Slaking Characteristics of Geomaterials in Direct Shear Test

Slaking characteristics of geomaterials in direct shear test
一面せん断試験による地盤材料のスレーキング特性の解明
Keshab Sharma
ケシャブシャーマ
September 2012
Slaking characteristics of geomaterials in direct shear test
一面せん断試験による地盤材料のスレーキング特性の解明
By
Keshab Sharma
ケシャブシャーマ
Supervisor: Associate Professor Takashi Kiyota
Co-supervisor: Professor Junichi Koseki
A Thesis submitted to Graduate School of Engineering, Department of Civil

Engineering, The University of Tokyo for the partial fulfillment of the
requirements for the degree of
Masters of Engineering
Department of Civil Engineering

Graduate School of EngineeringThe University of Tokyo
Japan
September 2012
ABSTRACT
Numerous slope stability and other geotechnical problems worldwide have been
reported due to the presence of mudstones. Similarly, in the world, infrastructure
developments have been extended to hillsides and mountainous areas because of population
growth, economics needs and other constraints. Soft sedimentary rock named mudstone is
the most often encountered geomaterials during the major construction works that are
undertaken in those areas.
In the past decade, there has been a significant increase in knowledge with respect to
the mechanics of geomaterials. Several fundamental issues have been solved, and important
achievements have been made in certain areas. However, most of these achievements are
either deal with rock or with soil because most of geotechnical engineers are generally used
to viewing geomaterials either as soils whose behavior is highly susceptible to the fabric
and water content of the intact material, or as a rock, with engineering behavior primarily
controlled by fissures and joints. However, mudstones have two inherent properties: firstly,
they are intermediate in behaviour between rock and soil, and secondly, they tend to
transgress from rocklike to soil-like materials within relatively short time-frames. So,
present experimental and theoretical methods used in geotechnical engineering practice are
not inadequate for assessing the stability of mudstones. In general, the laboratory
investigations on durability characteristics of mudstones are only made through the general
slaking test. However, studies examining the effects of slaking on strength and deformation
behaviour of mudstones are sparse.
This research began by investigating the landslides to explore the probability of
slaking induced landslides in soft sedimentary rock formations. Significant numbers of
landslide events were found to be occurred after moderate rainfall, sometimes followed by
drought. Similarly, intensive review of the studies that have been previously undertaken to
study the slaking of mudstones was done to ascertain the current state of knowledge
regarding this process.
At the beginning of the experiment, the conventional slaking tests were performed to
classify the tested material on the basis of slaking durability. In order to examine slaking
effects on strength and deformation characteristics of crushed mudstone (Hattian Bala and
Ishikawa), a series of direct shear tests were conducted on the crushed mudstone by
i
simulating cyclic wetting and drying under different stress conditions by using a modified
direct shear apparatus. In addition, a series of the monotonic loading tests on dry and
saturated specimens were also performed to compare strength and deformation
characteristics with those of the cyclic wetting and drying creep test. Similar tests were also
conducted on conventional granular material such as Chiba gravel and Toyoura sand and
Glass beads for the comparison of stress-strain behaviour after saturation and cyclic wetting
and drying with crushed mudstones. Sieve analysis was performed after each experiment
and degradation index was used to quantify the particle crushing due to slaking.
The strength and deformation characteristics of crushed mudstones under dry,
saturated and cyclic wetting and drying conditions were compared with standard granular
soils such as Silica sand (Toyoura), crushed sand stone (Chiba gravel) and Glass beads as
well. The saturated and the one with cyclic wetting and drying mudstone specimens exhibit
largely different stress-displacement features from those for dry specimens while the
difference in stress-strain-volume change behaviours between dry and saturated conditions
is insignificant for both Toyoura sand and Chiba gravel
The effects of stress ratio, density of specimen, initial water content before wetting,
slaking index and number of cyclic wetting and drying on slaking were investigated. It was
concluded that the slaking induced shear displacement is the function of various factors
such as stress ratio, density of specimen, initial water content before wetting, slaking index
and number of cyclic wetting and drying. Drying induced slaking also observed when water
content of the mudstone specimens became smaller than certain threshold value of water
content.
Key Words: Direct shear; Slaking; Cyclic wetting and drying; Stress ratio; Particles
crushing, Shear strength and deformation
ii
ACKNOWLEDGEMENTS
First, I would like to express my deepest gratitude to Associate Professor Takashi

Kiyota for his guidance, encouragement and assistance throughout the period of this
research for Master’s Thesis. I thank him for his seemingly infinite patience, and for always
being there when I needed him the most. He always provided me unflinching
encouragement and support in various ways. He not only guides for my research but also
cares and helps me as my parents in my daily life in Japan.
I would like to thank my co-supervisor Professor Junichi Koseki for his valuable
suggestion and inspiration. He gave me confidence to wrap the master thesis.
I am equally thankful to Professor Kazuo Konagai for his constant support and
caring professor, always asking us if we had any problems.
I am extremely obliged Dr. Hiroyuki Kyokawa for his kind help and direction,
which has helped me in conducting my research. He always helps me to learn more and
more about soil mechanics. I am deeply indebted to my colleagues Daisuke Okuno and
Wataru Munekata who kindly made me familiar to the direct shear apparatus.
I would like to thank Takesi Sato for his kind help to maintain of apparatus. I am
obliged to Hiroko Takashaki and Toshihiko Katagiri for their devoted help and managing
everything that I needed for my research.
I gratefully acknowledge Jina Lee for her crucial contribution from the beginning of
my thesis and friendly behaviour. I would like to express my sincere thanks to all members
of Kiyota and Konagai Laboratory for their direct and indirect co-operation, advices and
help during the period of my Master’s study and thesis work.
I am extremely obliged to Toru Asakura who took care of all the administrative
matters during my stay. You are very kind and cooperatives. You always take care of
foreign students and translate Japanese into English for convenience of us. I always miss
you wherever I am.
Further appreciation is also extended to the Kishi laboratory, IIS, for providing the
equipments to perform X-ray diffraction power test.
Many thanks to all the Nepalese colleagues of University of Tokyo for their
cooperation and kind supports. I am very grateful for my two year at University of Tokyo,
made possible by the scholarship funded by the Asian Development Bank. I would like to
iii
acknowledge members of Foreign Student Office (FSO) for their help and cooperation
before came to Japan as well during our life in Japan.
Finally, much appreciation goes to Mom and Dad, and the rest of my family as well
relatives; their support and belief in me have always inspired me in my endeavors.
iv
TABLE OF CONTENTS
ABSTRACT.............................................................................................................................i
ACKNOWLEDGEMENTS.................................................................................................iii
TABLE OF CONTENTS......................................................................................................v
1. INTRODUCTION……………………...……………………………………………..1.1
1.1. General……………………………………...…………………….………….……1.1
1.2. Typical geotechnical problems…………....…………...……………………….…1.1
1.2.1. Landslide……………………….............………………….....................….1.2
1.2.2. Earth embankments and other structures……………………...……………1.3
1.3. Problem statement and its significance………………..……………………..……1.5
1.4. Objectives and aims…………………………..…………………………..……….1.7
1.5. Organization of dissertation……....……………………..…….………..…………1.8
2. CASE STUDY ON RAINFALL INDUCED LANDSLIDES....................................2.1

2.1. General………………………………...…………………………………………..2.1
2.2. Infinite slope analysis…………………………………..…………………..……..2.2
2.3. Landslide events collection………………………………...……...………………2.6
2.3.1. Geology…………………………..……………………………...…………2.7
2.3.2. Precipitation………………………...…………………………...………….2.9
2.4. Distribution of landslides………………………...……………………………....2.12
2.4.1. Geology………………………..………………………………………….2.12
2.4.2. Precipitation……………………...…...……...……………………………2.13
2.5. Conclusions…………………………….………...………………………………2.15
3. LITERATURE REVIEW………………………..…………………………………..3.1
3.1. General…………………...……………………………………………………..…3.1
v
3.2. Slaking mechanism and process……………………..……………………………3.1
3.3. Measurement of slaking………………………….…………..……………………3.6
3.4. Swelling and Shrinkage of mudstones…………………...……………………....3.14
3.5. Effect of wetting and drying on the properties of geomaterials…...……..……...3.20
3.6. Strength and deformation characteristics of slakable materials………...….…….3.26
3.7. Conclusion…………..………….………………………………………………..3.33
4. EXPERIMENTAL SETUP……………………..……………………...…………….4.1
4.1. General…………………………………...…………………………………….….4.1
4.2. Materials……………………………..……………………………...…………….4.1
4.2.1. Sampling site…………………………..…………………………….……..4.2
4.2.2. Physical properties of test materials…………………..……………………4.4
4.2.3. General slaking test……………………..……………………………...…..4.8
4.2.4. Specimen preparation…………………..……………………………..…..4.12
4.2.5. Sieve analysis…………………………..……………………………...….4.12
4.2.6. X-ray diffraction (XRD) powder test…………………………..…………4.15
4.3. Modified direct shear apparatus……………………...………………………..…4.15
4.3.1. Application of direct shear……………………..…………………………4.16
4.3.2. Outline of apparatus…………………...…………………..………………4.17
4.3.3. General description of the apparatus…………………..………………….4.19
4.3.4. Moisture sensor………………………...…………………….……………4.22
4.4. Direct shear test procedure……………………..………………………………..4.27
4.4.1. Initial setting and loading…………………..…..…………………………4.27
4.4.2. Wetting…………………..………………………………………………..4.32
4.4.3. Drying………………………...……………………………………….…..4.33
5. TEST RESULTS…………………………………..………………………………….5.1
5.1. General……………………………...……………………………………………..5.1
5.2. General slaking test…………………………………...…………………………...5.2
vi
5.2.1. Slaking index test………………………………...……………..…………..5.2
5.2.2. Static slaking test………………………………...…………………..……..5.4
5.2.3. Slaking ratio test……………………..……………………………………..5.5
5.2.4. Water absorption………………………...……….…………………………5.5
5.3. Direct shear test with cyclic wetting and drying……………………...…………...5.6
5.3.1. Hattian Bala…………………..…………………………………………….5.9
5.3.2. Ishikawa………………...…………………………………………………5.28
5.3.3. Glass beads……………………..…………………………………………5.35
5.3.4. Chiba gravel and Toyoura sand…………...……………………………...5.39
5.4. Particle crushing and Degradation index……………...…………………………5.47
5.5. X-ray diffraction (XRD) powder test ……………………………...…………….5.55
6. RESULT ANALYSIS AND DISCUSSION……………………...………………….6.1

6.1. General ………………………...………………………………………...………..6.1
6.2. Effect of stress ratio (R)…………………………………………...……………....6.1
6.3. Effect of initial water content(W)……………………………………………..….6.6
6.4. Effect of slaking index (SI)…………………………...……………………...……6.8
6.5. Effect of density (ρ)…………………………..……………………………….…6.12
6.6. Effect of number of cyclic wetting and drying (N)……………..……………….6.12
6.7. Summary ……………………...………………………………………………....6.15
7. CONCLUSIONS AND RECOMMENDATIONS………...………………………...7.1
7.1. General………………………………...……………………..……………………7.1
7.2. Conclusions……………………...……………..………………………………….7.1
7.3. Recommendations…………...…………………………………………………….7.4
REFERENCES……………………………………………………………………………R1
APPENDIX 1……………………………………………...………………..……………..A1
vii
CHAPTER 1
INTRODUCTION
1.1 General
Geomaterials are defined as any material from geological origin. Geomaterials are of
enormous economic importance to the global construction industry. Similarly, in the world,
infrastructure developments have been extended to hillsides and mountainous areas because
of population growth, economics needs and other constraints. Soft sedimentary rock is the
most often encountered geomaterials during the major construction works that are
undertaken in those areas, involving deep excavations, cuts and fills etc. With this massive
infrastructure development on such young and fragile geology, a large number people are
being exposed to typical geotechnical failures which can be triggered by either earthquake
or rainfall or both. Where large scale works are to be constructed on the soft sedimentary
rock, it is necessary to understand the physical and mechanical properties of the mudstones
underlying the foundations and to decide whether the ground is suitable for construction for
long term. Similarly, we need to carry out the stability of the surrounding mudstones slope
if any.
In the past decade, there has been a significant increase in knowledge with respect to
the mechanics of geomaterials. Several fundamental issues have been solved, and important
achievements have been made in certain areas. However, most of these achievements are
either deal with rock or with soil because most of geotechnical engineers are generally used
to viewing geomaterials either as soils whose behavior is highly susceptible to the fabric
and water content of the intact material, or as a rock, with engineering behavior primarily
controlled by fissures and joints. However, soft sedimentary rocks such as mudstone, shale
etc. have two inherent properties: firstly, they are intermediate in behaviour between rock
and soil, and secondly, they tend to transgress from rocklike to soil-like materials within
relatively short time-frames. The behaviour of soft sedimentary rocks is totally dependent
on their environment that it experiences. Among different types of soft sedimentary rocks,
1.1
Chapter 1 Introduction
mudstone is the most sensitive against to surrounding environment. So, the detailed
knowledge on the change in behaviour of mudstones due to the surrounding environment is
becoming essential in geotechnical engineering practice for sustainable development and
better serviceability.
Because of the periodic fluctuation of ground water or reservoir level due to distinct
seasonal changes, geomaterials often suffered from alternate cyclic wetting and drying. In
case of mudstones, the successive cycles of wetting and drying generate a significant
physical degradation that takes place in very short periods of time. This phenomenon known
as slaking becomes seriously unfavorable for long term stability of natural and artificial
slopes, settlement of embankments, bearing capacity of foundation and strength of
pavement etc. The strength and deformation behaviour of such geomaterials becomes very
complex as different factors such as the water content before wetting, the number
successive cyclic wetting and drying, density of geomaterials and finally particles crushing
largely affects the overall behaviour. So, many geotechnical problems specifically on
mudstones have demonstrated that conventional methods typically used in geotechnical
engineering practice are not adequate for stability analysis of such unique geomaterials.
1.2 Typical geotechnical problems

1.2.1 Landslide
In 1999, the Tsaoling landslide was triggered by the Chi Chi earthquake in Taiwan.
However, the dip for the slip surface is too
gentle (about 140 ) to generate a landslide,
even if there were an earthquake tremor,
because the angle of internal friction at the
peak shear strength of landslide material
(shale and sandstone) in normal condition
is well enough to resist slide ( Chigira et.
al; 2003, Towhata et al., 2002; ). Shale is
weathered by slaking (Fig. 1.1), which is
assumed to be one of the major causes of
Fig.1.1. Slaking of the shale intercalated with
intermittent retrogressive development of sandstone which is fragmented but sandstone is
the landslides. not. (Chigira et. al., 2003)
1.2
Southwest China’s Guizhou Province experienced its worst drought in this century
in 2009/2010 as shown in Fig. 1.2. The drought had cut drinking water supplies to five
million people, and more than two million animals. Similarly, sixty percent of agricultural
land has been hit. A landslide occurred on 29th March, 2010 just after moderate rainfall
(about 15 mm/day according to TRRM data) followed by the worst drought in a century
(Fig. 1.3). Similarly, many more rainfall induced landslides occurred in 2010 in Guizhou
Province, China. However, the intensity and amount of rainfall was not higher than average
annual rainfall.
Precipitation in mm/day
140
120
Landslide Event date
100
80
60
40
20
0
29-Mar
2005
2006
2007
2008
2009
2010
Photo: http://feww.wordpress.com/2010/03/18/worst-
ever-drought-in-sw-china/ Time
Fig.1.2. Dry Agriculture farm due to worst Fig. 1.3 Precipitation in Guizhou from TRMM
drought
A large number of dry landslides occur every year. However most of these dry
landslides are out of scope of researchers or the media. These dry landslides are of shallow
depth and small in size. Because of their smaller size, generally no causalities and huge
economic loss are encountered. That is why these landslides are out of scope of many
researchers and media. An unexpected dry landslide occurred along the approach road to the
Tamakoshi hydropower project, Dolakha, Nepal on 12th December, 2011 as shown in figure
1.4. According to Nepalese media, no rainfall occurred and no construction activities were
done at that time in that landslide area. December is one of the driest months in context of
Nepal (Fig. 1.5). The cause of this dry landslide is still under investigation.
1.2.2 Earth embankment and other structures
The Ataturk Dam, Turkey is the fourth largest clay cored rock-fill dam in the world.
It was constructed in 1990. When the reservoir level started to rise, settlement problems
started to occur along the crest reaching considerable level (Cetin et al., 2000). Both vertical
and horizontal displacements are still taking place under more or less constant loading
1.3
conditions (Malla et. al., 2007). The vesicular basal rock used in the cross-section of the
dam was found to be weathered as shown in Fig. 1.6. However, the displacement rates are
gradually slowing down with time.
An earthquake of magnitude 6.5 occurred on 11th August 2009 in the Suruga Bay,
Japan. A large number of collapses occurred along Tomei expressway (Fig. 1.7). Bulging of
wall was also noticed before the earthquake possible due to the expansion of backfill
materials. Variation of ground water level or percolation of water may have accelerated the
slaking of backfill materials. Slaking reduced the strength of backfill materials and
ultimately was found to be one of the major causes of the damage of Tomei Expressway
(Takagi et al., 2010).
Fig.1.7. Collapsed highway embankment along Fig. 1.8 Slaked Mudstone of Muree
Tomei expressway, Japan (Takagi et al., 2010) formation around the middle portion
of landslides dam (Kiyota et al., 2011)
A huge landslide dam was formed by the 2005 Kashmir earthquake (7.6 Mw) in
Pakistan. The dam was formed of mainly
crushed mudstones and sandstones which
can easily slake (Fig. 1.8). The dam was
breached in 2010 just after moderate
rainfall followed by drought (Sattar et.
al., 2010; Kiyota et. al., 2011). These are
only few representations of typical
geotechnical failure. As already
mentioned, it is quite understandable that Fig.1.6. Weathered vesicular basalt blocks used
the geotechnical analysis and design in the rock-fill section of dam (Cetin et al., 2000)
1.4
practice for ordinary geomaterials rely on experience obtained from past failures.
Conventional simplified theories may not work well for such most abundant lithologies like
mudstone.
1.3 Problem statement and its significance

Figure 1.9 shows the
increment shear displacement
12 IM : Ishikawa Mudstone
Increment shear deformation, s (mm)

of various geomaterials due to HM: Hattianbala Mudstone
CG : Chiba Gravel
10 TS : Toyoura Sand IM
wetting under constant stress GB : Glass Beads
8
condition. From the Fig. 1.9, it
is clearly seen that the impact 6
of wetting upon shear 4
displacement is more 2 Saturation starting point HM CG TS GB
prominent in case of 0
mudstones as compare to other -2
0 5 10 15 20 25 30 35
geomaterials such as Sand Elapsed creep time, t (hour)
stone (Chiba gravel) and Silica Fig. 1.9 Effect of slaking on creep deformation
sand (Toyoura sand) etc. The on different geomaterials with R= 0.3
shear displacement of Hattian Bala mudstone (HM) having a slaking index (JGS 2125-
2006) of 1 and Ishikawa mudstone (IM) having a slaking index (JGS 2125-2006) of 3 is
about 5 times and 70 times greater than Chiba gravel (CG) respectively, due to wetting
under the same testing conditions. The details of Hattian Bala and Ishikawa mudstone will
be explained in Chapter 4. Many other authors (Dick et. al., 1994) mentioned that
mudstones deteriorate rapidly when subjected to changes in moisture content, which is
responsible for triggering numerous slope stability problems, excessive settlements and
other problems. Most of typical geotechnical problems described in previous section are
related to the slaking of geomaterials. The low durability of certain mudstones, coupled with
their physical breakdown due to slaking, is responsible for countless slope instability
problems (Flemming et al., 1970, Regues et al. 1995), coal mine roof falls, shale
embankment failures (Bragg and Zeigler 1975; Shamburger et al. 1975; Lutton 1977; Bragg
and Zeigler 1978; Strohm 1978; Dick and Shakoor, 1992 and Bhattarai et al., 2007) and loss
of bearing capacity of foundation (Mochizuki et al., 1985; Noda and Nishi, 1988).
1.5
The cyclic wetting and drying process is very common in many parts of the world
due to distinct seasonal changes. Similarly, the global warming promotes a strong
disturbance on the world-wide weather. Subjected to this impact, both the rainfall pattern
and the environmental temperature in the mudstone areas are changing obviously. The
number of high temperature days is increasing; the rainy season is gradually shortened.
Thus, the mudstone area all over the world is now under a very serious situation which it
has never been subjected to. After a drought has ended, and an intense rain has occurred, the
new threat of slaking induced landslides becomes present, as the drought before the rainfall
accelerates the slaking of mudstones. It is necessary for the long-term evaluation of the
slopes to know how the stabilities of the mudstone slopes are changed in accordance with
the slaking process.
Among the wide variety of geomaterials, mudstone is the most common group of the
rock material found in the earth’s crust, accounting for the majority of sedimentary rock
types (Tucker, 1981 and Jakobsen and Johansen, 2000). It represents approximately two-
thirds of the stratigraphic column and one-third of the total land area.
It is necessary to evaluate the durability of such rocks against wetting and drying,
their effects on physical and mechanical properties, and ultimately the stability of the
natural and cut slopes. Slaking of geomaterials was studied by many researchers (Terzaghi
and Peck, 1948; Ladd, 1960; Nakano, 1967; Eigenbrod, 1972; Franklin and Chandra, 1972;
Moriwaki 1974; Matsukura and Yatsu, 1982; Botts, 1986 and Yoshida et al., 1991). Nakano
(1967) conducted a research on the breaking of tertiary mudstone with slaking and observed
the changes in soil properties. Most of these researches are related with the change in
physical properties such as density, particle size distribution etc of mudstone due to slaking,
do not deal with mechanical properties. Only a few researchers tried to evaluate the impact
of slaking on the mechanical properties of geomaterials. However testing conditions were
different than reality. Because of the inherent properties of mudstone as described in 1.2,
the understanding of the mechanism of soft sedimentary rocks like mudstones under cyclic
environmental changes is still lacking.
Therefore, in order to predict the long term response of various natural slopes and
other geotechnical structures, a better understanding of the strength and deformation
characteristics of geomaterials undergoing cyclic wetting and drying induced deterioration
(slaking) is essential. So, the author tried to achieve a better understanding of slakable
geomaterials through this research.
1.6
1.4 Objectives and aims

A large number of methodologies have been developed for assessing slaking of
geomaterials. The details of existing methodologies for testing the susceptibility to slaking
are explained in chapter 3. The existing methodologies only describe physical appearance or
mass loss due to slaking under no stress condition. Similarly, a rapid drying process under
very high temperature is adopted. However, slaking typically occurs under stress conditions
in the field. Similarly, the drying process is very slow and steady in the field. It can be seen
that there is a large gap between the process involved in existing methodologies and the
field conditions. So, existing methods do not provide adequate information regarding the
effects of this slaking on the strength and stress-strain behavior of geomaterials in the field.
The author has found virtually no literature involving systematic research on the effects of
slaking under conditions of confinement.
Design and analysis of engineering structures, such as dams, highway embankments,
foundations, slopes, and underground excavations etc. require shear strength parameters
(cohesion, c and internal friction angle, Φ) of mudstones. Therefore, shear strength
parameter for slakable geomaterials after certain cycles of wetting and drying are either
scare or assumed. The relationship between the slaking index ratios with the strength
parameters of slakable geomaterials is still under study. Similarly, to understand the
geological hazards which may occur in the mudstone area after long drought, it is essential
to fully understand the basic slaking characteristics of mudstone under constant stress
conditions.
The main objectives of this research can be outlined as follow.
 To analysis the slaking characteristics of crushed mudstones in term of mechanical
properties by simulating cyclic wetting and drying under constant stress conditions
with a simple apparatus.
 To compare the strength and deformation characteristics of dry and saturated
specimens with those of the cyclic wetting and drying test.
 To compare the slaking characteristics of different geomaterials having different
slaking indexes.
 To explore probable slaking induced landslide by analyzing landslide event, geology
and precipitation.
1.7
The slaking of geomaterials can be an important tool in simulating the long term
behaviours of geomaterials experiencing cyclic wetting and drying phenomena. If the
findings of this research are incorporated in conventional soil mechanics models, it can be
very useful in the analysis of long term stability of natural slope, embankments and
foundations on soft sedimentary rock formation.
1.5 Organization of Dissertation

Primarily due to the transitional nature of some geomaterials, a somewhat
multidisciplinary knowledge base is necessary before one can fully understand the slaking
characteristics of geomaterials. The author therefore conducted an extensive literary review
concentrating on slaking induced geotechnical problems, factors affecting the slaking of
geomaterials and laboratory observations of the slaking characteristics of geomaterials. One
of the intentions of the author in writing this dissertation has been to present an insightful
review of many of the problems associated with slaking of geomaterials and occurrence of
possible slaking induced landslides. Considering the scope of work and objectives of this
research, the dissertation is presented in six chapters to explain details of previous relative
research, work don and the main outcomes of this research. A comprehensive layout of the
dissertation is as follow.
Chapter 1 is an introduction to this dissertation which briefly explains the synopsis
of this research. The typical geotechnical problems, objectives and aims as well as the scope
of work and limitations are described to reveal the research curriculum.
In Chapter 2, the author collects and reviews rainfall induced slides events and
classifies them on the basis of geology and precipitation intensity. Similarly, some basic
concepts and mechanisms of rainfall induced landslides are also discussed here.
Chapter 3 gives a brief literature review on different aspects of slaking such as
causes, mechanism and consequences. The understanding on the slaking mechanism and
impacts of cyclic wetting and drying on physical properties of mudstones is presented in
this chapter. Literature review on swelling and shrinkage of mudstone due to cyclic wetting
and drying are also summarized. Finally, existing studies on the strength and deformation
characteristics of slakable materials are also discussed.
The author has performed laboratory experiments involving direct shear tests on
different geomaterials and glass beads that have undergone various cycles of wetting and
drying under confinement. The experimental setups of these experiments are presented in
1.8
Chapter 4. The description of the physical properties of samples, their origins, general
slaking tests and XRD tests are also included in this Chapter 4.
The time histories of creep deformation during cyclic wetting and drying of all tests
are presented in details in Chapter 5. Results of the monotonic shear loading tests on dry
and saturated specimens are also presented to compare the strength and deformation
characteristics with those of the cyclic wetting and drying creep tests. The particles size
distribution curve before and after each experiments are also incorporated in this chapter.
From the detailed test results presented in Chapter 5, the effects of slaking on shear
strength and deformation are evaluated in Chapter 6. Similarly, a summary of slaking effect
on different geomaterials under different stress ratios is also presented in this chapter.
Finally, the conclusions in Chapter 7 present a concise summary of the major points
of the entire dissertation.
1.9
CHAPTER 2
CASE STUDY ON RAINFALL INDUCED
LANDSLIDES
2.1 General
Landslides constitute a major threat to both lives and property worldwide especially
in regions of residual soil subjected to heavy rainfall. In areas of steep terrain that
experience prolonged intense rainfall events, the mass instability of soil slopes continue to
affect large populations and cause economic loss. Especially, in mountainous areas,
landslides are the second most destructive natural hazard after earthquakes (Runqiu, 2009;
U.S. geological survey, 2000). The casualties and economic loss due to landslides increased
greatly in the last century, and most of the landslides resulted from global climate change
and human activities (Au, 1998). They are more frequent in young tectonic mountains such
as the Himalayas of Nepal (Yamagishi et al., 2000; Lin et al., 2002; Bhasin et al., 2002),
Rocky and Andes mountain chain in American continent (Parise and Wasowski, 1999;
Collison et al., 2000; Mauritsch et al., 2000) hills of Japan and Taiwan.
Although slope failure may develop due to various factors such as earthquake
shaking, loading of slope or removal of toe etc. Almost all traditional slope stability
analyses incorporate the rainfall influence by changing the ground flow patterns with
increasing pressure heads and often a rising ground table. For elongated slopes it is often
assumed that the groundwater table is parallel to the slope surface and simple rises to reduce
the stability of the slope.
Researchers working in various terrains have already observed and analyzed
different mechanisms for rainfall induced landslides. Only a few researchers and non-
scientific media have raised the issues of slope failure just after light/intense rainfall
followed by drought mainly in mudstone formations. In these studies, the researchers
(Spears and Taylor, 1971; Matsukura and Mizuna, 1986; Editorial, 2001 and 2007; Sadisun
et al., 2003 and Qureshi et. al., 2009) conclude that slaking induced landslide initiation is a
2.1
Chapter 2 Case study on rainfall induced landslides
complex problem involving the analysis of particle crushing due to slaking, suction force
and unsaturated soil shear strength etc. However to date, an analytical formulation that
incorporate the complete slaking phenomena and suction force within a traditional slope
stability framework has not been developed. In the past, many studies (Spears and Taylor,
1971; Matsukura and Mizuna, 1986; Editorial, 2001 and 2007 and Qureshi et. al.2009) were
conducted to investigate the weathering potential of rocks in relation to the slope stability
problems, but still the understanding is not reproducible. Weathering studies, in practical
assessment of slope failures, help to understand what stage the landscape has reached in the
weathering process. In the field of geotechnical engineering, recognition of the weathered
conditions of the slope is a critical issue in evaluating the slope failure hazards.
As already described in the Introduction Chapter, the global warming promotes a
strong disturbance on the world-wide weather. Subjecting to its impact, both of the rainfall
pattern and the environmental temperature in the mudstone area are obviously changing.
The number of high temperature days is increasing; the rainy season is gradually shortened.
Thus, the mudstone areas all over the world currently are in a very serious situation which it
has never been subjected to. After a drought has ended, and an intense rain has occurred, the
new threat of slaking induced landslides becomes present because the drought before
rainfall accelerates the slaking of mudstones. It is necessary for the long-term evaluation of
the slopes to know how the stabilities of the mudstone slopes are changed in accordance
with slaking the process.
2.2 Infinite slope analysis

A slope that extends for a relatively long distance and has a consistent subsurface
profile may be analysed as an infinite slope. The failure plane for this case is parallel to the
surface of the slope and the limit equilibrium method can be applied readily.
a) Infinite slopes in dry cohesionless soil

A typical section or “slice” through the potential failure zone of a slope in a dry
cohesionless soil is considered as shown in Fig. 2.1. All the forces acting on the slice are
presented by free body diagram.
2.2
Fig.2.1. Infinite slope failure in dry sand with free body diagram of forces
The weight of the slice of having a unit length is given by;

W  bh ………………………………………………… (2.1)
Where, γ is the unit weight of the dry soil, b is the width of slice and h is height of slice as
shown in Fig. 2.1
The normal (N) and tangential (T) force components of W acting on a slip surface with an β
angle are determined as
N  W cos  ……………………………………………..….. (2.2)
T  W sin  …………………………….………………….. (2.3)
The shear strength along the failure plane is given by:
S  N tan  ………………………………………….……... (2.4)
Where,  is the angle of internal friction.
The factor of safety (FOS) is defined as the ratio of available shear strength to strength
required to maintain stability. Thus, the FOS will be given by:
S N tan  W cos  tan  tan 
FOS     ….……… (2.5)
T W sin  W sin  tan 
2.3
From the equation (5), it is clear that the FOS is independent of the slope depth, h, and
depends only on the angle of internal friction,  and the angle of the slip surface, β. The
slope is said to have reached limit equilibrium when  = β and the FOS becomes 1.
b) Infinite slopes in cohesive soils with parallel seepage

If the seepage of a saturated slope in a cohesive soil is parallel to the surface of the slope
as shown in Figure 2.2, the same limit equilibrium concepts may be applied to determine the
FOS, which will now depend on the effective normal force (N'). In the saturated soil case,
effective shear strength parameters, (c' and φ') are used.
Fig.2.2. Infinite slope failure in a cohesive soil with parallel seepage with free body
diagram of forces
From the Fig.2.2.Weight of slice (W) and he effective normal forces (N’) are as following;
W  sat b h ………………………………………................. (2.6)
N' N U   sat b h cos    w b h cos  …………............. (2.7)
T  W sin  ………………………………………................. (2.8)

Similarly, the pore water force acting on the base of a typical slice having a unit length.
b ( h cos 2  )
U w
 b h cos  …………........... (2.9)
cos 
w
Where, b is width of assumed slice,  w unit weight of water, h is any depth less than or
equal to the depth of saturation and  is slope of slip surface.
2.4
The available shear strength, S, along the failure plane will depend on φ' and the
effective normal force, N' =N-U, where N is the total normal force. The total shear strength,
S is given by the following equation:
c' b
S  ( N  U ) tan  ' …….…………………….............. (2.10)
cos 
From the equation (2.10) and (2.8), the factor of safety (FOS);
S S c' cos   ( N  U ) tan 

b
FOS   ….….................... (2.11)
T W Sin
From equation (2.8), substituting the value of W into equation (2.11) and simplifying the
equation:
c'  (   w ) h cos 2 tan 
FOS  sat
…….……................. (2.12)
 sat h cos  Sin
For the case of a saturated cohesionless soil, c' = 0, Then the equation (2.12) becomes
(   w ) tan 
FOS  sat
…………….…….……................. (2.13)
 sat tan 
From equation 2.13 it is obvious that FOS is again independent of the slip surface depth, h.
Similarly, in the case of cohesionless soil, it is seen from equation (2.11) that both strength
parameters c’ and φ' have a vital role in stability of slope.
Botts (1986) and Tovar and Colmenares (2011) found a more than 70 % reduction in
the angle of internal friction (φ) of mudstone due to slaking. In addition, c was reduced by
about 40 % after 4 cycles of wetting and drying. Similarly, Yoshida et al. (1997), Sadisun et
al. (2003), Bhattarai et al. (2007), Youn and Tonon (2010), and many other researchers
showed a significant reduction in shear strength parameters after a number of cycles of
wetting and drying. According to Spears and Taylor; 1971, the shear strength parameters for
the shallowest fissile mudstone and shale samples show a decrease in the angle of internal
friction of up to 37 % and a drop in cohesion of around 93 %.
From the equations 2.5, 2.12 and 2.13, it is obvious that the slaking of mudstone
could slope failure trigger. Similarly, Matsukura and Mizuna (1986) and Qureshi et al.,
(2009) explained that slaking has a great influence on the value of c and φ, which in turn
determine the slope angle and of course, the stability of slopes.
However, drying is possible only up to shallow depths. So, slaking can induce
shallow landslides (small scale) and the speed of landslide is very slow as compared to
2.5
heavy rainfall induced landslide. Because of these facts, slaking induced landslides may not
cause any casualties and large economic loss. So, no media and researchers have focused on
this type of disaster.
2.3 Landslide events collection

Thousands of rainfall induced landslides occur every year in the world. So, the
number of landslide events considered in this research is almost negligible as compared to
the total number of landslides which occur. The main objective of this chapter is to explore
landslide events preceded by significant drought. These landslides may be due to slaking of
soft sedimentary rocks.
In order to analyse a landslide in this research, the author tried to collect several
primary elements and secondary elements to describe the event and landslide characteristic,
when available. The both primary and secondary elements are:
Primary elements
1. Location details (town/village, province, and country etc.)
2. Time (event date and local time)
3. Trigger (rainfall pattern and rainfall intensity)
4. Geology
Secondary elements
5. Latitude and longitude
6. Type and relative size of the event (e.g., landslide, debris flow, mudslide,
rock avalanche etc.)
7. Impact information (e.g., Casualties and economic loss etc.)
The primary and secondary elements for each landslide event were obtained mainly
from landslide related paper (Kirschbaum et al., 2009; Huang and le, 2011 and more),
online news media, and hazard databases, including: United States Geological Survey
(USGS), Disaster management website, ministry of home affairs , Nepal, Landslide blogs,
Global disaster data base, International Consortium on Landslides website, International
Landslide Centre, University of Durham, other online regional and national newspaper
articles and media sources. After identifying the landslide event, precipitation data and
geological information were collected from Tropical Rainfall Measuring Mission (TRMM)
and available geological map respectively, details of which are discussed later.
2.6
2.3.1 Geology
As previously mentioned, the location of landslide events in terms of latitude and

longitude was also collected from different sources. The latitude and longitude were used to
extract the geological information of landslides events from available geological map.
Figures 2.3-2.5 show some typical geological maps of China, Africa and Nepal respectively
which contain more than 50 groups of rock. However in order to analyse the landslide
distribution using the rock type, similar types of rock, based on its formation and age have
been grouped together, six groups of rock were obtained as follows as follows:
a) Granite with other intrusive rocks
b) Volcanic and plutonic rocks
c) Mudstone, sand, conglomerates and chert etc.
d) Metamorphic
e) Other ( do not lie within the above mentioned categories or are a complex
combination of these rocks)
f) Unidentified
Fig. 2.3 Geological map of China
2.7
Fig. 2.4 Geology map of Africa region
Fig. 2.5 Geological map of Nepal
2.3.2 Precipitation
The precipitation record of the landslide area, obtained from the Tropical Rainfall
Measuring Mission (TRMM) satellite is shown in Fig. 8. The Tropical Rainfall Measuring
Mission (TRMM) is a joint U.S.-Japan satellite mission to monitor tropical and subtropical
precipitation and to estimate its associated latent heat. TRMM was successfully launched on
27 November 1997 from the Tanegashima Space Center in Japan. TRMM estimates the
2.8
rainfalls over tropical to subtropical regions with 0.25 x 0.25 resolution. Details of the
algorithm can be found at http://trmm.gsfc.nasa.gov/3b42.html. Data are open to public via
their website. Ground base validation or calibration with rain gauge data from
corresponding landslide area is required to obtain more accurate precipitation data from
TRMM (Wolff et al., 2005; Robinson et al., 2000 and Islam et al., 2010). This part of
research is only for motivation or background of this research. Time constraint is also
another major factor to use crude precipitation data from TRMM to analyze the rainfall
induced landslide on the basis of precipitation intensity per day. Ground base validation or
calibration with rain gauges data from corresponding sites should be done for detailed and
further research on the same topic. Similarly, use of KBBI (Keetch-Byram Drought Index)
or DDSLR (Dry Day Since Last Rainfall) is one of best tools to quantify the drought or dry
days. A precipitation data set for at least 5 years before the landslide event was analysed in
this studies. Figures 2.7-2.9 are some typical examples of precipitation patterns derived
from TRMM.
Classification of rainfall intensity
The classification of rainfall intensity per hour or per day is a complex and difficult
task. Rainfall intensities are classified from different points of view such as agriculture,
flooding and landslides etc. Moreover, it largely depends on location and other climatic
conditions. So, many countries and organizations develop their own classification assuming
different threshold values. Table 1-3 show some typical rainfall intensity classifications
used in India and China. As shown in the table, these thresholds vary considerably from one
country to another, which means that it would be difficult to obtain a universal classification
on the basis of the different thresholds.
Table 1 Rainfall intensity classification in India
S.N. Intensity (mm/day) Classification
1 <35 mm Light rain
2 35-74.9 mm Rather heavy rain
2 75-124.9 mm Heavy rain
4 >125 mm Very heavy rain
2.9
Table 2 Rainfall intensity classification in China
1 <10 mm Light rain
2 10-25 mm Moderate rain
2 25-50 mm Heavy rain
4 50-100 mm Heavier rain
5 >100 mm Extremely heavy
For simplicity and uniformity, the author classified the rainfall intensity per day for
this research. However, this classification is not based on any scientific logic.
Table 3 Rainfall intensity classification used in this research
1 <30 mm Light rain
2 30-60 mm Moderate rain
2 60-100 mm Heavy rain
4 >100 mm Very Heavy rain
100
80
60
40
20
0
2006
2007
2008
2009
2010
11 Ja n, 2011
Time
Fig. 2.6 Daily precipitation in Dana Point, Pacific coast highway, USA from TRMM
2.10
100
80
60
40
20
25-May
2003
2004
2006
2007
2002
2005
Time
Fig 2.7 Daily precipitation in Flenglexiang, Sichuan Province, China from TRMM
140
120
100
80
60
40
20
1-Aug
2000
2001
2002
1998
1999
2003
Time
Fig. 2.8 Daily precipitation in Mankamana, Gorkha, Nepal from TRMM
2.4 Distribution of landslides

After analyzing the precipitation from the TRMM and geology from available
geological maps of more than 300 random landslides event from all over the world, the
following scenarios were found.
2.4.1 Geology
As mentioned in Chapter 2.2.1, in order to study the effect of rock type on landslide
hazard, similar types of rock, based on their formation, were grouped together to obtain 6
groups.
2.11
Percentage of landslides occurring in each group is as follows. Group 1: (Granite

with other intrusive rock such as basalt, andesine, diorite etc.) has a contribution of about 15
% on total landslide event. Group 2: (Volcanic and plutonic) is responsible for about 5 %.
Group 3: (Mudstone, sand, conglomerates and chert etc.) represents largest proportion
among other groups. About 40 % of landslides occurred on soft sedimentary rock
formations i.e. Mudstone, sand, conglomerates and chert etc. Group 4: (Metamorphic rocks)
is responsible for about 11 % of total rainfall induced landslides. Group 5: (Other: all the
rock which not lie on group 1-4 and very complex geological formation to identify fall in
this group) is responsible for 20 % of total rainfall landslides. Finally, Group 6:
(Unidentified) corresponds to those landslides event with lack of geological information.
The graphical representation of the landslide distribution on rock types is presented in Fig.
2.9 Mudstone
Granite
Volcanic
Unidetified: 9%
Metamorphic
Others
Unidentified
Others: Mudstone, Shale,

20% and Conglomerate:
40%
Metamorphic:
11%
Granite: 15%
Volcanic: 5%
Fig. 2.9 Landslides classified by frequency occuring in each rock group
Safaei et al. (2002) also analyzed the distribution of 422 landslides in Mazandaran
province, Iran. The author mentioned that more than 37 % of landslides are related to soft
sedimentary rock formations. Similarly, Soralump et al. (2007) showed that about 44.69 %
of total rainfall induced landslides in Thailand occurred in soft sedimentary rock formations.
The output of this case study of rainfall induced landslides is very close to Safaei (2002)
and Soralump et al. (2007). However the other author considered only a part of the world.
2.12
2.4.2 Precipitation
The main objective of this research is to evaluate the slaking characteristics of

geomaterials. It is known that soft sedimentary rocks are highly prone to slaking
susceptibility. That is why, the author tried to classify the rainfall induced landslides on soft
sedimentary rock formation on the basis of rainfall intensity per day. However, the daily
rainfall intensity is not the only influential factor for triggering landslides. The rainfall
intensity per hour and rainfall duration etc should also be considered. As previously
mentioned, due to various reasons only rainfall the intensity per day is considered. The
classification of rainfall intensity is described in Chapter 2.2.2.
It is found that about half of rainfall induced landslides on soft sedimentary rock
formations are triggered by moderate and light rainfall. About 17 % of landslides on soft
sedimentary rock formations were triggered by light rainfall, while 30 % were caused by
moderate rainfall. Similarly, the remaining half of landslides events were triggered by heavy
and very heavy rainfall. The graphical representation of landslide distribution on the basis
of rainfall intensity per day is presented in Fig. 2.10.
Low rain
(30mm/day):
16.68%
Heavy rain
(100mm/day):
Moderate rain
54.15%
(60mm/day):
29.17%
Fig. 2.10 Landslides classified by frequency occuring on the basis of rainfall intensity
2.13
Figure 2.11 shows the rank of rainfall intensity at the time of landslide event. About
42 % landslides were occurred due to the highest precipitation within five years duration.
Conventionally, it is believed that the rainfall induced landslides cause by heavily rainfall. It
is found that about 12 % rainfall induced landslides on soft sedimentary rock formations are
triggered by lowest rainfall intensity. So, the traditional explanation for rainfall induced
landslides is insufficient regarding the mechanism landslide due to lowest rainfall intensity.
The term slaking should be introduced to define the mechanism and threshold of landslide
initiation.
Fig. 2.11 Landslides classified on the basis of ranking of rainfall intensity at

the time of landslide considering five years precipitation
2.5 Conclusions
Conventionally, it is believed that the rainfall induced landslides cause by heavily
rainfall. Such a landslide event is analysed by susceptibility of hill slopes to generation of
positive pressure heads under a saturated regime, in association with water storage capacity
of regolith zone and/or subsurface water flow paths. However, about 50 % of total rainfall
induced landslides in soft sedimentary rock were found to be occurred after moderate and
light rainfall. So, the traditional explanation for rainfall induced landslides is insufficient
2.14
regarding the mechanism and threshold of landslide initiation. Some landslides most
probably occur under an unsaturated regime, resulting only from a decrease in shear
strength of the soil (Krahn et al., 1989; Terlien, 1997) especially in soft sedimentary rock
formation due to slaking of mudstones. However, detailed field and laboratory
investigations of landslide areas are necessary to inspect the subsurface structures and to
obtain geotechnical properties of slope materials. Then, real mechanism of land even could
be explained precisely.
2.15
CHAPTER 3
LITERATURE REVIWEW
3.1 General
As mentioned in the Introduction chapter, many researchers have been investigating
the slaking behaviour of geomaterials, especially mudstones for many years. Considering
the various aspects of slaking, this chapter provides a brief overview of the existing
literature on slaking of geomaterials. Since the focus of this study is on the slaking
behaviour of geomaterials and its effects on strength and deformation, research works
published on various aspects of slaking of geomaterials, especially mudstone are reviewed
here. Once the current level of knowledge has been examined, the direction of future
research can be more easily determined and focused.
In this chapter, first, the mechanism by which the slaking is believed to occur will be
investigated by studying various research papers on this subject, as well as summarizing
how the slaking process will be defined. Cyclic wetting and drying as well as swelling and
shrinkage are always associated with slaking or deterioration of mudstones, so previous
research on the impact of cyclic wetting and drying as well swelling and shrinkage of
geomaterials are also presented here. Finally, research papers on the strength and
deformation characteristics of slakable materials are studied and few results of these papers
are summarized here.
3.2 Slaking mechanism and process

The term ‘slaking’ is defined as ‘the softening or crumbling and disintegration of the
geomaterials upon cyclic drying and wetting; specifically the breaking up of dried
geomaterials or indurated soil when immersed in water or fully or partially saturated with
water, or the breaking up of clay-rich sedimentary rocks when exposed to air or experienced
drying (Nakano, 1967; Moriwaki, 1974; Surendra et al., 1981 and Vallejo et al., 1993). For
the present study, slaking is defined as the softening or breakdown and dispersion of rocks,
3.1
Chapter 3 Literature review
especially mudstones, in response to alternate wetting and drying or even only in water
immersion. Some mudstones at natural water content slake when immersed in water. Others
when immersed will remain stable with regard to slaking. However, if these materials are
first dried and then rewetted, drying induced slaking can occur. Nakano (1970) presented
evidence that some materials will not slake as long as the water content remains above a
certain threshold value.
Nakano (1967), Taylor (1988), Huppert (1988), Dick and Shakoor (1992), Moon and
Beattie (1995), Bell et al. (1997), Koncagul and Santi (1999), Gokceoglu et al. (2000),
Dhakal et al. (2002), Lashkaripour and Boomeri (2002), Lashkaripour and Ghafoori (2002)
and among other researchers studied factors affecting the slaking or durability of different
geomaterials such as mudstone, sandstone etc. According to those studies, the slaking of
geomaterials depends upon many factors such as mineralogy, physical and chemical
properties (fabric, porosity, mineral alignment, micro-fractures, density) and geology
(lithology, cementation) etc. Sadisun et al. (2005) stated that the main factors controlling
slaking are expandable clay minerals, non-clay minerals (pyrite) and soluble minerals. Santi
and Koncagul (1996) proposed different modes of slaking of shale and the main cause for
each type. Surface slaking, swelling slaking, body slaking and Dispersion slaking are the
four modes of slaking which result from the presence of Calcium-illite/Calcium-kaolinite,
Sodium-montmorillonite, Calcium-montmorillonite and Sodium-kaolinite respectively.
As already mentioned, many researchers have speculated on the slaking mechanism,
few have carried out extensive investigations concerning these possible slaking
mechanisms. According to them, slaking is a very complex process as it depends on many
factors as mentioned above. Therefore, there is very little understanding of the slaking
mechanism due to cyclic wetting and drying or even due to a single immersion. However,
there are different mechanisms discussed in the geotechnical literature, each of which offers
valuable insight into the slaking mechanism and process. There are two primary schools of
thought for the slaking mechanism and process due to cyclic wetting and drying or even
single emersion or prolonged drying.
a) Pore-air compression
Pore-air compression is the predominant slaking mechanism in mudstones
composed primarily of non-expansive clay minerals such as kaolinite. Terzaghi (1936),
3.2
Moriwaki (1974), Bell (1992) and Vallejo et al. (1993) and many others researchers
purposed a slaking mechanism in terms of pore-air compression.
In alternate drying and wetting, when mudstones become dry, air is drawn into
macro-pores and a high suction pressure develops. This in turn results in increased shearing
resistance of the individual fragments by virtue of the high contact pressures. The bulk of
the voids are filled with air under extreme desiccation conditions (Figure 3.1).
When the rock is then rapidly immersed in water, water will be pulled into the each
macro-pore, as a result of capillary forces developed due to suction, and the air that
originally filled the micro-pores will be subjected to compression (Fig. 3.1b) Tensile failure
of mineral skeleton along the weakest planes may occur as a result of which the significant
a. b.
Water Air under pressure
Continuous macro-pore
c.
d 2 Water air d 2
u p dTs
4 4
Fig. 3.1 Forces acting on the micro and macro pore in mudstone (Means and Parcher, 1963)
surface area to be exposed increases (Taylor and Spears, 1970). The air pressure developed
in the macro-pores depends on the capillary pressures, which themselves are related to the
surface tension of the water (Seedsman, 1986) and the pore radius (Vallejo, et al., 1993).
The system of forces acting at the interface between the air and the water in a macro-pore is
shown in Fig. 3.1c (Means and Parcher, 1963).
Considering equilibrium conditions, the relationship between pore pressure and pore
diameter can be explained by the equations below.
3.3
d 2 d 2 ……………..... (3.2.1)
dTs  p u0
4 4
Where, d is the diameter of the macro-pore, Ts is the surface tension of water acting
on the meniscus, p is the air pressure inside the macro- pore and u is the pore water
pressure.
From the equation 3.2.1, the following relationship can be obtained
4Ts
pu ……………………………… (3.2.2)
d
An analysis of the equation 3.2.2 shows that the air-filled pore air pressure, p in the
macro-pore increases as the pore diameter, d decreases. Thus, the smaller the diameter of
the macro-pore, the larger the pressure will be. In summary, slaking of mudstones by air
compression will be more significant in those mudstones containing smaller micro-pores
(Means and Parcher, 1963 and Vallejo et al., 1993). Youn and Tonon (2010) also
summarized four different factors that affect the slaking of clay bearing rocks by pore air
compression as following.
1) Air pressure entrapped in pore spaces and between clay particles
2) Osmotic swelling pressure of expandable clay minerals
3) Pre-existing fissures in rocks
4) Gradual removal of cementation by alternate wetting and drying
However, slaking by pore air compression is only possible, when the pressure build
up occur rapidly (Seedsman, 1986). Low permeability and the presence of expansive clays
may restrict the movement of the wetting front. Given enough cycles of drying and wetting,
breakdown can occur as a result of air pressure. This process can reduce the mudstone into
gravel-size particles (Bell, 1992). Bell (1992) explained failure occurring in consolidated
and poorly cemented rocks in terms of swelling pressure and capillary suction pressure.
This failure happens during the saturation process, when the swelling pressure, or internal
saturation swelling pressure (σs), developed by capillary suction pressures, exceeds the
tensile strength. Taylor (1988) attributed, slaking in less indurated mudstones, to a
combination of increasing air pressure as water invades narrow capillaries, and tensile
failure of weak inter-crystalline bonds due to drying induced pore water suctions. According
to Moon and Beattie (195), pore-air pressure compression, release of residual stresses which
exceeds the rock strength and the loss of cohesion of clay due to the absorption of water on
3.4
to the surfaces of clay may be important mechanisms applicable to the slaking of

geomaterials. The reliability of this mechanism can be demonstrated by the absence of
slaking when slake testing is performed under a vacuum.
b) Hydration of clay mineral

Nakano (1967) explained the mechanism of slaking process in terms of the hydrated
clay mineral present within the rock structure. This is an important hypothesis to consider,
primarily because clay mineral must be present within the rock for it to be slaked in this
way. According to Nakano, the slaking/ weakening of mudstone is due to a kind of chemical
dissolution mainly by the formation of hydrogen bond between originally adsorbed water
molecule and newly adsorbed water molecule around clay particles (Fig. 3.2). It is
δ-
δ + δ+ o
H H
δ+ δ+
o H H
+ δ+
δ H
H -
δ-
oδ
o δ+ δ+
H H
Fig. 3.2 Hydrogen bond between originally adsorbed and newly one water molecule
(Nakano, 1967)
presumed that a part of the free Gibb’s energy which evolved due to the hydrogen bond
formation is used as a mechanical destructive energy in slaking. Similarly, when the water
molecules are held around clay particles through the hydrogen bond, the basal spacing of
expansive clay minerals like montmorillonite increases, causing strain in mudstone and
destructing it.
Huggett (2003) and Scherer (2006) also explored the importance of clay mineral
hydration in the weathering of rocks due to cyclic wetting and drying. Hugget and Scherer
did not define the weathering of rocks due to cyclic wetting and drying as slaking. However,
many authors refer to this process as slaking. As mentioned above, there are a number of
3.5
papers (Yatsu, 1988 and Hall and Hall, 1996) suggesting that clay mineral do not need be
present for slaking of mudstones to occur. If these tests are correct, there must be another
mechanism for the slaking of mudstone. This is not to say that idea behind slaking is
incorrect. Indeed, there is overwhelming evidence that slaking does occur. Rather, it is
important to consider the possibility that the clay mineral expansion does not tell the
complete story; there is more to process for slaking.
3.3 Measurement of slaking
Several types of slake durability tests have been developed and used by researchers
worldwide to assess the slaking susceptibility of geomaterials. Some tests are based on
specific mechanisms of rock disintegration, resulting from physical breakdown by wetting
and drying. Some tests are performed to measure the amount of tightly held surface, pore
and capillary water, as an indicator of the resistance to slaking. Some tests are qualitative
while some are semi-quantitative or quantitative.
The following slaking related tests will be discussed here:

1. Jar slake test (Vellejo et al., 1993 and USBR, 2006)
2. Slake durability test (Franklin and Chandra, 1972)
3. Slaking index test (JGS 2125-2006)
4. Slaking ratio test (NEXCO-100, 2006)
5. Rate of slaking (Morgenstern and Eigenbrod, 1974)
6. Wet-dry deterioration test (Withiam and Andrews, 1982)
7. Emerson crumb test (Emersion, 1967)
8. Water absorption ((Anwar et al., 2000)
1. Jar slake Test (Vellejo et al., 1993 and USBR, 2006)

The jar slake test is a simple and rapid test to evaluate the durability of weak rocks,
such as mudstone. A rock piece about 50-100 grams, is oven-dried at a temperature 105° C
for 24 hours before immersion in water. Rock deterioration pattern during the test is
observed, and the slaking behaviors as well as the final state are recorded as a jar slake
index (Ij) as classified in table 4. This test is performed to give a qualitative assessment of
the rock’s slaking characteristics after immersion.
3.6
Table 3.1 Jar Slake Ranking (Vellejo et al., 1993)
Jar Slake index (Ij) Behaviour after immersion
1 Degrades to a mud
2 Breaks rapidly and totally reduces to flakes
3 Breaks slowly and forms few chips
4 Breaks rapidly and develops several fractures only
5 Breaks slowly and develops few minor cracks
6 No change
2. Slake durability test (Franklin and Chandra, 1972)

The slake durability of rocks is an important parameter when investigating the
engineering behavior of a rock mass (Franklin and Chandra 1972; Onodera et al. 1974;
Crosta 1998; Koncagul and Santi 1999; Gokceoglu et al. 2000; Dhakal et al. 2002; Singh et
al. 2005), especially weak and soluble rocks such as shale, clay-bearing rocks, travertine
and weak limestones, as it represents the degradability. The slake durability test,
recommended by both the American Society for Testing and Materials (ASTM) (ASTM,
1996) and the International Society for Rock Mechanics (ISRM) (ISRM, 2007), is the most
frequently performed test, originally developed by Franklin and Chandra(1972).
The test uses a steel drum with a 2-mm size mesh. Ten rock pieces are taken, each
weighing 40–60 g, with a total weight of approximately 450–550 g. After oven-drying the
sample at a temperature 105° C for 16
hours (until constant weight is achieved),
it is placed in the drum with a 2 mm size
mesh (Fig. 3.3) and the drum is sunk in
water and rotated at 20 rpm for a period
of 10 minutes. After this process, the
material retained in the 2 mm mesh drum Fig. 3.3 Slake durability equipment (ASTM, 1992)
is collected and oven-dried at 105° C for
24 hours and then cooled for 20 minutes before recording the final weight after the first
cycle. This procedure is repeated. The durability of rock is classified on the basis of Id2 as
3.7
described in Table 3.2, where Id2 is the ratio of the dry weight remaining in the 2 mm mesh
drum to the initial dry weight, expressed as a percentage. The standard method recognized
by ASTM and ISRM is based on the two cycles of drying and wetting, however many
researchers (Taylor, 1988; Moon and Beattie, 1995; Ulusay et al., 1995; Bell et al., 1997;
Gokceoglu et al. 2000; Yagiz and Akyol, 2008 and Yagiz, 2010) recommend more cycles of
wetting and drying to evaluate rocks of higher durability.
Table 3.2: Gamble’s slake durability classification (Gamble, 1971)
% Id1 (Dry weight %Id2 (Dry weight

S.N. Group Name
basis) basis)
1 Very high durability >99 >98
2 High durability 98-99 95-98
3 Medium high durability 95-98 85-95
4 Medium durability 85-95 60-85
5 Low durability 60-85 30-60
6 Very low durability <60 <30
Many researchers tried to develop a relationship between Id2 and strength

parameters. Gemici (2001) established a relation between Id2 and point load strength as
shown in Fig. 3.4. Sadisun et al. (2004) proposed a relation between important clay contents
which play an important role in slaking and Id2 (Fig.3.5). The slaking index and smectite
content relationship could be better used to subdivide the argillaceous rock into individual
rocks. Yilmaz and Karacan (2005) performed the regression analyses between Id2 and void
ratio (e) and the best-fitted relation was plotted for mudstone in doline formation. The test
results in Fig. 3.6 have shown a very good correlation with the correlation coefficient (R) of
0.96. Yagiz (2010) established the relationships between the slake durability indices and
modulus of elasticity (E), Uniaxial compressive strength (UCS) and shear velocity (Vp) etc
for some carbonate rocks as shown in Fig. 3.7 and 3.8. It is very useful to estimate strength
parameter from very simple and popular test like slake durability test. Gokceoglu et al.
(2000) also performed a lot of experiments to find the relation between Id4 and uniaxial
compressive strength. The outcomes of the experiments are as shown in Fig. 3.9.
3.8
Fig. 3.4 Relation between slaking index and point load strength (Gemici, 2001)
Fig 3.5 Relation between clay content Fig. 3.6 Slake durability index versus effective
and slaking index (Sadisun et al., 2004) porosity (Yilmaz and Karacan, 2005)
3.9
Fig. 3.7 Relationship between the E and Id4 Fig. 3.8 Relationship between the UCs and Id4
(Yagiz, 2010) (Yagiz, 2010)
Fig. 3.9 Relationship between Id4 and Uniaxial compressive strength (sc) of the marls
(Gokceoglu et al. (2000)
3.10
3. Slaking index test (JGS 2125 -2006)

The Slaking index test method is developed and adopted by Japanese Geotechnical
society (JGS). The details of the Slaking index test are explained in Chapter 4.
4. Slaking ratio test (NEXCO- 100, 2006)

The Slaking index test method is developed and adopted by Nippon Expressway
Corporation Limited (NEXCO). The details of the Slaking ratio test are explained in
Chapter 4.
5. Rate of slaking (Morgenstern and Eigenbrod, 1974)

It consists of saturating the mudstone sample generally in a funnel with a filter paper
for at least 2 hours. After 2 hours, free water is removed through the filter paper. Then, the
rate and amount of water absorbed by the mudstone is determined. The rate and amount of
water absorption can reflect porosity, void geometry and fractures. The slaked portion of the
sample is separated and Atterberge limits of slaked sample are found, which implicitly
reveal the proportion of clay size materials. Morgenstern and Eigenbrod, (1974) relate the
degree of slaking to liquid limit as given in Table 3.3.
Table 3.3 Description of the degree of slaking (Morgenstern and Eigenbrod, 1974)
S.N. Amount of Slaking Liquid limit (%)

1 Very low <20
2 Low 20-50
3 Medium 50-90
4 High 90-140
5 Very high >140
6. Wet and dry deterioration test (Withiam and Andrews, 1982)

As in the slake durability test, this test also consists of several cycles of alternate
wetting and drying. Wetting is carried out by soaking rock pieces in water or chemical
solutions (e.g. sodium sulphate). The test is considered to involve a cyclic wet-dry process,
the rate of water adsorption, and the rate of slaking (Withiam and Andrews, 1982). Both the
sodium sulphate soundness test (ASTM D 5240-92) and the ASTM test of the durability of
rock under wetting and drying conditions (ASTM D 5313-92) are classified as ‘wet-dry
deterioration tests’.
3.11
Sodium sulphate soundness test (ASTM D 5240-92) involves the oven-drying of 5

rock lumps having a volume of 10 cm3 and then immersed in a sulphate solution. This
process completes one cycle. Five successive drying and immersion cycles are required to
complete one Sodium sulphate soundness test (ASTM D 5240-92). This test is very popular
for testing aggregates used in road pavement and concrete.
The percentage of soundness loss is calculated as follows:
Wi  Wf
Average soundness loss  %
Wi
Where, Wi = oven-dried cumulative mass of all rocks prior to testing, and Wf = oven-
dried cumulative mass of the largest remaining pieces of all rocks after testing.
In case of wet and dry test (ASTM D 5313-92), at least 5 rock specimens are used,
the size of which should be as possible but not less than 125 mm. Each specimen should be
64 mm thick and cut normal to bedding or any potential planes of weakness. The test
consists of oven-drying the rock sample and placing the sample on a thin layer of sand in a
container filled with water for a period of 12 hours. The sample is removed from water and
then dried in an oven at a temperature of 60-70° C for at least 6 hours. The same process of
wetting and drying should be repeated for 80 cycles. Like in the Sodium sulphate soundness
test (ASTM D 5240-92), only the largest remaining piece of each rock specimen is used in
the calculation. The resistance of the rock sample to the accelerated weathering can be
evaluated. The percentage of loss is calculated as follows:
Wi  Wf
Average soundness loss  %
Wi
Where, Wi = oven-dried cumulative mass of all rocks prior to testing, and Wf =
oven-dried cumulative mass of the largest remaining pieces of all rocks after testing.
The test is very useful for determining the durability of rock bed for erosion control
and riprap testing. However, there is not any guideline to classify rocks on the basic of
average soundness loss value.
7. Emersion crumb test (Emersion, 1967)

In this test, specimens from the field directly are immersed in water. Ingles and
Metcalf (1972) recommended using good quality water. Slaking is observed first (slaking or
no slaking), and subsequent dispersion (dispersion or flocculation) is observed in the finest
slaked fragments. In this test, the slaking behaviour is defined by breakage of the specimen,
3.12
while the dispersion behaviour is derived from the diffusion of the soil particles over the
specimen or covering the bottom of the beaker. The slaking behaviour of the specimens can
be grouped into eight classes (Emerson, 1967).
8. Water absorption test (Anwar et al., 2000)

The rate and amount of water absorption also reveal the durability of rock (Anwar et
al., 2000). Water absorption is thus also very important. The details of water absorption are
explained in Chapter 4.
3.4 Swelling and Shrinkage of mudstones

Focusing on mudstones, a few researchers described the observed cyclic swelling
and shrinkage behavior of mudstones or soft rocks. According to Seedsman (1993), when
moisture in soil or rock decreases, air enters into the pores and when moisture returns by
capillarity, air is captured and its pressure increases in the internal pores which lead to a
higher swelling pressure. This higher swelling pressure leads to slaking. Suction pressure
increases with the decrease in water content during the drying phase. Higher negative
suction pressure causes the failure of weak crystalline bond of mudstones which also
ultimately leads to particle breakage (slaking).
As described above, the slaking process relates to the swelling mechanism of the
clay mineral, which is attributed to the ability of the rock to take up water between grains
and in interlayer structures which causes the increase in volume (Moriwaki, 1974; Anwar et
al., 2000). Similarly, many authors also relate the shrinkage to slaking (Nakano, 1970;
Surendra et al., 1991; Vallejo et al., 1993 and Anwar, 2000). Swelling is a combination of
physico-chemical reactions involving water and stress relief (ISRM, 1983).
Moriwaki (1974), Einstein (1996) and Youn and Tonon (2010) recognized that
swelling (then slaking) in mudstones is caused by one or a combination of three
mechanisms: mechanical (compressed entrapped air), osmotic and hydration of ion around
the clay surface (Moriwaki, 1974; Einstein, 1996). According to ISRM (1983), Einstein
(1996) and Doostmohammadi et al. (2008), the physico-chemical reaction with water is
usually the major contribution to swelling but it usually takes place simultaneously with, or
following, stress relief. As previously mentioned, the slaking of geomaterials is
consequence of swelling and shrinkage. So, all factors affecting the slaking, of course,
influence the swelling and shrinkage of geomaterials. These factors are mineralogy,
3.13
lithology, ground characteristics, hydrology, stress state and especially weathering

conditions (Moriwaki, 1974, Pejon and Zuquette, 2002; Moosavi et al., 2006).
Gysel (1987), Einstein (1996), Barla (1999) and Hawlader et al. (2003 and 2005)
explained the swelling behaviour of mudstones by only considering the first saturation
period and neglects the shrinking stage due to the following drying season. Typically in the
field, mudstones are subjected to cyclic wetting and drying with distinct seasonal change or
with ground water variation or reservoir level changes etc. The research on cyclic wetting
and drying induced swell-shrink behavior and its effect on the strength and deformation is
thus very important. In reality, these types of research are very rare. For the soft rock
containing expansive clays, many researchers such as Chen et al. (1985), Chen and Ma
(1987), Subba Rao and Satyadas (1987) and also Basma et al. (1996) have addressed the
problem of cyclic swelling and
shrinking. It was inferred that when
clays are repeatedly subjected to full
swell, then desiccated to their initial
water content, they gradually display
less expansion. On the other hand,
Osipov et al. (1987), Day (1994) and
also Basma et al. (1996) showed that the
swelling potential increases with the
number of wetting and drying cycles
when the samples are allowed to fully
shrink up to a water content equal to or
less than the shrinkage limit. These
studies show the influence of the
shrinking stage on the next swelling
period. Zhang et al. (2010) performed
free swelling tests on argillaceous (Cox)
and clay stones (OPA) by wetting the
samples with water vapor. He measured
strains at a low relative humidity of RH
Fig. 3.10 Free swelling strains measured on
= 23% and 100% relative humidity for COx and OPA samples (Zhang et al., 2010)
at various humidities
3.14
two months and eight months respectively. After reaching equilibrium during drying at RH
= 23%, the remaining water contents in both samples became asymptotic at the water
content about 1.7%, but the shrinkages behaviour are different (Fig. 3.10). A larger
shrinkage of 1.6% was observed on the OPA sample with more clay content, about 5 times
that of Cox sample with less clay content. The tests also showed that the mudstones can take
up great amounts of water, over 14%–18%, much more than that of about 7% in natural and
saturated states. The increase of water content yielded a large volume expansion up to 8% –
12%. More water uptake and volume expansion could be expected if wetting continued to
reach equilibrium. Additionally, it is to be pointed out that the swelling of the sedimentary
mudstones is pronouncedly anisotropic due to bedding planes. A larger swelling strain
occurs in direction perpendicular to the bedding plane (Fig.3.10). The anisotropic swelling
is so strong that it is able to break down the mudstones along the bedding planes. Figure 9
shows the pictures of the samples after swelling.
Huang et al. (1995) performed some cyclic wetting and drying tests on shales and
approximated the relation between maximum swelling potential (i.e. maximum pressure and
strain) between the first and the second test cycles. From the Fig. 3.11, it is seen that the
maximum swelling pressure measured in the second cycle test decreased significantly, and
the time needed to reach a peak pressure reduced as well.
Al-Shamrani and Al-
Mhaidib (1999) performed a
series of experiments to
measure swelling under
different confining stress
conditions. Figure 3.12 shows
that swelling decreases with
the increase in confining
pressure. According to many
researchers swelling reveals
the slaking of geomaterials. So,
slaking of geomaterials may Fig. 3.11 swelling strain in the first cycle and second
also be related to confining cycle wetting of shale (Huang et al. (1995)
pressure. Similarly, the same authors did research on the influence of swelling on the shear
strength of shale. It can be seen from Figure 3.13 that the amount of swell has a significant
3.15
influence on the shear strength of shale. For instance, shear strength decreases by about
50% when the shear was initiated at a swell percent of 25% of the ultimate vertical swell.
Fig. 3.12 Relation between confining Fig. 3.13 Deviator stress at failure versus
pressure and swelling (Al-Shamrani and Al- swelling in percent for shale samples (Al-
Mhaidib, 1999) Shamrani and Al-Mhaidib, 1999)
Furthermore, the shear strength is approximately 10% of the shear strength of the
non-swelled soil when the sample was sheared after reaching the ultimate vertical swell.
Pejon and Zuquette (2002)
described the impact of cyclic
wetting and drying on the swelling
behavior of mudrocks. They
demonstrated that accumulated
maximum axial swelling strain of
different mudrocks varied
according to the number of wetting
and drying cycles. From the Fig.
3.14, the trend of increasing
Fig. 3.14 Maximum swelling strain with respect to
swelling strain with the cycles is wetting and drying cycles (Pejon and Zuquette, 2002)
seen clearly. However, samples air-
dried to constant weight experienced bigger and faster swelling than the samples air-dried to
natural moisture content after each wetting cycle (Fig. 3.15a and b). The texture, structure,
clay content, bulk density, specific gravity, dry density, initial moisture content, initial void
ratio, porosity, degree of saturation, methylene blue value (vb), cation exchange capacity
3.16
(CEC), clay activity index (Acb), carbonate content and mercury porosity are effective
parameters for swelling rate (Pejon and Zuquette , 2002)
Anwar et al. (2000) conducted the free swelling and slaking tests to examine the
behavior of swelling of 24 specimens which consisted of sandstone, shale, coaly-shale,
sandy-shale and mudstone. All these specimens showed similar trend in swelling, in which
sandstones were less than 1 % but shales were in the order 2- 12 %. From the Fig. 3.16a, it
Fig. 3.15 Swelling strain of samples after being dried to (a) natural moisture in air
(b) completely dried (Pejon and Zuquette, 2002)
a) Fi
Fig. 3.16 a) time histories of swelling b) effect g.3
of Montomorilonite on swelling strain
(Anwar et al., 2000) .19
.
is seen that the largest swelling occurred in the initial stage, and then it was followed by
Co
slow growth of the expansion. It showed that the mudstones
mp are highly prone to swelling
which ultimately lead to slaking. Similarly he found ara that the expandable clay mineral
tiv
content may effect upon the swelling strain as shown in Fig. 3.16 b. This fact is significant
e
in the slaking process. str
ess
-
str
ain
plo 3.17
ts
Doostmohammadi et al. (2007) performed some cyclic swelling tests under

oedometric condition on mudstone samples. The swelling pressure of the samples was
measured in the axial direction over the time. The maximum swelling pressure increased
with each cycle (Fig. 3.17). The increasing rate of swelling pressure was decreased after the
third cycle and it finally reached an almost steady state.
As Al-Shamrani and Al-Mhaidib (1999) presented, Lee et al. (2002) also concluded
that the larger the confining stress, the smaller the swelling strain is (Fig. 3.18). The
swelling strain may approach zero at higher confining stress.
Fig. 3.17 The trend of swelling pressure in equal times of cycles

(Doostmohammadi et al., 2007))
Fig. 3.18 Swelling strain under different confining stress

(Al-Shamrani and Al-Mhaidib (1999)
3.18
3.5 Effect of wetting and drying on the properties of

geomaterials
Many researchers have been seeking to evaluate the effect of cyclic wetting and
drying on geomaterials. Most of their studies are done on soft sedimentary rock i.e.
mudstone etc. Some studies related to cyclic wetting and drying have already been
discussed in previous chapter. Table 3.4 provides the list of some researches done in
laboratory simulating wetting and drying process. The material used, methodology adopted
and outcomes of these researchers are also mentioned in Table 3.4. Of course, each research
mentioned in Table 3.4 was done by individual research groups. However, these research
studies have some similarity from different perspectives. Laboratory studies can initially be
grouped on the basis of objectives, methodology adopted and final outputs.
Franklin and Chandra (1972); Gokceoglu et al. (2000); Sri-in and Fuenkajorn (2007)
and Erguler and Ulusay, (2009) performed cyclic wetting and drying experiments to find the
slake durability index, which is used to evaluate the slaking susceptibility of mudstone. The
details of the slaking durability index were described in Chapter 3.3. Rocks containing
higher carbonate are less susceptible to disintegration during cyclic wetting and drying.
Similarly, rock containing higher amount of clays can easily undergo disintegration due to
cyclic wetting and drying. Hall and Hall (1996); Nicholson (2001); Kanyaya and Trenhaile
(2005); Wells et al. (2005); Gonzalez and Scherer (2006) and Subner and Loubser (2008)
applied a large number of cyclic wetting and drying cycles. The wetting period is relatively
very short as compared to the drying period. All these experiments were conducted under
isolated environmental conditions. Actually these experiments were conducted to study the
mechanism of wetting and drying. Hall and Hall (1996) investigated the relationship
between the way in which the rock types received moisture and the changes in the internal
structure of the rock. He found that the large number of cyclic wetting and drying has a
significant influence on particle crushing even in case of hard rock rather than mudstone.
Gonzalez and Scherer (2006) tried to develop a relationship between the number of cycle
and the volume change (swelling) of specimen. Similarly, Pradini et al. (1996) also
conducted the cyclic wetting and drying test to develop the relationship between the number
wetting and drying cycle and volume change (swelling). Gradual increase in swelling over
the time was observed. Matsukura and Yatsu (1982) and Huang et al. (1995) also monitored
the swelling of geomaterials due to cyclic wetting and drying. They found that the swelling
3.19
of geomaterials largely depends on the clay mineral and its type. Dick et al. (1994);
Gokceoglu et al. (2000); Pezen and Zuquette (2002) and Tovar and Colmenares (2011)
simulated cyclic wetting and drying and analyzed the mineralogical composition by using
XRD or SEM. Dick et al. (1994) found strong relationship between slaking and clay
content. Pezen and Zuquette (2002) found considerable breakdown of soft rock even if rock
has low clay content. Tovar and Colmenares (2011) showed that cyclic wetting and drying
generates the alternation and destruction of the rocks and an increase in void ratio. Okamoto
et al. (1984) found considerable change in both physical and mechanical properties of
mudstone after certain number of cyclic wetting and drying. The authors also concluded that
change in both physical and mechanical properties of mudstone depends on the degree on
dryness before immersion. Same results were described by Doostmohammadi et al. (2009).
Hachinohe et al. (1999) tried to evaluate the behaviour of geomaterials due to cyclic wetting
and drying in natural condition. Hachinohe et al. (1999) concluded that weathering rates
decrease logarithmically over time rather than linearly. Botts, 1998 found increase in
fissures in shale after cyclic wetting and drying leading to the decrease in shear strength.
Canton et al. (2001) conducted cyclic wetting and drying tests on consolidated mudstone.
Canton et al. (2001) showed increase in both water absorption and porosity.
3.20
Table 3.4 List of various studies on Effect of wetting and drying on the properties of geomaterials
S.N. Author Material Experiment procedure Measured parameter Obtained results

Franklin and Alternate Mudstones disintegration
` Mudstones Slake durability index
Chandra, 1972 wetting and drying cycles i.e. Slake durability index
Neogene All measured parameters are
Physical and mechanical test, Change in physical and
mudstone dependent on the degree of dryness
Okamoto et al., Absorption test and Swelling pressure mechanical properties,
2 cylinder 50 mm of the specimen, and do not vary
1984 measurement under simulating cyclic absorption, swelling and
dia. and 20-100 evenly according to drying time
wetting and drying slaking characteristics
mm height and temperature
Naturally dried for six month at room
shale and Tuff Swelling pressure,
temperature before experiment, Slaking rate depends on different
Matsukura and cube of one percentage of mass loss,
3 during experiment 9 hours dried in kinds of the clay mineral contained
Yatsu, 1982 edge about 50 Mass loss of different shale
vacuum desiccators and 15 hours in the rocks
cm and tuff
submerged in distilled water
Showed a strong relationship
Slake durability index and between percent expandable clays
4 Dick et al., 1994 Mudrocks slake durability test and XRD
mineralogical content and slake durability index for clay
stones
Percent of expandable clays and
2 cycles wetting and drying carried Maximum swelling
5 Huang et al., 1995 Shale cube slake durability index for clay
out at different temperature pressure
stones
3.22
Hall and Hall, Sandstones/dole 145 cycles of wetting and drying Change in sample dry mass Significant influences on crushing
6
1996 rite block carried out using different method and water absorption of particles
3 cycles by Non-contact laser Bulk density and Surface

Pardini et al., Smetitic Minimal effect on bulk density and
7 profiling (3 days wetting and 12 days micro-topography
1996 mudrocks block surface micro-topography
drying ) variations
Fissure measurement of
Up to 80 % of rock strength
Clay shale Slake durability test (only one cycles, slaking under confining
8 Botts, 1998 decreased even after 1 cycles of
cylinder 0-28 days drying) pressures and application
wetting and drying
of critical state model
Weathering rates decrease
Hachinohe et al., Sandstones and Weathering profile observed in the
9 Residual strength ratio logarithmically over time rather
1999 mudstones core field by using needle penetrometer
than linearly
Marls, clay Durability of clay-bearing rocks
Gokceoglu et al., 4 cycles of wetting and drying by Uniaxial compression test
10 stone and correlates well with amount of
200 multiple slake durability testing and Mineralogy by XRD
sandstones etc expandable clay content
40, 80 cycles wetting and drying on Weight loss, fracture
Lime stones different rocks (12 hours wetting and density, effective porosity Change in effective porosity and
11 Nicholson, 2001
and chalk 6-10 hours drying), Salt weathering and mercury intrusion amount of water for saturation
test porosimetry
mudstones 5, 10 and 20 cycles of wetting and Total mineralogy, Micro- Increase in both water absorption
12 Canton et al. 2001
blocks drying using sand bath morphology, salt chemistry capacity and porosity
3.23
11 cycles wetting and drying Grain size and pore size

Considerable breakdowns of
Pezen and French (duration of wetting and drying distribution, Ion exchange
13 particles even if rock has low
Zuquette, 2002 mudrocks core depended on moisture level) and capacity and mineralogical
mineral content
SEM measured composition
Basalts/Sandsto 700-930 cycles wetting and drying
Kanyaya and Rate of down wearing in Down wearing rates decrease with
14 nes/Argillite using tidal simulator in laboratory
Trenhaile, 2005 the field by MEM elevation
cores and cubes environment
5000 cyclic wetting and drying
Particle crushing and Significant changes in rock
15 Wells et al., 2005 Schist cube simulated tropical climate (5 minute
moisture content morphology
wetting and 67 minute drying)
Portland Ratio of wet and dry
Gonzalez and 30 minute wetting and 60 minute Gradual increase in swelling over
16 brownstone modulus and free swelling
Scherer, 2006 drying applied by automatic machine time
plate ratio
Change in sample dry Moisture amplitude has not
52 cycles wetting and drying using
Subner and Sandstones mass, change in water significant impact on degree of
17 different moisture amplitude (2-6 min
Loubser, 2008 block absorption, porosity and slaking/weathering that takes place
wetting and 48 hours drying)
saturation coefficient over an intermediate time period
Swelling potential of mudstones is
Swelling strain of
Mudstones quite different when the water
mudstones under different
Doostmohammadi cylinder 50 mm Simulating cyclic drying and wetting content of the sample changes
18 dead pressure with respect
et al., 2009 dia. and 20 in Swelling pressure test apparatus periodically as opposed to the case
to number of cyclic wetting
height where the sample is under constant
and drying
water content
3.24
Drying and wetting cycles,

Drying and wetting cycle (120 hours Change in shear strength generate alternation and
Tovar and Argillaceous
19 for each wetting and drying), and mineralogical destructuration of the rocks,
Colmenares, 2011 Rocks
Fluorescence test and SEM test composition increase void ratio; are main causes
of shear strength reduction
Seven rock
Multiple-cycle slakes durability High values of durability from the
cube
Gokceoglu et al., testing, X-ray diffraction (XRD) Slake durability, rock usually corresponds their
20 (sedimentary,
2000 analysis before and after durability mineralogical composition higher carbonate contents.
volcanic and
test and Uniaxial compression testing. Decrease in SDI after 3 or 4 cycles
metamorphic)
sedimentary,
usually correspond to their higher
and 6 cycles of drying and wetting (24 Slake durability index,
Sri-in and carbonate seems a better way to
21 metamorphic hours at 105 0C for drying and 24 change in internal angle of
Fuenkajorn, 2007 increase the number of cyclic
with different hours in 24 0C water) friction and weight loss
wetting and drying
clay
The comparison between the
Slake durability index,
disintegration index values of the
Clay bearing 1 cycle drying and wetting(24 hours disintegration index tests,
Erguler and laboratory specimens and the
22 rocks block at 105 0C for drying and 24 hours in slake durability rating
Ulusay, 2009 samples from the same outcrops
(30*30*40 cm) water), XRD diffraction (SDR) and Mineralogical
exposed to atmospheric conditions
constituent
for 1 year showed close agreement.
3.25
After reviewing the existing literature of research studies upon the effect of wetting
and drying on properties of geomaterials, a number of conclusions can be drawn. As
mentioned in the Introduction chapter, almost all cyclic wetting and drying related research
on geomaterials are done under atmospheric pressure condition (no stress condition).
However, geomaterials in the field are subjected to the cyclic wetting and drying under
confining stress conditions. Similarly, the rate of drying is very rapid and very high
temperature is applied as compared to natural temperature. The effect of rapid drying and
very high temperature is neglected. The impact of duration for drying and wetting is also
not taken into consideration in almost all research. ASTM, JGS and some other authors
developed standard procedures for evaluating the impact of cyclic wetting and drying on
geomaterials. However, it is found that only few researchers followed the standard
procedure. So, the outputs of these researches cannot be easily compared to one another,
making it difficult for all the information gathered to be integrated into one cohesive whole.
The experiments were conducted in isolation. It means only a small part of the rock is
considered under laboratory environment which may totally differ from real weather or
environment. It is understood that weathering processes do not act in isolation. It is not
only important to understand the response of each individual process, but also how all the
different processes interact with one another to cause the observed effect.
In spite of such limitations, most of the existing methodologies of cyclic wetting and
drying are very simple. The results obtained from these experiments will largely help to
evaluate the response of geomaterials under cyclic wetting and drying.
3.6 Strength and deformation characteristics of slakable

materials
Botts (1986) did both theoretical and experimental assessments of the effects of
slaking on the engineering behavior of clay shales. He found the drastic decreases in
strength resulting from slaking as shown in Fig. 3.19. Figure 3.19 shows the axial stress-
axial strain-volume change relationships for the Pierre shale specimens tested under 207
kPa (30 Psi) confining pressure. Here in the figure, 30/4-23 corresponds to the test
performed under 30 Psi confining pressure simulating 4 days of drying and 23 days of
wetting and so on. Drying times for these samples varied from 0 to 28 days. These results
3.26
show that the shear strength of Pierre shale samples is significantly decreased with longer
drying times.
Comparison of sample 30/4-1A and 30/4-23 indicate that shear strength is reduced
with longer wetting time; however drying times seem to be the most influential factor. As
observed in sample 30/28-3, the peak strength of the Pierre shale has been reduced by 80 %
to a value very near the residual strength of unaltered samples with only a single cycle of
drying and wetting. Similarly, Young's Modulus was found to be almost constant for
samples with similar drying times, but greatly reduced with increased drying time. Young's
Modulus of Elasticity varied from about 193,000 kPa for virgin Pierre shale to 34,000 kPa
for samples which had undergone 28 days
of drying (i.e. an 80 % increase in sample
compressibility with slaking). In contrast, as
seen in Figure 3.19, the volume change (the
ratio of axial strain to volumetric strain)
remained almost constant for all samples,
regardless of the degree of softening.
Tovar and Colmenares (2011)
performed a series of direct shear test on
fresh argillaceous rock and argillaceous
rocks experiencing number of cyclic drying
and wetting. Figure 3.20 shows the
reduction in maximum shear strength and
allows us to observe the effect of wetting
and drying cycles. He found that drying and
wetting cycles have a significant effect on
the shear strength, decreasing the angle of
friction and increasing the intercept of the
envelope with the passage of wetting and
drying cycles as shown. The shear strength Fig 3.19 Comparative stress-strain plots
showing drastic softening of Pierre shale after
reduction was significant (70 %) after 4
slaking (Botts, 1986)
cycles of wetting and drying cycles.
Figure 12. Comparative stress-strain
Bhattarai et al. (2007) investigated the shear properties of fresh mudstone, remolded
plots showing drastic softening of Pierre
mudstone and mudstone after one and four cycles of drying and wetting. He reported that
shale after slaking
3.27
significant reductions in peak shear strength were observed after both one cycle and four
cycles of wetting and drying. The shear strength of the specimens after four cycles of
wetting and drying was very close the shear strength of the fully softened specimen as
shown in Fig. 3.21.
Figure 3.22 shows the
shear stress–shear displacement–
volume change relationships for
the specimens made of mudstones
aggregates. It is seen that for all
the applied vertical stress, the
immersed specimens give smaller
peak shear stress than those
without immersion. Reduction in
shear stress seems larger for a
Fig.3.20. Mohr-Coulomb envelopes for wetting and
smaller applied vertical stress. The drying cycles (N) applied (Tovar and Colmenares, 2011)
inclination of shear stress–shear
displacement curve in the early
stage of shearing, though not
clearly seen, also seems to
decrease due to immersion.
Regarding the volume change, it
is noticed that for a vertical
pressure of 80 kPa, the specimen
without immersion exhibits a
dilatant behavior but that the
immersed specimen loses such Fig.3.21. Shear strength envelopes for fresh, slaked and
dilative nature. For applied remolded mudstone specimen (Bhattarai et al., 2007)
vertical pressures of 320 and 640 kPa, the volumes of the specimens without immersion are
contractive and immersion does not cause much difference.
3.28
The effect of water content on the

creep behaviour of soft rock was studied
by Jongpradist and Horii (2007). Drained
triaxial creep tests of specimens with 80%,
40% and 10% saturation degree were
performed at the same stress state. The
graph of the axial creep strain as a function
of time is shown in Fig. 3.23. The
magnitude and rate of axial creep strain at
the same loading time is largest when the
saturation degree is 80% and decreases as
the saturation degree decreases.
Furthermore, creep leads to failure at about
800 minutes in the case of 80% saturation
degree, while this phenomenon is not
observed during the tests with 40 % and 10
% saturation. This means that creep of soft
rock is more significant with the increase
of water content. Yoshida et al. (1997)
Fig. 3.22 Shear stress–shear displacement–volume
computed the strength envelope for change relationships (Yoshida et al., 2004)
mudstone shown in Figure 3.24 along with
its variation with time due to softening. It shows that the stable slope now will be change to
unstable with the elapsed of time. This example illustrates the deterioration of mudstone due to
softening, especially weathering.
Chen (1997) performed uniaxial compressive tests on Gutingken mudstone, Taiwan
under different water contents. Obviously, the author found that the uniaxial compressive
decreases with increase in water content (Fig. 3.25). According to Fig.3.25, when the water
content increases from 0% to 9-10%, the uniaxial compressive strength of Gutingken
mudstone
3.29
Fig. 3.23 Results of creep test with different saturation degree at the
same stress (Jongpradist and Horii (2007))
state
reduces from 39.6 MPa to 3.5 MPa,
and the failure strain increases from
1.42% to 2.5%. Figure 3.26 shows
that the loosely packed specimens
composed of crushed mudstones
(US1 and US2) begin to settle
immediately after being submerged
in water. On the other hand, densely
packed specimens (US3, D1 and D2)
do not show any settlement. The
specimen D2 composed of dry
crushed mudstones exhibits
relatively very small swelling instead
of compression after saturation with
Fig. 3.24 Results of creep test with different saturation
water (Nakano et al., 1998). degree at the same stress state (Yoshida et al., 1997)
3.30
Fig. 3.25.Uniaxial compressive tests under Fig. 3.26 Settlement behaviour after
different water contents (Chen, 1997) submergence (Nakano et al., 1998)
The effect of drying and wetting on the engineering properties was investigated in
terms of peak deviator stress and elastic modulus under triaxial conditions by Youn and
Tonon (2010). Figure 3.27a shows stress–strain curves of the four tested clay-bearing rocks
in fresh conditions and Fig. 3.27b shows the stress–strain curves after one cycle of drying
and wetting. By comparing the stress–strain curves in Fig. 3.27a and b for the same
material, it is readily recognizable how sharply the elastic modulus decreased after the
drying–wetting cycle.
Mechanical anisotropy of compositionally layered shale may be weak, given a
relatively weak preferred orientation of illite (or other phyllosilicates). Ibanez and
Kronenberg (1993) found that the strength of soft rock largely depends on the orientation of
bedding plan with respect to the load applied (Fig. 3.28). The strengths of shale samples
compressed parallel, perpendicular, and at 45 ° to bedding are strongly dependent on
confining pressure applied on it (Fig. 3.29).
3.31
Fig. 3.27 Stress–strain curves: (a) fresh material (b) after 48 h of air-drying for Del
Rio Clay (DR), Eagle Ford Shale (EF), Taylor Marl (TM), and 24 h of air-drying
for Navarro Shale (NA) (Youn and Tonon, 2010)
3.32
Fig.3.28. Stress-strain curves for shale Fig.3.29. Comparison of the weakly

samples compressed perpendicular, defined strength anisotropy of the Wilcox
parallel and at 45 ° to bedding (Ibanez formation shale with that of Martinsburg
and Kronenberg, 1993)
slate (Ibanez and Kronenberg, 1993)
3.7 Conclusions
After the literature review, it is clear that mudstones are problematic geomaterials.
The current tests for slaking susceptibility all involve measuring the effects of wetting and
drying cycles under unconfined conditions. It is clear that these methods do not provide
adequate information regarding the effects of this slaking on the strength and deformation
behavior of clay shale in the field. The author has found virtually no literature involving
research on the effects of slaking on strength and deformation under confining stress
condition as in real field. So, the effects of slaking and reduction of strength and
deformation characteristics of mudstones is still a key concern for geotechnical engineer.
3.33
CHAPTER 4
EXPERIMENTAL SETUP
4.1 General
This chapter describes the selected material, its physical properties and origin,
samples preparation procedure, particle size distribution and general slaking test. Similarly,
the reason behind the choice of this particular material and apparatus, loading conditions
and the general experimental procedure are also presented in this experimental setup
section.
4.2 Materials
As discussed in Chapter 1, a lot of geotechnical engineering problems have been
observed in soft sedimentary rock formations mainly in mudstones. The term “mudstones”
refers to the fine-grained, siliciclastic sedimentary rocks (claystones, mudrocks, siltstones,
and shales) in which more than 50% of the particles are smaller than 0.06 mm in size (Folk
et al., 1970; Grainger, 1984 and Dick and Shakoor, 1992). The mudstones undergo both
mechanical and physical weathering due to cyclic wetting and drying which could lead to a
drastic loss of strength and stiffness, instability of natural slopes and excessive settlement of
embankments. The intension of using crushed mudstone is to accelerate the slaking rate in
the laboratory and also for necessary adjustments to the size of direct shear box.
10mm
Fig. 4.1 Hattian Bala mudstone Fig. 4.2 Ishikawa mudstone Fig.4.3 Chiba gravel
5.1
Chapter 4 Experimental setup
Four types of geomaterials shown in Figure 4.1-4.4 were used for evaluating slaking
characteristics of geomaterials from different parts of Japan and Pakistan. Out of these four
geomaterials, two materials are mudstones from Pakistan and Ishikawa, Japan. The
remaining two materials were Chiba gravel (Chiba gravel) and Toyoura sand (Silica sand)
which are very popular granular materials for geotechnical experiment. Only few a
experiments were performed on Chiba gravel and Toyoura sand to compare their strength
and deformation characteristics with that of crushed mudstones. Similarly few tests on
completely non slakable material (Glass beads) were also carried out to compare slaking
characteristics with geomaterials (Fig. 4.5). Out of these materials, crushed mudstone from
Hattian Bala, Pakistan was used as main material for this research.
Fig. 4.4 Toyoura sand Fig. 4.5 Glass beads (4 mm dia.)
4.2.1 Sampling site

Muzaffarabad
a) Hattian Bala mudstone PAKISTAN
The Hattian Bala mudstone used in this
investigation was obtained from the
earthquake induced landslide dam, formed by
the 2005 Kashmir earthquake, which is located
southeast of Muzaffarabad, Pakistan (Fig 4.6
and 4.7). The earthquake induced landslide
dam was suddenly breached on 9th February,
2010 just after moderate rainfall preceded by
drought. Slaking of mudstone was assumed to
be one of the major causes of the failure
Fig. 4.6 Location of sampling site for
(Sattar et. al., 2010, Kiyota et. al., 2011).
Hattian Bala mudstone
4.2
The site is geographically located approximately 34.14800° North latitude and

73.722183° East longitude (Fig. 4.6). From a geological point of view, the source area is
formed of Miocene aged Murree formation (Dunning et al., 2005 and Mirza et. al., 1996
etc.), composed of alternate layers of mudstone and sandstones with minor intercalations of
limestone and conglomerates indicating its fluvial deposition environment. The fine grained
mudstones are mostly deep red in color (Fig.4.1), being indicative of high iron contents.
Kiyota et al. (2010) and Aziz et al. (2010) also used Hattian Bala mudstones for their
research.
N 0 500 1000
m
Landslide source
Landslide dam
Landslide source
Landslide
area dam
Quickbird (Date 27/10/2005)
Fig. 4.7 Quick Bird view (27th Oct. 2005) of Hattian Bala landslide
dam (Kiyota et al., 2011)
b) Ishikawa
The mudstone blocks collected form Ishikawa as shown in Fig. 5.8 was later crushed
in the laboratory to specific particles sizes as a necessary adjustment to the apparatus
dimension.
The site is geographically located approximately 36.4610° North latitude and
136.6251° East longitude. From a geological point of view, the source area is Sedimentary
rock formation formed during middle-upper Miocene epoch.
4.3
Ishikawa
JAPAN
5 cm
Fig. 4.8 Mudstone extracted from Ishikawa, Japan Fig. 4.9 Location map of Ishikawa
c) Chiba gravel
Chiba gravel is crushed well-graded angular sandstone from a quarry in Chiba,

Japan. It is a very common granular geomaterial used in geotechnical laboratories in
Japanese universities.
d) Toyoura sand
Toyoura sand originates from weathered granite in Toyoura, Yamaguchi prefecture,
Japan. It has been extensively used as a standard material in geotechnical laboratory
experiments in Japan for the last two decades e.g. Qui et al. (2000) and Wu et al. (2008), in
direct shear test, Tatsuoka et al. (1986) in Plane strain compression (PSC), Goto (1986), in
triaxial compression (TC) etc.
4.2.2 Physical properties of test materials

The particle size distribution (PSD) curves and physical properties of the crushed
mudstones, Chiba gravel, Toyoura sand and Glass beads are presented in Fig 4.10- 4.14.
However, the particle size distribution (PSD) and physical properties of crushed mudstones
do not match each other because of the effect of manual crushing, the large volume of
samples among other factors. The details of other physical properties such as density, void
ratio etc are mentioned in Chapter 5 and 6. The effects of particle side distribution on
strength and deformation (Igwe et. al., 2007; and Iwasaki and Tatsuoka, 1997) are neglected
4.4
in this research. The index properties were determined by following JGS standards. SI is the
slaking index of material (details about SI are discussed in Chapter 3), Gs is the specific
gravity of solids, D50 represents average particles size, Uc corresponds to the coefficient of
uniformity (ratio of D60 and D10) and FC is percentage of particles finer than 2 mm sieve for
coarser material and for finer material i.e. Toyoura sand, FC is percentage of particles finer
than 0.75 mm sieve.
Weight of partices finer than (%)
100
80
60
40
20
Crushed Mudstone
0 (Hattian Bala)
2 3 4 5
Particles size (mm)
Fig. 4.10 Physical properties and particles size distribution of the test
materials from Hattian Bala, Pakistan
4.5

100
80
60
40
20
Crushed mudstone
0 (Ishikawa)
2 3 4 5
Particles size (mm)
materials prepared by crushed soft rocks from Ishikawa, Japan
100
Percent finer by weight (%)
80
60
40
20
0 Chiba Gravel
2 3 4 5
Particles size (mm)
Fig. 4.12 Physical properties and particles size distribution of Chiba gravel
4.6

100
80
60
40
Glass beads
20
0
2 3 4 5
Particles size (mm)
materials (Glass beads)
100
Percent finer by weight (%)
80
60
40
20
0 Toyoura sand
0.01 0.1 1
Particles Size (mm)
Fig. 4.14 Physical properties and particles size distribution Toyoura sand
4.7
4.2.3 General slaking test

a) Slaking index (JGS-2125)
In order to obtain the slaking index of the specimens, accelerated rock slaking tests
(JGS 2125-2006) on crushed mudstone (Hattian Bala and Ishikawa), Chiba gravel and Glass
beads were conducted.
In this test, there pieces of rock lumps which had approximately a volume of 50 cm3
were taken (Fig 4.15). Each piece of the test
sample was placed in a separate container and
was oven-dried at a temperature of 400C. The
distilled water was poured subsequently into the
container until the specimen was fully immersed
(approximately within 1 minute). The specimen
was immersed for 24 hours. This represents one
cycle. Specimens were then subjected to the
same treatment during 3 successive cycles. To Fig. 4.15 Fresh specimen before slaking test
(Hattian Bala mudstone)
find the Slaking index of tested material, all
samples were then described in terms of crack development and any disintegration that
might occur as listed in Table 4.1 provided by JGS- 2125. Sadisun et al. (2002 b) also
proposed a similar slaking classification. Photographs were taken before and after
immersion.
Table 4.1 Slaking classes (JGS 2125-2006)
4.8
b) Static slaking test (Santi, 1998 and Sadisun et al., 2002b)

The procedure of a static slaking index test is quite similar to the dry–wet cyclic
slaking test. In this test, six pieces of parent rock which were taken had a mass between 100
and 150 grams as shown in Fig. 4.16. Each piece of the test sample was then placed in a
separate beaker and oven-dried to a constant mass at 1050 C. The oven dried rock pieces
were cooled at room temperature. The distilled water was then poured into the beakers so
that the rock pieces were covered by at least 10 mm of water. After about 24 hour of
immersion, the samples were washed with water on a 2 mm standard sieve (No. 10 sieve).
The material retained on the sieve was then put again into the beaker, decanted and oven-
dried to a constant mass. The weight of the mass retained on 2 mm (no. 10 sieve) was
recorded. The percentage of loosened sample to initial oven dried mass is calculated and
recorded as a slaking index value (Is) for that cycle, or it can be defined by the following
equation.
Wx  W ' x
IS  ……………….… (4.1)
Wx
Where, Wx = total initial mass of oven dried material; Wx’ = total mass oven dried
material retained on the 2.00 mm sieve
Fig. 4.16 Six set of specimens for static slaking test of Ishikawa mudstone
On the basis of Is value after one cycle, the slaking susceptibility of geomaterials is
evaluated as Table 4.2.
4.9
Table 4.2 Class value of Is1
S.N. Is1 Slaking susceptibility

1 0-2 % Very low
2 2-10 % Low
3 10-25 % Medium
4 25-50 % High
5 50-85 % Very high
6 85-100 % Extremely high
c) Slaking ratio (NEXCO- 100, 2006)

In this test, three sets of specimens of rock lumps were taken. Each set had a mass of
3 kilograms. The specimens were prepared by removing lumps finer than 19 mm and larger
than 37.5 mm. Each set of specimens was kept in the oven until a constant was obtained at
105 0C. The oven dried rock lumps were cooled at room temperature. The distilled water
was then poured into the pan so that the rock pieces were covered by at least 10 mm of
water for 24 hours (Fig. 4.17). This represents one cycle and specimens were subjected to
the same treatment during 5 successive cycles.
Fig. 4.17 Wetting of the Mudstone specimen in water during a slaking ratio
test (Hattian Bala mudstone)
4.10
Then, the samples were washed with water on a 9.5 mm standard sieve. The material
retained on the sieve was then put again into the oven and dried to a constant mass. The
weight of mass retained on 9.5 mm was recorded. The percentage of retained sample on 9.5
mm sieve to initial oven dried mass is calculated and recorded as a slaking ratio value (Sr)
for that cycle, or it can be defined by the following equation.
Wx  W ' x
Sr  ……………………………… (4.2)
Wx
Where, Wx = total initial mass of oven dried material; Wx’ = total mass oven dried material
retained on the 9.00 mm sieve
d) Water absorption test

Slaking of mudstones due to cyclic wetting and drying is influenced by their ability
to absorb water (Saffet, 2000; Erguler and Ulusay, 2009 and Cao et. al., 2006 etc.).
Absorbed water is the water that, during submergence, fills the void spaces present in
mudstones. However, because of the fact that most mudstones contain a certain amount of
clay minerals which can adsorb water during submergence, absorption measurements
usually also include a certain amount of adsorbed water. Water absorption was determined
using a slightly modified procedure form of ASTM method C 97 (ASTM, 1987). The test
requires submerging three sets of dried mudstones samples in water with each sample
weighing at least 20-30 grams (Fig 4.18). To reduce the slaking of the tested samples, the
absorption test was initiated at the natural water content of the samples. After 48 hours, the
samples were removed from the water, surface dried, and weighed. The saturated samples
were then oven dried at 105° C for 24 hours and then weighed. The percent absorption was
calculated as follows:
Ww  Wd
Water absorption  % ……………………… (4.3)
Wd
Where, Ww is saturated weight of samples and Wd is dry weight of samples.

The average value calculated from three sets of samples was defined as the water
absorption of crushed mudstones.
4.11
Fig 4.18 Three sets of pycnometers inside the vacum to release air from the
pores (Ishikawa mudstone)
4.2.4 Specimen preparation

As already mentioned, Slaking characteristics of crushed mudstones (Hattian Bala
and Ishikawa) are analyzed in this study. There are two main reasons for selecting crushed
mudstones. Firstly, the soils or rocks with in a slip layer of landslides are generally in a
crushed state. Secondly, it greatly increases expose (surface) area of mudstones which will
accelerate the slaking process in the laboratory. Moreover, the proper ratio of the shear box
length to the mean particles size should be maintained (Jewell and Worth, 1987).
Both Hattian Bala mudstones and Chiba gravel are naturally crushed geomaterials.
But, Ishikawa mudstones were needed to be crushed to a specific grain size for the
preparation of the direct shear test specimen. Manual crushing with a wooden hammer was
carried out. Both naturally and artificially crushed mudstones were sieved to retrieve
particles finer than 4.75 mm and larger than 2 mm. The basic physical and index properties
of tested materials were determined.
4.2.5 Sieve analysis
The crushed mudstones used in this study are comprised of grains which are
susceptible to disintegration due cyclic wetting and drying. As described in the chapter 3.3,
4.12
the effect of slaking due to cyclic wetting and drying on crushed mudstones can also be
quantified by determining particles degradation after a certain number of cyclic wetting and
drying. The mechanical properties of granular materials are strongly dependent on the
particle size distribution (Igwe et. al., and Iwasaki et. al., 1977). It is necessary to quantify
the particles disintegration due to wetting and drying induced by using a proper index to
analyse the slaking characteristics of geomaterials. In order to quantify particles
disintegration due to slaking, sieve analyses were performed both before and after
experiment
Particles crushing and degradation index

Generally, the amount of fines (FC) measured after the experiment indicates the
extent of particles crushing. However, overall changes in particles sixe distribution cannot
be described adequately using a single aperture, such as a 2 mm sieve because it only
calculates the amount of particles becoming finer than 2 mm sieve size. Therefore, the
fragment size distribution approach was used in this study to better define the particle
disintegration. The sieves used for particles size distribution included the following sizes:
4.75, 4.0, 3.35, 2.80, 2.36 and 2.0 mm. According to the fragment size distribution approach
(Erguler and Shakoor et. al., 2009; Aziz et. al., 2010), a degradation index, ID is defined on
the basis of area under particles size distribution curve before and after experiment as
shown in Fig. 4.19a. Many researchers have used different kinds of indexes for particles
disintegration like grading index, IG (Wood et al., 2008) and relative breaking index, Br (
Einav et. al., 2007). Here, degradation index, ID has been developed from these previous
studies (Aziz et. al., 2010).
Figure 4.19b shows the advantages of ID upon FC to measure particles crushing
during the experiment. For the same value of FC different index, ID values and shape of
particles size distribution curved can be observed. There, degradation index, ID can be used
over all particles crushing for a given samples better than FC. The particles size distribution
curves of all specimens before and after experiment under different test conditions are
presented in the Chapter 5.
4.13

Degradation Index, ID
C D
100 = Area of ABDA
Area of ACDA Before experiment
80 0<ID<1
If ID= 0, No crushing
60 If ID= 1, All crushing
40
B
Before experiment
20
a)
0 A
1 2 3 4 5
Particles size (mm)
Fig 4.19 Definition of the degradation index, ID
C D
100
FC= 33 %, ID= 0.818
80
FC= 33 %, ID= 0.2642
60
FC= Finer than 2 mm
40 ID= Degradation index B
20
b)
0 A
1 2 3 4 5
Particles size (mm)
Fig. 4.19 Effect of shape of particles size distribution curve on the degradation index
4.14
4.2.6 X-Ray diffraction

X-ray diffraction analysis is the
standard method for identifying clay
mineralogy. The analysis provides semi-
quantitative determinations of the various
mineral constituents. The analysis was
performed using a two-theta X-ray
diffractometer. All analyses were conducted
using a copper K-alpha source, equipped with
a monochrome filter. Initial scans covered
Fig. 4.20 Sample preparation for X-Ray
from 2 degrees to 60 degrees two-theta.
Diffraction (Hattian Bala mudstone)
Oriented ceramic-tile type mounts of each
sample were prepared using the 2- micron (0.002 mm) and smaller size particles (Fig 4.20).
This technique required mixing 10 to 20 grams of powder mudstone sample that passed
sieve # 200 with 2 % sodiumhexametaphosphate and distilled water. The 2-micron size was
selected because it is the optimum size for separating the clay minerals from the non-clay
mineral components (Grim, 1969). This arrangement of x-ray scanning enabled the
identification of clay mineral species according to characteristic basal diffraction peak
positions, peak areas and relative intensities. The scans also provided the basis for
estimating the quantity of each clay mineral species, according to a semi-empirical method
developed by many researchers. Hattian Bala and Ishikawa mudstones were used for X-ray
diffraction analysis.
4.3 Modified direct shear apparatus

In the last few decades an enormous growth in the development of numerical tools
likes FEM, DEM etc. for complex geotechnical engineering analysis has been achieved. The
use of such tools demands advanced experiments to obtain the experimental characterization
of the mechanical behavior of materials and to simulate the field condition as accurate as
possible. Laboratory testing has provided the lifeblood for advances in modern geotechnical
engineering. There are a number of advantages that laboratory testing offers; the sample and
the failure modes can be observed physically which may help to understand anomalies and
explain variations in the test results, index tests of the same material can be performed on
4.15
the same specimens, stresses can be directed towards the actual stress path, drainage
condition can be controlled and stress, strain and pore water pressure can be measured from
low strain to failure. Experiments are the only means by which cause and effect can be
established. It has already been noted that an experiment differs from non-experimental
methods in that it enables us to study cause and effect because it involves the deliberate
manipulation of one variable, while trying to keep all other variables constant. However,
there are some drawbacks to laboratory experiment. In the laboratory, it is almost
impossible and expensive to prepare exact specimen with the structure and fabric as in the
field. All samples are disturbed to some extent, which affects their strength and stiffness. A
good quality laboratory requires a variety of specialized equipment and technicians with the
training and skills to properly use that equipment. Despite these drawbacks, there are an
increasing number of ways that laboratory testing can contribute to geotechnical practice.
For this research, modified direct shear apparatus was used for evaluating slaking
characteristics of geomaterials. The details of modified direct shear apparatus are discussed
later.
4.3.1 Application of direct shear

Direct shear test is simple in principle, quick and inexpensive. It is used especially to
determine the shear strength of both cohesive as well as non-cohesive soils. Direct shear
provides essential design parameters in the stability analysis of slope failures or foundation
where lateral displacements are involved in the soil mass (Hanzawa, 1992). The advantages
and disadvantages of direct shear tests are given below.
Advantages of direct shear testing are as follows:
 The test is relatively inexpensive and quick to perform. It is also easier to
interpret the results.
 Rapid drainage can be achieved due to smaller thickness of sample.
 It has been found that soil parameters  and c obtained by direct shear testing are
nearly as reliable as triaxial values obtained by triaxial testing. Typical values
obtained with the direct shear test are 1 to 2 degrees larger than values obtained
with the triaxial test.
 It is good for measuring residual strength values.
4.16
Disadvantages or limitations of the direct shear test (Holtz and Kovacs, 2003) are as
follows:
 Failure occurs along a predetermined failure plane which may not be weakest
plane.
 Non-uniform distribution of shear stress along the failure surface. Initial failure
occurs at the corners and ends of the box, and propagates towards the center.
 Area of sliding changes as the test progresses.
 There is an uncontrolled rotation of principal planes and stresses that occurs
between the start of the test and failure.
4.3.2 Outline of apparatus

A number of factors are to be considered in the design of a direct shear apparatus
(DSA). Indeed, the DSA has a number of inherent drawbacks as already mentioned, mostly
originating from inevitable non uniform stress and strain condition associated with a
progressive failure in the potential (horizontal) shear zone. Numerical analyses of the
deformation and failure of granular material in direct shear by finite element method (FEM)
or more recently by DEM showed that the principle axes of distributed contact force and
initial rupture zone may first develop diagonally, not horizontally from the specimen edges
(Cui et al., 2006).
To minimize the effects of these inherent drawbacks and to match as much closely a
“quasi-simple shear” mode in the potential horizontal shear zone, attempts have been made
to optimize the direct shear apparatus design by modifying the conventional type (Jewell et
al., 1987; Shibuya et al., 1997; Lings et al., 2004). After Shibuya et al. (1997) and Wu et al.
(2008), Fig. 2-3 summarizes the different types of DSA: the conventional direct shear
apparatus can be categorized into Type A (ASTM D3080-90) while Type B and Type C
refer to modified one in the Fig. 4.21.
The major problems with the conventional one concern: a) the rotation of the upper
box; b) the side wall frictions. That is, the normal load is applied to the centre of the top
loading platen that is not fixed against rotation. As a consequence, when subjected to lateral
shearing, the distribution of normal stress along the central horizontal shear plan becomes
inevitably biased (so does the shear stress) to maintain the equilibrium of moment within
the specimen , which results in a more progressive mobilization of the shear strength along
the central horizontal plan. The vertical load applied at the top loading platen becomes
4.17
different from the value acting on the shear plan. This is because of vertical friction acting
along the inner walls of upper shear box caused by the volume changes of the specimen is
due to the free vertical movement of the upper shear box. To minimize the above mentioned
problems in the modified direct shear apparatus (Type B and C in Fig. 4.21), the top loading
platen is fixed against rotation. Similarly, to minimize the effect of side wall friction in
Type C (the one which was used in this research), the vertical load should be measured at
the bottom of the lower shear box (No. 6 in Fig. 4.22) (Shibuya et al., 1997).
W W W
Not fixed
Loading plate Fixed Loading plate Fixed Loading plate
Fixed
Shear plane Shear plane Shear plane
Winner
Type A Type B Type C

Winner
Fig. 4.21 Different type of Direct shear box (Shibuya et al., 1997; Wu et al., 2008)
4.18
1 Bellofram cylinder
2 Shear load cell (LC-1)
3 Normal load cell (LC-2)

4 LVDT-1
5 LVDT-2
6 Normal load cells (LC-3&4)
7 Low-friction ball slide
8 Upper shear box
9 Lower shear box
10 Friction load cells (LC-5&6)
11 Sponge
12 Low friction supporting rail
13 Rigid stopping bar
14 Gear loading device
15 Top plate
16 Acrylic made water tank
Fig. 4.22 Schematic diagram of modified direct shear apparatus
4.3.3 General description of the apparatus

(1) Shear box
It consists of two metal boxes i.e. upper and lower shear box (No. 8 and 9 in Fig.
4.22) with smooth inner walls. Shear box has inside dimension 20 cm* 20 cm* 10.8 cm.
Jewell and Worth (1987) proposed that the proper ratio of shear box length to the mean
particle size should be in the range 50-280. According to this criterion, the current shear box
dimension is suitable for testing specimen with a mean particle size (D50) up to 4 mm.
4.19
However, the size of this shear box is too large as compare to the conventional one. So, the
ratio of shear box length to mean particles size derived from conventional shear box may
not be appropriate for this large modified shear box.
The lower box (No. 9) moves on a low-friction supporting rail (No. 12). The upper
box (No. 8) is fixed by means of two rigid bars
(No. 13) to prevent both vertical and horizontal
displacements. A piece of sponge tape (No. 11)
is glued to the upper periphery of the lower box
to prevent sand from spilling out from the
opening during shearing as well as to prevent
the inside volume of shear box from increasing
because of shear displacement so that the
volume change occurring in the specimen is due
solely to the dilation or contraction of sand. Fig. 4.23 Hole on the bottom of shear box
A suitable prescribed opening (as
described later) can be set up in between the lower and upper boxes by inserting an
appropriate number of spacers. A hole is provided on the bottom of shear box (Fig. 4.23) to
facilitate the supply of water to saturate the specimen and also for removing water from the
specimen.
(2) Vertical loading system

Vertical load is applied through air cylinder (No. 1). Normal load cell (No. 3) is used
to measure applied upper load. The apparatus has a possible feedback control on normal
load to impose any prescribed stress path in the shear stress-normal stress space. The
vertical movement of the top shear box is not free; the vertical load applied at the top
loading platen becomes different from the value acting on the shear plane due to the vertical
friction acting along the inner walls of the top shear box caused by the volume changes of
the specimen. To measure the exact loading on the shear plane with respect to the side wall
friction, Normal load cells (No. 6) are used to measure the vertical load at the bottom of the
lower shear box. Rate of consolidation can be controlled in an automated way.
4.20
(3) Shear loading system

The shear load is applied by a
high precision gear loading system
controlled by servo-motor (Fig. 4.24).
The working mechanism of loading
system is explained briefly here. The
rotation of gear No. 2 is ensured by an
A-C analog motor (No. 13) through
the speed reduction gear (No. 1). The
gear No. 2 is continuously rotating in
the same direction while transmitting
the rotation to the two bevel gears No.
3 and 4, where two bevel gears No. 3
Fig. 4.24 High precision gear loading device
and 4 are rotating in fixed but in
(Tatsuoka et al., 1994) used to apply shear load
opposite direction. Similarly, the (No. 13 in Fig 4.22)
rotation is transmitted to the shaft and loading piston either by one of the Electro-magnetic
clutches (EMCs) (No. 6 and 7) depending on the direction of shaft. When one EMC is
connected, the other one is unconnected and vice-versa, depending on the applied loading
direction. In this way, by using the gear loading device, the direction of shaft rotation can be
reversed without problems of backlash. However, in this research, only the loading
condition was used i.e. only forward rotation of shaft. Moreover, the displacement can be
precisely controlled with an accuracy of 1 μm in an automated way. It can smoothly switch
displacement between different loading and displacement stages i.e. load control loading
phases, or between load relaxation stages or between sustained loading and constant
displacement rate loading etc. Similarly, it can change the displacement rate stepwise or
gradually by a prescribed factor by removing/ adding an appropriate number of speed
reduction gears. The above mentioned two functions are primordial to evaluate viscous
properties of geomaterials accurately.
The importance of all these features in order to obtain reliable data of direct shear
tests on granular material was demonstrated by Shibuya et al. (1997), Qui et al. (2000) and
Wu et al. (2008). More details of the apparatus are given by Duttine et al. (2008, 2009).
Finally, in this study, another improvement was made concerning the instantaneous
measurement of water content of specimen inside shear box. A moisture sensor was inserted
4.21
into the lower shear box (Fig. 4.24) to measure the water content of specimen inside shear
box. The details of the moisture sensor are presented in 4.3.3..
4.3.4 Moisture sensor

Decagon’s ECH2O (EC-5)
sensor (Fig. 4.26) was used to
measure the water content of the
specimen inside the shear box. It is a
very simple and compatible
instrument for the direct shear box.
Decagon’s ECH2O (EC-5) sensor
measures the volumetric water
content of the soil by measuring the
dielectric constant of the soil, which Fig. 4.25 Moisture sensor inserted inside shear box
is a strong function of water content. However, not all soils have identical electrical
properties. Due to the variations in soil bulk density, mineralogy, texture, salinity and
surrounding environment, the dielectric constant of soil is no constant. The manufacturer,
Decagon recommends that ECH2O (EC-5) sensor users conduct a soil-specific calibration
for best possible accuracy in volumetric water content measurements. An independent
researcher (Czarnomski et al., 2005 etc.) found that resolution, precision and repeatability of
ECH2O (EC-5) sensors are excellent.
Fig. 4.26 Moisture sensor

To minimize the possible data errors as well as to a develop relationship between
water content by weight (w) and displayed Raw value (R in mv), the calibration of moisture
sensor under same conditions as in experiment was carried out. ECH2O (EC-5) calibration
generally followed the general procedure for calibrating capacitance sensors outlined by
Starr and Paltineanu (2002). Calibration graph of ECH2O (EC-5) sensor for Hattian Bala
4.22
and Ishikawa mudstones are shown in Fig. 4.27 and 4.28 and data are summarized in Table
4.3 and 4.4 respectively.
20
Water content by weight, w (%) 18
16 w = 0.03329*R - 18.94339
2
14 R = 0.9976
12
10
8
6
4
2
0
-2
400 600 800 1000 1200
Display Raw value, R (mv)
Fig. 4.27 Relationship between water content, W and display raw value, R (Hattian Bala)
Water content by weight, w (%)
80
w = 0.13448*R - 66.083
2
60 R = 0.9971
40
20
400 600 800 1000 1200

Display Raw value, R (mv)
Fig. 4.28 Relationship between water content, W and display raw value, R (Ishikawa)
4.23
Table 4.3 Calibration of Moisture sensor for Hattian Bala mudstone
Water
Sensor out put Wet Average Average
Dry soil Mass & Dry Soil content Volumetri
Sample soil Water Volumetric
Container mass + Volume soil bulk by c water
S.N. volume mass + content by water
Raw mass (g) container of water mass density mass content
(cm3) contain mass of soil content
m3/m3 constant, (g) (cm3) (g) (g/cm3) of soil (cm3/cm3)
er (g) (%) (m3/m3)
R (%)
0 0 571 0.000
29.06 52.177 110.01 107.64 2.37 78.58 1.506 3.016 0.045

1 0.091 673 3.013 0.045
29.23 52.177 109.33 106.99 2.34 77.76 1.490 3.009 0.045
29.12 52.177 115.24 109.7 5.54 80.58 1.544 6.875 0.106
2 0.202 803 7.335 0.113
29.04 52.177 114.92 108.71 6.21 79.67 1.527 7.795 0.119
29.04 52.177 119.9 110.76 9.14 81.72 1.566 11.185 0.175
3 0.408 1036 29.2 52.177 121.5 109.61 11.89 80.41 1.541 14.787 15.51 0.228 0.243
28.27 52.177 125.02 111.51 13.51 83.24 1.595 16.230 0.259
29.06 52.177 111.41 109.93 1.48 80.87 1.550 1.830 0.028
4 0.049 623 1.887 0.029
29.14 52.177 111.96 110.38 1.58 81.24 1.557 1.945 0.030
29.35 52.177 107.92 105.52 2.4 76.17 1.460 3.151 0.046
5 0.083 663 29.15 52.177 112.68 110.14 2.54 80.99 1.552 3.136 3.119 0.049 0.047
29.53 52.177 111.09 108.66 2.43 79.13 1.517 3.071 0.047
29.28 52.177 111.08 107.42 3.66 78.14 1.498 4.684 0.070
6 0.116 702 29.42 52.177 111.16 107.56 3.6 78.14 1.498 4.607 4.614 0.069 0.070
29.33 52.177 113.19 109.54 3.65 80.21 1.537 4.551 0.070
27.47 50.941 111.11 106.56 4.55 79.09 1.553 5.753 0.089
7 0.14 731 5.708 0.088
27.74 50.941 111.8 107.29 4.51 79.55 1.562 5.669 0.089
4.24
27.91 50.941 108.38 104.04 4.34 76.13 1.494 5.701 0.085

28.32 50.941 114.15 108.14 6.01 79.82 1.567 7.529 0.118
8 0.188 787 28.19 50.941 114.6 108.35 6.25 80.16 1.574 7.797 7.709 0.123 0.121
28.03 50.941 113.29 107.12 6.17 79.09 1.553 7.801 0.121
4.25
Table 4.4 Calibration of Moisture sensor for Ishikawa mudstone
Sensor out put Average

Average
Wet soil Dry soil Mass & Dry Soil Water Water Volumetric
Sample Volumetric
Raw Container mass + mass + Volume of soil bulk content by content water
S.N. volume water
m3/m3 constant, mass (g) 3
container containe water mass density mass of by mass content
(cm ) content
R (g) r (g) (cm3) (g) (g/cm3) soil (%) of soil (m3/m3)
(m3/m3)
(%)
1 0 466 0.000 0 0.000 0.000
29.12 52.177 66.14 65.5 0.64 36.38 0.697 1.759 0.012
2 0.006 494 1.76 0.012
29.04 52.177 66.23 65.59 0.64 36.55 0.701 1.751 0.012
29.04 52.177 67.95 65.444 2.506 36.40 0.698 6.884 0.048
3 0.03 563 29.2 52.177 67.85 65.3 2.55 36.1 0.692 7.064 7.20 0.049 0.051
28.27 52.177 68.04 65.32 2.72 37.05 0.710 7.341 0.052
29.06 52.177 69.24 65.12 4.12 36.06 0.691 11.425 0.079
4 0.039 594 11.20 0.078
29.14 52.177 69.4 65.42 3.98 36.28 0.695 10.970 0.076
29.35 52.177 71.21 65.37 5.84 36.02 0.690 16.213 0.112
5 0.055 630 29.15 52.177 71.2 65.33 5.87 36.18 0.693 16.224 16.28 0.113 0.112
29.53 52.177 71.21 65.34 5.87 35.81 0.686 16.392 0.113
4.25
29.28 52.177 72.15 65.43 6.72 36.15 0.693 18.589 0.129

6 0.073 654 29.42 52.177 73.25 65.39 7.86 35.97 0.689 21.852 21.15 0.151 0.146
29.33 52.177 73.95 65.6 8.35 36.27 0.695 23.022 0.160
27.47 52.177 75.85 65.3 10.55 37.83 0.725 27.888 0.202
7 0.111 719 27.74 52.177 76.55 65.42 11.13 37.68 0.722 29.538 30.08 0.213 0.217
27.91 52.177 77.74 65.43 12.31 37.52 0.719 32.809 0.236
28.32 52.177 94.13 65.54 28.59 37.22 0.713 76.814 0.548
8 0.256 1060 28.19 52.177 94.92 65.34 29.58 37.15 0.712 79.623 79.17 0.567 0.565
28.03 52.177 95.75 65.43 30.32 37.4 0.717 81.070 0.581
4.26
4.4 Direct shear test procedure

4.4.1 Initial setting and loading
Experimental setup was done carefully as following.
1. A 10 mm think sponge was glued to the upper periphery of the lower shear box
(Fig. 4.29). The upper box along with the appropriate number of spacers was
also installed on lower shear box (Fig. 4.30). The two boxes were tightly
maintained by temporary bolts to avoid any movement of the boxes during the
specimen preparation and compaction (Fig. 4.30).
2. The moisture sensor was inserted into the lower shear box. The position of the
moisture sensor was carefully maintained (Fig. 4.24).
3. A porous stone was laid at the bottom of lower shear box. In addition, filter
paper was also laid on the porous stone as shown in Fig. 4.31. The porous stone
was used only for saturated and cyclic wetting and drying experiments.
4. The specimen was divided into 7 sub-layers to obtain uniform compaction and to
prevent from particles segregation. Crushed mudstone was slowly poured from
negligible height (See fig. 4.32). The specimens were not thoroughly compacted
to prevent from particle breakage and to simulate the natural condition in the
slope. However, a wooden parallelepiped was used (See Fig. 4.33) to carefully
compact each layer of the specimen to achieve the prescribed height, if each
layer height was found more than prescribed value.
5. After the final layer, the specimen was carefully leveled. Filter paper was first
laid and the porous stone was laid later. But, in case of the drying monotonic
loading test, the final layer was completed by placing a surrounding cardboard
mould to collect the surplus crushed mudstones. After light compaction, the top
end of the specimen was carefully leveled by sliding a sharp metal bar to remove
the excessive crushed mudstone towards the cardboard mould. The cardboard
was removed and specimen was covered by filter paper (Fig. 4.34).
6. The specimen density was calculated from the mass in shear box. The difference
between the initial (before experiment) calculation, and the final calculation
(after specimen preparation) was very small.
4.27
7. The DS box was slowly translated under the vertical piston. The upper box was
attached to rigid bars and rigid bars were tightly attached to the vertical bars of
the apparatus using nuts and bolts (Fig. 4.35).
8. The load piston of the gear loading device was temporarily connected to the
shear box frame to prevent the sliding of the lower shear box. The vertical piston
with loading plate was lowered by manually decreasing the air pressure in the
cylinder until it reached very close to the specimen.
9. The temporary bolts used to connect the lower and the upper shear box, and the
spacer used to maintain a fixed gap between upper and lower shear boxes were
removed very carefully.
10. The load piston of the gear loading device which was temporarily connected to
shear box frame to prevent sliding of lower shear box, was disconnected from
the shear box frame.
Spacer
Fig. 4.29 Attached sponge to the lower Fig. 4.30 Assembled upper and lower
periphery of the upper shear box shear box with prescribed spacing
4.28
2nd filter paper
1st pore
stone
Fig. 4.31 Pore stone and filter paper Fig. 4.32 Deposition of materials of prescribed
mass
Fig. 4.33 Compaction of specimen Fig. 4.34 Covering specimen after final
layer
Horizontal rigid bar
Fig. 4.35 Final setting before loading applied
11. The connection between the vertical piston and the top cap was checked and
made tight. Then, all load cells and both horizontal and vertical LVDTs (Linear
4.29
variable differential transducer) were checked and calibrated. Starting from this
stage, the readings of vertical LVDT and vertical LCs (load cell) were recorded
(in a before consolidation process (b4con) file.)
12. Then, the vertical piston was lowered by manually decreasing the air pressure in
cylinder. An initial pre-load of about 80 N was usually applied. This pre-load
was of negligible influence on vertical deformation and consolidation.
13. The vertical LVDT (No. 5 in Fig. 4.22) was positioned and initialized.
14. Consolidation was started gradually at a typical rate of 1 kPa per minute until the
normal stress (σv) reached the nominally prescribed value. The Typical normal
stress reached at the end of consolidation for the tests performed in this study
was 5 kPa. As the vertical displacement rate was recorded, it could be checked
that the vertical displacement rate became almost null and that the vertical
compression eventually stopped after several minutes.
In this study, the loading process during the test consisted of three stages as shown in Fig.
4.36
Initial loading
Before the start of initial loading, the following steps were done:
1) Connect the load piston of the gear loading device to the DS box frame.
2) Turn on the motor control unit and servo-control unit.
3) Setup the horizontal LVDT (No. 4 in Fig. 4.23)
4) Reset the reading of all displacement transducers and friction load cells to zero.
5) Arrange the settings in computer according to prescribed stress ratio or stress
path (Fig. 4.32). The loading process is fully automated. All the different steps of
loading history (stress path, Fig. 4.36) can be simply written in a text file that
can be directly implemented from the program interface by using computer as
follows:
a. Monotonic loading at a constant vertical stress starting/ending at a given
shear stress or shear displacement. (In this experiment, initial shear stress
was applied up to R* σv, σv = 5 kPa)
4.30
b. Anisotropic loading (both shear and vertical loading) starting/ending at

given shear stress/ vertical (In this experiment, shear stress was applied
up to R* σv, σv = 50 kPa).
c. Sustained loading stage starting at a given shear stress level for a
specified time period,
d. Constant acceleration test for a specified time period until a given shear
displacement rate and starting at a given shear stress or shear
displacement
6) Both shear stress, τ, and vertical stress, σv, were applied to the dry specimen
gradually keeping their ratio, R (=τ /σv, σv=50 kPa), constant. The value of R
represents the inclination of slopes. The shearing process is then fully automated.
In this research, the value of R was used as 0.3, 0.5 and 0.7 which are equivalent
to 16.70, 26.560 and 34.990 slopes respectively. Similarly, the value of σv was
fixed to 50 kPa. The value of σv depends on the thickness of the overburden soil
or depth of the slip surface etc. As mentioned in Chapter 2, the slaking induced
landslides are generally shallow depth landslides. So, to represent the condition
of shallow landslides the value of 50 kPa was used for σv.
Creep loading (with cyclic wetting and drying)

The Sustained loading stage
Monotonic loading
was started at a given shear stress
t = 15, 25 & 35kPa
Shear stress (t)
level for a specified time period.

The Time period depends on the
g 0 .7 )
din 0.5 & Creep with cyclic wetting
type of materials used and the lo a ,
tial 0.3 and drying 3 times
I ni R =
nature of experiment. tio
s ra
res
1) In case of dry monotonic (S t
Initial condition
loading (ML) tests, after
Vertical stress (s) s = 50kPa
the prescribed τ and σv (=
Fig. 1 Stress path for MS001, MS002 & MS003
50 kPa) values were Fig.4.36 Stress path
reached, sustained loading was continue until the stabilization of both shear and
vertical deformations was achieved. It took several hours for the stabilization of
both shear and vertical deformations.
4.31
2) In case of other tests simulating cyclic wetting and drying, after the stabilization
of both shear and vertical deformations, a prescribed number of cyclic wetting
and drying was carried out under constant stress conditions. Details of wetting
and drying process are discussed later.
Monotonic loading
After creep loading (with prescribed cyclic wetting and drying), finally a monotonic
shear loading was applied at a constant rate of s (0.2 mm/min) to evaluate the stress-strain
characteristics of geomaterials under constant stress condition (σv = 50 kPa) until the
specimen’s residual stage was reached.
4.4.2 Wetting
A large container (Fig. 4.37) was added to the direct shear apparatus for water
pounding to saturate the specimen through the sides of the shear box. Similarly, two
cylinders were used as shown in Fig 4.38 to
maintain proper water head for smooth supply
of water inside the specimen. One cylinder was
used to supply water inside the specimen
through the hole (Fig. 4.23) provided on the
bottom of the shear box while another cylinder
was used to supply water outside the shear box.
In addition, one small tank was also used for
continuous supply of water to both cylinders as
shown in Fig. 4.23.
First the water was supplied in a large
Fig. 4.37 Large container to saturate the
container (outside of shear box) up to about 10 specimen inside shear box
mm height. Then, the wetting was carried out
by pouring the distilled water from the bottom of the lower shear box. When the water level
reached the opening level inside the specimen, water was supplied to the large container
again up to the level higher than the opening to prevent the outflow of mudstones particles
from the shear box. Again, the water was supplied from the bottom of the shear box until
the specimen was fully immersed. The water flow should be controlled carefully not to
scour the specimen during wetting. It took about 30 min for wetting the specimen.
4.32
The wetting process always started after stabilization of both shear and vertical
displacements due to either the loading or drying processes. Similarly, the water content
during drying should also be almost constant before starting a new wetting process.
Fig. 4.38 Experimental setup to saturate specimen inside the shear box
4.4.3 Drying
When both shear and vertical displacements due to immersion led to stabilize, water
was drained out through the pipes connected to the holes at the bottom of the shear box and
outside of the shear box (in between wall of large container and shear box). When
gravitational water was drained out almost completely, the dry air was pumped from the
bottom of the shear box by using a dry pump (Fig. 4.39). Gravitational water is free water
moving through the soil by the force of gravity. It is largely found in the macro-pores of soil
and it drains rapidly through the water table in all soils except the most compacted ones.
The shear box was covered by silica gel (Fig. 4.40) as shown in Fig. 4.41. Both the shear
box and water containers were also covered by a vinyl sheet to control the humidity around
the shear box as shown in Fig.4.42, while room temperature was maintained at 300 C. The
water content of the specimen was measured frequently using a moisture sensor through the
Pro-Check. The room temperature and humidity were monitored as shown in Fig.4.41.
4.33
Fig. 4.39 Dry air pump to supply dry air Fig. 4.40 Silica Gel
Fig. 4.41 Shear box covered by silica gel Fig. 4.42 covered by vinyl sheet
and temperature and humidity measurement
4.34
CHAPTER 5
TEST RESULTS
5.1 General
To describe the slaking effects on the strength and deformation characteristic as well as
particles crushing of mudstones, a series of direct shear tests with other preliminary
experiments were conducted. Similar tests were also conducted on conventional granular
material such as Chiba gravel and Toyoura sand. Non-slakeable Glass beads was used for few
tests as well.
This chapter presents the preliminary data obtained from the slaking test and the main
experimental results obtained from the direct shear tests. Materials properties, test conditions,
procedures, and stress paths were explained in details in the preceding chapters. All the test
results are summarized in the main body of the dissertation. The results presented in this
chapter will be helpful in comparison slaking characteristics of non-conventional geomaterials
obtained from crushed soft rocks (Hattian Bala and Ishikawa mudstone) with the non-slakable
conventional materials (Chiba gravel, Toyoura sand and Glass beads).
Along with the properties of various test materials used in this study, the effect of the
slaking index, density of the specimen, water content before wetting and stress ratio are the key
parameters for explaining and comparison of the test results. Similarly, the influences of the
cyclic wetting and drying on the physical and mechanical properties of various geomaterials
used in this study will be discussed in the succeeding chapter.
5.1
`Chapter 5 Test results
5.2 General slaking test

5.2.1 Slaking index test
In order to obtain the slaking index of the specimens, accelerated rock slaking tests
(JGS 2125-2006) on crushed mudstones (Hattian Bala and Ishikawa), Sand stone and Glass
beads were conducted.
a) b)
Fig. 5.1 Hattian Bala mudstone piece a) before slaking test b) after slaking test with minor
cracks
a) b)
Fig. 5.2 Ishikawa mudstone piece a) before slaking test b) Crumbled mudstone after
slaking test
5.2
a) b)
Fig. 5.3 Sand stone a) before slaking test b) after slaking test (Kiyota et al., 2011)
As described in Chapter 4.2.3, JGS 2125-2006 guideline identifies slaking classes of the
sandstone and the mudstone specimens with respect to time during the first drying-wetting
cycle. Figures 5.1, 5.2 and 5.3 show the overall changes of the mudstone and sand stone
specimens before and after completing the tests (i.e. after the three drying-wetting cycles).
Small cracks and small bubbles appeared on surfaces of the Hattian Bala mudstone
specimens 30 minutes after the first pouring while the Ishikawa mudstone specimens converted
into number of fragments. The Hattian Bala mudstone specimens however did not exhibit any
further changes during the subsequent wetting/drying cycles, and their shapes were kept intact
(estimated slaking class = 1). However, the Ishikawa mudstone specimen crumbled and original
shape could not recognized after the three drying and wetting cycles (estimated slaking class =
3).
Similarly, Sandstone specimens showed little sign of slaking (estimated slaking class =
0), while small cracks and small bubbles appeared on surfaces of the mudstone specimens 30
minutes after the first pouring. The slaking index of Glass beads specimens were also evaluated
and found as level 0.
5.3
5.2.2 Static slaking test

In order to obtain the slaking index value of the specimens, Static slaking test (Santi,
1998 and Sadisun et al., 2002b) on crushed mudstones (Hattian Bala and Ishikawa), were
conducted.
The slaking index values of the specimens obtained were as follows:
Table 5.2 Static slaking test (Hattian Bala)
After experiment
Specimen Initial dry
(retained weight on
No. weight, gm
2 mm sieve), gm
1 120.35 120.05
2 141.46 140.73
3 149.01 148.89
4 127.35 127.34
5 129.81 129.61
6 134.26 134.26
802.24 800.88
The static slaking index of Hattian Bala mudstone was found as 0.17.
Table 5.2 Static slaking test (Ishikawa)
After experiment
Specimen Initial dry
(retained weight on
No. weight, gm
2 mm sieve), gm
1 120.35 117.25
2 141.46 136.58
3 149.01 145.38
4 127.35 124.36
5 129.81 124.98
6 134.26 130.58
802.24 779.13
The static slaking index of Hattian Bala mudstone was found as 2.88.
Higher static slaking index value indicates the higher vulnerability to slaking.
5.4
5.2.3 Slaking ratio test

In order to obtain the slaking ratio of the specimens, slaking ratio test (NEXCO- 100,
2006) on crushed mudstones (Hattian Bala and Ishikawa) were conducted.
The slaking ratios of the specimens obtained were as follows:
Table 5.3 Slaking ratio test (Hattian Bala)
Initial weight, After experiment

Specimen
gm (19 mm - (retained weight on
No.
37.5 mm) 9.5 mm sieve), gm
1 3000 2895.25
2 3000 2902.65
3 3000 2919.26
9000 8717.16
The slaking ratio of Hattian Bala mudstone was found as 96.85.
Table 5.4 Slaking ratio test (Ishikawa)
Initial weight After experiment

Specimen
(gm) (19 mm - retained on 9.5 mm
No.
37.5 mm) sieve
1 3000 40.5
2 3000 50.12
3 3000 59.78
9000 150.4
The slaking ratio of Hattian Bala mudstone was found as 1.67.
Higher slaking ratio indicates the durability of mudstone to slaking.
5.2.4 Water absorption

Slaking of mudstones due to cyclic wetting and drying is influenced by their ability to
absorb water (Saffet, 2000; Erguler and Ulusay, 2009 and Cao et. al., 2006 etc.). Absorbed
water is the water that, during submergence, fills the void spaces present in mudstones.
In order to obtain the water absorption value of the specimens, water absorption test
(ASTM method C 97) on crushed mudstones (Hattian Bala and Ishikawa) were conducted.
5.5
The water absorption of the specimens obtained was as follows:
Table 5.5 Water absorption (Hattian Bala)

Weight of
Weight of saturated Water
Container Weight of dry sample and
container, sample and absorption,
no. container, w2 (gm)
W1 (gm) containers, W%
w3 (gm)
1 29.18 105.15 103.59 2.05
2 29.22 104.38 102.85 2.04
3 29.16 103.85 101.95 2.54
Average water absorption,
2.21
W %
Table 5.6 Water absorption (Ishikawa)
Weight of
Weight of saturated Water
Container weight of dry sample and
containers, sample and absorption,
no. container, w2 (gm)
W1 (gm) container, W%
w3 (gm)
1 29.28 74.65 47.52 59.80
2 29.32 73.26 45.98 62.08
3 28.53 69.58 43.25 64.14
Average water absorption,
62.01
W %
Higher water absorption value indicates the higher vulnerability to slaking.
5.3 Direct shear test with cyclic wetting and drying

Stress-strain-volume change characteristics from monotonic direct shear tests on
crushed mudstone (Hattian Bala and Ishikawa) as well as other materials (Chiba gravel,
Toyoura sand and Glass beads) at constant vertical stress (σv=50 kPa) under dry, saturated and
cyclic wetting and drying conditions are presented this section. The test results obtained from
the cyclic wetting and drying under constant creep loading will be interpreted and discussed to
examine their behaviour on natural slope or when used as construction materials for
embankments. All these tests were performed under strain controlled conditions at constant
5.6
relatively slow rate 0.2 mm/min to a maximum shear displacement of 15 mm. The basic
properties of tested materials are listed in Table 5.7.
Table 5.7 Basic properties of tested materials

Stress
Initial Initial Density Test condition
ratio
S.N. Sample density void before ML during creep and
during
(g/cm3 ) ratio(e) ((g/cm3 ) ML
creep, R
1 PreM001 0.3 1.482 0.741 1.508 Creep1) & ML3)
2 PreM002 0.5 1.529 0.688 1.549 Creep1) & ML3)
3 PreM003 0.7 1.553 0.661 1.577 Creep1) & ML3)
Creep & ML under
4 PreM005 0 1.563 0.651 1.573 dry condition
Creep & ML under
Creep & ML under
6 PreM007 0.5 1.558 0.655 1.567 dry condition
Creep & ML under
Creep & ML under
Creep (dry and
9 PreM010 0.3 1.479 0.744 1.518 wetting) & ML3)
Creep (dry and
10 PreM011 0.5 1.529 0.688 1.549 wetting) & ML3)
Creep (dry and
11 PreM013 0.7 1.529 0.688 1.569 wetting) & ML3)
12 PreG014 0.3 1.558 0.656 1.589 Creep1) & ML2)
Creep & ML under
13 PreG015 0 1.584 0.577 1.604 dry condition
Creep (dry and

14 PreG016 0.5 1.599 0.561 1.598 wetting) & ML3)
Creep Creep (dry and

15 PreM017 0.8 1.593 0.568 wetting) & ML3)
failure
Creep & ML under
16 PreG018 0.5 1.529 0.688 1.596 dry condition
5.7
Creep (dry and

17 PreG019 0.5 1.589 0.572 1.595 wetting) & ML3)
Creep & ML under
18 PreG020 0.3 1.587 0.573 1.613 dry condition
Creep (dry and
19 PreG021 0.3 1.589 0.572 1.601 wetting) & ML3)
Creep & ML under
Creep & ML under
Excessive Creep (dry and
22 PreM024 0.5 0.683 2.366 displacem wetting) & ML2)
ent
Creep & ML under
Creep (dry and
24 PreM026 0.3 0.68 2.384 0.724 wetting) & ML3)
25 PreM027 0.3 0.68 2.38 0.713 Creep2) & ML3)
26 PreM028 0.5 0.68 2.38 1.54 Creep2) & ML3)
Creep & ML under
Creep (dry and
28 PreM030 0.5 1.536 0.68 1.587 wetting) & ML3)
Creep & ML under
29 PreC010 0.5 1.566 0.648 1.542 dry condition
Creep (dry and
30 PreC011 0.5 1.529 0.792 1.536 wetting) & ML3)
Creep & ML under
31 PreT050 0.5 1.529 0.792 1.536 dry condition
Creep (dry and
32 PreT052 0.5 1.549 0.709 1.555 wetting) & ML3)
1) Creep with cyclic wetting and drying for 3 times, 2) Creep with cyclic wetting and drying for 2 times 3) ML
under saturated condition
5.8
5.3.1 Hattian Bala

In this study, three cycles of wetting and drying were carried out on Hattian Bala
mudstone specimens under three different anisotropic consolidation conditions to analysis
the deformation and strength characteristics due to slaking by using a modified direct shear
apparatus. A series of the monotonic loading tests on dry and saturated specimens were also
performed to compare strength and deformation characteristics with those of the cyclic wetting
and drying creep test. As already described, the slaking index (JGS 2132) of the mudstone was
evaluated as level 1, while the slaking ratio (NEXCO-110, 2006) was 96.85 %. Similarly, the
index properties of Hattian Bala mudstone were explained in Chapter 4.
Figure 5.4-5.7 show the typical instantaneous response of creep deformations and water
content of the specimen during the wetting and drying cycles for three specimens; PreM003,
PreM002 and PreM001. PreM002, PreM003 and PreM001 experienced three cycles of wetting
and drying while PreM028 experienced only one and half cycles. The time t= 0 corresponds to
the start of creep loading. For all specimens, the creep shear and vertical displacements
converged on constant values after several hours of initial loading.
Each cyclic wetting and drying creep test took about one month to complete. Influence
of wetting in the first cycle upon shear displacement appears to be significant for all specimens
(Fig. 5.4a -5.7a). For the second wetting processes, the increment of shear displacements are
relatively small, almost 1/10 times the increment of shear displacement in the first wetting (Fig.
5.4a-5.6a). But, in case of PreM001, the increment of shear displacement during second wetting
is almost negligible (Fig. 5.6a). Similarly, for the third wetting processes, the increments of
shear displacements are equal to zero for all specimens. Nakano et al. (1998) and Panabokke
and Quirk (1956) reported that the slaking level of clay aggregates became higher as the initial
water content of the specimen became lower. Therefore, it can be understood that the maximum
displacement was observed during the first wetting because the specimen in this study was
prepared by oven-dried crushed mudstone.
In addition, from Fig. 5.4a- 5.6a, the creep shear displacement caused by wetting seems
to be decreased with progress of cycle, almost zero during the third wetting, even the water
content of specimens before wetting is relatively lower about 0.7 %. This may be attributed to
the specimen densification due to previous wetting and drying processes (Nakano et al., 1998).
5.9
Figure 5.4b-5.7b show considerable vertical displacement that occurred in each wetting
step. This expansive behavior of crushed mudstone would consist of two phases, swelling
caused by water absorption of expansive clay mineral and dilatancy due to shearing. Expansion
of clay minerals also cause cracking in the grains which ultimately lead to disintegration.
Cardoso and Alonso, 2009 also explained swelling induced particles disintegration in
mudstones. However, expansive clay mineral like montmorillonite of smectite group was not
found clearly from X-ray diffraction (XRD) analysis of Hattian Bala mudstones.
The vertical displacement, however, in case of R= 0.3, is almost negligible except in the
first wetting because of having higher water content (more than 3 %) of specimen before the
second and third wettings (Fig. 5.6).
Figure 5.4-5.7 also show that water content decreases gradually during the first and
second drying processes. Initially, no appreciable creep deformation is found to occur at higher
water contents. When the water content becomes about 2.5 %, both vertical and shear
displacements occur progressively with water loss and finally tend towards an asymptotic value
at water content of about 0.8 %. However, in case of PreM003, the water content decreases
gradually during drying process but become almost constant when water content is about 3 %.
No response of drying process upon both shear and vertical displacement is found in PreM003
(Fig. 5.6).
One of the noticeable behaviors observed in these experiments is a quite large creep
deformation during the drying processes (Fig. 5.4-5.7). Slaking phenomena cause loss of intra-
particles cementation and tensile failure of the weakly crystalline bonded granular materials
due to drying induced pore water suctions. (Czerewko and Cripps, 200; Moropoulou et al.,
2004; Karoglou et al., 2005; Soe et al., 2010). Reduction of pore water which lead to shrinkage
and disaggregation of fabric especially around discontinuities during dry process.
Consequently, such an evolution of soil grains produces rounded particles with relatively high
sphericity and smooth circumference which sequentially decreases the interlocking behaviour
(angle of internal friction) of granular medium. Particles slides on each other during shrinkage
causing shear deformation simultaneously. The strength and stiffness gained during drying
process are opposing to the further increase in shear displacement with drainage and
evaporation of pore water which lead to shrinkage and disaggregation of fabric especially
5.10
Increment vertical displacement, v (mm)

2.5 3
25
Hattian Bala (PreM003), SI = 1, in=1.558 gm/cm
st nd rd
2.0 1 wetting 2 wetting 3 wetting 20
st nd
1 drying 2 drying
Water content, w (%)

1.5 15
1.0 10
Initial state
0.5 w = 2.5 % 5
0.0 0
Vertical displacement
R= 0.7 Water content
-0.5 -5
0 96 192 288 384 480 576 672 768
a) Elapsed creep time, t (hour)
Increment shear displacement, s (mm)
3
12 Hattian Bala (PreM003), SI = 1, in=1.558 gm/cm 24
st nd rd
1 wetting 2 wetting 3 wetting
10 20
st nd

1 drying 2 drying
8 16
6 12
4 8
2 w = 2.5% 4
0 Shear displacement 0
Initial state R= 0.7 Water content
-2 -4
0 96 192 288 384 480 576 672 768
b) Elapsed creep time, t (hour)
Fig 5.4 Time histories of water content, shear and vertical displacement (a) shear (b)
vertical content under cyclic wetting and drying for 3 times for R= 0.7 (PreM003)
5.11

6 3
Hattian Bala (PreM002), SI = 1, in= 1.531 gm/cm
nd rd
5 1st wetting 2 wetting 3 wetting 20

st nd
4 1 drying 2 drying
15
3
10
2
1 5
w = 2.5%
0 Shear displacement 0
Initial state R = 0.5 Water content
0 96 192 288 384 480 576
2.5 25
3
Hattian Bala (PreM002), SI = 1, in= 1.531 gm/cm
nd rd
2.0 1st wetting 2 wetting 3 wetting 20
st nd

1 drying 2 drying
1.5 15
Contraction
1.0 10
0.5 5
w = 2.5%
0.0 0
Initial state Expansion
Ver. displacement
-0.5 R = 0.5 Water content -5
0 96 192 288 384 480 576
b) Elapsed time creep (hour)

5.12

3
2.4 Hattian Bala (PreM001), SI = 1, in= 1.486 gm/cm 24
st nd rd
2.0 st nd
20

1 drying 2 drying
1.6 16
1.2 12
0.8 8
0.4 4
0.0 Shear displacement 0

Initial state R = 0.5 Water content
-0.4 -4
0 96 192 288 384 480 576
Increment Vertical displacement, v (mm)
2.5 Hattian Bala (PreM001), SI = 1,  = 1.486 gm/cm3 25

in
st nd rd
2.0 st nd
20
1 drying 2 drying

1.5 15
Contraction
1.0 10
0.5 5
0.0 Initial state 0

Expansion Ver. displacement
-0.5 -5
R = 0.3 Water content
0 96 192 288 384 480 576
5.13

3.5 Water content 21
3.0 Shear displacement 18
2.5 15

2.0 12
Hattian Bala (PreM028)
1.5 9
SI = 1 3
1.0 in= 1.529 gm/cm 6
0.5 3
0.0 0
R = 0.5
-0.5 -3
0 96 192 288 384 480 576 672
Increament vertical displacement, v (mm)
1.5 Vertical displacement

20
Water content
16

1.0
Hattian Bala (PreM028) 12
SI = 1 3
in= 1.529 gm/cm 8
0.5
w = 2.60 % 4
0.0 0
-4
R = 0.5
-0.5
0 96 192 288 384 480 576 672
vertical content under cyclic wetting and drying for 2 times for R= 0..5 (PreM028)
5.14
Macro pore
Micro pore
Crushed mudstone
Fig. 5.8 Macro and micro pores system
Fig. 5.9 Relationship between water content and drying induced

vertical deformation (Braudeau et al. 2004)
around discontinuities. Therefore the drying induced deformation of crushed mudstones is very
complex which could be challenging and crucial importance for strategic geotechnical
structures.
5.15
As seen from the Fig. 5.4, 5.5 and 5.7 both shear and vertical displacement during
drying are composed of two well defined curvilinear parts. Figure 5.8 shows schematic sketch
of crushed mudstone particles which consist of two pore systems, inter (macro) and intra-
primary (micro) porosity (Asaoka et al., 1997; Nakano et al., 1998 and Braudeau et al. 2004). It
was assumed that water leaves the crushed mudstones from the inter-pedal (macro) pores
causing shrinkage of inter-pedal (macro) pores (first part of curvilinear). When the inter-pedal
(micro) pores empty out, intra-primary (micro) pores begin to shrink, losing its water content
(second part of curvilinear) as shown in Fig. 5.9.
Figures 5.10 -5.15 show the shear stress-deformation-volume change relationship for
the all dry, saturated and the one with cyclic wetting and drying history specimens. Here
“volume change” denotes the ratio of shear and vertical displacement, and its negative value
indicates dilation. The specimens of PreM001, PreM002 and PreM003 were subjected to the
alternative step of loading history including three cycles of alternative wetting and drying while
the specimens PreM0010, PreM011 and PreM013 were subjected to alternative step of loading
including saturation only. Similarly, PreM007, PreM008 and PreM009 were subjected original
loading history mentioned in Chapter 4. The time t=0 corresponds to the start of initial loading
(anisotropic consolidation).
As described in previous chapter, initially, both shear stress, τ, and vertical stress, σ v,
were applied to the dry specimen gradually keeping their ratio, R (= τ/σv, σv = 50 kPa),
constant. Figures 5.10a, 5.12a and 5.14a show stress-strain behaviours under stress ratios 0.7,
0.5 and 0.3 respectively, where stress ratios 0.7, 0.5 and 0.3 are equivalent to slope ground
having 350, 270 and 170 inclination respectively. The shear displacements from the origin, O to
A (Fig. 5.10a, 5.12a and 5.14a ) correspond to the initial loading. It shows that large shear
deformation was obtained when stress ratio, R (= τ/σv) is high.
When both shear and vertical displacements due to initial loading had reached an
almost constant value, the first wetting was carried out by supplying distilled water. This creep
loading process represent a situation that slope ground is saturated by rain fall under constant
stress condition. The shear displacements from A to B (Fig. 5.10a, 5.12a and 5.14a ) are the
combination of three probable phenomena such as creep, slaking and wetting induced
displacement under constant loading. In addition, different deformation response was observed
5.16
for consecutive specimens under similar condition during initial loading and saturation of
various specimens. This is possibly due to the irregular rearrangement of soil particles during
specimen preparation, particles disintegration of some grains and or different
collapsing/swelling direction upon saturation
As already mentioned, three specimens PreM001, PreM002 and PreM003 were
subjected to three cycles of wetting and drying under constant loading. So, the shear
displacements from B to C (Fig 5.10a, 5.12a and 5.14a) represent the cyclic wetting and drying
induced displacement.
The segments A to D (Fig 5.10a, 5.12a and 5.14a) of the dry specimens are the stress-strain
relationship during monotonic shear loading under constant stress (σv = 50 kPa) condition.
Similarly, the portions B to D are the stress-strain relationship during monotonic loading under
constant stress (σv = 50 kPa) for the saturated specimens while the segments C to D represent
the stress- strain behaviours during monotonic loading after three cycles of wetting and drying.
Figure Fig 5.10b, 5.12b and 5.14b show the shear deformation-volume change
relationship for all specimens under three different stress conditions. All specimens showed
contractive behaviour during initial loading. In addition, all specimens exhibit swelling
behaviours during first wetting. PreM002 and PreM003 undergo shrinkage during the drying
step and swelling in the wetting step as aforementioned (Fig 5.10b and 5.12b). However, shear
deformation- volume change relationship is not clearly visible during the cyclic wetting and
drying for the specimen PreM001 under stress ratio 0.3 (Fig. 5.14b).
Finally, all specimens under stress ratio 0.7 and 0.5 exhibited dilative behavior during
monotonic shear loading. However, it is seen that the residual state is no longer unique. For the
specimens under stress ratio 0.3, it is noticed that for the specimens with dry condition
exhibited a dilative behavior but that the saturated and the one with three cycles of wetting and
drying specimens loses such dilative nature. The volumes of full saturated specimens are
contractive nature.
Similarly, Figures 5.16 shows the shear stress-deformation-volume change relationship
for the specimen (PreM028) experienced one complete cycle of wetting and drying and
monotonic shear loading was applied under dry condition. PreM028 was also subjected to the
5.17
alternative step of loading history including one cycle of alternative wetting and drying. The
time t=0 corresponds to the start of initial loading (anisotropic consolidation).
As described in previous paragraphs for other test, the shear displacements from the
origin, O to A (Fig. 5.16a) correspond to the initial loading maintaining stress ratio 0.5. When,
the both shear and vertical displacement due to initial loading became almost constant, the
wetting was carried out as explained in Chapter 4. The segment A-B (Fig. 5.16a) probably is
the results of creep, slaking and wetting induced displacement under constant loading during
one complete cycle of wetting and drying. Then, the monotonic loading was applied to evaluate
stress-strain characteristics of Hattian Bala mudstone under dry condition after one complete
cycle of wetting and drying.
Figure 5.16b shows the shear deformation-volume change relationship during
monotonic loading. The specimen (PreM028) showed contractive behaviour during initial
loading. In addition, it exhibited swelling behaviour during wetting while shrinkage behaviour
during drying process. Finally, it exhibited dilatative behaviour during monotonic shear loading
under dry condition after one complete cycle of wetting and drying.
Figures 5.11, 5.13, 5.15 and 5.17 show the relationship between the shear stress ratio, R,
the shear displacement, s, together with the associated volume change (i.e. the vertical
displacement, v) for the Hattian Bala mudstone specimen. The values of s and v were set at zero
at the beginning of monotonic loading. In addition, the values of friction angles, φ d, (assuming
apparent cohesion cd =0) at peak strength were also indicated in these figures. The saturated
and the one with cyclic wetting and drying mudstone specimen exhibit largely different stress-
displacement features from those for dry specimen. The specimens were prepared at low initial
density therefore peak and residual stress state (strain softening) even in dry conditions are not
clearly visible. The initial stress-strain response becomes stiffer under both saturated and cyclic
wetting and drying tests as compared to dry tests. This is probably due to aging and viscous
effects on granular material because saturated and the one with cyclic wetting and drying test
specimens experienced relatively longer creep loading (Prisco and Imposimato, 1996; Tatsuoka
et al., 2000 and 2001; Nawir et al., 2002). So, both clear peak stress ratio and post peak stress
softening appeared from monotonic shear loading on both saturated and the one with cyclic
wetting and drying test conditions.
5.18
50 Hattian Bala
SI= 1
R= 0.7
Shear stress, t kPa)
40 D
A C
B
30
20
3
10 Dry (PreM008) ( in gm/cm )
3
Saturated (PreM013) (in gm/cm )
Three cycle (PreM003, c.f. Fig.
Fig. 5.3)
5.4)
0 O 3
(in   gm/cm )
0 2 4 6 8 10 12 14 16
a)
Shear displacment, s (mm)
4
1- Initial loading Hattian Bala
3 2 - Wetting SI= 1
Vertical displacement, v (mm)
3- Drying 2
R= 0.7
2 4- Monotonic loading
2 3
1
0 4 4
1
-1
-2
Dry (PreM008)
-3 Saturated (PreM013)
Three cycles (PreM003, c.f. Fig. 5.3)
5.4)
-4
0 2 4 6 8 10 12 14 16
b) Shear displacement, s (mm)
Fig 5.10 a) stress-deformation b) volume changes characteristics under R= 0.7

during experiment (Hattian Bala mudstone)
5.19
1.00 Dry (PreM008) ( = 1.565 gm/cm )

3
3
Saturated (PreM013) ( = 1.569 gm/cm )
0.95 After three cycle (PreM003, ( = 1.577 gm/cm
3
0.90 p = 42.61
0
0
0.85 p = 38.90
Stress ratio, R
0.80
0
p = 37.56
0.75
Hattian Bala
0.70 SI = 1
0.65
0.60 0 2 4 6 8 10 12 14
Shear displacement, s (mm)
a)
3
Vertical displacement, n (mm)
3
Dry (Prem008) ( = 1.565 gm/cm )
3
Saturated (Prem013)( = 1.569 gm/cm )
2 3
After three cycle(PreM003)( = 1.577 gm/cm )
1
Contraction
0
Hattian Bala
SI= 1
-1 R = 0.7
-2 Expansion
-3
0 2 4 6 8 10 12 14
Fig. 5.11 Effect of slaking on a) stress-deformation b) volume changes

characteristics of Hattian Bala mudstone specimen under R= 0.7
5.20
50
Hattian Bala
SI= 1
40 D
Shear Stress, t kPa) R= 0.5
30
A C
B
20
3
Dry (PreM007) (in gm/cm )
3
10 Saturated (PreM011) (in  gm/cm )
3
Saturated(PreM030) (in gm/cm )
0 After three cycles (PreM002, c.f. Fig.
Fig. 5.4)
5.5)
O (in gm/cm )
3
0 2 4 6 8 10 12 14 16
a) Shear displacement, s (mm)
4
Hattian Bala 1- Initial loading
3 SI= 1 2 - Wetting
R = 0.5 3- Drying
2 2 4- Monotonic loading
1
4
0 3
1
-1
Dry (PreM007)
After three cycles (PreM002, c.f. Fig.
Fig. 5.5)
5.4)
-4
0 2 4 6 8 10 12 14 16
Fig 5.12 a) stress-deformation b) volume changes characteristics under R= 0.5
during experiment (Hattian Bala mudstone)
5.21
1.0
Hattian Bala p = 42.61
0
SI= 1
0.9
0
p = 36.80
0
p = 36.89 p = 37.59
Stress ratio, R
0
0.8
0.7
3
Dry (PreM007)(in1.549 gm/cm )
0.6 3
Saturated (PreM011) in1.567 gm/cm )
3
Saturated (PreM030)(in1.549 gm/cm )
After three cycles (PreM002)
0.5 3
(in1.587 gm/cm )
0 2 4 6 8 10 12 14
3 3
Dry (PreM007)( gm/cm )
3
Saturated (PreM011)( gm/cm )
2 3
Saturated (PreM030)( gm/cm )
3
After three cycles (PreM002)( gm/cm )
1
Contraction
0
-1
Hattian Bala Expansion

-2
SI= 1
R=0.5
-3
0 2 4 6 8 10 12 14

5.22
50 Hattian Bala
SI= 1
R= 0.3
40 D
Shear stress,t kPa) 30
20
A 3
Dry (PreM009) (in gm/cm )
10 B C Saturated (PreM010)(in gm/cm )
3
Three cycles (PreM001, c.f. Fig. 5.5)

5.6)
0 (in gm/cm )
3
O
0 2 4 6 8 10 12 14 16
3
Dry (Prem009)
Saturation (Prem010)
2 After three cycles (Prem001, c.f.
c.f. Fig.
Fig.5.6)
5.5)
1
1- Initial loading
0 2 - Wetting
3- Drying
4- Monotonic loading
-1
-2 Hattian Bala
SI = 1
-3 R = 0.5
-2 0 2 4 6 8 10 12 14 16
Fig 5.14 a) stress-deformation b) volume changes characteristics

under R= 0.3 during experiment (Hattian Bala)
5.23
1.0
Hattian Bala p = 41.35
0
0.9 SI = 1
0
p = 38.46
0.8
Stress ratio, R
0.7 0
p = 36.13
0.6
0.5
3
0.4 Dry(fresh)(PreM009) (in gm/cm )
3
Saturation(PreM010) (in gm/cm )
3
0.3 After three cycles (PreM001) (in gm/cm )
0 2 4 6 8 10 12 14
3
3
Dry (Prem009) (in gm/cm )
3
2 Saturation (Prem010)(in gm/cm )
3
After three cycle (Prem001)(in gm/cm )
1
Contraction
0
Hattian Bala
-1 SI = 1
Expansion R = 0.5
-2
-3
-2 0 2 4 6 8 10 12 14 16

5.24
60
R = 0.5
Shear stress, t (kPa) 50
40
D
30
A B
20
Hattian Bala (PreM028)
10 SI = 1
ρin = 1.529 gm/cm3
0 O
-2 0 2 4 6 8 10 12 14 16
4
1- Initial loading
R = 0.5
3 2- Wetting
4 3- Drying
2 4 - Monotonic loading
1 2
0 3
1
-1
-2 Hattian Bala (PreM028)
SI = 1
-3
ρin = 1.529 gm/cm3
-4
-2 0 2 4 6 8 10 12 14 16
Fig 5.16 a) stress-deformation b) volume changes characteristics under

R= 0.5 during experiment (Hattian Bala mudstone)
5.25
1.2
0 p= 46.68
0 Hattian Bala
1.1 p= 46.10
SI = 1
1.0
Stress ratio, R
0.9
0
0.8 p= 36.47
0.7
0.6 3
Dry (PreM029) ( = 1.558 gm/cm )
3
0.5 Saturated (PreM011, c.f. Fig. 5.13)( = 1.549 gm/cm )
3
Dry after one cycle (PreM028) ( = 1.564 gm/cm )
0.4
0 2 4 6 8 10 12 14
4
3
Dry (PreM029)( = 1.558 gm/cm )
3
3 Saturated (PreM011, c.f. Fig. 5.13)( = 1.549 gm/cm )
3
Dry after one cycle (PreM028) ( = 1.564 gm/cm )
2
Hattian Bala
1 Contraction
SI = 1
0 R = 0.5
-1
-2 Expansion
-3
-4
0 2 4 6 8 10 12 14

5.26
It is quite clear that the peak stress ratio of the specimens experienced three cycles of
wetting and drying is higher than that of saturated specimens. This may be attributed to the
specimen densification due to previous wetting and drying processes and aging and viscous
effects due to relatively longer creep loading as compared to saturated specimen. In case of
PreM030 (saturated test) and PreM002 (cyclic wetting and drying test) which had almost same
density, the peak stress ratio of PreM030 (saturated test) is higher than PreM002 (cyclic
wetting and drying test) even if PreM002 experienced longer creep time (Fig. 5.13a). The peak
shear strength of the saturated and the one with cyclic wetting and drying test specimens is
reduced by about 25 % as compared to the dry test specimens. So, it is concluded that the
stress-strain response of crushed mudstones show higher peak stress ratio under dry condition
with enormous decrease in peak shear stress after saturation essentially due to slaking induced
particles disintegration.
As already discussed, all specimens except PreM001 and PreM010, under the three
different condition exhibited dilative behavior during monotonic shear loading (Fig. 5.11b,
5.13b and 5.15 b). However, it is seen that the residual state is no longer unique. Generally, the
position of residual state is unique under constant stress. The position of residual state varies
with the degree of saturation. Monotonic shear loading was applied under fully saturated
condition after one and three cycles of wetting and drying; there is a considerable difference in
the position of residual states. So, this result indicates that particles crushing due to cyclic
wetting and drying causes the difference in position of residual state. The PreM001 and
PreM010 exhibited dilatative behaviour followed by contraction. This is attributed to the
relatively low density of the specimen before monotonic loading (see Table 5.1).
Fig. 5.17 shows the stress-strain-volume change behaviour of three specimens namely;
PreM029, PreM011 and Prem028. The specimen PreM029 correspond to dry and fresh
(without any cyclic wetting and drying in laboratory) specimen. The specimen PreM011 was
subjected to wetting under constant load condition, while the PreM028 subject to one complete
cycle of wetting and drying. From the figure, it is quite clear that the reduction in peak shear
strength during wetting is almost recovered during drying phase. However, there is small
reduction is peak shear stress after one complete cycle of wetting and drying as compared to
fresh specimen. Similarly, density, aging and viscous effects are neglected. The specimen with
5.27
one complete cycle of wetting and drying specimen experience relatively large creep time as
compared to fresh specimen. This may be one of the reasons to recover the peak shear stress
again. However, the resistivity to the slaking of Hattian Bala mudstone is also high. In general,
it is concluded that the slaking effect on strength parameters due to cyclic wetting and drying is
not found significant for the Hattian Bala mudstone specimen.
5.3.2 Ishikawa mudstone

The influence of the slaking on the mechanical properties and particle disintegration by
simulating two cycles of wetting and drying under constant sustained loading were
investigated.
Figure 5.18 shows the typical instantaneous response of creep deformations and water
content of the specimen during the wetting and drying cycles for the Ishikawa mudstone
specimens; PreM027. The time t= 0 corresponds to the start of creep loading. For all
specimens, the creep shear and vertical displacements converged on constant values after
several hours of initial loading.
Each cyclic wetting and drying creep test took about one month to complete. Influence
of wetting in the first cycle upon shear displacement appears to be significant for the Ishikawa
mudstones specimens (Fig. 5.18a). Similarly, significant positive vertical displacement (about 3
%) found during saturation process. Positive value of vertical deformation is taken as
contraction. The vertical displacement of Ishikawa mudstone specimens during saturation being
positive is indicative of slaking induced disintegration of soil particles as well as collapse of
relatively open structure of these materials having high initial void ratios. For the second
wetting processes, the increment of shear displacement is insignificant as compared to
increment shear displacement during first wetting. However, substantial negative vertical
displacement occurs during second wetting processes while the vertical displacement during
first wetting was positive (Fig. 5.18a).
Figure 5.18 also shows that water content decreases gradually during drying processes.
Initially, no appreciable creep deformation is found to occur at higher water contents. When the
water content became about 62 %, both vertical and shear displacements occur progressively
with water loss and finally tend towards an asymptotic value at water content of about 12 %.
5.28
As in the Hattian Bala mudstone specimen, both shear and vertical displacement during
drying are composed of two well defined curvilinear parts.
The observations made in first wetting are in strong agreement with a study by Neves
and Pinto (1988) on prediction of collapse settlement due to saturation rock-fill, mainly in
embankment dams. Therefore, it is inferred that slaking induced deterioration of soil grains
which ultimately leads to enormous compression upon sustained loading can be critical for the
embankments constructed with crushed mudstones, for soil structure interaction of foundation
placed on such soils as well as progressive slope failures.
Shear displacement caused by wetting seems to be decreased with progress of cycle,
almost zero during the second wetting. This may be attributed to the specimen densification
due to previous wetting and drying processes (Nakano et al., 1998) and relatively higher water
content about 12 % before second wetting (Nakano et al., 1998 and Panabokke and Quirk,
1956).
The vertical displacement of Ishikawa mudstone during second wetting is contrary to
the vertical displacement during first wetting (Fig. 6.6b) Considerable negative vertical
displacement (expansion) occurred during second wetting. This is attributed to the densification
of specimen during previous wetting and drying.
Relatively large creep deformation during the drying processes was observed which is
similar to the Hattian Bala mudstones specimens.
Shear stress-deformation-volume change relationship for the all dry, saturated and the
one with cyclic wetting and drying history specimens are shown in Fig. 5.19 and 5.20. Here
“volume change” denotes the ratio of shear and vertical displacement, and its negative value
indicates dilation. The specimens of PreM027 was subjected to the alternative step of loading
history including two cycles of alternative wetting and drying while the specimens PreM026
was subjected to alternative step of loading including saturation only. Similarly, and PreM009
were subjected original loading history mentioned in Chapter 4. The time t=0 corresponds to
the start of initial loading (anisotropic consolidation).
5.29
Increament shear displacement, s (mm)

st nd
10 1 wetting st
1 drying 2 wetting 100
8 80

W = 62 %
6 60
Ishikawa (PreM027)
4 40
SI = 3
ρin = 0.68 gm/cm3
2 20
0 R = 0.3 Shear deformation 0

Water content
0 96 192 288 384 480 576
Increament vertical displacement, v (mm)
st st nd
10 1 wetting 1 drying 2 wetting 100
8 80

W
w = 62 %
6 60
Ishikawa (Prem027)
4 SI = 3 40
ρin = 0.68 gm/cm3
2 20
R = 0.3 Vertical displacement 0
0
Water content
0 96 192 288 384 480 576
Fig 5.18 Time histories of water content, shear and vertical displacement (a) shear
(b) vertical content under cyclic wetting and drying for 2 times for R= 0.3 (PreM027)
5.30
60
R = 0.3
50
Ishikawa mudston
SI = 3
Shear stress, t (kPa)
40 3
D
= 0.68 gm/cm
30
20 B
A C
10
Dry (PreM025)
One cycle (PreM026)
0 O Three cycles (PreM027, c.f. Fig. 5.18)
-2 0 2 4 6 8 10 12 14 16
10
R = 0.3
4
8 Ishikawa
SI= 3 3
6 in = 0.68 gm/cm 5
1- Initial loading
st
4 2- 1 wetting
st
3- 1 drying
2 nd
2 3 4- 2 wetting
5
0
1
-2 Dry (PreM025)
Saturated (PreM026)
-4 Three cycles (PreM027, c.f. Fig. 5.18)
-2 0 2 4 6 8 10 12 14 16
b)
Shear displacement, s (mm

R= 0.3 during experiment (Ishikawa mudstone)
5.31
1.2
1.1
1.0
0.9
Stress ratio, R
Ishikawa mudstone
0.8
SI = 3
0.7
0.6
0.5
3
0.4 Dry (Prem025) ( = 0.687 gm/cm )
3
Saturated (Prem026)( = 0.724 gm/cm )
3
0.3 After two cycles (Prem027)( = 0.74 gm/cm )
0 2 4 6 8 10 12 14
a)
4
3
Expansion
2
Ishikawa
1 SI = 3
0 R = 0.3
-1
Contraction
-2 3
Dry (PreM025) ( = 0.687 gm/cm )
3
-3 Saturated (PreM026) ( = 0.724 gm/cm )
3
After two cycles (PreM027) ( = 0.74 gm/cm )
-4
0 2 4 6 8 10 12 14

characteristics of Ishikawa mudstone specimen under R= 0.3
5.32
Figure 5.19a shows stress-strain behaviours of Ishikawa mudstone dry, saturated and
the one with cyclic wetting and drying specimen. As described in previous chapter, initially,
both shear stress, τ, and vertical stress, σ v, were applied to the dry specimen gradually keeping
their ratio, R (= τ/σv, σv = 50 kPa), constant. For the Ishikawa mudstone specimens, the creep
load was applied maintaining stress ratio equal to 0.3 which is equivalent to slope ground with
170 inclinations. The shear displacements from the origin, O to A (Fig. 5.19a) correspond to the
initial loading.
When both shear and vertical displacements due to initial loading had reached an
almost constant value, the first wetting was carried out by supplying distilled water. This creep
stress condition. The shear displacements from A to B (Fig. 5.19a) are the combination of three
probable phenomena such as creep, slaking and wetting induced displacement under constant
loading. In addition, different deformation response was observed for consecutive specimens
under similar condition during initial loading and saturation of various specimens. This is
possibly due to the irregular rearrangement of soil particles during specimen preparation,
particles disintegration of some grains and or different collapsing/swelling direction upon
saturation
As already mentioned, the specimen PreM027 was subjected to two cycles of wetting
and drying under constant loading. So, the shear displacements from B to C (Fig 5.19a)
represent the cyclic wetting and drying induced displacement. The segments A to D (Fig 5.19a)
of the dry specimens are the stress-strain relationship during monotonic shear loading under
constant stress (σv = 50 kPa) condition. Similarly, the portions B to D are the stress-strain
relationship during monotonic loading under constant stress (σv = 50 kPa) for the saturated
specimens while the segments C to D represent the stress- strain behaviours during monotonic
loading after three cycles of wetting and drying.
Figure 5.19b shows the shear deformation-volume change relationship for all specimens
under different conditions. All specimens showed contractive behaviour during initial loading.
In addition, the positive vertical displacement (contraction) was found during the first wetting.
PreM027 exhibited shrinkage behaviour during the drying process with increase in shear
displacement. The vertical displacement of Ishikawa mudstone during second wetting is
5.33
contrary to the vertical displacement during first wetting. Considerable negative vertical
displacement (expansion) occurred during second wetting.
Stress-strain-volume change characteristics from monotonic direct shear tests on
crushed mudstone specimens of Ishikawa material at constant vertical stress (σv=50 kPa) under
dry, saturated and cyclic wetting and drying conditions are presented in Fig. 5.20.
Figures 5.20 show the relationship between the shear stress ratio, R, and the shear
displacement, s, together with the associated volume change (i.e. the vertical displacement, v)
for the Ishikawa mudstone specimen. The values of s and v were set at zero at the beginning of
monotonic loading. In addition, the values of friction angles, φd, (assuming apparent cohesion
cd =0) at peak strength were also indicated in these figures. The saturated and the one with
cyclic wetting and drying mudstone specimen exhibit largely different stress-displacement
features from those for dry specimen. In case of saturated (PreM026) and the one with cyclic
wetting and drying (PreM027) specimen, neither clear peak stress ratio nor post-peak stress
softening appeared and the R-s curve converged monotonously (Fig. 5.20a). Consequently, the
stress level at which significant plastic (or irreversible) deformation develops (i.e. the start of
yielding or yield stress) is different between these two specimens. The peak shear strength of
the saturated and the one with cyclic wetting and drying test specimens is reduced by about 45
% as compared to the dry test specimens. Relatively large inherent void ratios, e (2.3), slaking
induced disintegration during wetting and particle rearrangement of weak soil grains are the
most responsible factors for significant reduction in peak shear strength during wetting.
The dry specimen (PreM025) exhibited the dilative behaviour during monotonic shear
loading while the saturated and the one with cyclic wetting and drying specimen (PreM026 and
PreM027) continued to exhibit contractive behaviour (5.20b). Such stress-displacement
characteristics are true to typical soft soils, indicating that the slaking process causes the
examined mudstone material to disintegrate and therefore changes its mechanical features
remarkably. The test, which experienced two cycles of wetting and drying (PreM027), seems
stiffer than the saturated one. This may be attributed to the specimen densification during
previous wetting and drying processes and aging and viscous effects due to relatively longer
creep loading as compared to saturated specimen (Prisco and Imposimato, 1996; Tatsuoka et
al., 2000 and 2001; Nawir et al., 2002).
5.34
Ishikawa mudstone showed relatively higher slaking index (level 3) and very small
slaking ratio, which indicate highly susceptible to slaking. Similarly, large shear displacement
and volumetric compression during saturation of crushed mudstones (Ishikawa) as well as a
drastic loss of strength parameters upon submergence are revealed in this study. It is concluded
that the observed soil behaviour can be critical for the slope stability. In addition, it can be
critical for the embankments constructed with similar materials and bearing capacity of the
place on such soft sedimentary rocks having higher slaking index or ratio.
5.3.3 Glass beads

The slaking index (JGS 2132) of the Glass beads was evaluated as level 0, while the
slaking ratio (NEXCO-110, 2006) was 100 %. Similarly, the index properties of Glass beads
were explained in Chapter 4. The influences of cyclic wetting and drying on Glass beads were
evaluated by simulating two cycles of wetting and drying under constant sustained loading.
Figure 5.21 shows the typical instantaneous response of creep deformations of the Glass
beads specimen (PreG0014) during the wetting and drying cycles. The time t= 0 corresponds to
the start of creep loading. For all specimens, the creep shear and vertical displacements
converged on constant values after several hours of initial loading.
The water was supplied to the Glass beads specimen after 5 hours creep loading. The
Influence of wetting in the first cycle upon both shear and vertical displacement appears to be
negligible (Fig. 5.21). Then, after 18 hours of creep loading under saturated condition, the
water was removed from the specimen. The water content of the specimen decreased gradually
and became almost zero after 4 days. As in the wetting, the influences of drying upon both
shear and vertical displacements were not noticed. Similarly, for the second wetting processes,
the increments of shear and vertical displacements were again almost zero.
Figure 5.22 and 5.123 show stress-strain-volume change behaviours of Glass beads
under both dry and saturated conditions for the stress ratio 0.5. As described in previous
chapter, initially, both shear stress, τ, and vertical stress, σv, were applied to the dry specimen
gradually keeping their ratio, R (= τ/σv, σv = 50 kPa) constant. The shear displacements from
the origin, O to A (Fig. 5.22a) corresponds to the initial loading. When both shear and vertical
displacements due to initial loading had reached an almost constant value, the saturation was
5.35

2.0
1st wetting 1st drying 2nd wetting
1.5
Glass beads (PreG014)

1.0 3
in = 1.584 gm/cm
R = 0.3
0.5
0.0
Water content was almost zero
before 2nd wetting
-0.5
0 24 48 72 96 120 144
2.0
Increment vertical diplacement, v (mm)
1st wetting 1st drying 2nd wetting

1.5
Glass beads (PreG014)

1.0 3
in = 1.584 gm/cm
R = 0.3
0.5
0.0
Water content was almost zero
before 2nd wetting
-0.5
0 24 48 72 96 120 144
b)
Elapsed creep time, t (hour)
vertical content under cyclic wetting and drying for 3 times for R= 0.3 (PreG014)
5.36
Glass beads
30 R = 0.3
Shear stress, t (kPa)
20
D
A
10
3
0 O Dry (PreG020) ( = 1.589 gm/cm
3
Saturated (PreG021)  = 1.587 gm/cm )
-2 0 2 4 6 8 10 12 14 16
3
1- Initial loading
Vertcial displacement, v (mm)
Glass beads
2- Saturation
2 R = 0.3 3- Monotonic loading
1 2 3 Contraction
0
1
-1
Expansion
-2
Dry specimen(PreG020)
Saturated specimen (PreG021)
-3
0 2 4 6 8 10 12 14 16
R= 0.3 during experiment (Glass beads)
5.37
0.7
0 Glass beads
p = 28.84 p = 29.47
0
0.6
Stress ratio, R
0.5
0.4
0.3
3
Dry(PreG020)( = 1.613 gm/cm )
3
Saturated (PreG021)( = 1.601 gm/cm )
0.2
0 2 4 6 8 10 12 14 16
3
Glass beads
2 R= 0.3
1
Contraction
0
Expansion
-1
-2 3
Dry(PreG020) ( = 1.613 gm/cm )
3
Saturated (PreG021) ( = 1.613 gm/cm )
-3
0 2 4 6 8 10 12 14 16
Fig. 5.23 Effect of saturation on a) stress-deformation b) volume changes
characteristics of Glass beads specimen under R= 0.3
5.38
carried out by supplying distilled water from the bottom of shear box. This creep loading
process represent a situation that slope ground is saturated by rain fall under constant stress
condition. The segments A to D (Fig 5.22a) of the specimens are the stress-strain relationship
during monotonic shear loading under constant stress (σv = 50 kPa) condition..
Figure 5.22b shows the shear displacement-volume change relationship for both
specimens under dry and saturated conditions under stress ratio 0.5. Both specimens showed
contractive behaviour during initial loading while these specimens exhibited dilative behavior
during monotonic shear loading.
It can be seen from the Fig. 6.23a that the difference in stress-strain behaviours between
dry and saturated conditions is insignificant under different stress condition for the Glass beads
specimen. Clear peak and residual stress states or strain softening under dry as well as saturated
conditions are observed. It is certainly due to the durable (non slakable) nature of grains which
are unaffected by the presence of water.
However, the volume change behaviour during monotonic loading under saturated
condition is different than that of dry condition (5.23b). Strangely, the saturated specimens
showed more dilative behaviour as compared to dry specimen during monotonic loading. So,
details investigation by using other apparatus like Tri-axial test is required to explain the
volume change behaviour of glass beads under dry and saturated conditions.
5.3.4 Chiba gravel and Toyoura sand

The influences of saturation on sand stone specimens (Chiba gravel) and silica sand
(Toyoura sand) were evaluated by applying water to the dry specimens under sustained loading
condition.
Figure 5.23 and 5.24 shows the typical instantaneous response of creep deformations of
the Chiba gravel (PreC011) and Toyoura sand (PreT052) specimens during saturation. The time
t= 0 corresponds to the start of creep loading. For all specimens, the creep shear and vertical
displacements converged on constant values after several hours of initial loading. The water
was supplied to the specimens after 17 hours creep loading. The Influence of saturation upon
both shear and vertical displacement appears to be negligible (Fig. 5.23 and 5.24).
Figures 5.25 and 5.26 show stress-strain-volume change behaviours of Chiba gravel and
Toyoura sand under both dry and saturated conditions respectively. As described in previous
5.39
sections, initially, both shear stress, τ, and vertical stress, σv, were applied to the dry specimen
gradually keeping their ratio, R (= τ/σv, σv = 50 kPa) constant. The shear displacements from
the origin, O to A (Fig. 5.25a and 5.26a) correspond to the initial loading. When both shear and
vertical displacements due to initial loading had reached an almost constant value, the
saturation was carried out by supplying distilled water from the bottom of shear box. This creep
stress condition. The segments A to D (Fig 5.25a and 5.26a) of the specimens are the stress-
strain relationship during monotonic shear loading under constant stress (σv = 50 kPa)
condition.
Figures 5.25b and 5.26b show the shear displacement-volume change relationship of the
specimens under dry and saturated conditions. Both specimens showed contractive behaviour
during initial loading while these specimens exhibited dilative behavior during monotonic shear
loading.
It can be seen from the Fig. 5.27 and 5.28 that the difference in stress-strain-volume
change behaviours between dry and saturated conditions is insignificant for both Toyoura sand
and Chiba gravel. Clear peak and residual stress states or strain softening under dry as well as
saturated conditions are observed. It is certainly due to the durable (non-slakable) nature of
grains which are unaffected by the presence of water. The smaller value of degradation index
(ID) and Fines content (FC) (Table 5.12 and 5.13) after the experiments also indicates the
durability of grain. However, some differences in stress-strain-volume change behaviours could
be noticed between dry and saturated test even though the difference is very small. The
presence of pore water in saturated conditions can cause small variations in the soil skeleton
and might be possible reason of negligible difference in stress-strain-volume change behaviours
between dry and saturated tests. In general, it can be said that the slaking effect on stress-strain
change behaviour of conventional granular materials is almost negligible. The specimens were
prepared at high initial density therefore, volumetric expansion (positive dilatancy) during
monotonic loading (shearing) for the Toyoura sand and Chiba gravel specimens was observed.
The shear and vertical displacements during wetting were also found to be insignificant.
5.40

2.0
Chiba gravel (PreC011)
3
in = 1.529 gm/cm
1.5
R = 0.5
1.0
0.5
Saturation started
0.0
-10 0 10 20 30 40 50
2.0
Chiba gravel (PreC011)
3
in = 1.529 gm/cm
1.5
R = 0.5
1.0
0.5
Saturation started
0.0
-10 0 10 20 30 40 50
b)
Fig 5.23 Increment a) shear b) vertical displacement during wetting

under R = 0.5 (Chiba gravel)
5.41

Toyoura sand (PreT052)
0.6 ρ = 1.459 gm/cm3
R = 0.5
0.4
0.2
Saturation started
0.0
-0.2
0 10 20 30 40 50
Increment vertical displacement, s (mm)
Toyoura sand (PreT052)

0.6 ρ = 1.459 gm/cm3
R = 0.5
0.4
0.2
Saturation started
0.0
-0.2
0 10 20 30 40 50
b)
Fig 5.24 Time histories of shear and vertical displacement (a) shear (b) vertical
displacement under R= 0.5 (Toyoura sand)
5.42
50
40 D
Shear stress,t (kPa)
30
Chiba gravel
3
A in = 1.529 gm/cm
20
R = 0.5
10
Dry specimen (PreC010)
0 O Saturated specimen (PreC011)
-2 0 2 4 6 8 10 12 14 16
4
Chiba gravel 1- Initial loading
3 3 2- Saturation
in = 1.529 gm/cm
2 R = 0.5
3 Contraction
1
0
2
1
-1
-2 Expansion
-3 Dry specimen (PreC010)
Saturated specimen (PreC011)
-4
-2 0 2 4 6 8 10 12 14 16
R= 0.5 during experiment (Chiba gravel)
5.43
50 Toyoura sand
R = 0.5
40
Shear stress, t(kPa)
30 D
A
20
10
3
0 Dry (PreT050)(in = 1.528 gm/cm )
O 3
Saturated (PreT052)(in= 1.549 gm/cm )
-10
-2 0 2 4 6 8 10 12 14 16
4
Toyoura sand 1- Initial loading
3 2-Wetting
R = 0.5
2
1 2 3
Contraction
0 1
-1
Expansion
-2
3
-3 Dry (PreT050) (in=1.528 gm/cm )
3
Saturated (PreT052) (in=1.549 gm/cm )
-4
-2 0 2 4 6 8 10 12 14 16

R= 0.5 during experiment (Toyoura sand)
5.44
1.0 0 0
p = 42.89 p = 43.13
0.9
0.8
Stress ration, R
0.7
Chiba gravel
0.6
0.5 3
Dry (PreC010)( = 1.532 gm/cm )
3
Saturated (PreC011)( = 1.532 gm/cm )
0.4
-2 0 2 4 6 8 10 12 14 16
4
3
Dry (PreC010)( = 1.532 gm/cm )
3 3
Saturated (PreC011) ( = 1.529 gm/cm )
2
Contraction
1
Chiba gravel
0 R= 0.5
-1
-2
-3 Expansion
-4
-2 0 2 4 6 8 10 12 14 16

R= 0.3 during experiment (Glass beads)
5.45
1.0
Toyoura sand
Vertical displacement, v (mm) 0.9 R = 0.5
0.8
0.7
0.6
0.5 3
Dry (PreT050)( = 1.536 gm/cm )
3
Saturated (PreT052)( = 1.555 gm/cm )
0.4-2 0 2 4 6 8 10 12 14 16
a)
4
Toyoura sand
3 R = 0.5
2
Contraction
1
0
-1
Expansion
-2
3
-3 Dry (PreT050)( = 1.536 gm/cm )
3
Saturated (PreT052)( = 1.555 gm/cm )
-4-2 0 2 4 6 8 10 12 14 16
b)

R = 0.3 during experiment (Toyoura sand)
5.46
From these discussion and the test results presented in this section, it is concluded that
the stress-strain behaviour and volume change characteristics of Toyoura sand and Chiba gravel
during monotonic shear loading are unaffected by presence of water i.e. no slaking effect on
Chiba gravel and Toyoura sand. Similarly, Fioraqvante and Capoferri (1997), Wichtmann et al.
(2005) and Youn et al. (2008) have reported negligible influence of saturated and dry condition
on strength and stiffness as well as volume change characteristics of conventional granular
materials consisting of durable grains.
5.4 Particle crushing and Degradation index

The mechanical properties of crushed mudstones are strongly affected by the slaking
induced deterioration because the conventional soil mechanics parameter like angularity, mean
grain size and grain size distribution etc., of crushed mudstones are no more constants.
Therefore, the particles size distribution under working conditions should be known to assess
the effects of deterioration on mechanical behaviour of the material. The index used in this
study for quantification of slaking induced disintegration of crushed mudstones which is
somehow related to slaking behaviour of crushed mudstones, was explained in Chapter 4. The
slaking susceptibility of each mudstone upon saturation or cyclic wetting and drying can be
quickly assessed by comparing the degradation index (ID) and fines contents (FC) at the end of
dry, saturated and the one with cyclic wetting and drying tests.
Particle size distribution (PSD) curves are essential for qualitative and quantitative
measurement of slaking susceptibility. Fineness (FC) and degradation index (ID) also useful for
quantitative comparison of stress-strain response and volume change characteristics between
dry, saturated and the one with cyclic wetting and drying conditions of various crushed
mudstones (Aziz, 2010). Considering importance of degradation index (ID), the particles size
distributions of all the specimens before and after experiments were determined.
The particle size distribution (PSD) curves for the dry, saturated and the one with cyclic
wetting and drying specimens tested under same stress ratio are presented in the same graphs.
The Fig. 5.16 shows the particles size distribution (PSD) curves for the dry, saturated and the
one with three cycles of wetting and drying Hattian Bala mudstones specimens tested under
stress ratio 0.7.
5.47
The particle size distribution (PSD) after dry test for Hattian Bala mudstone shows
some disintegration with average fines (finer than 2 mm) content and degradation index of 1.65
% and 0.06 respectively(Fig. 5.16). This is essentially due to crushing of relatively weak
particles under vertical stress and monotonic shear loading.
Similarly, particles are crushed due to immersion and cyclic wetting and drying. It can
also be defined as slaking induced disintegration. After experiment, it is seen that about 7.7 %
particles by mass become finer than 2.0 mm after the saturated test, where as in the case of
cyclic wetting and drying test 10.55 % particles by mass become finer than 2.0 mm
respectively. Similarly, the degradation indexes were 0.12 and 0.16 for saturated and cyclic
wetting and drying test (Fig 5.16).
Figure 5.17 shows the particles distribution curves (PSD) after the dry, saturated and the
one with cyclic wetting and drying test for Hattian Bala mudstone specimen under stress ratio
0.5. About 1.5 %, 5.24 % and 8.24 % particles by mass become finer than 2.0 mm after dry,
saturated and cyclic wetting and drying test. The degradation indexes after dry, saturated and
cyclic wetting and drying tests were about 0.07, 0.11 and 0.15 respectively.
The particles distribution curves (PSD) for the tests under stress ratio 0.3 are presented
in the Fig. 5.18. The average fineness (finer than 2 mm) content after each experiment is also
mentioned in the same figure. Figure 5.19 shows the particles distribution curves (PSD) after
the dry, saturated and the one with two cycles of wetting and drying test for Ishikawa
mudstone. The particle size distribution (PSD) after dry test for Ishikawa mudstone shows
some disintegration with average fines (finer than 2 mm) and degradation index of 3.92 % and
0.07 respectively. This is basically due to crushing of relatively weak particles under vertical
stress and monotonic shear loading as in the Hattian Bala mudstone. After saturated and two
cycles of wetting and drying tests, about 25.68 % and 27.28 % particles by mass were found
finer than 2 mm which were probably due to slaking of crushed mudstone. Similarly, the
degradation indexes after saturated and the one with cyclic wetting and drying tests were about
0.40 and 0.42 respectively.
Finally, the Fig. 5.20 show the particles distribution curves (PSD) after one complete
cycle of wetting and drying for the Hattian Bala mudstones specimen (PreM028). Average
5.48
fines (finer than 2 mm) content and degradation index of the spacemen after one complete
cycles of wetting and drying are 7.13 % and 0.13 respectively (Fig. 5.16).
The degradation index (ID) and fines contents (FC) of Hattian Bala, Ishikawa and Chiba
gravel specimens after experiments are summarized in Table 5.8-5.13. Ishikawa mudstone with
slaking index as level 3 showed (Table 5.10 and 5.11) the highest slaking induced
disintegration with an average ID and FC of 0.4 and 25.68 % respectively after single
immersion. Hattian Bala mudstone with slaking index as level 1, consisting of relatively
durable grains, showed the least effects of slaking on particles crushing with an average value
of ID and FC of 0.15 and 9 % respectively (Table 5.8 and 5.9). Similarly, Chiba gravel with
slaking index as level 0, consisting of durable grains, showed almost no effect of slaking on
particles crushing (Table 5.8 and 5.9). The degradation index (ID) and fines contents (FC) can
be compared with slaking index (SI) and slaking ratio. So, from the above results, it is cleared
that the material having relatively lower slaking index (SI) or higher slaking ratio shows higher
degradation index (ID) and fines contents (FC).
In addition, Particles size distribution (PSD) after cyclic wetting and drying test for the
both Ishikawa and Hattian Bala mudstone specimens also showed some more disintegration of
particles as compared to saturation test, which is basically due to particles crushing (slaking)
during cyclic wetting and drying. However, the differences of degradation index (ID) and fines
contents (FC) between saturation test and cyclic wetting and drying test is relatively small.
Panabokke and Quirk (1956) reported that the disintegration of clay aggregates became higher
as the water content of the specimen before wetting became lower. Therefore, it can be
understood that the maximum particles crushing occurred during the first wetting because the
specimen in this study was prepared by oven-dried crushed mudstones.
Moreover, small deterioration of grains under dry conditions is also observed in each
specimen, basically due to the crushing of relatively low durable grains under normal stress as
well as shearing action. Most of deterioration of grains in the tests took place during saturation
and partially during drying stages before shearing. To verify the above conclusion, the
particles size distribution analyses of crushed mudstones (Hattian Bala) freely submerged in
water (under atmospheric pressure only) were also carried out. It can be observed that the
5.49
After creep with cylic wetting

Percentage weight finer than (%) 100 wetting 3 times and ML (PreM003)
(FC= 10.55 & ID= 0.16)
80 Before test
Creep and ML under saturated
60 condtion (PreM013)
(FC= 7.7 & ID=0.12) Creep and ML under dry
(FC= 1.65 & ID= 0.06)
20
Hattian Bala
SI= 1
0 R = 0.7
2 3 4 5
Grain size (mm)
Fig 5.29 Particles size distribution curves of PreM008, Prem013 and PreM003 after
dry, saturated and cyclic wetting and drying test under R= 0.7 (Hattian Bala)

Percentage weight finer than (%)
100 wetting 3 times and ML (PreM002)

(FC= 8.24 % & ID= 0.15)
80 Before test
Creep and ML under saturated
(FC= 5.4 % & ID= 0.11)
40 Creep and ML under dry
condtion (PreM007)
(FC= 1.5 % & ID= 0.07)
20
Hattian Bala
SI = 1
0 R = 0.5
2 3 4 5
Grain size (mm)
5.50
After creep with cylic wetting and

100 drying 3 times and ML (PreM001)
(FC= 7.09 %)
80
Before test
Creep and ML under
60 saturated condtion (PreM010)
(FC= 6.65 %) Creep and ML under
40 dry condtion (Prem009)
(FC= 1.51%)
20
Hattian Bala
SI= 1
0 R = 0.3
2 3 4 5
Grain size (mm)

100 wetting 3 times and ML (PreM027)

(FC= 27.28 % & ID= 0.42)
80
Creep and ML under
saturated condtion (PreM025) Before test
60 (FC= 25.68 % & ID=0.40)
40 Creep and ML under dry

condtion (PreM026)
(FC= 3.92 % & ID= .07)
20
Ishikawa
0 SI= 3
R = 0.3
2 3 4 5
Grain size (mm)
dry, saturated and cyclic wetting and drying test under R= 0.3 (Ishikawa)
5.51

100 After creep with cylic wetting
and drying 1 times and ML (PreM028)
80 (FC= 7.13 % & ID= 0.13)
Before test
60
40
20
Hattian Bala
SI= 1
0 R = 0.5
2 3 4 5
Grain size (mm)
Fig. 5.33 Particles size distribution curves of PreM028 after one complete
cycle and dry monotonic loading (R= 0.5)
increment of fines at no stress conditions are comparable to values observed for saturated
monotonic shear tests (Fig. 5.34).
Similarly, the increment of fines during drying also observed during this experiment.
The test PreM028 was conducted by applying monotonic loading applied under dry condition
after one complete cycle of wetting and drying. The degradation index (ID) and the Fines
content is slightly higher than saturation tests (Table 1 and 2). The increment of degradation
index and Fines contents in PreM0028 test indicates the drying induced slaking. Thus, it is
inferred that the slaking phenomena causes surface deterioration of soil particles during drying
and splitting along the weak planes which primarily completes during saturation stage
(Czerewko and Cripps, 2001).
From the test results of Hattian Bala and Ishikawa mudstones, it can be observed that
degradation index (ID) of a given crushed mudstones has a clearly dependency on slaking
characteristics of geomaterials. It can be concluded that higher the slaking index or lower
slaking ratio reveal the drastic loss of strength during saturation and increase in disintegration
index (ID).
5.52
Table 5.8 Degradation index (ID) of Hattian Bala mudstone specimens
Test condition Stress ratio, R
0.7 0.5 0.3
Dry 0.06 0.07 0.06
Saturated 0.12 0.11 0.12
1 Cycle - 0.13 -
3 Cycles 0.16 0.15 -
Table 5.9 Fineness contents (FC) of Hattian Bala mudstone specimens
Stress ratio, R
Test condition
0.7 0.5 0.3
Dry 1.65 1.5 1.51
Saturated 7.7 5.4 6.65
1 Cycle 7.13
3 Cycles 10.55 8.24 7.09
Table 5.10 Degradation index (ID) of Ishikawa mudstone specimens
Stress ratio, R
Test condition
0.7 0.5 0.3
Dry - 0.08 0.07

Saturated - - 0.4
Cyclic - - 0.42
5.53
Table 5.11 Fineness contents (FC) of Ishikawa mudstone specimens
Stress ratio, R
Test condition
0.7 0.5 0.3
Dry - 3.8 3.52
Saturated - 0 25.68
Cyclic - 27.28
Table 3.12 Degradation index (ID) of Chiba gravel specimens
Stress ratio, R
Test condition
0.7 0.5 0.3
Dry - - .01
Saturated - - .01
Table 4.13 Fineness contents (FC) of Chiba gravel specimens
Stress ratio, R
Test condition
0.7 0.5 0.3
Dry - - 0.2
Saturated - - 0.32
5.54
Percentage weight of particle finer (%)

100
80
60
Before experiment
40 1 day drying after 1st wetting
2 days drying after 1st wetting
3 days drying after 1st wetting
20
1 day drying after 2nd wetting
2 days drying after 2nd wetting
0
3 days drying after 2nd wetting
1 days drying after 3rd wetting
-20
2 3 4 5 6
Grain size (mm)
Fig. 5.34 Particle size distribution distributions curves of Hattian Bala mudstones
after submerge in water
Finally, the phenomena of slaking induced particle crushing discussed in this research is
contrary to the conventional studies of stress-induced particles crushing performed by Hardin
(1985), Feda (2002), Einav (2007), Kikumoto et al. (2010) and many others.
5.5 X-ray Diffraction (XRD) power test

X-ray diffraction power test was used in order to determine the mineralogy of the rocks.
Based on the results of the XRD analysis of Hattian Bala mudstones specimen, the Hattian Bala
mudstone specimen demonstrated that the specimen generally comprised clay minerals,
calcium and silicate. It is believed that the Calcium hiders the slaking of mudstones. It behaves
as cementing materials. Similarly, the Ishikawa mudstones specimen showed Lithium Gallium
silicate.
5.55
Hattian Bala
Ishikawa
2θ
Fig. 5.35 X-ray diffraction result of Hattian Bala and Ishikawa mudstones
5.56
CHAPTER 6
RESULT ANALYSIS AND DISCUSSION
6.1 General
To describe the slaking effects on the strength and deformation characteristic as well
as particles crushing of mudstones, a series of direct shear tests with other preliminary
experiments were conducted. Similar tests were also conducted on conventional granular
material such as Chiba gravel and Toyoura sand. Non-slakeable Glass beads specimens
were used for few tests as well.
Along with the properties of various test materials used in this study, the effect of
the slaking index, density of the specimen, water content before wetting and stress ratio will
be discussed in this chapters. Similarly, the influences of the cyclic wetting and drying on
the physical and mechanical properties of various geomaterials used in this study are
discussed in this chapter.
6.2 Effect of stress ratio (R)

In this study, three cycles of wetting and drying were carried out on Hattian Bala
mudstone specimens under different stress ratio to analysis the deformation and strength
characteristics due to slaking by using a modified direct shear apparatus. Similarly, two
cycles of wetting and drying were carried out on Ishikawa mudstones and Glass beads
specimens under stress ratio 0.3. The test results obtained from the cyclic wetting and
drying under constant creep loading will be interpreted and discussed to examine their
behaviour on natural slope or when used as construction materials for embankments etc.
The results obtained from the cyclic wetting and drying tests are summarized in Fig.
6.1 and 6.2, showing increment values of creep shear and vertical displacement at each
stress ratio, R. The wetting-induced maximum creep shear deformation of 3.4 mm was
observed at R= 0.7 during the first wetting (see Fig. 4a) from the Hattian Bala mudstone
specimen. Wetting-induced creep failure was observed at R= 0.8 on the same material.
6.1
Chapter 6 Results analysis and discussion
Similarly, Kiyota et al. (2011) also reported that wetting-induced creep failure was observed
at R= 0.8 under σv= 100 kPa on the same material as the one in this study. Therefore, it
seems that the creep shear displacement during wetting increases with increase in the value
of R, which would indicate high risk of slaking-induced instability at steep slopes.
The vertical displacement, however, in case of R= 0.3, is almost negligible except in
the first wetting because of having higher water content (more than 3 %) of specimen before
the second and third wettings. The vertical displacement in the wetting is affected by stress
ratio, R values during creep shear loading and the progress of wetting and drying cycles.
As shown in Fig. 6.1, the creep shear displacements during drying increase with the
increase in the value of stress ratio, R. However, in the case of R= 0.3, the lowest water
content during the drying step is quite larger (about 3 %). This may be reason for relatively
small creep displacement during the drying process at R= 0.3.
Figure 6.2 shows the relationship between stress ratio, R and reduction in peak angle
of internal friction due to wetting. From the figure it is clearly seen that the reduction in
peak angle of internal friction due to wetting is independent of stress ratio, R.
As already mentioned in previous chapter, the degradation index was evaluated after
each experiment under different stress condition. The amount of particles crushing due to
slaking is independent of stress ratio, R (Fig. 6.3).
Similarly, the results obtained from the cyclic wetting and drying tests on Ishikawa
mudstone specimens are summarized in Fig. 6.2, showing increment values of creep shear
and vertical displacement at stress ratio 0.3. The wetting-induced maximum creep shear
deformation of 3.9 mm was observed at R= 0.3 during the first wetting (see Fig. 6.2a).
Wetting-induced creep failure was observed at R= 0.5 on the same material. Therefore, it
seems that the creep shear displacement during wetting increases with increase in the value
of R, which would indicate high risk of slaking-induced instability at steep slopes.
6.2

Wetting Hattian Bala
st
1 wetting
Drying
st
3 1 SI = 1
nd
2
rd
3 st
1 drying
2
nd
2 drying
1
nd
2 wetting
0 rd
3 wetting
0.3 0.4 0.5 0.6 0.7
a) Stress ratio, R
2.5
st
Wetting 1 drying
2.0 Drying
1
st Contraction
1.5 2
nd nd
2 drying
rd
1.0 3
Hattian Bala
0.5
SI = 1
0.0 st
1 wetting
-0.5 nd
2 wetting
-1.0 rd
Expansion 3 wetting
-1.5
0.3 0.4 0.5 0.6 0.7
Stress ratio, R
b)
Fig. 6.1 Increment value of creep a) Shear b) Vertical displacement at each stress
ratio, R
6.3
Reduction in peak angle of friction

SI= 1
40
due to wetting (%)
20
0
0.2 0.3 0.4 0.5 0.6 0.7 0.8
Stress ratio, R
Fig. 6.2 Relationship between R and change in peak angle of internal friction
1.0
after three cycle of wetting and drying

0.8 SI= 1
Degradation index (ID)
0.6
0.4
0.2
0.0
-0.2
0.2 0.3 0.4 0.5 0.6 0.7 0.8
Stress ratio, R
Fig. 6.3 Relationship between R and change in peak angle of internal friction
6.4
4
Increment shear displacement, s (mm) st
1 wetting
st
3 1 drying
Lead to creep failure
during first wetting
2
Ishikawa mudstone
SI= 3
nd
1 2 wetting
0.3 0.4 0.5 0.6 0.7

a)
Stress ratio, R
4 st
1 drying
3 st
1 wetting Contraction
2
Ishikawa mudstone
SI= 3
1 nd
2 wetting
0
Expansion
-1
0.3 0.4 0.5 0.6 0.7
b) Stress ratio, R
Fig. 6.4 Increment value of creep a) shear b) vertical displacement at each
stress ratio, R
6.5
6.3 Effect of water content before wetting

The creep shear displacement during first wetting was significant under each stress
ratio. However, the creep shear displacement caused by wetting seems to be decreased with
progress of wetting and drying cycle (Fig. 6.1a). The water content before each wetting was
not same. The Figures 6.5 and 6.6 show the relationship between water content before
wetting and wetting induced shear and vertical displacement respectively. The shear
displacement during wetting became higher as the water content before wetting of specimen
became lower (Fig. 6.5). Nakano et al. (1998) and Panabokke and Quirk (1956) also
reported that the slaking level of clay aggregates became higher as the initial water content
of the specimen became lower. Therefore, it can be understood that the maximum
displacement was observed during the first wetting because the specimen in this study was
prepared by oven-dried crushed mudstone. The lowest water content during drying step is
quite large (about 3 %) in case of R= 0.3. This may be reason for almost negligible creep
shear displacement during the second wetting process at R = 0.3 (Fig. 6.1a).
Similarly, in case of Ishikawa mudstone specimens, the creep shear displacement
during second wetting is relatively very small because of higher water content (about 12%)
before wetting.
The vertical displacement is almost negligible when water content before wetting is
more than 3 % (Fig. 6.6). The vertical displacement during wetting also became higher
when water content before wetting became lower. However, the vertical displacement of
oven dried specimen (water content about 0 %) during wetting is seemed to be smaller as
compared to the vertical displacement occurred during wetting when water content before
wetting was about 1 % (Fig. 6.7). This may be due to dilatancy effect due to significant
shear deformation occurred when oven dried specimens are used. Collapse of weak grains
during first wetting also causes reduction in negative vertical displacement during first
wetting.
6.6
4.0
Increment shear displacment, v (mm)

3.5
 = 1.568 attian Bala mudstone
SI= 1
3.0
3
2.5 Density ( in gm/cm
2.0
1.5
1.0  = 1.538
0.5  = 1.585
 = 1.591
0.0 R = 0.5
 = 1.556  = 1.563
R = 0.7
-0.5
0.0 0.2 0.4 0.6 0.8 1.0
Water content before wetting, w (%)
Fig. 6.6 Relationship between water content before wetting

and increment creep shear displacement
Increment vertical displacment, v (mm)
1.0
 = 1.591
 = 1.556  = 1.563 Hattian Bala mudstone
0.8 SI= 1
 = 1.585
Expansion
0.6
0.4
 = 1.568
 = 1.538
0.2
R = 0.5  = 1.50
0.0
R = 0.7 Density ( in gm/cm
3
 = 1.50
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Water content before wetting, w (%)
Fig. 6.7 Relationship between water content before wetting

and increment creep vertical displacement
6.7
6.4 Effect of slaking index (SI)

Figure 6.8 shows the increment shear displacement of various tested materials due to
saturation under same anisotropic stress condition i.e. under stress ratio 0.3 and 0.5. The
shear displacement of Hattian Bala mudstone specimens having slaking index as level 1 was
about 4 mm under R = 0.5 while, the Ishikawa mudstone having slaking index as level 3
specimens led to creep failure during wetting (Fig. 6.8). From the Fig. 6.8, it is seen that the
Ishikawa mudstone (SI = 3) specimens showed highest shear displacement during wetting.
Similarly, the Hattian Bala mudstone (SI = 1) specimen also showed considerable shear
displacement under same condition while other materials such as Chiba gravel and Glass
beads (SI = 0) did not show significant shear displacement during wetting.
Large amount of particles crushing as compared to other specimens (Fig. 6.9) during
wetting is also responsible for excessive shear displacement for the Ishikawa mudstone
specimens. The shear displacement of Hattian Bala mudstone specimen is about one tenth
of Ishikawa mudstone specimen because of consisting relatively strong grains. This
argument can be justified by comparing Degradation index (ID) and Slaking ratio (SI) of the
tested specimens. The Chiba gravel, Toyoura sand and Glass beads consist of strong grain
which can not deteriorate by slaking. That’s why; the shear displacement during wetting is
negligible as compared to mudstones specimens. Finally, it is inferred that slaking induced
deformation and deterioration of soil grains which ultimately leads to enormous shear
displacement upon sustained loading can be critical progressive slope failures.
The drying induced shear displacement of Ishikawa mudstone is also higher than
that of Hattian Bala mudstones specimen during first drying process (Fig. 6.1a and 6.4a),
even the water content of Ishikawa mudstone specimen is higher than Hattian Bala
mudstone specimen at the end of drying process and stress ratio lower than Hattian Bala
specimen. As already described in previous sections, the deterioration of the particle begin
from the surface of the particle during drying process (Moropoulou et al., 2004; Karoglou et
al., 2005; Soe et al., 2010). Higher void ratios increase the exposed area of individual
particles which lead to easily deterioration of particles. The tensile failure of the weakly
crystalline bonded granular materials due to drying induced pore water suctions is also
another cause of drying induced slaking (Czerewko and Cripps, 2001). So, higher void
ratios of Ishikawa mudstone specimen also lead to higher shear displacement during drying
phase as in wetting phase.
6.8
Finally, it is quite clear that with the increase in Slaking index (SI) there is drastic
increase in shear displacement upon wetting and drying as well. The relationship between
Slaking index (SI) and increment shear displacement during wetting could become very
important for the practical application.
4.5
R = 0.3 s > 15 mm
Increment shear displacment, s (mm)
4.0 under R = 0.5

R = 0.5
3.5
Ishikawa ( = 0.694)
3.0
2.5
Chiba (   1.536 )
2.0
1.5
1.0
0.5
Hattian Bala ( = 1.50)
0.0 3
Glass beads ( = 1.601) Density ( in gm/cm
0 1 2 3 4
Slaking index (SI)
Fig. 6.8 Relationship between Slaking index (SI) and
increment shear displacement during wetting
0.5
Ishikawa (SI = 3)
0.4
Degradation index (I D)
0.3
0.2
0.1 Hattian Bala (SI = 1)
0.0
0 1 2 3 4
Slaking index (SI)
Fig. 6.9 Relationship between Slaking index (SI) and degradation index (ID)
6.9
The specimens of Ishikawa mudstone (SI=3) showed vertical compression during

first wetting, which is attributed to the sensitivity of material to slaking induced
deterioration (Fig. 6.10). In contrast the crushed mudstones having slaking index as level 1
from Hattian Bala, Pakistan showed vertical expansion during saturation probably to
presence of expansive clay minerals. Expansion of clay minerals also cause cracking in the
grains which ultimately lead to disintegration. Cardoso and Alonso, 2009 also explained
swelling induced particles disintegration in mudstones. However, expansive clay mineral
montmorillonite of smectite group was not found clearly from X-ray diffraction (XRD)
analysis of two geomaterials such as Ishikawa and Hattian Bala mudstones. In addition,
Chiba gravel, Toyoura sand and Glass beads specimens having slaking index as level 0 have
shown relatively very small vertical settlement during wetting being an indicative of durable
soil grains against the slaking.
1.0
R= 0.3
0.5 Hattian Bala (  1.5)
Contraction
0.0
Chiba (  1.54)and
-0.5 Glass beads (  1.6)

Expansion
-1.0
-1.5
-2.0
-2.5
3
Density ( in gm/cm Ishikawa (  
-3.0
0 1 2 3 4
Slaking index (SI)
Fig. 6.10 Relationship between Slaking index (SI) and vertical

displacement during wetting
As already discussed, the vertical displacement of Ishikawa mudstone having

slaking index level 3 during saturation being positive (contraction) is indicative of slaking
induced disintegration of soil particles as well as collapse of relatively open structure of
these materials having high initial void ratios. So, relatively denser packing was achieved in
the specimen after first wetting. These observations are in strong agreement with a study by
Neves and Pinto (1988) on prediction of collapse settlement due to saturation rock-fill,
6.10
mainly in embankment dams. Therefore, it is inferred that slaking induced deterioration of

mudstone stones having higher degree of slaking index (SI) which ultimately leads to
enormous compression upon sustained loading can be critical for the embankments
constructed with crushed mudstones, for soil structure interaction of foundation placed on
such soils as well as progressive slope failures. It is also concluded that the mudstone
specimen with smaller slaking index (SI) exhibits expansive behaviour during first wetting
while the mudstone specimens with higher slaking index (SI) shows the contractive
behaviour.
The reduction in peak angle of internal friction (φp) of Ishikawa mudstone (SI = 3)
specimens during wetting is higher than the reduction in peak shear strength of Hattian Bala
(SI= 1) mudstone specimen. The reduction in peak angle of internal friction (φp) is
insignificant in the case of that specimen having slaking index level 0 (Chiba gravel and
Glass beads). Relatively large inherent void ratios, e (2.3) and slaking induced
disintegration during wetting cause the large reduction in peak strength of Ishikawa as
compare to Hattian Bala mudstone specimens. Similarly, it is seemed that the potential to
disintegrate due to slaking (reduction in strength parameter) is dependent of slaking index of
specimen. It is concluded that with the increase in Slaking index (SI), there is tremendous
reduction in the mechanical properties of crushed mudstones.
50
Reduction in peak angle of friction ( 
R=0.3
40 Ishikawa ( = 0.694)
due to wetting
30
20
10 Hattian Bala ( = 1.50)
0 Glass beads ( = 1.601) 3

Chiba (   1.536 ) Density ( in gm/cm
0 1 2 3 4
Slaking index (SI)
Fig. 6.9 Relationship between Slaking index (SI) and reduction in peak angle
of friction (φp) due to wetting
6.11
6.5 Effect of density (ρ)

The specimens prepared at high initial density showed smaller shear displacement
during wetting. The shear displacement of the specimen with initial density of about 1.50
gm/cm3 was about 0.9 mm while 0.3 mm shear displacement occurred in case of the
specimen with initial density of about 1.60 gm/cm3 under R = 0.5 (Fig. 6.10).
1.0
SI= 1
0.8
R = 0.5
0.6
0.4
0.2
0.0
1.50 1.52 1.54 1.56 1.58 1.60
3
Density,  gm/cm )
Fig. 6.10 Relationship between density of the specimen and increment

shear displacement during first wetting.
From about discussion, it is concluded that the degree of slaking decreases with
increase in density of specimen.
6.6 Number of cyclic wetting and drying (N)

The creep shear displacement caused by wetting seems to be decreased with increase
in number of wetting and drying cycles of cycle, almost zero during the third wetting (Fig.
6.12). This may be attributed to the specimen densification due to previous wetting and
drying processes (Nakano et al., 1998) and higher water content before wetting. Similarly,
as already discussed, the higher shear displacement during first wetting also may be due to
specimens prepared by oven dried crushed mudstone. Finally, it is concluded that the shear
displacement during wetting decreases with increase in number of
6.12
Contrary to shear displacement, the vertical displacement was found increasing with
increase in number of cyclic wetting and drying.
3.5
w= 0 %,  = 1.568
R = 0.5
3.0 R = 0.7
2.5 Hattian Bala mudstone

SI= 1
during wetting
2.0
1.5
w= 0 %,  = 1.538
1.0
0.5 w= 0.71 %,  = 1.585

w= 0.85 %,  = 1.591
0.0
w= 0.70 %,  = 1.556
w= 0.82 %,  = 1.563
-0.5
1 2 3
Number of cycles (N)
Fig. 6.11 Relationship between number of wetting and drying cycles (N)
and increment shear displacement during wetting.
0.4
attian Bala mudstone

0.2 SI= 1
Contraction
0.0
 = 1.568
during wetting
-0.2 Expansion
-0.4
 = 1.538
-0.6
 = 1.585
-0.8  = 1.591
 = 1.556
 = 1.563
-1.0 R = 0.5
R = 0.7 3
Density ( in gm/cm
-1.2
1 2 3
Fig. 6.12 Relationship between number of wetting and drying cycles (N)
and increment vertical displacement during wetting.
Figure 6.13 shows the relationship between the number of wetting and drying cycles
and peak angle of internal friction (φ). The influence of cyclic wetting and drying on peak
6.13
angle of friction (φ) is seemed insignificant. However, the gradual decrease in peak angle
of friction (φ) is found on enlarge view of Fig. 6.12 (Fig. 6.13).
dry Hattian Bala mudstone

50 dry
SI = 1
Peak angle of friction( p)
40
sat sat sat
30
20
10
0
0 1 2 3 4 5 6
Fig. 6.13 Relationship between number of wetting and drying

cycles (N) and increment vertical displacement during wetting.
50
48 SI = 1
dry
Peak angle of friction( p)
dry
46
D e c re a
se s g r a d
44 u a lly
42
40 D e c re a
se s g r a d
u a lly
38
sat sat
sat
36
0 1 2 3 4 5 6
Fig. 6.14 Enlarge view of Figure 6.13
6.14
6.7 Summary
The test results obtained from general slaking tests and direct shear tests on various
materials were interpreted and anlysed with the with the objectives of examining the factor
that influences the slaking behaviour of geomaterials. Along with different types of granular
soils, the effects of stress ratio, R, initial water content before wetting, slaking index (SI),
density of the specimen (ρ) and number of cyclic wetting and drying (N) on slaking of
crushed mudstones were investigated.
From the results, it is seen that the slaking phenomenon is a very complex behaviour
which is dependent of various factors such density of specimen, water content before
wetting, slaking index (SI) of the specimen, number of cyclic wetting and drying,
mineralogical composition etc. The conclusion of this research can be made as following.
dS  f (SI , w,  , R, N , I D )
Where, dS: Increment shear displacement, SI: Slaking index, w= Water content
before wetting, R = Stress ratio, N = Number of cyclic wetting and drying and ID=
Disintegration index.
However, the factors inside the functions are also interdependent to each other.
Similarly, there are other various factors such as mineral content, stress and deformation
histories, drying history etc which influence the increment shear displacement due to
slaking. So, the effects of slaking, quantification of deformation and disintegration and its
relationships with reduction of strength and deformation characteristics of non-conventional
materials from crushed mudstones is still a key concern for researchers.
6.15
CHAPTER 7
CONCLUSIONS AND RECOMMENDATIONS
7.1 General
The engineering properties of granular soils are conventionally believed to be
unaffected by the cyclic wetting and drying or the presence of water. This is basically due to
durable nature of grains of the conventional granular materials such silica sand (Toyoura sand)
and sand stone (Chiba gravel) investigated in various geotechnical laboratories. Nonetheless,
large scale availability of soft rock deposits in many parts of the world, massive infrastructure
developments have been extended to hillsides and mountainous areas because of population
growth, economics needs and other constraints. Soft sedimentary rock names mudstone is the
most often encountered geomaterials during the major construction works that are undertaken
in those areas. Slaking effects on these geomaterials become vital due to their sensitivity to
slaking upon distinct seasonal changes. Detailed knowledge of slaking effects on such granular
soils is very essential in geotechnical analysis and design for safety over the lifetime of
geotechnical structures.
To achieve the proposed objectives of this research, intensive experiments on both
conventional (Chiba gravel and Toyoura sand) and non-conventional (Hattian Bala and
Ishikawa mudstone) geomaterials by using direct shear apparatus were performed. A series of
direct shear tests simulating cyclic wetting and drying under different stress condition were
performed. Similarly, about 300 landslide events were studied on basis of precipitation and
geology.
7.2 Conclusions
In the beginning of this research, about 300 rainfall induced landslide events were
collected and analysed on the basis of precipitation and geology of landslides area. About 50 %
A.1
Chapter 7 Conclusions and recommendations
of total rainfall induced landslides in soft sedimentary rock were found to be occurred after
moderate rainfall. Similarly, number of rainfall induced landslide followed by drought was
found significant. So, the conventional explanation for rainfall induced landslides is insufficient
regarding the mechanism of all so called rainfall landslide landslides. Some landslides most
probably occur under an unsaturated regime even after light rainfall, resulting only from a
decrease in shear strength of the soil especially in soft sedimentary rock formation due to
slaking of mudstones. However, detailed field and laboratory investigations of landslide areas
are necessary to inspect the subsurface structures and to obtain geotechnical properties of slope
materials. Then, real mechanism of land even could be explained precisely
As hypothesized, the strength and deformation characteristics of crushed mudstones
under dry, saturated and cyclic wetting and drying condition are not in agreement with usual
geotechnical approach towards various standard granular soils such as Silica sand (Toyoura)
and sand stone (Chiba gravel). The stress strain behaviour and volume change characteristics of
conventional material (Toyoura sand, Chiba gavel and Glass beads) are unaffected by both
presence of water and cyclic wetting and drying, where as the mechanical behaviour of non-
conventional materials (Hattian Bala and Ishikawa crushed mudstone) is conditioned by the
number of cyclic wetting and drying or the presence of water. It is mainly due to the slaking
behaviour of mudstones under cyclic wetting and drying or even under single immersion.
The effects of stress ratio (R), initial water content (w) before wetting, dry density of the
specimens (ρ), slaking index (SI) and number of cyclic wetting and drying (N) on strength,
deformation and particles crushing of specimens were investigated.
It is seen that the creep shear displacement during wetting increases with increase in the
stress ratio (R), which would indicate high risk of slaking-induced instability at steep slopes.
However, there are no significant effects of the stress ratio (R) on the peak shear strength and
particles crushing due to slaking.
The shear strength, deformation and disintegration characteristics of crushed mudstone
are dependent of initial water content (w) before wetting. Higher slaking level is observed as
the initial water content (w) of the specimen before wetting became lower.
With the increase in Slaking index (SI) there is drastic increase in the shear
displacement upon wetting and drying and decrease in the peak shear strength as well.
6.2
The specimens prepared at high initial density (ρ) show smaller shear displacement
during wetting.
The creep shear displacement caused by wetting seems to be decreased with increase in
number of wetting and drying cycles of cycle, which may be attributed to the specimen
densification due to previous wetting and drying processes.
One of the noticeable behaviors observed in these experiments is a quite large creep
displacement during the drying processes. The creep displacements during drying increase with
the increase in the value of stress ratio (R) and Slaking index (SI). Slaking phenomena cause
loss of intra-particles cementation and tensile failure of the weakly crystalline bonded granular
materials due to drying induced pore water suctions. Drainage and evaporation of pore water
which lead to shrinkage and disaggregation of fabric especially around discontinuities during
dry process. Consequently, such an evolution of soil grains produces rounded particles with
relatively high sphericity and smooth circumference which sequentially decreases the
interlocking behaviour (angle of internal friction) of granular medium. Particles slides on each
other during shrinkage causing shear deformation simultaneously.
The strength and stiffness gained during drying process are opposing to the further
increase in shear displacement with reduction of pore water which lead to shrinkage and
disaggregation of fabric especially around discontinuities. Therefore the drying induced
deformation of crushed mudstones is very complex which could be challenging and crucial
importance for strategic geotechnical structures.
Finally, it is seen that the slaking phenomenon is a very complex behaviour which is
dependent of various factors such density of specimen (ρ), water content (w) before wetting,
slaking index (SI) of the specimen, number of cyclic wetting and drying (N), mineralogical
composition etc. The conclusion of this research can be made as following.
dS  f ( SI , w,  , R, N , I D )
Where, dS: Increment shear displacement, SI: Slaking index, w= Water content before
wetting, R = Stress ratio, N = Number of cyclic wetting and drying and ID= Disintegration
index.
However, the factors inside the functions are also interdependent to each other.
Similarly, there are other various factors such as mineral content, stress and deformation
6.3
histories, drying histories etc which influence the increment shear displacement due to slaking.
So, the effects of slaking, quantification of deformation and disintegration and its relationships
with reduction of strength and deformation characteristics of non-conventional materials from
crushed mudstones is still a key concern for researchers
7.3 Recommendations
The author expected large shear deformation with the progress of cyclic wetting and
drying and collapse of specimen after certain number of cyclic wetting and drying. However
the result observed is contrary to the expectation. This may be attributed to size of crushed
mudstone used in the specimen and higher water content before wetting with progress of
wetting and drying cycles. The level of slaking is dependent of size of crushed mudstones and
initial water content before wetting. It is advised that further tests should be performed on
relatively larger crushed mudstones. Similarly, the direct test apparatus should be modified to
reduce water content significantly during drying phase and drying duration as well.
Further tests should be performed to investigate the possible effects of slaking index
(SI), initial density (ρ), confining stress (σv) etc, which enhance analytical description of
slaking effects on strength and deformation behaviour of geomaterials. Some simple laboratory
tests like oedometer test are recommended to explore possible mechanism of slaking and its
effects on strength and deformation.
Large number of rainfall landslide events should be analysed on the basis of
precipitation intensity and pattern and geology of landslide area. Drought period should be
quantified by using some parameters like DDSLR (Dry Days Since Last Rainfall). Similarly,
detailed field and laboratory investigations of landslide areas are recommended to inspect the
subsurface structures and to obtain geotechnical properties of slope materials. Then, real
mechanism of land even could be explained precisely. Finally, a simple model test of
slopes/embankments composed of crushed mudstones under cyclic wetting and drying
condition is advised.
6.4
REFERENCES
Al-Shamrani, M. A. and Al-Mhaidib, A. I. (1999): Prediction of potential

swell of expansive soils using a triaxial stress pathi cel, Quarterly Journal of
Engineering Geology. 32, 45-54.
Al-Shamrani, M. A. and Al-Mhaidib, A. I. (2006): Influence of swell on
shear strength of expansive soils, Geotechnical Special Publication, American Society of
Civil Engineers, Journal of the Geotechnical Engineering
Anwar, H. Z., Simada, H., Ichinose, M. and Matsui, K. (2000): Slaking
phenomenon of the expandable clay bearing rock, Proc. of Indonesian Association of
Geologists, The 29th Annual Convention, Bangdung, Indonesia, 21-22 Novemeber,
2000,66-75
Au, S. W. C. (1998): Rain-induced slope instability in Hong Kong,
Engineering Geolology, 51(1), 1–36
Aziz, M. (2010): PhD dissertation, The University of Tokyo
Barla, M. (1999): Tunnels in Swelling Ground - Simulation of 3D stress
paths by triaxial laboratory testing. Ph.D Thesis, Politecnico di Torino.
Basma, A. A., Al-Homoud, A. S., Malkawi, A. I. H. and AL-Bashabsheh, M.
H. (1996): Swelling–shrinkage behavior of natural expansive clays, Applied Clay
Science, 11, 211–227.
Bell, F. G., Entwisle, D. C. and Culshaw M. G. (1997): A geotechnical
survey of some British coal measures mudstones, with particular emphasis on durability,
Engineering Geology, 46, 115- 29.
Bhasin R., Grimstad, E., Larsen, J. O., Dhawan, A. K., Singh, R., Verma, S.
K. and Venkatachalam, K. (2002): Landslide hazards and mitigation measures at
Gangtok, Sikkim Himalaya. Engineering Geology, 64(4), 351–368
A.1
References
Bhattarai, P., Marui, H., Tiwari, B., Watanabe, N. and Tuladhar, G. R.

(2007): Depth-wise variation of physical and mechanical properties of mustone in
relation to weathering-Cases in several landslides in Niigata prefecture. Journal of the
Japan Landslide Society, 44 (2), 79-89
Botts, M. (1986): The Effects of Slaking on the Engineering Behavior of
Clay Shales, PhD Thesis, University of Colorado (http://vast.uah.edu/publications.)
Botts, M. E. (1998): Effects of Slaking on the Strength of Clay Shales: A
Critical State Approach. Proc. of the 2nd International Symposium on the Geotechnics
of Hard Soils / Soft Rocks, Naples, Italy, October 1998.
Bragg, G. H., Jr., and Zeigler, T. W. (1975): Evaluation and remedial
treatment of shale embankments. Design and construction of compacted shale
embankments. Rep. No. FHWA-RD-75-62, U.S. Federal Highway Administration,
Washington, D.C.
Bragg, G. H., Jr., and Zeigler, T. W. (1978): Technical guidelines, Design
and construction of compacted shale embankments, Rep. No. FHWA-RD-78-141, U.S.
Federal Highway Administration, Washington, D.C.
Braudeau, E., Mohtar, R.H. and Chahinian, N. (2004): Estimating soil
shrinkage parameters, Development in Soil Science, 30, 225-240.
Canton, Y., Sole-benet, A., Queralt, I. and Pini, R. (2001): Weathering of
gypsum-calcareous mudstone under semi-arid environment at Taberanas, SE Spain:
Laboratory and field based experimental approach, Catena, 44(2), 111-132
Cao, Y., Huang, R., Feng, T, Zheng, H. and Lv, H. (2006): Study on the
slaking characteristics of soft rock engineered slopes at a hydroelectric station in
Southwest of China, The Geological Society of London, IAEG2006 Paper number, 695
Cetin, H., Laman, H. and Ertunc, A. (200): Settlement and slaking problems
in the world’s fourth largest rock-fill dam, the Ataturk Dam in Turkey, Engineering
Geology, 56(3-4), 225-242.
Chen, F.H. and Ma, G.S. (1987): Swelling and shrinkage behavior of
expansive clays, Proc. of 6th International Conference on Expansive Soils, New Delhi,
India, 127–129.
Chen, W. M. (1997): The influence of water content on the mechanical
behaviors of mudstone and its application on fill back engineering. Master Degree
Thesis, Department of Civil Engineering, National Cheng Kung University.
R.2
References
Chen, X.Q., Lu, Z.W., He, X.F. (1985): Moisture movement and
deformation of expansive soils, Proc. of the 11th International Conference on Soil
Mechanics and Foundation Engineering, San Francisco, USA, 2389–2392.
Chigiraa, M., Wangb,W., Furuyac T. and Kamai, T. (2003): Geological
causes and geomorphological precursors of the Tsaoling landslide triggered by the 1999
Chi-Chi earthquake, Taiwan, Engineering Geology, 68, 259–273.
Collison, A., Wade S., Griffiths, J. and Dehn, M. (2000): Modelling the
impact of predicted climate change on landslide frequency and magnitude in SE
England. Engineering Geology, 55(3), 205–218.
Cui, L. and O’Sullivan, C. (2006): Specimen size and scale effects of direct
shear box tests of sands, Geotechnical Testing Journal, 29(6), 507-516.
Czarnomski, N., Moore, M., Pypker, T., Licata, J., and Bond, B. (2005):
Precision and accuracy of three alternative instruments for measuring soil water content
in two forest soils of the Pacific Northwest, Canadian Journal of Forest Research,
35(8), 1867-1876.
Day, R.W. (1994): Swell–shrink behavior of compacted clay, Journal of
Geotechnical Engineering, 120, 618–623.
Dhakal G., Yoneda T., Kato M. and Kaneko, K. (2002): Slake durability and
mineralogical properties of some pyroclastic and sedimentary rocks, Engineering
Geology, 65, 31-45.
Dick, J. C. and Shakoor, A. (1992): Lithological controls of mudrock
durability, Quarterly Journal of Engineering Geology and Hydrogeology, 25, 31-46.
Dick, J. C., Shakoor, A., and Wells, N. A. (1994): A geological approach
toward developing a mudrock-durability classification system, Canadian Geotechnical
Journal, 31, 17–27.
Doostmohammadi, R., Moosavi, M and Arabi, B. N. (2008): A model for
determining cyclic swell-shrink behaviour of argillaceous rock, Applied clay science, 42
(1-2), 81-89.
Dunning, S. A., Mitchell, W. A., Rosser, N. J. and Petley D. N. (2007): The
Hattian Bala rock avalanche and associated landslides triggered by the Kashmir
Earthquake of 8 October 2005, Engineering Geology, 93, 130–144
R.3
References
Duttine, A., Tatsuoka, F., Kongkitkul, W. and Hirakawa, D. (2008): Viscous

behavior of unbound granular materials in direct shear, Soils and Foundations, 48 (3),
297-318.
Duttine, A., Tatsuoka, F., Lee, J. and Kongkitkul, W. (2009): Viscous
properties of Toyoura sand over a wide range of strain rate and its model simulation,
Soils and Foundations, 49 (2), 221-247.
Editorial (2001): Weathering and geomorphology, Geomorphology, 41, 1-3.
Editorial (2007): Studies in weathering and slope movements-and
introduction, 87, 101-103.
Einav, I. (2007): Breakage mechanics-part I : Theory, Journal Mechanics
and Physics of Solids, 55(6), 1274–1297
Einstein, H.H. (1996): Tunnelling in difficult ground-swelling behaviour and
identification of swelling rocks, Rock Mechanics and Rock Engineering, 28, 113–124.
Erguler Z. A. and Ulusay R. (2009): Water induced variations in mechanical
properties of clay bearing rocks, International Journal of Rock Mechanics and Mining
Sciences, 46, 355-370.
Erguler, Z. A. and Shakoor, A. (2009): Quantification of fragment size
distribution of clay bearing rocks after slake durability testing, Environmental and
Engineering Geosciences, XV (2), 81-89.
Folk, R. L., Andrew, P. B., and Lewis, D. W. (1970): Detrital sedimentary
rock classification and nomenclature for use in New Zealand, New Zealand Journal of
Geology and Geophysics, 13(4), 937-968.
Franklin J. A. and Chandra, R. (1972): The slake durability test,
International Journal of Rock Mechanics, Mineral Sciences and Geomechanics, 9, 325-
341.
Gemici, U. (2001): Durability of shale in Narldere, Izmir, Turkey, with an
emphasis on the impact of water on the slaking behaviour, Environmental Geology, 41,
430-439.
Gillott, J. E. (1970): Fabric of Leda clay investigated by optical, electron-
optical and X-ray diffraction methods, Engineering Geology, 4, 133-153.
Gokceoglu C., Ulusay, R. and Sonmez, H. (2000): Factors affecting the
durability of selected weak and clay bearing rocks from Turkey, with particular
R.4
References
emphasis on the influence of the number of drying and wetting cycles. Engineering
Geology, 57, 215-237.
Gonzalez, I. J. and Scherer, G. (2006): Evaluating potential damage to stones
from wetting and drying cycle, Measuring, Monitoring and Modeling Concrete
Properties, 7, 685-693.
Goto, S. (1986): Strength and deformation characteristics of granular
materials in triaxial tests, Ph. D. thesis, University of Tokyo, Japan (in Japanese)
Grainger, P. (1984): The classification of mudrocks for engineering
purposes, Quaternary Journal and Engineering Geology, 17(4), 381-387.
Grim, R. (1969): Clay mineralogy: McGraw-Hill Book Company, 596.
Gysel, M. (1987): Design of tunnels in swelling rock. Rock Mechanics and
Rock Engineering, 20, 219–242.
Hachinohe, S., Hiraki, N. and Suzuki, T. (1999): Rates of weathering and
temporal changes in strength of bedrocks in marine terrace in Boso, Peninsula, Japan,
Engineering Geology, 55, 29-43.
Hall, K. and Hall, A. (1996): Weathering by wetting and drying: some
experimental results, Earth surface process and landforms, 21: 364-376.
Hanzawa, H. (1992): A new approach to determine soil parameters free from
regional variation in soil behavior and technical quality, Soil and foundations, 32 (1),
71-84.
Hawlader, B. C., Lee, Y. N. and Lo, K. Y. (2003): Three-dimensional stress
effects on time-dependent swelling behaviour of shaly rocks, Canadian Geotechnical
Journal, 40(3), 512-535
Hawlader, B. C., Lo, K.Y., Moore, I. D. (2005): Analysis of tunnels in shaly
rock considering three-dimensional stress effects on swelling, Canadian Geotechnical
Journal, 42, 1–12.
Holtz, R.D. and Kovacs, W.D. (2003): An Introduction to Geotechnical
Engineering, Civil Engineering and Engineering Mechanics Series. Pearson Education
Taiwan Ldt., 733.
Huang, R. and le, W. (2011): Formation, distribution and risk control of
landslides in China. Journal of Rock Mechanics and Geotechnical Engineering, 3 (2),
97–116.
R.5
References
Huang, S. L., Speck, R. C. and Wang, Z. (1995): The temperature effect on

swelling of shales under cyclic wetting and drying, International Journal of Rock
Mechanics and Mining Sciences & Geomechanics Abstract, 32 (3), 227-236.
Hugett, R. J. (2003): Fundamentals of geomorphology, Routledge, London.
Huppert F. (1988): Influence of microfabric on geomechanical behavior of
Tertiary fine grained sedimentary rocks from Central North Island, New Zealand.
Bulletin of the International Association of Engineering Geology, 38, 83-94.
Ibanez, D. W. and Kronenberg, K. A. (1993): Experimental deformation of
shale: Mechanical properties and microstructural indicators of mechanisms,
International Journal of Rock Mechanics and Mining Sciences & Geomechanics
Abstracts, 30(7), 723-734.
Igwe, O., Sassa, K. and Wang F. (2007): The influence of grading on the
shear strength of loose sands in stress-controlled ring shear tests. Landslides, 4, 43-51.
ISRM (1983): Characterisation of swelling rock, Commission on Swelling
Rock. Pergamon Press, Oxford, UK.
Iwasaki, T. and Tatsuoka, F. (1997): Effects of grain size and grading on
dynamic shear moduli of sands, Soil and Foundations, 17(3), 19-25.
Jakobsen, M. and Johansen, T. A. (2000): Anisotropic approximations for
mudrocks: A seismic laboratory study. Geophysics, 65 (6), 1711-1725
Jewell, R. A. and Worth, C. P. (1987): Direct shear tests on reinforced sand,
Geotechnique, 37(1), 53-68.
Jongpradist, P. and Horii, H. (2007): Influence of water content on and
relationship between strength and creep behaviors of soft rock – experimental
characterization. Rock Mechanics, Fuenkajorn & Phien-wej (eds) © 2007. ISBN 978
974 533 613 1
Kanyaya, J. I. and Trenhaile, A. S. (2005): Tidal wetting and drying on shore
platforms: An experimental assessment: Geomophology, 70, 129-144.
Kirschbaum, D. B., R. Adler, Y. Hong, S. Hill and Lerner-Lam, A. L.
(2009): A global landslide catalog for hazard applications – Method, Results and
Limitations, Journal of Natural Hazards DOI: 10.1007/s11069-009-9401-4.
Kiyota, T., Konagai, K., Sattar, A., Kazmi, Z. A., Okuno, D. and Ikeda, T.
(2011): Breaching failure of a huge landslide dam formed by 2005 Kashmir earthquake,
Soils and Foundations, 51(6), 1179-1190.
R.6
References
Koncagul, E.C. and Santi, P. M. (1999): Predicting the unconfined

compressive strength of the Breathitt shale using slake durability, shore hardness and
rock structural properties, International Journal of Rock Mechanics and Mining
Sciences, 36, 139–53.
Ladd, C. C. (1960): Mechanisms of swelling by compacted clay. Highway
Research Board Bulletin, abstract, 245, 10–26.
Lashkaripour G.R. and Boomeri M. (2002): The role of mineralogy on
durability of weak rocks, Pakistan Journal of Applied Sciences, 2(6), 698-701.
Lashkaripour, G. R. and Ghafoori, M. (2002): Minelalogical controls of
mudrock durability. Engineering Geology for Developing Countries - Proceedings of
9th Congress of the International Association for Engineering Geology and the
Environment. Durban, South Africa, 20 September, 2002
Lings, M. L. and Dietz, M. S. (2004): An improved direct shear apparatus for
sand, Geotechnique, 54(4), 245-256.
Lutton, R. J. (1977): Slaking indexes for design. Design and construction of
compacted shale embankments. Rep. No. FHWA-RD-77-1, U.S. Federal Highway
Administration, Washington, D.C.
Malla, S., Wieland, M. and Straubhaar, R. (2007): Assessment of Long-Term
Behaviour of Ataturk Dam, Proceeding othe the 1st national symposium and exposition
on dam safety, Turkey, 28-30 May, 2007.
Matsukura, Y. and Mizuno, K. (1986): The influence of weathering of the
geotechnical properties and slope angles of mudstone in the Mineoka earth-slide area,
Japan, Earth surface professes and landforms, 11, 263-273.
Matsukura, Y. and Yatsu, E. (1982): Wet-dry slaking of Tertiary shale and
tuff. Transaction, Japanese Geomorphological Union, 3(1), 25–39.
Mauritsch H. J., Seiberl, W., Arndt, R., Romer, A., Schneiderbauer, K. and
Sendlhofer, G. P. (2000): Geophysical investigations of large landslides in the Carnic
Region of southern Austria. Engineering Geology, 56(3–4), 373–388
Means, R. E. and Parcher, J. V. (1963): Physical properties of Soils, Charles
E. Merril Books Inc, Columbus, Ohio.
Mirza, M. A. (1996): Geological map of Azad Jammu and Kashmir,
Geological Survey of Pakistan (GSP).
R.7
References
Mochizuki, A., Mikasa, M., and Kawamoto, S. (1985): Investigation of

settlement of clay fill at a housing development site.’’ Tsuchi-To-Kiso, 33(4), 25–32 (in
Japanese)
Moon V.G., Beattie A.G. (1995): Textural and micro-structural influences on
the durability of Waikato Coal Measures mudrocks. Quarterly Journal of Engineering
Geology, 28, 303-312.
Moosavi, M., Yazdanpanah, M. J. and Doostmohammadi, R. (2006):
Modeling the cyclic swelling pressure of mudrock using artificial neural network.,
Engineering Geology, 87, 178–194.
Morgenstern, N. R., Eigenbrod, K. D. (1974): Classification of argillaceous
soils and rocks, Journal of the Geotechnical Engineering Division, ASCE, 100 (10),
1137-1156
Moriwaki, Y. (1974): Causes of slaking of argillaceous materials. Ph. D
dissertation, Department of Civil Engineering, University of California, Berkeley.
Nakano, M., Asaoka, A. and Constantinscu, T. D. (1998): Delayed
compression and progressive failure of the assembly of crushed mudstones due to
slaking, Soil and Foundations, 38(4), 183-198
Nakano, R. (1967): On weathering and changes of properties of tertiary
mudstone related to landslide. Soils and Foundation, 3, 1-14.
Noda, T. and Nishi, M. (1988): A study of soft-rock embankment
deformation due to water increase, Proc., JSCE, 391(8), 77–86 (in Japanese).
Okamoto, R., Kojima, K. and Yoshinaka, R. (1981): Distribution and
engineering properties of weak rocks in Japan. Weak Rock, 2, p.1269–1283.
Osipov, V. I., Bik, N. N. and Rumjantseva, N. A. (1987): Cyclic swelling of
clays, Applied Clay Science, 2, 363–374.
Panabokke, C. R. and Quirk, J.P. (1987): Effect of initial water content on
stability of soil aggregates in water, Soil Science, 83(3), 185-196.
Parise, M, and Wasowski, J. (1999): Landslide Activity Maps for Landslide
Hazard Evaluation: three case studies from Southern Italy, Natural Hazards 20(2–3),
159–183.
Pejon, O. J. and Zuquette, L. V. (2002): Analysis of cyclic swelling of
mudrocks, Engineering Geology, 67, 97-108.
R.8
References
Pradini, G., Vigna, G., Pini, R., Regues, D. and Gallart, F. (1996): Structure
and porosity of smectitic mudrocks as affected by experimental wetting-drying cycles
and freezing-thawing cycles, Catena, 27 (3/4), 149-165
Qiu, J. Y., Tatsuoka, F. and Uchimura, T., (2000): Constant pressure and
constant volume direct shear tests on reinforced sand, Soils and Foundations, 40 (4), 1-
17.
Qureshi, M. U., Towhata, I. Yamada, S. and Aziz M. (2009): Field and
laboratory tests on risk of slope failure due to weathering of rock materials. Geophysical
Research Abstracts11, EGU2009-7133 , 11
Regues, D., Pardini, G. and Gallart, F.(1995): Regolith behavior and physical
weathering of clayey mudrock as dependent on seasonal weather conditions in a badland
area at Vallcebre, Eastern Pyrenees, Elsevier Science,25, 199-212.
Runqiu, H. (2009): Some catastrophic landslides since the twentieth century
in the southwest of China, Landslides, 6, 69–81.
Sadisun, I. A., Bandono, Shimada, H., Ichinose, M. and Mastui, K. (2003):
Slope instability of road cuts due to rock slaking. Proc. of 12th Asian regional
conference on soil mechanics and geotechnical engineering, 747-750.
Sadisun, I. A., Shimada, H., Ichinose, M. and Mastui, K. (2004): Textural
and mineralogical properties of argillaceous rocks in relation to their propensity of
slaking, engineering geology for sustainable development in mountainous area, 168-
172.
Sadisun, I. A., Shimada, H., Ichinose, M. and Mastui, K. (2005): Study on
the physical disintegration characteristics of Subang claystone subjected to a modified
slaking index test, Geotechnical and Geological Engineering, 23, 199–218.
Sadisun, I.A., Shimada, H. and Ichinose, M. (2002): Improved procedures
for evaluating physical deterioration of argillaceous rocks, In: Y. Lim, C. Tang, X. Feng
and S. Wang (eds.), Proceedings of 2nd Intl. Conf on New Development in Rock Mech.
and Rock Eng., P. R. China, Rinton Press, 36–39.
Safaei, M., Omar, H., Huat, K. B. and Yousof, B. M. Z. (2002): Relationship
between Lithology Factor and landslide occurrence based on Information Value (IV)
and Frequency Ratio (FR) approaches-Case study in North of Iran. Geotechnical
Engineering Electronic Journal, 17, 79-90.
R.9
References
Saffet, Y. (2011): Correlation between slake durability and rock properties

for some carbonate rocks. Bull Eng Geol Environ (2011), 70:377–383, DOI
10.1007/s10064-010-0317-8
Santi P.M. and Koncagul, E. C. (1996): Predicting the mode, susceptibility
and rate of weathering of shales. In: Matheson G., ed., Design with residual materials -
geotechnical and construction considerations. ASCE Geotechnical Special Publication,
63, 12-26.
Santi, P.M. (1998): Improving of jar slake, slake index, and slake durability
tests for shales, Environmental & Engineering Geoscience, 6(3), 385–396.
Sattar, A., Konagai, K., Kiyota, T., Ikeda, T. and Johansson, J. (2010):
Measurement of debris mass changes and assessment of the dam-break flood potential
of earthquake-triggered Hattian landslide dam, Landslides, DOI: 10.1007/s10346-010-
0241-9.
Scherer, G. (2006): Internal stress and cracking in stone and masonry. In: MS
Konsta-Gdoutos (Editor), measuring, monitoring and concrete properties, ed. M.S.
Konsta-Gdoutos (Springer, Dordrecht, Netherlands), 633-641
Seedsman, R. W. (1986): The behavior of clay shales in water, Canadian
Geotechnical Journa, 23, 18–22.
Seedsman, R.W. (1993): Comprehensive rock engineering, Characterizing
clay shales, Pragamon press Oxford, 3,151–164.
Shamburger, J. H., Patrick, D. M., and Lutton, R. J. (1975): Survey of
problem areas and current practices. Design and construction of compacted shale
embankments, Rep. No. FHWA-RD-75-61, U.S. Federal Highway Administration,
Washington, D.C.
Shibuya, S., Mitachi T., and Tamate, S. (1997): Interpretations of direct
shear box testing of sands as quasi-simple shear, Geotechnique, 47 (4), 769-790.
Singh, T. N., Verma, A. K., Singh, V. and Sahu, A. (2005): Slake durability
study of shaly rock and its predictions, Environmental Geology, 47, 246–253.
Soralump, S. (2010): Corporative of geotechnical approach for landslide
susceptibility mapping in Thailand, Proc. of International conference on slope 2010:
Geotechnique and Geosythentics for slope, 27-30, July, Chiangmi, Thailand
R.10
References
Spears, D. A. and Taylor, R. K. (1971): The influence of weathering of the

composition and engineering properties of in situ coal measures rocks. Intenrational
Journal of Rock mechanics and mineral Science, 9, 729-756
Sri-in, T. and Fuenkajorn, K. (2007): Slake durability index and strength
testing of some rocks in Thailand, Rock Mechanics, Fuenkajorn & Phien-wej (eds), 145-
159
Starr, J. L., and Paltineanu, I. C. (2002): Methods for Measurement of Soil
Water Content: Capacitance Devices. In J.H.Dane, and G.C.Topp (ed.) Soil Science
Society of America, Inc., 463-474.
Strohm, W. E., (1978): Field and laboratory investigations, Phase III. Design
and construction of compacted shale embankments. Rep. No. FHWA-RD-78-140, U.S.
Federal Highway Administration, Washington, D.C.
Subba Rao, K.S. and Satyadas, G. G. (1987): Swelling potential with cycles
of swelling and partial shrinkage. 6th International Conference on Expansive Soils, New
Delhi, India, 137–142.
Surendra, M., Lovell, C. W. and Wood, L. E. (1981): Laboratory studies of
the stabilization of non-durable shale. Transportation Research Record, 79, 33-40.
Taheria, A., Sasakia, Y., Tatsuokaa, F. and Watanabe, K. (2012): Strength
and deformation characteristics of cement-mixed gravelly soil in multiple-step triaxial
compression, Soils and Foundations, 52 (1), 126–145..
Takagi, M., Yokata, S., Suga K., Yasoda S. and Ota H. (2010): The actual
situation of the slope of earthfill that collapsed by an earthquake disaster in the Tomei
Expressway Makinohara district, 193-196
Tatsuoka, F., Sakamoto, M., Kawamura, T. and Fukushima, S. (1986):
Strength and deformation characteristics of sand in plane strain compression at
extremely low pressures, Soil and Foundation, 26-1, 65-8
Taylor, R. K. (1988): Coal Measures mudrocks: composition, classification
and weathering processes. Quarterly Journal of Engineering Geology and
Hydrogeology, 21, 85-99.
Taylor, R. K. and Spears, D. A. (1970): The breakdown of British coal
measures rocks. Int J Rock Mech Min Sci, 7, 481–501.
Terzaghi K. (1936): Stability of slopes in natural clay. Proc. Int. Conf. Soil
Mech., Harvard, 1, 161-165.
R.11
References
Terzaghi, K., and Peck, R. B. (1948): Soil mechanics in engineering

practice, Wiley, New York, 300–302.
Tovar, R. D. and Colmenares, J.E. (2011): Effect of drying and wetting
cycles on the shear strength of argillaceous rocks. Unsaturated Soils – Alonso & Gens
(eds), Taylor & Francis Group, London, ISBN 978-0-415-60428-4, 1471-1476.
Towhata, I., Yamazaki, H., Kanatani, M., Lin, C.E. (2002): Direct shear tests
on rock specimens of Tsaoling earthquake-induced landslide site and simple stability
analysis. In: Sassa, K. (Ed.), Proc. Int’l. Symp. Landslide risk mitigation and protection
of cultural and natural heritage, Kyoto, 123– 138.
Tucker, M. E. (1981): Sedimentary Petrology: An Introduction. (59), 986-
996.
U.S. Geological Survey (2000): Landslide hazards. USGS Fact Sheet Fs-
071-00.
Ulusay, R., Arikan, F., Yoleri, M.F. and Caglayan, D. (1995) Engineering
geological characterization of coal mine waste material and evaluation in the context of
back analysis of spoil pile instabilities in a strip mine SW Turkey, Engineering Geology,
40, 77–101.
USBR (2006): Engineering geology field manual, Technical Service Center,
U. S. Department of Interior Bureau of Reclamation.
Vallejo, L. E., Wlesh, R. A., Lovell, C.W. and Robinson, M. K. (1993): The
influence of fabric and composition on the durability of shale. In: Rock for Erosion
Control, ASTM STP 1117:15-70.
Wells, T., Binning, P. and Willgoose, G. (2005): The role of moisture cyclic
in the weathering of quartz chlorite schist in a tropical environment: finding of a
laboratory simulation, Earth surface process and landforms, 30, 413-428.
Wood, D. M. and Maeda, K. (2008): Changing grading of soil: effect on
critical states, Acta Geotechnica, 3, 3-14.
Wu, P. K., Matsushima, K. and Tatsuoka, F. (2008): Effects of specimen size
and some other factors on the strength and deformation of granular soil in direct shear
tests, Geotechnical Testing Journal, ASTM , 31 (1), 45-64.
Yagiz, S. (2010): Correlation between slake durability and rock properties
for some carbonate rocks, Bull Eng Geol Environ, DOI 10.1007/s10064-010-0317-8.
R.12
References
Yamagishi, H (2000): Recent Landslides in Western Hokkaido, Japan. Pure

Applied Geophysics, 157(6–8), 1115–1134.
Yamaguchi, H., Yoshida, K. Kuroshima, I. and Fukuda, M. (1988): Slaking
and shear properties of mudstones. Rock mechanics and power plants, Ramana (ed.),
133-144
Yatshu, E. (1988): The Nature of weathering: An introduction. Sozosha,
Tokyo
Yilmaz, I. and Karacan, E. (2005): Slake durability and its effect on the
dolomite formation in the gypsum, Environmental Geology, 47, 1010-1017
Yoshida, N. Asce, M. and Hosokawa, K. (2004): Compression and shear
behavior of mudstones aggregates. Journal of Geotechnical and Geo-environmental
engineering, 130 (5), 519-525
Yoshida, N., Enami, K. and Hosokawa, K. (2002): Staged compression-
immersion direct shear test on compacted mudstone, J. Test. Eval., 30 (3), 239-244.
Yoshida, N., Morgenstern, N. R., and Chan, D. H. (1991): Finite element
analysis of softening effects in fissured, over consolidated clays and mudstones,
Canadian Geotechnical Journal, 28, 51–61.
Yoshida, N., Nishil, M. Kitamural, M. and Adachi, T. (1997): Analysis of
mudstone deterioration and its effect on tunnel performance. International Journal of
Rock Mechanics and mineral Science, 34, 3-4, Paper No. 353.
Youn, H. and Tonon, F. (2010): Effect of air-drying duration on the
engineering properties of four clay-bearing rocks in Texas, Engineering Geology, doi:
10.1016/j.enggeo.2010.06.012.
Zhang, C. L., Wieczorek, K. M. and Xie, L. (2010): Swelling experiments on
mudstones, Journal of Rock Mechanics and Geotechnical Engineering, 2 (1): 44–51
R.13
APPENDIX I
List of Landslides
S.N. Name Date Location Provinance Longitude Lattitude

Sichuan,
1 Shuangliu landslide 6/6/2000 Gulin 106.175 7 27.929 8
China
Yunnan,
2 Yingjiang landslide 8/14/2000 Yingjiang 97.982 2 24.759 1
China
3 Qianjiangping landslide 7/13/2003 Zigui Hubei, China 110.602 3 30.975 7
Ganchuanxiang Chongqing,
4 5/13/2003 Xiushan 109.102 1 28.574 6
landslide China
Sichuan,
5 Tiantaixiang landslide 9/5/2004 Xuanhan 108.057 1 31.435 5
China
Guizhou,
6 Yingpancun landslide 9/1/2004 Shuicheng 105.014 3 26.731 2
China
Lianglongxiang Sichuan,
7 6/30/2004 Xingwen 105.081 8 28.282 6
landslide China
8 Luotancun landslide 7/20/2004 Tongdao Hunan, China 109.491 9 26.076 8
Guizhou,
9 Yanjiao rockfall 12/3/2004 Nayong 105.248 4 26.697 4
China
10 Shuidonggou rockfall 5/9/2005 Jixian Shanxi, China 110.679 0 36.090 1
11 Qixingjie landslide 6/23/2005 Jian’ou Fujian, China 118.305 7 27.019 6
Fenglongcun debris Zhejiang,
12 9/1/2005 Wencheng 119.882 5 27.848 7
flow China
Qingshan’ao debris
13 6/25/2006 Longhui Hunan, China 110.747 1 27.580 2
flow
Zhejiang,
14 Shimoxia debris flow 8/11/2006 Qingyuan 119.318 5 27.560 4
China
Zhejiang,
15 Pingtoucun debris flow 8/11/2006 Qingyuan 119.307 5 27.540 4
China
16 Zhangjia landslide 7/15/2006 Yongxing Hunan, China 113.185 8 26.359 7

Sichuan,
17 Luomapu debris flow 7/14/2006 Yanting 101.883 6 27.6851
China
Sichuan,
18 Shiqiaotou rockfall 6/18/2006 Kangding 102.186 3 30.114 4
China
19 Hesancun debris flow 7/14/2006 Longhai Fujian, China 117.554 4 24.384 0
Zhongxilinchang
20 7/14/2006 Zhangpu Fujian, China 117.559 4 24.284 7
landslide
Xianwukuang debris
21 7/15/2006 Yizhang Hunan, China 113.318 7 25.646 3
flow
Shannxi,
22 Gaoloucun landslide 10/6/2006 Huaxian 109.663 6 34.448 4
China
A.1
Appendix

Sichuan,
23 Qiantuocun landslide 5/10/2007 Leibo 103.332 9 28.065 0
China
Sichuan,
24 Miaozigou debris flow 5/24/2007 Jiulong 101.679 2 28.610 0
China
Sichuan,
25 Fenglexiang rockfall 5/25/2007 Shimian 102.544 3 29.324 9
China
Sujiahedianzhan Yunnan,
26 7/19/2007 Tengchong 98.257 8 25.339 2
andslide China
Sichuan,
27 Tianwanhe debris flow 8/10/2007 Shimian 102.119 3 29.396 3
China
28 Taipingzhen landslide 6/15/2007 Badong Hubei, China 110.330 4 29.911 9
29 Gaoyangzhai landslide 11/20/2007 Badong Hubei, China 110.328 8 30.622 7
30 Shang’ancun rockfall 6/13/2008 Lvliang Shanxi, China 111.150 4 37.578 7
31 Huashitou landslide 8/9/2008 Maguan Yunnan,China 104.532 5 22.909 7
Sichuan,
32 Renjiaping debris flow 9/24/2008 Beichuan 104.442 6 31.810 0
China
33 Banpocun landslide 4/22/2008 Yuncheng Shanxi, China 110.616 4 35.750 9
34 Yangtiyan landslide 4/26/2009 Zhaotong Yunnan,China 105.054 0 27.859
35 Jiuzhou landslide 5/16/2009 Lanzhou Gansu, China 103.801 9 36.083 4

Chongqing,
36 Jiweishan landslide 6/5/2009 Wulong 107.432 9 29.232 2
China
Guizhou,
37 Minluocun landslide 6/8/2009 Liping 108.759 1 26.082 8
China
Guizhou,
38 Santuancun landslide 6/10/2009 Liping 109.420 6 26.114 7
China
Guizhou,
39 Huoshicun landslide 6/20/2009 Tongzi 106.649 6 28.074 0
China
Guangxi,
40 Chang'anzhen rockfall 7/3/2009 Rong'an 109.385 4 25.265 1
China
Sichuan,
41 Dafengtan landslide 7/12/2009 Xuanhan 108.251 8 31.627 3
China
42 Xinguan debris flow 7/17/2009 Wenxian Gansu, China 104.427 8 33.026 2
Sichuan,
43 Mushurexi debris flow 7/17/2009 Xiaojin 102.233 1 30.792 8
China
Yunnan,
44 Xiaowan landslide 7/20/2009 Fengqing 100.031 2 24.729 2
China
Xiangshuigou debris Sichuan,
45 7/23/2009 Kangding 102.179 0 30.180 7
flow China
Tiexijingguansuo
46 7/25/2009 Huaihua Hunan, China 109.977 3 27.086 5
landslide
47 Nanyueshan landslide 7/25/2009 Huaihua Hunan, China 110.015 7 27.111 6
48 Liaojiaolou landslide 7/25/2009 Huaihua Hunan, China 110.022 4 27.102 7
49 Tianduan landslide 7/25/2009 Huaihua Hunan, China 110.000 1 27.072 9
50 Sanchaxi landslide 7/25/2009 Huaihua Hunan, China 110.023 9 27.098 5
A.2
Appendix

51 Xinhua landslide 7/25/2009 Huaihua Hunan, China 110.0117 27.099 0
52 Shuping landslide 7/25/2009 Hongjiang Hunan, China 110.1686 29.190 1
Shanghengling
53 7/25/2009 Hongjiang Hunan, China 110.028 5 27.126 9
landslide
Sichuan,
54 Houziyan landslide 8/6/2009 Hanyuan 102.781 9 29.313 1
China
55 Huangbi landslide 8/9/2009 Qianshan Jiangxi, China 117.693 8 27.899 4
Zhejiang,
56 Qingliangfeng landslide 8/13/2009 Lin’an 119.003 7 30.086 9
China
Zhejiang,
57 Sanjiezhen landslide 8/15/2009 Chengzhou 120.841 7 29.718 9
China
Sichuan,
58 Maoniugou debris 8/15/2009 Luding 102.178 1 30.0284
China
Zhejiang,
59 Qianwukeng debris 8/16/2009 Quzhou 118.753 5 29.125 4
China
60 Zuobie debris flow 8/16/2009 Nimu Tibet, China 90.080 9 29.339 8
Sichuan,
61 Wenquanxiang debris 8/17/2009 Yanbian 101.867 3 26.687 6
China
62 Machi landslide 9/5/2009 Huating Gansu, China 106.620 0 35.276 3
Sichuan,
63 Longdagou debris flow 9/12/2009 Luding 102.162 8 29.5553
China
64 Xiaogou landslide 9/14/2009 Kangxian Gansu, China 105.564 7 33.206 7
65 Zhangjiazui rockfall 11/16/2009 Zhongyang Shanxi, China 111.200 0 37.446 0
66 Guoguang landslide 11/28/2009 Hengshan Hunan, China 112.652 8 27.411 1
67 Dengfushan landslide 1/8/2010 Nanjing Jiangsu, China 118.769 9 31.992 7

Guizhou,
68 Yangyandi landslide 1/18/2010 Dejiang 108.144 0 28.546 3
China
Guangxi,
69 Dengcun rockfall 2/26/2010 Sihui 112.589 7 23.334 9
China
Shannxi,
70 Shuanghuyu rockfall 3/10/2010 Zizhou 110.027 8 37.614 5
China
Guizhou,
71 Pingzhai rockfall 3/29/2010 Zhijin 105.779 0 26.674 2
China
72 Hongtaoshan rockfall 4/12/2010 Nanping Fujian, China 118.199 4 26.638 8
73 Zengchong landslide 4/23/2010 Qichun Hubei, China 115.585 0 30.297 5
74 Huangni’ao debris flow 5/6/2010 Liling Hunan, China 113.475 9 27.667 3
75 Yanyacun landslide 5/17/2010 Baokang Hubei, China 110.953 8 31.749 7
76 Hefang landslide 5/23/2010 Dongxiang Jiangxi, China 116.629 6 28.221 8
77 Anshancun landslide 6/1/2010 Qianshan Jiangxi, China 117.826 0 28.171 1
78 Chencun landslide 6/2/2010 Rongxian Guanxi, China 110.745 1 22.7933
Guangxi,
79 Jili landslide 6/3/2010 Laibing 109.189 2 23.642 2
China
80 Dayetiekuang landslide 6/12/2010 Huangshi Hubei, China 114.892 4 30.223 4
Sichuan,
81 Shuangjigou landslide 6/14/2010 Kangding 102.289 4 30.435 4
China
82 Yanta landslide 6/14/2010 Nanping Fujian, China 118.185 4 26.644 7
A.3
Appendix

Sichuan,
83 Chenjiawan debris flow 6/16/2010 Jiulong 101.703 5 28.381 8
China
84 Gaotangling debris flow 6/20/2010 Guiyang Hunan, China 112.729 4 25.614 7
85 Guixi landslide 6/20/2010 Guixi Jiangxi, China 117.360 9 28.303 6 0

Guizhou,
86 Guanling landslide 6/28/2010 Guanling 105.330 2 25.956 4
China
87 Shenhe landslide 7/6/2010 Zhushan Hubei, China 110.223 1 32.108 6

Sichuan,
88 Toutang landslide 7/13/2010 Xuyong 105.413 3 28.219 2
China
89 Mingshuiquan landslide 7/14/2010 Xianning Hubei, China 114.338 4 29.715 4
90 Dalingtou rockfall 7/15/2010 Poyang Jiangxi, China 117.012 8 29.6228
91 Damucun landslide 7/15/2010 Xianning Hubei, China 114.616 0 29.821 8
92 Guojiaba landslide 7/16/2010 Zigui Hubei, China 110.684 5 30.912 7

Muchenggou debris Sichuan,
93 7/16/2010 Litang 102.990 3 31.516 8
flow China
Sichuan,
94 Niumachang landslide 7/7/2010 Nanjiang 106.614 6 31.874 3
China
Sichuan,
95 Wangshuiyan landslide 7/17/2010 Suining 105.681 8 30.428 1
China
Huagongchang debris Sichuan,
96 7/17/2010 Hanyuan 102.680 7 29.353 5
flow China
Shannxi,
97 Muzhucun landslide 7/18/2010 Ankang 108.875 9 32.275 6
China
Shannxi,
98 Qiyan landslide 7/18/2010 Ankang 108.725 7 32.562 2
China
Sichuan,
99 Chengnancun rockfall 7/18/2010 Tongjiang 107.258 2 31.907 4
China
Sichuan,
100 Fenghuangjie landslide 7/19/2010 Pingchang 107.148 2 31.442 9
China
Sichuan,
101 Weiganba landslide 7/19/2010 Wanyuan 107.836 2 31.763 3
China
Sichuan,
102 Jinshaba landslide 7/20/2010 Mianning 101.873 2 28.452 9
China
Sichuan,
103 Xinqiaocun landslide 7/20/2010 Longping 105.263 2 29.200 5
China
104 Qinlingcun landslide 7/23/2010 Guangyuan China 105.849 0 32.568 4
Shannxi,
105 Qiao’ergou landslide 7/24/2010 Shanyang 110.055 1 33.486 2
China
106 Diangoucun rockfall 7/24/2010 Huangting Gansu, China 106.636 4 35.218 8
Sichuan,
107 Tei'amocun debris flow 7/24/2010 Ganluo 102.788 1 29.198 2
China
108 Hutang landslide 7/25/2010 Loudi Hunan, China 112.240 5 27.617 2
Yunnan,
109 Mugudianzhan debris 7/26/2010 Gongshan 98.78 2 27.596 4
China
A.4
Appendix

Sichuan,
110 Ermanshan landslide 7/27/2010 Hanyuan 102.743 9 29.322 5
China
111 Yingjunzhen landslide 8/4/2010 Changchun Jilin, China 125.474 7 43.873 3
112 Zhouqu debris flow 8/8/2010 Zhouqu Gansu, China 104.371 8 33.789 3
113 Nanyugou debris flow 8/10/2010 Zhouqu Gansu, China 104.410 7 33.723 8
114 Chenjiawan landslide 8/11/2010 Tianshui Gansu, China 105.722 4 34.576 4
115 Mashicun debris flow 8/12/2010 Chengxian Gansu, China 105.656 4 33.990 2
116 Zhuanghecun landslide 8/12/2010 Zhangchuan Gansu, China 106.139 3 35.023 0
117 Pingwangcun landslide 8/12/2010 Zhangchuan Gansu, China 106.140 6 35.021 5
Sichuan,
118 Wenjiagou debris flow 8/13/2010 Mianning 104.117 1 31.551 7
China
Sichuan,
119 Sanhecun debris flow 8/13/2010 Jiuzhaigou 104.213 4 33.189 0
China
Sichuan,
120 Shenxigou debris flow 8/13/2010 Dujiangyan 103.628 7 31.092 6
China
Sichuan,
121 Sezucun debris flow 8/13/2010 Danba 101.880 3 31.108 7
China
Sichuan,
122 Yazhacun-1 debris 8/14/2010 Jiuzhaigou 103.933 7 33.314 5
China
Sichuan,
123 Dangduocun debris 8/14/2010 Baiyu 99.113 2 31.380 7
China
Long’andianzhan Sichuan,
124 8/17/2010 Songpan 103.713 2 32.401 3
rockfall China
Sichuan,
125 Yazhacun-2 debris 8/18/2010 Jiuzhaigou 103.932 7 33.312 1
China
Sichuan,
126 Yatungou debris flow 8/18/2010 Jiuzhaigou 103.861 9 33.295 0
China
Yunnan,
127 Dongyuegu debris 8/18/2010 Gongshan 98.775 0 27.601 8
China
Sichuan,
128 Zhangjiagou landslide 8/20/2010 Jianyang 104.244 5 30.312 7
China
Sichuan,
129 Pan’angou debris flow 8/23/2010 Xiaojin 102.085 5 30.755 3
China
Shannxi,
130 Huayang debris flow 8/24/2010 Huayin 109.999 9 34.413 1 0
China
131 Dashifang landslide 9/1/2010 Baoshan China 98.986 3 25.574 4
132 Songraocun debris 9/6/2010 Bomi Tibet, China 95.471 2 29.911 9
133 Rangyuhe landslide 9/15/2010 Fangxian Hubei, China 111.043 7 32.038 0
Qinghai,
134 Guomaricun debris 9/20/2010 Tongren 102.043 5 35.570 0
China
Guangdong,
135 Lingtoucun debris 9/21/2010 Yangcun 111.359 7 22.103 2
China
Sichuan,
136 Maliuwa debris flow 9/21/2010 Beichuan 104.436 9 31.751 0
China
137 Patal Landslide 8/26/2006 Achham Nepal 81.3005 29.005
138 Baglung landslide 7/12/2006 Baglung Nepal 83.6659 28.2135
139 Barangay 97 Landslide 3/16/2011 Tacloban City Phillipines 124.9693 11.2196
140 Chittagong landslide 6/11/2007 Chittagong Bangaladesh 91.7951 23.3713
A.5
Appendix

141 Cocuta landslide 3/27/2011 North Columbia USA -72.877 7.4319
142 Mastamandu landslide 6/8/2009 Dadeldhula Nepal 81.3005 29.2269
143 Dharla landslide 8/15/2007 Himanchal India 77.072 30.906
144 Doba Syedan landslide 3/21/2007 Jhelum Pakistan 73.4432 34.338
145 Eungella landslide Mackay Austrailia 148.4822 -21.0979
146 Ramche landslide 8/16/2006 Rasuwa Nepal 85.1786 28.021
147 Attabad landslide 1/4/2010 Attabad Pakistan 74.64941 36.31507
148 La Paz landslide 2/27/2011 La Paz Bolivia -68.1466 -16.4988
149 Dana point landslide 1/11/2011 California USA -117.69412 33.4671
150 Gulmi landslide 8/20/2003 Gullmi Nepal 83.26752 28.08387
151 Oas landslide 1/20/2011 Albay Phillipines 123.4949 13.2584
152 Pasco landslide 1/10/2011 Pasco Peru -72.6931 -12.8576
153 Taplejung landslide 8/8/2002 Taplejung Nepal 87.6666 27.35
154 Ayuttaha landslide Aug-11 Ayuttaha Thailand 100.5235 14.316284
155 Krabi Lanslide 3/29/2011 Krabi Thailand 99.1026 8.124038
157 Mae hong son landslide 3/31/2011 Mae hong son Thailand 87.89 18.73817
158 NW Oregon highway 1/18/2011 Tillamook USA -123.8245 45.4114
159 Bududa landslide 1/10/2010 Bududa Uganda 34.3317 1.0099
160 Darchula landslide 2/5/2010 Darchula Nepal 80.545 28.844
161 Mount Elegon landslide 5/4/2010 Bumwalukani Kenya 37.5152 0.5123
162 Kilimanjaro landslide 11/11/2009 Kilimanjaro Tanzania 37.8088 4.13287
163 Tawang landslide 11/28/2010 Arunanchal India 91.99012 27.5514
164 Gurdi lanslides 9/9/2006 Tanahu Nepal 84.2279 27.9134
165 Hallar Bridge landslide 1/5/2007 Kotli Pakistan 73.70622 33.545227
166 Sungai Sariak landslide 1/8/2007 Sumatra Indonesia 100.9833 0.0667
167 Bandar Seri, landslide 1/8/2007 Begawan Brunei 114.917 4.98333
168 Nuwara Eliya 1/13/2007 Nuwara Eliya Sri Lanka 80.757715 6.965843
169 Sangihe island 1/15/2007 Jakarta Indonesia 125.540125 3.542435
170 Dongala landslide 1/31/2007 Central Sulawesi Indonesia 119.762 -0.6811
171 Banten landslide 2/6/2007 Cadasari Indonesia 106.1 -6.25
172 Semarang landslide 2/19/2007 Central Java Indonesia 110.141 -7.28
173 Rawalakot landslide 2/25/2007 Kashmir Pakistan 73.7666 33.8666
174 Doba landslide 3/19/2007 Kashmir India 74.3333 33.8778
175 Muzaffarabad landslide 3/20/2007 Kashmir Pakistan 73.494 33.5708
176 Islamabad lanslide 3/21/2007 Kashmir Pakistan 73.038023 33.703333
177 Salyana landslide 3/21/2007 Salyana Nepal 81.28859 27.608918
178 Badakhshan landslide 3/28/2007 Bamyan Afaghanistan 69.169616 34.535309
179 Galle landslide 5/4/2007 Galle Sri Lanka 80.217393 6.032751
180 Samarouli landslide 5/8/2007 Jambu-Kashmir India 75.231 33.242
181 Rangmati 6/11/2007 Rangamati Bangaladesh 92.145249 22.63678
182 Indira Bypass 6/13/2007 Amdo Golai India 88.408898 26.732372
183 Sevoke 6/15/2007 Sevoke India 88.526 26.8096
A.6
Appendix

184 Chiayi landslide 6/24/2007 Chiayi County Taiwan 120.4256 23.4535
185 Ratnagiri landslide 6/25/2007 Ratnagiri India 73.5189 16.9648
186 Kozhikode landslide 6/25/2007 Kozhikode India 75.8275 11.2738
187 Wayanad landslide 6/25/2007 Wayanad India 76.074 11.7095
188 Kurnool landslide 6/25/2007 Kurnool India 78.0501 15.8294
189 Cuddappa landslide 6/25/2007 Cuddappa India 78.8205 14.4702
190 Prakasam landslide 6/25/2007 Prakasam India 80.6128 16.6252
191 Karimnagar landslide 6/25/2007 Karimnagar India 79.1492 18.431
192 Vasco landslide 7/30/2007 Vasco India 73.8277 15.4014
193 Lawngtlai landslide 7/30/2007 Lawngtlai India 92.899 22.533
194 Saiha landslide 7/30/2007 Saiha India 92.967 22.479
195 Joshimath landslide 7/30/2007 Joshimath India 79.56 30.557
196 Chamoli landslide 7/30/2007 Chamoli India 79.349 30.4
197 Uttarkashi landslide 7/30/2007 Uttarkashi district India 78.439 30.733
198 Dhading landslide 8/10/2007 Dhading Nepal 84.914 27.863
199 Khotang landslide 8/10/2007 Khotang Nepal 86.85 27.016
200 Surkhet slide 8/10/2007 Surkhet Nepal 81.601 28.601
201 Vernon County lanslide 8/18/2007 WI USA -91.225567 43.59595
202 Cody landslide 8/18/2007 WY USA -109.8661 44.465862
203 Rochester landslide 8/18/2007 MN USA -92.4671 44.0304
204 LaCrosse landslide 8/19/2007 WI, hwy 35 USA -91.219367 43.6796
205 Brownsville landslide 8/19/2007 Minnesota USA -91.280567 43.68896
206 Duli Village landslide 8/24/2007 Rukum Nepal 82.616667 28.616667
207 Neta landslide 8/24/2007 Gulmi Nepal 83.097064 28.087699
208 Argha landslide 8/24/2007 Pyuthan Nepal 83.083333 27.916667
209 Hattisurey landslide 8/24/2007 Kalimpong India 88.430782 27.071335
210 Kabilas landslide 7/31/2003 Chitwan Nepal 84.3500 27.6300
211 Koteswora landslide 7/30/2003 Nuwakot Nepal 84.3083 27.7061
212 Jalbhir landslide 7/30/2003 Chitwan Nepal 84.2578 27.7009
213 Baijnath landslide 7/30/2003 Far western India 79.6082 29.9262
214 Chhip Chhipe landslide 7/30/2003 Tanahu Nepal 84.2500 27.9167
215 Rupendehi landslide 7/30/2003 Rupendehi Nepal 83.4167 27.6667
216 Bara landslide 7/30/2003 Bara Nepal 85.0000 27.0333
217 Siraha landslide 7/30/2003 Siraha Nepal 86.2000 26.6500
218 Krishna Bhir lanslide 7/30/2003 Chitwan Nepal 84.7931 27.7922
219 Gaidakot landslide 7/30/2003 Nawalparasi Nepal 83.6741 27.5352
220 Ratanpur landslide 7/30/2003 Nawalparasi Nepal 83.7062 27.5060
Bhagar Village
221 7/30/2003 Nawaparasi Nepal 83.7623 27.6143
landslide
222 Damauli landslide 8/1/2003 Tanahu Nepal 84.2693 27.9684
223 Hadikhola landslide 8/1/2003 Makwanpur Nepal 84.8400 27.5947
224 Manakamana landslide 8/1/2003 Gorkha Nepal 84.4720 28.0381
225 Tigray landslide 9/11/2003 Tigray Ethiopia 39.5000 13.5000
A.7
Appendix

226 Painreman landslide 1/10/2003 West Kalimantan Indonesia 109.331 -0.023
227 Petropolis landslide 1/11/2003 Rio de Janeiro Brazil -43.1823 -22.5046
228 Borneo landslide 1/11/2003 Borneo Indonesia 109.1489 0.2424
229 Minas Gerais landslide 1/17/2003 Minas Gerais Brazil -43.1756 -22.9100
230 Garut landslide 1/29/2003 Java Indonesia 107.8918 -7.2250
231 Cantilan 1/31/2003 West Java Indonesia 108.4330 -7.1000
232 Sarawak 2/10/2003 Sarawak Malaysia 114.9442 4.0188
233 Dasu landslide 2/18/2003 Kohistan Pakistan 75.3167 35.6167
234 Ghorvodor landslide 2/19/2003 Khatlon Tajikistan 69.0595 37.8644
Neelum Valley
235 2/26/2003 Neelum Valley Pakistan 73.5750 34.4588
landslide
236 Tolti landslide 3/4/2003 Skardu Pakistan 76.1000 35.0333
237 Doda Landslide 3/6/2003 Jammu India 75.5464 33.1517
238 Bawngkawn 5/4/2003 Bawngkawn India 92.7084 23.7417
239 Noabadi landslide 5/5/2003 Akhaura Bangladesh 91.2189 23.8617
240 Lawngtlai landslide 5/5/2003 Lawngtlai India 92.9000 22.5330
241 Baotou landslide 5/11/2003 Guizhou China 104.6750 25.4000
242 Gatara landslide 5/15/2003 Muranga Kenya 37.1500 0.7060
243 Mbarara landslide 5/16/2003 Mbarara Uganda 30.7500 -0.4167
244 Xiangjiang landslide 5/17/2003 Hunan China 112.9500 27.1012
245 Dothupitiya landslide 5/17/2003 Ratnapura Sri Lanka 80.6000 6.6855
246 Gandoh landslide 1/6/2008 Doda, Kashimir India 75.5464 33.1517
247 Gedangan landslide 1/6/2008 Malang Indonesia 112.6278 -7.9720
248 Karnataka landslide 8/9/2008 Karnataka India 74.1100 15.1500
249 Edakumeri, landslide 8/10/2008 Subrahmanya India 75.6050 12.6445
250 Monkey Hill landslide 8/10/2008 Maharashtra India 73.4700 18.7200
251 Gaiparnath landslide 8/10/2008 Chambal ravines India 75.7200 25.0800
252 Hamang landslide 8/10/2008 Ilam Nepal 87.9190 26.9122
253 Hualoy landslide 8/11/2008 West Seram Indonesia 128.4000 -3.3000
Hassan-
254 Yedakumeri landslide 8/12/2008 India 75.6600 12.8450
Mangalore
255 Ayanur landslide 8/12/2008 Karnataka India 75.4520 13.9860
256 Jog falls landslide 8/12/2008 Karnataka India 74.8122 14.2290
257 Myawaddy landslide 8/13/2008 Karen State Burma 97.9380 16.6240
258 Papum Pare landslide 8/14/2008 Arunachal India 94.0000 27.1630
259 Shimla landslide 8/14/2008 Shimla, Mandi India 77.1200 31.0000
Chandibhanjyag
260 8/14/2008 Chitwan Nepal 84.4320 27.7950
landslide
Jammu and
261 Riasi landslide 8/15/2008 India 74.9277 33.0000
Kashmir
Luang Prabang
262 8/15/2008 Luang Prabang Laos 102.5160 18.0900
landslide
263 Bharta landslide 8/15/2008 Kalikot Nepal 81.7557 29.1964
264 Bobadan landslide 8/16/2008 Mogoke Burma 96.4800 22.9000
265 Portarlington landslide 8/16/2008 Portarlington Ireland -7.2068 53.1468
A.8
Appendix

266 Sadi landslide 8/16/2008 Pyuthan Nepal 82.8332 28.0993
267 Dong Thap landslide 8/16/2008 Dong Thap Vietnam 105.5740 10.5780
268 Gathaithi landslide 11/3/2009 Murang Kenya 37.2000 -0.4333
269 Gikingo landslide 11/3/2009 Imenti Kenya 37.6561 0.0571
270 Kangema landslide 11/3/2009 Murang'a Kenya 36.9667 -0.6833
271 Rawang landslide 11/3/2009 Kuala Lumpur Malaysia 101.6352 3.2909
Templer's Park
272 11/3/2009 Jalan Ipoh Malaysia 101.6314 3.2971
landslide
273 Bukhel landslide 11/5/2009 Lalitpur Nepal 85.3500 27.5000
274 Bac Tra My landslide 11/5/2009 Quang Nam Vietnam 108.4671 15.5721
275 St Cyrus landslide 11/6/2009 Aberdeenshire Scotland -2.4170 56.7709
276 Sakhon landslide 11/6/2009 Narathiwat Thailand 101.5081 6.1700
277 Jalan landslide 11/7/2009 Batu Caves Malaysia 101.6826 3.2711
278 Moo 6 landslide 11/7/2009 Yala Thailand 100.9417 6.8241
279 Kawthoung 10/2/2009 Tanintharyi Myanmar 98.5465 9.9988
280 Puddipadi 10/3/2009 Kerala India 75.9957 11.4816
281 Balussery 10/3/2009 Kerala India 75.8289 11.4477
282 Kawthaung 10/3/2009 Tanintharyi Myanmar 98.5465 9.9988
283 Benguet 10/3/2009 Benguet Philippines 120.8512 16.5364
284 Ampukao 10/3/2009 Benguet Philippines 120.6890 16.3452
285 Kyokyezo 10/3/2009 Kabale Uganda 29.9821 -1.2509
286 Pardawaras 10/4/2009 Sedayu Indonesia 104.4480 -5.5208
287 Sedayu 10/4/2009 Semaka Indonesia 104.4842 -5.4721
288 Way Kerap 10/4/2009 Tanggamus Indonesia 104.7608 -5.5160
289 Beckel 10/4/2009 Benguet Philippines 120.5855 16.4564
290 Chhinga Veng 10/5/2009 Aizawl India 92.7173 23.7374
291 Lower Kitma 10/5/2009 Baguio City Philippines 120.5970 16.4020
292 Su-hua Highway 10/5/2009 Yilan Taiwan 121.7195 24.6929
293 western Bhutan 10/6/2009 Western Bhutan Bhutan 90.4997 27.0524
294 Dadeldhura 10/6/2009 Dadeldhura Nepal 80.6750 29.2833
295 Shera 7/1/2009 Mugu Nepal 82.5068 29.7225
296 Dhapade 7/2/2009 Tanahun Nepal 84.2279 27.9447
297 Damauli 7/2/2009 Tanahun Nepal 84.2805 27.9586
298 Deurali 7/2/2009 Kaski Nepal 83.9806 28.2039
299 Umlazi 2/6/2009 Durban South Africa 30.8660 -29.9540
300 Abidjan 6/12/2009 Abidjan Ivory Coast -4.0276 5.3363
A.9

Slaking Characteristics of Geomaterials in Direct Shear Test

Uploaded by

Copyright:

Available Formats

Slaking Characteristics of Geomaterials in Direct Shear Test

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

Slaking Characteristics of Geomaterials in Direct Shear Test

Uploaded by

Copyright:

Available Formats

Slaking characteristics of geomaterials in direct shear test

Supervisor: Associate Professor Takashi Kiyota

Co-supervisor: Professor Junichi Koseki

A Thesis submitted to Graduate School of Engineering, Department of Civil

Department of Civil Engineering

First, I would like to express my deepest gratitude to Associate Professor Takashi

2. CASE STUDY ON RAINFALL INDUCED LANDSLIDES....................................2.1

6. RESULT ANALYSIS AND DISCUSSION……………………...………………….6.1

7. CONCLUSIONS AND RECOMMENDATIONS………...………………………...7.1

1.2 Typical geotechnical problems

1.2.2 Earth embankment and other structures

1.3 Problem statement and its significance

Increment shear deformation, s (mm)

of wetting upon shear 4

displacement is more 2 Saturation starting point HM CG TS GB

1.4 Objectives and aims

1.5 Organization of Dissertation

2.2 Infinite slope analysis

a) Infinite slopes in dry cohesionless soil

The weight of the slice of having a unit length is given by;

b) Infinite slopes in cohesive soils with parallel seepage

N' N U   sat b h cos    w b h cos  …………............. (2.7)

T  W sin  ………………………………………................. (2.8)

equal to the depth of saturation and  is slope of slip surface.

S S c' cos   ( N  U ) tan 

2.3 Landslide events collection

As previously mentioned, the location of landslide events in terms of latitude and

Fig. 2.3 Geological map of China

Fig. 2.4 Geology map of Africa region

Fig. 2.5 Geological map of Nepal

Classification of rainfall intensity

Table 1 Rainfall intensity classification in India

S.N. Intensity (mm/day) Classification

1 <35 mm Light rain

2 35-74.9 mm Rather heavy rain

2 75-124.9 mm Heavy rain

4 >125 mm Very heavy rain

Table 2 Rainfall intensity classification in China

S.N. Intensity (mm/day) Classification

1 <10 mm Light rain

2 10-25 mm Moderate rain

2 25-50 mm Heavy rain

4 50-100 mm Heavier rain

5 >100 mm Extremely heavy

Table 3 Rainfall intensity classification used in this research

S.N. Intensity (mm/day) Classification

1 <30 mm Light rain

2 30-60 mm Moderate rain

2 60-100 mm Heavy rain

4 >100 mm Very Heavy rain

Fig. 2.8 Daily precipitation in Mankamana, Gorkha, Nepal from TRMM

2.4 Distribution of landslides

Percentage of landslides occurring in each group is as follows. Group 1: (Granite

Others: Mudstone, Shale,

Fig. 2.9 Landslides classified by frequency occuring in each rock group

The main objective of this research is to evaluate the slaking characteristics of

Fig. 2.11 Landslides classified on the basis of ranking of rainfall intensity at

3.2 Slaking mechanism and process

Water Air under pressure