Soils and Foundations 2014;54(3):426–438

The Japanese Geotechnical Society

Soils and Foundations

www.sciencedirect.com
journal homepage: www.elsevier.com/locate/sandf

Modelling the compaction curve of fine-grained soils$

Yasir Al-Badrana,b,c,n, Tom Schanzb,c,1
a
Highway & Transportation Department, College of Engineering/Al-Mustansiriya University, Baghdad, Iraq
b
Chair of Foundation Engineering, Soil- and Rock Mechanics, Civil & Environmental Engineering Faculty, Ruhr-Universität Bochum, Germany
c
Building IC E5/ 117, Universitätsstrasse 150, D-44780, Bochum, Germany
Received 23 September 2011; received in revised form 25 September 2013; accepted 24 January 2014
Available online 17 May 2014

Abstract

A theoretical model for the compaction curve of fine-grained soils at various compaction efforts for the entire range of water content is
presented in this study. The prediction method is based on the assumption that the compaction curve represents the state surface at the yield state
in an unsaturated condition. Thus, for each applied compaction effort, the compaction curve relates to one yielding point on the saturated normal
consolidation line (NCL). For a given soil, the model requires the NCL, Src, and one point from any compaction curve to predict the compaction
curves for different compaction efforts. Moreover, the lines of equal suctions on the compaction curves can be determined if the SWCC, the
wetting path, is known. The model introduced here provides additional theoretical understanding of the soil's volume change behavior of the
compaction curve. The model was verified in two ways: first it was verified quantitatively, by experimental results, and second it was verified
qualitatively, by examining the relationships from other models in the literature. The model was further applied to experimental data reported in
the literature on previous static and dynamic compaction tests. The results show that the model fits the experimental data very well. Finally,
a simple chart, based on this model and using only liquid limits, is presented to estimate γdmax and OMC quickly.
& 2014 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved.

Keywords: Modelling; Compaction curve; Compacted soils; Volume change; Yield state; Fine-grained soils; Unsaturated soil; Suction

1. Introduction for fine-grained soils. It is assumed that the compaction curves

The aim of this study is to model the compaction curves at
various compaction efforts for the entire range of water content
☆ which includes all points from different initial conditions. The
International geotechnical classification numbers: A01, D03, D09.
n
Corresponding author. Tel.: þ49 234 32 26078; fax: þ 49 234 32 14150.
E-mail addresses: yasir.al-badran@rub.de,
yaser_albadran@yahoo.com (Y. Al-Badran), tom.schanz@rub.de (T. Schanz).
1
Tel.: þ49 234 32 26135; fax: þ49 234 32 14236.
Peer review under responsibility of The Japanese Geotechnical Society.
reproduce the volumetric yielding surface or the state surface
at the yield state in an unsaturated condition (Al-Badran and
Schanz, 2009a; Al-Badran, 2011). The state surface at the
yield state in an unsaturated condition is a unique surface
position of the yielding surface is defined by the NCL's
constant degree of saturation, Sr-lines (Al-Badran, 2001,
2011). The concept of the Sr-lines is modified according to
the microstructural consideration of Nagaraj et al. (2006a,
2006b). Consequently, each point of the compaction curve
represents a normal consolidated or yield state and results in an
increase in the applied net stress or the compaction effort until
a specific value under a constant water content condition can
be reached for soil in an initially loose state. Thus, for each

http://dx.doi.org/10.1016/j.sandf.2014.04.011
0038-0806/& 2014 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved.
 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438
compaction effort applied, the compaction curve relates to one
yielding point of the saturated NCL. The parameters involved
in this model are functions of the void ratio of the saturated
NCL (eNCL) of the effective stress equivalent to the compaction
effort, the degree of saturation Sr, the critical degree of
saturation Src (the degree of saturation at the optimum moisture
content, OMC), the parameter that controls the influence of the
degree of saturation to increase the void ratio under a constant
compaction effort (R), and the parameter that controls the
position of the maximum value for the void ratio under each
net stress or compaction effort (M). However, for a given soil,
the model requires the NCL, Src, and one point from any
compaction curve to predict the compaction curves for
different compaction efforts. The prediction of any further
compaction curve for a specific compaction effort requires
only the eNCL of the equivalent effective stress. In addition, a
simple chart, based on this model and using only liquid limits,
will present a quick estimation of γdmax and optimum moisture
content OMC. Moreover, the lines of equal suctions on the
compaction curves can be determined if the soil–water
characteristic curve, SWCC (wetting path) is known.
2. Literature review of compaction modelling
Previous studies on soil compaction models can be divided
into two main groups:
(1) Models that use empirical correlations for both cohesive and
cohesionless soils, which relate optimum water content,
OMC, and maximum dry density or unit weight γdmax, to
soil properties, such as the Atterberg index (liquid limit, LL,
and plastic limit, PL), the specific gravity of solids (Gs), and
the grain-size distribution (Davidson and Gardiner, 1949;
Turnbull, 1948; Ring et al., 1962; Ramiah et al., 1970; Jeng
and Strohm, 1976; Livneh and Ishai, 1978; Wang and Huang,
1984; Korfiatis and Manikopoulos, 1982; Al-Kafaji, 1993;
Basheer and Najjar, 1995; Najjar and Basheer, 1996; Blotz
et al., 1998; Sridharan and Nagaraj, 2005; Gurtug and
Sridharan, 2002; Jesmani et al., 2008).
(2) Models for the entire range of compaction curves describ-
ing the compaction characteristics (Joslin, 1959; Rethati,
1988; Hilf, 1991; Howell et al., 1997; Nagaraj and
Bindumadhava, 1992; Nagaraj, 1994; Pandian et al.,
1997; Basheer, 1998, 2001; Li and Sego, 2000; Nagaraj
et al., 2006a; Honda et al., 2007; Kurucuk et al., 2008;
Horpibulsuk et al., 2008a, 2008b).
An early work by Joslin (1959) yielded 26 typical standard
Proctor curves (named Ohio's curves). Rethati (1988) pre-
sented an empirical approach to modelling the compaction
curves based on fitting a simple quadratic equation to the water
content, w, and the dry density, γd, relationship (w–γd). Hilf
(1991) and Howell et al. (1997) used second-, third-, and
fourth-degree polynomial equations to model the compaction
curves. Nagaraj and Bindumadhava (1992) and Nagaraj (1994)
developed a semi-empirical phenomenological model to
427
determine the w–γd relationship separately for the dry of
optimum, DOP, and the wet of optimum, WOP, based on
the LL and the Gs for each CE. Pandian et al. (1997) presented
the linear relationships of w  LL, [w/(Sr) 0.5]–LL, [w (Sr) 2]–
LL, for the standard Proctor test. Basheer (1998) developed
models for both the compaction characteristics and the
compaction curves. Li and Sego (2000) derived an equation
with both soil and compaction effort parameters to predict the
complete compaction curves of fine-grained soils for all w 4 0.
Basheer (2001) suggested empirical models using both statis-
tical regression and ANNs to predict the compaction curves of
cohesive soils based on the soil properties and compaction
efforts. Nagaraj et al. (2006a) introduced an ideal pore model
(linear relationship) to estimate the compaction curves of fine-
grained soils quickly for different compaction efforts for DOP
and WOP. Based on their ideal model, state parameters w/Sr 0.5
and w/Sr 2 were proposed for DOP and WOP, respectively.
Honda et al. (2007) presented a method to determine the
maximum dry density–optimum moisture content (w–γd)
relationship based on the unsaturated normal consolidation
lines (NCL) of the constant water content condition behavior.
Kurucuk et al. (2008) presented an approach to predict the soil
compaction curve during undrained loading. They investigated
the effect of soil suction, stiffness, and pore air pressure on the
shape of the compaction curve.
Several studies (Joslin, 1959; Nagaraj and Bindumadhava,
1992; Nagaraj, 1994; Nagaraj et al., 2006a) yielded a set of
compaction curves that are nearly parallel for all soils for any LL.
Others (Basheer, 2001; Horpibulsuk et al., 2008a, 2008b) have
demonstrated that the compaction curves of natural soils are
rarely parallel to each other (some of them have flatter curves and
others have sharp peaks).
3. Material used
The materials used in this study are pure bentonite 100B,
and a mixture of bentonite and quartz sand, 30% bentonite and
70% sand mixture, 30B (Agus, 2005). The basic properties
investigated in this study were performed based on ASTM
standards (ASTM, 1997) and DIN standards (DIN, 1987). The
1.8
100B 100B-MP
100B-SP
Dry unit weight, (gm/cm3)
1.6 100B-RMP
100B-Pcom0.248
s=1 MPa
s=35 MPa
1.4
1.2
ZAV -
Src = 90%
1
S = 35 MPa
S = 1 MPa
Sr = 30% Sr = 50%
Sr = 70%
0.8
0 20 40 60
Moisture content, (%)
Fig. 1. Dynamic and static experimental compaction curves for different CE,
and points of equal suction for 100B mixtures.
428
2.2
Dry unit weight, (gm/cm3)
2 S = 6.5 MPa
1.8
1.6
S = 0.7 MPa
1.4
30B Sr = 30%
1.2
0 10
Moisture content, (%)
Fig. 2. Dynamic and static experimental compaction curves for different CE,
and points of equal suction for 30B mixtures.
basic properties of the bentonite (100B) are Gs ¼ 2.75, LL ¼
89.1%, PL ¼ 35.9%, shrinkage limit (SL)¼ 14%, and total
specific surface area (SSA) ¼ 400 m2/g, while for sand, Gs ¼
2.65, and total SSA ¼ 0.25 m2/g. The Gs ¼ 2.695 and LL ¼ 30%
for 30B.
For each soil mixture, different compaction efforts were
performed (dynamic: 248 [RP ¼ SP/2.5], 593 [SP], and
2694 [MP] kJ/m3), and static compaction: s¼ 248 kPa) with
different water contents. RP, SP, and MP are the Reduced, the
Standard, and the Modified Proctors, respectively. The total
suction of the compaction tests was measured at the end of the
tests with a Chilled Mirror device. Figs. 1 and 2 show the
experimental results of the compaction curves and the points of
equal suction for the soils used in the investigation.
4. Description of the model
The main assumption of the model is that the compaction
curves reproduce the state surface at the yield state in an
unsaturated condition, as proposed by Al-Badran and Schanz
(2009a) and Al-Badran (2011). It was shown that, for the same
soil, all unsaturated yielding points from different initial
conditions can be linked to a unique state surface of degree
of saturation and water content. The position of the state
surface at the yield state is defined by the NCL of the constant
degree of saturation, Sr-lines (Al-Badran, 2001, 2011), as
shown in Fig. 3. The concept of the Sr-lines is modified
according to the microstructural consideration of Nagaraj et al.
(2006a, 2006b), as shown in Fig. 4b. The modifications are
that the Sr-lines are not parallel, but they have slightly different
slopes, and empirical equations are proposed to determine the
volumetric yielding at each degree of saturation, as shown in
Eqs. (2)–(6), which will be explained later.
The model presents the relationship among the void ratio,
the compaction effort, and the degree of saturation according
to the Hypothesis 1 for Sr-lines. Then, to connect the void
ratio, the net stress (or compaction effort), and the degree of
saturation with suction, Hypotheses 2 and 3 are presented.
Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438
30B-SP
30B-MP
30B-RP
30B-Pcom0.248
s=0.7 MPa
s=6.5 MPa
ZAV
Src = 85%
Sr = 70%
Sr = 50%
20 30
Fig. 3. Constant degree of saturation lines (Sr-lines) that represent the NCL of
unsaturated state (Al-Badran, 2001).
Hypothesis 1. The volume change behavior on the compac-
tion curve (state surface) is governed by the degree of
saturation, and for each degree of saturation there is a straight
line representing the normal consolidation line for this degree
of saturation (Sr-lines).
The saturated NCL for the initially slurry state specimen is
adopted in this model to determine parameter eNCL. Hypothesis 1
was verified by performing a constant degree of saturation test
(for details, see Al-Badran and Schanz, 2009a; Al-Badran, 2011),
in which the degree of saturation was kept constant by
controlling the net vertical stress and suction. Moreover, the
contour lines for the constant degrees of saturation for the yield
state, after back calculations from the unsaturated compression
curves, also show straight lines (Lawton et al., 1989; Al-Badran,
2001; Honda et al., 2007; Sun et al., 2007).
The work of Nagaraj et al. (2006a, 2006b) supports the
relation between the saturated NCL and the Sr-lines from the
microstructural consideration, as shown in Fig. 4. It was
assumed that the change in the microstructure of the unsaturated
soil at a constant water content, due to compression under an
applied net pressure (P  ua) or applied suction (ua – uw), is
due to a change in the radius of the air–water interface in the
micropore enclosed by the saturated clay clusters. To represent
these changes, parameter w/√Sr (or e√Sr) is a convenient
parameter to employ to account for the effects of the degree of
saturation in the range of 40–90%.
The line for the Red Soil in Fig. 4a represents the void ratio–
Net stress data of any degree of saturation by normalizing the
unsaturated NCL's for the Red Soil in Fig. 4b. The lines for the
constant degree of saturation (Sr-lines) in Fig. 4b show that
there is a straight line for each degree of saturation and that the
slope of these lines increases as the degree of saturation
increases, starting from the slope of the saturated NCL.
Therefore, the new yielding volume change model for the
unsaturated soil connects the saturated NCL and the unsatu-
rated NCL (Sr-lines) using empirical equations, as with other
unsaturated models, but here it is based on a microstructural
consideration.
The model assumed that maximum dry density γdmax for all
compaction efforts occurs at the same (constant) degree of
 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438
1.5
e√Sr 1 1
Red Soil
0.5
Brown Soil
Black Cotton Soil
0 0
10 100
Equilibrium pressure, [kPa]
Fig. 4. (a) Nagaraj et al. (2006a) concept ‘each soil has one NCL by normalization’ (after Nagaraj et al., 2006a), (b) Sr-lines concept ‘straight NCL for each degree
of saturation’ for Red Soil.
saturation, and that this saturation is the critical degree of
saturation. The critical degree of saturation, Src, is the degree
of saturation at the air entry value when the air phase state
starts to be continuous and the water phase discontinuous. This
condition (change of phase state) also occurs at the optimum
water content, OMC (Pandian et al., 1997); therefore, Src is
assumed to be equal to the degree of saturation at the OMC.
The critical degree of saturation, Src, is a boundary limit
between the saturated and the unsaturated zones (transition and
residual zones). The soil in the saturated zone (Sr 4 ¼ Src)
behaves as fully saturated soil which is close to the behavior of
the soil in WOP, while the soil in the unsaturated zone
(Sr o Src) behaves as unsaturated soil which is close to the
behavior of soil in DOP. Moreover, it is assumed that the value
for the critical degree of saturation, Src, for fine-grained soil is
not largely affected by the change in compaction energy. This
is because the water at the air entry value, AEV, is under
considerable stress (SP and MP) and is almost located in the
micro-pores size which is not largely affected or remains
unchanged with the application of compaction energy, as can
be observed in the pore-size distributions for the mixture of
50% bentonite–50% sand (Arifin, 2008). The experimental
results of this study support this assumption, as seen in Figs. 9
and 10. The experimental results of Romero (1999) provide
good evidence of such behavior. The critical degree of
saturation, Src, equal to 90%, gives good predictions of the
compaction curves shown in Fig. 12, and this value is very
close to that obtained from the SWCC in Romero (1999).
A change in the number of meniscus water bridges in the
unsaturated soil (as the degree of saturation changes), produces
a change in their stabilising effect, and hence, leads to the
possibility of yielding and plastic compression during the
increase in the degree of saturation (wetting) (Wheeler et al.,
2003). However, in addition to the effect of net vertical stress
(or compaction effort), there is a strong relationship between
the volume change and suction, considered by the change in
the degree of saturation, according to the following hypothesis:
429
1.5
e
Red Soil
0.5 Sr =100%
Sr =90%
Sr =69%
Sr =49%
1000 10 100 1000
Equilibrium pressure,[KPa]
Hypothesis 2. If the fine-grained soils are kept in a drying or
wetting state for the unsaturated state water content and dry
density relationship located under the critical degree of
saturation, there is no density effect on the gravimetric water
content–suction relationship.
It is worthwhile to mention that Hypothesis 2 is valid only
for the dry of optimum, DOP. Therefore, for water contents
less than the OMC, the water content–dry density relationship
for constant suction represents a vertical line (different dry
densities at a constant water content that corresponds to the
wetting path). For the wet of optimum, WOP, the contour line
for the constant suction can be calculated by adopting a linear
relationship between Sr and suction to consider the effect of
density on the wetting path, as seen in Figs. 9 and 10.
The results of contour lines of equal suction in the
compaction curves of the experimental program, seen in
Figs. 1 and 2, and the results of Delage and Graham (1995),
Gens et al. (1995), Li (1995), Romero (1999), Tarantino and
Tombolato (2005), and Agus (2005) support Hypothesis 2.
The results showed that for the dry side of the compaction
curve, when the soils are in an unsaturated state, the lines of
constant suction are almost vertical. In other words, for fine-
grained soils on the dry side of the compaction curve, the
suction is a function of the water content regardless of density.
Romero (1999) attributed this behavior to the effects of the
loading mechanism on the macroporosity that does not contain
free water, since at low moisture content levels (on dry side of
the compaction curve), water is mainly absorbed or contained
in the less bonded diffuse layers on the intra-aggregate scale.
The degree of saturation that changes on the dry side of the
compaction curve is associated with microporosity changes
under a constant intra-aggregate water content. When the
matric suction values approach the saturation zone, on the
wet side of the compaction curve, however, the loading
mechanism affects the inter-aggregate water and the contours
of equal suction try to incline in order to converge to the limit
430 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438
condition of the matric suction equal to zero at the full
saturation condition (Sr ¼ 100%), as shown in Fig. 12. How-
ever, the contour lines of equal suction in the compaction
curves of Romero (1999), that pass the optimum moisture
content (OMC), may need some correction. Some of the
contour lines were established using only two different
positions (one at the wet side and the other at the dry side)
making a straight line, which does not necessarily represent the
actual behavior. The structure and the behavior at the wet side
are different from those at the dry side; thus, it is important to
use several points from different locations to draw the contour
lines of equal suction when the line passes from the dry side to
the wet side.
This model assumes that when the applied suction is higher
than the air entry value the pore water will be located only in
the fine pores or pods (Backer and Frydman, 2009). These fine
pores or pods are not affected as the density changes (i.e., the
diameter is unchanged). Therefore, at high suction levels
(higher than the air entry value or in the dry of optimum),
there is no density effect on the gravimetric water content–
suction relationship. When the applied suction is lower than
the air entry value, the pore water is believed to be located at
both the fine pores (pods) and the large pores. Therefore, at
low suction levels (lower than the air entry value or in the wet
of optimum), both density and suction have an effect on the
gravimetric water content value. In the saturated zone (lower
than the air entry value or in the wet of optimum) the model
assumes, depending on the experimental results of this study, a
linear relationship between the degree of saturation and
suction. In other words, at the wet of optimum, the suction
is a function of the gravimetric water content and the dry
density, while at the dry of optimum, the suction is a function
of only the gravimetric water content.
It is important to clarify that the counter lines of equal
suction in the compaction curves can be established by the
wetting process, because each point on the curve represents the
compaction under a constant water content which results in an
increase in the degree of saturation. To consider the effect of
hysteresis, Hypothesis 3 is adopted.
Hypothesis 3. The dry side of optimum in the compaction
curve represents the first wetting path. Therefore, the gravi-
metric water content–suction relationship in this part follows
the wetting path.
The compaction curve can be related to the state surface at
the yield state, which will be explained in the following
paragraph. For the compaction process, the volume change
behavior responds according to the two independent stress
variables, namely, net stress and suction. Generally, to prepare
soil for a compaction test, the soil is firstly mixed with water
and then the mixture disintegrates to make a loose soil
structure in an unsaturated state. To unify all the states, the
value of 1 kPa will be considered as the preconsolidation stress
for this initial loose unsaturated soil. This initially unsaturated
loose soil will produce volumetric yielding (state surface)
when the following two conditions are ensured: (i)
considerable stress ( 41 kPa) and (ii) sufficient time. Any
applied compaction effort value of more than 1 kPa on the
DOP will produce volumetric yielding (state surface) and the
soil structure will immediately reach the equilibrium. The
states of soil, normally consolidated (state surface) and over
consolidated, come from the difference between the value of the
applied stress or energy to the soil skeleton and the yield stress of
the soil in a specific condition (which depends on the dry density,
the water content, or the degree of saturation). When the applied
stress or energy 4 ¼ the yield stress of the soil, the soil is then
normally consolidated or at a yield state (state surface), and when
the applied stress or energyothe yield stress of the soil, the soil
is then over consolidated. In the case of compaction, the applied
stress or energy is specific and constant (during the test) and the
initial yield stress of the soil (before compaction) is relatively low
(due to the open structure of the loose soil), so the yield stress of
the soil will increase by an in increase in density, as a result of
compression under the applied stress or compaction effort, until it
reaches the same value as the applied stress or the compaction
effort. Therefore, all the points on the compaction curve in the
DOP represent the normally consolidated or the yield state.
The initial value for suction remains almost constant
because the water content is constant throughout the process.
At WOP, however, the process is a combination of compaction
and consolidation. Therefore, a sufficient amount of time is
needed to complete the volume change process. In other
words, the experimental results of the compaction curve at
WOP do not immediately reach the complete yield state. As a
result, on WOP, there are some differences between the
predicted (yielding) and the experimental results. With time,
suction will increase in the case of compacted expansive soils
due to incomplete hydration (Agus et al., 2010; Schanz et al.,
2010). This may lead to a change in the state of the soil from
the yield state to the over consolidated state (hardening due to
drying). The unsaturated soil exhibits a very small amount of
volume change during this process. Therefore, the volume
change for the long term will be neglected. Additionally, in
DOP, both dynamic and static compaction techniques give the
same horizontal alignment of soil particle groups (Mitchell,
1993). However, the alignment of the soil particles in WOP
will be affected by the compaction type, which is not taken
into account in this model. Therefore, the model does not
consider the dynamic or the static compaction explicitly. From
the above, the compaction curve in the DOP can be considered
as reproducing the state surface, while for the compaction
curve in the WOP, some difference is expected between the
predicted (state surface) and the experimental results.
5. Locations of Sr-lines on the state surface for constant
compaction effort
The locations of Sr-lines can be determined by performing
double oedometer tests (DOT) for two samples showing the
same initial state (preferring high void ratio and low water
content to intersect a large number of Sr-lines). The results of
Wheeler and Sivakumar (1995) and Casini et al. (2007)
support the Sr-lines concept. By analyzing unsaturated volume
 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438
1.5
II
I III
5
4 soil causes a reduction in dry density (or an increase in the
1 6
void ratio
3 54 void ratio of point 2 in Fig. 5, till reaching the maximum
1 7
2 8
0.5
compression (P=248 KPa)
compaction (CE=248 KJ.m/m3)
Proposed model
0 increases when the applied net stress or compaction effort
0 25 50
Degree of saturation, Sr (%)
Fig. 5. Experimental and proposed models for the void ratio–degree of
saturation relationship at normally consolidated or yield state under 248 kPa
constant net vertical stress or 248 kJ m/m3 compaction effort condition for 30B
soil with five characteristic intervals of degree of saturation (Src ¼ 85%).
change results in a normally consolidated state (compression
curves and static compaction data for different levels of net
stress and water content for 30B and 100B soils), Al-Badran
(2001) concluded: For constant applied vertical net stress or
compaction effort, the range of degree of saturation (from fully
dry to fully saturated condition) shows five different intervals,
as shown in Fig. 5 for 30B soil. Soils samples (for 30B and
100B soils) with different water contents were tested in
compression and static compaction tests. These mixtures were
left for one week to reach equilibrium. Each sample was
initially placed in its loose state, and then a constant net stress
was applied for the required period to reach equilibrium with
respect to deformations. This time required to achieve equili-
brium was shorter for intervals I, II, and III than that for
intervals IV and V. The water content was kept constant during
the tests by covering the cell or the compaction mold with a
plastic bag. Besides the initial weight, after finishing the tests,
the final weights, the sample dimensions, and suction were
measured. A wide range of applied net stress was covered in
these tests to gain a comprehensive understanding of the soil
behavior. Fig. 5 shows the experimental and the proposed
models for the void ratio–degree of saturation relationship at
the normally consolidated or yield state under 248 kPa/kJ
m/m3 constant net vertical stress or compaction effort condi-
tion for the 30B soil with five characteristic intervals of degree
of saturation (Src ¼ 0.85). The experimental data in Fig. 5 are
presented as points and numbered in order to facilitate the
distinction between the different intervals.
The first interval starts from a zero degree of saturation in
which the void ratio is unchanged (constant) due to the
absence of any bonding forces between soil particles, from
points 1 to 2 in Fig. 5. Then, by increasing the degree of
saturation, the second interval comes about when water attracts
the soil particles through the action of surface tension forces of
the air–water interaction, from points 2 to 5 in Fig. 5. The
experimental results show, as in Fig. 5, that the degree of
saturation remains almost unchanged throughout the second
interval. The soil stiffness in the second interval increases due
431
to surface tension forces (suction) hardening of air–water
V
IV
interaction. The hardening of the soil sample due to surface
tension forces of air–water interaction inside the unsaturated
void ratio) as the water content increases, void ratio of point
10
void ratio in the second interval, emax, which equals the void
9
ratio of point 5 in Fig. 5. It is important to mention that each
eNCL point in the second interval, i.e., points 2–5 in Fig. 5, resulted
from individual tests. The experimental results show that the
degree of saturation at the maximum void ratio, Sr(emax),
75 100
increases. Afterwards, the third interval sets off in which the
void ratio decreases as the degree of saturation increases, from
points 5 to 9 in Fig. 5. The third interval is the main part of the
volumetric yielding behavior and for this interval Nagaraj et al.
(2006b) established Eq. (1). Eq. (1) shows unique paths on (e/
eL)√Sr versus (p  ua) plots for different soils and for different
conditions (saturated and unsaturated).


e pffiffiffiffiffi
Sr ¼ 1:00 0:21 logðp  ua ÞE3 ð1Þ
eL
where e is the yield void ratio for both saturated and unsaturated
void ratios, eL is the void ratio at the liquid limit state (eL ¼ Gs x
LL), Sr is the degree of saturation, and p ua is the net stress.
The third interval continues until reaching the critical degree
of saturation (Src) with the lowest void ratio or the maximum
dry density, point 9 in Fig. 5, that can be reached under a
specific net stress or compaction effort. Then, the fourth
interval is reached, from points 9 to 10 in Fig. 5 and the void
ratio increases again as the degree of saturation increases until
reaching the full saturation condition (the fifth interval, point
10 in Fig. 5). For this interval, the void ratio is equal to the
void ratio at the saturated NCL for the specific net stress
(eNCL). Equations governing each interval are shown below:
The first interval is from a zero degree of saturation to the
degree of saturation at the maximum void ratio Sr (emax), from
points 1 to 2 in Fig. 5. In this interval, the void ratio is
unchanged (constant) due to the absence of any bonding forces
between soil particles. Then, by increasing the degree of
saturation, the second interval comes about when water attracts
the soil particles through the surface tension forces.
euns  I ¼ edry ð2Þ
where edry is the void ratio of the fully dry soil.
The second interval represents the degree of saturation at the
maximum void ratio, Sr(emax), from points 2 to 5 in Fig. 5.
The void ratio in this interval, euns-II, can be calculated as
Gs w
Gs w ¼ Srðe maxÞ euns -euns  II ¼ ð3Þ
Srðe maxÞ
During the second interval, the experimental results of this
study (see Figs. 1 and 2 or 9 and 10) show that the degree of
saturation remains almost unchanged. The experimental results
show that the water content at emax (w(emax)) can be predicted
432 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438

according to the following empirical equation.
  
eNCL
wðe maxÞ ¼ Gs þ U 100 Gs M
Gs
where eNCL is the void ratio of the saturated NCL and M is the
parameter that controls the position of emax.
The behavior of soil in the second interval gives an
additional peak to the compaction curve on the dry side.
Several experimental studies show that in some cases the
compaction curve has double peaks (e.g., the results of Hindi,
1967; Lee and Suedkamp, 1972; Ellis, 1980; Razouki et al.,
1980; Dixon et al., 1985; Benson et al., 1997; Sun et al., 2007;
Razouki et al., 2008).
The third interval covers the range from the degree of
saturation at the maximum void ratio, Sr(emax), to the critical
degree of saturation, Src, from points 5 to 9 in Fig. 5. The
following unique equation (Eq. (5)) can be used for low to
high plasticity fine-grained soils (Fig. 6), which is equivalent
to the approach used by Nagaraj et al. (2006a, 2006b) as in
2
SP-100B1-Exp.
SP-100B1-Eq.5
SP-30B1-Exp.
SP-30B1-Eq.5
Void ratio
Eq. (1), and Horpibulsuk et al. (2008a, 2008b):
eNCL Src
euns ¼ ð5Þ
ð4Þ ðSrÞR
where R is the parameter that controls the degree of water
sensitivity (slope of compaction paths) on the dry and wet
sides of optimum.
The fourth interval ranges from the critical degree of
saturation up to the fully saturated state (Sr=100%), from
points 9 to 10 in Fig. 5. All the degrees of saturation in the
fourth interval are greater than the critical degree of saturation;
therefore, the current degree of saturation is used in Eq. (5)
instead of the critical degree of saturation:
eNCL Sr
euns ¼ -euns ¼ eNCL ðSr Þ1  R ð6Þ
ðSrÞR
The fifth interval is defined by a degree of saturation equal to
100%, point 10 in Fig. 5; the void ratio is equal to the void
ratio at the saturated NCL (i.e., eNCL).
6. Model steps
Fig. 7 summarizes the different steps to predict the
compaction curve. The curve of compaction effort equivalent
to effective stress P1 can be estimated based on point B that

represents the eNCL corresponding to effective stress P1. From

1 point B, by using an additional model parameter (i.e., Src, R,
and M), the void ratio for any degree of saturation or water
content can be calculated (according to Eqs. (2) to (6) for Sr-
lines). The next step will be the conversion of the void ratio to
0
the dry density to obtain the compaction curve. Parameters Src,
20 40 60 80 100 R, and M are constant for each soil and each parameter requires
Degree of saturation, Sr (%) only one experimental point to be determined. To predict the
Fig. 6. Comparison between exp. results of standard compaction tests and
curve for other compaction efforts, such as that equivalent to
Eq. (5) for 100B (Src ¼90% and R¼ 0.43) and 30B (Src ¼ 85% and R ¼ 0.5) effective stress P2, it requires no more than updating parameter
soils. SP is the Standard Proctor test. eNCL which corresponds to effective stress P2, point A. Fig. 8

saturation line
B
B
Void ratio

A
A
Void ratio

NCL

P1 P2
log effective stress Moisture content

ZAV
Dry unit weight

CE2 = P
A

CE1 = P1 B

Moisture content

Fig. 7. Sketch showing the steps using the proposed model to predict the compaction curve.
 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438 433

represents the predicted compaction curves of different com-

paction efforts using the state surface model for the 30B soil.

7. Verification of model

The new compaction curve model has been verified in two

ways: first quantitatively, by experimental results, and second
qualitatively, by examining the relationships from other
models in the literature.

7.1. Experimental tests and materials used

The compaction curve model hypotheses were verified by

analyzing the residuals between the predicted and the experi- Fig. 9. Dynamic and static compaction curves for different compaction efforts
mental compaction curves (static and dynamic). The predicted for 100B mixture, the points are the experimental results and the lines are the
predicted results.
curve is determined by using Eqs. (2)–(6) for all five intervals
of volumetric yielding behavior. These equations require the
value of the void ratio of saturated NCL eNCL at the net stress
equivalent to each compaction effort, the critical degree of
saturation, R, M, and the degree of saturation. The net stress
equivalent to each compaction efforts is 248 kPa for RP,
593 kPa for SP, and 2694 kPa for MP. The critical degrees of
saturation, Src, are 90% and 85% for 100B and 30B,
respectively. The R values (Eq. (5)) are 0.43 and 0.5, and
the M values (Eq. (4)) are 0.4 and 0.24 for 100B and 30B,
respectively.
In the water content–dry density relationship, the contour
lines for equal suctions are vertical in DOP (Hypothesis 2).
For WOP, the contour lines for constant suction are calculated
by adopting a linear relationship between the degree of
saturation and suction to consider the effect of density on the
wetting path. Fig. 10. Dynamic and static compaction curves for different compaction
Figs. 9 and 10 show the experimental and the predicted efforts for 30B mixture, the points are the experimental results and the lines are
the predicted results.
compaction curves, and the contour line locations of constant
suction for the 100B and 30B mixtures. The results show that suction values seem to follow a scanning path due to drying or
the model agrees well with the experimental results of the a reduced water content, while taking a specimen from the
compaction tests. The predicted constant suction lines almost compaction mold until measuring the suction in a chilled
perfectly match the experimental data with a slight shift mirror device.
towards a lower water content due to the predicted results
representing the wetting path (Hypothesis 3). The measured
7.2. Verifying the relationships of previous compaction
curve models

The proposed model has the ability to explain most of the

Fig. 8. Predicted compaction curves for different compaction efforts using the
volumetric yielding surface model for 30B soil.
relationships given by other compaction models, such as the
linear relationships between Atterberg limits (liquid limits LL
and PL), Gs, water content w, degree of saturation Sr, and the
void ratio of constant water content, e(cons w), and between the
compaction effort on a logarithmic scale (Al-Kafaji, 1993;
Pandian et al., 1997; Blotz et al., 1998; Sridharan and Nagaraj,
2005; Gurtug and Sridharan, 2002; Nagaraj et al., 2006a,
2006b; Jesmani et al., 2008). All the above relationships can
be explained by the semi-linear relationship between void ratio
and net stress in the saturated NCL under constant or different
values of the liquid limit (LL) for the same soil or different
types of soils, respectively. It is important to mention that all
434 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438

the presented relationships came from a regression analysis 1.2

Horpibulsuk et al, 2008a-(Eq.8)
(fitting) of experimental results.
Model-Eq. 14
Honda et al. (2007) presented equations based on the 1
assumptions that the unsaturated normal compression lines in

OMC/OMCsp
the constant water content condition (although the degree of
saturation is increasing) are straight in the void ratio–ln (s) 0.8
plot, as long as the degree of saturation is not close to 100%.
This assumption, according to the proposed model, is correct
0.6
because the compression lines in the yielding state follow the
Sr-lines according to its degree of saturation. The positions of
the Sr-lines are fixed for each soil. Therefore, all the normal 0.4
compression lines for different initial dry densities (under the 100 1000 10000
Copaction Effort, CE (kJ.m/m3 )
same water content) follow one straight line according to
Fig. 11. Normalization of OMC–compaction effort relationship.
Gs U w ¼ Sr Ue-euns ¼ Gs U w=Sr ð7Þ
rewritten as Eq. (14).
Horpibulsuk et al. (2008a) observed that, even though the
OMC values are different for different soils, their rate of eNCL ¼ eLL ð1:122 0:2343 U log sÞ ð13Þ
decrease with compaction effort is almost the same for all
fine-grained soils, which is evident because the ratios for OMC
OMC/OMCSP are almost the same for the same compaction ¼ 2:375 0:496 logðCEÞ ð14Þ
OMCSP
effort:
OMC As shown in Fig. 11, Eq. (14) is close to Eq. (8). It is
¼ 2:09 0:39 logðCEÞ ð8Þ believed that the difference between Eqs. (8) and (14) is a
OMC SP
result of the fact that Eq. (13) is a fitting equation for only 11
where OMCSP is the OMC for a standard proctor. different soils with a limited range of LL (36% o LL o 160%)
Eq. (8) can be explained by the relationship between NCL and a limited range of stress (about 800 kPa). However, Eq. (8)
and the liquid limit (LL) by using different values for the was generated by fitting 24 different soils with LL (17% o
liquid limit to cover the range of fine-grained soils. The LL o 25%) and a larger range of compaction effort (356–
optimum moisture content (OMC), according to Eq. (6) is 2694 kJ m/m3).
the following:
1
OMC ¼ eNCL ðSrc Þ2  R ð9Þ
Gs
8. Applications
OMC ¼ eNCL K ð10Þ
The model was applied to several datasets from the literature
2R with a large range of liquid limit values (37–180%). Four
ðSrc Þ
K¼ ð11Þ different soils were chosen (Table 1): natural Boom clay [Boom]
Gs
(Romero, 1999), pure calcigel (calcium-type) bentonite [100B2],
where K is constant for each soil (Gs and Src are constant for bentonite–sand mixture [50B2], and [30B2] (from Agus, 2005;
the same soil). Thus, the ratio between OMC/OMCSP can be Arifin, 2008). Fifteen experimental datasets of compaction
calculated as curves for the four soils were collected; nine static compaction
OMC eNCL curves were for the natural Boom clay and two dynamic
¼ ð12Þ compaction curves were for each bentonite–sand mixture.
OMC SP eNCLð593kPaÞ
The critical degrees of saturation were determined directly
Now, if the general equation for NCL, presented by Nagaraj from the experimental results. The one-dimensional saturated
and Srinivasa Murthy (1986) (Eq. (13)) is used, Eq. (12) is NCL and SWCC for the natural Boom Clay were taken from
Romero (1999). The one-dimensional saturated NCL for
100B2 was taken from Baille et al. (2010), and for the 50B2
Table 1 and 30B2 mixtures, the one-dimensional saturated NCL test
Properties of the soils used in the application. results were taken from Al-Badran and Schanz (2009b). The
Soil symbol Gs LL% Src% R M
SWCC of all the bentonite–sand mixtures were obtained from
Arifin (2008).
Boom 2.7 58 90 0.35 0.35 Figs. 12–15 present the predicted and the measured compac-
100B2 2.8 180 90 0.04 – tion curves for the above selected soils. The predicted results
50B2 2.725 74 86.5 0.15 –
show a good agreement with the experimental results for both
30B2 2.695 37 85 0.45 –
compaction and suction data.
 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438 435

Fig. 12. Static compaction curves and contour lines of equal matric suction for
natural Boom clay (exp. Data from Romero, 1999), the points are the Fig. 15. Dynamic compaction curves and contour lines of equal suction for
experimental results and the lines are the prediction results. 30B2 soil (exp. Data from Agus, 2005; Arifin, 2008), the points are the
experimental results and the lines are the prediction results.

24 70

Optimum moisture content, (%)

MP-Max. Dd-model SP-Max. Dd-model

Max. Dry density, (kN/m3)

22 RP-Max. Dd-model MP-OMC-model 60
SP-OMC-model RP-OMC-model
20
50
18
40
16
30
14
20
12
10 10

8 0
20 40 60 80 100 120
Liquid limit, LL (%)

Fig. 13. Dynamic compaction curves and contour lines of equal suction for
100B2 soil (exp. Data from Agus, 2005; Arifin, 2008), the points are the
experimental results and the lines are the prediction results.
Fig. 14. Dynamic compaction curves and contour lines of equal suction for
50B2 soil (exp. Data from Agus, 2005; Arifin, 2008), the points are the
experimental results and the lines are the prediction results.
9. A stepwise procedure for fast estimation of optimum
moisture content and maximum dry unit weight
A simple chart based on liquid limit values for the straight
forward estimation of the maximum dry density (γdmax) and the
Fig. 16. Optimum moisture content (OMC)-Max. Dry unit weight (γdmax) chart
using the yielding model depending upon LL.
optimum moisture content (OMC), without performing labora-
tory Proctor tests, is presented in this section. This chart
depends on Eq. (6), with Sr ¼ Src to determine the maximum
dry density and Eq. (9) to determine the optimum moisture
content. Eq. (13) from Nagaraj and Srinivasa Murthy (1986) is
used to predict the NCL from the LL to identify the value for
eNCL with different compaction efforts. To simplify the
model, it is assumed that the specific gravity, Gs, taken as
equal to 2.7, and based on the experimental results, the
following values are adopted: Src ¼ 87% and R ¼ 0.5 for the
range of liquid limit from 20% to 120%. Fig. 16 shows
the relationship between the maximum dry density (γdmax) and
the optimum moisture content (OMC) for a large range of
liquid limit values (20–120%) for Reduced Proctor RP (CE ¼
356 kJ m/m3), Standard 1 Proctor SP (CE ¼ 593 kJ m/m3), and
Modified Proctor MP (CE ¼ 2694 kJ m/m3). The maximum dry
density for each compaction effort decreases with an increas-
ing liquid limit. The behavior of the optimum moisture content
is the reverse of that of the maximum dry density. The rate of
increase in the maximum dry density increases slightly as the
compaction effort increases.
Data for 147 different fine-grained soils (LL: 17–110%)
from 24 reported pieces of literature (Blotz et al., 1998; Daniel
and Benson, 1990; Benson and Trast, 1995; Daniel and Wu,
436 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438

24
10. Conclusions

Optimum moisture content,

MP-Max. Dd-model
Max. Dry density, (kN/m3 )

MP-Max. Dd-Exp. data 60

20
MP-OMC-model A theoretical model for the compaction curve of fine-grained
MP-OMC-Exp. data
soils at various compaction efforts for the entire range of water

(%)
40 contents is presented. The basic parameters of the new model
16
are NCL, the compaction effort, the degree of saturation, and
12
20 the critical degree of saturation, the parameter that controls the
influence of the degree of saturation to increase the void ratio
8 0
under constant compaction effort, and the parameter that
0 20 40 60 80 100 120 controls the position of the maximum value of the void ratio
Liquid limit, LL (%) under each net stress or compaction effort. The model can
describe the lines of equal suction on the compaction curves.
24 The new model was validated by comparing the predicted

Optimum moisture content,

SP-Max. Dd-model
results with the experimental results using dynamic and static
Max. Dry density, (kN/m3 )

SP-Max. Dd-Exp. data 60

SP-OMC-model
20 SP-OMC-Exp. data
compaction tests. The new model was further verified by
comparing the predicted results with the experimental data

(%)
40
16 reported by previous static and dynamic compaction tests.
The predicted results of the compaction curves using the new
20 model match the experimental data well. A simple chart, based
12
on this model and using only liquid limits, is presented to
8 0 estimate γdmax and OMC quickly. This chart covers a large range
0 20 40 60 80 100 120
of LL values (20–120%). The proposed chart has been validated
Liquid limit, LL (%)
using data for 147 different clay soils (LL: 17–110%). The chart
generally shows a good agreement with the experimental data.
24
Optimum moisture content,

RP-Max. Dd-model
Max. Dry density, (kN/m3 )

RP-Max. Dd-Exp. data 60

RP-OMC-model
20 RP-OMC-Exp. data

References
(%)

16
12
8 0
0 20 40 60 80 100
Liquid limit, LL (%)
Fig. 17. Optimum moisture content (OMC)-Max. Dry unit weight (γdmax) chart
using the yielding model depending upon LL and the experimental results from
literature: (a) modified proctor (MP), (b) standard proctor (SP), and (c) reduced
proctor (RP).
1993; Al-Zoubi, 2008; Benson et al., 1997; Leong et al., 2007;
Horpibulsuk et al., 2006, 2008a, 2008b; Ellis, 1980; Gurtug
and Sridharan, 2002; Sawangsuriya et al., 2008; Tripathy and
Schanz, 2007; Pereira and Fredlund, 2000; Pandian et al.,
1997; Delage et al., 1996; Tarantino and De Col, 2008;
Hoffmann and Tarantino, 2008; Kabir and Taha, 2004; Sun
et al., 2007; Jotisankasa et al., 2007; Al-Khafaji, 1993;
Romero, 1999; Agus, 2005; Arifin, 2008) were used beside
the results of the present study to validate the proposed chart.
The total data comprise 211 peak compaction points, PCPs: 42
for MP, 147 for SP, and 22 for RP, respectively. Fig. 17
presents a comparison between the new chart for the peak
compaction points and the experimental data. The figure shows
that the new chart results generally show a good agreement
with the experimental data. Thus, the proposed chart can be
used for the fast estimation of the maximum dry density (γdmax)
and the optimum moisture content (OMC).
40
Agus, S.S, Schanz, T, Fredlund, D.G., 2010. Measurements of suction versus
20 water content for bentonite–sand mixture. Can. Geotech. J. 47, 583–594.
Agus, S.S., 2005. An Experimental Study on Hydro-mechanical Characteristics
of Compacted Bentonite–Sand Mixtures (Ph.D. Dissertation). Faculty of
Civil Engineering, Bauhaus-Universität Weimar, Germany.
120
Al-Badran, Y., 2001, Collapse Behavior of Al-Tharthar Gypseous Soil (M.Sc.
Thesis), Civil Engineering, Baghdad University, Iraq,.
Al-Badran, Y., 2011. Volumetric Yielding Behavior of Unsaturated Fine-
Grained Soils (Ph.D. Thesis), Ruhr-Universität Bochum, Germany.
Al-Badran, Y. Schanz, T., 2009a. Yielding surface model of volume change
for fine-grained soils.In: 4th Asia-Pacific Conference on Unsaturated Soils,
23–25 November 2009, Newcastle, Australia, pp. 863–871.
Al-Badran, Y., Schanz, T., 2009b. Soil–water characteristic curve and
consolidation for fine-grained soils. In: 4th Asia-Pacific Conference on
Unsaturated Soils, 23–25 November 2009, Newcastle, Australia, pp. 243–249.
Al-Kafaji, A.N., 1993. Estimation of soil compaction parameters by means of
Atterberg limits. Q. J. Eng. Geol. 26, 359–368.
Al-Zoubi, M.S., 2008. Swell characteristics of Natural and treated compacted
clays. Electron. J. Geotech. Eng. 13 (Band D).
Arifin, Y.F., 2008. T-H-M Behavior of Compacted Bentonite–Sand Mixtures:
An Experimental Study (Ph.D. thesis). Faculty of Civil Engineering,
Bauhaus-Universität Weimar, Germany.
ASTM, 1997. Soil and Rock. ASTM International, West Conshohocken, PA
(Annual Book of Standards. Volumes 04.08 and 04.09).
Backer, R., Frydman, S., 2009. Unsaturated soil mechanics: critical review of
physical foundations. Eng. Geol. J. 106, 26–39.
Baille, W., Tripathy, S., Schanz, T., 2010. Swelling pressure and one-
dimensional compressibility behavior of bentonite at high pressure. Appl.
Clay Sci., 1–10. http://dx.doi.org/10.1016/j.clay.2010.01.002.
Basheer, I.A., 2001. Empirical modeling of the compaction curve of cohesive
soils. Can. Geotech. J. 38, 29–45.
Basheer, I.A., Neuromechanistic-Based Modeling and Simulation of Constitu-
tive Behavior of Fine-Grained Soils (Ph.D. Dissertation), Kansas State
University, Manhattan, Kans, 1998.
 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438
Basheer, I.A., Najjar, Y.M., 1995. A neural network for soil compaction. Numerical
Models in Geomechanics, Numerical Methods in Geomechanics, vol. 5.
NUMOG. A.A. Balkema, Rotterdam, The Netherlands, 435–440.
Benson, C., Trast, J., 1995. Hydraulic conductivity of thirteen compacted
clays. Clays Clay Min. 43 (6), 669–681.
Benson, C.H., Gunter, A.G., Boutwell, G.P., Trautwein, S.J., Berzanskis, P.H.,
1997. Comparison of four methods to assess hydraulic conductivity. J.
Geotech. Geoenviron. Eng. 123 (10), 929–937.
Casini, F., Vassallo, R., Mancuso, C., Desideri, A., 2007. Interpretation of the
behavior of compacted soils using Cam-clay extended to unsaturated
conditions. In: 2nd International Conference Mechancis of Unsaturated
Soil,7–9 March, Weimar, Germany.
Blotz, L.R., Benson, C., Boutwell, G., 1998. Estimating optimum water
content and maximum dry unit weight for compacted clays. J. Geotech.
Eng. 124, 907–912.
Daniel, D., Benson, C., 1990. Water content-density criteria for compacted soil
liners. J. Geotech. Eng. 116 (12), 1181 (1130).
Daniel, D., Wu, Y., 1993. Compacted clay liners and covers for arid sites. J.
Geotech. Eng. 119 (2), 223–237.
Davidson, D. T., Gardiner, W. F., 1949. Calculation of standard proctor
density and optimum moisture content from mechanical, analysis, shrink-
age and factors and plasticity index. In: Highways Research Board
Proceedings of the Annual. Meeting, vol. 29, pp. 447–481.
Delage, P., Graham, J., 1995. Understanding the behaviour of unsaturated soils
requires reliable conceptual models: state of the art report. In: Alonso, E.E.,
Delage, P. (Eds.), In Proceedings of the first international conference on
unsaturated soils. Rotterdam, Balkema, pp. 1223–1256.
Delage, P., Audiguier, M., Cui, Y., Howat, M., 1996. Microstructure of a
compacted silt. Can. Geotech. J. 33, 150–158.
DIN. Baugrund und Grundwasser. Benenen und Beschreiben von Boden und
Fels, 1987, Deustche Institut für Normung e.V., Beuth Verlag GmbH, Berlin.
Dixon, D.A., Gray, M.N., Thomas, A.W., 1985. A study of the compaction
properties of potential clay-sand buffer mixtures for use in nuclear fuel
waste disposal. Eng. Geol. 21, 247–255.
Ellis, C. I., 1980. Soil compaction at low moisture content-field trials in Sudan.
In: 7th Regional Conference for Africa on Soil Mechanics and Foundation
Engineering, Accra, Ghana, 1–7 June.
Gens A., Alonso E.E., Suriol J., and Lloret A. (1995) Effect of the structure
and volumetric behaviour of a compacted soil. In: Alonso and Delage
(Eds.), In Proceedings of the 1st International Conference on Unsaturated
Soils, Paris, France, Vol. 1: 83–88.
Gurtug, Y., Sridharan, A., 2002. Prediction of compaction characteristics of
fine-grained soils. Geotechnique 52 (10), 761–763.
Hilf, J.W., 1991. Compacted fill. In: H.-Y. Fang (Eds).Foundation Engineering
Handbook.. Chapman and Hall Ltd., London, UK, pp, 249–316.
Hindi, A.R. O., 1967. Identification, physical properties and compaction of
swelling soils in western Egypt. Bull. College Eng. 1 (1), 15–38.
Hoffmann, C., Tarantino, A., 2008. Effect of grain size distribution on water
retention behavior of well graded coarse material, unsaturated soils: advances
in geo-engineering. In: Toll, D.G., Augarde, C.E., Gallipoli, D.,
Wheeler, S.J. (Eds.), Proceedings of the 1st European Conference on
Unsaturated Soils. Leiden: CRC Press/Balkema, Durham, UK, pp. 291–298.
Honda, M., Ohno, S., Iizuka, A., Kawai, K., Ohta H. A study on yield stress in
unsaturated clay. In: 3rd Asia Conference on Unsaturated Soils, China,
2007, pp. 219–222.
Horpibulsuk, S., Rachan, R, Katkan, W., 2006. Prediction of compaction
curves at various energies using a one point tests. In: International Symposium
of Lowland Technology, September 14–16, 2006 in Saga, Japan.
Horpibulsuk, S., Katkan, W., Apichatvullop, A., 2008a. An approach for
assessment of compaction curves of fine- grained soils at various energies
using a one point test. Soils Found. 48 (2), 115–126.
Horpibulsuk, S., Katkan, W., Naramitkornburee, A., 2008b. Modified Ohio's
curves: a rapid estimation of compaction curves for coarse- and fine-
grained soils. Geotech. Test. J. 32 (1), 1–12.
Howell, J.L., Shackelford, C.D., Amer, N.H., Stern, R.T., 1997. Compaction of
sand-processed clay soil mixtures. Geotech. Test. J. 20 (4), 443–458.
Jeng, Y.S., Strohm, W.E., 1976. Prediction of the Shear Strength and
Compaction Characteristics of Compacted Fine- Grained Cohesive Soils.
437
United States Waterways Experiment Station, Soil and Pavement Labora-
tory, Vicksburg, Miss.
Jesmani, M, Manesh, A.N., Hoseini, S.M. R., 2008. Optimum water content
and maximum unit weight of clayey gravels at different compactive efforts.
Electron. J. Geotech. Eng. 13 (L).
Joslin, J.C., 1959. Ohio's Typical Moisture–Density Curves. 239. American
Society for Testing and Materials, Special Technical Publication STP
111–118.
Jotisankasa, A., Ridley, A., Coop, M., 2007. Collapse behavior of compacted
silty clay in suction-monitored oedometer apparatus. J. Geotech. Geoen-
viron. Eng. 133 (7), 867–877.
Kabir, M.H., Taha, M.R., 2004. Assessment of physical properties of a granite
residual soil as an isolation barrier. Electron. J. Geotech. Eng. 9 (Bundle B).
Korfiatis, G.P., Manikopoulos, C.N., 1982. Correlation of maximum dry
density and grain size. J. Geotech. Eng. 108, 1171–1176.
Kurucuk, N., Kodikara, J., Fredlund, D. G., 2008. Theoretical modelling of the
compaction curve. In: Unsaturated Soils Conference, Advances in Geo-
Engineering, Durham, UK, 2008, pp. 375–379.
Lawton, E.C., Fragaszy, R.J., Hardcastle, J.H., 1989. Collapse of Compacted
Clayey Sand. Journal of Geotech. Engrg. , ASCE, 115 (9), 1252–1267.
Lee, P.Y., Suedkamp, R.J., 1972. Characteristics of Irregularly Shaped
Compaction Curves of Soils. National Academy of Sciences, Washington,
DC (Highway Research Record No. 3811-9).
Leong, E. C., Wee, H. S., Raharajo, H. One-dimensional compressibility of
unsaturated soils. In: 3rd Asian Conference on Unsaturated Soil, China,
2007, pp. 237–242.
Li, H., Sego, D.C., 2000. Eq. for complete compaction curve of fine-grained
soils and its applications. In: Shanklin, D.W., Rademacher, K.R., Talbot., J.R.
(Eds.), Construction and Controlling Compaction of Earth fills, vol. 1384.
American Society for Testing and Materials, Special Technical Publication
STP, pp. 113–125.
Li, Z.M. (1995) Compressibility and collapsibility of compacted and saturated
lossial soils. Proc. 1st Int. Conf. On Unsaturated Soils, Paris. EE. Alonso
and P. Delage (eds.), Balkema, 1:139-144.
Livneh, M., Ishai, I., 1978. Using indicative properties to predict the density–
moisture relationship of soils. Transp. Res. Record 690, 22–28.
Mitchell, J.K., 1993. Fundamentals of Soil Behavior, 2nd edition Wiley, New
York.
Nagaraj, T.S., Lutenegger, A., Pandian, N.S., Manoj, M., 2006a. Rapid
estimation of compaction parameters for field control. Geotech. Test. J.
29 (6), 1–10.
Nagaraj, T.S., 1994. Analysis and Prediction of Compaction Characteristics of
Soils. Indian Institute of Science, Bangalore, India.
Nagaraj, T.S., Bindumadhava, J., 1992. Analysis of compaction characteristics
of fine grained soils. In: Indian Geotechnical Conference, Calcutta, vol. 1,
pp. 441–444.
Nagaraj, T.S., Srinivasa Murthy, B.R., 1986. A critical reappraisal of
compression index Eq.s. Geotechnique 36 (1), 27–32.
Nagaraj, T.S., Schanz, T., Prasad, K.N., 2006b. Generalized State Parameter
for Constitutive Relations. Indian Geotechnical Society, ING, Chennai,
India (14–16 December).
Najjar, Y.M., Basheer, I.A., 1996. Utilizing computational neural networks for
evaluating the permeability of compacted clay liners. J. Geotech. Geol.
Eng. 14, 193–212.
Pandian, N.S., Nagaraj, T.S., Manoj, M., 1997. Reexamination of compaction
characteristics of fine-grained soils. Geotech. Test. J. 47 (2), 363–366.
Pereira, J., Fredlund, D.G., 2000. Volume change behavior of collapsibility
compacted Gneiss soil. J. Geotech. Geoenviron. Eng. 126 (10), 907–916.
Ramiah, B.K., Viswanath, V., Krishnamurthy, H.V., 1970. Interrelationships
of compaction and index properties. In: Proceedings of the 2nd Southeast
Asian Conference on Soil Engineering, Singapore, pp. 577–587.
Razouki, S.S., Haik, M., Spyridon, G., Mosa, S., Wazir, H., Majead, G.,
Ibrahim, E., 1980. On the compaction and CBR behavior of Basrah soil.
Bull. College Eng. 4 (1), 107–133.
Razouki, S., Kuttah, D., Al-Damluji, O., Nashat, I., 2008. Using gypsiferous
soil for embankments in hot desert areas. In: ICE, Construction Materials,
vol. 6, CM2, pp. 63–71.
Rethati, L., 1988. Probabilistic Solutions in Geotechnics. Elsevier, New York.
438 Y. Al-Badran, T. Schanz / Soils and Foundations 54 (2014) 426–438
Ring, G.W., Sallgerb, J.R., Collins, W.H., 1962. Correlation of compaction
and classification test data. Highway Res. Board Bull. 325, 577–587.
Romero, E., 1999. Characterisation and T-H-M behavior of unsaturated Boom
clay: an experimental study (Ph.D. thesis), Uni. Polytecnica de Catalunya.
Sawangsuriya, A., Edil, T., Bosscher, P., 2008. Modulus-suction-moisture
relationship for compacted soils. Can. Geotech. J. 45, 973–983.
Schanz, T., Arifin, Y., Agus, S., Khan, M., 2010. Time effect on the total
suction of bentonites. Soil Found. J. 50 (2), 195–202.
Sridharan, A., Nagaraj, H.B., 2005. Plastic limit and compaction characteristics
of fine-grained soils. Ground Improv. 9 (1), 17–22.
Sun, D.A., Sheng, D., Xu, Y.F., 2007. Collapse behavior of unsaturated
compacted soil with different initial densities. Can. Geotech. J. 44 (6),
673–686.
Tarantino, A., De Col, E., 2008. Compaction behavior of clay. Géotechnique
58 (3), 199–213.
Tarantino, A., Tombolato, S., 2005. Coupling of hydraulic and mechanical
behaviour in unsaturated compacted clay. Géotechnique 55 (4), 307–317.
Tripathy, S., Schanz, T., 2007. Compressibility behavior of clays at large
pressures. Can. Geotech. J. 44, 355–362.
Turnbull, W.J. Computation of the optimum moisture content in the moisture–
density relationship of soils. In: Proceedings of the 2nd International
Conference on Soil Mechanics and Foundation Engineering, Rotterdam,
vol. 4, 1948, pp. 256–262.
Wang, M.C., Huang, C.C., 1984. Soil compaction and permeability prediction
models. J. Environ. Eng. 110, 1063–1083.
Wheeler, S.J., Sivakumar, V., 1995. An elasto-plastic critical state framework
for unsaturated soils. Géotechnique 45 (1), 35–53.
Wheeler, S.J., Sharma, R.J., Buisson, M.S.R., 2003. Coupling of hydraulic
hysteresis and stress–strain behaviour in unsaturated soils. Géotechnique
53 (1), 41–54.