................... 64
Table 4.7 Friction angles of 0.55 mm glass beads ........................................................................ 64
Table 4.8 Friction angles of fine sand ........................................................................................... 64
Table 4.9 Friction angles of coarse sand. ...................................................................................... 65
Table 5.1 Physical characteristics of the tested materials. ............................................................ 84
Table 5.2 Test results of glass beads. ............................................................................................ 89
Table 5.3 Test results of fine sand. ............................................................................................... 91
Table 5.4 Test results of coarse sand. ........................................................................................... 92
Table 5.5 Friction angles and normal stress axis intersection values of the shear strength
diagrams for the three granular materials tested at different vibration intensities. ...... 98
Table 6.1 Physical characteristics of the tested materials. .......................................................... 128
Table 6.2 Shear strengths and strength losses at pre-peak and residual strength states for 0.55 mm
glass beads and fine sand............................................................................................ 130
Table 6.3 Test results of 0.55 mm glass beads and sand sheared over smooth glass surface. .... 145
xii
1. Chapter 1: Introduction
1.1 Statement of Problem
Granular soils are frequently exposed to vibrations due to earthquakes, blasting, construction
operations, machinery and vehicle traffic. The strength behavior of a soil during vibration not
only depends on the vibration characteristics, such as acceleration, frequency and amplitude, but
also on the physical properties of the soil, such as moisture content, grain size distribution,
particle shape, dry density or void ratio, cohesion and internal friction angle, as well as the
density and mineralogy of the soil particles.
There are two important aspects that contribute to the unique properties of granular
materials: ordinary temperature plays no role on their mechanical behaviour, and the interactions
between grains are dissipative because of static friction and the inelasticity of collisions (Jaeger
et al., 1996). Granular material can behave like solids, fluids and even gas under different
conditions. When sufficient energy is supplied to a granular material in a vibrating system, the
granular material can exhibit fluid-like behavior. This transition from a solid state to a liquid
state (fluidization) takes place when vibrational acceleration, a, exceeds a certain critical value.
When granular materials are subjected to strong enough vibration intensities, various
phenomena, such as compaction, mixing, localized excitations, convective flow, size segregation
and surface wave formation, can be observed. Although the above mentioned processes have
been well investigated, it should be mentioned that there are very few, if any, studies done on the
effect of normal stress on the vibrational fluidization of granular materials.
1
If large displacements are anticipated, the design of an earth structure, such as an earth or
tailing dam, has to take into account the residual shear strength characteristics of the soils that
underlie or/and constitute the structure. In this case, if the soil is subjected to vibration,
depending on the magnitude of the vibrations, it may undergo some loss of strength, thus
resulting in excessive deformation and compromising the stability of the structure. Therefore, it
is of paramount importance to predict and consider the effects of vibrations on the residual
strength of soils.
Despite the knowledge gained on this issue in the last few decades, there is still a lack of
understanding on the mechanism of particle interactions and shear deformation due to vibration.
Therefore, there is a need to study the shear strength and deformation behavior of soils due to
vibration. In particular, it is specifically interesting to determine the effect of vibration on the
residual shear strength of soils, that is, the effect of vibration on the critical state, which is a
unique state independent of the initial density (void ratio) of soil that can be reached upon
sufficiently large shear deformations.
The strength of a granular material is determined by the strength characteristics of the
material at the shear zone/band defined by a certain combination of boundary conditions and the
mechanical properties of the material. Therefore, to better understand the mechanism of the
strength loss of granular materials, it is important to investigate the stress and deformation
characteristics of the shear zone during its initiation and propagation through the granular media.
The thickness of the shear zone is usually 8 to 10 times the mean grain diameter (Roscoe, 1970;
Muhlhaus and Vardoulakis, 1987; Bardet and Proubet, 1992).
Direct shear testing (used in this research project) has successfully been used to evaluate
the shear strength characteristics of soils for many decades. Despite some of the disadvantages of
2
the testing, such as lack of control of pore pressure, failure at predefined planes and non-uniform
stress conditions in the tested samples, it is one of the most common shear strength evaluation
tests used in geotechnical laboratories. The popularity of direct shear testing arises from the
simplicity of the setup and ease of testing procedures.
1.2 Research Objectives
The main objectives of the research project are as follows.
1. Design and build a laboratory apparatus and develop a testing methodology that can be used
to investigate the effect of vibration on the shear strength characteristics of granular materials
at different strength states, normal (confining) stresses and vibrational accelerations.
2. Investigate the behavior of granular materials at their residual strength states under the
influence of vibration with different intensities.
3. Investigate the impact of vibration on the normal (overburden) stress and determine the
amount of strength loss of the granular materials due to the normal stress changes when
subjected to vibrations of different acceleration values.
4. Evaluate the post-vibration behavior of granular materials.
5. Investigate the impact of vibration on the peak strength of dense granular materials by
subjecting them to vibration at pre-peak states.
6. Determine the particle size and shape effect on the behavior of the granular materials
subjected to vibration.
7. Investigate the behavior of the shear zone in the granular materials before, during and after
the application of vibration at the pre-peak and residual strength states.
3
8. Use the experimental results to develop a comprehensive framework of the behavior of
granular materials during and after vibration.
The developed framework will relate all of the mentioned experimental results and reveal
the mechanism of the strength behavior of granular soils during and after vibration. The
framework will be used to develop a constitutive model for granular materials subjected to
vibration. This will make it possible to more accurately predict the strength loss, as well as
compression and shear deformations of granular soils for the anticipated or existing vibrations.
The framework will allow for a more realistic and reliable design of structures on/in
granular soils subjected to vibration. Of special interest will be the contribution of the research
data to improve design and stability analyses of tailing dams in seismically active areas. It is also
anticipated that the research project will assist towards a better understanding of the extreme
mobility of big rock avalanches.
1.3 Scope of the Thesis
The following steps have been carried out to achieve the research project objectives.
 A new vibrating direct shear apparatus is built. The apparatus has been designed and built to
investigate the effect of vibration on soils at different strength states, especially at the critical
state, as well as to evaluate the impact of vibration on the deformation properties of soil under
different normal, shear and vibrational loading conditions. The new apparatus represents a
modification of the conventional WYKEHAM FARRANCE direct shear apparatus, such that
4
soil samples can be tested in both strain and stress controlled modes under different
vibrational frequencies, amplitudes and accelerations.
 Experiments have been carried out on granular materials at relatively low vibrational
accelerations. The materials are tested in the strain-controlled mode at different vibration
intensities and normal stresses to investigate the effect of vibration on the residual strength
and deformation characteristics of the tested materials. The particle shape effect on the
strength loss during vibration is evaluated, as well as the impact of vibration on the normal
(overburden) stress that acts on the samples.
 Experiments are implemented on granular materials at relatively high vibrational
accelerations. To investigate the mechanism of the vibrational fluidization of dry granular
media, three different types of granular materials are tested by using the vibrating direct shear
apparatus at greater vibrational accelerations. The effects of normal stress and vibrational
acceleration on the shear strength characteristics and the fluidization of granular media are
evaluated.
 Experiments that focus on the behavior of the shear zone in granular materials subjected to
vibrations are carried out. The behavior of the shear zone before, during and after the
application of vibration is investigated at the pre-peak and residual strength states of two
different granular materials. The effects of the particle shape, normal (confining) stress and
vibrational acceleration on the shear zone are also studied.
5
1.4 Organization of the Thesis
This thesis has been prepared in a paper-based format. A brief outline of the main chapters are
provided below.
Chapter 3 is a description of the setup and testing procedures of the new vibrating direct
shear apparatus that is designed and built to evaluate the strength and deformation characteristics
of soils under a wide range of vibrational accelerations. The results of two sets of sand samples
tested in stress and strain controlled modes are provided, which prove the workability and
reliability of the new apparatus.
Chapter 4 is a summary of the test results of eighty samples from four different granular
materials which are tested with the new vibrating direct shear apparatus at four normal stresses
and different vibrational accelerations. Strength loss vs. vibration acceleration plots at different
normal stresses, as well as the peak, residual and vibro-residual shear strengths obtained for the
four granular materials at different intensities of vibration, are provided. The results show the
effect of vibration on the critical state of the dry granular media; particularly, a reduction in the
residual friction angles of the tested materials with an increase in the intensity of vibration. The
effect of particle shape on the strength loss due to vibration and the effect of vibration on the
normal stress are also discussed.
Chapter 5 is a summary on the experiments conducted to investigate the vibrational
fluidization of granular media. The test results of sixty samples from three different granular
materials which are tested by using the modified direct shear apparatus at six normal stresses and
a wide range of vibration accelerations are provided. The peak, residual and vibro-residual shear
strength envelops obtained for the tested materials at different vibration intensities are presented.
6
It is shown that an increase in vibration intensity reduces the friction angle of the granular
materials, and increases the value of the normal stress, the vibro-fluidizational limit, below
which, the granular material is fluidized. The particle shape effect on strength loss due to
vibration is demonstrated, as well as a reduction in friction angle and increase in vibro-
fluidizational limit with an increase in vibration acceleration.
An outline of the results of the laboratory experiments conducted with glass beads and
sand by using the modified vibrating direct shear apparatus and different testing procedures is
provided in Chapter 6. The behavior of the shear zone in the granular materials before, during
and after the application of vibration at the pre-peak and residual strength states is discussed.
Three zones are identified in a sheared granular material: A – zone unaffected by the shear of the
granular media, B – the shear zone portion that has been developed due to the shear of the
material, but does not contribute to the critical state of a thinner shear zone, C, at which the
actual shear takes place. Also, the shear strength and deformation characteristics of the granular
materials affected by vibration at their pre-peak and residual strength states are presented and
discussed.
7
1.5 References
Bardet, J. P., and Proubet, J. (1992). “Shear-Band Analysis in Idealized Granular Material”. J.
Eng. Mech., 118, pp. 397-415.
Jaeger, H. M., Nagel, S. R. and Behringer, R. P. (1996). Granular solids, liquids, and gases. Rev.
Mod. Phys. 68, 1259
Muhlhaus, H. B., and Vardoulakis, I. (1987). "The thickness of shear bands in granular
materials." Geotechnique, 37, pp. 271-283.
Roscoe, K. H. (1970). “Tenth Rankine lecture: The influence of strains in soil mechanics”,
Geotechnique, 20, pp. 129-170.
8
2. Chapter 2: Literature Review
Granular soils are often exposed to vibrations that are natural and human-made in origin.
Examples of vibration types include earthquakes, blasting, machinery, vehicle traffic and wind.
Depending on their intensity, vibrations may result in partial or complete strength loss, which
would lead to significant deformation of soils.
For example, when vibration is applied in the vertical direction, fluidization takes place at an a
that is greater than 1 g (Huan, 2008). Further increases in the vibration acceleration changes the
behavior of the granular material to that of a gas. Jaeger et al. (1996) described the three
aforementioned states of granular materials.
phenomena, such as compaction (e.g. Barkan (1962), Ayer and Soppet (1965/1966)), swelling
(e.g. Poschel and Rosenkranz (1998)), mixing (e.g. Alexeev et al. (2000)), localized excitations
(e.g. Umbanhowar et al. (1996), Tsimring and Aranson (1997)), convective flow (e.g. Laroche et
al. (1989), Evesque and Rajchenbach (1989), Gallas et al. (1992), Pak et al. (1995) and Huan
(2008)), Bourzutschky and Miller (1995), Wassgren (1997) and Liffman et al. (1997)), size
9
segregation (e.g. Knight et al. (1993), Cooke et al. (1996) and Alexeev et al. (2000)) and surface
wave formation (e.g. Pak and Behringer (1993), Melo et al., (1993 and 1995), Clement et al.
(1996), Brone and Muzzio (1997), and Mujica and Melo (1998)) can be observed. Although the
above mentioned processes have been well investigated, there are very few, if any, studies done
on the effect of normal stress on the vibrational fluidization of granular materials.
Richards et al. (1990) proposed the concept of “dynamic fluidization” which takes into
consideration the effect of earthquake accelerations on dry granular soils. The imposed
accelerations at some critical level change the state of the soil, which causes general
plastification, such that the soil becomes, in a sense, an anisotropic fluid. They assumed that the
main trigger of the fluidization is the inertial forces that act between the particles of a granular
soil. They showed that fluidization mainly depends on horizontal and not vertical accelerations.
Another distinguishing feature of fluidization is that when it occurs, flow takes place, if at all, in
finite increments rather than continuously, with increments that correspond to the acceleration
pulses of an earthquake above a critical value (Richards et al., 1990).
Other researchers who investigated the vibro-fluidization of granular materials are Savage
(1988), Fauve et al. (1989), Doudy et al. (1989), Zik et al. (1992), Lan et al. (1995), Warr et al.
(1995), Goldstein et al. (1995), Luding (1995), Ristow et al. (1997), Tennakoon et al. (1998),
Falcon et al. (1999), Sunthar et al. (2001), Moon et al. (2004) and Gotzendorfer et al. (2006).
There has been significant progress in understanding the effects of vibration on the strength
and deformation properties of soils since the first half of the last century (Housner, 1954, 1959;
Barkan, 1962; Richart, 1970; Seed and Idriss, 1982, 1983; Idriss and Boulanger, 2008). A
number of different experiments with vibration application have been conducted on cohesive and
cohesionless soils that generated valuable data which have led to some important conclusions
10
(Savchenko, 1958; Pyke et al., 1975; Robertson and Campanella, 1985, 1986; Wartman et al.,
2005; Meehan et al., 2008). All of these findings significantly help to improve the design of
different structures subjected to dynamic loads (Barkan, 1962; Seed, 1966; Richart, 1970; Seed
and Idriss, 1982; Das, 2011; Sangroya and Choudhury, 2013).
Pokrovsky et al. (1934) were the first to experimentally investigate the influence of
vibrations on the internal friction of sand and showed that the coefficient of internal friction
decreases with an increase in the kinetic energy of vibration. These results were later supported
by experiments performed by Barkan (1962). Nowadays, there are different types of laboratory
apparatuses to measure the strength properties of soils under vibrational loads. Depending on the
design and technical characteristics, each of the tests has its advantages and limitations in
simulating field conditions. The following are some examples of the above mentioned techniques
and the investigations performed by using these techniques.
Levshinsky designed a stress-controlled vibrating shear apparatus that generates cyclic
shear loads (horizontal vibration) on soil samples. It was later used by Preobrazhenskaya and
Savchenko in 1958 to determine the shearing strength of soils with vibration (from Meschyan,
1992). Youd (1970) investigated shear strength reduction and density changes of granular
materials under vibration by mounting a direct shear apparatus onto a shaker table. Some
examples of work done with the use of shaking table tests are that by Arango and Seed (1974),
Wartman et al. (2005), and Toyota et al. (2004), who used the technique to investigate vibration
induced deformations in slopes. Meschyan and Badalyan modified a stress-controlling torsional
shear apparatus into a stress-controlling vibrating torsional shear apparatus that allows the testing
of thin cylindrical samples under vibrational torsional shear loads generated by a vibrator
(Meschyan, 1978). Melosh et al. (1995) built a rotational viscometer to quantitatively investigate
11
the relationship between stress and strain rate in a bucket of strongly vibrated dry, fine sand. A
similar setup, with the use of a penetrating cone instead of a sanding drum (Melosh et al., 1995),
was used to evaluate the strength of soils under vibration (Kutergin, 1989). Shibata et al. (1970),
who conducted experiments on sand, used a vibrating triaxial apparatus.
Of special interest is the effect of vibration on the critical state (residual shear strength) of
granular soils, a unique state independent of the initial density (void ratio) of soil that can be
reached upon sufficiently large shear deformations. There is very little research done in this
regard, an example of which is the work done by Youd (1968), who experimentally investigated
the effect of vibration on the critical state of granular media and found that both the shear
strength and void ratio are reduced by vibration.
The strength of a granular material is determined by the strength characteristics of the
material at the shear zone/band defined by a certain combination of boundary conditions and the
mechanical properties of the material. Therefore, to better understand the mechanism of the
strength loss of granular materials, it is important to investigate the stress and deformation
characteristics of the shear zone during its initiation and propagation through the granular media.
The thickness of the shear zone is usually 8 to 10 times the mean grain diameter (Roscoe, 1970;
Muhlhaus and Vardoulakis, 1987; Bardet and Proubet, 1992).
Some examples of the numerous studies carried out on the shear zone in granular media are
provided below.
Bardet and Proubet (1992) investigated the emergence, inclination, and thickness of shear
bands in idealized granular materials and showed that they are similar to those of real materials.
Aidanpaa et al. (1996) used a torsional shear cell to study the shear layers of uniform spheres,
and observed that the shear zone thickens and dilates with an increase in the shear speed,
12
accompanied by a transition from a single layer to many layers of shearing. Bora (1984)
experimentally investigated the shear failure mechanism in granular materials, and found that the
friction angle is dependent on effective confining contact stress within a dilative range regardless
of the drainage conditions and amount of particle crushing. Frost et al. (2002) studied the shear
failure behavior of granular–continuum interfaces on a selection of sand–continuum material
interfaces and through the use of discrete element modeling. Borja (2003) used a geological and
mathematical framework to classify deformation patterns in granular media. Frost et al. (2004)
evaluated the interface behaviour of granular soils by carrying out physical and numerical
experiments. Rechenmacher (2005) evaluated local displacements and strains that are associated
with shear band growth and evolution in sands by testing the plane strain with the use of digital
image correlation (DIC). The magnitude of the shear and rotational strains was found to vary
along the length of the shear bands, which lends support to the idea of fluctuating buildup and
collapse of the “force chains”. Alshibli and Alramahi (2006) investigated the evolution of local
strains during the shearing of particles of a granular material, and showed that when compared to
particle sliding, rotation is imperative in the shearing resistance of granular materials. Abriak and
Caron (2006) conducted an experimental study of shear in granular media, and showed that
granular media behavior depends on the local friction (grain–grain friction). Sadrekarimi (2008)
studied the shearing behavior of loose and dense sands by implementing constant volume ring
shear tests. Widulinski et al. (2010) performed comparative modeling of shear localization in
granular bodies with the finite and discrete element methods (FEM and DEM). Liu (2010)
conducted laboratory experiments and numerical simulations to investigate the failure
characteristics and micro-mechanical behaviors of granular soil slopes. Cox and Budhu (2010)
studied the grain shape of granular materials, and through the use of light microscopy,
13
determined that there are several grain shape parameters. They then related these parameters to
the dilatancy of the granular materials. Mesarovic et al. (2013) analyzed the changes in the
topology of a granular assembly during deformation by using the graph theory and showed that
the elementary mechanism of diffuse deformation consists of intermittent flips, an increasing
number of which directly stems from dilatancy, as well as that shear band formation is associated
with the massive rolling of particles.
Direct shear testing has successfully been used to evaluate the shear strength characteristics
of soils for many decades. Despite some of the disadvantages of the testing, such as lack of
control of pore pressure, failure at predefined planes and non-uniform stress conditions in the
tested samples, it is one of the most common shear strength evaluation tests used in geotechnical
laboratories these days. The popularity of direct shear testing arises from the simplicity of the
setup and ease of testing procedures.
Many researchers have studied the applicability and limitations of direct shear testing
through experimental investigations and numerical modelling. One of the more early studies on
direct shear testing through the use of finite element analysis was performed by Potts et al.
(1987). They demonstrated that, despite the strongly non-uniform stresses and strains in the shear
box before failure, strains and stresses in the final failure zone are surprisingly uniform. Liu et al.
(2005) investigated the interface friction through direct shear tests. Two improvements were
made to reduce the friction of the sample material at the inner surface of the upper shear box,
which caused the real shear strength to increase for a dense sample and decrease for a loose
sample. Bagherzadeh-Khalkhali and Mirghasemi (2009) performed numerical modeling and
experiments by using direct shear testing to study its suitability and limitations for testing coarse-
grained soils. Li and Aydin (2010) investigated the fluctuations in vertical displacement and
14
shear stress with different sized glass beads during direct shearing. Nam et al. (2011) used
multistage direct shear tests to determine the shear strength of unsaturated soils, which allowed
the matric suction to be independently controlled, and compared the results with samples which
underwent conventional direct shear testing. Härtl and Ooi (2011) investigated, through direct
shear tests, how particle shape and interparticle friction would influence bulk friction by using
DEM. They showed that packing density has less influence than particle interlocking on the bulk
friction. Kang et al. (2012) performed 3D discrete element simulations in conjunction with image
processing of the pore geometry to determine the pore size distribution and orientation in dilative
and contractive assemblies in direct shear testing.
Despite the knowledge obtained on this topic in the last few decades, there is still a lack of
Therefore, granular media strength and deformation behavior under the impact of vibration
warrants further research work.
15
3. Chapter 3: A Direct Shear Apparatus with Vibrational Loading
3.1 Abstract
For some geotechnical design projects where soils are exposed to dynamic loads (vibration), it
becomes necessary to evaluate the strength and deformation characteristics of the soil under
existing and/or anticipated vibrations. In order to investigate the effect of the vibrations on the
strength and deformation properties of soils, representative samples should be collected and
tested in laboratories and subjected to vibration of expected magnitudes. In this case, it is
important that the laboratory equipment used is able to simulate field conditions as close as
possible to provide the necessary parameters that can successfully be used in the design. A
vibrating direct shear apparatus has been developed based on the conventional direct shear
apparatus to evaluate the strength and deformation characteristics of soils (granular and
cohesive) under a wide range of vibrational accelerations and frequencies. The apparatus makes
it possible to test soils in both stress and strain controlled modes. The design of the apparatus is
such that it allows modification of the most commonly used direct shear apparatuses into
vibrating ones. The new apparatus has been built and tested to prove its workability and
reliability.
16
3.2 Introduction
Soils are frequently exposed to dynamic loads (vibrations) due to natural and human-made
causes. Examples of vibration types include earthquakes, blasting, machinery and vehicle traffic.
Depending on their intensity, vibrations may result in partial or complete strength loss, which
leads to significant deformation of soils. Of specific interest is the effect of vibration on the
critical state (residual shear strength) of soils, a unique state independent of the initial density
(void ratio) of the soil that can be reached upon sufficiently large shear deformations.
Pokrovsky et al. (1934) were the first to experimentally investigate the influence of
vibrations on the internal friction of sand and showed that the coefficient of internal friction
decreases with an increase in the kinetic energy of vibration. These results were later supported
by experiments performed by Barkan (1962). Nowadays, there are different types of laboratory
apparatuses to measure the strength properties of soils under vibrational loads. Depending on the
design and technical characteristics, each of the tests has its advantages and limitations in
simulating field conditions. The following are some examples of the above mentioned techniques
and the investigations performed by using these techniques.
Levshinsky designed a stress-controlled vibrating shear apparatus that generates cyclic
shear loads (horizontal vibration) on soil samples. It was later used by Preobrazhenskaya and
Savchenko in 1958 to determine the shearing strength of soils with vibration (from Meschyan,
1992). Youd (1970) investigated shear strength reduction and density changes of granular
materials under vibration by mounting a direct shear apparatus onto a shaker table. Some
examples of work done with the use of shaking table tests are that by Arango and Seed (1974),
Wartman et al. (2005), and Toyota et al. (2004), who used the technique to investigate vibration
induced deformations in slopes. Meschyan and Badalyan modified a stress-controlling torsional
17
shear apparatus into a stress-controlling vibrating torsional shear apparatus that allows the testing
of thin cylindrical samples under vibrational torsional shear loads generated by a vibrator
(Meschyan, 1978). Melosh et al. (1995) built a rotational viscometer to quantitatively investigate
the relationship between stress and strain rate in a bucket of strongly vibrated dry, fine sand. A
similar setup, with the use of a penetrating cone instead of a sanding drum (Melosh et al., 1995),
was used to evaluate the strength of soils under vibration (Kutergin, 1989). Shibata et al. (1970),
who conducted experiments on sand, used a vibrating triaxial apparatus.
It should be mentioned that presently, there are very few, if any, laboratory equipment that
can be used to determine the effect of vibration on soil at different states, especially at the critical
state, as well as to evaluate the impact of vibration on the deformation properties of soil under
different normal, shear and vibrational loading conditions.
A vibrating direct shear apparatus has been designed and built that meets the requirements
mentioned above. The new apparatus represents a modification of a conventional WYKEHAM
FARRANCE direct shear apparatus, such that soil samples can be tested in both strain and stress
controlled modes under different vibrational frequencies, amplitudes and accelerations. One of
the main advantages of this apparatus is that vibration can be applied at the residual state. Also,
the modifications can easily be implemented in most existing direct shear apparatuses in
conventional geotechnical laboratories.
The changing of a conventional direct shear apparatus into a vibrating direct shear
apparatus requires the placement of an actuator between the top half of the shear box and the
load cell to apply shear loading onto the sample as shown in Fig. 3.1. The actuator generates
vibration in the horizontal (shear) direction by horizontally expanding and contracting, and
induces vibration on the top half of the shear box. The actuator can be electromagnetic,
18
piezoelectric or pneumatic. By using a control panel connected to the actuator, the frequency
and force of vibrations can be changed. To measure acceleration, frequency and amplitude of
vibrations, an accelerometer is attached to the top half of the shear box. To measure the
acceleration, frequency and amplitude of vibrations in the vertical direction, another
accelerometer is attached to the top of the loading plate which is subjected to a normal load. The
readings can be recorded by using an oscilloscope, data-logger or a PC.
Porous plate
Accelerometers Loading plate

Load cell
Soil Actuator
sample
Shear box
Figure 3.1 Schematic of the vibrating direct shear apparatus
19
3.3 Strain controlled vibrating direct shear apparatus
Presently, most direct shear apparatuses use a load cell to measure the shearing resistance of a
soil (see Fig. 3.2a). However, there are some that still use a proving ring, as shown in Fig. 3.2b.
Modifications can be made in both types of devices that use a load cell and/or a proving ring.
There are advantages and limitations in both cases. A proving ring allows more compression
under loading and therefore enables the actuator to apply larger amplitudes of vibration onto the
sample. A load cell delivers high stiffness which requires a larger actuator.
In strain controlled testing with a load cell, see Fig. 3.3a, the long shafts (4) that are
attached to the force transducer (5) must be replaced with shorter shafts (14) to accommodate the
actuator (13), which is mounted between the shear box (3) and the load cell (5). Depending on
the testing conditions, an additional load cell can be placed between the control panel (1) and the
shear box (3) in order to measure the shear load which is less affected by the actuator (13).
In order to modify the strain controlled direct shear apparatus (Fig. 3.2b) into a strain
controlled vibrating direct shear apparatus with a proving ring for shear load measurement (Fig.
3.3b), the main body of the direct shear apparatus (7) has to be extended with an extension (15)
for sufficient space to accommodate the location of the actuator (13) between the shear box (3)
and the proving ring (12).
20
1 2 3 4 5 4
6
(a)
7
8
10
9
11
9
(b) 12
Figure 3.2 (a) – Conventional strain controlled direct shear apparatus with load cell & (b) – with
proving ring.
Dial gages (or LVDTs) that measure vertical and shear displacements are not shown for
simplicity. 1 – Control panel of the direct shear apparatus; 2 – Shaft that pushes the shear box
(3); 3 – Shear box with soil sample; 4 – Shaft attached to load cell (5); 5 – Load cell that
measures the shear force; 6 – Screw for zero setting of shear load before starting a test; 7 – Main
body of the direct shear apparatus; 8 – Frame that transfers normal load to the soil sample (3); 9
– Legs of the direct shear apparatus that support the main body (7); 10 – Lever that provides
21
normal load to the soil sample; 11 –Weights to apply the normal load; 12 – Proving ring to
measure the shear force.
(a) 3 5
13 14
(b)
3 13 12
15
Figure 3.3 (a) – Strain controlled vibrating direct shear apparatus with load cell and (b) – with
proving ring.
13 – Actuator that generates vibrations; 14 – Short shaft attached to the load cell (5); 15 –
Extension of the main body (7) in Fig. 3.2 of the direct shear apparatus to accommodate the
proving ring.
22
A photograph of the prototype of the strain controlled vibrating direct shear apparatus is
shown in Fig. 3.4. Note that both the proving ring and a load cell (on the right hand side of the
proving ring) are used together for this apparatus, as well as another load cell is placed on the left
hand side of the shear box.
Figure 3.4 Prototype of the strain controlling vibrating direct shear apparatus
23
3.4 Stress controlled vibrating direct shear apparatus
In stress controlled testing, two pulley (17) carrying cables (16) are placed at the corners of
the main body (7) on both sides of the zero setting screw (6) (see Fig. 3.5a). One end of the
cables (16) is attached to the hanger (18) that carry the weights (19) and the other end is attached
to two pins (20) mounted on both sides of the bottom half of the shear box (3). In this case, the
control panel (1) of the direct shear apparatus is not used and the bottom half of the shear box (3)
is disconnected from the shaft (2), thus preventing movement during the testing. The bottom half
of the shear box (3) is moved by the shear load generated by the weights (19) placed on the
hanger (18) (see Fig. 3.5a). Fig. 3.5b shows a schematic of a modified stress controlling vibrating
direct shear apparatus with both a proving ring and a load cell. A photograph is shown in Fig.
3.6.
In this modification, the actuator (13) generates vibration shear loads, which, depending on
the intensity of the vibration and the shear load, may cause shear deformation or failure of the
soil.
24
(a)
20 16
17
16
18
19
(b)
Figure 3.5 (a) – Stress controlling vibrating direct shear apparatus with a load cell and (b) – with
a proving ring.
16 – Cable that transfers shear loads to the sample by pulling the bottom half of the soil sample;
17 –Pulley carrying cable (16); 18 – Hanger attached to cables (16) and carrying the weights
(19); 20 – Pins mounted to the shear box (3) in Fig. 3.3 that connect to the cables (16).
25
A photograph of the prototype of the stress controlling vibrating direct shear apparatus is
given in Fig. 3.6. Note that both the proving ring and load cell (on the right hand side of the
proving ring) are used. Also, the shear box is disconnected from the gearbox shaft. An
additional dial gauge is used to measure shear deformation.
Figure 3.6 Prototype of stress controlling vibrating direct shear apparatus.
26
3.5 Testing of the Apparatus
3.5.1 Strain Controlled Testing Mode
To determine the workability of the strain controlled vibrating direct shear apparatus, dry fine
sand samples that were angular in shape were tested to obtain both static and dynamic shear
strength characteristics. A picture of a representative sample taken by a microscope and the
particle size distribution of the tested material are provided in Figs. 3.7 and 3.8, respectively. The
density of the sand ρ = 1.46 g/cm3, density of solids ρs = 2.65 g/cm3 and void ratio e = 0.815.
The size of the shear box was 60 mm x 60 mm x 32 mm (W x L x H) and loading was applied at
a shear rate of 0.61 mm/min under eight normal stresses (8.4, 23.2, 36.3, 50, 77.3, 118.2, 159.1
and 200 kPa). After reaching the residual shear strength, the samples were subjected to vibration
(frequency = 140 Hz; horizontal acceleration ≈ 0.4 g; vertical acceleration ≈ 0.15 g) and changes
in shearing resistance (dubbed “vibro-residual strength” here) were recorded. Two linear variable
differential transformers (LVDTs) were used to measure the vertical and horizontal
displacements, two load cells to measure the shear resistance of the samples at the bottom and
top halves of the shear box, and two uniaxial accelerometers to measure the vertical and
horizontal vibration accelerations of the samples (see Fig. 3.4). Measurements were taken at a
frequency of 1 kHz by using the NI CompactDAQ System, which was connected to a computer
that logged the data with NI LabVIEW software (National Instruments). The test results are
shown in Table 3.1 and Fig. 3.9.
27
Figure 3.7 Photograph of sand taken by using a microscope
100
90
80
70
60
%
50
finer
40
30
20
10
0
0.01 0.1 1
mm
Figure 3.8 Particle size distribution of the tested sand samples
28
Table 3.1 Measured peak, residual and vibro-residual strengths of sand samples
Sample # Normal stress Peak strength Residual strength Vibro-residual strength

σ (kPa) τp (kPa) τf (kPa) τfd (kPa)
1 8.4 15 11 5
2 23.2 30 21 14
3 36.3 38 29 21
4 50 51 37 30
5 77.3 78 56 43
6 118.2 106 80 61
7 159.1 148 113 91
8 200 173 134 110
Friction angle 400 330 290
As shown in Table 3.1 and Fig. 3.9, vibration reduces the residual friction angle by about
40 and cohesion1 by about 4.5 kPa. An example of the typical plots of shear resistance, vertical
deformation, and horizontal and vertical accelerations are shown in Fig. 3.10. From the shear
resistance plot, it is seen that vibration causes the residual shear strength to drop to the vibro-
residual strength. When vibration is stopped, the residual strength returns to the pre-vibration
value. This means that the decrease in residual strength is temporary and only occurs during the
application of vibration. Note that the normalized displacement in the plots of Fig. 3.10 is
defined as d/L, where d is the horizontal displacement and L is the initial length of the
specimen.
1
Note that this is not real cohesion since sand has zero cohesion. The cohesion intercept is due to assumption that a
straight line can be used to approximate the shear strength envelop. The actual shear strength envelop is non-linear,
especially at low stress, due to particle interlocking and dilation.
29
Shear Strength Diagrams
200
180
y = 0.837x + 9.5
R² = 0.997
160
140 y = 0.652x + 5.3

R² = 0.998
120
y = 0.546x + 0.9
τ, kPa
100 R² = 0.997
80 Peak
60 Residual
Vibro-Residual
40
Linear (Peak)
20
Linear (Residual)
0 Linear (Vibro-
0 20 40 60 80 100 120 140 160 180 200 Residual)
σ, kPa
Figure 3.9 Static and dynamic shear strength envelops of sand
Depending on the objective of the laboratory testing program, soil samples can be
subjected to vibration at different states (pre-peak, at peak, post-peak and residual states). Also
depending on the technical characteristics of the actuator, the acceleration, amplitude and
frequency can be changed in order to simulate existing or expected field conditions.
30
(a)
Normalized displacement
(b)
(c)
(d)
Figure 3.10 Shear stress (a), vertical displacement (b), and horizontal (c) and vertical (d)
acceleration responses versus normalized horizontal displacement of sand under a normal stress
of 50 kPa.
31
3.5.2 Stress Controlled Testing Mode
To verify the workability of the stress controlled vibrating direct shear apparatus, five dry sand
samples were tested at a normal stress of 50 kPa. The samples were subjected to a shear stress
equal to 50%, 60%, 70%, 80% and 90% of the static shear strength () determined at a normal
stress of 50 kPa. After being loaded with the appropriate shear load, the samples were subjected
to vibration (frequency = 140 Hz; horizontal acceleration ≈ 0.4 g; vertical acceleration ≈ 0.15-0.2
g) and the shear deformations of the samples were recorded. The results are shown in Fig. 3.11.
As seen in Fig. 3.11, depending on the magnitude of the shear load, the samples experience
different shear deformations due to vibration. It is seen that the samples fail at 0.9 τ while the soil
would undergo shear deformations at different rates for the other tests.
32
Figure 3.11 Displacement and time response of stress controlled direct shear testing with
vibration at different levels of peak stresses with normal stress of 50 kPa
Typical plots of the shear resistance, vertical and horizontal (shear) deformations, as well
as horizontal and vertical accelerations versus time are shown in Fig. 3.12. There is some “noise”
in the horizontal (shear) deformation plots which is due to the technical characteristics of the
LVDT used to measure the shear displacement of the samples and the frequency of the data
logging relative to the output signal frequency of the sensor.
33
(a)
(b)
(c) (d)
(e) (f)
Figure 3.12 Shear stress (a), horizontal (c) and vertical displacement (e), horizontal (d) and
vertical (f) acceleration responses versus time of sand in stress controlled tests at 50 kPa normal
stress and shear stress equal to 50% of the static shear strength
34
3.6 Conclusions
The simple design of the vibrating direct shear apparatus discussed here makes it possible to
apply modifications to most of the existing direct shear (and simple shear) apparatuses that are
readily available in many geotechnical laboratories. Prototypes for a stress and strain controlled
vibrating direct shear apparatus have been built and tested. The new apparatus has been proven
in terms of its workability and efficiency in studying the vibrational responses of sand.
The new vibrating direct shear apparatus can be successfully used to determine the strength
and deformation characteristics of fine and granular soils (up to the size of coarse sand) under
different accelerations and frequencies of vibration. Depending on the testing mode (strain or
stress controlled) the characteristics of soils can be determined under drained conditions at
different shear rates, normal and shear stresses, as well as different stress states.
3.7 References
Arango, I. and Seed, H. B. (1974). “Seismic stability and deformation of clay slopes.” J.
Geotech. Engrg. Div., 100(2), pp. 139–156.
Barkan, D. D. (1962). “Dynamics of Basis and Foundations”. Translated from Russian by L.
Drashevska, New York, McGraw-Hill, 434 p.
Kutergin, V. N. (1989). The property change patterns of clayey soils under vibration. (in Russian)
Moscow: Nauka, 143 p.
35
Melosh, H. J. and Girdner K. K. (1995). Rheology of vibrated granular materials: Application to
long-runout landslides, EOS, 76.
Meschyan, S. R. (1978). “Short-term and long-term strength of clayey soils”. (in Russian)
Moscow, Nedra, 207 p.
Meschyan, S. R. (1992). “Rheological processes in clayey soils”. (in Russian) Yerevan,
“Hayastan”, 395 p.
National Instruments website: http://www.ni.com/labview/ (accessed in September, 2013)
Pokrovsky, G. I., Ehrlich, A. A., Laletin, N. V. and Lush, F. A. (1934). New Methods of
Investigation of the Compressibility and Internal Friction in Soils. Vestnik Voyenno-
Inzhenernoy Akademii RKKA, no. 6.
Shibata, T. and Yukitomo, H. (1970). Shear strength of sand under a vibrating load. Bulletin of
the Disaster Prevention Research Institute, 19(3): pp. 27-37.
Toyota, H., Towhata, I., Imamura, S. and Kudo, K. (2004). "Shaking Table Tests on Flow
Dynamics in Liquefied Slope", Soils and Foundations, Vol. 44, No. 5, pp. 67 – 84.
Youd, L. T. (1970). Densification and Shear of Sand during Vibration. ASCE, Journal of the Soil
Mechanics and Foundations Division, Vol. 96, No. 3, May/June 1970, pp. 863-880.
Wartman, J., Seed, R. B. and Bray, J. D. (2005). “Shaking table modeling of seismically induced
deformations in slopes.” J. Geotech. Geoenviron. Eng., 131(5), pp. 610–622.
36
4. Chapter 4: Effect of Vibration on the Critical State of Dry Granular Soils
4.1 Abstract
In order to investigate the effect of vibration on the critical state of dry granular soils, eighty
samples from four different granular materials are tested at four normal stresses under different
vibrational frequencies and accelerations. The experiments are performed on a modified
vibrating direct shear apparatus with vibrational loading in the horizontal (shear) direction.
Strength loss vs. vibration acceleration plotting at different normal stresses, as well as that for the
peak, residual and vibro-residual shear strengths, have been carried out for the four granular
materials at different intensities of vibration. It has been determined that an increase in the
intensity of vibration reduces the friction angle of the granular materials. The effect of particle
shape on the strength loss due to vibration has been observed, as well as the effect of vibration
on normal stress.
4.2 Introduction
Granular soils are frequently exposed to vibrations due to earthquakes, blasting, construction
operations, machinery and vehicle traffic. The strength behavior of a soil during vibration not
only depends on the vibration characteristics, such as acceleration, frequency and amplitude, but
also on the physical properties of the soil, such as moisture content, grain size distribution,
particle shape, dry density or void ratio, cohesion and internal friction angle, as well as the
density and mineralogy of the soil particles.
37
There has been significant progress in understanding the effects of vibration on the strength
and deformation properties of soils since the first half of the last century (Housner, 1954, 1959;
Barkan, 1962; Richart, 1970; Seed and Idriss, 1982, 1983; Idriss and Boulanger, 2008). A
number of different experiments with vibration application have been conducted on cohesive and
cohesionless soils that generated very valuable data which have led to some important
conclusions (Savchenko, 1958; Pyke et al., 1975; Robertson and Campanella, 1985, 1986;
Wartman et al., 2005; Meehan et al., 2008). All of these findings significantly help to improve
the design of different structures subjected to dynamic loads (Barkan, 1962; Seed, 1966; Richart,
1970; Seed and Idriss, 1982; Das, 2011; Sangroya and Choudhury, 2013).
Despite the knowledge gained on this issue in the last few decades, there is still a lack of
Therefore, there is a need to study the shear strength and deformation behavior of soils due to
vibrations. In particular it is specifically interesting to determine the effect of vibration on the
residual shear strength of soils, that is, the effect of vibration on the critical state, a unique state
independent of the initial density (void ratio) of the soil that can be reached upon sufficiently
large shear deformations.
If large displacements are anticipated, the design of a structure, such as an earth or tailing
dam, has to take into account the residual shear strength characteristics of the soils that underlie
or/and constitute the structure. In this case, if the soil is subjected to vibration, depending on the
magnitude of the vibrations, it may undergo some loss of strength, thus resulting in excessive
deformation and compromising the stability of the structure. Therefore, it is of paramount
importance to predict and consider the effects of vibrations on the residual strength of soils.
38
4.3 Previous Studies on Vibrational Loading on Soils
There are two important aspects that contribute to the unique properties of granular materials:
ordinary temperature plays no role on their mechanical behaviour, and the interactions between
grains are dissipative because of static friction and the inelasticity of collisions (Jaeger et al.,
1996). Pokrovsky (1934) was the first to experimentally investigate the influence of vibration on
the internal friction of sand. He showed that the coefficient of internal friction depends on the
kinetic energy of vibration. As the energy increased, the coefficient decreased, thus approaching
a value that was 25% to 30% smaller than that observed before vibration. These results were
later supported by experiments performed by Barkan (1962), who tested sand at a vibration
frequency of 140 sec-1 and amplitude of 0.5 - 0.15 mm. It was concluded that vibration has
considerable effects on the shearing resistance of soil and the internal friction is lower than the
static friction during vibration. Increase in acceleration will decrease the internal friction, which
asymptotically approaches to a limit value, depending on the properties of the soil. Analogous
experiments conducted on sand with a moisture content of 10% to 12 % showed that the moist
sand subjected to vibration has a smaller decrease in the internal friction than dry sand. This was
attributed to the capillary forces between the soil particles with a moisture content of 10% to
12%.
Additional confirmation of the above observations was provided by Savchenko (1958) who
revealed that at a constant vibrational frequency, the coefficient of the internal friction of sand
continuously decreases as the amplitude increases. The dependence of tan φ on the frequency of
vibration is more complicated, and as the test results showed, there exist frequency ranges that
correspond to small and large changes in tan φ of sand. Savchenko (1958) also tested two sets of
medium grained sand at different moisture contents (2 – 24 %), constant vibration amplitude of
39
0.35 mm and two vibration frequencies of 144 sec-1 and 250 sec-1, respectively. For both sets of
samples, the largest decrease in tan φ was observed at approximately a moisture content of 13%
(Barkan, 1962). Barkan (1962) concluded that the principal vibration parameter which
determines the effect of vibrations and shocks on the compaction of soils is the acceleration, or
rather, the inertial force, which acts on the soil particles during vibration. Metcalfe et al. (2002)
and Huan (2008) also confirmed that dimensionless acceleration 2/g is a key vibration
parameter, where A is the vibration amplitude,  is the frequency and g is the gravitational
acceleration.
To study the influence of particle size on the effect of vibration, Savchenko (1958)
conducted experiments on four different sizes of sand at vibrational frequencies of 144 sec-1 and
250 sec-1 and a constant vibration amplitude of 0.35 mm. The results revealed that the effect of
vibration on internal friction in sand is directly proportional to the diameter of the sand grains
(Barkan, 1962). On the other hand, Maslov (1959) performed a series of experiments on sands
and concluded that vibration does not cause any changes in the coefficient of the internal friction
of sand, and that the changes in its shear resistance are due to a decrease in the normal stress
caused by vibration. Shibata and Yukitomo (1969) carried out triaxial vibrating tests under
drained conditions. The results revealed that the dynamic strength increases with increases in the
density of sand, and that the influence of vibration frequency on the strength of sand is rather
remarkable; higher frequency indicates lower shear strength.
An interesting concept of “dynamic fluidization” was proposed by Richards et al. (1990).
They took into consideration the effect of earthquake accelerations on dry granular soils. They
hypothesized that the imposed accelerations at some critical level change the state of the soil,
which causes general plastification, such that the soil becomes, in a sense, an anisotropic fluid.
40
They assumed that the main trigger of the fluidization is the inertial forces that act between the
particles of a granular soil. They showed that fluidization mainly depends on horizontal and not
vertical accelerations. Another distinguishing feature of fluidization is that when it occurs, flow
takes place, if at all, in finite increments rather than continuously, with increments that
correspond to the acceleration pulses of an earthquake above a critical value (Richards et al.,
1990).
The authors distinguished three stages of dynamic fluidization: initial, intermediate and
general. Initial fluidization is recognized as a threshold, above which, significant loss of soil
shearing resistance may occur. For a dry sand in its neutral condition, the Ko initial fluidization
takes place at <0.3 and at even lower values for soils with a smaller friction angle. For saturated
loose granular soils, it is postulated that the initial fluidization will initiate liquefaction because
of collapse of the soil structure under shear flow. At the intermediate stage of dynamic
fluidization, the soil continues to lose its shearing resistance and provides support of the external
loads from internal shearing generated by the inertial forces. When the soil reaches the general
stage of dynamic fluidization, its shear strength is mobilized based on a broad range of
orientations, and the soil within these orientations behaves like a viscous fluid (Richards, 1970).
Youd (1968) performed laboratory experiments by mounting a direct shear apparatus onto
a shaking table to show the effects of vibration on the shear strength and void ratio of dry
granular materials. It was concluded that each vibratory equilibrium void ratio2 is also the critical
void ratio when the sample is sheared under the same vibration. The critical void ratio and
coefficient of internal friction were both considerably reduced during vibration.
2
Vibratory equilibrium void ratio is defined as the ultimate minimum void ratio for a sample densified at a
particular vibration.
41
Of special interest are the theories of “mechanical fluidization” (Davies, 1982) and
“acoustic fluidization” (Melosh, 1979) which attempt to explain the large runout distance of big
rock avalanches. The essence of mechanical fluidization is the concept that high energy input
into a granular mass causes high impulsive contact pressure between individual grains such that
they become statistically separated and the mass dilates. The internal shearing resistance is
thereby reduced as shown by Bagnold (1954) and Bjerrum et al. (1961), and the mass may flow
under gravity when dilated. The high relative velocity between the base of the high-speed debris
flow and the bedding material of the flow channel causes high speed shearing with high energy,
which results in dilation as shown by Bagnold (1954) for a granular material subjected to
unidirectional shearing (Davies, 1982). On the other hand, Hungr (1981) conducted high
velocity ring shear tests and did not observe any effects of “mechanical fluidization”. Several
types of materials were tested, including two sizes of relatively coarse sand, mixtures of sand and
rock flour, polystyrene beads and sand in water, under different velocities and normal stresses.
All of the materials showed straight linear residual strength envelopes with zero cohesion and
unique angles of residual friction, which were minimally influenced by the shearing speed
(Hungr, 1981).
The acoustic fluidization theory proposed by Melosh (1979) is another attempt to explain
the large runout distance of sturzstroms observed on earth, as well as on the moon and Mars.
Based on this theory, a thin layer of material can be fluidized by strong enough sound waves at
the bottom interface of moving debris, if its volume is large enough to retain enough acoustic
energy to maintain fluidization. An important aspect of the flow process is that sound (acoustic
energy) is created as debris moves, and that the sound must have short wave-lengths compared to
the dimensions of the fluidized rock debris. Melosh (1996) also suggested that the overburden
42
pressure in some of the faults can be relieved by acoustic fluidization and allow pressure to slide
at a low average stress (Melosh, 1979). Later, Sornette and Sornette (2000) found inconsistency
in Melosh’s theory of acoustic fluidization in explaining fault motion which nulls the results, and
although they provided alternatives, they also indicated that the significance of acoustic
fluidization remains questionable.
As evident from the studies summarized above, there is currently no comprehensive model
that can fully explain the mechanism of the effect of vibrations on the shear strength
characteristics of soils. Moreover, some of the findings seem to contradict each other. Therefore,
the shear strength characteristics of soils under the influence of vibration remains open for
further studies.
4.4 Testing Equipment and Procedures
In order to investigate the effects of vibration on the residual shear strength of granular material,
a direct shear apparatus was modified into a vibrating direct shear apparatus as shown in Figs.
4.1 and 4.2. The modifications include installation of an electromagnetic actuator (11) between
the proving ring (3) and the shear box (2), as well as two load cells (10 and 12) to measure the
shear forces at the top and bottom halves of the shear box (2). An extension (13) of the main
body (5) is installed to accommodate the actuator (11). The actuator (11) consists of two
electromagnets, and the frequency and force can be changed by using a control panel. The
vibration generated by the actuator is in the horizontal shearing direction.
The measuring equipment consisted of two linear variable differential transformers
(LVDTs) that measured the vertical and horizontal displacements, two load cells (10 and 12),
43
and two uniaxial accelerometers that measured the vertical and horizontal vibration accelerations
on the soil samples. One of the accelerometers was placed on top of the loading plate (measured
vertical vibration accelerations), and the other accelerometer was attached to the top half of the
shear box in the direction of the shear (measured horizontal vibration accelerations). The output
signals were acquired by using the NI CompactDAQ System, which was in turn, connected to a
PC that logged the data with NI LabVIEW software.
(a) 1 (b)
2 3 10
4 11
12
5
13
6
8
7
9 7
vibrating direct shear apparatus.
1 – Control panel of the direct shear apparatus; 2 – Shear box with a soil sample; 3 – Proving
ring; 4 – Screw for zero setting of shear load before starting a test; 5 – Main body of the direct
shear apparatus; 6 – Frame that transfers normal load to the soil sample placed in the shear box
(2); 7 – Legs of the direct shear apparatus that supports the main body (5); 8 – lever that provides
a normal load to the soil sample; 9 – Weights that define a normal load on the soil sample; 10 –
Load cell; 11 –Actuator; 12 – Load cell; and 13 – Extension of the main body (5) of the direct
shear apparatus.
44
The testing procedures were carried out in accordance with ASTM D3080/D3080M
(Standard Test Method for Direct Shear Test of Soils Under Consolidated Drained Conditions),
with additional vibration applied for a short period of time at the pre-peak and residual strength
states. Depending on the objective of the test, vibrations were applied while the sample was
being sheared, as well as when shearing was terminated. The frequency and force of the
vibrations were adjusted to the required magnitude and kept constant for the set of soil samples
tested.
Figure 4.2 Photograph of the modified direct shear apparatus
45
4.5 Tested Materials
To evaluate the effect of vibration on the residual strength of granular material, four different dry
granular materials (a total of 80 samples) were tested. The materials were 0.1 mm and 0.55 mm
glass beads, as well as fine and coarse sands. These materials were selected to investigate the
effect of particle shape and size on the vibro-residual strength characteristics at different
vibrational accelerations. Particularly, different glass bead sizes (0.1 and 0.55 mm) were chosen
to determine how the different inertial forces (due to different sized particles), generated among
the particles through vibration, contributed to the vibro-residual strength. On the other hand,
coarse sand (well-rounded) and fine sand (angular) samples were chosen in addition to the glass
beads in order to evaluate the effect of particle roundness on the vibro-residual strength of the
materials, particularly to observe the influence of particle friction during the induced vibration.
The samples (6 cm x 6 cm x 3.2 cm (W x L x H) in size) were tested under normal stresses
of 23, 50, 118 and 200 kPa in strain-controlled mode at a shear rate of 0.61 mm/min. A vibration
frequency of 140 Hz was used for all of the samples. From each type of granular material, five
sets of samples (four samples in each set) were tested at five different vibration intensities. The
maximum impact forces of the actuator electromagnets were 1.61, 3.22, 3.71, 5.18 and 7.14 N.
Note that these impact forces are the designed forces provided by actuator. The actual impact
forces may vary during the testing. They were measured by the accelerometers and load cell.
All of the data were recorded at a rate of 1 kHz.
The vibration frequency value used in the text (140 Hz) was set by the control panel of the
actuator, and double-checked by the observation of the recorded horizontal vibration acceleration
sine wave, where it corresponded to the acceleration peaks. It should be mentioned that there is
also vibration noise present in the recorded horizontal acceleration plots that has different
46
frequencies. The vibration noise is the secondary vibration frequency that is combined with the
recorded primary frequency at low frequency level. Although these secondary vibration
frequencies have not been analyzed and only the frequency corresponding to the peak
acceleration values has been used for simplicity, it is realised that they shouldhave some effects
on the measured strength and deformation characteristics of the tested materials.. To filter this
noise effect a Fast Fourier transform analysis can be done.
The physical characteristics of the tested materials are provided in Table 4.1. The particle
size distributions of the four materials are given in Fig. 4.3. Representative samples of the
materials were examined under a microscope to determine the shape of the particles (see Figs.
4.4a, 4.4b, 4.4c, and 4.4d). It is seen in Fig. 4.4a that the 0.1 mm glass beads are not round with
irregular shape grains and some are smaller than 0.1 mm. Therefore the “0.1 mm glass beads”
should be treated like a mixture of fine sand sized 0.1 mm and smaller. The 0.55 mm glass beads
are basically the ideal spherical shape, see Fig. 4.4b. From Figs. 4.4c and 4.4d, it can be seen that
the fine sand particles have an angular shape and the coarse sand is well rounded. The shear
strength diagrams and the residual friction angles of the four materials are shown in Fig. 4.5.
Table 4.1 Physical characteristics of the glass beads and sands
Material ρs ρ e n
Glass Beads 0.1 mm 2.65 1.49 0.779 0.438
Glass Beads 0.5 mm 2.65 1.566 0.692 0.409
Fine Sand 2.65 1.46 0.815 0.449
Coarse Sand 2.65 1.755 0.510 0.338
47
Figure 4.3 Particle size distribution of the tested materials
48
(a) (b)
(c) (d)
Figure 4.4 Pictures of the glass beads and sand under a microscope at magnification of 50x. (a)
0.1 mm glass beads; (b) 0.55 mm glass beads; (c) fine sand; and (d) coarse sand
49
(a)
(b)
(c)
(Continued on the next page)
50
(d)
φr =27.5o
Figure 4.5 Shear strength diagrams of 0.1 mm glass beads (a), 0.55 mm glass beads (b), fine sand
(c) and coarse sand (d).
4.6 Test Results under Vibration
A summary of the test results for the 0.1 mm and 0.55 mm glass beads, and the fine and coarse
sands are given in Tables 4.2, 4.3, 4.4 and 4.5, respectively. Fig. 4.6 shows a typical plot of shear
stress, vertical displacement, horizontal and vertical accelerations versus the normalized
displacement of fine sand. The shearing resistance of the material was measured by using a load
cell mounted at the top half of the shear box (see Fig. 4.1b (12)). Vertical displacements were
measured from an LVDT attached to the loading plate of the shear box. Horizontal and vertical
accelerations were obtained from an accelerometer attached to the top half of the shear box in the
direction of shearing and a second accelerometer that measured the vertical accelerations of the
loading plate, respectively. All of the data except for vertical displacement were collected at a
rate of 1 kHz. Due to the technical characteristics of the LVDT, vertical displacements were
measured at a rate of 2 Hz filtered from the original rate of 1 kHz.
51
As seen in Fig. 4.6, vibration is applied when the soil reaches the critical state. The shear
resistance plot shows that during vibration, there is an immediate strength loss, Δτ, for the
residual to vibro-residual states. It is believed that the strength loss occurs in a very short time,
probably in a few milliseconds. The vibro-residual strength remains practically constant during
vibration. When vibration is terminated, the shear strength of the material gradually increases
and reaches the residual quasi-static strength value. Note that the normalized displacement in
Figs. 4.6 and 4.7 is defined as d/L, where d is the horizontal displacement and L is the initial
length of the specimen.
All of the samples have experienced contraction and decrease in volume, due to vibration
as shown in the vertical displacement plot in Fig. 4.6. It is seen that contraction takes place
during vibration and the volume change is permanent or plastic. The initial volume before
vibration cannot be restored when vibration stops.
An example of the effect of an increase in horizontal vibration acceleration on the residual
shear strength of fine sand is provided in Fig. 4.7 (a, b and c), which shows the plots of fine sand
samples tested at vibration forces of 3.71, 5.18 and 7.14 N of the electromagnets and horizontal
accelerations of 0.27, 0.37 and 0.47 g, respectively. The example demonstrates that at a constant
normal stress (118 kPa) and vibration frequency, an increase in vibration acceleration by 0.1 g
results in greater strength loss.
The strength loss process can be briefly outlined in the following way: when vibration is
applied to a sample of granular material sheared at its residual strength state, momentary loss of
contacts between the grains takes place at the shear zone. This disturbs the stress chains in the
sample, thus causing reduction in shear resistance. It should also be stated that since the
amplitude of vibration is much smaller than the diameter of grains comprising the most of the
52
tested materials, the movement of particles in the shear zone can be combination of rotational
and translational motions.
The peak, residual and vibro-residual shear strength envelops of the 0.1 and 0.55 mm glass
beads, as well as fine and coarse sand for five different vibration forces are shown in Figs. 4.8,
4.9, 4.10 and 4.11, respectively. The peak, residual and vibro-residual friction angles of the four
materials are given in Tables 4.6, 4.7, 4.8 and 4.9. As seen in Figs. 4.8 to 4.11 and acceleration
values in Tables 4.2 to 4.5, increase in the vibration force results in an increase of the vertical
and horizontal accelerations and amplitudes (not provided in Tables 4.2 to 4.5), which in turn,
results in a greater loss in the residual strength at a given normal stress. From the tables and
figures, it is seen that vibration reduces the residual friction angle.
53
Figure 4.6 Shear stress, vertical displacement, horizontal and vertical accelerations versus
normalized displacement of fine sand subjected to a normal stress of σ=50 kPa, vibration
frequency of 140 Hz and vibration force of 5.18 N.
54
(a)
(b)
(c)
Figure 4.7 Shear stress plots of three fine sand samples tested at normal stress of 118 kPa and
different vibration accelerations
55
Table 4.2 Test results of 0.1 mm glass beads
Normal Peak Residual Vib.-Res. Vertical Horizontal Strength

Force
Stress Strength Strength Strength Acceleration Acceleration Loss
σ τp τf τfd Δτ
N g g
kPa kPa kPa kPa kPa
23 21 19.5 18 0.03 0.1 1.5
50 38.5 35 33 0.03 0.11 2
1.61
118 82.5 71 68 0.04 0.12 3
200 137.5 113 109.5 0.04 0.12 3.5
23 21 16.5 15 0.09 0.22 1.5
50 38.5 29 26.5 0.08 0.23 2.5
3.22
118 83.5 67 63 0.07 0.22 4
200 135.5 114 108 0.065 0.2 6
23 21 18 15.5 0.075 0.25 2.5
50 38 31.5 28 0.09 0.28 3.5
3.71
118 85.5 70 65 0.08 0.24 5
200 132 111.5 105 0.07 0.23 6.5
23 21.5 17 14 0.11 0.42 3
50 38 32.5 26 0.11 0.38 6.5
5.18
118 84 73 63 0.105 0.325 10
200 136 121 108 0.09 0.3 13
23 22 17 13 0.14 0.5 4
50 39.5 36 26.5 0.13 0.43 9.5
7.14
118 83.5 73 56 0.15 0.43 17
200 135 115 94 0.13 0.38 21
56
Table 4.3 Test results of 0.55 mm glass beads

Force
N g g
kPa kPa kPa kPa kPa
23 17.5 13.5 12.5 0.04 0.11 1
50 32 25.5 24 0.035 0.11 1.5
1.61
118 69 55.5 53.5 0.035 0.08 2
200 111 82 79 0.035 0.09 3
23 17 14 12.5 0.07 0.175 1.5
50 30.5 24.5 22.5 0.07 0.175 2
3.22
118 65 51.5 47.5 0.07 0.175 4
200 101 84 78 0.07 0.17 6
23 18 14 11.5 0.1 0.28 2.5
50 32 25.5 22.5 0.085 0.23 3
3.71
118 64 54 48.5 0.09 0.23 5.5
200 103 82 73 0.085 0.23 9
23 17 14 10.5 0.105 0.325 3.5
50 31 25.5 21 0.105 0.325 4.5
5.18
118 62.5 51.5 43 0.105 0.325 8.5
200 106 81 68 0.09 0.325 13
23 18 13.5 9 0.16 0.46 4.5
50 32.5 25 19 0.16 0.46 6
7.14
118 61.5 53 42 0.16 0.42 11
200 106 83 66.5 0.14 0.43 16.5
57
Table 4.4 Test results of fine sand

Force
N g g
kPa kPa kPa kPa kPa
23 27 21 19.5 0.045 0.15 1.5
50 53.5 40 38 0.045 0.12 2
1.61
118 110 87.5 85.5 0.05 0.12 2
200 178.5 139 137 0.045 0.11 2
23 26.5 20.5 17.5 0.1 0.26 3
50 49 38.5 35 0.1 0.27 3.5
3.22
118 107.5 83 77.5 0.09 0.27 5.5
200 170.5 137.5 131 0.07 0.23 6.5
23 30 22 18.5 0.09 0.25 3.5
50 51.5 37 33.5 0.1 0.27 3.5
3.71
118 107.5 81 75 0.09 0.27 6
200 169 143 134.5 0.08 0.25 8.5
23 29.5 22 17.5 0.1 0.37 4.5
50 51 39.5 33.5 0.115 0.38 6
5.18
118 108.5 82.5 71.5 0.105 0.37 11
200 171 134 121 0.1 0.32 13
23 30 21 14.5 0.17 0.5 6.5
50 50.5 37.5 27.5 0.155 0.49 10
7.14
118 108 83 65 0.145 0.47 18
200 177 138 116 0.145 0.45 22
58
Table 4.5 Test results of coarse sand

Force
N g g
kPa kPa kPa kPa kPa
23 25 18.5 17 0.045 0.12 1.5
50 42 34.5 33 0.04 0.12 1.5
1.61
118 88.5 67.5 66 0.045 0.1 1.5
200 145 108.5 106.5 0.04 0.1 2
23 24.5 18.5 16.5 0.075 0.2 2
50 43 34 31.5 0.065 0.2 2.5
3.22
118 89 67.5 64.5 0.07 0.19 3
200 145.5 111 107 0.07 0.18 4
23 24 18.5 16 0.085 0.23 2.5
50 44.5 33.5 30.5 0.075 0.22 3
3.71
118 93.5 70 66.5 0.075 0.21 3.5
200 148.5 116 111 0.075 0.2 5
23 24 18 14.5 0.12 0.35 3.5
50 43 34.5 29.5 0.09 0.33 5
5.18
118 91.5 71 65.5 0.1 0.28 5.5
200 148 110 102 0.08 0.28 8
23 23.5 18 12 0.2 0.59 6
50 44.5 32 25.5 0.145 0.4 6.5
7.14
118 91 68.5 59 0.135 0.39 9.5
200 147 109 95 0.1 0.39 14
59
(a) Shear Strength Diagram (1.61 N) (b) Shear Strength Diagram (3.22 N)
140 y = 0.6578x + 5.5778 140 y = 0.6479x + 6.2892
120 y = 0.5265x + 8.1579 120 y = 0.5549x + 2.3818
100 y = 0.5153x + 6.7589 100 y = 0.5302x + 1.2978
τ, kPa
τ, kPa
80 80
60 60 Peak
Peak
40 40
20 Residual 20 Residual
0 0
0 100 200 Vibro-Residual 0 100 200 Vibro-Residual
σ, kPa σ, kPa
(c) Shear Strength Diagram (3.71 N) (d) Shear Strength Diagram (5.18 N)
140 y = 0.6323x + 7.3208 140 y = 0.6505x + 6.2901
120 y = 0.5322x + 5.7317 120 y = 0.5887x + 3.3301
100 100 y = 0.5355x + 0.4082
y = 0.5102x + 3.5059
τ, kPa
80
τ, kPa
80
60 Peak 60 Peak
40 40
0 0
σ, kPa σ, kPa
Shear Strength Diagram (7.14 N)

(e)
140
y = 0.6387x + 7.5702
120
y = 0.547x + 6.7785
100
y = 0.4542x + 2.9772
τ, kPa
80
60 Peak
40
20 Residual
0
0 100 200 Vibro-Residual
σ, kPa
Figure 4.8 Shear strength envelops of 0.1 mm glass beads tested at: (a) vibration force of 1.61 N;
(b) vibration force of 3.22 N; (c) vibration force of 3.71 N; (d) vibration force of 5.18 N and (e)
vibration force of 7.14 N.
60
120 120
y = 0.5291x + 5.6587 y = 0.476x + 6.8482
100 100
y = 0.3887x + 6.1285 y = 0.3959x + 4.8049
80 80
y = 0.378x + 5.3017 y = 0.3699x + 3.9705
τ, kPa
τ, kPa
60 60
40 Peak 40 Peak
20 Residual 20
Residual
0 0
0 100 200 Vibro-Residual 0 100 200 Vibro-
σ, kPa σ, kPa Residual
120 120
y = 0.478x + 7.5255 y = 0.4993x + 5.3191
100 100
y = 0.3851x + 6.232 y = 0.3769x + 6.1548
80 80
y = 0.3476x + 4.8926
τ, kPa
τ, kPa
y = 0.3222x + 4.1277
60 60
Peak Peak
40 40
20 Residual 20
Residual
0 0
0 100 200 Vibro- 0 100 200 Vibro-
Residual Residual
σ, kPa σ, kPa
(e) Shear Strength Diagram (7.14 N)

120
y = 0.4904x + 6.5636
100
y = 0.3927x + 5.2345
80
τ, kPa
y = 0.324x + 2.4503
60
40 Peak
20 Residual
0
σ, kPa
Figure 4.9 Shear strength diagrams of 0.55 mm glass beads tested at: (a) vibration force of 1.61
N; (b) vibration force of 3.22 N; (c) vibration force of 3.71 N; (d) vibration force of 5.18 N and
(e) vibration force of 7.14 N.
61
200 y = 0.8491x + 9.2491 200
180 180 y = 0.8162x + 8.5904
160 y = 0.6675x + 6.6229 160 y = 0.6603x + 5.3299
140 y = 0.6655x + 4.9429 140
120 120 y = 0.6398x + 2.712
τ, kPa
τ, kPa
100 100
80 Peak 80 Peak
60 60
40 40
Residual Residual
20 20
0 0
σ, kPa σ, kPa
200 200
180 y = 0.7877x + 12.504 180 y = 0.803x + 11.511
160 160
140 y = 0.686x + 3.6931 140 y = 0.6321x + 7.7095
120 y = 0.656x + 1.25 120 y = 0.5825x + 3.9328
τ, kPa
τ, kPa
100 100
80 Peak 80 Peak
60 60
40 40
0 0
σ, kPa σ, kPa

200
180 y = 0.8344x + 9.8084
160
140 y = 0.6636x + 5.009
120
τ, kPa
y = 0.5755x - 0.5028
100
80 Peak
60
40
20 Residual
0
σ, kPa
Figure 4.10 Shear strength diagrams of fine sand tested at: (a) vibration force of 1.61 N; (b)
vibration force of 3.22 N; (c) vibration force of 3.71 N; (d) vibration force of 5.18 N and (e)
62
160 160
140 y = 0.6803x + 8.6244 140 y = 0.683x + 8.734
120 y = 0.5034x + 8.0391 120 y = 0.5184x + 7.0806
100 y = 0.5007x + 6.6809 100 y = 0.5076x + 5.2538
τ, kPa
τ, kPa
80 80
60 Peak 60
Peak
40 40
0 0
σ, kPa σ, kPa
160 160
y = 0.7024x + 8.9615 y = 0.7014x + 8.0659
140 140
120 y = 0.5496x + 5.7741 120 y = 0.5174x + 7.7989
100 y = 0.5362x + 3.5891 100 y = 0.4945x + 4.5375
τ, kPa
τ, kPa
80 80
60 Peak 60 Peak
40 40
0 0
0 100 200 Vibro-Residual 0 100 200 Vibro-
σ, kPa σ, kPa Residual

160
y = 0.6935x + 8.706
140
120 y = 0.5156x + 6.4737
100 y = 0.4694x + 1.9874
τ, kPa
80
60 Peak
40
20 Residual
0
σ, kPa
Figure 4.11 Shear strength diagrams of coarse sand tested at: (a) vibration force of 1.61 N; (b)
vibration force of 3.22 N; (c) vibration force of 3.71 N; (d) vibration force of 5.18 N and (e)
63
Table 4.6 Friction angles of 0.1 mm glass beads.
ϕ (o)
Force (N)
Peak Residual Vibro-residual
1.61 33.3 27.8 27.3
3.22 32.9 29 27.9
3.71 32.3 28 27
5.18 33 30.5 28.2
7.14 32.6 28.7 24.4
Table 4.7 Friction angles of 0.55 mm glass beads
ϕ (o)
Force (N)
1.61 27.9 21.2 20.7
3.22 25.5 21.6 20.3
3.71 25.5 21.1 19.2
5.18 26.5 20.7 17.9
7.14 26.1 21.4 17.9
Table 4.8 Friction angles of fine sand
ϕ (o)
Force (N)
1.61 40.3 33.7 33.6
3.22 39.2 33.4 32.6
3.71 38.2 34.5 33.3
5.18 38.8 32.3 30.2
7.14 39.8 33.6 29.9
64
Table 4.9 Friction angles of coarse sand.
ϕ (o)
Force (N)
1.61 34.2 26.7 26.6
3.22 34.3 27.4 26.9
3.71 35.1 28.8 28.2
5.18 35 27.4 26.3
7.14 34.7 27.3 25.1
4.6.1 The Effect of Strength Loss Due to Changes in Normal Stress Caused by Vibration
Changes in the shearing resistance can be due to changes in the normal stress during vibration,
per Maslov (1959). In this study, vertical acceleration is measured to calculate the normal stress
that acts on a sample during vibration. From Newton’s second law of motion (F = m a), when the
vertical acceleration of the loading plate and the mass that acts on the loading plate are known,
the force due to vibration can be calculated. By using the area (A) of the sample, the reduction in
normal stress can be calculated from Δσ = F/A. If the Δσ value is known and the residual shear
strength diagram (equation) of the material is used, the strength loss Δτ` due to reduced normal
stress Δσ is obtained.
The measured total residual strength loss Δτ and the calculated residual strength loss Δτ`
due to normal stress reduction caused by vibration are provided in Figs. 4.12 to 4.15, where Δτ is
equal to the total strength loss, Δτ` is the strength loss due to reduction in normal stress σ, for
samples tested at normal stresses of σ = 23, 50, 118 and 200 kPa, denoted by 1, 2, 3 and 4
respectively in the figures.
65
22
20 Δτ at 1.61 N
18 Δτ' at 1.61 N
16 Δτ at 3.22 N
14 Δτ' at 3.22 N
Strength 12
Loss Δτ at 3.71 N
(kPa) 10 Δτ' at 3.71 N
8
Δτ at 5.18 N
6
4 Δτ' at 5.18 N
2 Δτ at 7.14 N
0 Δτ' at 7.14 N
1 2 3 4
beads.
18
Δτ at 1.61 N
16
Δτ' at 1.61 N
14
Δτ at 3.22 N
12
Δτ' at 3.22 N
Strength 10
Loss Δτ at 3.71 N
(kPa) 8
Δτ' at 3.71 N
6
Δτ at 5.18 N
4
Δτ' at 5.18 N
2
Δτ at 7.14 N
0
Δτ' at 7.14 N
1 2 3 4
beads.
66
22
20 Δτ at 1.61 N
18 Δτ' at 1.61 N
16 Δτ at 3.22 N
14 Δτ' at 3.22 N
Strength 12
loss Δτ at 3.71 N
(kPa) 10 Δτ' at 3.71 N
8
Δτ at 5.18 N
6
4 Δτ' at 5.18 N
2 Δτ at 7.14 N
0 Δτ' at 7.14 N
1 2 3 4
Figure 4.14 Total strength loss and strength loss due to reduced normal stress for fine sand.
14
Δτ at 1.61 N
12
Δτ' at 1.61 N
10 Δτ at 3.22 N
Δτ' at 3.22 N
Strength 8
loss Δτ at 3.71 N
(kPa) 6 Δτ' at 3.71 N
4 Δτ at 5.18 N
Δτ' at 5.18 N
2
Δτ at 7.14 N
0 Δτ' at 7.14 N
1 2 3 4
Figure 4.15 Total strength loss and strength loss due to reduced normal stress for coarse sand.
67
As seen in Figs. 4.12 to 4.15, the residual strength loss Δτ` due to a reduction in normal
stress Δσ caused by vibration for all of the materials is very small compared to the total strength
loss Δτ. Therefore, it can be concluded that although there is some reduction in normal stress due
to vibration, the contribution to the loss in residual strength is minimal and the strength loss is
mainly due to the material fluidization.
4.6.2 The Effects of Acceleration on Residual Strength Loss
Based on the data from Tables 4.2 to 4.5, the total residual strength loss is plotted against
horizontal acceleration for the 0.1 mm and 0.55 mm glass beads, and fine and coarse sands, and
the results are shown in Figs. 4.16 to 4.19. As seen in Figs. 4.16, 4.18 and 4.19, the effect of
horizontal acceleration on the strength loss for the 0.1 mm glass beads, and fine and coarse sands
is nonlinear under higher normal stresses of 50, 118 and 200 kPa. The relationship is practically
linear at a low normal stress of σ = 23 kPa. Also, the plots become more nonlinear with an
increase in the normal stress from 50 to 200 kPa. On the other hand, Fig. 4.17 shows that the
relationship between strength loss and horizontal acceleration for the 0.55 mm glass beads is
practically linear under all normal stress values (23, 50, 118 and 200 kPa). This can probably be
explained by the absence of irregular shape particles in this material (see Fig. 4.4b). It was
discussed earlier that the 0.1 mm glass beads are not round, but consisted of angular and irregular
particles. This can be considered parallel to a mixture of fine angular sand and glass beads with a
size up to 0.1 mm in diameter.
As mentioned above, the irregular shape of the particles that comprised the tested 0.1 mm
glass beads, and fine and coarse sands, results in nonlinearity of horizontal acceleration vs.
residual strength loss at higher normal stresses (50, 118 and 200 kPa). It is apparent from Figs.
68
4.16, 4.18 and 4.19 that unlike the plots of the samples tested at a normal stress of 23 kPa, the
samples tested at higher normal stress values demonstrate less strength loss at low horizontal
accelerations than they would if their plots were linear. This phenomenon can be explained by
the additional friction among particles due to their irregular shape. The irregular shape of the
particles generates greater shear resistance during their rotational and translational motions
induced by relatively low vibration intensities.
On the other hand, the 0.55 mm glass beads practically have a round shape and unlike the
other materials, they do not generate additional friction due to the irregular shape of the grains.
Here, the round shape of the particles results in a linear pattern of the horizontal acceleration vs.
residual strength loss at all normal stress values (23, 50, 118 and 200 kPa) as seen in Fig. 4.17.
It should also be noted that at higher horizontal acceleration, the strength loss of the 0.1
mm glass beads, and the fine and coarse sand samples notably increases. This implies that there
is a certain value of the horizontal acceleration, above which, the effect of the irregular shape of
the particles becomes insignificant.
Figs. 4.16 to 4.19 show estimates of strength loss due to vibration for different normal
stresses and accelerations.
69
Figure 4.16 Total strength loss vs. horizontal acceleration of 0.1 mm glass beads
22
20 Glass Beads
18 0.55 mm φr=21.2°
16
14
12
Δτ
23 kPa
kPa 10
50 kPa
8
118 kPa
6 200 kPa
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Horizontal g
Figure 4.17 Total strength loss vs. horizontal acceleration of 0.55 mm glass beads
70
22
20 Fine Sand
φr=33.5°
18
16
14
12 23 kPa
Δτ
kPa 10 50 kPa
118 kPa
8
200 kPa
6
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Horizontal g
Figure 4.18 Total strength loss vs. horizontal acceleration of fine sand
22
20 Coarse Sand
φr=27.5°
18
16
14
12 23 kPa
Δτ
kPa 10 50 kPa
8 118 kPa
6 200 kPa
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Horizontal g
Figure 4.19 Total strength loss vs. horizontal acceleration of coarse sand
71
4.7 Conclusions
A new experimental technique and apparatus have been introduced in this paper to study the
effect of vibration on the critical state strength of granular material. If the material is subjected
to vibration after reaching the critical state, there is a decrease in volume and shearing resistance
of the material. A decrease in volume implies compaction and a denser state, which usually lead
to higher shear strength under static conditions. However, the shearing resistance of the material
is reduced during vibration. Based on the test results of the 0.1 and 0.55 mm glass beads and the
fine and course sands, the following conclusion can be made.
1. Upon application of a sufficiently strong vibration at the critical strength state of a granular
material, an immediate strength loss takes place for the residual to vibro-residual strengths,
after which the vibro-residual strength remains practically constant during vibration.
2. Vibration does not have any permanent post-vibrational effects on the residual strength of
granular material, provided that there is sufficient shear displacement of the material after
termination of the vibration.
3. Initially dense or loose granular material compresses when vibration is applied at its critical
strength state.
4. At a constant vibration frequency, an increase in vibration force results in an increase in the
vertical and horizontal vibration accelerations and amplitudes, which in turn, results in a
greater residual strength loss at the given normal stress.
5. Sufficiently strong vibration reduces the residual friction angle; that is, the magnitude of the
residual strength loss increases with an increase in the normal stress.
6. The residual strength loss Δτ` due to the normal stress reduction caused by vibration for all
of the granular materials tested is small compared to the total loss in residual strength Δτ.
72
Therefore, it can be concluded that although normal stress reduction caused by vibration (in
the given range of accelerations) contributes to residual strength loss, the strength loss is
mainly due to material fluidization.
7. For a given test condition, the relationship between the residual strength loss and horizontal
acceleration is nonlinear for irregularly shaped particles. The nonlinearity increases with an
increase in confining (normal) stress.
8. For a given test condition, the relationship between the residual strength loss and horizontal
acceleration is linear for glass beads (0.55 mm) with spherical shape particles under all
normal stresses tested.
4.8 References
Bagnold, R. A. (1954). “Experiments on the Gravity-Free Dispersion of Large Solid Spheres in a
Newtonian Fluid Under Shear”. Proc. Royal Soc. London 225. pp. 49–63.
Bjerrum, L., Kringstad, S. and Kummeneje, O. (1961). “The Shear Strength of a Fine Sand”.
Proc. 5th Int. Conf. Soil Mech. and Found. Eng. pp 29–37.
Bougie, J. L. (2004) Continuum Simulations of Fluidized Granular Materials. The University of
Texas, Austin.
Das, B. M. (2011). “Principles of Soil Dynamics”, Second Edition. Cengage Learning, 563
73
Davies, T. R. H. (1982). “Spreading of Rock Avalanche Debris by Mechanical Fluidization”.
Rock Mechanics 15; Springer-Verlag, pp. 9-24.
Housner, G. W. (1954). “Geotechnical problems of destructive earthquakes”. Geotechnique,
December.
Housner, G. W. (1959). “Behavior of structures during earthquakes”. Journal of the Engineering
Mechanics Division, ASCE, 85, No. EM-4.
Huan, C. (2008) NMR Experiments on Vibrofluidized and Gas Fluidized Granular Systems. PhD
Thesis, University of Massachusetts Amherst.
Hungr, O. (1981). “Dynamics of Rock Avalanches and Other Types of Slope Movements”. PhD
Thesis, University of Alberta, 506 p.
Idriss, I. M. and Boulanger, R. W. (2008). “Soil liquefaction during earthquakes”. Earthquake
Engineering Research Institute, Oakland, California.
Ingale, R. A. (2008). Dynamics of vibrated granular matter. PhD Thesis, The City University of
New York.
Mod. Phys. 68, 1259.
Youd, L. T. (1968) Reduction of critical void ratio during steady-state vibration. International
symposium on wave propagation and dynamic properties of earth materials, Albuquerque,
N. Mex., 1967, Proc.
74
Maslov, N. N. (1959). “Stability Conditions of Saturated Sands”. (Russian: Usloviya
Ustoychivosti Vodonasishennikh Peskov). M. – L. Gosenergoizdat. 328 p.
Meehan, C. L., Boulanger, R. W., and Duncan, J. M. (2008). "Dynamic Centrifuge Testing of
Slickensided Shear Surfaces", Journal of the Geotechnical Engineering Division, American
Society of Civil Engineers (month), Vol. 134, No. 8, pp. 1086-1096
Melhus, M. F. (2011). Effects of Noise and Vibration on the Solid to Liquid Fluidization
Transition in Small Dense Granular Systems under Shear. PhD Thesis. Northwestern
University, Evanston, Illinois.
Melosh, H. J. (1979). “Acoustic Fluidization: A New Geologic Process?” Journal of Geophysical
Research, v. 84: pp. 7513-7520.
Melosh, H. J. (1996). Dynamical weakening of faults by acoustic fluidization, Nature, 379, pp.
601-606.
Metcalfe, C., Tennakoon, S. G. K., Kondic, L., Schaeffer, D. G. and Behringer, R. P. (2002).
Granular Friction, Coulomb Failure, and the Fluid-Solid Transition for Horizontally
Shaken Granular Materials. Physical Review E - Statistical, Nonlinear, and Soft Matter
Physics.
Pokrovsky, G. I., Ehrlich, A. A., Laletin, N. V. and Lush, F. A. (1934). New Methods of
Investigation of the Compressibility and Internal Friction in Soils. Vestnik Voyenno-
Pyke, R., Seed, H. B. and Chan, C. K. (1975). “Settlement of sands under multidirectional
shaking”. Journal of Geotechnical Engineering Division, ASCE, 101, GT-4, pp.379-398.
75
Richards, R., Jr., Elms, D. and Budhu, M. (1990). ”Dynamic Fluidization of Soils.” Journal of
Geotechnical Engineering, 116(5), pp. 740–759.
Richart, F. E. (1970). “Vibrations of Soils and Foundations”. Prentice-Hall, 414 p.
Robertson, P. K. and Campanella, R. G. (1985). “Liquefaction potential of sands using the
CPT”. Research Report R69-15, Dept. of Civil Engineering, Massachusetts Institute of
Technology, Cambridge, Mass.
Robertson, P. K. and Campanella, R. G. (1986). “Estimating liquefaction potential of sands
using the flat plate dilatometer”. Geotechnical Testing Journal, 9, No. 1, pp. 38-40.
Sangroya, R. and Choudhury, D. (2013). “Stability Analysis of Soil Slope Subjected to Blast
Induced Vibrations Using FLAC3D”, Geo-Congress 2013, pp. 472-481
Savchenko, I. A. (1958). “Effect of Vibrations on Internal Friction in Sand”. Symposium, Soil
Dynamics, Moscow, Gosstroiizdat, No. 2.
Seed, H. B. (1966). “A method for earthquake resistant design of earth dams”. Journal of the Soil
Mechanics and Foundation Division, ASCE, 92, No. SM-1, pp. 13-41.
Seed, H. B. and Idriss, I. M. (1982). Ground motions and soil liquefaction during earthquakes.
Monograph series, Earthquake Engineering Research Institute, Berkeley, California.
Seed, H. B. and Idriss, I. M. (1983). Evaluation of liquefaction potential using field performance
data. Journal of Geotechnical Engineering, ASCE, 109, No. 3, pp. 458-482.
Shibata, T. and Yukitomo, H. (1969). “Liquefaction Process of Sand during Cyclic Loading”.
Soils and Foundations, Vol. 3, pp. 54-69.
76
Sornette, D. and Sornette, A. (2000). “Acoustic Fluidization for Earthquakes?”. Bulletin of the
Seismological Society of America, 90, 3, pp. 781-785.
Wassgren, C. R. (1997). Vibration of granular materials. PhD Thesis, California Institute of
Technology, Pasadena, California.
77
5. Chapter 5: Vibrational Fluidization of Granular Media
5.1 Abstract
To investigate the vibrational fluidization of granular media, sixty samples of three different
granular materials are tested at six normal stresses and a wide range of vibration accelerations.
The experiments are conducted on a modified vibrating direct shear apparatus, where vibration is
horizontally applied in the direction of shear. The peak, residual and vibro-residual shear
strength envelops are obtained for the tested materials at different vibration intensities. It has
been determined that increasing vibration intensity reduces the friction angle of the granular
materials, as well as increases the value of the normal stress, σf, below which, the granular
material is fluidized. The particle shape effect on strength loss due to vibration has also been
observed, and a reduction in the friction angle and increase in the σf with an increase in vibration
acceleration are found.
5.2 Introduction
78
For example, when vibration is applied in the vertical direction, fluidization takes place at an a
that is greater than 1 g (Huan, 2008). Further increases in the vibration acceleration changes the
behavior of the granular material to that of a gas. Jaeger et al. (1996) described the three
aforementioned states of granular materials.
phenomena, such as compaction (e.g. Barkan (1962), Ayer and Soppet (1965/1966)), swelling
(e.g. Poschel and Rosenkranz (1998)), mixing (e.g. Alexeev et al. (2000)), localized excitations
(e.g. Umbanhowar et al. (1996), Tsimring and Aranson (1997)), convective flow (e.g. Laroche et
al. (1989), Evesque and Rajchenbach (1989), Gallas et al. (1992), Pak et al. (1995) and Huan
(2008)), Bourzutschky and Miller (1995), Wassgren (1997) and Liffman et al. (1997)), size
segregation (e.g. Knight et al. (1993), Cooke et al. (1996) and Alexeev et al. (2000)) and surface
wave formation (e.g. Pak and Behringer (1993), Melo et al., (1993 and 1995), Clement et al.
(1996), Brone and Muzzio (1997), and Mujica and Melo (1998)) can be observed. Although the
above mentioned processes have been well investigated, there are very few, if any, studies done
on the effect of normal stress on the vibrational fluidization of granular materials.
Richards et al. (1990) proposed the concept of “dynamic fluidization” which takes into
consideration the effect of earthquake accelerations on dry granular soils. The imposed
accelerations at some critical level change the state of the soil, which causes general
plastification, such that the soil becomes, in a sense, an anisotropic fluid. They assumed that the
main trigger of the fluidization is the inertial forces that act between the particles of a granular
soil. They showed that fluidization mainly depends on horizontal and not vertical accelerations.
Another distinguishing feature of fluidization is that when it occurs, flow takes place, if at all, in
79
finite increments rather than continuously, with increments that correspond to the acceleration
pulses of an earthquake above a critical value (Richards et al., 1990).
Other researchers who investigated vibro-fluidization of granular materials are Savage
(1988), Fauve et al. (1989), Doudy et al. (1989), Zik et al. (1992), Lan et al. (1995), Warr et al.
(1995), Goldstein et al. (1995), Luding (1995), Ristow et al. (1997), Tennakoon et al. (1998),
Falcon et al. (1999), Sunthar et al. (2001), Moon et al. (2004) and Gotzendorfer et al. (2006).
The objective of the conducted laboratory experiments outlined below is to investigate the
effect of vibration intensity, normal (confining) stress and particle shape on the fluidization
phenomenon of dry granular media. The following series of tests is a continuation of an
experimental program conducted earlier with the use of similar testing equipment and
procedures.
In order to investigate the vibrational fluidization of granular materials, a direct shear apparatus
was modified into a vibrating direct shear apparatus (Taslagyan et al., 20141) as shown in Figs.
5.1 and 5.2. The modifications include installation of an electromagnetic actuator (11) between
the proving ring (3) and the shear box (2), as well as two load cells (10 and 12) for shear force
measurement at the top and bottom halves of the shear box (2). An extension (13) is added to the
main body (5) to accommodate all of the additional components. The actuator (11) has two
electromagnets, which generate vibrations in the horizontal shearing direction. The frequency
and intensity of the vibrations were controlled by using a control panel (not shown in Fig. 5.1).
(LVDTs) that measured the vertical and horizontal displacements, two load cells (10 and 12),
80
and two uniaxial accelerometers that measured the vertical and horizontal vibration accelerations
on the soil samples. One of the accelerometers was placed on top of the loading plate (measured
vertical vibration accelerations), and the other accelerometer was attached to the top half of the
shear box in the direction of the shear (measured horizontal vibration accelerations). The output
signals were acquired by using the NI CompactDAQ System, which was in turn, connected to a
PC that logged the data with NI LabVIEW software.
(a) 1 (b)
2 3 10 11
4
12
5
13
6
8
7
9 7
Figure 5.1 (a) – Strain controlled direct shear apparatus (LVDTs that measured the vertical and
shear displacements are not shown for simplicity) and (b) – modified strain controlled vibrating
direct shear apparatus.
81
shear apparatus.
tested.
Figure 5.2 Modified direct shear apparatus.
82
In order to investigate the effect of vibrating on the shear zone in granular materials, three
different dry granular materials (a total of 60 samples) were tested on the modified direct shear
apparatus. The tested materials were 0.55 mm glass beads, as well as fine and coarse sands. The
samples filled a shear box with dimensions of 6 cm x 6 cm x 3.2 cm (W x L x H), after which, a
constant compaction effort was applied to the samples which brought them to a dense state and
similar density. The samples were tested at normal stresses of 8, 23, 36, 50, 118 and 200 kPa in
strain-controlled mode at a shear rate of 0.61 mm/min. The vibration frequency was 60 Hz for all
of the samples that were being tested. Four sample sets of the 0.55 mm glass beads (six samples
each), as well as three sample sets of the fine and coarse sand were tested at different vibration
intensities. The sensor data were logged at a frequency of 1 kHz.
The highest normal stress did not exceed 200 kPa to avoid grain crushing. Low normal
stresses (8, 23 and 36 kPa) were used to obtain realistic experimental evidence of material
fluidization at low normal (confining) stresses.
The physical characteristics of the tested materials are provided in Table 5.1. The particle
size distributions of the three materials are given in Fig. 5.3. Representative samples of the
materials were observed under a microscope to determine the shape of the particles (see Figs.
5.4a, 5.4b and 5.4c). As can be seen in the photographs in Fig. 5.4a, the 0.55 mm glass beads
have a practically ideal spherical shape. From Figs. 5.4b and 5.4c, it can be concluded that the
fine sand particles have an angular shape and the coarse sand is well rounded.
83
Table 5.1 Physical characteristics of the tested materials.
Material ρs ρ e n
Glass Beads 2.65 1.566 0.692 0.409
Fine Sand 2.65 1.460 0.815 0.449
Coarse Sand 2.65 1.755 0.510 0.338
The three materials were chosen because of their different particle shapes in order to
investigate the particle shape effect on the vibrational fluidization of granular materials. A
material with round and angular particle shape was selected to obtain a wide range of residual
friction angles (φr from 21.2o to 32.2o), and the well rounded coarse sand (with φr = 29.7o) was
chosen as a material with an intermediate friction angle to determine the reliability of the
approximation of the experimental results for granular materials which have a residual friction
angle in the range of 21.2o to 32.2o.
100
90
80
70
60
%
50 Fine Sand
finer
40 Coarse Sand
30 Glass Beads 0.55 mm
20
10
0
0.01 0.1 1 10
mm
Figure 5.3 Particle size distribution of the tested materials.
84
(a) (b)
(c)
Figure 5.4 Photographs of the glass beads and sand samples under a microscope at magnification
of 50x. (a) - 0.55 mm glass beads; (b) - fine sand; and (c) - coarse sand
(scale bar on the right bottom corner is 0.1 mm in length)
85
5.5 Test Results
The results of the laboratory experiments conducted on the glass beads, and fine and coarse
sands are summarized in Tables 5.2, 5.3 and 5.4, respectively, where h-av is the average relative
horizontal acceleration, is the normal stress, p is the peak strength, r – residual strength, rv –
vibro-residual strength, h –relative horizontal acceleration, v –relative vertical acceleration,
and strength loss for the residual to the vibro-residual strengths.
The relative acceleration, , mentioned above, is a dimensionless parameter and defined
as:
  2/g [1]
where A is the vibration amplitude,  is the vibration frequency and g is the gravitational
acceleration.
The values given in Tables 5.2, 5.3 and 5.4 were taken from the plots of the sensor
readings. An example of such is provided in Fig. 5.5. The shear resistance plots were obtained
from the load cell that measured the shear force at the top half of the shear box (see Figs. 5.1 (b)
– 12). The vertical displacement plots were constructed from the measurements of an LVDT
located on the loading plate of the shear box. The horizontal and vertical acceleration plots were
created from an accelerometer that measured the horizontal accelerations at the top half of the
shear box in the direction of shear, and a second accelerometer that measured the vertical
accelerations of the loading plate, respectively.
86
All of the data, except for vertical displacement, were collected at a frequency of 1 kHz.
Due to the technical characteristics of the LVDT, vertical displacements were plotted based on 2
Hz data filtered from the original 1 kHz records.
As seen in Fig. 5.5, vibration is applied when the samples have reached their residual
strength. The shear resistance plot shows that upon the application of vibration, an immediate
strength loss of shearing resistance Δτ takes place for the residual to the vibro-residual strengths.
When vibration is terminated, the strength of the sample gradually increases and reaches the
residual strength value. All of the tested samples experienced compression (compaction) due to
vibration (e.g. see Figs. 5.5, 5.11 – 5.19). Note that the normalized displacement in the plots of
Figs. 5.5, 5.11 – 5.19 is defined as d/L, where d is the horizontal displacement and L is the
initial length of the specimen.
The peak, residual and vibro-residual shear strength diagrams of the glass beads, as well as
those of the fine and coarse sand samples tested at different vibration intensities, are provided in
Figs. 5.6 (a, b, c, d), 5.7 (a, b, c) and 5.8 (a, b, c), respectively. The residual and vibro-residual
friction angles of the three materials tested at different vibration intensities are given in Table
5.5.
87
Figure 5.5 An example of test results of a fine sand sample tested at σ=200 kPa, =60 Hz and
h=2.3.
88
Table 5.2 Test results of glass beads.
h-av  p r rv h v 
8 8 6 2 0.4 0.04 4
23 16 12.5 7 0.55 0.08 5.5
36 22 19 12.5 0.35 0.15 6.5

0.44
50 29 24 17 0.5 0.15 7
118 63.5 53 38 0.44 0.13 15
200 107 85 66 0.4 0.11 19
8 8 6 0 1.55 0.3 6
23 16 12.5 4 1.4 0.2 8.5
36 23 18.5 6.5 1.55 0.2 12

1.38
50 30 24 10 1.4 0.18 14
118 65.5 54.5 32 1.2 0.42 22.5
200 107 91 58 1.2 0.35 33
8 8 6 0 2.2 0.45 6
23 16 13 1 2 0.32 12
36 24 19 4.5 2.2 0.32 14.5

2.12
50 29.5 23 8 2.1 0.22 15
118 65 54 26 2.2 0.41 28
200 107 88 53 2 0.41 35
8 8 6 0 3.5 0.38 6
23 17 13 0 3.5 0.7 13
36 24 19 0 3.4 0.9 19
3.40
50 31 25 4 3.4 0.75 21
118 63 50 17 3.3 0.75 33
200 106 86 40 3.3 0.65 46
89
It is apparent from Figs. 5.6, 5.7 and 5.8 that increasing of vibration intensity results in a
reduction of the residual friction angle of all the materials. From these figures, it can also be seen
that besides reducing the friction angles of the materials, vibration results in the intersecting of
the shear strength envelop with the normal stress axis at the non-zero vibro-fluidizational limit,
σf, which increases with an increase in the vibration intensity. This implies that at a given
vibration acceleration, a granular material can fluidize at normal stresses up to the σf value, and
above the σf value, remains in the solid state, but has a smaller friction angle (vibro-residual
friction angle, φvr) than the residual one, φr. Thus, Coulomb’s equation for dry granular material,
 = σ tan φ, can be written as:
 = (σ – σf) tan φvr if σ > σf [2]
=0 if σ ≤ σf
where  – shear stress, σ – normal stress, σf – vibro-fluidizational limit (intersection point of the
vibro-residual shear strength with the normal stress axis), and φvr –friction angle of vibro-
residual strength.
Alternatively, the σf parameter can be expressed as:
σf = /tanφvr + σ where σ > σf [3]
90
Table 5.3 Test results of fine sand.
h-av  p r rv h v 
8 12.5 9 5 0.43 0.07 4
23 25 19 14.5 0.43 0.14 4.5
36 36.5 26.5 22.5 0.4 0.13 4

0.42
50 46 37.5 32.5 0.41 0.13 5
118 103 81 71.5 0.43 0.18 9.5
200 166 130 118.5 0.4 0.08 11.5
8 13 9.5 0 2.5 0.22 9.5
23 25 20.5 8 2.5 0.4 12.5
36 35.5 31 14 2.6 0.58 17

2.42
50 46.5 37.5 19.5 2.3 0.4 18
118 102.5 83.5 51 2.3 0.75 32.5
200 170.5 130 91 2.3 0.7 39
8 13.5 9.5 0 3.5 0.4 9.5
23 24 20 0 3.9 0.65 20
36 39.5 30 5 4 0.85 25
4.25
50 46 34.5 8 4.5 0.7 26.5
118 112.5 87 38 4 0.7 49
200 178 128 62 4.5 1.3 66
91
Table 5.4 Test results of coarse sand.
h-av  p r rv h v 
8 11 7.5 1.5 0.44 0.04 6
23 21.5 16.5 9.5 0.47 0.08 7
36 30 25 19 0.47 0.1 6
0.44
50 40.5 32 22.5 0.46 0.15 9.5
118 88 71 62.5 0.42 0.15 8.5
200 144 119 97 0.4 0.21 22
8 11 8 0 2.3 0.52 8
23 21 16 0 2.6 0.52 16
36 29.5 24 7 2.7 0.49 17

2.50
50 39.5 31.5 11 2.8 0.49 20.5
118 88 71.5 36 2.3 0.52 35.5
200 141.5 117.5 76 2.3 0.75 41.5
8 10.5 8 0 4.6 1.1 8
23 21 17 0 4.2 1.1 17
36 31 24.5 0 4.1 1.2 24.5

4.28
50 40 32.5 4 4.5 0.8 28.5
118 89 71 27.5 4.2 1 43.5
200 145.5 115.5 52 4.1 1.4 63.5
92
(a) Glass Beads 0.55mm (0.44 g) Peak
Residual
200
Vibro-Residual
180
Linear (Peak)
160 Linear (Residual)
140 Linear (Vibro-Residual)
120
τ, kPa
y = 0.5142x + 3.6364
100
y = 0.4112x + 3.4389
80
y = 0.3298x - 0.1583
60
40
20
0
0 50 100 150 200
σ, kPa
(b) Glass Beads 0.55mm (1.38 g) Peak

Residual
200
Vibro-Residual
180
Linear (Peak)
120
τ, kPa
y = 0.5154x + 4.2164
100
y = 0.443x + 2.3
80
60 y = 0.3106x - 4.423
40
20
0
0 50 100 150 200
σ, kPa
93
(c) Glass Beads 0.55mm (2.12 g) Peak
Residual
200
Vibro-Residual
180
Linear (Peak)
120
τ, kPa
y = 0.5136x + 4.3464
100
y = 0.4271x + 2.8694
80
60
y = 0.2917x - 6.4086
40
20
0
0 50 100 150 200
σ, kPa
(d) Glass Beads 0.55mm (3.4 g) Peak

Residual
200
Vibro-Residual
180
Linear (Peak)
120
τ, kPa
y = 0.5023x + 5.0801
100
y = 0.4091x + 3.5079
80
60
40 y = 0.2414x - 9.2751
20
0
0 50 100 150 200
σ, kPa
.
Figure 5.6 Shear strength diagrams of glass beads tested at (a) - h0.44, (b) - h1.38, (c) -
h2.12 and (d) - h3.4.
94
(a) Fine Sand (0.42 g)
200
180
y = 0.8008x + 6.7789
160
140 y = 0.6322x + 4.6627

120 y = 0.5889x + 1.3849
τ, kPa
100
80
Peak
60 Residual
40 Vibro-Residual
20 Linear (Peak)
Linear (Residual)
0
0 50 100 150 200 Linear (Vibro-Residual)
σ, kPa
(b) Fine Sand (2.42 g)

200
180
y = 0.8211x + 5.9726
160
140 y = 0.6264x + 6.5878

120
τ, kPa
100
y = 0.4697x - 3.4721
80
60 Peak
40 Residual
Vibro-Residual
20
Linear (Peak)
0
Linear (Residual)
0 50 100 150 200
σ, kPa Linear (Vibro-Residual)
95
(c) Fine Sand (4.25 g)
200
180 y = 0.8694x + 5.8862
160
140 y = 0.6276x + 5.9985

120
τ, kPa
100
80
60 y = 0.3577x - 7.8786
Peak
40
Residual
20
Vibro-Residual
0 Linear (Peak)
0 50 100 150 200
Linear (Residual)
σ, kPa
Linear (Vibro-Residual)
Figure 5.7 Shear strength diagrams of fine sand tested at (a) - h0.42, (b) - h2.42 and (c) -
h4.25.
(a) Coarse Sand (0.44 g)

200
180
160
140 y = 0.6941x + 5.5135
120 y = 0.5778x + 3.2786

τ, kPa
100 y = 0.5021x - 1.0703
80
Peak
60 Residual
40 Vibro-Residual
20 Linear (Peak)
Linear (Residual)
0
0 50 100 150 200 Linear (Vibro-Residual)
σ, kPa
96
(b) Coarse Sand (2.5 g)
200
180
160
140 y = 0.6846x + 5.4492
120 y = 0.573x + 3.2092

τ, kPa
100
80
y = 0.4209x - 10.011
60 Peak
40 Residual
Vibro-Residual
20
Linear (Peak)
0
Linear (Residual)
0 50 100 150 200
σ, kPa Linear (Vibro-Residual)
(c) Coarse Sand (4.28 g)

200
180
160
y = 0.7041x + 5.1198
140
120 y = 0.5592x + 4.2056

τ, kPa
100
80
60 y = 0.3193x - 11.332
Peak
40
Residual
20
Vibro-Residual
0 Linear (Peak)
0 50 100 150 200
Linear (Residual)
σ, kPa
Linear (Vibro-Residual)
Figure 5.8 Shear strength diagrams of coarse sand tested at (a) - h0.44, (b) - h2.5 and (c) -
h4.28.
97
Evidence of the presence of both a vibro-fluidizational limit, σf, and a reduced friction
angle due to vibration can possibly be explained by the bifurcation process that takes place
during sufficiently strong vibrations, which results in the fluidization of the top layer of a
granular material, while the bottom layer remains in a solid state. This phenomenon was
described in detail by Metcalfe et al. (2002), who performed experiments on granular materials
by subjecting them to horizontal vibrations and concluded that normal stress has a significant
impact on the thickness of the vibro-fluidized layer of the granular media.
Table 5.5 Friction angles and normal stress axis intersection values of the shear strength
diagrams for the three granular materials tested at different vibration intensities.
Material h-av  f kPa
0 32.2 0
0.42 30.5 7.4

Fine Sand
2.42 25.1 19.6
4.25 19.7 38.1
0 29.7 0
0.44 26.7 2.1

Coarse Sand
2.50 22.8 23.8
4.28 17.7 35.5
0 21.2 0
0.44 18.3 9.7
Glass Beads 1.38 17.3 24
2.12 16.3 32.4
3.40 13.6 51
98
35
30
y = -2.8795x + 31.969
25
Glass Beads
y = -2.5917x + 28.903
20
Fine Sand

Coarse Sand
15
Linear (Glass Beads)
y = -1.9735x + 20.237
10 Linear (Fine Sand)
Linear (Coarse Sand)
5
0
0 1 2 3 4 5
Relative horizontal acceleration,  h
Figure 5.9 Friction angles of the tested materials at different vibration intensities.
The final results provided in Table 5.5 are graphically presented in Figs. 5.9 and 5.10. It is
seen from Fig. 5.9 that the reduction of the residual friction angles of the tested granular
materials with an increase in vibration intensity is practically linear. The only noticeable
deviation from the linear pattern of the φ vs. h relationship is observed at a h value of less than
1. It is also apparent that the plot of the well rounded coarse sand correlates well with those of
the angular fine sand and glass beads. This means that the plots presented in Fig. 5.9 can be used
for the approximation of the residual friction angle changes of sand sized dry granular materials
(which have a residual friction angle that ranges from 21o to 32o) when subjected to vibration in
the given range of horizontal acceleration values.
99
80
70
Vibro-fluidizational limit (kPa)
60
50 Glass Beads
Fine Sand
40
Coarse Sand
30 Linear (Glass Beads)
Linear (Fine Sand)
20
Linear (Coarse Sand)
10
0
0 1 2 3 4 5
Relative horizontal acceleration,  h
Figure 5.10 σf values at different vibration intensities. (Note that σf = 0 at h = 0. These points
represent a static condition).
As shown in Fig. 5.10, the plots of the σf vs. h relationship for all of the tested materials
are linear. At a given vibration intensity, the glass beads have a greater σf magnitude than the
coarse and fine sands, the plots of which practically coincide. It can be assumed that there is a
certain level of roundness, above which the σf vs. h plot will shift from that of the coarse and
fine sand to the one of the glass beads.
Since vibration generated by the actuator is in the horizontal (shear) direction, the main
parameter that defines vibration intensity used in this work is horizontal acceleration. The
vertical accelerations measured on the loading plate are generated due to the horizontal
vibrations and their magnitude depends on the boundary conditions and the materials tested. That
100
is why the v values provided in Tables 5.2, 5.3 and 5.4 are somewhat random at given h-av.
Despite this fact, it is apparent from Tables 5.2, 5.3 and 5.4 that the v values generally increase
with an increase in the magnitude of the horizontal acceleration (h). Therefore, it should be
mentioned that although vertical accelerations are not used in the data analysis in this paper, they
contribute to the strength loss of the tested materials, and show a well defined trend of their
magnitude change which is related to the horizontal acceleration intensity.
Below are some examples that show the effect of vibration intensity on the shearing
resistance and compressive deformations of the glass beads. Figs. 5.11, 5.12, 5.13 and 5.14 show
the test results of glass bead samples tested at a normal stress of 200 kPa and h equal to 0.4, 1.2,
2 and 3.3, respectively. It can be seen that an increase in the horizontal vibration acceleration
results in not only greater strength loss, but also greater compression deformations. Note that at
high vibration intensities (Figs. 5.13 and 5.14) excessive vibro-compaction of the samples leads
to post-vibrational peaks, which with further shear deformation, decrease to the residual strength.
It can also be seen that after sufficiently strong vibration, the development of post-vibrational
peak strength is accompanied with post-vibration dilation (see Figs. 5.14 – 5.19). On the other
hand, the shear resistance plots of the fine and coarse sands do not show any post-vibrational
peaks at high normal stress values (σ = 118 kPa and 200 kPa), which can be explained by the
irregular shape of the grains that did not allow the samples to undergo sufficient compaction at
high confining stresses, even at horizontal vibration accelerations greater than h > 4.
The irregular pattern of the shear resistance plots at the post-peak strength state (see Figs.
5.11, 5.12, 5.13 and 5.14) is due to the arranged alignment of the glass beads on the shear surface
that takes place when the tested samples approach their residual strength. When the glass beads
101
which have an arranged alignment snap over each other, a sudden decrease in the shearing
resistance takes place, after which, the shear stress starts to increase until it reaches its residual
value. This phenomenon has only been observed for glass beads at higher normal stresses, σ =
118 kPa and 200 kPa (see Fig. 5.5 for comparison purposes).
The effect of normal (confining) stress on the strength and deformation characteristics of
granular materials subjected to similar vibration intensities (horizontal vibration accelerations)
can be seen in Figs. 5.14 – 5.19. The greatest compressive deformations induced by vibration are
observed at the lowest normal stress, σ = 8 kPa. An increase in normal stress from 8 to 200 kPa
leads to a reduction in compression. This phenomenon is also apparent from the magnitude of the
post-vibrational peak of the shear resistance plots. Higher normal stress means lower post-
vibrational peak. Therefore, the mechanism here is similar to the initial dilation of the tested
samples, where the dilation magnitude reduces with an increase in normal stress (see Figs. 5.14 –
5.19).
102
samples tested at h = 0.4.
103
104
samples tested at h = 2.
105
106
glass beads tested at σ = 8 kPa.
107
108
109
110
111
On the other hand, as seen from Figs. 5.14, 5.18, and 5.19, the magnitude of the shear
strength loss due to vibration increases with an increase in the normal stress, which in turn,
defines the friction angle reduction from the residual to the vibro-residual value.
Figs. 5.14 – 5.19 show that strong vibrations (h = 3.3 to 3.5) bring the samples to a denser
state than their state prior to testing. It is also seen that compression caused by vibration is
greater than dilation due to the shearing of the samples. This results in well defined post-
vibrational peaks in the shear resistance plots.
As mentioned earlier, the development of post-vibrational peak strength is accompanied
with post-vibration dilation. It is apparent from Figs. 5.14 to 5.19 that the magnitude of post-
vibration dilation is smaller than the initial dilation of the samples. Therefore, it can be
concluded that post-vibration dilation is caused by the expansion of the shear zone of the
samples, while the difference between the initial and post-vibration dilation is due to the
compression of the material beyond the shear zone.
It can also be seen that some of the samples showed an increase in shear resistance during
vibration (e.g. Fig 5.14), which can be explained by the ongoing densification of the shear zone
material.
The experimental results mentioned above can contribute to a better understanding of long
run-out distance of large rock avalanches. The attempts to explain this phenomenon were made
by Melosh (1979) and Davies (1982). It has been observed that the run-out distance of large rock
avalanches depends on their volume, such that the larger volume of an avalanche results in
longer run-out distance.
112
The evidence that sufficiently strong vibration decreases the friction angle of a granular
material from residual to vibro-residual value, such that greater vibration intensity results in
lower vibro-residual friction angle, correlates well with the concepts of “acoustic fluidization”
(Melosh, 1979) and “mechanical fluidization” (Davies, 1982) attempting to explain the long run-
out distances of large rock avalanches: the resistance of granular media at the shear interface is
affected by the vibration generated by the collisions of the avalanche material with the stationary
base, such that larger volume of a rock avalanche having greater kinetic energy generates greater
vibration intensity, which will, in turn, result in a longer run-out distance of the rock avalanche.
5.6 Conclusions
Based on the physical observations of the mechanical response of sand and glass beads under
vibrational shearing, the following conclusion can be made.
1. Upon vibration application, an immediate strength loss of the granular material takes
place from the residual to vibro-residual strengths. If the granular material experiences
sufficient shear deformation after the termination of vibration, its strength reaches the
residual strength value.
2. All of the tested samples have experienced compression (compaction) due to the applied
vibration. An increase in the horizontal vibration acceleration results in greater
compression deformation.
3. An increase in the vibration intensity results in a reduction of the residual friction angle
of all the tested granular materials.
113
4. Besides reducing the friction angles of the tested materials, vibration causes the
intersecting of the shear strength envelop with the normal stress axis at a certain vibro-
fluidizational limit value, σf, which increases with an increase in vibration intensity. This
implies that at a given vibration acceleration, a granular material can fluidize at normal
stresses up to the σf value, and above this σf value, remains in a solid state. However, the
granular material has a smaller friction angle (vibro-residual friction angle, φvr) than the
residual friction angle, φr. Thus, the Coulomb’s equation for dry granular material,  = σ
tan φ, can be written as:  = (σ + σf) tan φvr.
5. The reduction of the residual friction angles of the tested granular materials with an
increase in vibration intensity is practically linear. The only noticeable deviation from the
linear pattern of the φ vs. h relationship is observed at a h value less than 1.
6. The φ vs. h plots can be used for the approximation of changes in the residual friction
angle of sand sized dry granular materials (with residual friction angles from 21o to 32o)
when subjected to vibration in the given range of the horizontal acceleration values (0 <
h < 4.5).
7. The plots of the σf vs. h relationship for all of the tested materials are linear. At a given
vibration intensity, the glass beads have a greater σf magnitude than the coarse and fine
sands, the plots of which coincide. It can be assumed that there is a certain level of
roundness, above which, the σf vs. h plot will shift from that of the coarse and fine sand
to the one of the glass beads.
8. Excessive vibro-compaction of the samples leads to post-vibrational peaks, which with
further shear deformation, decreases to the residual strength. The development of post-
vibrational peak strength is accompanied with post-vibration dilation.
114
9. The shear resistance plots of the fine and coarse sands, unlike that of the glass beads, do
not show any post-vibrational peaks at high normal stress values (σ = 118 kPa and 200
kPa), which can be explained by the irregular shape of the grains that did not allow the
samples to undergo sufficient compaction, even at horizontal vibration accelerations
greater than h > 4.
10. The largest compressive deformation of the granular materials induced by vibration is
observed at the lowest normal stress, σ = 8 kPa. An increase in the normal stress from 8
to 200 kPa leads to compression reduction. This phenomenon is also apparent from the
magnitude of the post-vibrational peak of the shear resistance plots. A higher normal
stress means a lower post-vibrational peak.
11. It appears that the essence of acoustic fluidization, which attempts to explain the large
runout distance of big rock avalanches (Melosh, 1979), and vibrational fluidization
observed in the above mentioned experiments, is the same: the resistance of granular
media at the shear interface is affected by the vibration generated by the collision of the
grains.
5.7 References
Alexeev, A., Royzen, V., Dudko, A., Goldshtein, A. and Shapiro, M. (2000). "Mixing and
Segregation in Vertically Vibrated Granular Layers". Solid Mechanics and Its
Applications, Volume 81, 2000, pp 129-139
115
Ayer, J. E. and Soppet, F. E. (1965/1966): Vibratory Compaction, Part I: Compaction of
Spherical Shapes and Part II: Compaction of Angular Shapes. J. Am. Ceramic Soc. 48,
180-183 and 49, 207-210.
Bourzutschky, M. and Miller, J. (1995): "Granular" Convection in a Vibrated Fluid. Phys. Rev.
Lett. 74, 2216-2219.
Brone, D. and Muzzio, F. J. (1997). Size segregation in vibrated granular systems: A reversible
process. Phys. Rev. E. 56[1], pp. 1059-1063.
Clement, E., Vanel, L., Rajchenbach, J. and Duran, J. (1996). Pattern formation in a vibrated
granular layer. Phys. Rev. E, 53, 2972-2975.
Cooke, W., Warr, S., Huntley, J. M. and Ball R. C. (1996). “Particle size segregation in a two-
dimensional bed undergoing vertical vibration”, Phys. Rev. E, 53, 2812.
Doudy, S., Fauve, S. and Laroshe, C. (1989). Subharmonic instabilities and defects in a granular
layer under vertical vibrations. Europhys. Lett., 8[7], 621
Evesque, P. and Rajchenbach, J. (1989). “Instability in a sand heap”, Phys. Rev. Lett., 62, 44.
Falcon, E., Wunenburger, R., Evesque, P., Fauve, S., Chabot, C., Garrabos, Y. and Beysens, D.
(1999). Cluster formation in a granular medium fluidized by vibrations in low gravity.
Phys. Rev. Lett., 83:440.
Fauve, S., Douady, S. and Laroche, C. (1989): Collective behaviours of granular masses under
vertical vibrations. J. Physique C3, 187-191.
Gallas, J. A. C., Herrman, H. J. and Sokolowski, S. (1992): Convection Cells in Vibrating
Granular Media. Phys. Rev. Lett. 69, 1371-1374.
116
Goldshtein, A., Shapiro, M., Moldavsky, L. and Fichman, M. (1995). Mechanics of collisional
motion of granular materials. 2. Wave propagation through vibrofluidized granular layers.
J. Fluid Mech., 287, p. 349
Gotzendorfer, A., Tai, C., Kruelle, C. A., Rehberg, I. and Hsiau, S. S. (2006), `Fluidization of a
vertically vibrated two-dimensional hard sphere packing: A granular meltdown', Phys.
Rev. E 74, 011304.
Huan, C. (2008) NMR Experiments on Vibrofluidized and Gas Fluidized Granular Systems. PhD
Thesis, University of Massachusetts Amherst.
Mod. Phys. 68, 1259
Knight, J. B., Jaeger, H. M. and Nagel, S. R. (1993). “Vibration-induced size separation in
granular media: the convection connection”, Phys. Rev. Lett., 70, 3728.
Lan, Y. and Rosato, A. D. (1995). “Macroscopic behavior of vibrating beds of smooth inelastic
spheres”, Phys. Fluids, Vol. 7, No. 8, pp. 1818-1831.
Laroche, C., Douady, S. and Fauve, S. (1989): Convective Flow of Granular Masses under
Vertical Vibration. J. Physique 50, 699-706.
Liffman, K., Metcalfe, G. and Cleary, P. (1997): Convection due to horizontal shaking. in:
(Behringer and Jenkins (1997)). 405-408.
Luding, S. (1995). Granular materials under vibration: simulations of rotating spheres. Phys. Rev.
E, 52:4442.
117
Melo, F., Umbanhowar, P. and Swinney, H. L. (1993). “Transition to parametric wave patterns in
a vertically oscillated granular layer”, Phys. Rev. Letts., 72, 172.
Melo, F., Umbanhowar, P. and Swinney, H. L. (1995). “Hexagons, kinks, and disorder in
oscillated granular layers”, Phys. Rev. letts., 75, 3838.
Research, v. 84: pp. 7513-7520.
Granular Friction, Coulomb Failure, and the Fluid-Solid Transition for Horizontally
Shaken Granular Materials. Physical Review E - Statistical, Nonlinear, and Soft Matter
Physics.
Moon, S. J., Swift, J. B., and Swinney, H. L. (2004). Role of friction in pattern formation in
oscillated granular layers. Phys. Rev. E, 69(031301):1.
Mujica, N. and Melo, F. (1998). Solid-liquid transition and hydrodynamic surface waves in
vibrated granular layers. Phys. Rev. Lett., 80:5121.
Pak, H. K. and Behringer, R. P. (1993). “Surface waves in vertically vibrated granular
materials”, Phys. Rev. Lett. 71, 1832
Pak, H. K., Van Doorn, E. and Behringer, R. P. (1995). “Effects of ambient gases on granular
materials under vertical vibration”, Phys. Rev. Lett., 74, 4643.
Poschel, T. and Rosenkranz, D. E. (1998). Experimental study of horizontally shaken granular
matter – the swelling effect. Lecture Notes in Physics, Volume 503, 1998, pp 96-109
118
Geotechnical Engineering, 116(5), pp. 740–759
Ristow, G. H., Strasburger, G. and Rehberg, I. (1997). Phase diagram and scaling of granular
materials under horizontal vibrations. Phys. Rev. Lett., 79(5):833.
Savage, S. B. (1988). Streaming motions in a bed of vibrationaly fluidized dry granular material.
J. Fluid Mech., 194:457.
Sunthar, P. and Kumaran, V. (2001). Characterization of the stationary states of a dilute
vibrofluidized granular bed. Phys. Rev. E, 64:041303.
Tennakoon, S. G. K. and Behringer, R. P. (1998). Vertical and horizontal vibration of granular
materials: Coulomb friction and a novel switching state. Phys. Rev. Lett., 81(4):794–797.
Tsimring, L. and Aranson, I. (1997). Localized and cellular patterns in a vibrated granular layer.
Phys. Rev. Lett., 79(2):213–216.
Umbanhowar, P. B., Melo, F. and Swinney, H. L. (1996). “Localized excitations in a vertically
vibrated granular layer”, Nature (London) 382, 793
Warr, S., Huntley, J. M. and Jacques, G. T. H. (1995): Fluidization of a two-dimensional
granular system: Experimental study and scaling behavior. Phys. Rev. E, 52, 5583-5595.
Wassgren, C. R. (1997). Vibration of granular materials. PhD Thesis, California Institute of
Zik, O., Stevans, J. and Rabin, Y. (1992). “Mobility of a sphere in vibrated granular media”,
Europhys. Lett., Vol. 17, No. 4, pp. 315-319.
119
6. Chapter 6: Effect of Vibration on the Shear Zone in Dry Granular Materials
6.1 Abstract
The following paper outlines the results of laboratory experiments conducted on glass beads and
sand by using a modified vibrating direct shear apparatus. The behavior of the shear zone in the
granular materials before, during and after the application of vibration is investigated at the pre-
peak and residual strength states. Three zones are identified in the sheared granular material: A –
the zone that is unaffected by the shear of the granular media, B – the shear zone portion that has
been developed due to the shear of the material, but does not contribute to the critical state of the
thinner shear zone C, at which the actual shear takes place. Also, the shear strength and
deformation characteristics of the granular materials affected by vibration at their pre-peak and
residual strength states are presented and discussed.
6.2 Introduction
The strength of a granular material is determined by the strength characteristics of the material at
the shear zone/band defined by a certain combination of boundary conditions and the mechanical
properties of the material. Therefore, to better understand the mechanism of the strength loss of
granular materials, it is important to investigate the stress and deformation characteristics of the
shear zone during its initiation and propagation through the granular media. The thickness of the
shear zone is usually 8 to 10 times the mean grain diameter (Roscoe, 1970; Muhlhaus and
Vardoulakis, 1987; Bardet and Proubet, 1992).
120
Some examples of the numerous studies on shear zones in granular media are provided
below.
Bardet and Proubet (1992) investigated the emergence, inclination, and thickness of shear
bands in idealized granular materials and showed that they are similar to those of real materials.
Aidanpaa et al. (1996) used a torsional shear cell to study the shear layers of uniform spheres,
and observed that the shear zone thickens and dilates with an increase in the shear speed,
accompanied by a transition from a single layer to many layers of shearing. Bora (1984)
experimentally investigated the shear failure mechanism in granular materials, and found that the
friction angle is dependent on effective confining contact stress within a dilative range regardless
of the drainage conditions and amount of particle crushing. Frost et al. (2002) studied the shear
failure behavior of granular–continuum interfaces on a selection of sand–continuum material
interfaces and through the use of discrete element modeling. Borja (2003) used a geological and
mathematical framework to classify deformation patterns in granular media. Frost et al. (2004)
evaluated the interface behaviour of granular soils by carrying out physical and numerical
experiments. Rechenmacher (2005) evaluated local displacements and strains that are associated
with shear band growth and evolution in sands by testing the plane strain with the use of digital
image correlation (DIC). The magnitude of the shear and rotational strains was found to vary
along the length of the shear bands, which lends support to the idea of fluctuating buildup and
collapse of the “force chains”. Alshibli and Alramahi (2006) investigated the evolution of local
strains during the shearing of particles of a granular material, and showed that when compared to
particle sliding, rotation is imperative in the shearing resistance of granular materials. Abriak and
Caron (2006) conducted an experimental study of shear in granular media, and showed that
granular media behavior depends on the local friction (grain–grain friction). Sadrekarimi (2008)
121
studied the shearing behavior of loose and dense sands by implementing constant volume ring
shear tests. Widulinski et al. (2010) performed comparative modeling of shear localization in
granular bodies with the finite and discrete element methods (FEM and DEM). Liu (2010)
conducted laboratory experiments and numerical simulations to investigate the failure
characteristics and micro-mechanical behaviors of granular soil slopes. Cox and Budhu (2010)
studied the grain shape of granular materials, and through the use of light microscopy,
determined that there are several grain shape parameters. They then related these parameters to
the dilatancy of the granular materials. Mesarovic et al. (2013) analyzed the changes in the
topology of a granular assembly during deformation by using the graph theory and showed that
the elementary mechanism of diffuse deformation consists of intermittent flips, an increasing
number of which directly stems from dilatancy, as well as that shear band formation is associated
with the massive rolling of particles.
Direct shear testing has successfully been used to evaluate the shear strength characteristics
of soils for many decades. Despite some of the disadvantages of the testing, such as lack of
control of pore pressure, failure at predefined planes and non-uniform stress conditions in the
tested samples, it is one of the most common shear strength evaluation tests used in geotechnical
laboratories these days. The popularity of direct shear testing arises from the simplicity of the
setup and ease of testing procedures.
Many researchers have studied the applicability and limitations of direct shear testing
through experimental investigations and numerical modelling. One of the more early studies on
direct shear testing through the use of finite element analysis was performed by Potts et al.
(1987). They demonstrated that, despite the strongly non-uniform stresses and strains in the shear
box before failure, strains and stresses in the final failure zone are surprisingly uniform. Liu et al.
122
(2005) investigated the interface friction through direct shear tests. Two improvements were
made to reduce the friction of the sample material at the inner surface of the upper shear box,
which caused the real shear strength to increase for a dense sample and decrease for a loose
sample. Bagherzadeh-Khalkhali and Mirghasemi (2009) performed numerical modeling and
experiments by using direct shear testing to study its suitability and limitations for testing coarse-
grained soils. Li and Aydin (2010) investigated the fluctuations in vertical displacement and
shear stress with different sized glass beads during direct shearing. Nam et al. (2011) used
multistage direct shear tests to determine the shear strength of unsaturated soils, which allowed
the matric suction to be independently controlled, and compared the results with samples which
underwent conventional direct shear testing. Härtl and Ooi (2011) investigated, through direct
shear tests, how particle shape and interparticle friction would influence bulk friction by using
DEM. They showed that packing density has less influence than particle interlocking on the bulk
friction. Kang et al. (2012) performed 3D discrete element simulations in conjunction with image
processing of the pore geometry to determine the pore size distribution and orientation in dilative
and contractive assemblies in direct shear testing.
Of special interest is the effect of vibration on the shear zone in granular media (Youd,
1968). In this work, the deformation and strength characteristics of two granular materials with
different grain shapes are experimentally investigated by applying vibration at their pre-peak and
residual strength states. The post-vibrational strength and volumetric changes of the materials are
also evaluated, as well as a case in which the materials are sheared over a smooth glass surface.
The laboratory experiments are performed on a modified vibrating direct shear apparatus, the
description of which together with the test results will be provided in the following sections.
123
A modified vibrating direct shear apparatus was used to conduct the laboratory experiments (see
Figs. 6.1 and 6.2). The modifications included the installation of an electromagnetic actuator (11)
between the shear box (2) and proving ring (3), as well as two load cells (10 and 12) for shear
force measurement at the top and bottom halves of the shear box (2). An extension (13) was
added to the main body (5) to accommodate all of the mentioned components. The actuator (11)
incorporated two electromagnets, which generate vibrations in the horizontal shearing direction.
The frequency and intensity of vibrations were controlled by a control panel (not shown in Fig.
6.1).
(LVDTs) that measure vertical and horizontal displacements, two load cells (10 and 12), and two
uniaxial accelerometers that measure the vertical and horizontal vibration accelerations on the
soil samples. One of the accelerometers was placed on top of the loading plate (measured vertical
vibration accelerations), and the other accelerometer was attached to the top half of the shear box
in the direction of the shear (measured horizontal vibration accelerations). The output signals
were acquired by using the NI CompactDAQ System, which was in turn, connected to a PC that
logged the data with NI LabVIEW software.
124
(a) (b)
1 10
2 3 11
4
12
5
13
6
8
7
9 7
vibrating direct shear apparatus.
shear apparatus (LVDTs that measured the vertical and shear displacements are not shown for
simplicity).
125
tested.
Each of the tested samples generated a set of plots, such as those provided in Figs. 6.6, 6.8
and 6.10. As seen from the figures, each plot set consists of four plots: shear resistance vs.
normalized displacement, vertical displacement vs. normalized displacement, horizontal
acceleration vs. normalized displacement and vertical acceleration vs. normalized displacement,
as well as a short outline on the right hand side related to the type of material tested, normal
stress, and vibration and material strength properties.
Figure 6.2 Modified direct shear apparatus.
126
In order to investigate the effect of vibrating on the shear zone in granular materials, two
different dry granular materials (40 samples) were tested on the modified direct shear apparatus.
The tested materials were 0.55 mm glass beads and fine sand samples. The two materials were
chosen with different particle shapes in order to investigate the particle shape effect on the shear
zone of the granular materials under vibration. The samples filled a shear box with dimensions of
6 cm x 6 cm x 3.2 cm (W x L x H), after which, a constant compaction effort was applied to the
samples which brought them to a dense state and similar density. Some of the samples were
tested in the loose state, which was achieved by pouring the granular materials through a narrow
neck funnel, and no consequent compaction. The samples were tested at normal stresses of 8, 23,
36, 50, 118 and 200 kPa in strain-controlled mode at a shear rate of 0.61 mm/min. The highest
normal stress did not exceed 200 kPa to avoid grain crashing. The vibration frequency was 140
Hz, and the force of the actuator electromagnets was set at 7.14 N and kept constant to generate
the same vibration intensity for all of the samples tested. The sensor data were logged at a
frequency of 1 kHz.
The physical characteristics of the tested materials are given in Table 6.1, where ρs – is the
density of the solids, ρ – bulk density, e – void ratio, n – porosity and φr – residual friction angle.
The particle size distributions of the four materials are given in Fig.6.3. Representative samples
of the materials were observed under a microscope to determine the shape of the particles (see
Figs.6.4a and 6.4b). As can be seen in the photographs in Figs. 6.4a and 6.4b, the 0.55 mm glass
beads have a practically ideal spherical shape, and the fine sand particles have an angular shape.
127
Table 6.1 Physical characteristics of the tested materials.
Material ρs ρ e n φr
Glass Beads 2.65 1.566 0.692 0.409 21.90
Fine Sand 2.65 1.460 0.815 0.449 31.50
100
90
80
70
60
%
50
finer Fine Sand
40
Glass Beads 0.55 mm
30
20
10
0
0.01 0.1 1
mm
Figure 6.3 Particle size distribution of the tested materials.
128
(a) (b)
Figure 6.4 Photographs of the glass beads and sand samples under a microscope at
magnification of 50x. (a) 0.55 mm glass beads and (b) fine sand.
(scale bar on the right bottom corner is 0.1 mm in length)
6.5 Test Results
6.5.1 Pre-Peak and Residual Strength States of Granular Materials under Vibration
Four samples of the 0.55 mm glass beads and four samples of the fine sand were tested to
investigate the effect of vibration on their pre-peak and residual strengths and deformation
properties. Vibration of the same intensity was applied for 15 sec while the samples were being
sheared both at pre-peak and residual strength states. The test results are summarized in Table
6.2 and Figs. 6.5a and 6.5b. The relative acceleration (horizontal and vertical), , given in Table
6.2, is a dimensionless parameter and defined as:   2/g, where A is the vibration amplitude,
 is the vibration frequency and g is the gravitational acceleration.
129
Table 6.2 Shear strengths and strength losses at pre-peak and residual strength states for 0.55 mm
glass beads and fine sand.
Material Normal Peak Residual Vib.-Res. Residual Pre-Peak Horizontal Vertical

Stress Strength Strength Strength Strength Loss Strength Loss Acceleration Acceleration
σ τ τ τ Δτ Δτp
p r vr
h v
kPa kPa kPa kPa kPa kPa
23 18 14 9 5 4 0.42 0.13
0.55 mm 50 31.5 26 18 8 4 0.41 0.13
Glass Beads 118 68.5 55.5 44.5 11 4 0.41 0.14
200 110.5 85 69 16 3 0.40 0.13
23 29 22.5 16 6.5 7.5 0.48 0.14
50 51 38.5 31 7.5 6 0.48 0.14
Fine Sand
118 105.5 90 71 19 6.5 0.46 0.14
200 164 134 119 15 8 0.42 0.14
As can be seen from Table 6.2, vibration of the glass bead samples at the pre-peak strength
state does not affect their peak strength. The magnitude of the pre-peak strength loss was
practically the same (≈4 kPa) for all normal stresses and smaller (especially at high normal
stresses) than the strength loss at the residual strength state.
The same pattern of strength loss was observed in the case of the fine sand samples, with
the exception that the pre-peak strength loss was somewhat greater (the horizontal accelerations
as well) than that of the 0.55 mm glass beads, and ranged from 6 to 8 kPa. The peak strength was
also not affected by vibration.
It is assumed tough that vibration of a greater intensity, capable of causing compaction of
the materials, would densify the samples, which would in turn, result in an increase in their peak
strength.
130
It is seen from Figs. 6.5a and 6.5b that vibration reduces the residual friction angle values
of the tested glass beads and fine sand. Note that this is not real cohesion since granular materials
have zero cohesion. The cohesion intercept is due to the assumption that a straight line can be
used to approximate the shear strength envelop. The actual shear strength envelop is non-linear,
especially at low normal stresses, due to particle interlocking and dilation.
Two examples of the fine sand and glass bead samples tested at a normal stress of 50 kPa
are provided in Figs. 6.6 and 6.7, respectively. The vertical displacement plots in Figs. 6.6 and
6.7 show no deformation during pre-peak vibration. The only exception from the eight samples
of the glass beads and sand is a fine sand sample tested at a normal stress of 23 kPa, where the
sample experiences compressive deformation at the pre-peak strength state. It has to be
mentioned that all 40 tested samples (both types of materials) demonstrated compressive
deformations upon application of vibration at the residual strength state. Note that the normalized
displacement in the plots of Figs. 6.6, 6.7, 6.14 and 6.15 is defined as d/L, where d is the
horizontal displacement and L is the initial length of the specimen.
131
(a) Glass Beads
200 Peak
180
Residual
160
Shear Stress, kPa
140 Vibro-Residual
120 y = 0.525x + 5.8
100
y = 0.402x + 5.8
80
y = 0.343x + 1.6
60
40
20
0
0 50 100 150 200
Normal Stress, kPa
(b) Fine Sand

200
180
y = 0.764x + 12.7
160
Shear Stress, kPa
140 y = 0.641x + 8.6

120 y = 0.583x + 2.2
100
80 Peak
60
Residual
40
20 Vibro-Residual
0
0 50 100 150 200
Normal Stress, kPa
Figure 6.5 Shear strength diagrams of 0.55 mm glass beads (a) and fine sand (b).
The compressive deformation and relatively high value of strength loss at the pre-peak
strength state of the fine sand sample tested at a normal stress of 23 kPa showed that
compression of the fine sand at the given vibration intensity can take place only at low normal
stresses. High normal stresses prevent the sand particles from sliding along each other during
vibration at the pre-peak strength state.
132
In comparing the vertical deformation plots of the 0.55 mm glass beads and fine sand
samples, the absence of compressive deformations due to vibration at the pre-peak strength state
of the glass bead sample tested at a normal stress of 23 kPa can be explained by the difference in
the particle shape of the glass beads (spherical) and the fine sand (angular) as shown in Figs. 6.4a
and 6.4b.
133
Figure 6.6 Vibration of fine sand at pre-peak and residual strength states.
134
Figure 6.7 Vibration of 0.55 mm glass beads at pre-peak and residual strength states.
135
It is interesting to note that vibration at the residual strength state densifies the samples,
thus decreasing their void ratio to a certain e value that stays constant after vibration termination,
so that e < ecritical. This should lead to a strength level greater than the residual strength, but in
fact, strength reaches the critical state (residual strength) and remains the same at a constant e
value, as can be seen in Figs. 6.6 and 6.7. This phenomenon can possibly be explained in the
following way: A dense granular material sample dilates when it is sheared, thus developing a
shear zone. The thickness of this shear zone reaches a certain maximum value, which is greater
than the actual shear zone thickness required for the sample to remain at the critical state. Since
the material in the shear zone is looser than in the rest of the sample, vibration causes
compression of the shear zone portion which no longer contributes to the developed critical state.
It is assumed that the absence of post-vibrational vertical deformations (dilation) can be
attributed to the shearing that takes place when the vibration is applied. The ongoing shear
deformations keep the actual shear zone at its critical state void ratio (ecritical) during vibration,
thus preventing any further post-vibrational vertical deformations. It is also assumed that if a
sample was subjected to the same intensity of vibration at the critical state, while not being
sheared, the vibration would densify the actual shear zone to a certain void ratio (e < ecritical). This
would increase the magnitude of compression induced by vibration, thus resulting in dilation
with further shearing of the sample.
In other words, the post-vibration dilation of a granular material depends on the vibration
intensity and the shear rate of the test. This means that post-vibration dilation can be caused not
only by the termination of shearing during vibration, but also an increase in vibration intensity
and lowering of the shear rate of the test.
136
6.5.2 Post-Vibrational Strength and Volumetric Changes of Granular Materials
As a result of the observations mentioned above, additional experiments were performed in order
to investigate the impact of vibration on the post-vibrational strength and volumetric change
characteristics of granular materials. These were particularly carried out to demonstrate and
prove the assumption that the shear zone thickness in granular materials consists of two
parts/zones: one that develops before the material reaches the critical state and the second
(thinner than the initially developed thickness of the shear zone) that keeps the material in the
critical state, so that the first part/zone does not contribute to the critical state maintained by the
second thinner zone. A schematic drawing of the mentioned shear zone formation in a granular
material is given in Fig. 6.8, where A denotes the zone unaffected by the shear of the granular
medium, B is the shear zone portion that has been developed due to the shear of the material, but
does not contribute to the critical state of shear zone C, at which the actual shear takes place.
Figure 6.8 Schematic cross-section of the shear zone in a granular material.
137
Two materials, fine sand and 0.55 mm glass beads, were tested in dense and loose states
for this purpose. The vibration applied at the residual strength state had a frequency of 140 Hz
and vibration force of 7.14 N.
Dense samples were prepared in the same way as all of the previously tested samples, by
filling the shear box with a certain amount of material and compacting the material. The loose
samples were prepared by slowly filling the shear box with the granular material through a
narrow neck funnel, with no consequent compaction, in order to achieve the loosest possible
state. A total of 24 samples were tested: 12 samples of fine sand and 12 samples of 0.55 mm
glass beads. Six samples of each material, in dense and loose states, were tested at normal stress
values of 8, 23, 36, 50, 118 and 200 kPa. Unlike the previously tested samples, these samples
were not sheared during the application of vibration, that is, shearing was terminated before and
started after vibration. This was done to allow the shear zones B and C (see Fig. 6.8) to compress
equally, so that the void ratio (or density) of the sample at the shear zone is not affected by
shearing during the vibration. Four examples of the tests conducted on dense and loose 0.55 mm
glass beads and fine sand at a normal stress of 23 kPa are given in Figs. 6.9 – 6.12.
After termination of vibration, the samples of dense fine sand tested at normal stresses of 8
and 23 kPa demonstrated dilation and shear resistance greater than those at the residual strength
state. Then after a certain amount of shear deformation, the vertical displacement of the samples
reached a constant value which was also the case for the shear resistance while approaching the
residual strength value. The sample tested at σ=36 kPa did not demonstrate any changes in post-
vibrational vertical displacement. On the other hand, the samples tested at σ=50, 118 and 200
kPa after some of the shear deformations started to compress, and higher normal stresses resulted
in greater compression. This was probably due to some tilting of the loading plate, which usually
138
takes place in direct shear testing when shearing granular material at relatively high normal
stresses, especially at high shear displacements.
In the case of the dense 0.55 mm glass beads (note that vibration was applied twice to the
samples tested at σ=23, 36, 50 and 118 kPa), the samples that demonstrated post-vibration
dilation and strength increase were those tested at σ=8, 23 and 36 kPa. The samples tested at
normal stresses of 50a and 118 kPa had no post-vibrational vertical deformation changes, and the
sample tested at σ=200 kPa showed some compression close to the end of the shear test.
139
Figure 6.9 Test results of dense 0.55 mm glass beads tested at σ = 23 kPa.
140
Figure 6.10 Test results of loose 0.55 mm glass beads tested at σ = 23 kPa.
141
Figure 6.11 Test results of dense fine sand tested at σ = 23 kPa.
142
Figure 6.12 Test results of loose fine sand tested at σ = 23 kPa.
143
The test results of the loose fine sand and 0.55 mm glass beads show that there is similar
behavior after vibration termination: samples tested at low normal stresses dilated to a certain
degree with the associated changes in shear resistance.
As mentioned in the previous section, the samples of dense fine sand and 0.55 mm glass
beads neither compressed nor dilated at the pre-peak vibration application, except for one fine
sand sample tested at a normal stress of 23 kPa. This means that the vibration, with the
mentioned characteristics, cannot cause any volumetric changes in the tested granular materials
at their initial void ratio (or density). Since the same materials (with the same initial density)
were tested with the same vibration characteristics, then all the volumetric changes are due to the
dilation of the shear zone (due to shearing) and compression (due to vibration). Therefore, the
post-vibration dilation of the samples observed can only be attributed to the void ratio increase of
the shear zone due to the post-vibration shear of the material. As can be seen in Figs. 6.9 – 6.12,
the magnitude of the post-vibration dilation is less than that of the compression caused by the
vibration. Therefore, it can be inferred that the post-vibration dilation is due to the actual shear
zone C (see Fig. 6.8). This proves the previous assumption that the shear zone in a granular
material consists of two zones: B and C, as shown in Fig. 6.8.
The absence of the post-vibration dilation of the granular materials at higher normal stress
values implies that the dilation of shear zone C at higher normal stresses takes place by means of
a reduction in the void ratio of shear zone B (see Fig. 6.8).
In addition to the above described experiments, four more dense samples of the 0.55 mm
glass beads and four samples of the fine sand were sheared over a smooth glass surface in order
to provide additional support to the assumption zones A, B and C exist as outlined in Fig. 6.8.
144
The samples were tested at normal stresses of 23, 50, 118 and 200 kPa. Vibration was applied to
the samples for 30 sec at their residual strength state.
The test results of the eight samples are summarized in Table 6.3 and Figs. 6.13a and
6.13b. Two examples of the fine sand and glass beads tested at a normal stress of 50 kPa are
given in Figs. 6.14 and 6.15, respectively.
Table 6.3 Test results of 0.55 mm glass beads and sand sheared over smooth glass surface.
Material Normal Peak Residual Vib.-Res. Residual Horizontal Vertical

Stress Strength Strength Strength Strength Loss Acceleration Acceleration
σ τp τr τvr Δτ
h v
kPa kPa kPa kPa kPa
23 - 3 1.5 1.5 0.33 0.17
0.55 mm 50 6 5 3.5 1.5 0.32 0.22

Glass Beads 118 12 9 7.5 1.5 0.3 0.15
200 21.5 16 14.5 1.5 0.3 0.14
23 - 3 1.5 1.5 0.33 0.16

50 6 5.5 4 1.5 0.32 0.18
Fine Sand
118 14.5 12.5 11 1.5 0.32 0.13
200 25.5 22 20.5 1.5 0.33 0.19
As expected, the residual friction angles of the materials tested over the smooth glass
surface are much less than those of the materials themselves, which are 4.1o for the glass beads
and 6.2o for the fine sand. Both materials have no peak strength at σ = 23 kPa (see Table 6.3),
which implies that dense granular materials do not generate peak strength when sheared over a
smooth surface at low normal stress values. The peak friction angles of the glass beads and fine
sand are 5.9o and 7.3o, respectively.
145
As can be seen from Table 6.3 and Figs. 6.13a and 6.13b, vibration lowers the residual
strength of both materials by the same magnitude, Δτ=1.5 kPa, at all normal stresses. Therefore,
in this case, there are no changes in the residual friction angles of the tested materials.
(a) 0.55 mm Glass Beads

30
25
Shear Stress, kPa
y = 0.103x + 0.6
20
y = 0.072x + 1.2
15 y = 0.072x - 0.3
10
Peak
5 Residual
Vibro-residual
0
0 50 100 150 200
Normal Stress, kPa
(b) Fine Sand

30
25 y = 0.128x - 0.2
Shear Stress, kPa
y = 0.108x + 0.2
20 y = 0.108x - 1.3
15
10
Peak
5 Residual
Vibro-residual
0
0 50 100 150 200
Normal Stress, kPa
Figure 6.13 Shear strength diagrams of 0.55 mm glass beads (a) and fine sand (b).
146
The examples presented in Figs. 6.14 and 6.15 show that vibration does not generate any
compressive deformations of the samples. In fact, it caused a negligible dilation that took place
instantly with no further changes in vertical deformation. A similar behavior of vertical
deformation can be observed in the case of the glass beads tested at normal stresses of 23 and 50
kPa, as well as fine sand samples tested at σ = 23, 50 and 118 kPa. The rest of the samples did
not experience any vertical deformations upon the application of vibration.
It should also be mentioned that vibration did not cause any post-vibrational changes in the
shear resistance of all eight samples. Apparently, the densities of the dense granular materials
were not affected by the vibration, and neither were their post-vibrational residual strengths.
Therefore, it can be concluded that when a granular material is sheared over a smooth glass
surface, the void ratio of the material at the shear surface/zone remains unchanged during and
after the application of vibration. This implies that in this case, the entire material of each of the
tested samples can be classified as Zone A as described earlier in the text and presented in Fig.
6.8.
147
Figure 6.14 An example of test results of fine sand sheared over smooth glass surface.
148
Figure 6.15 Example of test results of 0.55 mm glass beads sheared over smooth glass surface.
149
6.6 Conclusions
 Vibration of sufficiently strong intensity reduces the residual friction angles of the
granular materials in this study.
 All of the tested samples (both types of materials) demonstrate compressive deformation
upon application of vibration at the residual strength state.
 Vibration of a given intensity applied at the pre-peak strength state of the granular
materials does not affect their peak strengths.
 The magnitude of the pre-peak strength loss of the granular materials caused by vibration
is generally the same at all normal stresses, and smaller (especially at higher normal
stresses) than the strength loss at the residual strength state.
 It is assumed that vibration of a greater intensity, capable of causing compaction of the
granular materials, would densify the samples, which would in turn, result in an increase
in their peak strength.
 All of the tested samples do not experience any volumetric changes due to vibration at
their pre-peak strength state, except for one sample of fine sand tested at a normal stress
of 23 kPa which showed some compression.
 Dense granular material dilates when sheared, thus developing a shear zone. The
thickness of this shear zone reaches a certain maximum value, which is greater than the
actual thickness of the shear zone required for the sample to remain at the critical state.
 Three zones are identified in a sheared granular material: A – zone unaffected by shear of
the granular medium, B – shear zone portion that has been developed due to the shear of
150
the material, but does not contribute to the critical state of shear zone C, at which the
actual shear takes place.
 The concept of the A, B and C zones correlates well with the finite element analysis
results obtained by Potts et al. (1987).
 As expected, the residual friction angles of the materials tested over a smooth glass
surface are much less than those of the materials themselves.
 Dense granular materials do not generate peak strength when sheared over a smooth
surface at low normal stress values.
 The strength loss of the granular materials caused by vibration has the same magnitude at
all normal stresses, which means that vibration does not reduce the residual friction
angles of the materials sheared over a smooth glass surface.
 When a granular material is sheared over a smooth glass surface, the void ratio of the
material at the shear surface/zone remains unchanged during and after the application of
vibration.
151
6.7 References
Abriak, N. E. and Caron J.-F. (2006). “Experimental study of shear in granular media”.
Advanced Powder Technol., Vol. 17, No. 3, pp. 297–318.
Aidanpaa, J. O., Shen, H. H. and Gupta, R. B. (1996). “Experimental and Numerical Studies of
Shear Layers in Granular Shear Cell”. J. Eng. Mech., 122, pp. 187-196.
Alshibli, K. A. and Alramahi, B. A. (2006). “Microscopic Evaluation of Strain Distribution in
Granular Materials during Shear”. J. Geotech. Geoenviron. Eng.132, pp. 80-91.
ASTM D5311. “Standard Test Method for Load Controlled Cyclic Triaxial Strength of Soil”.
Bagherzadeh-Khalkhali, A. and Mirghasemi, A. A. (2009). “Numerical and experimental direct
shear tests for coarse-grained soils”. Particuology, 7, pp. 83–91.
Bardet, J. P., and Proubet, J. (1992). “Shear-Band Analysis in Idealized Granular Material”. J.
Eng. Mech., 118, pp. 397-415.
Bora, P. K. (1984). “Shear Failure in Granular Media”. J. Eng. Mech. 110, pp. 582-598.
Borja, R. I. (2003). “Deformation bands in granular media”. Geomechanics: Testing, Modeling,
and Simulation. pp. 394-399.
Cox, M. R. and Budhu, M. (2010). “Grain Shape Quantifications and their Relationship to
Dilatancy”. GeoFlorida 2010: Advances in Analysis, Modeling & Design, pp. 540-549.
Frost, J. D., DeJong, J. T. and Recalde, M. (2002). “Shear failure behavior of granular–
continuum interfaces”. Engineering Fracture Mechanics, 69, pp. 2029–2048.
152
Frost, D., Hebeler, G. L., Evans, T. M. and DeJong, J. T. (2004). “Interface behaviour of
granular soils”. Earth and Space, pp. 65-72.
Härtl, J. and Ooi, J. Y. (2011). “Numerical investigation of particle shape and particle friction on
limiting bulk friction in direct shear tests and comparison with experiments”. Powder
Technology, 212, pp. 231–239.
Kang, D. H., Choo, J. and Yun, T. S. (2012). “Evolution of pore characteristics in the 3D
numerical direct shear test”. Computers and Geotechnics, 49, pp. 53–61.
Youd, L. T. (1968) Reduction of critical void ratio during steady-state vibration. International
Li, Y. R. and Aydin, A. (2010). “Behavior of rounded granular materials in direct shear:
Mechanisms and quantification of fluctuations”. Engineering Geology, 115, pp. 96–104.
Liu, S. H., Sun, D., Matsuoka, H. (2005). “On the interface friction in direct shear test”.
Computers and Geotechnics, 32, pp. 317–325.
Liu, S. (2010). “Failure Mechanism of Granular Soil Slopes”. Earth and Space 2010:
Engineering, Science, Construction, and Operations in Challenging Environments, ASCE,
pp. 3724-3752.
Mesarovic, S. Dj., Padbidri, J. M. and Muhunthan, B. (2013). “Micromechanics of dilatancy,
critical state and shear bands in dense granular matter”. Poromechanics V, ASCE, pp.
1067-1074.
153
Nam, S., Gutierrez, M., Diplas, P. and Petrie, J. (2011). “Determination of the shear strength of
unsaturated soils using the multistage direct shear test”. Engineering Geology, 122, pp.
272–280.
Potts, D. M., Dounias, G. T., and Vaughan, P. R. (1987). "Finite element analysis of the direct
shear box test". Geotechnique, 37(1), pp. 11-23.
Rechenmacher, A. L. (2005). “Grain scale processes associated with shear banding in sands”.
Geomechanics II, pp. 342-355.
Roscoe, K. H. (1970). “Tenth Rankine lecture: The influence of strains in soil mechanics”,
Sadrekarimi, A. (2008). “Shearing Behavior of Sands in Terms of Compressibility Mechanisms”.
GeoCongress 2008: Characterization, Monitoring and Modeling of GeoSystems. pp. 197-
204.
Widulinski, L., Kozicki, J. and Tejchman, J. (2010). “Comparative Modeling of Shear
Localization in Granular Bodies with FEM and DEM”. GeoShanghai 2010 International
Conference, Soil Behavior and Geo-Micromechanics, pp. 198-203.
154
7. Chapter 7: Summary, Conclusions and Recommendations
7.1 Summary
The focus of the accomplished research project is the influence of vibration on the strength
properties of dry granular media. The granular materials used in the experimental program are
dry and free of cementation so as to better explore the mechanism of vibration impacts on the
shear strength and deformation characteristics of the tested materials from the standpoint of
vibrational fluidization.
In this research work, a new vibrating direct shear apparatus has been designed and built to
evaluate the strength and deformation characteristics of soils under a wide range of vibrational
accelerations. The apparatus is used throughout the research project to investigate the behavior of
dry granular materials at their residual strength states under the influence of vibration with
different intensities. The impact of vibration on the normal (overburden) stress and the peak
strength of granular materials is evaluated. Experiments have also been conducted to study the
post-vibrational behavior of granular materials, as well as the particle shape effect on the
strength loss and deformation of the granular media during vibration. Finally, the behavior of the
shear zone in the granular materials before, during and after the application of vibration at the
pre-peak and residual strength states is investigated.
Chapters 3, 4, 5 and 6 are summaries of the experimental results of about 200 tested
samples with different granular materials where the new vibrating direct shear apparatus is used
with different testing modes, procedures, normal stresses and vibration intensities. A number of
plots have been presented that show the behavior of the granular materials under the impact of
155
vibration in different testing conditions. The Mohr-Coulomb equation for granular materials has
been modified to account for the effect of vibrational fluidization. A new pattern of the shear
zone in granular media has been provided which outlines the deformation and shear resistance
before, during and after the application of vibration. Finally, some valuable conclusions have
been made and will be presented in the next section.
7.2 Conclusions
The following conclusions have been made based on the results of the conducted research
project.
1. The new vibrating direct shear apparatus can be successfully used to determine the strength
and deformation characteristics of fine and granular soils (up to the size of coarse sand)
under different accelerations and frequencies of vibration.
2. Upon application of a sufficiently strong vibration at the critical strength state of a granular
material, an immediate strength loss takes place from the residual to the vibro-residual
strengths. An increase in vibration intensity results in a larger residual strength loss at a
given normal stress.
3. Sufficiently strong vibration reduces the residual friction angle of granular materials. An
increase in vibration intensity results in lower vibro-residual friction angles.
4. Initially dense or loose granular material compresses when vibration is applied at its critical
strength state. An increase in horizontal vibration acceleration results in greater compression
deformation.
156
5. Although normal stress reduction caused by vibration contributes to the residual strength
loss in granular materials, the strength loss is mainly due to the material fluidization.
6. At low vibrational accelerations, the relationship between the residual strength loss and
horizontal acceleration is nonlinear for the granular materials that have irregularly shaped
particles. The nonlinearity increases with an increase in confining (normal) stress. On the
other hand, the relationship between strength loss and horizontal acceleration is linear for
glass beads with spherical shaped particles.
7. Besides reducing the friction angles of granular materials, vibration causes the intersection
of the shear strength envelop with the normal stress axis at a certain vibro-fluidizational
limit value, σf, which increases with an increase in vibration intensity. This implies that at a
given vibration acceleration, a granular material can fluidize at normal stresses up to the σf
value, and above this σf value, it remains in a solid state. Thus, the Coulomb’s equation for
dry granular material,  = σ tan φ, can be written as:  = (σ + σf) tan φvr, where φvr is the
vibro-residual friction angle of the material.
8. In a wide range of horizontal vibration acceleration values (0 < h < 4.5), the reduction of
the residual friction angles of the tested granular materials with an increase in vibration
intensity is practically linear. The only noticeable deviation from the linear pattern of the
relationship between φ and h is observed at h values less than 1. Thus, the φ vs. h plots
can be used for the approximation of the changes in residual friction angle of sand sized dry
granular materials (with residual friction angles between 21o to 32o) when subjected to
vibration in the given range of horizontal acceleration values.
9. The plots of the relationship between σf and h for all of the tested materials are linear. At a
given vibration intensity, the glass beads have a greater magnitude of σf than the sands with
157
well rounded and angular particle shapes, the plots of which practically coincide. It can be
assumed that there is a certain level of roundness, above which, the σf vs. h plot will shift
from that of the sands to the one for glass beads.
10. The largest amount of compressive deformation of the granular materials induced by
vibration is observed at the lowest normal stress. An increase in normal stress leads to a
compression reduction. This phenomenon is also apparent from the magnitude of the post-
vibration peak of the shear resistance plots. Higher normal stresses mean lower post-
vibration peaks.
11. When vibration is applied at the pre-peak strength state of a dense granular material, it does
not lower the peak strength of the material. The vibration may increase the peak strength if
its intensity is strong enough to densify the material.
12. It appears that the essence of acoustic fluidization which attempts to explain the large runout
distance of big rock avalanches (Melosh, 1979), and vibrational fluidization observed in the
above mentioned experiments is the same: the resistance of granular media at the shear
interface is affected by the vibration generated by collision of the grains.
13. When a dense granular material is sheared, it dilates, thus developing a shear zone. The
thickness of this shear zone reaches a certain maximum value, which is greater than the
actual thickness of the shear zone required for the sample to remain at the critical state.
Thus, three zones are identified in a sheared granular material: A – the zone unaffected by
the shear of the granular medium, B – the shear zone portion that has been developed due to
the shear of the material, but does not contribute to the critical state of shear zone C, at
which the actual shear takes place. The concept of the A, B and C zones correlates well with
the finite element analysis results obtained by Potts et al. (1987).
158
7.3 Recommendations
Recommendations for future work are as follows.
 The effect of vibration frequency on the strength and deformation characteristics of granular
media should be experimentally investigated. It is recommended that experiments be
conducted on granular materials by subjecting them to different frequency vibrations at their
critical strength states. Depending on the technical characteristics of the vibrational actuator,
which is used as part of the vibrating direct shear apparatus, vibration should be applied in the
widest possible range of frequency (approximately 5 to 250 Hz) by keeping the vibration
amplitude constant, such that the magnitude of the vibrational accelerations is controlled by
the frequency of vibration.
 Granular materials can be tested at different moisture contents to investigate the effect of
moisture content on the strength and deformation characteristics of the granular media under
the influence of vibration.
 Numerical simulations should be carried out by using 3D DEM software to investigate the
effect of vibration on the critical void ratio and shear strength of granular materials, and relate
the results to the critical state theory of granular materials.
o The numerical model can be calibrated by using the experimental results of the 0.55 mm
glass beads in this work (see Chapters 4 and 5).
o Numerical simulations can be implemented with a wide range of vibrational accelerations,
amplitudes and frequencies.
o Numerical simulations can be carried out to quantify and visualize the A, B and C shear
zones (see Chapter 6) in granular media.
159
o Numerical simulations can be implemented to evaluate the effect of particle size and shape
on the behavior of granular materials during vibration, by creating particles from combined
and different sized spheres.
160
References
Abriak, N. E. and Caron, J. F. (2006). “Experimental study of shear in granular media.”
Advanced Powder Technol., Vol. 17, No. 3, pp. 297–318.
Aidanpaa, J. O., Shen, H. H. and Gupta, R. B. (1996). “Experimental and Numerical Studies of
Shear Layers in Granular Shear Cell.” J. Eng. Mech., 122, pp. 187-196.
Alexeev, A., Royzen, V., Dudko, A., Goldshtein, A. and Shapiro, M. (2000). "Mixing and
Segregation in Vertically Vibrated Granular Layers." Solid Mechanics and Its
Applications, Volume 81, 2000, pp 129-139
Alshibli, K. A. and Alramahi, B. A. (2006). “Microscopic Evaluation of Strain Distribution in
Granular Materials during Shear.” J. Geotech. Geoenviron. Eng.132, pp. 80-91.
Arango, I. and Seed, H. B. (1974). “Seismic stability and deformation of clay slopes.” J.
Geotech. Engrg. Div., 100(2), pp. 139–156.
ASTM D5311. “Standard Test Method for Load Controlled Cyclic Triaxial Strength of Soil.”
Ayer, J. E. and Soppet, F. E. (1965/1966). “Vibratory Compaction, Part I: Compaction of
Spherical Shapes and Part II: Compaction of Angular Shapes.” J. Am. Ceramic Soc. 48,
180-183 and 49, 207-210.
Bagherzadeh-Khalkhali, A. and Mirghasemi, A. A. (2009). “Numerical and experimental direct
shear tests for coarse-grained soils.” Particuology, 7, pp. 83–91.
Bagnold, R. A. (1954). “Experiments on the Gravity-Free Dispersion of Large Solid Spheres in a
Newtonian Fluid Under Shear.” Proc. Royal Soc. London 225. pp. 49–63.
161
Bardet, J. P., and Proubet, J. (1992). “Shear-Band Analysis in Idealized Granular Material.” J.
Eng. Mech., 118, pp. 397-415.
Barkan, D. D. (1962). “Dynamics of Basis and Foundations.” Translated from Russian by L.
Bjerrum, L., Kringstad, S. and Kummeneje, O. (1961). “The Shear Strength of a Fine Sand.”
Proc. 5th Int. Conf. Soil Mech. and Found. Eng. pp 29–37.
Bora, P. K. (1984). “Shear Failure in Granular Media.” J. Eng. Mech. 110, pp. 582-598.
Borja, R. I. (2003). “Deformation bands in granular media.” Geomechanics: Testing, Modeling,
and Simulation. pp. 394-399.
Bougie, J. L. (2004). “Continuum Simulations of Fluidized Granular Materials.” The University
of Texas, Austin.
Bourzutschky, M. and Miller, J. (1995). "Granular Convection in a Vibrated Fluid." Phys. Rev.
Lett. 74, 2216-2219.
Brone, D. and Muzzio, F. J. (1997). “Size segregation in vibrated granular systems: A reversible
process.” Phys. Rev. E. 56[1], pp. 1059-1063.
Clement, E., Vanel, L., Rajchenbach, J. and Duran, J. (1996). “Pattern formation in a vibrated
granular layer.” Phys. Rev. E, 53, 2972-2975.
Cooke, W., Warr, S., Huntley, J. M. and Ball, R. C. (1996). “Particle size segregation in a two-
dimensional bed undergoing vertical vibration,” Phys. Rev. E, 53, 2812.
Cox, M. R. and Budhu, M. (2010). “Grain Shape Quantifications and their Relationship to
Dilatancy.” GeoFlorida 2010: Advances in Analysis, Modeling & Design, pp. 540-549.
162
Das, B. M. (2011). “Principles of Soil Dynamics,” Second Edition. Cengage Learning, 563.
Davies, T. R. H. (1982). “Spreading of Rock Avalanche Debris by Mechanical Fluidization.”
Rock Mechanics 15; Springer-Verlag, pp. 9-24.
Doudy, S., Fauve, S. and Laroshe, C. (1989). “Subharmonic instabilities and defects in a granular
layer under vertical vibrations.” Europhys. Lett., 8[7], 621
Evesque, P. and Rajchenbach, J. (1989). “Instability in a sand heap,” Phys. Rev. Lett., 62, 44.
Falcon, E., Wunenburger, R., Evesque, P., Fauve, S., Chabot, C., Garrabos, Y. and Beysens, D.
(1999). “Cluster formation in a granular medium fluidized by vibrations in low gravity.”
Phys. Rev. Lett., 83:440.
Fauve, S., Douady, S. and Laroche, C. (1989). “Collective behaviours of granular masses under
vertical vibrations.” J. Physique C3, 187-191.
Frost, J. D., DeJong, J. T. and Recalde, M. (2002). “Shear failure behavior of granular–
continuum interfaces.” Engineering Fracture Mechanics, 69, pp. 2029–2048.
Frost, D., Hebeler, G. L., Evans, T. M. and DeJong, J. T. (2004). “Interface behaviour of
granular soils.” Earth and Space, pp. 65-72.
Gallas, J. A. C., Herrman, H. J. and Sokolowski, S. (1992). “Convection Cells in Vibrating
Granular Media.” Phys. Rev. Lett. 69, 1371-1374.
Goldshtein, A., Shapiro, M., Moldavsky, L. and Fichman, M. (1995). “Mechanics of collisional
motion of granular materials. 2. Wave propagation through vibrofluidized granular layers.”
J. Fluid Mech., 287, p. 349
163
Gotzendorfer, A., Tai, C., Kruelle, C. A., Rehberg, I. and Hsiau, S. S. (2006). “Fluidization of a
vertically vibrated two-dimensional hard sphere packing: A granular meltdown.” Phys.
Rev. E 74, 011304.
Härtl, J. and Ooi, J. Y. (2011). “Numerical investigation of particle shape and particle friction on
limiting bulk friction in direct shear tests and comparison with experiments.” Powder
Technology, 212, pp. 231–239.
Housner, G. W. (1954). “Geotechnical problems of destructive earthquakes.” Géotechnique,
Volume 4, Issue 4, 01 December 1954, pages 153 –162.
Housner, G. W. (1959). “Behavior of structures during earthquakes.” Journal of the Engineering
Mechanics Division, ASCE, 85, No. EM-4.
Huan, C. (2008). “NMR Experiments on Vibrofluidized and Gas Fluidized Granular Systems.”
PhD Thesis, University of Massachusetts Amherst.
Hungr, O. (1981). “Dynamics of Rock Avalanches and Other Types of Slope Movements.” PhD
Thesis, University of Alberta, 506 p.
Idriss, I. M. and Boulanger, R. W. (2008). “Soil liquefaction during earthquakes.” Earthquake
Engineering Research Institute, Oakland, California.
Ingale, R. A. (2008). “Dynamics of vibrated granular matter.” PhD Thesis, The City University
of New York.
Jaeger, H. M., Nagel, S. R. and Behringer, R. P. (1996). “Granular solids, liquids, and gases.”
Rev. Mod. Phys. 68, 1259
Kang, D. H., Choo, J. and Yun, T. S. (2012). “Evolution of pore characteristics in the 3D
numerical direct shear test.” Computers and Geotechnics, 49, pp. 53–61.
164
Knight, J. B., Jaeger, H. M. and Nagel, S. R. (1993). “Vibration-induced size separation in
granular media: the convection connection.” Phys. Rev. Lett., 70, 3728.
Kutergin, V. N. (1989). “The property change patterns of clayey soils under vibration.” (in
Russian) Moscow: Nauka, 143 p.
Lan, Y. and Rosato, A. D. (1995). “Macroscopic behavior of vibrating beds of smooth inelastic
spheres.” Phys. Fluids, Vol. 7, No. 8, pp. 1818-1831.
Laroche, C., Douady, S. and Fauve, S. (1989). “Convective Flow of Granular Masses under
Vertical Vibration.” J. Physique 50, 699-706.
Li, Y. R. and Aydin, A. (2010). “Behavior of rounded granular materials in direct shear:
Mechanisms and quantification of fluctuations.” Engineering Geology, 115, pp. 96–104.
Liffman, K., Metcalfe, G. and Cleary, P. (1997). “Convection due to horizontal shaking.” in:
(Behringer and Jenkins (1997)). 405-408.
Liu, S. H., Sun, D., Matsuoka, H. (2005). “On the interface friction in direct shear test.”
Computers and Geotechnics, 32, pp. 317–325.
Liu, S. (2010). “Failure Mechanism of Granular Soil Slopes.” Earth and Space 2010:
Engineering, Science, Construction, and Operations in Challenging Environments, ASCE,
pp. 3724-3752.
Luding, S. (1995). “Granular materials under vibration: simulations of rotating spheres.” Phys.
Rev. E, 52:4442.
Maslov, N. N. (1959). “Stability Conditions of Saturated Sands.” (Russian: Usloviya
Ustoychivosti Vodonasishennikh Peskov). M. – L. Gosenergoizdat. 328 p.
165
Meehan, C. L., Boulanger, R. W. and Duncan, J. M. (2008). "Dynamic Centrifuge Testing of
Slickensided Shear Surfaces," Journal of the Geotechnical Engineering Division, American
Society of Civil Engineers (month), Vol. 134, No. 8, pp. 1086-1096
Melhus, M. F. (2011). “Effects of Noise and Vibration on the Solid to Liquid Fluidization
Transition in Small Dense Granular Systems under Shear.” PhD Thesis. Northwestern
University, Evanston, Illinois.
Melo, F., Umbanhowar, P. and Swinney, H. L. (1993). “Transition to parametric wave patterns in
a vertically oscillated granular layer.” Phys. Rev. Letts., 72, 172.
Melo, F., Umbanhowar, P. and Swinney, H. L. (1995). “Hexagons, kinks, and disorder in
oscillated granular layers.” Phys. Rev. letts., 75, 3838.
Research, v. 84: pp. 7513-7520.
Melosh, H. J. and Girdner, K. K. (1995), “Rheology of vibrated granular materials: Application
to long-runout landslides.” EOS, 76.
Melosh, H. J., (1996). “Dynamical weakening of faults by acoustic fluidization.” Nature, 379,
pp. 601-606.
Mesarovic, S. Dj., Padbidri, J. M. and Muhunthan, B. (2013). “Micromechanics of dilatancy,
critical state and shear bands in dense granular matter.” Poromechanics V, ASCE, pp.
1067-1074.
Meschyan, S. R. (1978). “Short-term and long-term strength of clayey soils.” (in Russian)
Moscow, Nedra, 207 p.
166
Meschyan, S. R. (1992). “Rheological processes in clayey soils.” (in Russian) Yerevan,
“Hayastan”, 395 p.
“Granular Friction, Coulomb Failure, and the Fluid-Solid Transition for Horizontally
Shaken Granular Materials.” Physical Review E - Statistical, Nonlinear, and Soft Matter
Physics.
Moon, S. J., Swift, J. B., and Swinney, H. L. (2004). “Role of friction in pattern formation in
oscillated granular layers.” Phys. Rev. E, 69(031301):1.
Mujica, N. and Melo, F. (1998). “Solid-liquid transition and hydrodynamic surface waves in
vibrated granular layers.” Phys. Rev. Lett., 80:5121.
Nam, S., Gutierrez, M., Diplas, P. and Petrie, J. (2011). “Determination of the shear strength of
unsaturated soils using the multistage direct shear test.” Engineering Geology, 122, pp.
272–280.
Pak, H. K. and Behringer, R. P. (1993). “Surface waves in vertically vibrated granular
materials.” Phys. Rev. Lett. 71, 1832
Pak, H. K., Van Doorn, E. and Behringer, R. P. (1995). “Effects of ambient gases on granular
materials under vertical vibration.” Phys. Rev. Lett., 74, 4643.
167
Pokrovsky, G. I., Ehrlich, A. A., Laletin, N. V. and Lush, F. A. (1934). “New Methods of
Investigation of the Compressibility and Internal Friction in Soils.” Vestnik Voyenno-
Poschel, T. and Rosenkranz, D. E. (1998). “Experimental study of horizontally shaken granular
matter – the swelling effect.” Lecture Notes in Physics, Volume 503, 1998, pp 96-109
Potts, D. M., Dounias, G. T., and Vaughan, P. R. (1987). "Finite element analysis of the direct
shear box test." Geotechnique, 37(1), pp. 11-23.
Pyke, R., Seed, H. B. and Chan, C. K. (1975). “Settlement of sands under multidirectional
shaking.” Journal of Geotechnical Engineering Division, ASCE, 101, GT-4, pp.379-398.
Rechenmacher, A. L. (2005). “Grain scale processes associated with shear banding in sands.”
Geomechanics II, pp. 342-355.
Geotechnical Engineering, 116(5), pp. 740–759
Richart, F. E. (1970). “Vibrations of Soils and Foundations.” Prentice-Hall, 414 p.
Ristow, G. H., Strasburger, G. and Rehberg, I. (1997). “Phase diagram and scaling of granular
materials under horizontal vibrations.” Phys. Rev. Lett., 79(5):833.
Robertson, P. K. and Campanella, R. G. (1985). “Liquefaction potential of sands using the
CPT.” Research Report R69-15, Dept. of Civil Engineering, Massachusetts Institute of
Technology, Cambridge, Mass.
Robertson, P. K. and Campanella, R. G. (1986). “Estimating liquefaction potential of sands
using the flat plate dilatometer.” Geotechnical Testing Journal, 9, No. 1, pp. 38-40.
168
Roscoe, K. H. (1970). “Tenth Rankine lecture: The influence of strains in soil mechanics.”
Sadrekarimi, A. (2008). “Shearing Behavior of Sands in Terms of Compressibility Mechanisms.”
GeoCongress 2008: Characterization, Monitoring and Modeling of GeoSystems. pp. 197-
204.
Sangroya, R. and Choudhury, D. (2013). “Stability Analysis of Soil Slope Subjected to Blast
Induced Vibrations Using FLAC3D.” Geo-Congress 2013, pp. 472-481
Savage, S. B. (1988). “Streaming motions in a bed of vibrationaly fluidized dry granular
material.” J. Fluid Mech., 194:457.
Savchenko, I. A. (1958). “Effect of Vibrations on Internal Friction in Sand.” Symposium, Soil
Dynamics, Moscow, Gosstroiizdat, No. 2.
Seed, H. B. (1966). “A method for earthquake resistant design of earth dams.” Journal of the Soil
Mechanics and Foundation Division, ASCE, 92, No. SM-1, pp. 13-41.
Seed, H. B. and Idriss, I. M. (1982). “Ground motions and soil liquefaction during earthquakes.”
Monograph series, Earthquake Engineering Research Institute, Berkeley, California.
Seed, H. B. and Idriss, I. M. (1983). “Evaluation of liquefaction potential using field
performance data.” Journal of Geotechnical Engineering, ASCE, 109, No. 3, pp. 458-482.
Shibata, T. and Yukitomo, H. (1969). “Liquefaction Process of Sand during Cyclic Loading.”
Soils and Foundations, Vol. 3, pp. 54-69.
Shibata, T. and Yukitomo, H. (1970). “Shear strength of sand under a vibrating load.” Bulletin of
the Disaster Prevention Research Institute, 19(3): pp. 27-37.
169
Sornette, D. and Sornette, A. (2000). “Acoustic Fluidization for Earthquakes?” Bulletin of the
Seismological Society of America, 90, 3, pp. 781-785.
Sunthar, P. and Kumaran, V. (2001). “Characterization of the stationary states of a dilute
vibrofluidized granular bed.” Phys. Rev. E, 64:041303.
Tennakoon, S. G. K. and Behringer, R. P. (1998). “Vertical and horizontal vibration of granular
materials: Coulomb friction and a novel switching state.” Phys. Rev. Lett., 81(4):794–797.
Toyota, H., Towhata, I., Imamura, S. and Kudo, K. (2004). "Shaking Table Tests on Flow
Dynamics in Liquefied Slope." Soils and Foundations, Vol. 44, No. 5, pp. 67 – 84.
Tsimring, L. and Aranson, I. (1997). “Localized and cellular patterns in a vibrated granular
layer.” Phys. Rev. Lett., 79(2): 213–216.
Umbanhowar, P. B., Melo, F. & Swinney, H. L. (1996). “Localized excitations in a vertically
vibrated granular layer.” Nature (London) 382, 793
Warr, S., Huntley, J. M. and Jacques, G. T. H. (1995). “Fluidization of a two-dimensional
granular system: Experimental study and scaling behavior.” Phys. Rev. E, 52, 5583-5595.
Wassgren, C. R. (1997). “Vibration of granular materials.” PhD Thesis, California Institute of
Widulinski, L., Kozicki, J. and Tejchman, J. (2010). “Comparative Modeling of Shear
Localization in Granular Bodies with FEM and DEM.” GeoShanghai 2010 International
Conference, Soil Behavior and Geo-Micromechanics, pp. 198-203.
170
Youd, L. T. (1968). “Reduction of critical void ratio during steady-state vibration.” International
Youd, L. T. (1970). “Densification and Shear of Sand during Vibration.” ASCE, Journal of the
Soil Mechanics and Foundations Division, Vol. 96, No. 3, May/June 1970, pp. 863-880.
Zik, O., Stevans, J. and Rabin, Y. (1992). “Mobility of a sphere in vibrated granular media.”
Europhys. Lett., Vol. 17, No. 4, pp. 315-319.
171
APPENDIX
Part 1
The plots provided in this section were used in Chapter 3. Following are the test results of 8
samples of fine sand tested in strain-controlled mode at 8.4 kPa, 23.2 kPa, 36.3 kPa, 50 kPa, 77.3
kPa, 118.2 kPa and 200 kPa normal stresses, as well as 5 samples of fine sand tested in stress-
controlled mode at 5 different shear loads, 50 kPa normal stress, 140 Hz vibration frequency and
constant vibration intensity.
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
Part 2
samples of fine sand, 20 samples of coarse sand, 20 samples of 0.1 mm glass beads and 20
samples of 0.55 mm glass beads tested in strain-controlled mode at 23 kPa, 50 kPa, 118 kPa and
200 kPa normal stresses, 5 different vibration intensities and 140 Hz vibration frequency.
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
Part 3
samples of fine sand, 18 samples of coarse sand and 24 samples of 0.55 mm glass beads tested in
strain-controlled mode at 8 kPa, 23 kPa, 36 kPa, 50 kPa, 118 kPa and 200 kPa normal stresses,
different vibration intensities and 60 Hz vibration frequency.
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
Part 4
The plots provided in this section were used in Chapter 6. Following are the test results of a few
sample sets.
Set 1: 4 samples of fine sand and 4 samples of 0.55 mm glass beads tested in strain-controlled
mode over a smooth glass surface at 23 kPa, 50 kPa, 118 kPa and 200 kPa normal stresses,
constant vibration intensity and 140 Hz vibration frequency.
Set 2: 4 samples of fine sand and 4 samples of 0.55 mm glass beads tested in strain-controlled
mode at 23 kPa, 50 kPa, 118 kPa and 200 kPa normal stresses, constant vibration intensity and
140 Hz vibration frequency applied at the pre-peak and residual strength states of the samples.
Set 3: 12 samples of loose and dense fine sand samples, as well as 12 samples of loose and dense
0.55 mm glass beads tested in strain-controlled mode at 23 kPa, 50 kPa, 118 kPa and 200 kPa
normal stresses, and no shearing during the application of vibration having constant intensity and
140 Hz frequency.
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369

Taslagyan Karen 201409 PHD

Uploaded by

Copyright:

Available Formats

Taslagyan Karen 201409 PHD

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Taslagyan Karen 201409 PHD

Uploaded by

Copyright:

Available Formats

The Shear Strength of Granular Soils under the Influence of Vibration

A thesis submitted in partial fulfillment of the requirements for the degree of

Department of Civil and Environmental Engineering

© Karen Taslagyan, 2014

structure. Therefore, it is of paramount importance to predict and consider the effects of

sufficiently large shear deformations.

This thesis presents a summary of an experimental program performed to investigate the

strength states is investigated.

gratitude to my co-supervisor, Professor Norbert R. Morgenstern, for his important comments

always been encouraging and stimulating.

building the laboratory apparatus used in this research.

Futures (Graduate Student Scholarship).

1.2 Research Objectives ............................................................................................................. 3

1.3 Scope of the Thesis .............................................................................................................. 4

1.4 Organization of the Thesis ................................................................................................... 6

1.5 References ............................................................................................................................ 8

2. Chapter 2: Literature Review ..................................................................................................... 9

3.2 Introduction ........................................................................................................................ 17

3.3 Strain controlled vibrating direct shear apparatus .............................................................. 20

3.4 Stress controlled vibrating direct shear apparatus .............................................................. 24

3.5 Testing of the Apparatus .................................................................................................... 27

3.5.1 Strain Controlled Testing Mode .................................................................................. 27

3.5.2 Stress Controlled Testing Mode .................................................................................. 32

3.6 Conclusions ........................................................................................................................ 35

3.7 References .......................................................................................................................... 35

4.2 Introduction ........................................................................................................................ 37

4.3 Previous Studies on Vibrational Loading on Soils ............................................................ 39

4.5 Tested Materials ................................................................................................................. 46

4.6 Test Results under Vibration .............................................................................................. 51

4.6.2 The Effects of Acceleration on Residual Strength Loss .............................................. 68

4.7 Conclusions ........................................................................................................................ 72

4.8 References .......................................................................................................................... 73

5. Chapter 5: Vibrational Fluidization of Granular Media ........................................................... 78

5.2 Introduction ........................................................................................................................ 78

5.3 Testing Equipment and Procedures .................................................................................... 80

5.4 Tested Materials ................................................................................................................. 83

5.5 Test Results ........................................................................................................................ 86

5.6 Conclusions ...................................................................................................................... 113

5.7 References ........................................................................................................................ 115

6.2 Introduction ...................................................................................................................... 120

6.3 Testing Equipment and Procedures .................................................................................. 124

6.4 Tested Materials ............................................................................................................... 127

6.5 Test Results ...................................................................................................................... 129

6.6 Conclusions ...................................................................................................................... 150

6.7 References ........................................................................................................................ 152

7. Chapter 7: Summary, Conclusions and Recommendations ................................................... 155

7.2 Conclusions ...................................................................................................................... 156

7.3 Recommendations ............................................................................................................ 159

References ................................................................................................................................... 161

Part 2 ...................................................................................................................................... 187

Part 3 ...................................................................................................................................... 268

Part 4 ...................................................................................................................................... 329

Figure 3.1 Schematic of the vibrating direct shear apparatus ....................................................... 19

1.1 Statement of Problem

density and mineralogy of the soil particles.

effect of normal stress on the vibrational fluidization of granular materials.

vibration. In particular, it is specifically interesting to determine the effect of vibration on the

sufficiently large shear deformations.

The strength of a granular material is determined by the strength characteristics of the