KSCE Journal of Civil Engineering (2015) 19(4):893-903 Geotechnical Engineering

Copyright ⓒ2015 Korean Society of Civil Engineers

DOI 10.1007/s12205-013-1341-6 pISSN 1226-7988, eISSN 1976-3808
www.springer.com/12205
TECHNICAL NOTE

Behavior of Compacted Lime – (Well-graded) Soil Columns:

Large Scale Tests and Numerical Modelling
Mohammad Reza Malekpoor* and Gholamreza Poorebrahim**
Received June 8, 2013/Revised February 16, 2014/Accepted June 3, 2014/Published Online December 1, 2014

··································································································································································································································

Abstract

This study presents the results of large scale laboratory model tests to investigate the behavior of Compacted Lime-Well-graded
Soil (CL-WS) rigid stone columns in soft soils. Tests were carried out on composite specimens to evaluate the influence of different
parameters such as: the diameter of the column, the slenderness ratio, area ratio and the shear strength of the surrounding soil. Finite
element analysis has been also performed using PLAXIS software to compare the results of numerical and experimental modelling.
In order to assess the real behavior of these columns, some tests have been carried out in the field. Based on the results, it was
concluded that CL-WS columns increase the load carrying capacity of soft soils and reduce the settlement. In addition, the results
show the influence of model size on the stiffness of the specimens which means that the load carrying capacity decreases by
increasing the size of models. However, for specimens containing columns with diameter greater than 100 mm, the variations of
stiffness become negligible and hence the results can be used to extrapolate and predict the full size behavior of these columns. A
detailed comparison between the experimental and numerical modelling shows a very good agreement.
Keywords: soft clay, rigid stone column, lime, load intensity, settlement
··································································································································································································································

1. Introduction columns, if installed in loose sands, minimize the likelihood of

One of the major functions of geotechnical engineering is to
design, implement and evaluate ground improvement schemes
for infrastructure projects. During the last decades, significant
new technologies and methods have been developed and
implemented to assist the geotechnical specialist in providing
cost-effective solutions for construction on marginal or difficult
sites (Elias et al., 2006). The choice of appropriate method
depends on many factors such as: structural loads, type of the
structure, soil type, availability of equipments and material,
economical considerations and etc.
One of the techniques extensively used in improvement of soft
soils is stone columns or granular piles. This method consists of
drilling vertical holes in the ground which are then filled with
crushed stones or gravels to form columns or piles confined by
the soil. Because of the relatively high modulus of the columns
in comparison to the weak soil, a large proportion of the applied
vertical load is transferred to the columns (Malekpoor and
Toufigh, 2010).
Stone columns improve the performance of foundation on soft
ground by reducing the settlement to an acceptable level and by
increasing the load carrying capacity (Babu et al., 2012). Field
observations show that stone columns could also accelerate the
rate of consolidation of soft clays (Han and Ye, 1992). Stone
liquefaction of these deposits due to earthquakes because of their
tendency to dilate while shearing and dissipate the excess pore
pressures generated (Mitchell and Huber, 1985).
Stone columns are installed in wide variety of soils, ranging
from loose sands to soft compressible clays (Shivashankar et al.,
2010). The beneficial effects of stone columns on their load-
settlement characteristics of reinforcing soft clay deposits have
been widely demonstrated by several researchers (Engelhardt et
al., 1974; DiMaggio, 1978; Han and Ye, 2001; McKelvey et al.,
2004; Castro and Karstunen, 2010; Murugesan and Rajagopal,
2010; Cimentada et al., 2011; Deb et al., 2011; Fattah et al.,
2011).
It is well established that the stone columns derive their load
carrying capacity from the lateral confining pressure provided by
surrounding soil (Greenwood, 1970; Hughes et al., 1975; Barksdale
and Bachus, 1983). When stone columns are installed in very soft
soils or in layered soils where the top layer is very soft, they may not
derive significant load capacity owing to low lateral confinement
offered by soft soils, which led to excessive bulging and also
squeezing of soft clay into the void space of aggregates (Mckenna et
al., 1975). According to German regulations, the application of
stone columns is generally limited to soils with undrained shear
strength Cu ≥ 15–25 kPa (FGSV (Forschungsgesellschaft f.
Straßenund Verkehrswesen), 1979). Below this strength, the

*Ph.D. Candidate, Dept. of Civil Engineering, University of Kerman, Kerman 761-6914111, Iran (Corresponding Author, E-mail: mreza.malekpoor@yahoo.com)
**Assistant Professor, Dept. of Civil Engineering, University of Kerman, Kerman 761-6914111, Iran (E-mail: gh.r.poorebrahim@gmail.com)

− 893 −
 Mohammad Reza Malekpoor and Gholamreza Poorebrahim
lateral support provided by surrounding soil may be insufficient
to prevent column failure through excessive radial expansion
(Gniel and Bouazza, 2009). There are two other limitations for
using of the stone columns. One relates to the closer spacing of
the columns. It is suggested that for a significant improvement in
bearing capacity for stone column treated ground, approximately
25% of the ground should be replaced by the stones (Wood et al.,
2000). The other relates to the lower length to diameter ratio of
these columns (about 4 to 5 times of the diameter of the column
(Hughes and Withers, 1974; Mitra and Cahttopadhyay, 1999;
Samadhiya et al., 2008)).
In the last decades, these limitations have prompted investigation
into the use of rigid stone columns or rigid inclusions. Unlike
stone columns, rigid inclusions derive their stability without any
lateral confinement of the surrounding soil (ASIRI, 2012). These
columns have stiffness significantly greater than that of the
surrounding soil. Nonetheless, this stiffness may vary widely
depending on the type of inclusion developed, which can
include: lime column, vibro concrete column, metal section, etc.
(Simon, 2012).
The first applications of rigid stone columns date from the late
1970s, mainly in road embankments in Scandinavian countries
(Rathmeyer, 1975). A renewed interest in this technique
stemmed from a study about negative side friction by Combarieu
during that time period (Combarieu, 1974 and 1988). Rigid stone
columns appear to be best suited for strengthening the stone
column in locally weak zones (Barksdale and Bachues, 1983).
From the results of the laboratory triaxial compression tests,
Juran and Riccobono (1991) revealed that low-level cementation in
compacted sand columns can significantly improve the settlement
response and load carrying capacity (Juran and Riccobono,
1991). Rigid inclusion ground improvement is now a very cost-
effective foundation solution for common construction projects.
Several land mark applications punctuate its development and
illustrate that this basic concept can be applied equally effectively
to complex construction projects (Simon, 2012).
In spite of the extensive use of rigid stone columns as an
efficient and economical method for soil improvement, few
number of publications on the behavior and design of these
columns are reported in the literature (unlike other column-like
ground improvement techniques such as: piles, vertical sand
drains, etc).
The main objective of this research is to investigate soft soil
improvement using CL-WS rigid stone columns. Lime is used in
this study due to its relative cost-effectiveness compared with
other materials used in rigid stone columns such as cement,
metal sections, etc., besides its approved compatibility with soft
soils. The use of lime for soft soils stabilization is not new and
was studied by many researches (Broms and Boman, 1979; Bell,
1988; Locat et al., 1996; Rao and Rajasekaran, 1996; Matthew
and Rao, 1997; Rajasekaran and Rao, 1998; Zhou et al., 2003;
Rajasekaran and Rao, 2002; Rajasekaran, 2005).
A series of large scale laboratory model tests was carried out in
− 894 −
a circular unit cell tank with the CL-WS column at the center and
soft clay as the surrounding soil. The influence of parameters
such as the slenderness ratio (the ratio of the column Length to
its Diameter (L/D)), the variation of the column diameter (D), the
area ratio (the ratio of the area of the column (AS) to the total area
within the unit cell (A)), and the shear strength of the
surrounding soil have been studied. The results of tests were
compared with that of numerical modelling using PLAXIS
software. In order to compare the results of laboratory and
numerical modelling with the real behavior of these columns,
some tests have been carried out in the field using plate load test
apparatus.
2. Theoretical Considerations
For the development of a precise laboratory scale model, all
practical dimensions were reduced by an appropriate scale factor.
It was considered that a well designed testing program would
allow observation of key aspects of improved ground with CL-
WS columns. Special attention was paid to keep the key ratios
including the ratio of column length to column diameter, column
diameter to diameter of the entire specimen, column diameter to
aggregate size of used soil for constructing of the column,
identical in the laboratory modelling and actual field condition.
In the practical applications, the diameter of the stone columns
was chosen based on the design considerations and construction
method. This value generally varies between 60 to 100 cm for
conventional stone columns and 25 to 60 cm for rigid stone
columns. The other dimensional parameters such as length of the
column and the area ratio have been presented according to the
diameter of column. The stone columns are formed with typical
aggregates size of 2-75 mm (IS, 2003). Hence, the ratio of the
column diameter to the maximum particle size will be in the
range of 8-12. These remarks have been considered in the
experimental program.
Unit cell idealization was used to simplify the design of the
apparatus needed to assess the behavior of an interior column in
a large group of columns. For an infinitely large group of
columns subjected to a uniform loading applied over the area, the
behavior of each interior column may be simplified to a single
column installed at the center of a cylinder of soil representing
the column’s influence zone. Due to the symmetry of load and
geometry, lateral deformation can not occur across the boundaries
of the unit cell, and the shear stresses on the outside boundaries
of the unit cell must be zero (Barksdale and Bachus, 1983). The
unit cell can be physically modeled as a cylindrical shape
container having a smooth, rigid exterior wall symmetrically
located the column. Pribe proposed unit cell concept for
estimating the settlement of foundation resting on the infinite
grid of stone columns (Priebe, 1995). This concept has also been
used by many researchers (Alamgir et al., 1996; Ambily and
Gandhi, 2007; Gniel and Bouazza, 2009; Shivashankar et al.,
2011).
KSCE Journal of Civil Engineering
 Behavior of Compacted Lime – (Well-graded) Soil Columns: Large Scale Tests and Numerical Modelling
3. Experimental Investigation
3.1 Test Setup
Two experimental setups were conducted in this investigation.
In the first setup, primary tests were carried out to determine the
properties of used materials and to specify the construction
method of the CL-WS columns. In the second setup, main tests
were conducted on composite specimens to investigate the
performance of CL-WS columns on the behavior of soft soils.
The experiments were carried out on columns with the different
diameters and different slenderness ratios. The columns were
surrounded by soft soil with various shear strength in cylindrical
tanks of 300 to 1200 mm height, and a diameter varying from
160 to 470 mm to represent the required unit cell area of soft soil
around each column assuming triangular pattern of installation.
Therefore the tests were conducted with four different area
ratios: 5, 10, 15 and 20% which corresponds to spacing of 4.3D,
3D, 2.5D and 2D, respectively (where, D is the diameter of the
column). Experiments were carried out on the both floating and
end bearing columns. In the specimens containing floating
columns, a layer of soft soil with thickness of 2D, was used
beneath the column. The prepared specimens were kept in plastic
covers and tested after curing time of 60 days. In the field, the
entire of the CL-WS column treated ground will be subjected to
loading from the superstructure. The same was simulated in the
laboratory by loading the entire area of the specimen to study the
stiffness of improved ground. A 50 mm thick sand layer was
placed below loading plate to serve as a blanket. Two dial gauges
were fixed at 180º angles to each other for measuring the
settlement of specimens during the application of the load. A
typical test arrangement and overview of experimental testing
program are shown in Figs. 1 and 2, respectively.
The load intensity – settlement behavior of column- treated
Fig. 1. Laboratory Test Setups: (a) Floating Column, (b) End Bear-
ing Column
Vol. 19, No. 4 / May 2015
Fig. 2. Overview of Experimental Testing Program
soil has been studied by applying vertical load with the help of
loading frame through a proving ring at the constant strain rate of
0.7 mm/min. The load was applied over the entire area of the
composite specimen using a steel plate of 20 mm thickness and
diameter of 10 mm less than the inner diameter of the test tank. A
sand layer with thickness of 50 mm was placed beneath the
loading plate and the load was monitored at the equal interval of
settlement up to 15 mm.
In order to assess the actual behavior of CL-WS columns, two
groups of columns with the diameter of 100 and 150 mm, in
accordance with unit cell concept and assuming triangular
pattern, were made in the field and loaded using plate load
apparatus.
3.2 Preparation of Laboratory Specimens
An identical technique was used to prepare all of the
specimens. Before filling the tank with clay, polythene sheet was
laid on the internal walls of it to avoid any friction between clay
and walls of tank. To prepare each of test beds, the required clay
soil was air dried and checked for initial moisture content. The
additional water quantity required to achieve desired moisture
content was added and thoroughly mixed to form a uniform
paste. Clay was filled in the tank in layers with measured quantity
by weight. Each layer was subjected to uniform compaction with
a tamper to achieve 50 mm height and corresponding unit
weight. The CL-WS column was constructed by replacement
method. A thin open-ended steel pipe of various inner diameters
and wall thickness of 2mm was employed to construct the
column. After the bottom layer was prepared for a depth of twice
the diameter of the column, the steel pipe was placed at the
center of the soil bed, and construction of clay soil and column
were carried out simultaneously. The outer surface of the pipe
was lubricated by applying a thin layer of grease for easy
withdrawal of pipe without any significant disturbance of the
surrounding soil.
Construction of the columns was carried out in two stages;
first, specified amount of used materials with optimum moisture
content were mixed completely and then poured into the steel
− 895 −
 Mohammad Reza Malekpoor and Gholamreza Poorebrahim

pipe. Subsequently, they were compacted using 4.5 kg tamper to Table 2. Chemical Analysis of the Hydrated Lime
attain maximum density. The pipe was then raised in stages Component Oxides Composition (%)
ensuring a minimal of 50 mm penetration below the top level of Calcium Oxide (CaO) 73.70
the placed soil. The procedure was repeated until the construction Magnesium Oxide (MgO) 1.619
of the column and surrounding soil was completed to the full Silica (SiO2) 1.15
height. Iron Oxide (Fe2O3) 0.24
Alumina (Al2O3) 0.11
3.3 Properties of Materials Sulphur Trioxide (SO3) 0.015
The clay used is of CL classification, excavated from a Manganese (Mn) 0.005
construction site in Kerman-Iran. The properties of this soil are Chloride as NaCl 0.011
given in Table 1. The clay is one the problematic soils which Loss on ignition 23.15
exhibit high compressibility, reduced strength and high plasticity.
The method proposed in current study is aimed at improving the
engineering properties of this soil. The soil used for construction shown in Fig. 3. Normal hydrated lime with the chemical
of column is of SW-SC with clay content of approximately 11 properties given in the Table 2 was used to construct CL-WS
percent by weight. The grain size distribution of this soil is columns.

4. Finite Element Analysis

Table 1. Properties of Used Clay Soil
Parameter Value In this paper the emphasis was on the physical modelling.
Specific Gravity 2.72 However, in order to provide some quantitative support for the
Water content (%) 12.60 results, numerical analysis has been also performed. Indeed, the
Liquid limit (%) 42.00 numerical modelling has been used to confirm the load carrying
Plastic limit (%) 21.50 capacity as well as the load intensity-settlement behavior of
Shear strength (kPa) 17.00 composite specimens and to determine stress concentration ratio.
Compression index (CC) 0.17 The finite element analysis was carried out using PLAXIS
Cation exchange capacity software. The cylindrical unit cell idealization was adopted with
36
( m equiv./100 g of the soil)
Sulphates (%) 0.23
regard to radial symmetry around a vertical axis through the
Chlorides (%) 0.96 center of the CL-WS column. The Mohr-Coulomb material
pH 7.11 model was used for modelling the soft clay soil and linear elastic
model was used to describe the behavior of the column material.
The physical and mechanical properties of the used materials
were shown in Table 3. Modulus of elasticity of the clay was
determined by conducting consolidation test and Poisson’s ratio
was obtained from triaxial test. The vertical and horizontal
displacements were restricted at the base of the tank, and horizontal
displacements were restrained along the circumference of the
tank. The finite element mesh was generated using 15-node
triangular element for all the components. This mesh gives finer
distribution of nodes and thus much more accurate results than a
similar mesh composed of 6-node elements (Brinkgreve, 2002).
Trial calculations have shown that the global coarseness variation
of the elements has a very little effect on the results (approximately
0.4%). Thus, in the present research, the analyses were performed
Fig. 3. Grain Size Distribution for Well Graded Clayey Sand using medium mesh elements. The interaction between CL-WS

Table 3. Physical and Mechanical Properties of Materials used

Material Material Model W (%) C(kPa) Φ (deg) E (kPa) µ γd (kN/m3)
18.50 10 6.50 12000 0.42 13.70
13.00 17 8.00 12000 0.41 14.05
Clay Mohr-Coulomb’s
11.50 30 9.50 12000 0.39 14.30
9.50 40 10.50 12000 0.37 14.70
CL-WS Linear elastic - - - 200000 0.21 21.20

− 896 − KSCE Journal of Civil Engineering

 Behavior of Compacted Lime – (Well-graded) Soil Columns: Large Scale Tests and Numerical Modelling
column and the surrounding soil at their interface is totally
different from that of a conventional stone column and soft soil,
due to the migration of the lime or calcium ions into the soil
surrounding the column. In the PLAXIS software an interface
element is introduced by the interface strength factor (Rint). This
factor links the strength parameter of the interface element to the
strength parameters of the soil (Brinkgreve, 2002). In order to
measure Rint, cubic specimens comprised of CL-WS and soil was
constructed in the laboratory and then was subjected to the direct
shear test. Based on the results, Rint was considered 0.85.
5. Results and Discussion
5.1 Primary Laboratory Tests
To compare the strength behavior of the CL-WS specimens,
which were prepared with maximum dry density and optimum
moisture content, with the strength of the Lime-WS specimens,
which were made with 34% moisture content in slurry form,
specimens with dimensions of 100 mm in diameter and 200 mm
in height were constructed using different lime contents.
Unconfined Compression Strength (UCS) tests were carried out
on specimens after curing time of 60 days according to ASTM
standard (ASTM D2166, 2000), and the variation of the UCS
with lime content has been presented in Fig. 4. It has been
observed that the compacting of the Lime-WS specimens
increases the UCS noticeably. For instance, the improvement in
UCS of CL-WS specimens, as compared to Lime-WS specimens,
for 15 % lime content, is 134 %. Based on results drawn from the
above mentioned tests it can be concluded that in spite of Lime-
WS specimens, in CL-WS specimens with the lime content more
than 15% the increment in the strength is not noticeable and
hence composite specimens were constructed with columns
containing 15% lime content. It should be noted that the
percentage of the lime content in CL-WS specimens also
depends on the clay content of the soil used for constructing the
specimens. Increasing the clay content also increases the strength
of the specimens due to chemical reaction between lime and
Fig. 4. Variation of UCS of Lime- WS and CL-WS Specimens with
Lime Content
Vol. 19, No. 4 / May 2015
Fig. 5. Modified Compaction Test Results for CL-WS Specimens
Containing 15% Lime and 18% Clay Content
silica in clay. According to the previous study, the grain size
distribution of SW-SC soil, which is used for construction of
columns, is modified so that the clay content reaches 18%
(Malekpoor and Toufigh, 2010).
For determining the optimum moisture content, and also
maximum dry density used for construction of above mentioned
CL-WS specimens, compaction tests using modified effort were
carried out according ASTM standard (ASTM D1557, 2000).
Since the composite specimens are prepared with columns
containing 15% of lime content and 18% of clay content, the
results of compaction test for this case is given in Fig. 5.
5.2 Main Tests
5.2.1 Laboratory Tests and FEM Analysis
As mentioned earlier, for simulating the actual behavior of CL-
WS column treated ground in the field, load was applied over the
entire area of composite specimens for evaluating the improvement
of the stiffness of the treated ground. When the entire area is
loaded, failure does not take place even for a settlement as high
as 15 mm, due to the confining effect of the unit cell.
Figure 6 shows the typical relationship between load intensity
and settlement for improved and unimproved ground for different
slenderness ratios when the area ratio, diameter of the column
and cohesion of the surrounding soil equal to 10%, 100 mm and
17 kPa, respectively. In the specimens containing floating
columns, a layer of soft soil with thickness of 2D (200 mm) was
used beneath the column. In the case of end bearing columns, the
column was constructed on the rigid base of the tank.
It has been observed that CL-WS columns with different
slenderness ratios increase the load carrying capacity of the soft
soils. When the slenderness ratio is small (specimens containing
short columns), the load carrying capacity is high, and as the
slenderness ratio increases (specimens containing slender columns),
load carrying capacity of specimens decreases. However, as
compared to untreated clay bed, an improvement of 37% in load
intensity has been observed, when the clay bed is improved
using floating CL-WS column with slenderness ratio of 8. This
− 897 −
 Mohammad Reza Malekpoor and Gholamreza Poorebrahim

Fig. 6. Effect of the Slenderness Ratio on the Load Intensity – Set-

tlement of Composite Specimens: (a) Floating Columns, (b) Fig. 7. Effect of the Diameter of the Column on the Load Intensity
End Bearing Columns – Settlement of Composite Specimens: (a) Floating Col-
umns, (b) End bearing Columns

value is 167% for end bearing column with the same slenderness
ratio. The comparison of Figs. 6(a) and 6(b) shows that the
variation of the slenderness ratio is more effective on the
behavior of end bearing columns rather than floating columns. In
the end bearing columns, loads are transmitted to the rigid base
and bulging failure takes place when the slenderness ratio
increases.
The results of laboratory model tests and finite element analysis
are quantitatively compared regarding the load intensity
corresponding to the 15 mm settlement for different slenderness
ratios. Comparison of the results reveals that the differences
between them vary from 0.5 to 7%. This dissimilarity may arise
due to common errors in the laboratory testing as well as the Fig. 8. Variation of the Load Intensity (corresponding to 15 mm
potential shortcoming of the numerical modelling such as: settlement) with the Diameter of the Column (D)
uncertainties in the constitutive models used for the materials
especially CL-WS column, uncertainties in the evaluation of the
materials properties and the complicated behavior of the CL-WS the size of models on the behavior of CL-WS columns by
columns and the soft clay in their interface. varying the column diameter from 50 mm to 150 mm. Fig. 7
It should be noted that, a conventional stone columns which indicates a relationship between load intensity and the settlement
has a length greater than its critical length (about 4 to 5 times of for columns with area ratio equal to 10% and L/D = 6. Fig. 8
the diameter of the column) fails by bulging irrespective of shows the variation of load carrying capacity (corresponding to
whether the condition of column is end bearing or floating (IS, 15 mm settlement) with variation of the column diameter. As
2003). predicted, the load carrying capacity decreases with increasing
The experimental program is extended to evaluate the effect of the diameter, and the variations of load carrying capacity of

− 898 − KSCE Journal of Civil Engineering

 Behavior of Compacted Lime – (Well-graded) Soil Columns: Large Scale Tests and Numerical Modelling
Fig. 9. Effect of the Area Ratio (Ar) on the Load Intensity-Settle-
ment of Composite Specimens: (a) Floating Columns, (b)
End Bearing Columns
specimens reduces by increasing the size of the model. For
instant, the change of the load intensity (corresponding to 15 mm
settlement) for floating columns of 50 and 75 mm diameter is
32% and for columns with diameter of 125 mm and 150 mm is
11%. Regarding the load carrying capacity as well as load
intensity-settlement relationship, both experimental and numerical
modellings show a comparable behavior.
Since rigid stone columns are used with minimum diameter of
250 mm in practice, the results of large scale tests can be applied
to extrapolate and predict the full size behavior of these columns
with tolerable error.
Figure 9 illustrates the variation of load intensity versus
settlement for different area ratios, L/D = 6 and D = 100 mm.
Comparing the results shows that CL-WS columns increase the
stiffness of soft soils, even for area ratio of 5% (which
corresponds to spacing of 4.2D) and application of end bearing
columns further improves the bearing capacity of soft soils. As
compared with untreated specimens, an improvement of 45% in
load intensity has been observed when the clay bed is improved
with floating CL-WS column and for an area ratio of 5%. This
value is 145% for end bearing column and for the same area
ratio. As compared with respective untreated ground, the load
intensity has been increased by 59, 90 and 118%, when the soil is
Vol. 19, No. 4 / May 2015
Fig. 10. Variation of the Load Intensity (corresponding to 15 mm
settlement) with the Cohesion of the Surrounding Soil: (a)
Floating Columns, (b) End Bearing Columns
improved by floating column with area ratios of 10, 15 and 20%.
These ratios are 238, 351 and 396% for end bearing columns
under the same conditions. It can be observed that the higher area
ratio results in higher load intensity, However, when the area
ratio exceeds 15%, the increment rate of load intensity is
decreased. Comparison of the results of physical model tests and
numerical analysis predictions reveal that there is a close
agreement between them.
It should be noted that, for a significant improvement in
bearing capacity using conventional stone columns, an area ratio
of 25% was suggested (Wood et al., 2000). For important projects,
it is desired to carry out field trials to determine the most
optimum spacing of columns with regard to the required bearing
capacity of the soil, permissible settlement of the foundation,
loading pattern, column dimensions, etc. Accordingly, rational
decisions can be taken to tailor design of different type of stone
columns installation to achieve maximum performance at
optimum cost (Babu et al., 2012).
Figure 10 indicates the variations of load intensity (corresponding
to 15 mm settlement) with cohesion of the surrounding soil for
the various area ratios, L/D = 6 and D= 100 mm. Fig. 10(a)
shows the consistency of soft soil has a slight effect on the
− 899 −
 Mohammad Reza Malekpoor and Gholamreza Poorebrahim
behavior of the floating columns, and the load intensity increases
by increasing the shear strength of the surrounding soil. This is
due to the increase of the strength of the bottom layer of the
column and consequently increases the base resistance of the
column. As it can be seen in Fig. 10(b), the load carrying capacity
of end bearing columns is found to be almost independent of the
shear strength of the soil.
As mentioned before, the shear strength of the surrounding soil
has a significant influence on the load carrying capacity as well
as failure mechanism of conventional stone columns.
To quantitatively evaluate the improvement in stiffness of the
ground, due to installation of CL-WS columns, stiffness
improvement factor (β), defined as the ratio of load intensity of
the improved ground to that of untreated ground for the same
settlement (15 mm), is calculated and shown in Fig. 11. As
illustrated, the stiffness improvement factor is found to be
approximately independent of the shear strength of the surrounding
soil. For specimens which contained floating columns and area
ratio of 5%, β increases around 6% as the cohesion of the soil
varies from 10 to 40 kPa. A similar trend can be observed for
other area ratios (7, 8 and 6% increment corresponding to area
ratios of 10, 15 and 20%, respectively). On the other hand, for
Fig. 11. Variation of Stiffness Improvement Factor (β) with the
Cohesion of the Surrounding Soil: (a) Floating Columns,
(b) End Bearing Columns
− 900 −
end bearing columns, 6 to 16% reduction in β was seen while the
cohesion of the surrounding soil increases from 10 to 40 kPa and
the area ratio varies from 5 to 20%. Hence, it can be concluded
that the end bearing columns are desirably applicable to extremely
soft soils.
In the composite unit cell, the applied load is divided between
the column and surrounding soil according to their relative
stiffness values. Stress Concentration Ratio (SCR) is defined as
the ratio of the average stress tolerated by the column to average
stress experienced by the soil for the same applied load. Fig. 12
shows the stress concentration ratios obtained from the
numerical analyses versus cohesion of the surrounding soil for
different area ratios.
As expected, SCR is greater in the end bearing columns than
floating columns. The results demonstrated that in the floating
columns, SCR increases with the increase of the area ratio, and
the higher area ratio, the better the efficiency improvement. The
similar behavior has been observed in the developed model for
the soft soil improvement using conventional stone columns
(Ambily and Gandhi, 2007). The increase of the soil’ cohesion
further improves the tip resistance of the column rather than the
surrounding soil. Thus, SCR becomes greater as the shear
Fig. 12. Variation of Stress Concentration Ratio (n) with the Cohe-
sion of the Surrounding Soil: (a) Floating Columns, (b) End
Bearing Columns
KSCE Journal of Civil Engineering
 Behavior of Compacted Lime – (Well-graded) Soil Columns: Large Scale Tests and Numerical Modelling
Fig. 13.Comparison between the Results of Laboratory Model
Tests and Field Tests
strength of the soil is increased.
In the case of end bearing columns, the increase in the
cohesion of the surrounding soil causes the stress concentration
on the soil to be increased and therefore decreasing SCR. In
these columns, as the area ratio increases, SCR is slightly
decreased. By increasing the area ratio, the surrounding soil is
further confined between the column and rigid wall of the tank
and as a result the horizontal displacements of the soil are
decreased which caused the stiffness of the soil increased and
thus, the SCR is decreased.
5.2.2 Field Tests
As mentioned before, in order to compare the results of
laboratory and numerical modelling with the behavior of these
columns in the filed, two groups of seven CL-WS columns with
the diameter of 100 and 150 mm were prepared in the field.
Columns were arranged in a triangular pattern and the slenderness
ratio and area ratio were fixed to 6 and 10%, respectively. Tests
were conducted using plate load test apparatus and the results are
shown in Fig. 13. The results are nearly comparable and
informative in two ways. Firstly, it was observed that for columns
with the same dimensions, the load intensity corresponding to
15 mm settlement of field test is approximately 10 to14% less
than that of laboratory test. The key reason is that in the
laboratory, the specimen was prepared using unit cell concept in
the circular tank with rigid walls and the load was applied over
its entire area and hence the deformations of the surrounding soil
are totally restricted. But in the field tests, load was applied only
on the interior unit cell and thus the soil deformations in the
boundaries of the unit cells are not restricted. From this point of
view, the laboratory modelling represents the actual field
condition, where the entire area of improved ground is loaded by
the superstructures in the practical applications. The other reason
of the difference in the stiffness of the field and laboratory
models is that in the laboratory all the model properties and
loading conditions can be controlled and checked accurately
rather than in the field. The second feature of the results is the
decreasing of the load carrying capacity of the model tests with
Vol. 19, No. 4 / May 2015
increasing the scale which was discussed earlier.
6. Conclusions
Experimental investigation and numerical modelling were
carried out to study of the factors affecting the behavior of CL-
WS columns in soft soils. Based on the presented results and
discussion, the following conclusions may be drawn:
1. The results reveal that CL-WS columns improve the load
carrying capacity and decrease the settlement of the stabi-
lized soils. In contrast to the conventional stone columns,
CL-WS columns can be used as an appropriate ground
improvement technique in very soft soils.
2. It is demonstrated that the load intensity decreases by
increasing the slenderness ratio. The variation of the slender-
ness ratio is more effective on the behavior of the end bear-
ing columns rather than floating columns.
3. Meaningful influence of the model size has been observed in
the both experimental and numerical models, in such a way
that the load carrying capacity decreases by increasing the
size of models. However, for specimens containing columns
with the diameter greater than 100 mm and constant slender-
ness ratio, the stiffness variations become negligible, i.e. the
stiffness tends to converge to a fixed value and hence the
results can be used to extrapolate and predict the full size
behavior of these columns with tolerable error.
4. The performance of CL–WS column is remarkably enhanced
by increasing the area ratio. However, when the area ratio
exceeds 15%, the rate of increment of the load carrying
capacity is negligible.
5. The influence of shear strength of the surrounding soil has
been also evaluated on the behavior of CL-WS columns.
Results show that the consistency of soft soil has a slight
effect on the behavior of the floating CL-WS columns, and
the load intensity increases by increasing the shear strength
of the surrounding soil. In the end bearing columns, the load
carrying capacity is found to be almost independent of the
shear strength of the soil.
6. A detailed comparison between the results of the experimen-
tal study and numerical modelling for all cases was pre-
sented which indicated a close agreement for the two
methods.
7. The comparison of the result of the laboratory and the field
tests shows that the load carrying capacity of columns pre-
pared in the laboratory is 10 to14% larger than of the col-
umns prepared in the field with the same dimensions.
References
Alamgir, M., Miura, N., Poorooshasb, H. B., and Madhav, M. R. (1996).
“Deformation analysis of soft ground reinforced by columnar
inclusions.” Comput. Geotech., Vol. 18, No. 4, pp. 267-290.
Ambily, A. P. and Gandhi, S. R. (2007). “Behavior of stone columns
based on experimental and FEM analysis.” J. Geotech. Geoenviron.
− 901 −
 Mohammad Reza Malekpoor and Gholamreza Poorebrahim
Engg., ASCE, Vol. 133, No. 4, pp. 405-415, DOI: 10.1061/(ASCE)
1090-0241(2007) 133:4(405).
ASIRI (2012). Amelioration des sols par inclusions rigides, Presses des
Ponts et Chaussees (in French).
ASTM D1557 (2000). Test method for laboratory compaction
characteristics of soil using modified effort, West Conshohocken, PA.
ASTM D2166 (2000). “Standard test method for unconfined compressive
strength of cohesive soil, West Conshohocken, PA.
Babu, M. R. D, Nayak, S., and Shivashankar, R. (2012). “A critical
review of construction, analysis and behavior of stone columns.”
Geotech. Geol. Eng., Vol. 31, No. 1, pp. 1-22, DOI: 10.1007/
s10706-012-9555-9.
Barksdale, R. D. and Bachus, R. C. (1983). Design and construction of
stone columns, Federal Highway Administration, RD-83/026.
Bell, F. G. (1988). “Stabilization and treatment of clay soils with lime,
Part 1-Basic Principles.” Ground Engineering, Vol. 21, No. 1, pp.
10-15.
Brinkgreve, R. B. J. (2002). Plaxis user’s manual-version 8.2, Delft
University of Technology and PLAXIS b.v., The Netherlands.
Broms, B. B. and Boman, P. (1979). “Lime columns - A new foundation
method.” Journal of the Geotechnical Engineering Division, ASCE,
Vol. 105, No. 4, pp. 539-556.
Castro, J. and Karstunen, M. (2010). “Numerical simulations of stone
column installations.” Can. Geotech. J., Vol. 47, No. 10, pp. 1127-
1138, DOI: 10.1139/T10-019.
Cimentada, A., Costa, A. D., Canizal, J., and Sagaseta, C. (2011).
“Laboratory study on radial consolidation and deformation in clay
reinforced with stone columns.” Can. Geotech. J., Vol. 48, No. 1, pp.
36-52, DOI: 10.1139/T10-043.
Combarieu, O. (1974), “Effet d’accrochage et methode d’evaluation du
frottement negatif.” Bulletin de Liaison des Laboratoires des Ponts
et Chaussees nº71(mai-juin), pp. 93-107 (in French).
Combarieu, O. (1988), “Amelioration des sols par inclusions rigides
verticals-application a l’edification de remblais sur sols mediocres.”
Revue Francaise de Geotechnique, No. 44, pp. 57-59 (in French).
Deb, K., Samadhiya, N. K., and Namdeo, J. B. (2011).” Laboratory
model studies on unreinforced and geogrid-reinforced sand bed over
stone column-improved soft clay.” Geotextiles and Geomembranes,
Vol. 29, No. 2, pp.190-196, DOI: 10.1016/j.geotexmem.2010.06.004.
DiMaggio, J. A. (1978). Stone columns for highway construction, Technical
Report No. FHWA-DP-46-1, U.S. Department of Transport, Federal
Highway Administration.
Elias, V., Welsh, J., Warren, J., Lukas, R., Collin, J. G., and Berg, R. R.
(2006). Ground improvement methods, Volumes I and II, FHWA
NHI-06-019 and FHWA NHI-06-020, US Dept. of Transportation,
Federal Highway Administration.
Engelhardt, K., Flynn, W. A., and Bayuk, A. A. (1974). “Vibro-replacement
method to strengthen cohesive soils in situ.” Proc. of ASCE National
Struc. Eng. Meeting, Cincinnati, OH, USA.
Fattah, M. Y., Shlash, K. T., and Al-Waily, M. J. M. (2011). “Stress
concentration ratio of model stone columns in soft clays.” Geotechnical
Testing Journal, Vol. 34, No. 1, pp. 50-60, DOI: 10.1520/GTJ103060.
FGSV (Forschungsgesellschaft f. Straβenund Verkehrswesen) (1979).
Merkblatt für die untergrundverbesserung durch Tiefenrüttler,
FGSV, Klön, (in German).
Gniel, J. and Bouazza, A. (2009). “Improvement of soft soils using
geogrid encased stone columns.” Geotextiles and Geomembranes,
Vol. 27, No. 3, pp. 167-175, DOI: 10.1016/j.geotexmem.2008.11.001.
Greenwood, D. A. (1970). “Mechanical improvement of soils below
ground surface.” Proc. of Ground Engineering Conference,
− 902 −
Institution of Civil Engineers, London, England, pp. 11-22.
Han, J. and Ye, S. L. (1992). Settlement analysis of buildings on the soft
clays stabilized by stone columns, Int. Conference on Soil
Improvement and Pile Foundation, Nanjing, China, pp. 446-451.
Han, J. and Ye, S. L. (2001). “Simplified method for consolidation rate
of stone column reinforced foundation.” J. Geotech. Geoenviron.
Engg., Vol. 127, No. 7, pp.597-603, DOI: 10.1061/(ASCE) 1090-
0241(2001) 127:7(597).
Hughes, J. M. O. and Withers, N. J. (1974). “Reinforcing of soft
cohesive soils with stone columns.” Ground Engineering, Vol. 7, No. 3,
pp. 31-44.
Hughes, J. M. O., Withers, N. J., and Greenwood, D. A. (1975). “A field
trial of the reinforcing effect of a stone column in soil.” Geotechnique,
Vol. 25, No. 1, pp. 31-44.
IS: 15284 (2003). Indian standard code of practice for design and
construction for ground improvement-guideline, Part: 1, stone
columns, India.
Juran, I. and Riccobono, O. (1991). “Reinforcing soft soils with artificially
cemented compacted-sand columns.” J. Geotech. Geoenviron. Engg.,
ASCE, Vol. 117, No. 7. pp. 1042-1060, DOI: 10.1061/(ASCE)
0733- 9410(1991) 117:7(1042).
Locat, J., Tremblay, H., and Leroueil, S. (1996). “Mechanical and
hydraulic behavior of a soft inorganic clay treated with lime.” Can.
Geotech. J., Vol. 33, No. 4, pp. 654-669, DOI: 10.1139/T96-090-
311.
Malekpoor, M. R. and Toufigh, M. M. (2010). “Laboratory study of soft
soil improvement using lime mortar-(well graded) soil columns.”
Geotechnical Testing Journal, Vol. 33, No. 3, pp. 1-11, DOI:
10.1520/GTJ102291.
Matthew, P. K. and Rao, S. N. (1997). “Effect of lime on cation
exchange capacity of marine clay.” J. Geotech. Geoenviron. Engg.,
ASCE, Vol. 123, No. 2, pp. 183-185.
Mckelvey, D., Sivakumar, V., Bell, A., and Graham, J. (2004). “Modeling
vibrated stone columns on soft clay.” Geotech. Eng., Vol. 157, No. 3,
pp. 137-149, DOI: 10.1680/geng.2004.157.3.137.
McKenna, J. M., Eyre, W. A., and Wolstenholme, D. R. (1975).
“Performance of an embankment supported by stone columns in soft
ground.” Geotechnique, Vol. 25, No. 1, pp. 51-59.
Mitchell, J. K. and Huber, T. R. (1985). “Performance of a stone column
foundation.” J. Geotech. Engg., ASCE, Vol. 111, No. 2, pp. 205-223.
Mitra, S. and Chattopadhyay, B. C. (1999). “Stone columns and design
limitations.” Proc. of Indian Geotechnical Conference, Calcutta,
India, pp. 201-205.
Murugesan, S. and Rajagopal, K. (2010). “Studies on the behavior of
single and group of geosynthetic encased stone columns.” J. Geotech.
Geoenviron. Engg., ASCE, Vol. 136, No. 1, pp. 129-139, DOI:
10.1016/j.geotexmem.2006.05.001.
Priebe, H. J. (1995). “The design of vibro replacement.” Ground Eng.,
Vol. 28, No. 12, pp. 31-37.
Rajasekaran, G. (2005). “Sulphate attack and ettringite formation in the
lime and cement stabilized marine clays.” Ocean Engineering, Vol.
32, Nos. 8-9, pp.1133-1159.
Rajasekaran, G., and Rao, S. N. (1998). “Particle size analysis of lime
treated marine clays.” Geotechnical Testing Journal, Vol. 21, No. 2,
pp. 109-119, DOI: 10.1520/GTJ10749J.
Rajasekaran, G. and Rao, S. N. (2002). “Compressibility behavior of
lime-treated marine clay.” Ocean Engineering, Vol. 29, No. 5, pp.
545-559.
Rao, S. N. and Rajasekaran, G. (1996). “Reaction products formed in
lime-stabilized marine clays.” J. Geotech. Engg., ASCE, Vol. 122, No.
KSCE Journal of Civil Engineering
 Behavior of Compacted Lime – (Well-graded) Soil Columns: Large Scale Tests and Numerical Modelling

5, pp. 329-336, DOI: 10.1061/(ASCE) 0733-9410(1996) 122:5(329).
Rathmayer, H. (1975). “Piled embankment supported by single pile
caps.” Proc. of Istanbul Conference on Soil Mechanics and Foundation,
Istanbul, Turkey, Vol. 1, pp. 283-290.
Samadhiya, N. K, Maheswari, P., Basu, P., and Kumar, M. B. (2008).
“Load-settlement characteristics of granular piles with randomly
mixed fibers.” Indian Geotech. J., Vol. 38, No. 3, pp. 345-354.
Shivashankar, R., Babu, M. R. D., Nayak, S., and Manjunath, R. (2010).
“Stone columns with vertical circumferential nails-Laboratory model
study.” Geotech. Geol. Eng., Vol. 28, No. 4, pp. 695-706, DOI:
10.1007/s10706-010-9329-1.
Shivashankar, R., Babu, M. R. D., Nayak, S., and Rajathkumar, V. (2011).
“Experimental studies on behavior of stone columns in layered
soils.” Geotech. Geol. Eng., Vol. 29, pp. 749-757, DOI: 10.107/s10706-
011-9414-0.
Simon, B. (2012). “General report, session 5: Rigid inclusions and stone
columns.” Internal Symposium on Ground Improvement, Brussels,
Belgium.
Wood, D. M., Hu, W., and Nash, D. F. T. (2000). “Group effect in stone
column foundations: Model tests.” Geotechnique, Vol. 50, No. 6, pp.
689-698.
Zhou, C., Yin, J. H., and Ming, J. P. (2002). “Bearing capacity and
settlement of weak fly ash ground improvement using limefly ash or
stone columns.” Can. Geotech. J., Vol. 39, pp. 585-596, DOI:
10.1139/T02-011.

Vol. 19, No. 4 / May 2015 − 903 −