IMPROCE 2009

GROUND IMPROVEMENT TECHNIQUES – AN OVERVIEW

Dr.S.R. Gandhi *

1. INTRODUCTION

Though the term Ground Improvement has been familiar to Civil Engineers, the
design approach is still empirical, mostly based on past experience. Well defined
design procedure, constructions procedure and codal provisions are yet to be developed.
In the absence of above, execution of the same is difficult and sometimes lead to
contract disputes.

On the other hand, adequacy of Ground Improvement for supporting even large
structures has been proved beyond doubts. Application of Ground Improvement is not
only cheaper but reduces the construction time significantly. In last two decades several
major projects have been successfully built in the country adopting some of the ground
improvement techniques. For each project, field trials and ground monitoring has been
carried out. This has generated large amount of field data and built the confidence of
geotechnical designer for adopting ground improvement techniques.

In this lecture note, aspects such as occurrence of weak deposits, typical

prosperities weak deposits, methods of ground improvement, etc. have been discussed.
In addition, three case studies are described were one of the ground improvement
techniques has been adopted for a large project.

2. OCCURRENCE OF WEAK DEPOSITS

Weak deposits can be defined as deposit, which has one or more of the
following.

i. Low shear strength

ii. High compressibility
iii. Susceptibility to liquefaction

Such deposits are very common along the coastal region. Most of the marine
deposits are of recent origin and have not undergone much consolidation. As a result,
they have low shear strength or high compressibility. Typical places with such soft clay
deposits are Kandala, Nhava Sheva, Cochin, Ennore, Kakinada, Visakhapatnam, Haldia,
Calcutta, etc. Even some of the land deposits, particularly alluvial deposits along the
river belt have loose silt/sand to a large depth. Typical example is large area of gangetic
belt in Uttar Pradesh. Even man made deposits such as mine back-fill or land reclaimed
by filling can have inadequate strength properties requiring ground improvement.

Major problems associated with such deposits are:

i. Low bearing capacity

ii. Higher settlement (both total and differential)
iii. High seepage losses
* Professor, Department of Civil Engineering, IIT Madras, Chennai - 36 1
 IMPROCE 2009

iv. Liquefaction during earthquake

v. Instability of foundation excavations
vi. Higher earth pressures on retaining structures

In early days, areas having such deposit were avoided for construction. But with
scarcity of land in urban areas, we do not have choice and structures have to be built on
weak deposits. Pile foundation is of course a possible approach as it by pass the weak
deposit and transfer the load on next competent layer. Where thickness of deposit is
very large, pile foundation is uneconomical and time consuming.

Number of industrial plants in last one or two decades have successfully adopted
one of the ground improvement techniques. This has resulted in either elimination of
RCC piles or considerable reduction in number of RCC piles. Wherever ground
improvement technique has been adopted, it is observed that there is reduction in
construction cost and also the construction is faster. Most of the plants in India are
constructed based on Lumpsum Turn Key Project (LSTK). The work is awarded to the
bidder who provides the lowest prices but at the same time meeting all technical
requirements. The construction agencies always look forward to improve the ground
wherever possible to minimise the total cost.

Though ground improvement has been successfully adopted for several plants,
there is no well-defined codes or design procedures to select suitable type and for design
of ground improvement and design for the same. In most of the cases, the design is
based on field trials carried out followed by post improvement test. Theoretical analysis
for such improvement technique is rather limited.

Numbers of techniques have been developed to improve the weak deposits such
that required structure can be placed without pile foundation. Following section
describes typical properties of weak deposits, methods available for ground
improvement and finally few case studies on improvement carried out with different
techniques.

3. PROPERTIES OF WEAK DEPOSITS

Weak deposits can be broadly classified into two categories

(i) Cohesive deposit comprising of soft clay

(ii) Cohesionless deposit comprising of loose sand.

The type of problems and method of treatment are different for each category.
Typical properties are described below:

3.1 Soft Clay Deposits :

Marine deposits along coastal region or lacustrine deposits often remain

unconsolidated for a long time. Such deposits are of recent origin undergoing
consolidation under its self weight and in most of the cases they have flocculated
structure and moisture content close to the liquid limit. Typical properties of soft clay
are:-

2
 IMPROCE 2009

i. Moisture content ≅ Liquid limit

ii. Cohesion strength = 10 to 20 kPa
iii. SPT (N) Value = 0 to 2
iv. Void ratio = 0.8 to 1.2
v. Compression index (Cc) = greater than 1.0
vi. Co-efficient of Consolidation (Cv) = 1 to 2 m2/year

Major problems faced in placing structure over such deposits are -

a. Very low bearing capacity

b. High settlements (the rate of settlement is very slow and settlement may
continue for several years)
c. Instability of deep excavations
d. Deep seated slip failure
e. Lateral flow under surface loading leading to settlement and lateral force on
adjacent foundations.

3.2 Loose Sand Deposits

Quite often the alluvial deposits along the river flood pan as well as beach sand
and sand deposits in desert have low relative density. Since the particles are arranged in
loose configuration, slight shear strain may result in large volume reduction. Typical
properties of loose sand deposits are:
i.
Saturated unit weight = 19 kN/m3
ii. Dry unit weight = 14 kN/m3
iii. Void ratio = 0.85
iv. SPT (N) Value = 4 to 10
v. Angle of friction (φ) = 28o to 30o
vi. Compressibility modulus (Es) = 5000Kpa
vii. Permeability constant (k) ≅ 10-5m/sec

Most of the problematic deposits have uniform gradation due to the sorting
carried out during transportation by river, wind, etc. Well graded deposits on the other
hand have better engineering behaviour. Major problems faced in placing structures
over loose sand deposits are:

a. Low bearing capacity (mostly local shear failure instead of general shear failure)
b. Higher settlement (though settlement is high, unlike soft clay deposits, these
settlement is not time bound and takes place during construction itself.
c. Susceptibility to liquefaction
d. Higher seepage loss for water retaining structures
e. Higher earth pressures on retaining structures

Numbers of techniques have been developed in recent past to densify such weak
deposits which are discussed in the following sections.

3
 IMPROCE 2009

4. METHODS OF GROUND IMPROVEMENT

The methods of ground improvement that can be adopted depends basically on

nature of strata and purpose of improvement. The methods available are as follows

(a) For cohesive soils:

i. Vertical Drains
ii. Vacuum Dewatering
iii. Stone columns
iv. In-situ deep mixing

(b) For cohesionless soils:

i. Compaction piles
ii. Vibro compaction
iii. Stone-columns
iv. Dynamic compaction
v. Compaction by deep blasting
vi. Grouting

The final choice among the methods available will depend on overall economy
in the total foundation cost. In most of the cases ground treated with one of these
methods results in less foundation cost compared to the cost of pile foundation which is
to be otherwise adopted if no treatment is adopted. A brief procedure and salient feature
of the above methods are given below.

4.1 Vertical Drains and Preloading

This method is suitable for deep deposit of soft clay. The natural moisture
content in this stratum can be brought down substantially by installing vertical drains
with a preloading. The presence of vertical drains reduces the drainage path of water in
the pores of soil and thereby reduces the time required for consolidation. The spacing
of drains depends on the speed at which required improvement is to be achieved.

In earlier days, such vertical drains were installed by driving a close ended steel
pipe of 100-200 mm diameter upto the full thickness of such soft clay deposit. The pipe
is then filled with sand and withdrawn in stages to form a vertical sand drain. The pipe
is generally refused for installing other drains.

In the recent past, there has been number of different materials developed to
replace the sand drains. These are basically flexible plastic sections having thickness
varying from 5 to 10 mm and width from 100 to150 mm. The section has channels to
permit flow of water. The perimeter of the section is covered with a layer of geo-textile
to prevent the entry of soil particles into the channel. The advantage of such drain is
that it results in minimum remoulding of surrounding soil during installation. The
process of installation is also very fast. The machine is mounted on a crane and typical
drain upto 10 m depth can be installed in a period varying from 1-2 minutes including
the time for shifting the machine to the new location.

The percentage consolidation which can be achieved by such vertical drains can
be theoretically predicted from the design charts developed based on theory of three
4
 IMPROCE 2009

dimensional consolidation. Typically, a deposit which requires a period of over ten

years for 95% consolidation can complete the same consolidation within a short period
of 3 to 6 months when vertical drains are installed.

After the drains are installed, the magnitude of preload to be placed depends on
the required shear strength of the layer after improvement. Depending on the time
available for improvement, the degree of consolidation is worked out and the effective
over burden pressure (p) is computed. From plasticity index of the soil, the ratio Su / p
is determined. This ratio normally remains constant and therefore with increase in value
of p is determined. This ratio normally remains constant and therefore with increase in
value of p the shear strength after treatment increases.

4.2 Vacuum Dewatering

In vacuum dewatering technique also the layer is provided with vertical drains
as mentioned above. The surface is then covered with a layer of sand blanket 150-300
mm thick and covered with a thin plastic sheet which is suitably anchored along the
perimeter. Vacuum is then applied within the sand blanket which creates a suction
pressure upto 50-60 mm height of mercury and forces the pore water out from the
deposit through vertical drains. The area is also subjected to atmospheric pressure
which then acts as a preload. The vacuum pressure is maintained till the required
improvement is achieved. The period of vacuum application can be predicted in similar
way as discussed for the vertical drains.

4.3 Stone Column

Stone columns are cylindrical columns made below ground level which
comprises of granular material of large size varying from 25 to 100 mm. A hole is
made in the soft deposit by different techniques and then filled with stones in layers and
compacted to form the complete column. When a structure is placed over the area
treated by stone columns, majority of the load (80-90%) is transmitted to the stone
column because of their higher stiffness. Balance 10-20% of the load is taken by clay
deposit. With the help of this 10% of surcharge load, the soft clay is able to provide
adequate confinement to the cylindrical column. The maximum permissible actual
stress on the columns can be predicted from the known theory.

The area treated by stone columns can be used to support only flexible structures
such as embankment, oil storage tank, etc. because, the settlement even after treatment
with stone column can be large (50-200 mm). Without stone columns the settlement
could have been 3-4 times higher and also the bearing capacity would have been much
less. The stone columns are installed by different techniques, some of them are:

1. With bored piling equipment, a hole of required diameter is made with a steel
liner. Stones are the poured in stage of 2-3 m height, compacted with a hammer
at each level and the steel liner is simultaneously withdrawn as the height of
column increases. This method does not create much lateral displacement
within the soft clay, which can otherwise result in remoulding of the soil.

2. With a large size vibrator, a hole is formed in the ground using high pressure
water jets. After vibrator is penetrated upto required depth of column, the stones
5
 IMPROCE 2009

are poured through the annular gap between the vibrator and surrounding soil.
The high power vibration compacts the stones in stages. The stones are poured
in stages and compacted till entire column is formed. This technique is called
vibroplacement.

3. In vibrocompaction, the vibrator is forced into the ground without water jets.
The compaction is achieved by pushing soil particles to the surrounding soil.
The vibrator is withdrawn in stages and the volume left by the vibrator is
occupied by the surrounding soil. There is no necessity of stones to be added in
this process. At the end of compaction, the surface settles by few centimeters as
a result of compaction. This technique is fast but suitable only for cohesionless
deposits.

4. Rammed Column: In this technique, a steel liner with a closed end is pushed
into the ground either with a drop hammer or with a vibratory hammer. After
the liner penetrates the required depth, stones/sand is poured inside the liner
(and the liner is gradually withdrawn keeping the liner under continuous vibrator
or with light taping with the hammer.

4.3 In-situ Deep Mixing

In-situ deep mixing using hydraulically operated helical blade augers penetrates
the ground to the required depth. The hollow stem of the auger is used to inject cement
slurry/lime/any other stabilising compound into the ground. A pair of 2 or 3 augers is
operated simultaneously. After injecting the stabiliser into the ground, the augers are
rotated such that the soil along with the stabiliser is churned in-situ and mixed the
stabiliser thoroughly with the soil without necessity of taking out the soil. The auger
configuration can be chosen such that either a stabilised wall can be formed (to act as
barrier) or the entire area can be stabilised.

4.4 Thermal Treatment

In thermal treatment, the moisture from the soft clay is driven out by raising the
temperature of the deposit. This is done by passing hot gasses through the soil.
Inclined bore holes of suitable diameter are drilled along the slope and a fire is created
near the bottom of the holes. The bore holes act as a chimney and carry the hot flues.
This reduces the water content considerably and stabilises the slope. The deposit could
also be heated by electrical energy, but it works out to be more expensive. Thermal
treatment is used more for temporary purpose like stabilisation of excavation slope
during construction.

4.5 Compaction Piles

In olden days, compaction piles comprised of wooden tree trunks were driven
into ground at required spacing. The voids in the strata gets reduced equivalent to the
volume of the pile. The spacing between the piles can be chosen to achieve the required
improvement. There has been considerable development in installing sand compaction
piles in which a casing pipe with closed end is driven to the required depth, filled with
coarse sand in layers and the pipe is withdrawn compacting the sand. Installation of
compaction pile is very fast and effective in compacting the cohesionless deposits. The
6
 IMPROCE 2009

technique is not suitable for cohesive soil as the lateral displacement while driving the
pile can result in remoulding of surrounding soil.

4.6 Dynamic Compaction

In dynamic compaction, a heavy weight of 15-20 ton is raised to a height of

about 20 m above the ground level and freely dropped on the ground surface. The
impact on the ground transmits compression waves, which spreads radially from the
point of impact. The compression waves (P waves) travels fast and creates liquefaction
of the strata, which is saturated. It is followed by shear waves, which has less velocity
but higher energy. The shear wave reorients the soil particles, which are liquefied state
to denser state and compacts the configuration. Representative blows (5-10) are applied
at each treatment point. The spacing between the treatment point, number of impact at
each point, etc. can be varied depending on the required improvement. There is no
rational theory to arrive at spacings and usually it is decided by compacting a trial area.
Deposits upto 10-20 m depth can be successfully compacted by this technique.

Wherever the silt/clay content in deposit exceeds 15 to 20%, the method is

found to be less effective. After each blow, the pore pressure developed in the ground
takes long time to dessipate as its percentage fines is higher. Though the technique is
used in deposits such as clayey sand and silty sand, the efficiency is less compared to
compaction of sand without fines.

4.8 Compaction by Deep Blasting

Thick cohesionless deposits such as mine back fill or natural cohesionless

deposit can be effectively densified by deep blast. In this technique, a bore hole is made
to a depth approximately 3/4 or 2/3 of the total depth. Suitable explosive is lowered to
the bottom of the borehole and the space above is back filled with compacted soil
(known as stemming). The explosive is attached with cordex which is taken upto the
surface and connected to electric detonator. When the explosive is detonated, it creates
a cavity, which expands instantaneously. The cavity expansion generates shock waves.
Propagation of these waves densifies the soil to certain extent.

After the blast, a large conical depression is observed at the place of blast. The
water within the pores of the deposit gets expelled out and comes upto the ground and
forms a large pool of water in the cone of depression. The quantity of charge and
spacing between explosives can be suitably selected to achieve the required
densification. As a thumb rule, about 15 to 45 grams of explosive is required for
densification of every m3 of the deposit. Initial loose deposit which has relative density
of about 50% and susceptible to liquefaction can be densified to a relative density of
over 60% which is adequate to prevent liquefaction. The vibrations created during the
blast are within permissible limits for construction purpose.

4.9 Grouting

High pressure injection grouting technique have been developed which are
capable of injecting grout even in less permeable soil like clayey soil. A fluid stream of
high energy is produced by a nozzle at lower end of the drill rod which cut into the
surrounding soil. usually the stream consists of water or sand suspension. Grout
7
 IMPROCE 2009

columns are formed by continuously turning and retracting the drill rod. The diameter
of columns depends on the pressure adopted and permeability of the surrounding soil.
Typical application of ground improvement by grouting is to strengthen the existing
foundation, supporting tunnel roof, trench excavation in weak soil, create barrier for
contaminated soil, cut-off trench for reducing seepage loss, etc.
5. CASE STUDY ON GROUND IMPROVEMENT

Large numbers of major projects in the country have adopted one of the methods
described above. Few typical case studies are highlighted here. In each case, field
measurements have been made to find the property of the deposit before and after
treatment. The post treatment results show that the ground improvement was effective
to meet the requirement.

5.1 Improvement of Open Stack Ground behind 7th Cargo Berth at

Kandla Port
Large area measuring 250 m x 90 m is covered by 20 m deep soft marine clay.
The soft marine clay has average shear strength of 15-25 kN/m2. The area was required
to stack cargo with loading intensity upto 50 kN/m2. Providing pile foundation for such
large area is impractical and it will be very expensive. It was therefore decided to
improve the area with vertical drains and preloading. 200 mm diameter sand drains
have been provided with a centre to centre spacing of 1.5m in both directions. After
installation of drains a 500 mm thick sand blanket was provided over the area. The area
was then subjected to soil preload of 6 m height. The area was constantly monitored for
settlement, pore water pressure in the ground and lateral flow if any. The pore water
pressure measurements indicate that the pore pressure stabilised within a period of
about 6 months. The preload was then removed. Vane Shear Test carried out after
preload indicate improvement in shear strength from 20 kN/m2 to 45 kN/m2. The area
was then paved with designed pavement section. The berth is now operational for last
2-3 years and the open stacked ground has satisfactory performance. Typical details of
strata and pore pressure records will be discussed during lecture.
Subsequently, the 8th cargo berth with equal area has been treated similarly with
band drains and the results are found to be satisfactory.

5.2 Fertilizer Plant at IFFCO – Aonla

For a major fertiliser plant in Gangetic belt of Uttar Pradesh, the strata is
essentially silty sand having SPT(N) value from 4-7 upto 10 m depth. The nature of
strata and presence of ground water table close to the ground level caused threat
liquefaction during earthquake.

When the first phase of the plant was taken up, due to lack of experience on
foundation on this type of deposit, a conservative approach was adopted. It was decided
to provide RCC pile foundation comprising of 400 mm diameter driven cast-in-situ
piles. Even if the plant structures are supported on piles, during liquefaction, the pile
cannot have lateral stability upto the liquefaction zone and therefore the design required
high percentage of steel in the piles. To avoid this problem, each group of piles were
provided with 2 to 3 rows of sand compaction piles around the group upto 10 m depth to
prevent liquefaction and improve lateral capacity.
8
 IMPROCE 2009

Due to large nature of the project, about 10000 RCC piles were estimated and it
was decided to carry out detailed field monitoring to optimise the foundation.
Accordingly, numbers of vertical, lateral and uplift load tests were carried out on RCC
piles. This enabled to increase the pile capacity considerably and accordingly
significant reduction in number of piles could be achieved. In addition, number of in-
situ tests, like standard penetration tests were carried out in-between the compaction
piles installed with different spacing to check whether it meets the requirement to avoid
liquefaction. These detailed data helped to increase the spacing between the
compaction piles. As a result of these detailed monitoring, it was possible to reduce
RCC piles by about 2000 numbers and the compaction piles could also be reduced by
25% of the total number. This resulted in considerable economy and speedy completion
of the project.

When the second phase of the plant was taken up, the foundation design was
reviewed and based on earlier experience it was felt that the ground after compaction
alone meets all the requirements of bearing capacity, settlement, liquefaction, etc. and
hence there is no necessity of any RCC piles. Further before making the final choice, a
trial area in the expansion site was treated with 900 mm diameter vibroflotation stone
column at 3 different spacings namely 1.5 m, 2.0 m and 2.5 m centre to centre. Field
footing tests were carried out to check the vertical capacity and settlement. These
monitoring confirmed that RCC piles can be eliminated. Based on results of these field
tests, depending on the actual loading intensity of various structures, the spacings of
stone column was provided. Accordingly the phase II of the project was supported
entirely on stone columns which resulted in large savings and also reduced the
construction time considerably.

5.3 Dynamic Compaction of Reclaimed Area at NTPC Power Plant near

Kayamkulam

For a major Thermal Power Project near Kayamkulam in Kerala, large area
measuring about 600 m x 400 m, which was covered by backwater, was reclaimed using
hydraulic dredged fill. The sand deposit 3-5 m thick was in loose condition and which
was susceptible to liquefy. Average SPT ‘N’ value varies from 2-6. To avoid
liquefaction it was desirable to achieve minimum SPT ‘N’ value of ground level
increasing to 14 at 10 m depth.
After considering several options of ground improvement it was found that
dynamic compaction would be most effective solution. This method has additional
advantage of compaction deposit very fast. To decide the spacing of the treatment
point, a trial area measuring 100 m x 100 m was adopted. Compaction was carried out
using a pounding weight of 15t falling through a height of 20 m. Number of blows and
spacings was designed in such a way that each square meter of the area receives input
energy of 10 t.m. After compaction of the trial area, SPT test and Static Cone
Penetration Tests were carried out which showed that the improvement was not upto the
mark and it was necessary to increase energy input further. Accordingly, energy input
was increased from 10 t.m to 15 t.m per m2. The measurement of pore pressure
revealed that one week after the energy was imparted; the pore pressures were not
dissipating. Due to this, the compaction was not effective. It was concluded that large
time gap is required between the first phase and second phase of compaction. This time
gap considerably delays the project schedule. Typical data of the soil strata, SPT/SCPT
results, pore water pressure measurement will be discussed during the lecture.
9
 IMPROCE 2009

The post compaction results showed the required compaction could be achieved
with dynamic compaction, which was adequate to prevent the soil liquefaction.

6. SUMMARY

This lecture highlights the necessity of ground improvement, variety of methods

available for ground improvement, typical case studies, etc. With large number of field
test data now available, it is hope that more and more design procedure, codes of
practices will be available which will help the geotechnical designer. In view of the
limitations of the theory, detailed field monitoring is strongly recommended.

REFERENCES

1. Carillo, N.J. (1942),"Simple two and three dimensional cases in the theory of
consolidation of soils", J.Math. phys., 21(1), 1-5.
2. Gandhi, S.R. (1998), "Ground Improvement", QIP/CEP short term course on
Foundations on weak deposits, 21-25 Dec. 1998, IIT, Madras, pp. 1-13.
3. Gandhi, S.R., Dey, A.K., and Selvam, S. (1999) “Densification of pond ash by
blasting”, Journal of Geotechnical and Geoenvironmental Engineering,
American Society of Civil Engineers, Vol. 125, pp.889-899.
4. Gandhi, S.R. (1996) " In-situ Densification of Deposited Ash Slurry"
Proceedings of national ash ponds and ash disposal systems held at IIT Delhi,
March 1996. pp 211 - 221. Publication by Narosa publishing house.
5. Gandhi, S.R., Raju, V.S. and Vimal Kumar (1997), “Densification of Deposited
Ash Slurry”, Proc. of 13th International Conference on Solid Waste Technology
and Management, Widener University, Philadelphia, Vol.1, pp.4C1-4C8.
6. Greenwood, D.A. and G.H. Thomson (1984), " Ground Stabilization :Deep
compaction and Grouting", ICE works construction Guides, Thomas Telford
Ltd., London, 47 pp. (available through ASCE).
7. IS: 6403-1981 " Code of practice for Determination of Bearing capacity of
shallow foundations", Bureau of Indian Standards.
8. Landau, R.E., (1996), "Method of installation as a factor in sand drain
stabilization design". Highway Research Record, 133, Highway Research
Board, Washington, D.C., pp.75-97.
9. Leonards, G.A. (Ed) (1962), “Foundation Engineering”, McGraw-Hill Book
Company, New York.
10. Raju,V.S., and Gandhi S.R.,(1993)," Report on Analysis of Post compaction
results and Foundation recommendations", submitted to Indian farmer's co-
operative limited, New Delhi, I.I.T. Madras-36.
11. Rixner, J.J., Kraemer, S.R., and Smith, A.D. (1986)," Prefabricated vertical
drains : Vol.1, Engineering guidelines" Report No. FHWA/RD-86/168, Federal
Highway Admin., McLean, Va.
12. Sharma, BVR and Naresh, D.N. (2000) “Dynamic Compaction for Improvement
of Ground at a Power Project Site”, Civil Engineering and Construction Review,
May 2000, pp 12-16.
13. Yeung, A.T., (1997), "Design curves for prefabricated vertical drains", Jl. of
Geotechnical and Geoenvironmental Engineering, Vol.123, No.8. pp. 755-759.

10
 IMPROCE 2009

IDENTIFICATION AND CLASSIFICATION OF PROBLEMATIC SOILS

Dr. V.K. Stalin *

1. INTRODUCTION

Even though the soil is a weathered product of rocks, they do not truly represent
the behaviour as to that of its parent rock. Because of different process of weathering
such as physical, chemical and physio–chemical, the weathered derivatives of rock all
together behave differently. Physically weathered rock mass to some an extent closely
represents the parent rock materials, for instance gravel and sand. On the other hand,
chemically weathered rock mass result in the formation of totally different materials
such as clay. Because of the formation of new minerals, the soils such as clay exhibit a
different characteristics with respect to change in the environmental condition.

Needless to say, gravel and sand are regarded as a good quality material and
being so they are suitable for any civil engineering applications, except the situation
where permeability is to be controlled. Whereas clays, chemically weathered rock mass,
show volume change behaviour on the addition and removal of moisture content.

Because of the mineralogical composition, there is a considerable change exist

among the different clays. The dominant mineral groups present in clays are kaolinite,
montmorillonite and Illite. In general, montmorillonite mineral enriched clays show
poor shear strength, high compressibility, high volume change (both swelling and
shrinkage) and low permeability, for example block cotton soil. On the other hand,
kaolinite mineral enriched soil exhibit high shear strength, low compressibility, high
permeability and low volume changes, for example red soil. Where as Illitic mineral
enriched soils show the behaviour in between montmorillonite and kaolinite clays.

Apart from weathering process responsible for the soil behaviour, it is equally
important that the natural agents are also responsible for soil to ensure a specific
characteristics. If wind is the agent for transfer of weathered rocks, it results in aeoline
deposits which are cohesionless soils such as gravel and the sand. If water is the agent
for movement of weathered rock products, it results in the formation of alluvial deposit
and their suitability has construction material is varying from poor to fair. Other deposit
are glacial, marine , beach, etc. Residual deposits are the one which is not transported to
farther distances.

Suitability of any soil can be assessed based on its properties. The soil properties
include index and engineering properties. The index properties are specific gravity, void
ratio , liquid limit, plastic limit, shrinkage limit, relative density, dry density, porosity,
initial water content, grains size distribution etc. The engineering properties are shear
strength, compressibility and permeability. There are some specific properties of soil
also a geotechnical engineer will come across called expansivity and shrinkage.

Unlike other material, soil behaviour is influenced by many factors such as

mineralogy, water content, void ratio, soil structure, porefliuid characteristics (Ion
concentration , valancy of iorn , dielectric constant) , temperature, drainage, condition,
* Assistant Professor, Department of Civil Engineering, Anna University Chennai, Chennai - 25 11
 IMPROCE 2009

strain rate, aging etc. Beside the complexity of understanding soil, geotechnical
engineers made their best efforts to group the soil based on its specific response to
different environmental conditions.

With the efforts of researchers, soil can be classified as high compressible and
low compressible, expansive & non expansive, sensitive & insensitive, high plastic &
low plastic, very soft to stiff clay, loose and dense sand etc. In this note, the
identification and classification of different soils are presented in order to classify the
good and poor soil, otherwise called as Problematic soil.

2. IDENTIFICATION AND CLASSIFICATION OF SOILS

2.1 Liquid Limit

Liquid limit values of soils may be described as low, intermediate, high very
high or extra high plasticity as given below in Table 1.

Table 1. Plasticity Classification based on Liquid limit

Plasticity Classified as Liquid limit (%)

Low 20 to 35
Intermediate 35 to 50
High 50 to 70
Very high 70 to 90
Extra high Over 90

2.2 Plasticity Index

No uniform standard is adopted in classifying degree of plasticity of soils.

However, the classification given below in Table 2 is approximately the one which is
often used and hence is recommended.

Table 2. Plasticity Classification based on Plasticity Index

Soil classified as Plasticity Index (%)

Non – Plastic 0-5
Moderately Plastic 5-16
Plastic 16-35
Highly plastic Over 35

2.3 Shrinkage Limit

Shrinkage limit of soil is on indication of not only the shrinkage potential of

clays but also and indicative of swelling nature. Lesser the shrinkage limit higher the
shrinkage potential and swelling potential and vice versa. These 3 can be used for
classifying shrinkage nature of soil based on shrinkage limit.

12
 IMPROCE 2009

Table 3. Swelling and Shrinkage Classification based on Shrinkage Limit

Classified as Shrinkage table / Shrinkage limit

Swelling detail (%)
Very Low <1
Low 8 to 10
Intermediate 11 to 15
High 16 to 20
Very high Over 20

2.4 Indian Standard Classification System

Indian Standard Classification (ISC) system adopted by Bureau of Indian

Standards is in many respects similar to the Unified Soil Classification (USC) system.
However, there is one basic difference in the classification of fine-grained soils. The
fine-grained soils in ISC system are subdivided into three categories of low, medium
and high compressibility instead of two categories of low and high compressibility in
USC system. A brief outline of Classification and Identification of Soils for general
engineering purposes (IS : 1498-1970 is given below. ISC system classifies the soils
into 18 groups as per Tables 4,5 and 6. Soils are divided into three broad divisions:

1. Coarse-grained soils, when 50% or more of the total material by weight is

retained on 75 micron IS sieve.

2. Fine-grained soils, when more than 50% of the total material passes 75 micron
IS sieve.

3. If the soil is highly organic and contains a large percentage of organic matter
and particles of decomposed vegetation, it is kept in a separate category marked
as peat (Pt).

There are 18 groups of soils: ‘8’ groups of coarse- grained, ‘9’ groups of fine-
grained and one of peat. Basic soil components are given in Table 4.

1. Coarse-grained Soils :Coarse-grained soils are subdivided into gravel and sand.
The soil is termed gravel (G) when more than 50% of coarse fraction (plus 75 is
retained on 4.75 mm IS sieve, and termed sand (S) if more than 50% of the coarse
fraction is smaller than 4.75 mm IS sieve. Coarse-grained soils are further
subdivided as given in Table 5 into 8 groups.

2. Fine-grained Soils :The fine-grained soils are further divided into three
subdivisions, depending upon the values of the liquid limit:
a. Silts and clays of low compressibility—These soils have a liquid limit less than
35 (represented by symbol L).
b. Silts and clays of medium compressibility—These soils have a liquid limit
greater than 35 but less than 50 (represented by symbol I).
c. Silts and clays of high compressibility—These soils have a liquid limit greater
than 50 (represented by symbol H).
d. Fine-grained soils are further subdivided, in 9 groups as given in Table 6

13
 IMPROCE 2009

Table 4 : Basic Soil Components in ISC System

Soil Soil components Symbol Particle size range and description

i) Coarse Boulder None Rounded to angular, bulky, hard, rock, particle;
Grained average diameter more than 300 mm
components
Cobble None Rounded to angular, bulky, hard, rock particle;
average diameter smaller than 300 mm but retained
on 80 mm IS sieve

Gravel G Rounded to angular, bulky, hard, rock particles;

passing 80 mm IS sieve but retained on 4.75 mm IS
sieve Coarse : 80 mm to 20 mm IS sieve Fine : 20
mm to 4.75 mm IS sieve

Sand S Rounded to angular, bulky, hard, rock particle;

passing 4.75 mm IS sieve, but retained on 75 micron
sieve Coarse : 4.75 mm to 2.0 mm IS sieve Medium :
2.0 mm to 425 micron IS sieve Fine : 425 micron to
65 micron IS sieve
(ii) Fine-grained Silt M Particles smaller than 75 micron IS sieve; identified
components by behaviour, that is, slightly plastic or non-plastic
regardless of moisture and exhibits little or no
strength when air-dried.

Clay C Particles smaller than 75 micron IS sieve: identified

by behaviour, that is, it can be made to exhibit plastic
properties within a certain considerable strength
when air dried.

Organic matter O Organic matter in various sizes and stages of

decomposition.

2.4.1 Boundary Classifications

Most often, it is not possible to classify a soil into any one of 18 groups
discussed above. A soil may possess characteristics of two groups, either in particle size
distribution or in plasticity. For such cases, boundary classifications occur and dual
symbols are used.
a. Boundary classification for coarse-grained soils
The following boundary classification can occur:

1. Boundary classifications within gravel group or sand group can occur. The
following classifications are common.
GW-GP, GM-GC, GW-GM, GW-GC, GP-GM
SWSP, SMSC, SWSM, SWSC, SP—SM

2. While giving dual symbols, first write a coarser soil then a finer soil. Boundary
classification can occur between the gravel and sand groups such as
GW—SW, GPSP, GM—SM, and GC-SC

The rule for correct classification is to favour the non-plastic classification. For
example, a gravel with 10% fines, Cu = 20 and Ce = 2.0 and Ip = 6 will be classified as
GW—GM, and not GW—GC.
14
 IMPROCE 2009

Table 5 : Classification of Coarse – grained Soils (ISC System)

Group Typical
Division Subdivision Laborator Criteria Remark
symbol names
(1) Coarse Gravel Clean (1)GW Well Cu grater than Cc between When
grained (G) (more gravels graded 1 fines are
soils than half (Fines less gravels between
(More than of coarse than 5%) (2)GP Poorly Not meeting all gradation 5% to
half of fraction is graded requirements for GW 12%,
material is larger gravels border
larger than than 4.75 Gravels (3)GM Silty Atterberg Atterberg line
75-micron mm IS with gravels Limits Limits cases
IS sieve sieve) appreciable below plotting requiring
size) amount of A-line or above A-line dual
fines Ip less with Ip
(Fines than betwen 4
more than (4)GC Clayey Atterberg and 7 are
12%) gravels limits border-line
above A- cases
line and Ip requiring use
grater than of dual
7 symbol
GM—GC
Sand (S) Clean (5) Well- Cu greater than 6 Cc
(Mute sands SW graded between 1 and 3
than half (Fines less sands
of coarse than 5%) (6)SP Poorly- Not meeting all gradation
fraction is graded requirements for SW
smaller sands
than 4.75 Sands (7)SM silty Atterberg Atterberg's
mm IS with, sands Limits Limits
sieve) appreciable below A- plotting
amout of line or Ip above A-line
fines less than 4 with Ip
(Fines (8) SC Clayey Atterberg between 4
more than sands limits and 7 are
12%) above A borderline
line with t^ses
Ip greater requiring use
than 7 of double
symbols
SM—SC

(b) Boundary classification for fine-grained soils.

1. Within the same compressibility subdivision, such as ML-CL, ML—OL, CL-

OL; CI—Ml MI—OI, CI-OI; MH-CH, MH—OH; CH-OH. First write
a coarser soil when there is a choice and then a finer soil.
2. Between low and medium compressibility, such as ML—MI, CL—CI; OL-OI
3. Between medium and high compressibility ML-MH, CI-CH, OI-OH
4. Boundary Classification between coarse-grained and fine-grained soils.
Boundary classification can occur between a coarse-grained soil and a fine-
grained soil, such as SM-ML, SC—CL

15
 IMPROCE 2009

Table 6 : Classification of Fine – grained Soils (ISC System)

Group Laboratory Criteria
Division Subdivision Typical names Remarks
Symbols (see Fig 5.6)
2) Fine- Low- (1) ML Inorganic silts Atterberg Atterberg (1) Organic and
grained compressibility with none to limits plot limits inorganic soils plotted
soils' (L) (Liquid low plasticity below A- pioting in the same zone in
(more than Limit less than line or Ip above A- plasticity chart are
50% pass 35%) less than 7 line with IP distinguis-1 hed by
75^, IS between 4 odour and colour or
sieve) to 7 liquid limit test
(hatched after oven-drying. A
zone) ML- reduction in liquid
CL limit after
(2)CL Inorganic clays Atterberg
of low plasticity limits plot
above A-
line and Ip
greater
than 7
(3)OL Organic silts of Atterberg limits plot oven- drying to a
low plasticity below value less than
A-line three- fourth of the
liquid limit before
Intermediate (4) MI Inorganic silts Atterberg limits plot oven- drying is
compressibility of medium below positive identification
(i) plasticity A-line of organic soils.
(Liquid limit
greater than 35%
but less than
50%
(5) CI Inorganic clays Atterberg limits plot (2) Black cotton
of medium above soils of India lie
plasticity A-line along a band partly
above the A-line
and partly below tha
Aline
(6)OI Orgaic silts of Atterberg limits plot
medium below
plasticity A-line
High (7)MH Inorganic silts Atterberg limits plot
compressibiliy of high below
(//) (Liquid limit compressibility A-line
greater than
50%)
(8)CH Inorganic clays Atterberg limits plot See, plasticity chart
of high above (Fig. 5.6)
plasticity A-line
(9) OH Organic clays of Atterberg limits plot
medium to high below
plasticity A-line
(3) Highly (10) Pt Peat and other Readily identified by
organic highly organic colour, odour, spongy
soil soils feel and fibrous texture
3.0 FIELD IDENTIFICATION OF SOILS
The soils can be identified in the field by conducting the following simple tests.
The sample is first spread on a flat surface. If more than 50% of the particles are visible
to the naked eye (unaided eye), the soil is coarse-grained; otherwise, it is fine grained.
16
 IMPROCE 2009

The fine- grained particles are smaller than 75μ size and are not visible to unaided eye.
The fraction of soil smaller than 75μ size, that is, the clay and silt fraction, is referred to
as fines.
1. Coarse-grained soil: if the soil is coarse-grained, it is further identified by
estimating the percentage of (a) gravel size particles (4.75 mm to 80 mm), (b) sand
size particles, (75μ to 4.75 mm) and (c) silts and clay size particles (smaller than
75μsize). Gravel particles are larger than 4.75 mm size and can be identified
visually.
If the percentage of gravel is greater than that of sand, the soil is gravel; otherwise,
it is sand. Gravels and sands are further classified as clean if they contain fines less
than 5% and as dirty if they contain fines more than 12%. Gravels and sands
containing 5 to 12% fines are given boundary classification. The fine fraction of the
coarse-grained soils is identified using the procedure given below for fine- grained
soils to determine whether it is silty or clayey.
To differentiate fine sand from silt, dispersion test is adopted. When a spoonful of
soil is poured in a jar full of water, fine sand settles in a minute or so, whereas silt
takes 15 minutes or more.
2. Fine-grained soils: If the soil is fine-grained, the following tests are conducted for
identification on the fraction of the soil finer than the 425-micron IS sieve to
differentiate silt from clay.
a. Dilatancy (reaction to shaking) test: A small pat of moist soil of about 5 ml in
volume is prepared. Water is added to make the soil soft but not sticky. The pat is
placed in the open palm of one-hand and shaken horizontally, striking against the
other hand several times during shaking. If the soil gives a positive reaction, the
water appears on its surface which changes to a lively consistency and appears
glossy. When the pat is squeezed between the fingers, the water and gloss disappear
from the surface. It becomes stiff and ultimately crumbles. The rapidity with which
water appears on the surface during shaking and disappears during squeezing is used
in the identification of fine-grained soils (Tables 7). The larger the size of the
particles, the quicker is the reaction. The reaction is called quick if water appears
and disappears quickly. The reaction is termed slow if water appears and disappears
slowly. For no reaction, the water does not appear at the surface.
b. Toughness test: The pat used in the dilatancy test is dried by working and
remoulding until it has the consistency of putty. The time required to dry the pat
depends upon the plasticity of the soil.
The pat is rolled on a smooth surface or between the palms into threads of about
3 mm in diameter. The thread is folded and re- rolled to reduce the water is soil, due to
evaporation by heat of hand, until the 3 mm diameter thread just crumbles. The water
content at that stage is equal to the plastic limit and the resistance to moulding at that
stage is called the toughness.
After the thread crumbles, the pieces of the sample are lumped together and
subjected to kneading until the lump also crumbles. The tougher the thread at the plastic
limit and the stiffer the kneaded lump just before it crumbles, the higher is the
toughness of the soil. The toughness is low if the thread is weak and the soil mass
cannot be lumped together when drier than plastic limit. The toughness is high when the

17
 IMPROCE 2009

lump can be moulded drier than plastic limit and high pressure is required to roll the
thread. The toughness depends upon the potency of the colloidal clay.
Table 7 : Field Identification Tests
Test ML CL OL MI CI CI MH CH OH
(a) Quick None to Slow Quick None Slow Slow to None None to
Dilatancy very slow to slow none very slow
(b) None Medium Low None Medium Low Low to High Low to
Toughness medium medium
(c) Dry None Medium Low Low Medium Low to Low to High to Medium to
strength of low to high medium medium very high
high

c. Dry strength test: The pat of the soil is completely dried by air drying, sun drying
or oven-drying. The dry strength is determined by breaking the dried pat and
crumbling it between fingers. The dry strength is a measure of plasticity of the soil.
The dry strength depends upon the colloidal fraction of the soil. The strength
is termed high if the dried pat cannot be powdered at all; medium, if considerable
pressure is required; and low, if the dry pat can be easily powdered.
4.0 CHARACTERISTICS OF SOILS OF DIFFERENT GROUPS
General characteristics of the soils of various groups as classified by ISC system
and USC system are given in Table 8. The information given in the table 8 should be
considered as a rough guidance about the engineering properties of soils. For complete
information, the tests should be conducted and the engineering properties determined.
Table 8 : General Behavior of Soils
Soil Group Permeability Compressibility Shear Strength Workability
(a) Gravels
GW Pervious Negligible Excellent Excellent
GP Very pervious Negligible Good Good
GM Semi-pervious to Negligible Good Good
impervious
GC Impervious Very low Good to fair Good
(b) Sands
SW Pervious Negligible Excellent Excellent
SP Pervious Very low Good Fair
SM Semi-pervious to Low Good Fair
impervious
SC Impervious Low Good to fair Good
(C) Low & medium
Plasticitysilt &
clays
ML, MI Semi-pervious to Medium Fair Fair
impervious
CL, CI Impervious Medium Fair Good to fair
OL, OI Semi-pervious to Medium Fair Fair
impervious
(D) High- plasticity
silts & clays
MH Semi-pervious to High Fair to poor Poor
impervious
CH Impervious High Poor Poor
OH Impervious High Poor Poor

18
 IMPROCE 2009

5.0 CLASSIFICATION OF EXPANSIVE SOILS

Damages to structures, property and life resulting from swell-shrink

characteristics of expansive soils have been reported from many parts of the world in-
cluding India. India, Africa,_ Australia, Israel, South America and United States of
America possess vast tracts covered with such soils. Our Black Cotton soil is an
expansive soil. It extends nearly one-fifth of our country, chiefly in the states of Tamil
Nadu. Maharashtra, Gujarat, Madhya Pradesh, Southern Uttar Pradesh, Eastern
Rajasthan, Karnataka and parts of Andhra Pradesh

Most of the expansive soils found in India are black in colour and are good for
growing cotton. Hence they have acquired the name black cotton soils. .Some of these
soils are reddish brown and yellowish grey in colours. These soils are generally found
near the surface, with layer thickness varying from 0.5 m to 10.m, and sometime more
than 10 metres. Typical properties of black cotton soils are given in Table 9.

Once an expansive soil is encountered at the project site, the following items
need to be given specific attention viz ;

i. Swell – shrink characteristics of soil encountered.

ii. Thickness and depth of various underlying soil layers
iii. Depths of significant moisture variation.
iv. Local climate and hydrology.
v. Floor-foundation system; ability to accommodate and tolerate the soil behaviour.
vi. Methods of improving the soil behavior.

Table 9: Typical Characteristics of Black Cotton Soils (Expansive Soils)

Sl.No Characteristics Normal Range of Values

1. Grain size distribution
Gravel 0 to 10%
Sand 15 to 25%
Silt 15 to 30%
Clay 25 to 70%
2. Consistency limits
Liquid limit 50 to 120%
Plastic limit 30 to 60%
Plasticity index 30 to 60%
Shrinkage limit 8 to 18%
3. Soil classification CI-CH-MH
4. Specific gravity 2.6 to 2.75
5. pH value 8 to 9
6. Organic content 0.4 to 2.4%
7. Differential free swell (DFS)* 40 to 180%
8 Swelling pressure 0.5 to 8.0 kg/cm2
(≈ 50 to 800 kN/m2)

• A check list furnished in Table 10 would be useful in recognising expansive soils in

the field. If the answer to any of the question in the checklist is “yes”, soil may be
expansive indicating need for further testing. Soil showing shrinkage cracks in dry
19
 IMPROCE 2009

season, can definitely be taken as expansive soil. Table 11 shows the classification
of expansive clays based on liquid, plastic and shrinkage limits. These will show
the classification of expansive clays based on liquid, plastic and shrinkage limits. In
general, higher the plasticity characteristics, higher is the swelling nature provided if
the soils are having expansive montmorillonitic mineral.

Table 10 Checklists for Recognition of Expansive Soil

Answers
Sl.No Item to be checked
Yes No
1. Are the soils nearby the project area known to be expansive?
2. Are there evidences of cracks in walls, curbs, sidewalks,
3. pavement etc. in nearby construction?
4. Are there shrinkage cracks in the soil (in dry season)?
5. Does the soil behave as a very sticky soil, sticking to shoes,
6. types etc. when wet?
If you take a lump of dry soil and try to break it between
fingers, do you find the soil hard and difficult to break?
If you select a large lump of dry soil (say about one kg) and
7. raise it chest high and drop it on hard surface (pavement),
does the soil stay as lump (s), without breaking up into a
8. number of small pieces? (Breaking of thin edges or sharp
corners or breaking into 2 or 3-major-pieces because of
existence of shrinkage cracks shall be discarded).
9. If you take (or make) a wet soil with moisture content that
would permit easy molding with light finger pressure and
then remould and roll it to threads of approximately 3 mm in"
diameter, can a thread, a few cm. long, stand on its own
weight when held down at one end?
If you make a uniform paste of soil and water (water content
should be almost same as item 7 above) on the palm of your
hand and strike the lower part of your hand against the other
hand in rapid but short strokes (5 to 8 cm long) for 10 to 15
times, does the soil look almost the same as when you started
without shiny surface?
If you take a ball of wet soil (similar to that in item 8 above)
and drop it on a piece of smooth, dry glass plate from a height
of about 45 cm and turn the glass upside down (soil towards
ground), slightly tilted and tap the top of the glass plate with
your finger tips, does the soil remain stuck in the original
position on the glass instead of falling off or sliding?

Note: A” yes” answer to any of the above questions may indicate an expansive soil
and a need for further testing and study.

20
 IMPROCE 2009

Table 11. Classification Expansive Soils

% Colloids PI % SL % LL % Swelling Potential

<15 <18 <15 <39 Low
13-23 15-28 10-16 39-50 Medium
20-31 25-41 7-12 50-63 High
>28 >35 >11 >63 Very high

6.0 SENSITIVE SOILS

Some clays have a curious property called sensitivity, which means their
strength in a remolded or highly disturbed condition is less than that in an undisturbed
condition at the same moisture content. These highly sensitive clays are called quick
clays, are found in certain areas of Eastern Canada, parts of Scandinavia, and else
where. This behavior occurs because these clays have a very delicate structure that is
disturbed when they are remolded. The degree of sensitivity is defined by the parameter
S1, the ratio of undisturbed shear strength to the remolded shear strength.

S undisturbed
Si =
S remolded

Table 12 presents two systems of classifying sensitivity based on S1 Note the

difference between the criteria commonly used in the United States, where highly
sensitive clays are rate. with that used in Sweden, where they are common. Shear
failures in highly sensitive clays can be very dramatic because the strength loss makes
the failure propagate over a wide area. This sometimes produces large flow slides.
Sensitive clays also can recover from these strength losses through a process called
thixotropic hardening.

Table 12. Classification of sensitive soils (Holtz and Kovacs, 1981)

Classification Sensitivity, S1
United States Sweden
Low sensitivity 2-4 <10
Medium sensitivity 4-8 10-30
High sensitivity 8-16 30-50
Quick >16 50-100
Extra Quick >100

7.0 CLASSIFICATION OF SAND

There is a good correlation exists between ‘N’ value and angle of internal
friction of cohesionless soils. The state of sandy soils can be classified as very loose,
loose, medium, dense and very dense based on ‘ϕ’ value or ‘N’ value as seen is table
13.

21
 IMPROCE 2009

Table 13 Correlation between N and φ

N Denseness φ
0-4 Very Loose 25° - 32°
4-10 Loose 27° - 35°
10-30 Medium 30° - 40°
30-50 Dense 35° = 45°
> 50 Very Dense > 45°

8.0 CLASSIFICATION OF CLAYS

The SPT ‘N’ value is having a good linear relationship with unconfined
compressive strength of clays. The state of clay can be classified has very soft, soft,
medium, stiff, very stiff, and hard based on either ‘N’ value or unconfined
compressive strength as shown in table 14.

Table 14 Correlation between N and qu

N Consistency qu (kN / m2)

0-2 Very Soft < 25
2-4 Soft 25 – 50
4-8 Medium 50 – 100
8 - 15 Stiff 100 – 200
15 – 30 Very Stiff 200 – 400
> 30 Hard > 400

9.0 DISPERSIVE SOILS

Dispersion occurs in soils when the repulsive forces between clay particles
exceed the attractive forces thus bringing about deflocculation so that in the presence of
relatively pure water the particles repel each other to form colloidal suspensions. In non-
dispersive soil there is a definite threshold velocity below which flowing water causes
no erosion. The individual particles cling to each other and are only removed by water
flowing with a certain erosive energy. By contrast, there is no threshold velocity for
dispersive soil, the colloidal clay particles go into suspension even in quiet water and
therefore these soils are highly susceptible to erosion and piping. Dispersive soils have a
moderate to high clay material content but there are no significant differences in the
clay fractions of dispersive and non-dispersive soils, except that soils with less than
10% clay particles may not have enough colloids to support dispersive piping.
Dispersive soils contain a higher content of dissolved sodium (up to 12%) in their pore
water than ordinary soils. The clay particles in soils with high salt contents exist as
aggregates and coatings around silt and sand particles and the soil is flocculated

The sodium adsorption ratio (SAR) is used to quantify the role of sodium where
free salts are present in the pore water and is defined as:

Na
SAR = 0.5 (Ca + Mg )

22
 IMPROCE 2009

with units expressed in meq/litre of the saturated extract. There is a relationship

between the electrolyte concentration of the pore water and the exchangeable ions in the
adsorbed layers of clay particles. This relationship is dependent upon pH value and also
may be influenced by the type of clay minerals present. Hence, it is not necessarily
constant. Gerber and Harmse (1987) considered an SAR value greater than 10 indicative
of dispersive soils, between 6 and 10 as intermediate, and less than 6 as non-dispersive.
However, Aitchison and Wood (1965) regarded soils in which the SAR exceeded 2 as
dispersive. The presence of exchangeable sodium is the main chemical factor
contributing towards dispersive behaviour in soil. This is expressed in terms of the
exchangeable sodium percentage (ESP):

exchangeable sodium
ESP = x 100
cation exchange capacity

Where the units are given in meq/100 g of-dry clay. A threshold value of ESP of
10% has been recommended by Elges (1985), above which soils that have their free
salts leached by seepage of relatively pure water are prone to dispersion. Soils with ESP
values above 15% are highly dispersive (Bell and Maud, 1994a). Those with low cation
exchange values (15 meq/100 g of clay) have been found to be completely non-
dispersive at ESP values of 6% or below. Similarly, soils with high cation exchange
capacity values and a plasticity index greater than 35% swell to such an extent that
dispersion is not significant.

Unfortunately, dispersive soils cannot be differentiated from non-dispersive soils

by routine soil mechanics testing. Although a number of special tests have been used to
recognize dispersive soils, no single test can be relied on completely to identify them
(Bell and Maud, 1994a). These can be divided into physical and chemical tests. The
former include the crumb test, the dispersion or double hydrometer test, the modified
hydrometer or turbidity ratio test and the pinhole test. Craft and Acciardi (1984) found
that the crumb test land the pinhole test at times yield conflicting results from the same
samples of soil. Subsequently, Gerber and Harmse (1987) showed that the crumb test,
the double hydrometer test and the pinhole test were unable to identify dispersive soils
when free salts were present in solution in the pore water, which is frequently the case
with sodium saturated soils.

Serious piping damage to embankments and failures of earth dams have

occurred when dispersive soils have been used in their construction (Bell and Maud,
1994b). Severe erosion damage also can form deep gullies on earth embankments after
rainfall. Indications of piping take the form of small leakages of muddy coloured water
from an earth dam after initial filling of the reservoir. The pipes become enlarged
rapidly and this can lead to failure of the dam. Dispersive erosion may be caused by
initial seepage through an earth dam in areas of higher soil permeability, especially
areas where compaction may not be so effective such as around conduits,, against
concrete structures and at the foundation interface; or through desiccation cracks, cracks
due to differential settlement or those due to hydraulic fracturing.

In many areas where dispersive soils are found there is no economic alternative
other than to use these soils for the construction of earth dams. However, experience

23
 IMPROCE 2009

indicates that if an earth dam is built with careful construction control and incorporates
filters, then it should be safe enough even if it is constructed

10. COLLAPSIBLE SOIL

Collapsible soils, which are sometimes referred to as metastable soils, are

unsaturated soils that undergo a large volume change upon saturation. This volume
change may or may not be the result of the application of additional load. The behavior
of collapsing soils under load is best explained by the typical void ratio-pressure plot (e
against log p) for a collapsing soil, as shown in Figure 1. Branch ab is determined from
the consolidation test on a specimen at its natural moisture content. At a pressure level
of Pw , the equilibrium void ratio is e1. However, if water is introduced into the
specimen for saturation, the soil structure will collapse. After saturation, the equilibrium
void ratio at the same pressure level pw is e2 ; cd is the branch of e - log p curve under
additional load after saturation. Foundations that are constructed on such soils may
undergo large and sudden settlement if and when the soil under them becomes saturated
with an unanticipated supply of moisture. This moisture may come from several
sources, such as (a) broken water pipelines, (b) leaky sewers, (c) drainage from
reservoirs and swimming pools, (d) slow increase of groundwater, and so on. This type
of settlement generally causes considerable structural damage. Hence identification of
collapsing soils during field exploration is crucial.

The majority of naturally occurring collapsing soils are aeolin that is, wind-
deposited sand and/or silts, such as loess, aeolic beaches, and volcanic dust deposits.
These deposits have high void ratios and low unit weights and are cohesionless or only
slightly cohesive. Loess deposits have silt-sized particles. The cohesion in loess may be
the result of the presence of clay coatings around the silt-size particles, which holds
them in a rather stable condition in an unsaturated state. The cohesion may also be
caused by the presence of chemical precipitates leached by rainwater. When the soil
becomes saturated, the clay binders lose their strength and hence undergo a structural
collapse. In the United States, large parts of the Midwest and arid West have such types
of deposit. Loess deposits are also found over 15%-20% of Europe and over large parts
of China.

Figure 1 Nature of variation of void ratio with pressure for a collapsing soil

24
 IMPROCE 2009

Table 15 Relation of Collapse Potential to the Severity of Foundation Problems

(Clemence and Finbarr (1981)

Cp (%) Severity of problem

0-1 No problem
1-5 Moderate trouble
5-10 Trouble
10-20 Severe trouble
20 Very severe trouble

11. SUMMARY

Stability of any civil engineering structures lies primarily with the response of
soil under the influence of external loading. It is a must for any civil engineer to
understand the type of soil and their engineering characteristics prior to the use of same
for any applications. If there is no proper importance given to the soil before start of
construction activities in the beginning itself, then the rectification of damage to the
structure, because of soil movement, if any would be much higher than the cost of the
project itself.

REFERENCES

1. Braja M. Das (1999), “ Principles of Foundation Engineering”, PWS Publishing

Company.

2. Donald P. Coduto (1999), “Geotechnical Engineering Principles and

Practice”,Prentice Hall of India Private Limited.

3. Murthy V.N.S (2000), “Text Book of Soil Mechanics and Foundation Engineering”,
CBS Publications.

4. Nayak, V. (1996) Foundations Design Manual, Dhanpat Rai Publications (p) Ltd,
New Delhi.

25
IMPROCE 2009

26
 IMPROCE 2009

ROLE OF REMOTE SENSING IN THE IDENTIFICATION OF SOIL PROPERTIES

Dr. S. Sanjeevi *

1. INTRODUCTION

Regional maps showing the distribution of soil types based on the engineering
properties are of immense value to a geotechnical engineer. Such maps are usually
prepared after extensive field surveys and tests. Current engineering and geologic
practices for characterization of soil types and problematic soils involves time-
consuming and expensive engineering tests such as the tests to determine swelling
potential, X-ray diffraction analyses for mineralogical identification, penetration tests
etc. Certain soil properties are also dynamic in nature. Hence, to monitor the status of
such dynamic soil properties, there is a need for an efficient system that can evade long
procedures involved in the traditional methods.

Remote sensing could be of interest because it has the potential of rapid

identification of certain properties/parameters of soils and the constituent minerals in
soils. As remote sensing technology is advancing into sub-meter resolution and
hyperspectral imaging, there is a corresponding interest in the detailed analysis of
spectral features at a close range distances. The question arises whether this technology
could become a useful tool for identifying and evaluating expansive soils, either with
field instruments at specific sites such as outcrops or trench walls, or with remote
sensing data at different spatial scales.

Problematic soils include swelling soils. Swelling is a major engineering

problem and ranks high among the natural hazards. It is the process where the soil
changes in volume with changing moisture conditions and has been attributed to be
mainly due to the presence of clay minerals, which are susceptible to penetration of
their chemical structure by water molecules. Most swelling potential estimation
methods have aimed at recognizing the clay minerals and associated soil properties,
examples of which include the Atterberg limits, the cation exchange capacity (CEC) and
the coefficient of linear extensibility (COLE) tests. The above discussion illustrates the
importance of identifying the type of clay mineral in swelling potential estimation. It is
therefore necessary that a method that can directly ascertain the type of clay mineral and
increase our ability to classify soil swelling potential should be developed. Reflectance
spectroscopy offers such a possibility.

This write-up describes the rudiments of remote sensing, the spectral properties
of soils and the applications of optical remote-sensing techniques to detect and map
certain geotechnical properties of soils.

Reflectance spectroscopy offers a new science that can be used to derive

significant information about mineralogy with little or no sample preparation and is
based on the principle of absorption feature mapping, which is physically based and
exploits the fact that many minerals exhibit absorption bands that are diagnostic of
mineral type and composition (Mustard and Jessica 1999). The key features used in this
mapping are the position, shape and relative strength (depth) of the absorptions, which
* Professor, Department of Geology, Anna University Chennai, Chennai - 25 27
 IMPROCE 2009

are characterized and then compared and analysed in the context of spectral libraries.
Reflectance Spectroscopy has been established as an excellent tool for clay mineralogy
based on the diagnostic fingerprints in the short-wave infrared (SWIR). It offers a more
fundamental method on the basis of its capacity to identify the clay mineral type and
content and thus has the potential to increase the accuracy of the swelling potential
classification schemes.

Remote sensing
Remote sensing is the science (and to some extent, art) of acquiring information
about the Earth's surface without actually being in contact with it. This is done by
sensing and recording reflected or emitted energy and processing, analyzing, and
applying that information. In much of remote sensing, the process involves an
interaction between incident radiation and the targets of interest. This is exemplified by
the use of imaging systems where the following seven elements are involved.
1. Energy Source - the first requirement for remote sensing is to have an energy
source which illuminates or provides electromagnetic energy to the target of
interest.
2. Radiation and the Atmosphere - as the energy travels from its source to the
target, it will come in contact with and interact with the atmosphere it passes
through. This interaction take place again as the energy travels from the target to the
sensor.
3. Interaction with the Target - once the energy makes its way to the target through
the atmosphere, it interacts with the target depending on the properties of both the
target and the radiation.
4. Recording of Energy by the Sensor (D) - after the energy has been scattered by, or
emitted from the target, we require a sensor (remote - not in contact with the target)
to collect and record the electromagnetic radiation.
5. Transmission, Reception, and Processing (E) - the energy recorded by the sensor
has to be transmitted, often in electronic form, to a receiving and processing station
where the data are processed into an image (hardcopy and/or digital).
6. Interpretation and Analysis (F) - the processed image is interpreted, visually
and/or digitally, to extract information about the target which was illuminated.
7. Application (G) - the final element of the remote sensing process is achieved when
we apply the information we have been able to extract from the imagery about the
target in order to better understand it, reveal some new information, or assist in
solving a particular problem.
How Sensors Work: The Basics
Sensors collect and store data about the spectral reflectance of natural features
and objects, both of which reflect radiation. This radiation can be quantified on an
electromagnetic spectrum. The electromagnetic spectrum is a continuum of
electromagnetic energy arranged according to its frequency and wavelength. As the
electromagnetic waves are radiated through space, their energy interacts with matter and
one of three reactions occurs. The radiation will either be:

1. reflected off the object

2. absorbed by the object
3. transmitted through the object

28
 IMPROCE 2009

The total amount of radiation that strikes an object is referred to as the incident
radiation, and is equal to:

incident radiation = reflected radiation + absorbed radiation + transmitted radiation

Of the three types of radiation, remote sensing is primarily concerned with

reflected radiation. This is the same radiation that causes our eyes to see colours, causes
infrared film to record vegetation, and allows radar images of the earth to be created.

Spectral Reflectance

Blu Green Red

Figure 1. The Electromagnetic Spectrum and its

Spectral reflectance is the portion of incident EM radiation that is reflected by a

non-transparent surface. The fraction of energy reflected at a particular wavelength
varies for different features. Additionally, the reflectance of features varies at different
wavelengths. Thus, two features that are indistinguishable in one spectral range may be
very different in another portion of the spectrum. This is an essential property of matter
that allows for different features to be identified and separated by their spectral
signatures.

Figure 2. Spectral signatures of certain basic cover types on the Earth’s surface.

29
 IMPROCE 2009

A spectral signature is a unique reflectance value in a specific part of the

spectrum. For any given material, the amount of solar radiation that it reflects, absorbs,
transmits, or emits varies with wavelength. When that amount (usually intensity, as a
percent of maximum) coming from the material is plotted over a range of wavelengths,
the connected points produce a curve called the material's spectral signature (spectral
response curve). This important property of matter makes it possible to identify
different substances or classes and to separate them by their individual spectral
signatures, as shown in Figure 2. The reflectance characteristics of earth surface features
may be quantified by measuring the portion of incident energy that is reflected. This is
measured as a function of wavelength (l) and is called spectral reflectance, SR.

SR = ER (l) /EI (l)

Where: ER is reflected energy and EI is incident energy.

Spectral Reflectance of Vegetation

The spectral characteristics of vegetation vary with wavelength. A compound in

leaves called chlorophyll strongly absorbs radiation in the red and blue wavelengths but
reflect green wavelength. The internal structure of healthy leaves act as diffuse reflector
of near-infrared wavelengths. Measuring and monitoring the infrared reflectance is one
way that scientists determine how healthy particular vegetation may be. Leaves appear
greenest to us in summer and become red or yellow with decrease in chlorophyll
content in autumn.

Spectral Reflectance of Water

Majority of the radiation incident upon water is not reflected but is either
absorbed or transmitted. NIR radiations and longer wavelengths are absorbed more by
water than the visible wavelengths. Thus, water looks blue or blue-green due to stronger
reflectance at shorter wavelengths and darker if viewed at red or NIR wavelengths. The
factors affecting the variability in reflectance of a water body are depth of water,
materials within water, salinity and surface roughness of water.

Spectral Reflectance of Soil

The majority of radiation on a soil surface is either reflected or absorbed and

little is transmitted. Four main factors influence the soil reflectance in remote sensing
images: mineral composition, soil moisture, organic matter content and soil texture
(surface). The soil curve shows less peak and valley variations. The presence of
moisture in soil decreases its reflectance. Empirical research on the relationships
between soil properties and the characteristics of scattered radiation has focused
predominantly on the spectral distribution of the radiation.

Remote Sensing of Soils

There are several different areas involved in remote sensing of soils. While
satellite images provide the visible boundaries of soil types, microwave remote sensing
provides for a shallow penetration of soils, thereby rendering sub-surface moisture
conditions. Studying the spectral signatures of the soil types provide additional physical
30
 IMPROCE 2009

data. Finally, interpreting the remote sensing images and the spectral signatures allow
for the classification of the soil types.

Information about soils from remote sensing is obtained by observing : (i) Crop
cover and vegetation since there is a definite relation between crop or vegetation and
between crop development and soil; (ii) Bare soil surface since there is a definite
relation between soil surface and soil reflectance, and (iii) Topography / relief
difference because there is a definite relationship between topography and soil.

Radiance of exposed soil can be expressed as:

Lt = Lp + Ls + Lv

Where, Lt = at-sensor radiance of a pixel of exposed soil; Lp= atmospheric path

radiance; Ls= radiance reflected off the air-soil interface boundary layer; Lv = volume
scattering of EMR which penetrates a few mm to cm.

The main factors influencing soil reflectance are: Roughness and texture;
Organic matter content; Moisture condition (OH) and Mineralogical composition (OH,
CO32, Fe2-, Fe3-etc). The causes of specific absorption bands are (i) Electronic processes
(short wavelength; absorption bands - Fe: UV, 0.4 to 1.0 microns); (ii) Vibrational
processes (long wavelength, relatively narrow bands - OH: 1.45 and 1.95 microns, OH:
> 1.0 microns (Minerals containing OH, H2O).

Figure 3. Spectral reflectance curves for sandy loam with different textures.

Size and shape of the soil aggregate also influence the reflectance in the images.
The mineral composition of soils affects the reflectance spectrum. Increasing
reflectance of soils occurs from the visible to the shortwave infrared - with absorption
bands around 1.4 um and 1.9 um related to the amount of moisture in the soil. Radar
waves may not be able to penetrate soil if it is moist. On the soil reflectance spectra, this
soil moisture will develop parallel curves. Moisture of soil has an equal effect over the
31
 IMPROCE 2009

spectrum and the ration between the spectral bands. Spectral bands of red and near-
infrared bands are independent from the soil moisture. Organic matter is the third factor
that influences soil optical properties. Organic matter may indirectly affect the spectral
influence, based on the soil structure and water retention capacity. High organic matter
in soil may produce spectral interferences for band characteristics of minerals like Mn
and Fe.

Texture and roughness

Soil texture (roughness) also affects soil optical properties. Light is trapped in
the rough surfaces of the coarse soil particles. For example, if iron and lime are present,
a stronger reflectance is received than if the soil material was fine textured and dry.
Variations in soil reflectances occur where there is a change in distribution of light and
shadow areas with surface roughness areas. This factor is important in the thermal
infrared and microwave spectral domains.

Soil size and shape influence the reflectance properties. If the size of a soil
aggregate expands in diameter, a decrease in reflection will result. The shape is related
to the surface (texture); a smooth, even surface will probably result from a more
spherical soil aggregate, instead of a jagged soil aggregate.

Wavelength (microns) Wavelength (microns)

Figure 4. Spectral signatures of sand & clay sowing the influence of moisture on the spectral response.

A rough surface generally reflects less, due to self-shadowing effects and due to
multiple scaterring. If a surface is smooth (particles smaller than wavelength), then
specular reflection is important. If there is no return, then the surface is dark, unless the
sensor is correctly positioned and pointed in specular direction. Smooth soil surfaces
tend to be clayey or silty, often are moist and may contain strong absorbers such as
organic content and iron oxide. conversely, a rough surface scatters EMR and thus
appears bright.

Organic Matter

There is always decrease in reflectance as the amount of organic matter

increases in soils. When the OM is greater than 2%, there is a masking of other
absorption features in the spectra and hence no distinct absorption features are seen in
soils with abundant organic matter.
32
 IMPROCE 2009

4.00%
22.9%
45.10%

Figure 5. Spectral signature for three soil types with varying OM content.

Soil moisture and texture

Water ‘coats’ particles, filling air spaces and reducing the amount of multiple
scattered light. Hence, soils with more moisture will be darker in the VNIR and SWIR
than drier soils. Moist soils will also be darker in the SWIR region where water
absorption increases significantly with increasing wavelength. The depth of the water
absorption bands at 1.4, 1.9 and 2.7 microns can be used to determine soil moisture.
There are certain soils that have the tendency to hold water because of a suitable texture.
Clays hold more water more tightly than sand. Thus, the spectra of clay display more
prominent water absorption bands than the spectra of sand.

Stoner and Baumgardner (1981) identified five characteristic curve shapes on

the basis of a visual inspection of their soil reflectance spectra. Each shape was
associated with certain soil characteristics. The five associated soil classes were: organic
dominated and fine texture), minimally altered affected (organic carbon >2%, iron oxide
<1%), iron-affected (organic carbon <%, iron oxide>1 %< 4%), organic affected
(organic carbon >2%, iron oxides<1%), and iron-dominated (iron oxides >4%). It is
obvious that the analysis of Stoner and Baumgartner (1981) yielded a set of
information, which made it possible for a precise classification of the soil from their
reflectance spectra.

33
 IMPROCE 2009

Figure 6. Specular and volume reflectance in soils due to moisture content.

Sensing Problematic Soils

We are very much aware of the fact that swelling soils/expansive clays and clay-
shales cause billions of rupees of damage every year. Chen (1988) showed the
distribution around the world of reported instances of heaving, indicating that the
problem of expansive soil is widespread throughout the five continents. The cost of
post-construction mitigation and standard engineering soil tests for creation of regional
maps are immense. Smectite is the clay mineral group that has the greatest swelling
potential and is responsible for most of the severe swelling soil damage observed in
certain parts of the world.
To demonstrate the potential of remote sensing in the identification and mapping
of swelling clays, Chabrillat etc al (2002) acquired data sets from 1997 to 1999 with the
Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) and the Hyperspectral
Mapper (HyMap). Using a matched filtering algorithm, maps of exposed clay material
were produced despite a strong vegetation cover. Among those exposures, spectral
discrimination and identification of variable clay mineralogy such as smectite,
smectite/illite, and kaolinite, in decreasing order of swelling potential hazard, was
possible. This work demonstrated that, using recent instruments and well-established
methodologies, imaging spectrometry can be of practical help for the detection and
mapping of expansive clays.
The main limits to successful mapping are the mixing at the sub-pixel scale and
the degree of exposure of the target at the surface, although those limits have been
pushed forward in recent years by the development of new processing algorithms such
as the spectral mixture analysis and the improvement of the signal-to-noise ratio (SNR)
of the instruments, along with better spectral and spatial resolution.
Case study

Youssef (2007) used Landsat Thematic Mapper image along with field and
laboratory investigation to map the Pliocene clay in the East Sohag Area, Egypt. These
deposits have high swelling potential and hence they have a negative impact on the
urbanization expansion in the flood plain of Nile River. Figure 8 shows the satellite
image and the distribution of the clays in the area. Principal Component and Minimum
Noise Fraction techniques with the help of supervised classification were successful in
34
 IMPROCE 2009

the mapping of these deposits. The Pliocene clay ranges from silty clay to clay with
small amounts of sand. Laboratory analysis (Atterberg limits, saturation degree, and
CEC) indicate the swelling characteristics of the Pliocene clay. Different classification
systems were applied and all reveal the expansive nature of this soil, which should be
taken into account for future development in the area.

Figure.8 Color composite of the Principal Component Analysis 1, 2, and 3 in RGB

respectively for an area adjacent to Nile river in which the Pliocene clay appear in
light blue (a) image showing the whole area; (b) El Kawther area; (c) El-Diabat–El
Ahaywa area;and (d) Qasab–Abu Nafokh area.

Soil Salinity

Soil salinity is the principal cause of soil degradation and is one of the major
factors affecting biomass production. Saline soils are located mostly in arid and
semiarid regions, apart from being found in sub-humid and coastal zone areas. Na2CO3,
NaHCO3 and NaCl are the salts reported in soils. These are very soluble and mobile
components in the soil environment. Typically, saline soils are highly erosive and have
poor structures, low microbial activity, and other attributes not conducive to plant
growth.

The spectral signature of saline soils can be a result of the salt itself, or
indirectly, from other chromophores related to the presence of the salt (e.g., organic
matter, particle size distribution). Although the spectra of salt is featureless, Hick and
35
 IMPROCE 2009

Russell (1990) raised a hypothesis that there are certain wavelengths in the VNIR-
SWIR region that can provide more accurate information about saline- affected areas.
For examples, a laboratory spectra of mixtures of SiO2 and NaCl + MgCl2 showed
significant features associated with two of the water absorption bands (at around 1.4 and
1.9μm, figure 9.b), due presumably to the high affinity of the salt to water molecules.
Hirschfeld (1985) was able to predict the NaCl concentration in a water solution based
on four features in the SWIR region at 1.63, 1.70, 2.08, and 2.17 μm.).

Table. Physio-chemical and engineering characteristics of the

Pliocene clay in East Sohag Area, Egypt.

Vegetation is an indirect factor that facilitates that detection of salts in soil using
reflectance measurements. Gausman et al., (1970), for example, pointed out that cotton
leaves grown in saline soils had a higher chlorophyll content than that of leaves grown
in low – salt soil. The major influence of salt is on the structure of the upper soil
surface. In figure 9(a), saline and non-saline spectra in the VNIR region are given. As
can be seen, the saline soils are relatively higher in albedo than the non-saline soils. The
saline soils have crusted surfaces that tend to be smoother than the generally rough
surfaces of the non-saline soils. Although similar trends have been reported in other
soils, it should be noted that in relatively high salt content soils, the opposite behaviour
is also reasonable. This is due to the fact that salt is a very hygroscopic material which
tends to decrease the soil albedo as water content increases. Because sodic soils have no
direct significant spectral features in the VNIR-SWIR region, it is difficult to identify
such soils. Salt in water is most likely to affect the hydrogen bond in water molecules,
causing subtle spectral changes. This fact is supported by many scientists who report
that characterizing the salinity status in soils was feasible with high-resolution
laboratory spectra.

A relatively high number of spectral channels are required to identify an indirect

relationship between salinity and other soil properties that appear to consist of
chromophores in the VNIR-SWIR regions. Csilag et al. (1993) analyzed high-resolution
spectra from about 290 soils in the United States and Hungary against chemical
parameters, including clay and organic matter content, pH, and salt. They state that
36
 IMPROCE 2009

because salinity is such a complex phenomenon, it cannot be attributed to a single soil

property.

Fig 9(a) VNIR spectra of saline and non-saline Fig.9(b) SWIR spectra of Na and Mg chlorides
soils as measured in the field. (Everitt et al., with silica mixtures. (Hick and Russel, 1990).
1998).

Case study

The tsunami which struck Nagapattinam in south India during December 2004
greatly affected agriculture and shrimp culture in the district. The waves were reported
to be as high as 3 - 10meters in southern India and penetrated 300m to 3km inland. The
tsunami left behind a trail of devastation and a large area was inundated in the process.
Immediate relief will be supported by knowing the extent of area inundated. The sea
water which came inland further made the soil saline. To identify saline affected area
for effective rehabilitation, remote sensing techniques were resorted to. For this
purpose, the pre tsunami Landsat ETM+ and the post tsunami ASTER images were
made use of by Shilpa (2005). Table 1 lists the wavelength bands of the two images
used. Figure 10 shows that saline soils are highly reflective in the bands 4 and 6 of
ASTER SWIR. Further, vegetation is highly absorptive in the band 2 of ASTER VNIR
and so also is water. Hence, use of band 2 of ASTER VNIR will enable to bring about a
distinction between vegetation, water and salt affected soils.

The pre- and post-tsunami images were then fused. Image fusion is a technique
which combines two digital images of different spatial resolutions, acquired in different
wavelength regions. The resulting image contains spatial information equivalent to the
finer resolution image and spectral information from both the image sets. Because it
was desired to know the extent of salt affected soils prior to and after the tsunami,
spectral bands 3, 5, and 7 of the ETM+ image were fused with the band 2 of ASTER
image. The resulting maps indicated a large area of salt affected soils post-tsunami as
compared to the pre-tsunami soil maps.

37
 IMPROCE 2009

Table 1. Spectral ranges of ETM+ and ASTER sensors

ETM+ (Spectral resolution) ASTER (Spectral resolution)

B1 0.45-0.515μm
B2 0.525-0.605μm B1 0.52-0.6μm
B3 0.63-0.69μm VNIR B2 0.63-0.69μm
B4 0.75-0.9μm B3 0.76-0.86μm
B5 1.55-1.75μm B4 1.6-1.7μm
B5 2.145-2.185μm
B6 2.185-2.225μm
B7 2.09-2.35μm SWIR
B7 2.235-2.285μm
B8 2.295-2.365μm
B9 2.36-2.43μm

Salt affected soil

Good soil
Salt crust
Water
Vegetation

Figure.10. Spectral curves of vegetation, water and salt affected

soils in ASTER image. (After Al-Khaier 2003)

Iron in soils
Iron is the most abundant element on the earth as a whole and the fourth-most-
abundant element in the earth’s crust. The average Fe concentration in the earth’s crust
is 5.09 mass, and the average Fe3+/Fe2+ ratio is 0.53. Major Fe-bearing minerals in the
earth’s crust are the mafic silicates, Fe-sulphides, carbonates, oxides, and smectite clay
minerals. All Fe3+ oxides have striking colors, ranging among red, yellow, and brown,
due to selective light absorption in the visible range caused by transitions in the electron
shell. It is well known that even a small amount of iron oxides can change the soil color
significantly. A representative soil spectra with various amounts of total Fe2O3 is
presented in figure 12. The iron’s feature assignments in the VNIR region result from
the electronic transition of iron cations (3+, 2+), either as the main constituent (as in
iron oxides) or as impurities (as in iron smectite). Hunt et al. (1971) summarized the
physical mechanism that allows Fe2+ (ferrous) and the Fe3+ (ferric) to be spectrally
active in the VNIR region.
38
 IMPROCE 2009

Iron oxide content and species are strongly corrected with the soil weathering
process in both the short and long terms. Transformation of iron oxide in soil often
occurs during natural soil conditions. Hematite and goethite are common iron oxides in
soils, and their relative content in soils is strongly controlled by soil temperature, water,
organic matter, and annual precipitation. Hematite soils are reddish, and goethite soils
are yellowish brown. Their reflectance spectra also differ, as can be seen in figure 13.
Hematite (α-Fe203) has Fe3+ ions in octahedral coordination with oxygen. Goethite (α-
FeOOH) also has Fe3+ in octahedral coordination, but different site distortions, along
with oxygen ligand (OH), provide the main absorption features that appear near 0.9μm.

s
s c
c
s s
c c

(a (b (c
) ) )

Nagapatti
namm
Akkaraipett
ai

Vellar
river

(d (e (f
Highly salt Moderately salt Least salt affected Not affected
)
affected affected
) )
c – cloud, sc- shadow of
cloud

Figure 11. Soil salinity mapping using satellite images. (a) Original ETM+
multispectral image. (b) fused image. (c) Soil map from (b). (d) ASTER VNIR image.
(e) Fused image. (f) Soil map from (e).

39
 IMPROCE 2009

Figure 13. Spectra of representative iron oxide minerals in soils (After Grave et al., 1992)

Iron associated with the structures of clay minerals is also active in both the
VNIR and SWIR spectral regions. This can be seen in the spectral curves presented in
Figure 12. Based on the structural OH-Fe features of smectite in the SWIR region, Ben-
Dor and Banin (1990) were able to generate a prediction equation to account for the
total iron content in a series of smectite minerals. Stoner (1979) also observed a higher
correlation between reflectance in the region 1.55 to 2.32 μm and the iron content in
soils, whereas Coyne et al. (1989) found a linear relationship between total iron content
in montmorillonite and the absorbance measured in the spectral region 0.6 to 1.1μm.

Hyperspectral remote sensing, or imaging spectroscopy, consists of acquiring

images in many (>100) narrow, contiguous spectral bands (Goetz et al., 1985), thus
providing a continuous spectrum for each pixel, unlike multispectral systems that
acquire images in a few (<10) wide spectral bands. Hyperspectral images enable
spectral identification of minerals, rocks, or soils at the surface, with laboratory-like
reflectance spectroscopy at the remote sensing scale. Spectra from the Airborne Visible/
Infrared Imaging Spectrometer (AVIRIS) have been used to identify individual minerals
such as alunite, calcite, dolomite, kaolinite, and muscovite (e.g., Kruse, 1988), apart
from crustal/mantle rocks (Mustard & Pieters, 1987). It has also enabled the mapping of
units and subunits of carbonate, clay, and iron oxide minerals in sedimentary rocks
(e.g., Clark et al., 1992). The best results have been obtained from areas where the rock
surfaces are well exposed, little weathered, and the minerals occur in largely pure
concentrations.

Conclusions

This write-up has examined the extent to which remote sensing imagery can be
used for detection and mapping of soils in general, and problematic soils in particular. It
40
 IMPROCE 2009

is has been realised that for general information about soil types, multispectral satellite
image data can be used. For quantitative retrieval of soil parameters such as organic
matter, iron, moisture etc, detailed spectral measurements are needed. This can be
achieved by resorting to imaging spectrometry or hyperspectral remote sensing. While
predicting soil properties, analysis of one soil-type is simple and accurate. Predictive
models are not always applicable on several soil types. The other great limitation is the
presence of vegetation above soil, which hinders much of the soil-information
extraction process.

Hyperspectral images generate spectra equivalent to the laboratory spectra,

which could be used to detect and discriminate different clays, smectite, illite, and
kaolinite, related to variable swelling potential. It has also been observed that maps of
clay type derived from the imagery are accurate. The main limitations for the expansive
clay soils detection is in the case of a heavy vegetation cover (forest or green grass), and
when the reflectance of the soils is approximately 10% or less. Hyperspectral remote-
sensing methods, along with modest field and laboratory analyses, can facilitate the
construction of clay hazard maps, as long as the soils are adequately exposed.

REFERENCES

1. Baumgardner, M.F., L.F. Silva, L.L. Biehl and E.R. Stoner. (1985). Reflectance
properties of soils. Adv. Agron. 38:1–44.

2. Ben-Dor, E. and A. Banin. (1990). Near-infrared reflectance analysis of carbonate

concentration in soils. Appl. Spec. 44:1064-1069

3. Bowers, S.A., and A.J., Hanks, (1965). Reflection of radiant energy from soil. Soil
Science, 100: 130-138.

4. Chabrillat S, Goetz. A.F.H ., Krosley. L., and Olsen. H.W. (2002). Use of
hyperspectral images in the identification and mapping of expansive clay soils and
the role of spatial resolution. Remote Sensing of Environment. 82: 431–445.

5. Chen. F.H (1988). Foundations on expansive soils.. Elsevier, New York, NY

6. Clark, R.N., G.A. Swayze, and A. Gallagher, (1992), Mapping the Mineralogy and
Lithology of Canyonlands, Utah with Imaging Spectrometer Data and the Multiple
Spectral Feature Mapping Algorithm. Summaries of 3rd JPL Airborne Geosciences
Workshop, Volume 1: AVIRIS Workshop. JPL Publication 92-14, 11-13.

7. Coyne L.M., Bishop J.L., Scattergood.T., Bani.A., Carle. G., and Orenberg.J..
(1989). NIR correlation spectroscopy: quantifying iron and surface water in series of
variably cation-exchanged monmorillionite clays. In Spectroscopic Characterisation
of Minerals and their surfaces, ACS Symp. Ser. 415, L.M Coyne, S.W.S McKeever,
and D.F. Blake, eds. American Chemical Society, Washington, D.C., pp.407-429.

8. Csillag,F.,Pasztor,L. & Biehl,L.L(1993). Spectral band selection for characterization

of salinity status of soils. Remote Sensing of Environment. 43: 231-242.

41
 IMPROCE 2009

9. Gausman, H. W., Allen, W. A., Cardenas, R., and Richardson, A. J. (1970), Relation
of light reflectance to histological and physical evaluations of cotton leaf maturity,
Appl. Opt. 9:545-552.

10. Goetz, A. F. H., Vane, G., Solomon, J. E., & Rock, B. N. (1985). Imaging
spectrometry for Earth remote sensing. Science, 228, 1147–1153.

11. Hick, P. T., and Russell, W. G. R. 1990. “Some spectral considerations for remote
sensing of soil salinity,” Australian Journal of Soil Research, vol. 28, pp. 417-431.

12. Hirschfield, T. 1985. Salinity determination using NIRA. Applied Spectroscopy 39:
740-741.

13. Kruse, F. A., 1988, Use of Airborne Imaging Spectrometer data to map minerals
associated with hydrothermally altered rocks in the northern Grapevine Mountains,
Nevada and California: Remote Sensing of Environment, v. 24, no. 1, pp. 31-51.

14. Mustard, J. F., & Pieters, C. M. (1987). Abundance and distribution of ultramafic
microbreccia in Moses Rock dike: quantitative application of mapping
spectroscopy. Journal of Geophysical Research, 92, E617–E626.

15. Mustard. J.F and Jessica, M.S. (1999), Spectral analysis for earth science:
investigations using remote sensing data. In Remote Sensing for the Earth Sciences,
Manual of Remote Sensing, 3rd edn, vol.3, edited by A. N. Rencz (New York: John
Wiley), pp.251–305.

16. Ben-Do.E, Irons.J.R. and Epema. G.F Soil (1999). Soil Reflectance. In Remote
sensing for the Earth Sciences: Manual of Remote Sensing, 3rd Ed. Vol 3, edited by
Rencz. A.N. John Wiley and Sons Inc. New York.

17. Shilpa.S. (2005) Recent approaches to multi-sensor image fusion. Unpublished M.S
(By Research) thesis. Anna University. Chennai.

18. Stoner, F.R. (1979). Physicochemical, site, and bidirectional reflectance factor
characteristics of uniformly moist soils. Ph.D. dissertation, Purdue Univ., West
Lafayette, Indiana.

19. Stoner.E and Baumgardner.M (1981). Characteristics Variations in Reflectance of

Surface Soils. Soil Society of America Journal 45, 1161-1165.

20. Youssef. A.M. (2007) , Mapping the Pliocene Clay Deposits Using Remote Sensing
and its Impact on the Urbanization Developments in Egypt: Case Study, East Sohag
Area. Geotech Geol Eng (2008) 26:579–591.

42
 IMPROCE 2009

FIELD AND LAB TECHNIQUES FOR DETERMINATION OF DYNAMIC SOIL

PARAMETERS

Dr.A.Boominathan *

1. INTRODUCTION

The measurement of dynamic soil properties is a critical task in the dynamic

analysis of foundation for various buildings and structures subjected to vibratory and
seismic loads. In general, problems involving the dynamic loading of soils are divided
into small and large-strain amplitude responses. The strains in the soil are usually small
for machine foundations. Strong motion earthquakes and nuclear explosions can
develop large strain amplitudes of the order of 0.1% strain whereas some precision
equipment such as electron microscopes are sensitive to very small amplitude vibrations
as low as 0.0001% strain. Many of the available techniques are oriented towards
measurement of low strain properties and many others toward properties mobilized at
large strains. The selection of testing techniques for measurement of dynamic soil
properties requires careful consideration and understanding of the specific problem at
hand.

This note presents the basic concepts of dynamic behaviour of soils. The various
parameters used to define the dynamic response of soils and its evaluation through field
and laboratory tests are also discussed. The evaluation of the dynamic parameters of
single piles and pile groups are also discussed.

2. BASIC SOIL PROPERTIES

The dynamic behaviour of the soils is defined by the following soil parameters:

• Shear wave velocity

• Maximum shear modulus Gmax or Young’s modulus E, Poisson’s ratio
• Variation of modulus and damping with strain
• Barkan’s elastic constants
• Attenuation coefficients.

3. FIELD DYNAMIC SOIL TESTS

A number of tests can be performed in the field to measure soil properties under
in situ conditions. These tests allow the effects of in situ stress, thermal, chemical and
structural states which can be destroyed by sampling to be reflected in the measured
properties. Field tests often sample large volumes of soil and do so quite
economically.Field tests most commonly employed in geotechnical engineering can be
divided into those that measure low-strain properties and those that measure properties
at intermediate to high strains. The following tests are actually used to determine in situ
dynamic soil properties at various strain levels.

1. Seismic cross hole test

2. Seismic down-hole (up-hole) test
* Professor, Department of Civil Engineering, IIT Madras, Chennai - 36 43
 IMPROCE 2009

3. Seismic refraction test

4. Spectral analysis of surface waves (SASW) test
5. Multichannel Analysis of surface waves (MASW) test
6. Seismic cone test
7. Standard penetration test
8. Block vibration (Resonance Footing) test
9. Cyclic plate load test

3.1 Seismic cross-hole test

Low-strain tests that require borings include down-hole test, up-hole test, cross
hole test and suspension logger test. Among these methods the cross-hole test is widely
used to measure the shear wave velocity profile particularly at deep soil sites. The cross-
hole test makes use of more than one bore hole. A vibratory source is placed in one
boring and the receiver is placed at the same depth in each of the other bore holes. An
impulsive disturbance is applied at the source and the travel times to each of the
receivers is measured. The wave propagation velocity is measure by knowing the
distance between the source and the receiver. The cross-hole test is the best method to
measure the shear wave velocity of the strata at any depth and it can detect the low
velocity layers and it is widely used in earthquake engineering application. The cross-
hole test can be used to measure the wave velocity within individual soil layers. This
test is considered by many engineers to be the most reliable method of determining
dynamic shear modulus of soil. The test determines shear and compression wave
velocity profiles with depth. Other parameters, such as poisson’s ratios and moduli can
also be easily determined from the measured shear and compression velocities.

This test involves the creation of transient and/or steady state stress waves and
the interpretation of their behavior from measurements made at one or more different
locations. A source capable of generating shear and compression waves is lowered in
one of the boreholes, and a 3-component geophone/accelerometer receiver (clamped to
the casing walls) is lowered to the same depth in another borehole to sense the passage
of shear and compression waves.

Cross-hole tests will be carried out at particular location using three boreholes
(one source borehole and two receiver boreholes) aligned in line with each other. A thin
walled PVC casing will be installed in the boreholes to maintain the borehole integrity.
The gap between the boreholes and PVC casing will be grouted with bentonite or dry
sand to eliminate any air gaps between the PVC casing and the boreholes. The
boreholes will be surveyed using a suitable device for determining verticality. Typical
distances between adjacent boreholes are 3 to 5 m. Fig. 1 shows a field setup for
crosshole test measurements (Boominathan 2004).

The receiver boreholes are drilled to the total investigation depth. The source
borehole is advanced during testing at 1.5 m intervals (as per ASTM D 4428). If a
source containing an impactor and can be clamped to the total investigation depth prior
to testing. The vertical component of the receiver is used to capture the vertically
propagating shear waves (SV). The radial component senses the propagating
compression waves (P) and the tangential component senses the horizontally
propagating shear waves (SH). The hammer input and the receiver outputs are recorded
by a digital oscilloscope with plotter. The source borehole is advanced the next
44
 IMPROCE 2009

measurement depth and the process is continued until all desired measurements are
taken.

It is important that a seismic source is capable of generating waves of

predominantly one type and that is repeatable at each required energy level. Usually a
downhole hammering device and/or blows on a SPT hammer on a split spoon sampler
and also explosive charges are used to generate P and S waves in the source borehole.

Fig. 1 The seismic cross-hole test setup

The following dynamic elastic parameters will be calculated as follows:

(Vp / Vs ) 2 − 2
μd =
2( Vp − Vs ) 2 − 1
Dynamic Poisson’s ratio,

Dynamic shear modulus, Gs = ρ.Vs2

3(Vp / Vs ) 2 − 4

Dynamic Young’s modulus, Es = ρ.Vs2 (Vp − Vs ) − 1

2

where, ρ = the density of soil (kg/m3)

Vp = the velocity of the P-wave (m/s)
Vs = the velocity of the S-wave (m/s)
3.2 Up hole or Down hole wave propagation tests
Up hole and down hole tests are performed by using only one borehole (Fig. 2).
In the uphole method, the receiver is placed at the surface, and the shear waves are
generated at different depths within the borehole, while in the down hole method, the
excitation is applied at the surface, and one or more receiver are placed at different
depths with in the hole. Both the uphole and the downhole methods give average values
of wave velocities for the soil between the excitation and the receiver.
45
 IMPROCE 2009

Figure. 2 Seismic Downhole test setup

3.3 The seismic cone penetration test

This test combines the downhole method and cone penetration testing. To this
end, a small rugged velocity seismometer is incorporated inside the electronic
penetrometer during brief pauses in cone penetration testing (Fig. 3). A significant
advantage of the seismic cone is that with a single sounding test one obtains information
for the stratigraphy of the site, low strain moduli of the various layers, as well as the
(static) strength related parameters. A limitation of this method is that it may not be
appropriate for some types of soil such as those containing coarse gravel.

Figure. 3 Seismic cone test setup

46
 IMPROCE 2009

3.4 Seismic Refraction Survey (SRS)

The seismic refraction test involves measurement of the travel times of P-and S-
waves from an impulse source to a linear array of points along the ground surface at
different distances from the source. A typical test setup is shown in Fig. 4 an impulse
energy source, which can be mechanical or explosive, is located at or near the ground
surface (Krammer, 1996). A series of receivers, usually geophones, are placed in a
linear array. One receiver is located at the source. The output of all of the receivers is
recorded when the impulse load is triggered. From these recordings the arrival times of
the first waves to reach each receiver can be determined and plotted as a function of
source receiver distance, as shown in Fig. 4. If the receivers were all excited by a single
stress wave travelling along the ground surface at constant velocity, v (the test can be
performed and interpreted with p-or s-waves, so v could be vp or vs), the arrival time
distance plot would be a straight line of slope 1/v that passed through the origin, Fig. 4
Shows that this is clearly not the case-a different more complicated mechanism is at
work.

Figure 4. Seismic refraction test setup

The SRS has been made for general site investigation for larger projects. The
SRS evaluates the determination of elastic velocities of each layer, the thickness of each
layer, and the tip angle of each layer as large as the wave velocities increase in each
succeeding deeper loss.
The SRS energy large areas to be explosed economically but there is loss of
precision. The possible error is difficult to establish due to the wide varieties of possible
refraction paths.

3.5 Spectral Analysis of Surface Wave (SASW) test

Seismic reflection and seismic refraction tests are staples of conventional
geophysical exploration, and can provide information on subsurface layer thicknesses

47
 IMPROCE 2009

and propagation velocities without the need for soil borings. Nevertheless, these
techniques are not commonly used in geotechnical earthquake engineering practice
because they cannot detect soft layers below stiff layers and because s-wave velocities
can be more effectively evaluated using other techniques described below. A relatively
new technique that can be used to determine subsurface layer thicknesses and wave
propagation velocities, and also requires no borings, is the spectral analysis of surface
waves (SASW) test (Heisey et al., 1982; Nazarian and Stokoe, 1983; Stokoe et al.,
1994). In this test, two or more vertical receivers are placed on the ground surface in
line with a vibration source. The output of both receivers is recorded and transformed
into the frequency domain. The phase angles between the recorded responses can be
computed for each frequency in the transformation and used to compute an apparent
travel time of the surface waves from one receiver to the other. The surface wave phase
velocity can be computed as a function of frequency by knowing the distance between
the receivers. The variation of phase velocity with frequency (i.e., the dispersion) is a
function of the variation of stiffness with depth, and is therefore used for computing the
stiffnesses of the underlying soil layers. Because SASW profiling can be conducted
from the ground surface (i.e. without borings), can detect low-velocity layers and can be
used to considerable depth, it has seen increasing use in earthquake engineering
applications. A typical shear wave velocity profile obtained from SASW test is shown
in Fig 5.

Fig. 5 Profile of shear wave velocity in Pisa clay, as obtained from seismic tests
(Foti, 2003)

3.6 Multichannel Analysis of Surface Wave (MASW) test

The Multi channel Analysis of Surface Waves (MASW) test is an advanced and
a faster technique than SASW which enables to determine 1D and 2D shear wave
velocity profiles. MASW is a nondestructive seismic method to evaluate thickness of
soil strata and its shear wave velocity which indicates the stiffness of the strata. It
analyzes dispersion properties of certain types of seismic surface waves (fundamental-
48
 IMPROCE 2009

mode Rayleigh waves) propagating horizontally along the surface of measurement

directly from impact point to receivers. It gives this shear-wave velocity (Vs) (or
stiffness) information in either 1-D (depth) or 2-D (depth and surface location) format in
a cost-effective and time-efficient manner. The concept of MASW using multi-channel
seismic acquisition methods as commonly used for petroleum exploration has been
extended to estimate 1D shear wave velocities using a multi-channel approach of data
acquisition that permits the generation of a laterally continuous 2D shear wave velocity
cross-section. The MASW method derives Vs wave velocities for a layered earth model
by inverting Rayleigh wave phase velocities.

A typical 1D and 2D shear wave velocity profile obtained at a site in Chennai is

shown in Figure. 6.

Figure. 6 Shear wave velocity profiling for West Mogappair

(Uma Maheshwari et al, 2008a)

3.6 The Block Vibration test

A plain cement concrete block of M-15 concrete should be constructed in the

test pit of size 3 m x 6 m at the bottom and a depth preferably equal to proposed depth
of foundations. The test should be constructed above the ground water table. In case of
rock the test may be performed on the surface of rock bed itself. The size of the block
should be selected depending upon the sub-soil conditions. In ordinary soils it may be
1m x 1m x 1.5m and in dense soil it may be 0.75m x 0.75m x 1 m (IS:5249-1992). In
boulder deposits the height may be may be increased suitably. The block size should be
so adjusted that the mass ratio
49
 IMPROCE 2009

1− ε m
BZ = [( × )]
4 ρ ro 3 is always less than unity.

The concrete block should be cured for at least 15 days before testing.
Foundation bolts should be embedded into the concrete block at the time of testing for
fixing the oscillator assembly.

A typical block vibration test setup is given in Fig. 7. The mechanical oscillator
is mounted on the test block such that it generates purely vertical sinusoidal vibrations
and the line of action of vibratory forces passes through the centre of gravity of the
block. The mechanical oscillator was connected through a flexible shaft with DC motor
and speed control unit. Two acceleration pick-ups duly calibrated are mounted on the
block such that they sense vertical motion of the block. Choosing a suitable value of
angle of setting of eccentric masses, the oscillator was made to run at constant
frequency. Out put signals from pick-ups were monitored and recorded using Carrier
Frequency Amplifiers and Digital Storage Oscilloscope. The frequency of the oscillator
was then increased and the process was repeated. The tests were carried out for
minimum 3 different settings of eccentric masses. At each frequency, the oscillator was
made to run for sufficient time to record the steady-state response.

Figure. 7A typical block vibration test setup

Based on the Block resonance test data, the amplitude versus frequency curves is
plotted for each eccentricity. The frequency fnz, corresponding to the maximum vertical
displacement was taken as the natural frequency of the soil foundation block systems.
A typical amplitude vs. frequency curve obtained from block vibration tests carried out
at the 500MW Prototype Fast Breeder Reactor (PFBR) site at Kalpakkam is shown in
Figure 8 (Boominathan et al, 2000).

Coefficient of elastic uniform compression of soil

The coefficient of elastic uniform compression (Cu) of soil is given by the

following equation.

4π 2 f nz M
2

Cu = A
50
 IMPROCE 2009

where

fnz - Natural Frequency,

M - Mass of the block, exciter and motor
A - Contact area of the block with the soil.

From the values of Cu obtained for the test block of contact area A the value of
Cu1 for the foundation having contact area A1 may be obtained from the equation

A / A1
Cu1= Cu

The dynamic shear modulus G, can be established using the following equations

Cu = [1.13E / (1 −ν 2 )]*(1/ A )
G = E / 2(1 +ν )

Damping coefficient of soil

In case of vertical vibration test, the value of damping coefficient ε of soil is

given by the following equation

f 2 − f1
ε=
2 f nz

where,
Xm
f2, f1 - Two frequencies at which the amplitude is equal to the 2

Xm - Maximum amplitude
fnz - Frequency at which amplitude is maximum (resonant frequency).

3.7 Cyclic plate load test

The equipment for a cyclic plate load test is similar to that used in static plate
load test.

After equipment has been set up and arranged, the cycles of loading, unloading,
and reloading are continued until the estimated ultimate load has been reached. The
magnitude of the load increment is such that the ultimate is reached in five or six
increments.

The elastic rebound of the plate corresponding to each intensity of loading can
be obtained from the obtained during cyclic plate load tests. The load intensity versus
the elastic rebound is plotted. The value of coefficient of elastic uniform compression
Cu can be calculated from the following equation.

uniform compression ( p)
Cu = Elastic settlement ( S e )

51
 IMPROCE 2009

3.8 Standard penetration test

Imai (1977) developed a correlation between (uncorrected) N value and shear

wave velocity, Vs (m/s), in 943 recordings at four urban locations in Japan and
established the following relationship,

Vs = 91 N0.337

Prakash and Puri (1988) successfully applied the above relationship in

predicting dynamic soil properties at different depths.

Based on the shear wave velocity measured at various places in Chennai, a

correlation between SPT ‘N’ and shear wave velocity Vs was proposed for Chennai soils
(Uma Maheshwari et al, 2008b).

Vs = 95.64N 0.301 (r 2 = 0.83), All soils

Vs =100.53N 0.265 (r 2 = 0.84), Sandy soils
Vs = 89.31N 0.358 (r 2 = 0.93), Clayey soils

4. LABORATORY TESTS

The following laboratory tests are available to determine the dynamic properties
at various strain levels.

1. Resonant column test

2. Ultra sonic pulse test
3. Cyclic triaxial test
4. Cyclic direct shear test
5. Cyclic torsional shear test

Laboratory techniques such as the resonant column or cyclic torsional hallow

cylinder tests can be used to evaluate the shear modulus over a range of strain
amplitudes. The cyclic simple shear, cyclic triaxial tests can be used to measure the
liquefaction strength of loose saturated sands. The ultrasonic wave propagation test is
not used widely in soil dynamics but it is routinely used to measure the elastic
properties of rock. The result of laboratory techniques is mainly depends on the quality
of UDS samples and therefore these techniques are used for evaluation of dynamic
properties for major projects in India.

5 FACTORS AFFECTING DYNAMIC SHEAR MODULUS

Based on the study of dynamic elastic constants, the factors on which these
depend are (Prakash and Puri, 1988)
1. Type of soil and its properties (e.g. Water content and γ d ) and state of
disturbance.
2. Initial (sustained) static stress level or confining stress.
3. Strain level
4. Time effects
52
 IMPROCE 2009

5. Degree of saturation
6. Frequency and number of cycles of dynamic load.
7. Magnitude of dynamic stress.
8. Dynamic prestrain

6 SELECTION OF DESIGN PARAMETERS

6.1 Maximum Shear Modulus , Gmax

Since the seismic geophysical tests induce shear strain lower than about 3x10-4
%, the measured shear wave velocities can be used to compute Gmax as

Gmax =
ρ Vs 2

On the basis of analysis on the values of soil constants for the large amount of
data collected by Hardin, the maximum value of shear modulus, Gmax (at low shear
strain of 10-6) is expressed by the following equation (Hardin and Black, 1969).

(2.973 − e) 2
(σ o ) 0.5
Gmax = 1230 OCRk (1 + e)

Where,

OCR - over consolidation ratio

e - void ratio
k - factor depends on the plasticity index of clays
σo - mean effective confining stress in Psi, equals,
σ o = (σ x + σ y + σ z ) / 3
table gives values of K after Hardin and Drenevich, 1972

Plasticity index (PI) K

0 0.00
20 0.18
40 0.30
60 0.41
80 0.48
100 0.50

Other empirical relationship has been proposed for specific soil types, the maximum
shear modulus of sand can be estimated using the

Gmax = 1000 K2max (σ 0 )

1 0.5

Where K2max is determined from the void ratio and relative density and σ 0 in
1

lb/ft2 (Seed and Idriss, 1970)

53
 IMPROCE 2009

E K2max Dr % K2max
0.4 70 30 34
0.5 60 40 40
0.6 51 45 43
0.7 44 60 52
0.8 39 75 59
0.9 34 90 70

For fine-grained soils, preliminary estimates of the maximum shear modulus can
be obtained from plasticity index, over consolidated ratio and undrained strength using
the table prepared by Weiler (1988).

Over Consolidation Ratio,

OCR
Plasticity Index (PI)
1 2 5
15-20 1100 900 600
20-25 700 600 500
35-45 450 380 300

6.2 Correction for confining pressure

The value of the dynamic shear modulus G is affected by a number of

parameters out of which confining pressure, shear strain amplitude and relative density
are more important. In the range of strains associated with properly design machine
foundations, the effect of variation in strain on shear modulus is small and the values of
G for design purposes may be determined from the in-situ test values using the relation
given below.

G1 σ
= ( 01 ) m
G σ0

where,
G1 and G - Dynamic shear modulus for the prototype and from field
test Respectively
σ 01 , σ 0 - Mean effetive confining pressure, associated with
prototype foundation and the in-situ teat respectively.

M - Constant depending upon the type of soil, shape of grains,

etc. its value has been found to vary from 0.3 to 0.7 and
may on the average be taken as 0.5.

6.3 Strain level correction

In situations where high strain levels are associated as in case of analysis for
earthquake conditions, the effect of strain level shall be considered along with that of
confining pressure.

54
 IMPROCE 2009

In such case, the values of G from different field tests may be first be reduced to
same confining pressure (expected below the footing) and their variation with the strain
levels may be suited to obtain appropriate values corresponding to the expected strain
level.

6.4 Modulus reduction

The variation of modulus reduction (G/Gmax) with strain level is important for
carrying out ground response analysis and for selection of design parameters for
important foundations/structures subjected to dynamic loads. In order to plot G/Gmax vs.
strain, the modulus values determined from various dynamic tests can be converted to a
standard unit pressure of 1 MN/m2 using the following equation.
0.5
G 1 ⎛ σ o1 ⎞
=⎜ ⎟
G ⎝ σo ⎠

where, G1 and G are dynamic shear modulus for the pressure of 1 MN/m2 and
for test confining pressure σo1 respectively. Then the dynamic shear modulus G (for a
mean effective confining pressure of 1 MN/m2) versus magnitude of shear strain, γθ (G
vs. γθ) are plotted.

The normalized shear modulus (G/Gmax) values at any depth can be obtained
by dividing G value by the corresponding Gmax. This normalized shear modulus
(G/Gmax) vs. shear strain plot is also called as a modulus reduction curve. The modulus
reduction and damping curves for the Kalpakkam sand were obtained from various
dynamic field tests (Boominathan et al, 2000) and laboratory tests (Jaya et al 2008).
Commonly used damping curves for soil of different plasticity and for cohesionless
soils are shown in Fig. 8 (Vucetic and Dobry, 1991).

Fig. 8 Vucetic-Dobry model: (a) modulus reduction curves and (b) damping
curves.
7 DYNAMIC PILE PARAMETERS

Determination of group pile stiffness and damping is an important step in the

analysis of pile-supported structures subjected to dynamic loading due to earthquakes,
machinery etc. They are most often evaluated using the properties of a single pile and
are accounted for the group effect by means of the interaction factors.

The single pile stiffness and damping can be evaluated by the following two
approaches.

1. Theoretical methods
2. Insitu dynamic pile load tests

7.1 Estimation of dynamic pile parameters using theoretical methods

Elastic continuum model (Novak and El-Sharounby, 1982) and Beam-on-

Winkler-Foundation model (Gazetas et al, 1993) are widely used to determine the
55
 IMPROCE 2009

dynamic pile characteristics. PILAY and DYNA5 computer programs based on Novak’s
elastic continuum approach are currently used for this purpose. The maximum shear
modulus or shear wave velocity of soil layers is the input data for the above
programmes. The above dynamic soil properties is usually determined by conducting
seismic crosshole tests in the field.

7.2 Field Dynamic pile load tests

The Geotechnical Engineering Division of IIT Madras conducted in situ

dynamic tests on piles in various power plants and petro-chemical complex projects
throughout India, to find the natural frequencies of the soil-pile system, damping
coefficient and soil-pile stiffness under various modes of vibration (Boominathan and
Ayothiraman, 2006).

The lateral dynamic pile load tests are usually carried out on set of three piles as
per IS:9716-1981 “Guide for lateral dynamic tests on piles”. The vertical dynamic pile
load tests were carried out similar to vertical block resonance test as per IS: 5249-1992
“Determination of Dynamic Properties of soil - Method of Test”. A typical dynamic pile
load setup is shown in Fig. 9. A pile cap of 750 x 750 x 750 mm is casted above the pile
to fix the mechanical oscillator to induce vibrations.

Fig. 9 A typical dynamic pile load setup

Lateral Free vibration test

The lateral load was applied to the pile cap by rotating pulling screw duly
calibrated and then suddenly releasing the load with the help of clutch type attachment
(Fig. 10). The load cell of capacity 2 to 5 tonnes is used to determine the applied lateral
load. Natural frequency and damping coefficients was computed from the time history
of the acceleration.

56
 IMPROCE 2009

Fig. 10 Free vibration test setup

Lateral Forced vibration test

A mechanical oscillator mounted on the pile cap is used to generate the steady
state sinusoidal force in vertical and horizontal directions. Vibration response of piles in
both lateral and vertical directions is recorded from two HBM acceleration pickups
fixed on the pile cap. A Data Acquisition system consisting of carrrier frequency
system, Digital storage oscilloscope and plotter is used to monitor and record the time
history of accelerations. Using the free and forced vibration test data, dynamic force vs.
static amplitude can be plotted as per IS: 9716-1981. The slope of the above plot gives
soil- pile stiffness under dynamic loading conditions. A typical frequency response
curves and dynamic force vs. static amplitude plot obtained from dynamic pile load test
conducted at IOCL-Haldia refinery are shown in Fig. 11.

Fig. 11 A typical frequency response curves

57
 IMPROCE 2009

REFERENCES

1. ASTM D 4428 Standard test methods for seismic crosshole testing.

2. Arya, S, O’Neill, M and Pincus, G (1979) Design of structures and foundations for
vibrating machines, Gulf publishing co., Houston.
3. Boominathan, A. and Ayothiraman, R. (2006). “Dynamic Response of Laterally
Loaded Piles in Clay,” Geotechnical Engineering, ICE UK, Vol. 159, No. 3, pp. 233
- 241.
4. Boominathan, A. (2004). Seismic site characterization for nuclear structures and
power plants, Current Science, Vol. 87, 10, 25.
5. Boominathan, A. Koteswara Rao, P.C, Sivathanu Pillai, C and Hari, S (2000)
Measurement of Dynamic properties and evaluation of liquefaction potential of a
500 MW prototype fast breeder reactor in south of India, Proceedings of 12th world
conference on Earthquake Engineering, Auckland, New Zealand, Jan.30 - Feb.4,
2000 Vol.2, p.964 (1-8).
6. Foti, S. (2003) Small-strain stiffness and damping ratio of Pisa clay from surface
wave tests. Géotechnique, 53, 5, 455–461.
7. Gazetas, G (1991) Foundation vibrations, Foundation Engineers Handbook, ed. H.
Fang, Van Nonstrand Reinhold, New York.
8. Hardin, B.O and Black, W.L (1969), Cosure of vibration modulus of normally
consolidated clays, J. soil Mech. Foundation Divn, ASCE, 98 (SM6), 603-623.
9. Hardin, B.O and Drnvich (1972), Shear modulus and damping in soils, Design
equations and curves, J. Soil Mech. And Found. Divn, ASCE, 98 (SM 7), 667-692.
10. Imai, T (1977) Velocities of P and S waves in sub surface layers in ground in Japan,
Proc. of 9th Intl. Conf. Soil Mechanics and foundation, Tokyo, vol.2, 257-260.
11. IS:5249-1992 Determination of dynamic properties of soil – method of test
12. IS:9716-1981 Guide for lateral dynamic load test on pile
13. Jaya, V., Dodagoudar, G. R. and Boominathan, A. (2008). “Cyclic Behaviour of
Saturated Sands of Kalpakkam Region, Tamilnadu”, Indian Geotechnical Journal,
38 (2), 171-186.
14. Moore, P.J. (1985) Analysis and Design of Foundations for vibrations, Oxford and
IBH Publishing co.
15. Prakash, S. and Puri, V.K (1988) Foundations for Machines: Analysis and Design,
John Wiley Sons, New York.
16. Krammer, S.L (1996) Geotechnical Earthquake Engineering, Prentice Hall, New
Jersey.
17. Uma Maheshwari, R., Boominathan, A., & Dodagoudar, G.R., (2008).
“Development of empirical correltation between shear wave velocity and standard
penetration resistance in soils of Chennai” Proceedings of 14th World Conference on
Earthquake Engineering, Beijing, Paper No:04-01-0090.
18. Vucetic, M. and Dobry. R. (1991) Effect of soil plasticity on cyclic response,
Journal of Geotechnical Engineering, ASCE, 117, 89-107.
19. Weiler, W.A (1988). Small strain modulus of clay, Proc. ASCE speciality
conference on Earthquake Engineering and Soil Dynamics II:Recent advances in
ground motion evaluation, Geotechnical special publication.20, ASCE, New York,
331 pp.

58
 IMPROCE 2009

USE OF SWCC IN IDENTIFICATION OF PROBLEMATIC SOLIS AND

RECOMMENDATION OF SUITABLE FOUNDATION

Dr.K.Premalatha*

1.0 GENERAL

Terzaghi defined soil mechanics as the applications of the laws of mechanics

and hydraulics to engineering problems dealing with sediments and other
unconsolidated accumulations of solid particles produced by the mechanical and
chemical disintegrators of rocks, regardless of whether or not they contain an admixture
of organic constituents. In the above definition Terzaghi referred the saturated earthen
materials. The emerging appreciation of unsaturated soil in geotechnical engineering
practice requires refinement of Terzaghi’s basic definitions. The qualifier
“Unsaturated” leaves the same meaning as “partially saturated” and simply indicates
that the degree of pore water saturation in any value less than unity or a third phase of
matter is introduced into the two phase, saturated system.

To deal about unsaturated soil, apart from the principles of mechanics and
hydraulics, thermo dynamic principles describing equilibrium among gas, solid and
liquid phase, the transition of matter from one phase to another, and the adsorption or
desorption of one phase of matter from or onto an adjacent phase of different matter.
The force and energies associated with these multiphase interactions by their vary
nature separate unsaturated soil behavior from saturated soil behavior. In many practical
problems, where the hydrologic and stress-stain behavior of natural or engineered
system comprised of soil is strongly influenced by the presence absence or changes in
there interfacial, the traditional soil mechanics frame work often fails to satisfactorily
describe or predict the behavior of the system.

It is convenient to classify the various geotechnical engineering problem

involving unsaturated soil into there general phenomena, specifically, flow phenomena,
stress phenomana and deformation phenomena. Majority of practical engineering
problems generally involve all these phenomena concurrently and in coupled fashion.

1.1 Flow Phenomena

Flow phenomena requires mainly the application of hydraulics and interfacial

physics principles capillary potential and capillary conductivity is the driving force and
controlling material variable for capillary fluid flow. The pore water potential resulting
from dissolved chemical species, gravity, capillarity and short range of physio chemical
effects occurring at the solid liquid phase interface is known as chemical potential.
Many seepage related problem are effectively treated through the application of thermo
dynamic potential theory with little or involvement of solid mechanics.

1.2 Stress phenomena

The problem that requires consideration of both mechanical and chemical

problems equilibrium are classified as stress phenomena. The theories for describing the
* Assistant Professor, Department of Civil Engineering, Anna University Chennai, Chennai - 25 59
 IMPROCE 2009

state of stress and failure in unsaturated soil require consideration of the thermodynamic
properties of the pore water in terms of soil suction. Material variables such as grain
size and grain size distribution, state variable such as the degree of saturation and the
consequent inter particle force such as suction – induced effective stress suction stress.

1.3 Deformation Phenomena

Physical process characterized by large deformation or strain are classified as

deformation phenomena. In unsaturated soils, these deformations are very often caused
or governed by changes in the moisture condition of the soil. Important deformation
phenomena include compaction, multiphase consolidation compressibility and
collapsing soil behavior. The most unsaturated soil deformation phenomena are that of
swelling or shrinking soils. The inherent coupling between volume change, pressure
generation and moisture transport in expansive soil demonstrates the importance of the
combined roles of deformation, stress and fluid flow phenomena.

2.0 SOIL WATER CHARACTERSTIC CURVE

SWCC describes the constitutive relationship between soil suction and soil water
content. The general shape of SWCC for various soils reflects the dominating influence
of material properties including pore size distribution, grain size distribution, density,
organic material content, clay content and mineralogy on the pore water retaining
behavior. Understanding the general behavior of SWCC and its relationship to the
physical properties of soil that it describes is one of important critical component of
unsaturated soil mechanics.

Several methods are available for obtaining the SWCC for a particular soil.
SWCC may be determined directly or indirectly in the laboratory. Direct methods
include pressure plate, Bunchner funnel, Tensiometer and pressure membranes. These
methods measure the pore water pressure in the soil or impose a known air pressure to
soil and allow the water content to come to equilibrium with the imposed air pressure.
Among these methods conventional Pressure plate extractor (ASTMD 6836) is the most
common method. Indirect methods use measurement or indictor of water content or a
physical property that is sensitive to changes in water content. These experiments are
costly and time consuming. Several empirical equations have also been proposed to
obtain the SWCC from basic soil properties.

3.0 Expansive Soil

Expansive soils and their problems have been the focus of research from the
1950’s and several solutions are formal scattered in literature. These soils are mostly
found in arid and semi arid area and exhibit volumetric changes. The factors that
influence the shrink-swell potential are considered in three different groups. They are
the soil characteristics that influence the basic nature of the internal force field, the
environmental factors that influence the changes that may occur in the internal force
system and state of stress. All there factors are listed in Table 1.

60
 IMPROCE 2009

3.1 Identification of Expansive Soils

In engineering practice, the common identification scheme are based on standard

classification results, such as grain size analysis and Atterberg limits. There are number
of other methods are available and also provide valuable additional information.

Table 1. Factors that influence shrink – swell potential

Soil Properties Environmental Conditions Stress Conditions

Microscale factors Initial moisture condition Stress history
Clay mineralogy In-situ condition
Moisture variation Loading
Soil water chemistry • Climate Soil Profile
Soil Structure and fabric • Ground water
• Drainage and man made water
Soil suction sources
Macro scale factors • Vegetation
Plasticity • Permeability
• Temperature
Dry density

3.2 Classification of Expansive soil

The parameters determined from expansive soil identification tests have been
combined in a number of different classification schemes. Unfortunately, there has not
yet evolved a standard classification procedure, and a different scheme is used in
practically every different location.

The most confusing aspect of expansive soil classification is the lack of a

standard definition of swell potential. Not only do sample conditions vary in the
different swell tests used to evaluate swell potential but surcharge loading and other
testing factors vary over a wide range of values. Potential swell is the equilibrium
vertical volume change from an oedometer – type test, expressed as a percent of the
original height, of an undisturbed specimen from its natural water content and density to
a state of saturation under an applied load equivalent to the in situ overburden pressure.
In all cases, however, the term swelling potential refers to the relative capacity for
expansion of the different soils. The amount of swell that may be realized in the field is
a function of the environmental conditions. This will include whether the soil is left in
place or recompacted. It is important therefore to take into account the fact that two
soils may have the same swelling potential, according to their classification, but exhibit
very different amounts of swell. Many classification schemes provide an “expansion
rating” to provide a qualitative assessment of the degree of probable expansion.
Expansion ratings may be something such as ‘high,’ “medium,” and “low,” or “critical”
and “non-critical.”

The classification schemes provide one or more of the following ratings :

i. Ranges of values for either probable percentage of volume change, or probable

swelling pressure.

61
 IMPROCE 2009

ii. A qualitative expansion rating, i.e., low, medium, high, and very high expansion
potential.
iii. Some other classification parameter such as CH or A6 (in the USC and
AASHTO schemes, respectively). These ratings are specific to the classification
scheme, and defined for each individually.

Some of the classification methods used are

• Unified soil classification system (USCS)

• American Association of State Highways and Transportation
• Based on SL and linear shrinkage
• Based on % clay, WL, N.
• Based on IP.
• Based on IP and WS
• COLE classification chart.
• Based on WL, IP and insitu suction .
• Mineralogical classification based on activity and CEC.
• Differential free swell
• Based on Activity and % clay.

3.3 Prediction Methods

Prediction methods are separated into three board categories. They are loosely
described as theoretical methods, semi empirical methods and empirical method.
Application of any method must rely on testing procedure and analysis techniques
developed within a sound theoretical frame work. The state of practice continue to
include empirical procedure for predicting the heave in expansive soil. These procedure
are valid only if they are issued within the bounds of soil type, environment, and energy
application for which they are developed. Some of the methods are listed as Table 2 .

Table 2. Methods of prediction of volume change

Name of Method Year

1. Double oedometer method 1969
2. Sales and Serratosa method 1957
3. Volumeter method 1961
4. Mississippi method 1975
5. Sampson, Schuster and budge’s method 1965
6. Noble method 1966
7. Sullivan and Mcclell and 1969
8. Holtz method 1970
9. Navy method 1971
10. Direct model method 1973
11. Simple oedometer 1973
12. O.S.B.R method 1973
13. Fredlund, Hasan and Filson’s method 1980

62
 IMPROCE 2009

3.4 WCC OF EXPANSIVE CLAYS

To understand the influence of the degree of expansion, the composition and

after bergs limits the SWCC of some of the soils are presented in table3, and 4 The most
main mineral present in the different soil are also presented in table 5

Table 3 Grain Size Distribution Results

Location of Grave Sand Silt Clay
Sample / Type of l (%) (%) (%)
Soil (%)
Anna Nagar - 04 42 56
K.K.Nagar - 07 37 49
Thiruneermalai - 04 25 71
Korattur - 09 39 52
Thiruvanmiur - 10 37 53
Tharamani - 10 32 58
Thirumullaivoyal 6 21 18 55
Marine Clay 1 40 35 24
Red Soil - 21 61 18
Silty Soil - 29 70 01
Bentonite - - - 100

Table 4 Atterberg Limits and BIS Classification

Location of WL WP WS B.I.S.
Sample / Type of (%) (%) (%) Classi
Soil -fication
Anna Nagar 70.0 31.0 9.0 CH
K.K.Nagar 57.0 26.0 12.0 CH
Thiruneermalai 78.3 34.3 11.7 CH
Korattur 60.0 25.0 11.0 CH
Thiruvanmiur 64.0 27.0 10.0 CH
Tharamani 61.0 22.0 10.0 CH
Thirumullaivoyal 60.0 25.3 10.0 CH
Marine Clay 52.5 15.0 11.5 CH
Red Soil 34.0 22.0 18.0 ML
Bentonite 325.0 90.0 - -

Table 5 Most Main Minerals Present in the Soil

Location of Sample / Minerals
Type of Soil
Anna Nagar Illite
K.K.Nagar Montmorillinite
Thiruneermalai Vermiculite
Korattur Kaolinite
Thiruvanmiur Quartz
Tharamani Mica
Thirumullaivoyal Carbonate
Bentonite Montmorillinte
Silty Soil Feldspar

63
 IMPROCE 2009

The relation between the suction and water content were obtained for some of
the samples are Shown in Figure 1 to 5

Figure 1. Suction Versus Water Content (Thirumullaivoil Clay)

Figure 2. Suction Versus Water Content RelationShip of Natural Soils

Figure 3. Suction Versus Water Content of Natural Soils

Figure 4. Suction Versus Water Content of Bentonite,

Kaolinite and Thirneermalai Clay
64
 IMPROCE 2009

Figure 5. Suction Versus Degree of Saturation of Chennai Expansive Clay

3.4.3 Soil Water Characteristics Curve

Some of these curves of the tested soil samples are shown in figures 1 to 5 From
these curves it is possible to identify the different zones in the SWCC as described by
Vanapalli.et al (1998). This procedure involves, drawing a tangent line through the
inflection point on the straight line portion of the soil water characteristics curve. Three
zones of the curve are (i) Boundary effect zone (ii) Transition zone (iii) Residual
unsaturation zone. All the tested soils have all the three zones but soil type and the most
mineral present in the soil significantly varied the range of these zones.

The range of these zones decreases as the coarser fraction increases. All the
tested soil except bentonite attained the residual unsaturated state. Increase in the
density of soil also decreases the range of these zones. The boundary effect zone is due
to the initial saturation of the soil and is not seen for partially saturated soil samples.
The slope of the transition zone decreased with increase in liquid limit, plasticity index
and percentage of clay. The co-ordinates of the point of inflection at three zones are
obtained and listed in table 6. Fig 1 shows the zones and coordinates.

3.4.4 Soil Water Characteristics Curve Parameters

Table 6 summarizes the estimated air entry and residual suction value based on
the construction procedure defined in the figure 1.

The air entry value of silt is comparatively lower than all other tested soil. The
water content at the air entry value of silt is slightly higher than that of the kaolinite,
because of the presence of certain percentage of plastic fines. The air entry value and
the respective water content of bentonite reflects the influence of mineralogy and hence
the plasticity index and liquid limit of the bentonite. The plasticity index and percentage
of fines are higher for marine clay than the red soil, hence the air entry value and
respective residual suction values are higher than red soil.

65
 IMPROCE 2009

Table 6 SWCC parameters

Ψr Ψi wi wr
Location
kPa kPa (%) (%)
68.0
Anna Nagar 950 106.0 35.0
56.0
K.K.Nagar 690 100.0 27.0
58.0
Korattur 560 76.0 32.0
54.8
Thirumullaivoyal 540 80.0 24.0
61.0
Thiruvanmiur 800 116 28.0
58.5
Tharamani 763 70.0 29.0
38.8
Red Soil 220 75.0 16.3
49.5
Marine Clay 390 82.0 23.5
39.0
Silt 200 68.0 9.00
25.3
Kaolinite 585 70.0 16.7
168.
Bentonite 1400 110.0 115.0
3

The plasticity index, liquid limit and percentage fine also influences the residual
suction and residual water content. Increases in plasticity index, liquid limit and
percentage fine increases the residual suction and residual water content.

3.5 Design of Foundation in Expansive soil

It is necessary to understand that all parts of a building will not equally be

affected by the swelling potential of the soil. Beneath the centre of the building where
the soil is protected from sun and rain the moisture changes are small and soil
movements the least. Beneath outside walls, the movement is greater. Damage to
buildings is greater on the outside wall due to soil movements.

These general types of foundations are considered in Expansive soil. They are

i. Structure that are kept isolated from the swelling effects (under-reamed pile)
ii. Designing of foundations that will remain undamaged in spite of swelling –
(Stiffened raft)
iii. Elimination of swelling potential of soil – (Stabilization of soil)

All three methods are in use either singly or in combination but the first are by
far the most wide spared. The granular fill provided around the shallow foundation
mitigates the effects of expansion of the soils.

Different angular distortion depending on the behavior of super structure system

is recommended as follows

Behavior Tolerable deflection / Length ratio.

Rigid (Precast, reinforced) 1/1000.
Semi (ligid brick referred) ) 1/500 to 1/1000
Flexible (wood) 1/500.

66
 IMPROCE 2009

4.0 Dispersive Soil

Dispersive soils are clayey soils which are highly susceptible to erosion in
nature and contain a high percentage of exchangeable sodium ions, Na+. These soils can
erode rapidly even under negligible seepage velocities by segregation of individual
particles (dispersed in suspension). The tendency for dispersive erosion in a given soil
depends on variables such as mineralogy and chemistry of the clay, as well as dissolved
salts in the water in soil pores and in the eroding water. The depressiveness of the soil is
mainly a function of the concentration of sodium ions with in the soil structure.
Dispersive clays have a preponderance of sodium cations, whereas ordinary clays have
preponderance of calcium, potassium, and magnesium cations in the pore water.
Dispersive ness seems to be a result of sodium adhering to montmorillonite rather than
the Na+ present in pore water.

4.1 Determination of Dispersive Potential Of Soil

The dispersive potential of the are identified based on the following tests.

1. Double hyudrometer test

2. Pinhole test
3. Exchangeable Sodium Percentage, (ESP)
4. SAR Sodium Absorption Ratio, (SAR)

4.4.1 Double Hydrometer Test

The SCS (soil conservation service) laboratory dispersion test, also known as the
double hydrometer test, was one of the first methods developed to assess dispersion of
clay soil. The particle size distribution is to be determined using the standard
hydrometer test in which the soil specimen is dispersed in distilled water with strong
mechanical agitation and a chemical dispersant. Another parallel hydrometer test is to
be conducted on a duplicate soil specimen, without mechanical agitation and without
chemical dispersant. The “percent dispersion” is the ratio of the dry mass particles
smaller than 0.005-mm diameter of the second test to the first test and expressed as
percentage. The chemical dispersion agent used in the test is Sodium
hexametaphosphate and sodium carbonate. The double hydrometer result of a soil
sample is shown in figure 6.

Figure 6. Double hydrometer curves for soil

67
 IMPROCE 2009

If the percent dispersion is less than 30%, the soil is Non-dispersive, greater than
50% it is Dispersive and for in between values the potential in Intermediate.

4.1.2 Pinhole Test

The pinhole (BS1377: Part 5:1990: Clause6.2) test represents a direct,

qualitative measurement of the dispersibility and consequent colloidal erodability of
clay soils. A 1.0-mm-diameter hole is to be pierced through a 25-mm-long by35-mm-
diameter cylindrical soil specimen. Distilled water is allowed to percolate through the
pinhole under heads of 50, 180, and 380 mm (hydraulic gradients of approximately 2, 7,
and 15), and the flow rate and effluent turbidity are to be recorded. The 50, 180, and
380-mm heads result in flow velocities ranging from approximately 30 to 160 cm/s at
hydraulic gradients ranging from approximately 2 to 15. It is important that the test is to
be conducted on soil at its natural water content to avoid the effect of drying. The
resistance to erosion of the specimen was measured by the turbidity of the discharged
water, the rates of flow and the final hole diameter. Figure 3 shows the laboratory setup.

The ASTM classification for Pinhole test describing the quality of the discharge
is linked in table7. The pinhole test result is to be further verified by dissecting the soil
samples as shown in figure 7.

(a) Non dispersive (b) Dispersive

Figure 7 Pinhole test setup and view of bisected sample

Table 7 . ASTM classification for Pinhole test results

Final Hole
Head Cloudiness at the end Dispersive
Flow through size
(mm) of flow Classification
Specimen(ml/s) (mm)
50 1.3 dark 1.8 D1, High
50 1.21 dark 1.6 D1, High
50 1 Moderately dark 1.5 D2, Moderate
380 1.65 clear 1.0 ND2, Non dispersive

68
 IMPROCE 2009

4.1.3 Exchangeable Sodium Percentage

The presence of exchangeable sodium is the main contributing chemical factor

for dispersive behavior. The basic parameter to quantify this effect is ESP
(Exchangeable sodium percentage). The cation exchange capacity and the percentage of
exchangeable sodium are determined based on relevant I.S Code. The ESP for
Dispersive soil is greater than 10 and Non dispersive soil is less than 7

4.1.4 Sodium Adsorption Ratio

The sodium adsorption ratio of soil is to be determined based on the standard

Procedure . If the SAR is greater than 10 then the soil is dispersive in nature. If SAR is
less than 6 the soil is Non-dispersive. For in between values the degree of dispersion of
Intermediate.

4.2 Identification of Dispersive Soil Based on Swcc

4.2.1 Drying Curve

Figure 8 Varying Water content with applied Suction

From figure 8. it is observed that with the increment of suction the water content
tends to decrease .Here the water content gets reduced to 10.1 % , 10.4% & 10.9% for
soil1, soil2 & soil3 respectively at the suction of 1300 kPa, whereas for the dispersive
soil its get reduced to for 20% for the same suction. This curve is redrawn for degree of
saturation Vs suction and the parameters were determined as shown in figure 8.The
parameters obtained from the desorption curve is listed in Table 8.

69
 IMPROCE 2009

Table 8 Typical Results obtained from the drying curve

Description Soil 1 Soil 2 Soil 3 Soil 4

Air Entry value kN/m2 4.6kPa 5.5kPa 7.5kPa 10 kPa
Residual saturation (Sr) 32% 29% 25% 22%
Residual water content (wr) 27.5% 24.5% 19.01% 16%
Residual suction (kN/m2) 200 250 290 400
Slope of de-saturation zone 1.47 1.55 1.67 2

The air entryvalue, residual water content, residual suction and residual
saturation obtained are different for different soils according their composition and
minerology. The air entry value of dispersive soil is lesser than non-dispersive soil.
Increase in degree of dispersion decreses the air entry value, the residual section, and
increased the residual water content. Increase in dispersion decreased the dewatering
potential of the soil. For higher degree of dispersion the slope of dewatering zone is less
than 1.5 and for non-dispersive soil it is greater than 2.

Dewatering Potential

Figure 9 shows the relation between dewatering potential and suction. It is

understood form the graph that the dewatering rate versus suction for the samples.
Dewatering can be done for the dispersive soil but the potential is lower than the non
dispersive soil. For a non-dispersive soil the dewatering potential increases with
increase in suction, where as for all dispersive soil the potential decreases. The more the
degree of dispersion the less is the dewatering potential.

4.2.2 Wetting results

The results obtained from the Capillary wettings is shown in figure 10.

Fig.9. Dewatering poterteal Fig.10. Capillary wetting curve

Absorption Potential

From the figure 10, it is observed that the the absorption rate of dispersive soils
are comparatively lower than non-dispersive soil. It is also observed that after 10 hours
of wetting, there was a change in the absorption rate and the rate decreased irrespective
of the nature of soil. Increase in degree of dispersion decreases the absorption rate.

70
 IMPROCE 2009

4.4 Foundation for Dispersive soils

Foundation Design in Dispersive soils in very difficult task. The results from
laboratory or field tests are to used to predict the likely settlement that may occur under
severe condition. In many cases, deep foundation, such as pile, piers etc, are used to
transmit foundation loads to deeper bearing strata below the dispersive soil deposit. In
cases where it is feasible to support the structure on shallow foundation in or above the
dispersive soils; the use of continuous strip footings may provide a more economical
and safer foundation than isolated footings. Differential settlement between columns is
to be minimized and a more equitable distribution of stresses may be achieved with the
use of strip footing.

4.5 Other Treatment Methods

A great variety of treatment methods were tried in world wide some of them are

1. Moistening and compaction technique, with either conventional impact, or

vibratory rollers for shallow depths upto about 1.5m and Vibroflotation, stone
columns and displacement piles for deeper depths.
2. Heat treatment to solidify the soil
3. Chemical stabilization with use of sodium silicate of 82% or gypsum of about
2% or injection of carbon dioxide.
4. Prewetting accompanied with pre loading / surcharging It is reported from field
that Prewetting with dynamic compacters treatment and prewetting with 2%
Sodium Silicate, (provide concentration that reduces the potential for settlement)
and Prewetting with surcharge were successful.

5.0 REFERENCES

1. Premalatha, K. “Prediction of Heave Using Soil Water Characteristic and Analysis

of Stiffered Raft on Expansive Clays of Chennai”. Ph.D thesis, Department Civil
Engg, Anna University, Chennai – 25. 2001

2. Suriya, Pa “Characterisation of soil water characteristic curve and its use in

dewatering” M.E thesis, Department of Civil Engg, Anna University Chennai,
Chennai – 25. 2009.

3. Bhuvaneswari, “Quantification of the degree of dispersion of dispersive soil using

SWCC” M.E thesis, Department of Civil Engineering, Anna Univeristy, Chennai,
Chennai 25.2008

4. V.N.S Murthy, “Text book of Soil Mechanics and foundation Engineering.” CBS
Publishers and distributions, New Delhi. 2007 ISBN 81 – 239 – 1362 – 1.
www.google.com – Dispersive Soil.

5. J.D. Nelsan and D.J. Miller, “Expansive Soil” John wiley sons NJ 1992.
ISBN 0 – 471 – 18114 – 5.

6. W.Lu and W. Likos “Unsaturated Soil Mechanics” John wiley sons NJ 2004, ISBN
0 -471-44731-5.
71
IMPROCE 2009

72
 IMPROCE 2009

SHALLOW AND DEEP COMPACTION FOR MADE-UP

AND IN-SITU SOILS

Dr. V.K. Stalin*

1.0 INTRODUCTION

Many civil engineering projects require the use of soils as ‘fill’ material.
Whenever soil is placed as an engineering fill, it is nearly always necessary to compact
it to a dense state, so as to obtain satisfactory engineering properties which would not be
achieved with loosely placed material. Compaction on site is usually effected by
mechanical means such as rolling, ramming or vibrating. Compaction can be classified
as shallow and deep. For shallow compaction especially for made-up soil, rollers,
rammers and vibrating equipments are employed. Where as for deep compaction of in-
situ soils, the vibrofloation, deep dynamic compaction (dropping hammers) and basting
technique are adopted which depends largely on depth of in-situ soils to be improved,
influence of adjacent structures, soil condition etc., Control of the degree of compaction
is necessary to achieve a satisfactory result at reasonable cost. In this write up, the
details on various factors influencing compaction characteristics of soils, behaviour of
compacted soils and field compaction methods are briefly discussed.

2.0 PROCESS OF COMPACTION

Compaction of soil is the process by which the solid soil particles are packed
more closely together by mechanical means, thus increasing the dry density. It is
achieved through the reduction of the air voids in the soil, with little or no reduction in
the water content. This process must not be confused with consolidation, in which water
is sqeezed out under the action of a static load. The air voids cannot be eliminated
altogether by compaction, but with proper control they can be reduced to a minimum.
The effect of the amount of water present in a fine-grained soil on its compaction
characteristics, when subjected to a given compactive effort, is discussed below.

At low moisture content the soil grains are surrounded by a thin film of water,
which tends, to keep the grains a part when compacted (Figure.1a). The finer the soil
grains, the more significant is this effect. If the moisture content is increased, the
additional water enables the grains to be more easily compacted together (Figure.1b).
Some of the air is displaced and the dry density is increased. The addition of more
water, up to a certain point, enables more air to be expelled during compaction. At that
point the soil grains become as closely packed together as they can be (i.e the dry
density is at the maximum) under the application of this compactive effort (Figure. 1c).
When the amount of water exceeds that require to achieve this condition, the excess
water begins to push the particles apart (Figure.1d), so that the dry density is reduced.
At higher moisture contents little or no more air is displayed by compaction, and the
resulting dry density continues to decrease.

If at each stage the compacted dry density is calculated, and plotted against
moisture content, a graph similar to curve A in Figure 2. is obtained. This graph is the
moisture-density curve. The moisture content at which the greatest value of dry density
* Assistant Professor, Department of Civil Engineering, Anna University Chennai, Chennai - 25 73
 IMPROCE 2009

is reached for the given amount of compaction is the optimum moisture content (OMC),
and the corresponding dry density is the maximum dry density.

At this moisture content the soil can be compacted most efficiently under the
given compactive effort. The relationship between bulk (wet) density and moisture
content is shown by the dotted curve (W) in Figure.2. This type of plot is not generally
used, except perhaps as a guide during a compaction test before the moisture contents
are measured.

A typical compaction curve obtained form the British Standard ‘ordinary’

compaction test is shown in Figure.3. are curve A. If a heavier degree of compaction,
corresponding to the Standard ‘heavy’ compaction test, is applied at the each moisture
content, higher values of density, and therefore of dry density, will be obtained. The
resulting moisture density relationship will be a graph such as curve B in Figure. 3. The
maximum dry density is greater, but the optimum moisture content at which this occurs
is lower, than in the ‘ordinary’ test.

3.0 OBJECTIVE OF PROPER COMPACTION

Soils may be used as fill for any purpose, the must usual being:
1. To refill an excavation, or a void adjacent to a structure (such as behind a retaining
wall.)
2. To provide made-up ground to support a structur
3. As a sub-base for a road, railway or airfield runway.
4. As a structure in itself, such as an embankment or earth dam, including reinforced
earth
Compaction, by increasing the density, improves the engineering properties of
soils. The most significant improvements, and the resultant effects on the mass of fill as
a whole, are summarized in Table 1.
Table 1. Effect of Proper Compaction of Soils
Improvement Effect on mass of fill
Higher shear strength Greater stability
Lower compressibility Less settlement under state load
Higher CBR value Less deformation under repeated loads
Lower permeability Less tendency to absorb water
Lower frost susceptibility Less likelihood of frost heave

4.0 FACTORS INFLUENCING COMPACTION CHARACTERISTICS OF

SOILS
Following are the primary factors influencing compaction of soils.
4.1 Type of clay

Figure 4 shows the effect of clay content on the density and water content
relationship In general, higher the clay content (liquid limit) lower is the maximum dry
density and higher the optimum moisture content. However for sand there is no peak in
the proctor curve unlike clays (Figure 6).
74
 IMPROCE 2009

4.2 Compactive effort

Higher the compactive effort, higher is the maximum dry density and lesser is
the optimum moisture content as seen from the (Figure.5).

Figure 1 Representation of compaction of soil grains

Figure 2 Dry density – moisture contents Figure 3 Dry density – moisture curves
relationship for soils for various compactive effort

75
 IMPROCE 2009

Figure 4 Effect of clay content on density (Das 2006)

Figure 5. Effect of Compactive effort on dry unit weight Vs. moisture content
relation

76
 IMPROCE 2009

Figure 6 Proctor Compaction test results on a sand

(AASHTO test designation T-99).

4.3 Effect of Soil Structure / Water Content

Dry side of optimum assumes flocculated structure and wet side of optimum
assumes dispersed structure. Compressibility will be lower and permeability will be
higher on the dry side of optimum compare to wet side of optimum is primarily due to
effect of soil structure. In Figure.7 for example, the fabric at point C is more oriented
than at point A. Now, if the comparative effort is increased, the soil tends to become
more oriented, even dry of optimum. Again, referring to (Figure.7), a sample at point E
is more oriented than at point B, although the effect is less significant than dry of
optimum. This change is shown in the variation of permeability with molding water
content Figure.8a, where it can be seen that the permeability is about an order of
magnitude higher when this soil is compacted dry of optimum than when it is
compacted wet of optimum.
4.4. Organic content

In general organic content increases in soil the maximum dry density decrease
and OMC increases as seen in Figure 9a and 9b
5.0 ENGINEERING BEHAVIOUR OF COMPACTED COHESIVE SOILS

Table 2 is a summary of the efforts of wet versus dry of optimum compaction on

several engineering properties. The kind of compacting equipment or rollers used on a
job will depend on the type of soil to be compacted. Equipment is available to apply
pressure, impact, vibration, and kneading. A smooth wheel, or drum, roller supplies
100% coverage under the wheel, with ground contact pressures upto 380 kPa (55 psi)
and may be used on all soil types except rocky soils types except rocky soils. The most
common use for large smooth wheel rollers is for proofrolling subgrades and
compacting asphalt pavements. The pneumatic, or rubber-tired, roller has about 80%
coverage (80% of the total area is covered by tires) and tire pressures may be up to
about 700 kPa (100psi). A heavily loaded wagon with several rows of four to six closely
spaced tires is self-propelled or towed over the soil to be compacted. Like the smooth
wheel roller, the rubber-tired roller may be used for both granular and cohesive highway
fills, as well as for earth dam construction.
77
 IMPROCE 2009

Probably the first roller developed and perhaps the most common type of
compactor used today is the sheepfoot roller. This roller, as its name implies, has many
rounded or rectangular shaped protrusions of “feet” attached to a steel durm. Sheepfoot
rollers are usually towed in tandem by crawler tractors. Other rollers with protrusions
have also been developed to obtain high contact pressures for better crushing, kneading
and compacting of a rather wide variety of soils.

Figure 7. Effect of Compaction on soil structure (after Lambe, 1958)

Figure 8 (a) Change Permeability with molding water content (after Lambe, 1958)

78
 IMPROCE 2009

Figure 8 (b) Charge in compressibility with molding water content

(after Lambe, 1958)

79
 IMPROCE 2009

Figure 9 (a) Maximum dry unit weight Vs. Organic content for
all compaction test (DAS 2006)

Figure 9 (b). Effect of drying history and organic content on optimum moisture
content (Das 2006)

80
 IMPROCE 2009

Table 2. Compaction of soil properties between dry of optimum and wet of

Optimum compaction

Properties Comparison
1. Structure
a. Particle arrangement Dry side more random.
b. Water deficiency Dry side more deficient; thus imbibes more water, swells
more, has lower pore pressure.
c. Permanance Dry side structure sensitive to change
2. Permeability
a. Magnitude Dry side more permeability
b. Permanance Dry side permeability reduced much more by permeation
3. Compressibility
a. Magnitude Wet side more compressible in low pressure range, dry
side in high pressure range.
b.
Rate Dry side consolidates more rapidly
4.
Strength
a.
As molded:
b.
Undrained Dry side much higher
c.
Drained Dry side somewhat higher
a.
After saturation: Dry side somewhat higher if swelling prevented; wet
b.
Undrained side can be higher if swelling permitted
c.
Drained Dry side about the same or slightly greater
A.
Pore water pressure Wet side higher
at Failure
B. Stress-strain Dry side much greater
modulus
C. Sensitivity Dry side more apt to be sensitive
* After Lambe (1958)
Figure 10 summarises the applicability of various types of compaction
equipment as a function of soil type, expressed as a percentage of sand to clay. These
“zones” are not absolute, and it is possible for a given piece of equipment to compact
satisfactorily outside the given zone.
6.0 SHALLOW COMPACTION

The degree of compaction achieved in the field by a roller will depend on

several factors that include:
1. Thickness of lift
2. Area over which the pressure is applied
3. Intensity of pressure applied to the soil
4. Type of roller
5. Number of roller passes
The growth curve during the field compaction of a lean clay is shown in (Figure
11). It can be seen that the dry unit weight of the soil gradually increases with the
increase of the number of passes of a roller. After about eight to ten passes, the increase
in γd of the soil is rather negligible. In most cases, the maximum economically
attainable dry unit weight is achieved with about ten roller passes.
81
 IMPROCE 2009

Figure 10 Applicability of various types of compaction equipment for a given soil

type (modified after caterpillar Tractor Co., 1977)

Figure 11 Effect of number of passes on compaction of lean clay (After E.H Yoder
“Principles of Pavement Design,” Wiley, New York, 1959)

Vibratory rollers are particularly useful for compacting granular soils. Self-
propelled and towed vibratory roller of various sizes, weights, and vibration frequencies
are available. The vibrations are generally produced by rotating an off-center weight.
Vertical vibrations can be obtained by using two synchronized counter-rotating weights.
Figure 12 shows the compacted unit-weight profiles for 8-fit lifts of the dune sand for
2,5,15 and 45 roller passes. For field compaction work, the specification requires that
the granular soil be compacted to a certain minimum relative density at all depths.
Determination of the height of each lift depends on the type of roller and the economic
number of passes.
82
 IMPROCE 2009

6.1 Construction Control

The relationship between dry density and moisture content for soil subjected to a
given compaccitve effort, established by laboratory compaction tests, provides
reference date for the specification and control of soil placed as fill. On many projects
the laboratory compaction tests are supplemented by field compaction trials by using
the actual placing the compacting equipment which is to be employed for construction.

Sometimes it is necessary to adjust the natural moisture content of a soil to a

value at which it can be most effectively compacted, or at which it has the higher
strength. The required moisture content, and the dry density to be achieved, can be
assessed on the basis of the dry density – moisture content relationship derived form a
laboratory compaction tests on samples taken from the borrow area.

While compaction in situ needs to be of a sufficient degree to obtain the required

density, it is equally important not to over compact fine-grained soils. Overcompaction
not only is wasteful of effort, but should be avoided because over compacted soil, if not
confined by overburden, can readily absorb water, resulting in swelling, lower shear
strength and greater compressibility. Toes and sides of embankments are particularly
sensitive to this effect.

6.2 Design Parameters

When the compaction characteristics of a soil are known, it is possible to

prepare samples in the laboratory at the same dry density and moisture content as that
likely to be attained after compaction in the field. These samples can be subjected to
laboratory tests for the determination of their shear strength, compressibility and other
engineering properties. Design parameters derived from these tests enable the stability,
deformation and other characteristics of the fill to be assessed. They can also provide
the basis for the initial design of an embankment or earth dam.

More elaborate tests can be carried out on compacted samples to measure the
changes of pore pressure due to changing conditions of applied stress. During
construction, pore pressures can be monitored so as to ensure they do not at any time
exceed certain limiting values established by the tests. A specification for compacted fill
may require a certain ‘relative compaction’ (measured in terms of dry density) to be
achieved, within specified limits of moisture content. More usually a specification
defines the maximum air avoids permitted in the compacted soil within the so that air
voids lines can be added to the compaction test graphs.

83
 IMPROCE 2009

Figure 12. Compacted unit weight for 8ft (2.44m) lift height for 2,5,15 and 45
vibratory roller passes. (DAS 2006)

7.0 DEEP COMPACTION

Several special types of compaction techniques have been developed for deep
compaction of in-place soils, and these techniques are used in the field for large-scale
compaction works. Among these, the popular methods are vibraflotation, dynamic
compaction, and blasting. Details of these methods are provided in the following
sections.

84
 IMPROCE 2009

7.1 Vibroflotation
Vibroflotation is a technique for in situ densification of thick layers of loose
granular soil deposits. It was developed in Germany in the 1930s. The first
vibroflotation device was used in the United States about 10 years later. The process
involves the use of a Vibroflot (also called the vibrating unit), which is about 2.1m
(∼7ft) logn. (as shown in Figure 13) This vibrating unit has an eccentric weight inside it
and can develop a centrifugal force, which enables the vibrating unit to vibrate
horizontally. There are openings at the bottom and top of the vibrating unit for water
jets. The vibrating unit is attached to a follow –up pipe. Figure 13 shows the entire
assembly of equipment necessary for conducting the field compaction. The entire
vibroflotation compaction process in the field can be divided into four stages (Figure
14)
Stage 1 : The jet at the bottom of the Vibroflot is turned on and lowered into the
ground
Stage 2 : The water jet creates a quick condition in the soil and it allows the
vibrating unit to sink into the ground
Stage 3 : Granular material is poured from the top of the hole. The water from the
lower jet is transferred to the jet at the top of the vibrating unit. This
water carries the granular material down the hole.
Stage 4 : The vibrating unit is gradually raised in about 0.3 m (=1 ft) lifts and held
vibrating for about 30 seconds at each lift. This process compacts the soil
to the desired unit weight.
Table 3 Types of Vibroflot Units
Motor Type 75kW electric and 23kW electric
hydraulic
a.Vibrating tip
Length 2.1 m (7.0ft) 1.86m (6.11 ft)
Diameter 406 mm (16 in) 381 mm (15 in)
Weight 17.8 kN (4000 Ib) 17.8 kN (4000 Ib)
Maximum movement 12.5 mm (0.49 in) 7.6 mm (0.3 in)
when full
Centrifugal force 160 kN (18 ton) 89 kN (10 ton)
b. Eccentric
Weight 1.2 kN (260 Ib) 0.76 kN (170 Ib)
Offset 38 mm (1.5 in) 32 mm (1.25 in)
Length 610 mm (24 in) 390 mm (15.25 in)
Speed 1800 rpm 1800 rpm
C.pump
Operating flow rate 0-1.6m3/min (0-400 0-0.6m3/min(0-150
gal/min) gal/min)
Pressure 700-1050 kN/m2 (100-150 700-1050 kN/m2 (100-
Ib/in2) 150 Ib/in2)
d.Lower follow-up
pipe and extensions
Diameter 305 mm (12 in) 305 mm (12 in)
Weight 3.65 kN/m (250 Ib/ft) 3.65 kN/m (250 Ib/ft)
* After brown (1977)
85
 IMPROCE 2009

Figure 13 Vibrofloatation Techniques

Figure 14 Compaction process of vibroflotation

86
 IMPROCE 2009

The details of various types of Vibroflot units used in the United States are
given in Table 5.5. Note that 23kW (30-hp) electric units have been used since the latter
part of the 1940s. The 75kW (100-hp) units were introduced in the early 1970s. The
zone of compaction around a single probe varies with the type of Vibroflot used the
cylindrical zone of compaction has a radius of about 2m (=6 ft) for a 23kW (30-hp) unit.
This radius can extend to about 3m (=10ft) for a 75kW (100-hp) unit.

Compaction by vibroflotation is done in various probe spacing, depending on

the zone of compaction. This spacing is shown in Figure 15. The capacity for successful
densification of in situ soil depends on several factors, the most important of which is
the grain-size distribution of the soil and the type of backfill used to fill the holes during
the withdrawal period of the Vibroflot. The range of the grain-size distribution of in situ
soil marked Zone 1 in Figure 16 is most suitable for compaction by vibroflotation. Soils
that contain excessive amounts of fine sand and silt-size particles are difficult to
compact, and considerable effort is needed to reach the proper relative density of
compaction. Zone 2 in Figure 16 is the approximate lower limit of grain-size
distribution for which compaction by vibroflotation is effective. Soil deposits whose
grain-size distributions fall in Zone 3 contain appreciable amount of gravel. For these
soils, the rate of probe penetration may be slow and may prove un economical in the
long run.
The grain-size distribution of the backfill material is an important factor that
controls the rate of densification. Brown (1977) has defined a quantity called the
suitability number for rating backfill as

3 1 1
SN = 1.7 2
+ +
( D50 ) ( D20 )2 ( D10 ) 2

Where D50,D20, and D10, are the diameters (in mm) through which, respectively,
50,20, and 10% of the material passes.
The smaller the value of SN, the more desirable the backfill material. Following
is backfill rating system proposed by Brown (1977)
Range of SN Rating as backfill
0-10 Excellent
10-20 Good
20-30 Fair
30-50 Poor
>50 Unsuitable

7.2 Dynamic Compaction

Dynamic compaction is a technique that has gained popularity in the United

States for the densification of granular soil deposits. This process consists primarily of
dropping a heavy weight repeatedly on the ground at regular intervals. The weight of
the hammer used varies over a range of 80 to 360 kN (18 to 80 kip), and the height of
hammer drop varies between 7.5 and 30.5 m (2.5 and 100 ft). The stress waves
generated by the hammer drops aid in the densification. The degree of compaction
achieved at a given site depends on the following three factors :

87
 IMPROCE 2009

Figure 15 Probe spacing of vibroflotation

Figure 16 Range of Grain size distribution of soil for vibroflotation

1. Weight of hammer
2. Height of hammer drop
3. Spacing of locations at which the hammer is dropped

Leonards, Cutter, and Holtz (1980) suggested that the significant depth of
influence for compaction can be approximated by using the equation

D = (1 / 2) WH h

Where D = significant depth of densification (m)

WH = dropping weight (metric ton)
h = height of drop (m)

7.3 Blasting

Blasting is technique that has been used successfully in many projects (Mitchell,
1970) for the densification of granular soils. The general soil grain suitable for
88
 IMPROCE 2009

compaction by blasting are the same as those for compaction by vibroflotaton. The
process involves the detonation of explosive charges such as 60% dynamite at a certain
depth below the ground surface in saturated soil. The lateral spacing of the charges
varies from about 3 to 10 m (10 to 30 ft). Three to five successful detonations are
usually necessary to achieve the desired compaction. Compaction up to a relative
density of about 80% and up to a depth of about 20 m (60ft) over a large area can easily
be achieve by using this process. Usually, the explosive charges are placed at a depth of
about two-thirds of the thickness of the soil layer desired to compacted.

8.0 SUMMARY

Compaction of soil results in the overall improvement of made-up soil

especially for Construction of embankments, land reclamation projects, land fill design
etc. Shallow compaction equipments can be effectively used to achieve a required
density which in turn results is required design parameters such as bearing capacity,
settlement etc. On the other hand, it is also possible to densify the in-situ soil through
deep compaction equipments such as vibrofloat, blasting, deep dynamic compaction etc.
The achievement of desired properties as required in the design, is instantaneous in
compaction methods unlike other methods of Ground Improvement..

REFERENCES

1. Head K.H (1980), Manual of soil laboratory testing Vol.1. Soil Classification and
compaction tests, Pentech Press, London.

2. Seed HB, Mitchell J.K and Chan C.K (1960), Research Conference on Shear
Strength of Cohesive Soils, ASCE, Colorado, P 877 – 964.

3. Das, B.M (2006).Principals of Geotechnical Engineering, McGraw Hill Book Co.,

New York.

4. Sridharan, A (2000). “Compaction of soils. Proceedings of AICTE-ISTE Short

terms Training Programme on Ground Improvement Techniques, Mysore.

89
IMPROCE 2009

90
 IMPROCE 2009

VACUUM CONSOLIDATION AND BAND DRAINS FOR

HIGH COMPRESSIBLE CLAYS

Dr. R. G. Robinson*

1 INTRODUCTION

Geotechnical engineers very often encounter soft clays with very low shear
strength and high compressibility characteristics. The construction of infrastructure
projects in these clays is a challenging task, when the undrained shear strength is less
than 25 kPa, because of low bearing capacity. The settlements in these soils may extend
over a long period of time and hence any structure built on such soils need to undergo
constant repairs/remediation works. These soil deposits need to be stabilized before
commencing the construction activities so as to avoid the bearing capacity failures and
excessive settlements. There are different types of ground improvement techniques like
chemical stabilization, preloading, deep mixing, sand compaction piles, stone columns
etc.

Preloading methods are the most popular ground improvement techniques for
the treatment of soft clays. Preloading is the process of consolidating foundation soils
under an applied vertical stress prior to the construction of the proposed structure.
Usually the applied vertical stress, referred to as preload, exceeds the anticipated
foundation pressure. The preloading is conventionally achieved in the field in the form
of sand surcharge. The method of preloading can be used to eliminate all or part of the
post construction settlement. This may allow for economies in the choice of the
foundation system. The major beneficial effects of preloading are given in Fig. 1(a) and
(b). This method is environmental friendly, as it does not put any chemical admixtures
to the ground.

The rate of settlement of the clay deposits depend on the drainage path, which is
the maximum distance the water has to travel during the consolidation process. For
thicker clay deposits with low permeability, the time required for the required degree of
consolidation will be very large. Therefore, in order to reduce the consolidation time,
vertical drains are often used. The beneficial effects of vertical drains are illustrated in
Fig. 2. The vertical drains accelerate the settlements but do not reduce the final
movement.

In the past, sand drains consisting of boreholes filled with sand were extensively
used for accelerating the settlement. The holes were formed by driving, jetting or
auguring. Typical sand drains have diameters ranging from 200 to 450 mm with a
spacing of 1.5 to 6 m. The drains may be placed in triangular or rectangular pattern.
Currently, geosynthetic band drains in the form of strip (or band) are extensively used.
The band drains are of composite construction, a corrugated or studded inner core
wrapped in a filter fabric, normally a non-woven geotextile (Fig. 3). The drains are
typically about 100 mm wide and 2 to 6 mm thick. The advantages of Prefabricated
Vertical Drains (PVD) using geosynthetics are:

* Associate Professor, Department of Civil Engineering, IIT Madras, Chennai - 36 91

 IMPROCE 2009

• Easy, rapid installation possible

• Made of uniform material, easily stored and transported
• Equipment needed is lighter than the rigs used for equivalent sand drains
• Low cost compared to traditional sand drains

(a) Preloading reduces post construction (b) Preloading may allow

savings in settlements foundation costs

Figure. 1 Beneficial effects of preloading (after Hausmann, 1990)

Figure. 2 Beneficial effects of vertical drains (after Hausmann, 1990)

92
 IMPROCE 2009

Fig. 3 Typical core shapes of geosynthetic drains (Hausmann, 1990)

2 DESIGN CONSIDERATIONS

The main objective of preloading is to achieve the desired degree of

consolidation within the specified time. The design of preloading with vertical drains
requires the evaluation of drain and soil properties as well as the effect of installation.

In the one-dimensional consolidation, the settlement and pore water pressure

dissipation occur in the same direction. Terzaghi’s one-dimensional consolidation
theory may be used to analyse problems involving one-dimensional consolidation.
However, in the radial consolidation problems, the drainage occurs in the radial
direction while the settlement takes place in the vertical direction. In the vertical drain
problem, settlement may occur due to concurrent vertical and horizontal drainage. For
such cases, Carrillo (1962) has shown that

U = 1 − (1 − U v )(1 − U r ) . . .(1)

where U = average degree of consolidation for simultaneous vertical and radial

drainage
Uv = average degree of consolidation calculated on the assumption that only
vertical drainage exists
Ur = average degree of consolidation calculated on the assumption that only
radial drainage exists

The degree of consolidation under vertical drainage is well established in the

literature based on Terzaghi’s one-dimensional consolidation theory. A comprehensive
analysis of the radial drainage problem was presented by Barron (1948), who

93
 IMPROCE 2009

considered two fundamental cases such as free-strain case and equal-strain case. The
equal strain case is very widely used in practice.

Barron used the unit cell concept as illustrated in Fig. 4a and b. Fig. 4a shows a
typical triangular pattern of sand drains of diameter, dw, installed in a clay deposit of
thickness h. The zone of influence of each drain is hexagonal in plan. This hexagon can
be approximated as an equivalent circle of diameter de. The simplified equation for the
condition of equal vertical strain without smear is given as (Barron, 1948),

⎡ − 8Tr ⎤
U r = 1 − exp ⎢ ⎥
⎣ F ( n) ⎦ . . .(2)

cr t
Tr = 2
de . . .(3)
n 2
(3n − 1)
2
F ( n) = ln(n) −
and (n − 1)
2
4n 2 . . . (4)

in which, n is the drain spacing ratio given by

de
n=
dw . . .(5)

In Eqn. (3), cr is the coefficient of radial consolidation. The procedure for the
determination of coefficient of radial consolidation and experimental set-up required for
carrying out the experiments are discussed in the subsequent sections.

Cylindrical
Unit cell

Sand drain

dw
de

(a) (b)

Fig. 4 (a) Typical unit cell within a network of vertical sand drains and (b) Typical
unit cell
94
 IMPROCE 2009

)
3 DETERMINATION OF COEFFICIENT OF RADIAL CONSOLIDATION (cr

In order to use the theory for the economic design of vertical drains, a rational
assessment of coefficient of radial consolidation (cr) is essential. While many studies are
available for the determination of coefficient of consolidation under conditions of
vertical drainage, very few methods exist for the evaluation of radial coefficient of
consolidation. For the case of peripheral drain with free strain condition, McKinlay
(1961) observed a linear relationship between Ur versus Tr0.465 in the range of Ur
between 0 to 50%, where Ur is the degree of consolidation for the corresponding time
factor (Tr). With this as the characteristic feature, he suggested a method for the
determination of cr. For the case of central drain with free strain condition, Berry and
Tr
Wilkinson (1969) suggested the use of Ur versus plot to determine cr. The
characteristic feature used was a linear plot in the range of Ur from 20% to 60%. While
the above two methods are for the case of free strain conditions, Sridharan et al. (1996)
proposed a t method, where t is the time, for the determination of cr for the equal strain
T
condition. They observed a linear relation in the Ur versus r plot for 20% ≤ Ur ≤ 60%.
Guided by this, a plot of t versus the settlement S is used to determine cr. Another
simple graphical method is the log(S-So) - log t method (Robinson, 2009), where So is
the corrected zero reading. This method is based on an initial straight line portion
identified in the log Ur- log Tr plot, which when extended intersects the Ur =100 % line
for a time factor corresponding to a degree of consolidation of 66%.

If the degree of consolidation Ur (and hence Tr) corresponding to any real time t
from an experiment is known, then cr can be calculated using Eqn. (3). This can be
achieved using curve fitting procedures by identifying a characteristic feature of the
theoretical Ur-Tr relationship. The characteristic feature is identified in a plot of log Ur
versus log Tr. The variation of log Ur with log Tr is shown in Fig. 5 a typical values of n
= 5. The plot is linear, when the degree of consolidation is less than 20%. The initial
straight line portion of the log Ur versus log Tr plot is extended to intersect the Ur =
100% line. The intersection point (Point A) corresponds to a time factor of Tr = 0.126
and Ur = 65.9%, for n = 5. The degree of consolidation at the intersection point is found
to be approximately the same for all values of n (n ≤ 10). A value of Ur = 66% may,
therefore, be used for all practical purposes.

Sophisticated experimental set-up like Rowe’s cell (Rowe and Barden 1966) are
often used to perform radial consolidation tests. However, the conventional oedometer
cell may also be used after suitable modifications. A schematic of the conventional
consolidation cell, after modification, is shown in Fig. 6. The inner diameter of the
consolidation ring is 75 mm and its thickness is 25 mm. The top and bottom boundaries
of the clay were made impermeable by introducing rubber sheets of diameter equal to
the diameter of the consolidation ring and with a central hole equal to the diameter of
the sand drain. A sampling tube, having outer diameter of 15 mm and inner diameter of
14.5 mm, can be used to make the central hole so that n = 5. The hole needs to be filled
with clean fine sand. It is essential that the maximum particle size of the sand should be
of the order of 1/20th of the diameter of the drain (Shields and Rowe 1965). Incremental
load consolidation tests were conducted at a load increment ratio of 1.0. During the
consolidation process, the time-settlement data were recorded.

95
 IMPROCE 2009

From the time-settlement data, the value of cr can be evaluated by applying

Barron’s theory, using any of the curve fitting procedures. The procedure for the log(S-
So)-log t method is briefly described below.

1. Calculate the initial compression (So) using the following equation (Eqn. 6). from
the time-compression data by choosing two data points in the early stages of
consolidation.
⎛ S2 − So ⎞ ⎛ t2 ⎞
⎜ ⎟ = ⎜⎜ ⎟⎟
⎜ S −S ⎟
⎝ 1 o ⎠ ⎝ t1 ⎠ . . .(6)

2. Plot the time (t)- corrected settlement (S-So) in a log-log plot, similar to those shown
in Fig. 7.
3. Identify the initial linear portion and draw a line.
(i) Identify the linear secondary compression portion and draw a line and extend
it to intersect the initial straight line. The time at the point of intersection
(t66) corresponds to a degree of consolidation of 66%.
(ii) Calculate cr from the following equation,

2
(Tr ) 66 d e
cr =
t 66 . . . (7)

where, (Tr)66 is the time factor for Ur=66%. Variation of (Tr)66 with n is shown in Fig. 8.
(iii) Read the settlement S66 from the time-settlement curve as shown in Fig. 7.
(iv) Settlement corresponding to Ur = 100% can be calculated as:

S 66 − S o
S100 = S o +
0.66 . . .(8)

0.1
Degree of consolidation, Ur (%)

n=5

10
A

100
0.0001 0.001 0.01 0.1 1
Time factor, T r

Fig. 5 Log Ur-log Tr curves for n=5 showing the characteristic feature 96
(Point A corresponds to Tr=0.126 and Ur = 65.9%)
 IMPROCE 2009

0.001

Singapore marine clay

0.01 Kaolinite
S-S0, mm

0.1

10
0.01 0.1 1 10 100 1000 10000
Time, min

Fig. 7 log (S-So) versus log t plot

0.25

0.2

0.15
(Tr )66

0.1

0.05

0
0 2 4 6 8 10
n

Fig. 8 Variation of time factor at the intersection point (Tr)66

5. PRACTICAL CONSIDERATIONS FOR USING VERTICAL DRAINS

Though the principle and the design procedure involved in the design of vertical
drains are simple, several other factors need to be considered. This inclue the
installation effects, selection of design parameters and factors that control the selection
of PVDs, etc. (Chu et al. 2004).
97
 IMPROCE 2009

5.1 Basic design assumptions

The commonly used model for the design of vertical drains is the one suggested
by Barron’s theory for drain wells. While sand drains are circular in cross-section, the
pre-fabricated vertical drains (PVD) are rectangular. For design purposes, an equivalent
circular cross-section is adopted. Typically, Hansbo (1979) proposed the following
formulation for equivalent diameter of well.

2( w + t )
dw =
π . . . (9)

where, w and t are the width and thickness of PVD, respectively. When PVDs are
installed, an oversized mandrel is normally used to push the drains into the soil. The soil
around the PVD gets disturbed and the permeability of the soil in the disturbed zone,
called as the smear zone, is reduced. This process has been called as smear effect and
many design procedures and analytical models have been proposed in the literature to
consider the smear effect in the design. One such analytical model, proposed by Hansbo
(1981), assumes the smear zone to be an annulus of smeared clay around the vertical
drain as shown in Fig. 9. With ideal radial flow, the average degree of consolidation can
be calculated by:
⎛ 8c t ⎞
U h = 1 − exp⎜⎜ − h2 ⎟⎟
⎝ μd e ⎠ . . . (10)
⎛d ⎞ k ⎛d ⎞ 3 k
μ = ln⎜⎜ e ⎟⎟ + h ln⎜⎜ s ⎟⎟ − + πz (2l m − z ) h
And ⎝ ds ⎠ ks ⎝ dw ⎠ 4 qw
. . .(11)

where dw is the equivalent drain diameter; t is the time of consolidation; ds is the

diameter of the smear zone; kh is the permeability of the undisturbed soil in the
horizontal direction; ks is the permeability of the smeared soil; qw is the discharge
capacity of the PVD, lm is the maximum discharge length of the PVD. The uncertainties
involved in the design of PVD include (Chu et al. 2003):
a. The bias in the analytical model
b. Uncertainties in the parameters involved in Eqn . (10) and (11)
c. Variations in the quality of drain materials and uncertainties involved in
installation.

de/

Fig. 9 Smear zone around vertical drain

98
 IMPROCE 2009

Studies indicated that the analytical model proposed by Hansbo (1981) provide a
good estimation of degree of consolidation. It is very difficult to quantify the radius of
the smear zone, ds, as it is affected by many factors such as the shape, the size of the
mandrel, the type of soil and the sensitivity of the soil. Hansbo (1981, 1997) estimated
ds = (1.5 to 3.0)dw. This relation is commonly used in the design. Indraratna and
Redena (1998) observed based on large scale laboratory tests that ds could be as large as
(5 to 6)dw.

5.2 Requirement of discharge capacity

The efficiency of the drain is evaluated using a dimensionless parameter called

discharge factor defined as (Mesri and Lo, 1991):

qw
D=
k h l m2 . . .(12)

Studies suggested that the effect of well resistance can be ignored when the
value od D is greater than or equal to 7.85. Majority of the modern PVDs provide
sufficient discharge capacity and well resistance is not a serious issue.

5.3 Factors affecting the selection of PVD

The quality and suitability of the drain play a major role in the whole soil
improvement scheme involving vertical drains. A considerable saving could be resulted
without sacrificing the performance of drain if the control factors for the vertical drains
be identified and the design requirements are specified accordingly. Two such factors
are the compatibility of the filter with the soil to be improved and the tensile strength of
drain apart from the discharge capacity of the drain.

The filter material should have sufficient permeability and should meet the filter
criteria. The permeability of the filter is normally required to be at least one order of
magnitude higher than that of the soil. As brought out earlier, the discharge capacity of
the drain should be sufficient for effective performance of the drain. In addition, the
PVDs should have adequate tensile strength so that it can sustain the tensile load
applied to it during the installation. Therefore, the strength of the core, the strength of
the filter, the strength of the entire drain needs to be specified at both wet and dry
condition. It is quite common nowadays to specify the tensile strength of the whole
drain at both dry and wet conditions to be larger than 1 kN at a tensile strain of 10%
(Chu et al. 2004).

5.4 Instrumentation

The technique of vertical drains is appropriate to soft clays when the applied
load is above the preconsolidation pressure and secondary compression is not the main
component of settlement. Very often it may be difficult to predict the rate of
consolidation reliably due to various factors, including the permeability of the soil,
presence of permeable strata, soil disturbance at the time of installation, etc. Therefore,
proper field monitoring is necessary to evaluate the performance of the system. A
general guideline is as follows:

99
 IMPROCE 2009

• Install a combination of groundwater observation wells and piezometers to

provide a complete pore pressure profile prior to, during and after drain
installation.
• Install sufficient number of settlement gauges to monitor the settlements after
preloading
• Provide inclinometers where ever necessary

6 VACUUM PRELOADING

One of the limitations of surcharge preloading is that large quantity of material

is needed to impose the required surcharge pressure. Very often it may be difficult to
transport the required material if the site is in remote areas. In such situations, vacuum
preloading is of great use. Vacuum preloading technique also has other advantages over
the conventional preloading technique, as described below.

Vacuum preloading is the application of atmospheric pressure to form a

temporary surcharge for soft clays while applying a vacuum to the surface of the soil
beneath a membrane (Mitchell and Jardine 2002). Nowadays this technique has become
an effective alternative to the surcharge preloading technique and it has wide
applications where the hydraulic fill is used for the land reclamation purpose in addition
to the stabilization of soft clay deposits.

This method was first introduced by Kjellman in 1952. The method found its
first field application during the extension work at Philadelphia International Airport.
Afterwards several case histories of application of vacuum preloading technique
assisted with vertical drains have been reported, in the literature. In this form of
preloading method, vacuum is applied to the pore phase in a sealed membrane system.
The vacuum causes water to drain out from the soil and creates negative pore water
pressure in the soil. This leads to an increase in effective stress to the magnitude of the
induced negative pore water pressure, without the increase of total stress.

In order to achieve rapid pre-consolidation, vertical drains in the form of

prefabricated vertical drains are very often used along with the vacuum preloading. The
vertical drains will reduce the drainage path by promoting the flow in the horizontal
direction. The preloading technique assisted with vertical drains will ensure the
homogenous improvement of ground and it can reduce the primary consolidation
settlement (Bo et al. 2007). In vacuum preloading method, the vertical drains will help
to distribute the vacuum pressures to the deeper layers and drain out water from the sub
soil. Vacuum preloading with PVD substantially reduces the lateral displacement and
potential shear failure (Indraratna et al. 2005). Therefore the application of vacuum
pressure with prefabricated vertical drains has been widely used in land reclamation
projects to avoid the stability issues relating to the high surcharge embankment.
However, the maximum achievable vacuum pressure in the field is only about 80kPa
(Chai et al. 2005).

Schematic of the vacuum consolidation technique is illustrated in Fig. 10. The

stages in vacuum assisted consolidation project involve placing of a drainage blanket
above the soft ground in order to provide a working platform, installation of vertical
drains through the sand blanket, placing of corrugated flexible pipes horizontally to link
the PVDs to vacuum pump, excavation of trenches around the perimeter of the
100
 IMPROCE 2009

preloaded area to a depth of 50cm below the ground water table and filled with
bentonite slurry for the subsequent sealing of the impermeable membrane, installation
of impermeable membrane over the area and sealing it along the peripheral trenches and
followed by vacuum preloading using high efficiency vacuum pump.

The primary consideration for the effectiveness of the system includes air
tightness of the system, effectiveness of the seal between the membrane edge and
location of ground water level. The water table at the ground surface is most desirable
because if the water table is at a greater depth, then most of vacuum applied will go
towards raising the water table; resulting in a decrease in the expected effective stress.

Fig. 10 Schematic layout of vacuum applied vacuum consolidation

7 MECHANISM OF VACUUM CONSOLIDATION

The mechanism of vacuum preloading is very simple and has been well
explained in the literature based on the effective stress principle. Chu and Yan (2005)
explained the principle through the spring analogy and the same is reproduced here.

The consolidation process of soil under surcharge load has been well understood
and can be illustrated using the spring analogy as shown in Fig. 11(a). For the
convenience of explanation, the pressures in Fig. 11 are given in absolute values and pa
is the atmospheric pressure. As shown in Fig. 11(a), the instance when a surcharge
pressure, ∆p, is applied, the pore water takes the load and excess pore pressure gets
generated. Therefore, the initial excess pore water pressure, ∆u, is the same as the
surcharge ∆p. Gradually, the excess pore water pressure dissipates and the load is
transferred from water to the spring (i.e., the soil skeleton) in the model shown in Fig.
11(a). The amount of effective stress increment equals to the amount of pore water
pressure dissipation, ∆p - ∆u (Fig. 11(a)). At the end of consolidation, ∆u = 0 and the
101
 IMPROCE 2009

total gain in the effective stress is the same as the surcharge, ∆p. It should be noted that
the above process is not affected by the atmospheric pressure, pa.

The mechanism of vacuum preloading can also be illustrated in the same way
using the spring analogy shown in Fig. 11(b). When a vacuum load is applied to the
system shown in Fig. 11(b), the pore water pressure in the soil reduces. As the total
stress applied does not change, the effective stress in the soil increases. The instance
when the vacuum load, -∆u, is applied, the pore water pressure in the soil is still pa.
Gradually, the pore pressure is reducing and the spring starts to be compressed, i.e. the
soil skeleton starts to gain effective stress. The amount of the effective stress increment
equals to the amount of pore water pressure reduction, ∆u, which will not exceed the
atmospheric pressure, pa, or normally 80 kPa in practice.

For an idealised soil profile with the water table and a single drainage boundary
at the ground level, the distributions of pore water pressure and effective stress with
depth at a given time during consolidation can be plotted in Figures 12(a) and (b) for
surcharge and vacuum preloading respectively. Under surcharge load, the effective
stress equals to ∆σv- ut(z), where ∆σv is the surcharge and ut(z) is the excess pore water
pressure. As the pore water pressure increases with depth, the effective stress decreases
with depth as shown in Fig. 12(a). Under vacuum load, the effective stress equals to
σ’0+ u0(z)-ut(z), where σ’0 is the initial effective overburden stress, u0(z) the hydrostatic
pore water pressure, and ut(z) the pore water pressure. When the vacuum pressure is
applied from the ground level, ut(z) is smallest at the top. Therefore, the effective stress
will be the highest at the top (Fig. 12(b)). It should be pointed out that in the case of
vacuum preloading, the increment in effective stress cannot exceed 98 kPa, although the
effective stress in the soil can be higher than 98 kPa.

Chu and Yan (2005) have shown in their study that the above simplified model
depicts well the pore water pressures change process in soil under vacuum preloading.

Fig. 11 Spring analogy (a) under surcharge (b) under vacuum load
(Chu and Yan, 2005)
102
 IMPROCE 2009

Fig. 12 Pore water pressure and effective stress changes (a) under surcharge and
(b) under vacuum load (Chu and Yan, 2005)

8 COMPARISON OF VACUUM PRELOADING AND SURCHARGE

PRELOADING

The vacuum preloading technique has several advantages over the surcharge
preloading technique. Ground improvement with vacuum preloading does not require
any fill material and there is no need of heavy machinery. Construction period is
generally shorter. The increase in effective stress under vacuum preloading is isotropic.
Therefore, the corresponding lateral displacement is in the inward direction and there is
no risk of shear failure (Chai et al. 2005, Qian et al. 1992, Rujikiatkamjorn et al. 2007).
In addition, the bearing capacity of soil is nearly doubled in the case of soft clays and
70% of the total estimated settlement of design load will be eliminated (Lou et al,
1988). The overall cost of vacuum preloading is only about 2/3rd of that with surcharge
preloading (Qian et al. 1992). Therefore this is an economical and feasible method of
ground improvement.

The main difference between the two preloading techniques (surcharge

preloading and vacuum preloading techniques) in the response of the ground is in the
lateral movement of ground. The surcharge preloading technique will cause the outward
lateral movement of ground and will affect the stability of embankment. While vacuum
preloading will cause the inward lateral movement of ground, create the cracks around
the treated area and will affect the stability of adjoining structures. A balance between
103
 IMPROCE 2009

these two techniques will help to minimize the lateral yield of soil. Hence the Combined
Surcharge and Vacuum Consolidation technique can be used as an effective method to
optimize the soft clay foundation problems. In the combined preloading technique, the
consolidation starts with an excess pore-pressure equivalent to the applied surcharge
pressure and ends with a negative pore pressure equivalent to the applied vacuum
pressure. By using this method, the benefits of the two preloading techniques may be
achieved effectively. As described earlier, the maximum achievable vacuum pressure in
the field is only 80 kPa, so when a preload of higher than 80 kPa is required, a vacuum
pressure combined with fill surcharge is an ideal method in various preloading
techniques. Therefore, the combination of vacuum and surcharge load can effectively
shorten the preloading period, reduce the embankment height and counterbalance the
excessive lateral displacement.

9 REFERENCES

1. Barron, R. A. (1948). Consolidation of Fine –grained Soils by Drain Wells. Trans.

ASCE, Vol.113, 718-754.

2. Berry, P. L. and Wilkinson, W. B. (1969). The Radial Consolidation of Clay Soils,

Geotechnique, Vol. 19, No.2, pp. 253-284.

3. Bo, M.W., Arulrajah, A., and Nikraz, H. (2007). Preloading and Prefabricated
Vertical Drains Design for Foreshore Land Reclamation Projects: A Case Study.
Ground Improvement, Vol.11 (2), 67-76.

4. Carillo, N. (1962). Simple Two- and Three-Dimensional cases in theory of

consolidation of soils, J. Math. Phys., Vol. 21, No. 1, 1-5.

5. Chai, J.C., Hayashi, S., and Carter, J.P. (2005). Characteristics of Vacuum
Consolidation. Proceedings of 16th International Conference on Soil mechanics and
Foundation Engineering, 1167-1170.

6. Chu, J., Bo, M. W. and Choa, V. (2004). Practical considerations for using vertical
drains in soil improvement projects. Geotextiles and Geomembranes, Vo.. 22, 101-
117.

7. Chu, J., and Yan, W. (2005). Application of the vacuum preloading method in soil
improvement projects. Ground Improvement Case Histories, Vol. 3, Elsevier Geo-
Engineering Book Series, Indraratna, B. and Chu, J. (Eds), pp. 91-117.

8. Hansbo, S. (1979). Consolidation of Clay by Band-Shaped Vertical Drains. Ground

Engineering, Vol. 12, No. 5, pp. 16-25.

9. Hansbo, S. (1981). Consolidation of fine-grained soils by prefabricated drains. Proc.

10th International Conference on Soil Mechanics and Foundation Engineering, Vo.
3, Stockholm, pp. 677-682.

10. Hansbo, S. (1997). Aspects of vertical drain design: Darcian or non-Darcian flow,
Geotechnique, Vol. 47, pp. 983-992.

104
 IMPROCE 2009

11. Hausmann, M. R. (1990). Engineering Principles of ground Modification. McGraw-

Hill Publishing Co., New York.

12. Indraratna, B. and Redana, I. W. (1998). Laboratory determination of smear zone

due to vertical drain installation”, Journal of Geotechnical and Geoenvironmental
Engineering, ASCE, Vol. 124, No. 2, pp. 180-184.\

13. Indraratna, B., Sathananthan, I., Rujikiatkamjorn, C., and Balasubramaniam, A. S.

(2005) “Analytical and numerical modeling of soft soil stabilized by prefabricated
vertical drains incorporating vacuum preloading.” International Journal of
Geomechanics, ASCE, Vol. 5, No. 2, 114-124.

14. Mesri, G. and Lo, D. O. K. (1991). Field performance of prefabricated vertical

drains. Proc. of the International conference on Geotechnical Engineering for
Coastal Development, Yokohama, Vo. 1, pp. 231-236.

15. Robinson, R. G. (2009). Analysis of Radial Consolidation Test Data Using a log-log
Method, Geotechnical Testing Journal, ASTM, Vol. 32, No. 2, pp. 1-7.

16. Rowe, P. W. and Barden, L. (1966). A new consolidation cell, Geotechnique, Vol.
16, pp. 162-170.

17. Rujikiatkamjorn, C., Idraratna, B. and Chu, J. (2007). Numerical Modeling of Soft
soil Stabilized by Vertical drains, Combined Surcharge and Vacuum preloading for
a Storage yard. Canadian Geotechnical Journal, Vol.44, 326-342.

18. Kjellman, W. (1952). Consolidation of Clayey Soils by means of Atmospheric

Pressure. Proceedings of a conference on Soil Stabilisation, Massachusetts Institute
of Technology, Cambridge., 258-263.

19. Lou, Y. (1988). The Present and Future of Vacuum Preloading Technology.
Technical Bulletin, Nanjing Institute of Hydraulics.

20. McKinlay, D. G. (1961). A laboratory study of Consolidation in Clays with

Particular Reference to Conditions of Radial Porewater Drainage, Proceedings,
Fifth International Conference on SM & FE, Paris, Vol. 1, pp. 225-228.

21. Qian, J.H., Zhao, W.B., Cheung, Y.K., and Lee, P.K.K. (1992). The theory and
Practice of Vacuum Preloading. Computers and Geotechnique, Vol.13, 103-118.

22. Shields, D. H. and Rowe, P. W. (1965). Radial drainage oedometer for laminated
clays, Jl. of the Soil Mechanics and Foundations Division, ASCE, Vol. 91, SM1, pp.
15-23.

23. Sridharan, A., Prakash, K., and Asha, S.R. (1996). Consolidation Behavior of
Clayey Soils under Radial Drainage. Geotechnical Testing Journal, Vol.19 (4), 421-
431.

105
IMPROCE 2009

106
 IMPROCE 2009

GROUND IMPROVEMENT TECHNIQUES - STONE AND LIME COLUMNS

Dr.K.Ilamparuthi *

1.0 INTRODUCTION

Soft deposits can be found in the subsoil profile at number of places particularly
along the coast in narrow tidal plains and swamp area. In coastal Tamil nadu these
deposits occur in backwater areas of Kovalam, Vedaranyam, Pullicat and Pondichery.
The soft clays are geologically young sediments, which are unconsolidated deposits
with low shear strength and high compressibility. Further their natural moisture content
is likely to be higher than liquid limit with low 'N' value (<5) and static cone resistance
values of 200 to 1000 KN/m2. In these deposits conventional shallow foundation for
buildings and equipments may pose problems with respect to sizing of foundation and
foundation settlement.

The dearth of good construction sites has forced engineers in recent times to use
these poor sites either by providing deep foundation or by improving the ground.
Methods developed over recent years to improve the soft soil can be effectively used so
that the improved ground exhibits adequate bearing capacity and foundations
constructed on them would result in settlement within the tolerable limits.
2.0 PURPOSE OF GROUND IMPROVEMENT

The ground improvement technique is a special method aimed at increasing

bearing capacity, reduce compressibility, decrease permeability, modify dynamic
response, reduce, the risk of liquefaction potential of subsoil and increase the stability of
soil in slopes.
3.0 NECESSARY DATA

Following information of subsoil is essential to establish the need for ground

improvement at a site, for selection of suitable method and for the design using selected
method.
i. Subsoil profile and soil characteristics up to a depth of about twice the width of
the loaded area or up to hard strata if encountered earlier.
ii. Properties of subsoil such as index, shear strength, compressibility
characteristics etc.
iii. Static or dynamic cone penetration resistance profile upto the depth of soil to be
improved.
iv. Information with respect to nature of structure, area covered by it, intensity and
nature of load and the permissible distortion that the structure can withstand.

4.0 METHODS OF IMPROVING SOFT GROUND

Ground improvement is generally achieved by anyone of the following methods:
1. Stone column
2. Lime column
* Professor & Head, Division of Soil Mechanics and Foundation Engg, Anna University Chennai, Chennai - 25 107
 IMPROCE 2009

3. Sand drains With preloading

4. Sand compaction piles
5. Sand pads
6. Dynamic compaction
7. Grouting etc.

Among the methods indicated, stone and lime column methods are widely
adopted in field, hence principles involved in these two methods are briefing explained.

5.0 STONE COLUMN

This technique has been developed very recently for improving bearing capacity
and to reduce settlement of weak deposits like soft clays and loose sands. It has been
increasingly adopted in India. This method involves making bore holes in the weak
deposits and filling stone chips or gravel or mixture of these and compacting them to
create a column of desired strength. These are constructed by adopting any one of the
following methods.
i. Non-displacement Method
ii. Vibrofloated Stone Column.
5.1 Non-displacement Method

In this method a hole of required size is accomplished either by using bailer and
casing or using rotary drill. Initially the borehole is advanced using a bailer while its
sides are supported by a casing. After the casing has reached the required level, sound
and well-graded crushed stone of 75mm down size of 2 mm is placed in the borehole
and casing is withdrawn at a certain length to ensure continuous formation of stone
column. The loose charge below the bottom of the casing is then compacted by
operating a rammer of suitable weight and fall within the casing, so as to obtain
compaction energy of ground of 20 kN m per blow. The sequence of formation of stone
column is shown in Figure 1.
5.2 Vibrofloated Stone Column
In this method a hole is formed in the weak deposit using a vibrofloat unit. The
vibrofloat equipment comprises of a vibrofloat probe, accompanying power supply,
water pump, crane and front - end loader as shown in Figure 2.The vibrofloat is a poker
vibrator normally of diameter varying from 300 mm to 450 mm and about 2m to 3.5m
long weighing 2 to 4 tonnes depending upon the size (Figure 3).

By this method soil generally gets replaced by jetting water used with the vibro
float, size of the hole formed can be controlled to a certain extent, by regulating
intensity of vibration, pressure and rate of penetration of float. Diameter of holes
formed vary generally from 0.8m to 1.1m. After the vibrofloat has penetrated to the
desired level it is gradually withdrawn and crushed stone is poured. During the process
of withdrawal of the needle, vibration and jetting is continuously maintained to ensure
compaction of granular fill. The withdrawal of the float is made in short passes,
preferably of about lm to ensure proper compaction and uniformity along the entire
length of stone column. The sequence of constructing the column by vibrofloat is as
shown in Figure 4.

108
 IMPROCE 2009

5.3 Failure Mechanisms Of Single Stone Column

Failure mechanism of a single stone column loaded over its area significantly
depends upon the length of the column. For columns having length greater than its
critical length (that is about 4 times the column diameter) and irrespective whether it is
end bearing or floating, it fails by bulging (Figure.5 A). However, column shorter than
the critical length are likely to fail in general shear if it is end bearing on a rigid base
(Figure. 5B). and in end bearing if it is a floating column as shown in Figure 5 C.

In practice, however, a stone column is usually loaded over an area greater than
its own (Figure.6) in which case it experiences significantly less bulging leading to
greater ultimate load capacity and reduced settlements since the load is carried by both
the stone column and the surrounding soil.

The above failure mechanisms apply to stone columns installed in homogeneous

soils. Practical situations may rise where isolated zones of very soft cohesive soils may
result in significant bulging at both shallow and deep depths and hence, this should be
duly considered wherever necessary. The failure mechanisms of stone column in non –
homogeneous cohesive soil are as shown in Figure 7.

5.4 Load Carrying Capacity and Settlement of Stone Column Reinforced

Ground

The stone column is essentially a system of soil reinforcement with the

additional advantage of providing drainage path. Design of stone column reinforced
ground involves two aspects.

1. Estimation of load capacity of a stone column

2. Settlement Analysis.

5.4.1 Estimation of Load capacity of a stone column

Load capacity of the treated ground may be obtained by summing up the

contribution of each of the following components for wide spread loads, such as
tankages and embankments:

a) Capacity of the stone column resulting from the resistance offered by the
surrounding soil against its lateral deformation (bulging) under axial load,

b) Capacity of the stone column resulting from -increase in resistance offered by

the surrounding soil due to surcharge over it, and

c) Bearing support provided by the intervening soil between the columns

a. Capacity Based on Bulging of Column

Considering that the foundation soil is at failure when stressed horizontally due to
bulging of stone column, the limiting (yield) axial stress in the column is given by
the sum of the following:

109
 IMPROCE 2009

σv = σ rL K pcol 1
σv = (σ ro + 4Cu ) k pwl 2

where σv = limiting axial stress in the column when it approaches shear

failure due to bulging, and
limiting radial stress = σ ro + cu
( 4 )
σrL =

where Cu = undisturbed undrained shear strength of clay surrounding the

column, and
σro = initial effective radial stress = k o σ ro

where K0 = average coefficient of lateral earth pressure for clays equal to 0.6
or alternatively, as determined from the relationship Ko = 1 -sinφ,
where φ is the effective angle of internal friction of soil, and
σvo = average initial effective vertical stress considering an average
bulge depth as 2 times diameter of the column (Figure 5A), that
is σro = γ2D

where γ = effective unit weight of soil within the influence zone.

φ
K pcol = tan 2 (45o + c )
2

where
φc = angle of internal friction of the granular column material and it
may vary depending upon angularity, surface characteristics and
density of column material. As a broad guide, the φc may range
from 38° to 42° de-pending upon the compactness achieved
during. construction of stone columns.

Yield load = σv π D2/4

Safe load on column alone
Q1 = σV (π/4 D2 ) / 2 . . . .(2)

where 2 is the factor of safety.

b. Surcharge Effect

The increase in capacity of the column due to surcharge may be computed in

terms of increase in mean radial stress of the soil as follows:

q safe
Δσ ro = (1 + 2 K 0 )
3 . . . .(3)

where Δσro is the increase in mean radial stress due to surcharge, and q safe is the safe
bearing pressure of soil with the factor of safety of 2.5 (See IS 6403)

qsafe = Cu Nc / 2.5
110
 IMPROCE 2009

Increase in yield stress of the column = Kpcol Δσro

Allowing a factor safety of 2, increase in safe load of column, Q2 is given by the

following formula:

K p col Δσ ro As
Q2 =
2 . . . . (4)

The surcharge effect is minimum at edges and it should be compensated by

installing additional columns in the peripheral region of the facility.

c. Bearing Support Provided by the Intervening Soil

This component consists of the intrinsic capacity of the virgin soil to support a
vertical load which may be computed as follows:

Effective area of stone column including the intervening soil for triangular pattern
= 0.866 S2

Area of intervening soil for each column, Ag is given by the following formula:

πD 2
A g = 0.866 S2 −
4

Safe load taken by the

intervening soil, Q3 = qsafe Ag . . .. (5)

Overall safe load on each column

and its tributary soil = Q1 + Q2 + Q3

NOTE: Number of columns to be provided under a structure may be obtained if

the total load which the structure is subjected to and the reduction in settlements
required considering the permissible total and differential settlements for it are known.

5.4.2 Settlement Analysis

The settlement of stone column treated ground may be determined by either the
empirical approach or the concept of vertical average stress on the soft soil. In the
empirical method the settlement of treated ground is computed as a percentage of
untreated ground settlement. For a given spacing of stone column and strength of the
soft ground, the settlement of the reinforced ground can be obtained from the Figure 8.
In this method it is assumed that the column rests on sufficiently hard ground. In the
reduced stress method, the settlement of the treated ground depends upon two
parameters namely stress concentration factor, n, and the area replacement ratio, as.
Factor, 'n' depends on the relative stiffness of the stone column and generally varies
from 2 to 6 with usual values of 3 to 4. The area ratio, as is determined from the
diameter and spacing of stone columns provided. Consolidation settlement St of treated
ground is determined by the equation given below:

111
 IMPROCE 2009

St = βS
Where β = Settlement reduction factor
1
= (1 + (n − 1) a s )
σs
n = σc
σs
= Vertical stress in compacted stone column
σc
= Vertical stress in surrounding soil
A sc
as = (A sc + A s )
Asc = Area of stone column
As = Area of soil surrounding the stone column

6.0 LIME COLUMN

Lime has been used widely in soil stabilisation for several years and in recent
times quick lime has found wider application particularly in stabilisation of soil soil.
Lime stabilisation is achieved by two ways depending on nature of soil (soft or
desiccated), thickness of strata, depth of strata etc. Intrinsic mixing of lime with clay is
generally resorted to, when the thickness of layer is small and is available at shallow
depths, whereas lime column technique is adopted at locations where soft clay extends
to larger depth. The lime reacts with soil and increases the strength and decreases the
compressibility of soil. These reactions are hydration, cation exchange, puzzolanic and
carbonation. The substantial increase in strength, permeability and decrease in
compressibility are mainly attributed to the formation of cementation products such as
calcium silicate hydrate (CSH) and calcium aluminate hydrate (CAH) due to puzzolanic
reaction.

The stabilisation of thick deposits of soft clay by lime column or lime pile over-
comes some of the difficulties connected with pile foundation. This method was first
described by Broms and Boman (1975). In this method the columns made insitu by
mixing un-slaked lime with soft clay are used to support light structures. The lime and
different additives are mixed with the soil using an auger, thereby column of stabilised
soil with 0.5m diameter and upto a length 10m are formed with spacing of 2 to 4 times
the diameter. The shear strength of the soil is increased while the compressibility is
reduced due to chemical action between lime and soil. Measurements indicate that an
area stabilized with lime columns behave as a rigid block as long as the average shear
stress along the perimeter of the block, is less than the undrained shear strength of the
surrounding unstabilised clay.

6.1 Formation of Lime Column at Insitu

The lime columns are formed insitu by mixing the soil insitu with unslaked lime
using an auger, shaped like a giant eggbeater (Figure 9). Normally quick lime is added
upto 10% of dry weight of soil. The rotation rate is 60 rpm, when the tool is pushed
down into the soil corresponding to pitch of the inclined blades. At the required column
length, the rotation is reversed and the tool is withdrawn gradually. While withdrawing
112
 IMPROCE 2009

the auger the lime is forced into the soil through a hole in the auger using compressed
air. The auger is withdrawn slowly in order to mix the lime thoroughly with the soil and
to compact the lime-soil mixture. The amount of lime forced into the soil is recorded as
well as the injection pressure. The rotary table mast mounted on a special carrier is
shown in Figure 10.This equipment has a capacity to form 10m long lime column of
0.5m diameter in about 10 minutes.

6.2 Lime Column Applications

Lime column act as reinforcement and drain in the soil. The reinforcing effects
can be studied by varying the length, spacing and the diameter of lime columns. Lime
columns can be used to support light structures as shown in Figure 11. The total as well
as the differential settlements are thereby reduced, since the weight of the structure is
transferred to deeper and less compressible strata. The lime columns can also be used to
decrease the negative skin friction the pile which is supporting buildings (Figure 12) or
below bridge abutments. They will carry part of the weight of the soil which otherwise
will be transferred to the structural piles when the soil around the structure settles.
Further the lime column also prevents the lateral displacement of soil around bridge
abutments or substructure due to the lateral creep. Lime columns are also used in soft
clay to increase the stability of slopes and to reduce the lateral earth pressure on sheet
pile and retaining walls and in deep cuts to prevent failure by bottom heaving (Figure
13).

6.3 Quantum of Lime

The stabilisation effect of lime depends on the presence or absence of organic

matter, clay content and pore fluid characteristics. If the organic content is more, the
requirement of lime content is also more. In silt textured soil with clay fines deposited
in fresh water the stabilisation effect is maximum with smaller quantity of lime. The
stabilisation effect is retarded if the natural moisture content is very high. In marine
clays generally 6 to 10% of lime by dry weight of soil will be adequate.

6.4 Variation of Strength with Time

The increase in strength of lime treated soil increases with time. One third of
final strength can be achieved within a month of stabilisation. The fifty percent of final
strength can be achieved in one year after the installation of lime column. The reaction
of lime with clay is similar to concrete which also exhibits strength increase with time.
The full strength of lime treated ground can be achieved after several years of
installation.

6.5 Design Of Lime Column

6.5.1 Ultimate Strength

The ultimate strength of lime column depends upon natural moisture content,
initial strength of soft clay, amount of lime content and spacing of lime columns.
Eventhough the existing literature indicates that there is a substantial increase in shear
strength of lime treated ground, the ultimate strength of the soil is estimated through
field tests such as static cone penetration tests, screw plate load tests and vane shear
113
 IMPROCE 2009

tests. Further there is no established method to estimate the ultimate capacity of lime
column treated ground. The design of the lime column treated ground was carried out
based on settlement consideration rather than the strength.
6.6 Settlement

6.6.1 Differential Settlement

Structures are primarily damaged by differential settlements. For an area

stabilised with lime column settlement is caused by shear deformations in the
unstabilised soil between columns. The larger part of the weight of the structure will be
transferred to the surrounding soil through the periphery of the block enclosed by
columns as shown in Figure 14.
A small relative displacement of the block is required to mobilize the, shear
strength of the soil along the periphery. A large deformation of the order of 50mm to
100mm is necessary to mobilize the shear strength of the soil below the block when the
depth of soft strata below column is large. With time, larger part of the weight of the
structure will be transferred to the bottom layer through the block as the soil around the
block consolidates.
The differential settlement will be small as long as the perimeter shear is less
than the average shear strength of the surrounding soil. The perimeter shear S can be
calculated from the equation given below, by assuming that the load transferred through
the bottom of the stabilised area is negligible.

W Cu
S = <
2( B + L) H FOS . . . (5)
Where W = Weight of the structure
B = Width of the reinforced block
L = Length of the reinforced block
H = Height of reinforced block
Cu = Average undrained shear strength of surrounding soft clay
FOS = Factory of safety, 1.5.
S Cu
α= =
G B (FOS x G B ) . . . .(6)

Where GB = equivalent Shear mod ulus of soil

B
Gc
= (B − nD)
B = width of reinforced block
When the load is unevenly distributed it is advisable to concentrate the columns
at the part of structure with highest unit load. The load distribution below structure will
depend on the relative stiffness of the structure with respect to the soil. The differential
settlements from the shear deformation in the soil between columns can be estimated as
shown in Figure 15. The angle change, α at the edge of the reinforced block will, at low
stress levels, increase approximately linearly with shear stress, 'S'.

114
 IMPROCE 2009

D = Diameter of lime column

n = number of column rows
Gc = shear modulus of the unstabilised soil approximately equal to 100 Cu
The angle change, α should be less than (tan-1(1/250) which is the maximum
angle change most structures can tolerate without damage.
6.6.2 Total Settlement

Most of the structures can tolerate higher total settlements provided the
settlements are uniformly distributed. However when the total settlement exceeds
100mm serviceability facilities such as water and sewer connections may begin to
rupture. These difficulties can be controlled if settlements are taken care of during the
design stage itself. The maximum settlement below the centre of loaded area of lime
column supported ground can be calculated as below.
The total settlement is equal to the sum of the compression of the reinforced
block, h1 and compression of the unstabilised soil lying below the reinforced lime
column block, h2 as shown in Figure 16.
The compression of the reinforced block depends on the interaction between the
lime column and the soil enclosed between the columns as well as the length of lime
columns. The settlement of the reinforced block is generally computed in two ways.

1. The deformation of the block is large so that the creep limit 'Qr' of the column is
reached.
2. The applied load and deformation of the block are small so that the creep limit
of the column is not reached.
(i) Column has reached creep limit
In this method settlement can be estimated by dividing the applied load into two
parts q1 and q2 in which q1 is the part of the load resisted by the lime column and q2 is
the reminder of the load carried by the enclosed soil as shown in Figure 17. The load q1
carried by the column is dependent on the creep limit, which is 0.7 Cu of column of the
stabilised soil. The settlement h1 caused by the load q2 can be calculated from the
consolidation test on undisturbed samples. The reinforced layer block is divided into
layers and the stress increase for each layer is calculated from Boussinesq equation or
by 2:1 method.
(ii) Column has not reached creep limit

In this method the relative stiffness of the column with respect to the enclosed
unstabilised soil will govern the stress distribution. The settlement h1 of the reinforced
block will be governed by the compression Modulus 'Mc' of the column material
qH
h1 = c
Mc . . . (7)
Where qc = the average axial stress in the column
H = column length
Mc = 300 quc for short term loading
quc = the ultimate strength of the column.

115
 IMPROCE 2009

Further the above equation can be modified as follows by considering that the
soil between the columns and column settle equally.
qH
h1 =
( ρM c + (1 − ρ ) M s ) . . . (8)

Where q = intensity of load on the foundation

H = column length
Mc = compression modulus of column
Ms = compression modulus of soil
(= 250 Cu approximate or from consolidation test)
n Ac
ρ = Area ratio = BL
n = total number of columns
Ac = Area of cross section of each column
B = width of reinforced block
L = length of reinforced block (Or)
h1 = β x consolidation settlement of untreated ground
(9)

1
β=
ρ Mc
+ (1 − ρ )
Where Ms

The settlement h2 of layer lying below the lime column can be determined by
conventional method.

SUMMARY

Numerous instances arise where the soil at many sites at shallow depths are not
having required properties to support proposed structures. In some situations, poor soil
conditions may pose problem for the integrity of existing structures. In response to these
needs special techniques for the in-place treatment of soils have been developed and are
practical effectively. Among the several methods the techniques of stone column and
lime column (Deep Mixing Method) are popular since these methods require less time
to implement and improvement is quicker.

REFERENCES

1. Broms, B.B (1984), Stabilisation of Soft clay in Southeast Asia, 5th Int.
Geotechnical seminar in Case histories of soft Clay, Singapore, pp 163 – 168.
2. Moseley, M.P, (1983), Ground improvement, Blaucie Academic and
Professional, Water Cleddens Road, Bishopbriggs, Glasgow.
3. Nayak, v. (1996) Foundations Design Manual, Dhanpat Rai Publications (p) Ltd,
New Delhi.
4. IS15284(Part 1):2003, Design and Construction for Ground Improvement –
Guidelines: part-1 Stone columns, Bureau of Indian Standards, New Delhi

116
 IMPROCE 2009

Figure 1 Stages of construction of stone column by bored piling equipment (Non –

Displacement Method)

Figure 2 Vibroflotation equipment (After Brown, 1976)

117
 IMPROCE 2009

Figure 3 100 HP Vibroflot (After Brown, 1976)

Figure 4 Vibroflotation compaction process (after Brown, 1975)

118
 IMPROCE 2009

Figure 5 Failure Mechanisms of a Single Stone Column in a

Homogenous Soft Layer (IS 15284)

Fig. 6 Deferent Type op Loadings Applied to Stone Columns (IS 15284)

Figure 7 Stone Column Failure Mechanisms in Non – Homogenous Cohesive Soil

(IS 15284)
119
 IMPROCE 2009

Figure 8 Effect of stone column on anticipated foundation settlement

Figure 9 Installation steps of time column

120
 IMPROCE 2009

Figure 10 Rotary table mast during transport

Figure 11 Stabilisation of light structures with lime columns

Figure 12 Protection of pile structures with lime column

121
 IMPROCE 2009

Figure 13 Stabilisation of cuts with lime columns

Figure 14 Shear stress along periphery of soil block reinforced with lime columns

Figure 15 Differential Settlements

122
 IMPROCE 2009

Figure 16. Calculation of total settlement

Figure 17. Load Distribution

123
IMPROCE 2009

124
 IMPROCE 2009

GROUND MODIFICATION BY GROUTING

Dr. M. Muttharam*

1. INTRODUCTION

Grouting refers to the injection of suitable fluid (colloidal solution, chemicals

etc.) into voids, cracks, fissures of the ground or spaces between the ground and
adjacent structures, generally through boreholes and under pressure.

The main objectives of grouting are to produce a stronger, denser, and/or less-
permeable soil or rock; it may also simply serve to fill voids, which are otherwise
inaccessible and may prevent adequate stress transfer within the ground or from a
structure to the ground.

2 CATEGORIES OF GROUTING

Grouting techniques are classified according to the method used to introduce the
grout into the ground. However, other criteria could be used to differentiate grouting
methods, such as

• Type of grout material injected

• Typical applications,
• Layout of injection points
• Sequence of construction.

Distinguished by the mode of entry into the soil or rock, the basic categories of
grouting are (as shown in Fig. 1)

• Penetration grouting (intrusion, permeation)

• Displacement grouting
• Compaction grouting (including slab-jacking)
• Grouting of voids Jet grouting (replacement)
• Special grouting applications and techniques, including electro grouting

Penetration grouting describes the process of filling joints or fractures in rock or pore
spaces in soil with a grout without disturbing the formation. More specifically,
permeation grouting refers to the replacement of water in voids between soil particles
with a grout fluid at low injection pressure so as to prevent fracturing.

Displacement grouting is the injection of grout into a formation in such a manner as to

move the formation; it may be controlled, as in compaction grouting or uncontrolled, as
in high-pressure soil or rock grouting which leads to splitting of the ground, also called
hydro fracture.

In compaction grouting a very stiff (say 25-mm. slump) mortar is injected into loose
soils, forming grout bulbs, which displace and densify the surrounding ground, without

* Assistant Professor, Department of Civil Engineering, Anna University Chennai, Chennai - 25 125
 IMPROCE 2009

penetrating the soil pores. With slightly more fluid grant, thick fissures rather than
bulbs may form; this is sometimes referred to as squeeze grouting.

Fig. 1 Schematic representation of basic modes of grouting

(After Hausmann, 1990)

Jet grouting is a technique where high-speed water jets emanating from a drill bit cut
into alluvial soils; as the drill bit is withdrawn, grout is pumped through horizontal
nozzles and mixes with or displaces the soil. The original foundation material is thus
replaced with a stronger and/or more impermeable grout-soil mixture. Jet grouting may
be used to form cutoff walls, do underpinning, or form deep foundations similar to
grouted auger piles.

Electro grouting is a term used for promoting electrochemical hardening during electro
osmosis by adding chemicals, such as sodium silicate or calcium chloride, at the anode.
Under the influence of the electric field, these chemicals permeate the ground, flowing
in the direction of the cathode, while the anode becomes a grout injection pipe. An
overview of typical grouting applications is given in fig 2.

126
 IMPROCE 2009

Figure 2. Typical application of grouting (After Hausmann, 1990)

3 GROUT MATERIALS

3.1 Classification of Grout Materials

Three basic types of grout are differentiated according to composition:

Suspensions: Small particles of solids are distributed in a liquid dispersion medium.

Examples: cement and clay in water.

Emulsions: A two-phase system containing minute (colloidal) droplets of liquid in a

disperse phase. Example: bitumen and water. Also in this category are foams, created by
emulsifying a gas into the grout material, which could be cement or an organic
chemical. Foaming agents, such as additives, which increase surface tension, assist in
forming bubbles by agitation; alternatively, they may induce gas-forming chemical
reactions.

Solutions: Liquid homogeneous molecular mixtures of two or more substan. Examples:

sodium silicate, organic resins, and a wide variety of other so-called chemical grouts. A
difference may be made between colloidal solutions (e.g., silica or lignochrome gels)
and pure solutions (e.g., phenolic and acrylic resins, aminoplasts).

127
 IMPROCE 2009

Principal types of grouts as distinguished by Cambefort (1987) are list in Table

1, which also gives an indication of appropriate uses, construction controls, and relative
costs.

Table 1 Principal types of grout

State
Suspensions Liquids Aerated
Unstable Stable Chemical products Emulsions
Grout type Cement Bentonite De Sodium Sodium Organic Cement Organic
+ cement flocculated silicate silicate resins foams foams
bentonite hard diluted
gels gels
Range of Fissures Sands and gravels. k m/s
uses
High
>5 x 10-4 >10-4 >I0-4 >10-5 >10-6 Cavities Water
flows
Grouting Refusal
control pressure Limited quantities Filling

Relative 4.2 1 (cement

cost for the (deposit 200 kg;
products to with γd betonite 0.8-1 6 2-4 10-500 1.2 10
fill 1-m3 = 1.5) 30 kg)
voids
Source: After Cambefort (1987).

4 BASIC PROPERTIES.

Important basic characteristics of grouts are stability, setting time, and viscosity.

Stability. A grout is referred to as stable, if its particles remains in suspension of

solution until it has reached the destination in the ground. If sedimentation occurs as
soon as the grout is no longer agitated by the mixer or through turbulence in the grout
pipes, it would be considered unstable. The breaking of an emulsion and the exudation
of liquids from colloidal gels (syneresis) could also be considered a sedimentation
process

The settling out of solid particles from a suspension while the liquid component
travels further into the soil or rock mass is also referred to as "filtration," although the
process is not just related to the relative sizes of the particles and the voids, but more
importantly to the flow velocity of the grout.

Settling of particles out of suspension when the grout becomes stationary results
in part of the voids containing water rather than grout.

Setting time. Setting time is the time required for the grout to harden. Cement-based
grouts normally set within 4 to 24 h, depending on the additives used. Setting or gel
time can be critical for chemical grouts, which can set very rapidly, possibly within
minutes.

128
 IMPROCE 2009

Viscosity. Viscosity μ is defined by Newton's law of viscosity as the proportionality

factor relating the shear resistance τ in a fluid to the velocity gradient dv/dz. which
represents the rate at which one layer of fluid moves relative to an adjacent layer.

The viscosity of "evolutive" grouts, like silica gels, increases gradually until
they set. Acrylic and other resins and other "nonevolutive" grouts show constant
viscosity until they set almost instantaneously; their setting time, usually measured in
minutes, is controlled by the type of catalyst present and the temperature.

Ideally, a grout should have low viscosity, a controllable setting time, and high strength
once it is in the ground. In addition it should be nontoxic, permanent, and cheap.

In order that a grout may bring about the desired effect it is necessary that it
should have the correct fluid properties for injection into the formation, that its set
properties satisfy the design specifications, and that the transformation between the
fluid and final set states should be sufficiently rapid for displacement of the grout to be
unlikely under the others to which it will probably be subjected

Other factors which are important in particular circumstances include the

chemical stability of the grout components and their deterioration under exceptionally
high or low site temperatures.

A full description of a grout being evaluated for treatment of a particular soil

includes its initial density, fluid viscosity and shear strength, particle size distribution of
any solids, and changes in these that may occur during injection. The possibility of the
grout mixing with or being diluted by the groundwater, as well as the effect of the fluid
on the chemistry of the soil and groundwater, also need to be known.

Viscosity and rigidity are the two rheological properties, which govern the flow
of grout in the voids of the soil. In suspension grouts they are inversely proportional to
the water/solids ratio.
Other factors apart (e.g. dilution by groundwater or reaction with components it
carries), gel time depends on the concentration of activator, inhibitor and catalyst in the
grout formulation. With most chemical grouts, varying the concentration of one or more
of these three components can change the gel time. Increases in temperature can reduce
initial viscosity but the reductions are marginal and are quickly compensated by the
accelerated gelling process.
Grout, which has been injected successfully, must develop sufficient rigidity to
remain in place. In particular, it must be capable of resisting the hydraulic pressure
exerted by groundwater that has been sealed off by the grouting operation. Rigidity of
an ordinary clay grout cannot be improved by adding more of the basic clay
constituents, as viscosity would increase. This is usually done by adding bentonite (1-10
g/litre) or by using sodium silicate with a reagent. Sodium silicate tends to thicken the
grout and increases rigidity markedly.
The yield point of a grout is the lowest stress that must be applied to initiate
laminar flow. Consequently, it governs the minimum pumping pressure required to
inject grout. The yield point of many suspensions increases if they remain at rest and
diminishes when they are agitated.
129
 IMPROCE 2009

5 PENETRATION OF GROUTS
The ability of grout to penetrate soil depends on the particle size distribution,
permeability and porosity of the soil; the pressure being used for injection; the period of
injection; and the viscosity of the grout. Because grouts do not maintain the same
viscosity indefinitely, there is a limit to the distance any grout can penetrate before the
decreasing head available at the advancing interface between the grout and the pore
water, and the increasing viscosity of the grout as it sets, prevents further flow.
The penetration characteristics of cement or clay grouts or admixtures thereof
are very different from those of chemical grouts. When particulate grouts are injected
into porous soil, filtering may occur whereby the larger particles in the suspension tend
to separate out at the entrances to pores. For example, filtercake formation occurs when
cement grout is injected into uniform sands of less than 2 mm grain size.
Hence the ability of particulate grouts to penetrate a formation depends upon the
particle size of the suspended material in that the particles of the grout must be smaller
than the voids that require filling. This has been indicated in terms of a groutability
ratio, which has been defined as
D15 (soil)
N=
D 85 (grout ) . . . (1)
This ratio should exceed 25 if a grout is successfully to penetrate the formation
concerned. Grouting is not possible if the ratio is less than 11. However, Burwell (1958)
suggested that the D85 value of the grout should not be relied upon solely. He
recommended that when the limits of groutability are approached, the criterion should
be re-examined on the basis of

D10 (soil)
N=
D 95 (grout ) . . . .(2)
in order to be doubly sure that the grout is suitable. In the latter case grouting is
consistently possible when N values are above 11 but is impossible when they are less
than 5. Alternatively, the limits for particulate grouts may be taken as a 10:1 size factor
between the D15 of the grout and the D15 size of the granular soil concerned.
Soils containing less than 10% fines can usually be permeated with chemical
grouts. If the fines content exceeds 15%, effective chemical grouting may prove
difficult. Permeation grouting is not possible when the fines content is greater than 20%.
Then hydrofracture must be used. Particle fineness (given in cm2/g) also offers an
indication of groutability.
For better groutability another criterion to be satisfied: that is particle size of
grout material must not be more than 1/10th of D10 of soil strata further for cement
grouting particle size of cement should be equal to 0.051mm.
The travel distance from the grout pipe varies as one-third the power of pumping
pressure, pumping rate, soil I permeability, and soil porosity, this has been quantified in
the following manner.
1/ 3
⎛ Rgt ⎞
r = 0.62 ⎜ ⎟
⎝ n ⎠
130
 IMPROCE 2009

Where r = grout penetration radius, ft

t = grout gel time, min
n = porosity of soil formation
g = acceleration due to gravity, ft/s2
R = ratio of viscosity of water to viscosity of grout

Few guidelines to select the type of grout are given in Table 2.

Table 2 selection of grout

Soil Group Sandy gravel Medium to fine and Silty/clayey sand

Grain dia (mm) >0.5 0.02-0.5 <0.02
(d10)
Degree of dispersion s(cm-1) <100 100-1000 >1000
Permeability cm/s >1x10-1 10-1-10-3 10-3to10-6
Grout type Suspension or Colloidal True
Emulsion Solution Solution
Purpose strength Cement grout Hard Amino last
Silicates Phenoplast
Water tightness Clay-cement Soft silicate aminoplast or acrylamids
Bentonite-cement
Asphalt emulsion

Alternatively, permeability may be used to assess groutability. For example,

Littlejohn (1985) suggested that cement grouts cannot be used for treating soils with
permeabilities less than 5 x 10-3 m/s as cement particles would be filtered out during
injection. As far as clay grouts are concerned, the small sizes of the particles permit
them to penetrate voids in soil with a permeability of 10-5 m/s. The limits of groutability
of some grout mixes are given in Table 3.

Table 3 Limits of groutability of some grout mixes (after Caron et al. 1975)

Characteristics Coarse sands and Types of soils Silty or clayey sands,

gravels Medium to fine sands silts
Grain diameter d10 > 0.5mm 0.02 < d10 < 0.5mm d10 < 0.02mm
Specific surface S < 100 cm 100cm < 5 < 1000 cm S > 1000 cm
Permeability k > 10-3 m/s 100-3 > k > 10-5 m/s k < 10-5 m/s
Type of mix Bingham suspensions Colloid solutions (gels) Pure solutions (resins)
Consolidation grouting Cement Hard silica gels: Aminoplastic
(k > 10-2 m/s) double shot: phenoplastic
Aerated mix Joosten
(for k > 10-4m/s
Single shot:
Carongel
Glyoxol
Siroc
Impermeability grouting Aerated mix Bentonite gel Acrylamids
Bentonite gel Ligno-chromate Aminoplastic
Clay gel Light carongel Phenoplastic
Clay-cement Soft silicagel
Vulcanizable oils
Others (Terranier)

131
 IMPROCE 2009

6 GROUTING EQUIPMENT AND METHODS

Because of the different functions of grouting, the differences between grouts

themselves, and the differences between materials to be grouted, generalizations about
grouting equipment and methods are difficult to make. Nevertheless, an overview is
important.

6.1 Drilling of Grout Holes

Most grout holes are drilled with an EX bit, which results in diameters of
approximately 1½ in. The holes are usually drilled with rotary drilling rigs, and
diamond bits are used for rock.

In rock, the grout holes are always washed out and pressure-tested. The volume
of water inflow is measured and plotted against a log of the hole. A grouting plan is
then made after inspecting the adjacent holes also. The "Lugeon unit" is sometimes
used, 1 Lugeon unit being equal to a water take of 1 liter per meter of hole per minute at
10 bars of pressure. Since this is about 150 lb/in2 of pressure and normal grouting
pressures are much lower, a refined value is used as follows.

10 (bars )
Lugeon value = water take [1/( m.min)]
actual pressure (bars ) . . .(3)
150 (lb / in 2 )
= 0.0107 water take [ ft 3 / ( ft.min) ]
actual pressure (lb / in 2 ) . . .(4)

Materials allowing only 1 Lugeon of water take are solid enough as foundations
that grouting is not necessary, materials allowing 10 Lugeons generally need grouting,
and materials allowing 100 Lugeons have open joints and large voids that must be
grouted.

6.2 Injection Methods

1. Bottom-up grouting, shown in Fig.3a, drives the casing to full depth, withdraws
it to a specified elevation, and then injects the grout. The process continues
upward to the surface of the material being grouted.

2. Sealed-in sleeve pipe injection shown in Fig.3b allows for several injections at
the same elevation. Injection is done by a double packer, which is inserted in the
sleeve pipe to the desired location. Although the method is more complicated
than bottom-up grouting and the initial cost during drilling is greater, it allows
for independent drilling and grouting operations and the selection of different
grout viscosities as the grout take varies at different locations.

3. Simultaneous drilling and grouting, shown in Fig.3c, allows for injection during
rotary drilling. At a predetermined distance, the drill rod is withdrawn and grout
is injected into the soil or material through a separate drill rod. The process
continues from the top downward. It is only suitable for pervious, granular soils.

132
 IMPROCE 2009

Figure. 3. Various types of grout injection methods (After Caron et at, 1975)

6.3 Grout Plant Equipment

The mixing plants and delivery systems for suspension and solution grouts differ
mainly in their storage and mixing configurations. A typical cement (suspension)
grouting plant is shown in Fig.4. In essence it consists of a mixer, an agitator, a pump,
and piping connected to the grout holes.

For solution grouts, separate ingredients are stored in stationary tanks or tank
trucks and metered out (on a flow volume basis), mixed at junction points, and brought
to the intended grout pipe.

Figure 4 Typical cement (suspension) grouting plant. (USBR memo, 1957)

133
 IMPROCE 2009

6.4 Grout Injection Measurements:

The Basic information to be measured during the grouting process is weight for
suspension grouts or volume for solution grouts, and grout flow rate and pressure for
both. Accurate measurements of weight or volume are required to confirm mix
proportions. During injection, the flow rate of the properly mixed grout should be
constantly monitored and plotted against the grout pressure to understand what is
happening below ground. Typical response curves of flow rate versus pressure and flow
rate versus time are shown in Fig.5. By means of such curves, one can follow on a
particular job how the soil or rock is responding to the grout being injected
a. For permeation, or penetration, grouting, flow rate-versus-pressure and flow rate
- versus-time curves are parallel to each other (resulting in a nearly linear
pressure-versus-time response) at flow rates, which depend upon the porosity of
the soil or the sizes of the crack openings in the rock to be grouted. The flow
rate, of course, always decreases with time since the voids are being gradually
filled with the grout. The weight or volume of grout injected depends on the
structure. of the material being grouted.
b. For compaction, or controlled displacement, grouting, the response is very sim-
ilar except that generally lower flow rates are to be expected. As with
permeation grouting, those rates decrease with time as the in situ formation
being grouted becomes densified within the limits of the equipment and grout
being used.
c. For hydro fracturing, or uncontrolled displacement grouting, the responses are
markedly different than for the other two functions. Here a pressure is always
arrived at where the tensile strength of the soil or rock fails and a lens of grout
shoots into the vacated space.

7 GROUTING DESIGNS AND LAYOUTS

Depending upon the material to be grouted, (i.e. whether it is rock or soil), its
quality. (i.e., in terms of fracturing or density), and the purpose for grouting (i.e., for
seepage control or strengthening), a number of grouting designs and layouts are
possible.

Figure 5 Grout project control curves showing how soil or rock is responding to
injection.
(a) Permeation (penetration) grouting; (b) Compaction (Controlled displacement);
(c) hydrofracturing (uncontrolled displacement) grouting.
134
 IMPROCE 2009

8 GROUT MONITORING SCHEMES

Since most grouting occurs at subsurface elevations where its behavior can
usually only be postulated (i.e., often only guessed at!), the normal way of monitoring a
grouting program has been to measure the quantities of materials injected. Flow rates,
and grout pressures. By monitoring we mean making a positive assessment of the
results of the injected grout. Table 4 summaries the grout monitoring procedures at
various stages of grouting.In general, the following surface-deployed methods have
been used to assess the completeness and adequacy of grouting: Standard penetration
testing, Cone penetration testing, Seismic refraction, Electrical resistivity, Ground-
probing radar and Inductance surveys.In addition, the following borehole-deployed
methods have also been used: Piezometers, Pressure meter testing, cross - hole seismic
testing and Cross - hole radar.

Table 4 grout monitoring procedure at various times during the grouting process
(after Huck and Waller)

Grouting Minor monitoring effort only Intense monitoring effort

activity
Prior to grouting Inspect equipment
Set elevation survey points
Establish monitoring plan and
procedures
During drilling
Grout materials Certificates of compliance
Trial grout mixes
During grout Monitor injection pressure and flow Monitor and record injection
- rate
Grout samples for gel time and chart) Grout samples for gel
storage
Plot grout-take log Heave
measurements on survey points
After grouting Final heave survey
Final review and signoff
Inspect equipment
Set elevation survey points
Establish monitoring plan and
procedures
Conduct pre grout radar and
cross-hole acoustic surveys
Certificates of compliance
Trial grout mixes Independent
laboratory tests
pressure and flow rate (strip
time and storage Plot grout-take
log Heave measurements on
survey points In situ
deformation measurements as
appropriate (inclinometer and/
or in situ strain) Pore-pressure
data In situ resistivity Acoustic
emission monitoring for hydro
fracturing
Final heave survey Post grout
radar and acoustic survey Final
review and signoff.

9 SUMMARY

If the soil present at a site is not suitable for engineering construction, ground
improvement is the best choice to make use of that site. The choice of method of ground
improvement depends on various factors. Among the various ground improvement
methods, grouting is a popular method to tackle various situations.
135
 IMPROCE 2009

REFERENCES

1. F.G. Bell, “Engineering Treatment of Soils”, 1993

2. Robert.M. Koerner, “Construction and Geotechnical Methods in Foundation

Engineering”, 1985.

3. Manfred. R. Hausmann, “Engineering Principles of Ground Modification”, 1990.

4. Camberfort. H, “Injection des solos”, vol.1, “Principles and Methods”, vol 2,

Applications Eyrolles, Paris. 1964.

5. Camberfort. H, “The Principles and Applications of Grouting”, Quart. J.Eng.Geol,

vol 10, 197, pp 44 – 63.

6. Caron. C, T.F. Herbst and P. Cattin, “Injectioms”, Chapter 9 in Foundation

Engineering Hand Book, H.F. Winterkorn and H.Y. Fang (eds), Van Nostrand
Reinhold, Newyork, 1975, pp 337 – 353.

7. Huck. P.J. and M.J. Waller, “Quality Control for Grouting”, Foundation
Engineering Hand Book, H.F. Winterkorn and H.Y. Fang (eds), Van Nostrand
Reinhold, Newyork, 1975, pp 781 – 792.

136
 IMPROCE 2009

GEOSYNTHETICS - A SMART MATERIAL FOR GEOTECHNICAL ENGINEERS

Dr. K. Ilamparuthi *

1. INTRODUCTION

The diverse nature of sub-soil conditions such as soft clays, highly compressible
organic clays, collapsible deposits and expansive clays creates a variety of problems for
the construction engineers. In general these deposits are termed as problematic deposits
(poor soils) because of their low bearing capacity, low permeability and high
compressibility. The increase in activity of building industry over last two decades to
keep phase with industrial growth has resulted demand for construction of Civil
Engineering Structures in poor sub-soil conditions (marginal lands). Geotechnical
Engineer is often consulted to deal with problematic soils and unlike in other branches
of engineering he cannot replace totally the non-functional or damaged part, but has to
remedy it. Variety of ground improvement techniques are in practice to stabilize
problematic soils. But, highly variable soil conditions at each location and the limitation
of traditional ground improvement methods have lead the Geotechnical Engineer to
search for an unique ground improvement method. One such alternative is application
of new class material known as Geosynthetics. The innovation of this material is a boon
to practicing Geotechnical Engineer in solving variety of problems associated with
problematic soils.

2. GEOSYNTHETICS AND THEIR PROPERTIES

Geosynthetics are polymer-based materials that are synthesized for use with
poor geological materials like soil and rock or any other Geotechnical engineering
related materials to improve their behaviour or modify the functional characteristics.
Geosynthetics are commercially available in different forms. They are Geotextiles,
Geogrids, Geomembranes, Geocomposites, Geonets, Geomats, Geowebs etc.

Geotextiles are product of textile industry and include non-woven, woven,

knitted and stitched bonded textiles. The textile manufactured from polymers and fibre-
glass are more durable and easy to manufacture in factories as per specifications hence
are more commonly used in Geotechnical applications. Woven geotextiles are
manufactured from circular or elliptical shaped fibres arranged essentially at right
angles to one another in varying configurations, the general structure is shown in Figure
1. Non-woven geotextiles are produced by random arrangement of short or continuous
fibres bonded together by heat or physical entanglement. The structures of non-woven
geotextiles are shown in Figure 2.

Woven geotextiles are generally used for reinforcing the soil. As a reinforcement
the geotextile provides additional strength to the soil. Depending on the fibres used and
the method of weaving technique the strength of woven fabrics can be as high as 1100
KN/m at 5% elongation. The non-woven geotextiles are used for filtration and drainage
works because of their high porosity. Even though they are thin and low strength they
can act as separators. The filtration characteristics of non-woven geotextiles made them

* Professor & Head, Division of Soil Mechanics and Foundation Engg, Anna University Chennai, Chennai - 25 137
 IMPROCE 2009

distinct from that of woven type. Wovens have uniform size/shape openings whereas
the non-wovens are characterized by a wide range of pore size/shapes.

Geogrids are formed either by the cross laying of strips which are subsequently
bonded together at their cross over points or by the punching and drawing of polymer
sheets. Geogrids are open structure frequently presenting more as 30 percent open area.
The open structure (apertures) of geogrids helps them to interlock with surrounding
soil/aggregate. They are characterised by high strength and modulus at very low
elongation. They are of two types. Viz. Uniaxial and biaxial geogrids (figure 3), with
high strength in one or both directions. They are used for reinforcing the soil.

Geomembrane is a thin polymer sheet composed of materials of low

permeability to control fluid migration. Geocomposites generally consists of high
strength fibres set within a polymer matrix or encased within a polymer skin. The fibres
provide tensile strength properties for the material while the matrix of skin provides the
geometrical shape and protects the fibre from damage. There are two common types of
geocomposite structures, strips and bars as shown in Figure 4.

Geonets/Geomeshes are extruded polymer meshes resemble like geogrids. The

strength of geonets is extremely lower hence they are used to drain water in the
horizontal plane. They also provide space between non-woven geotextiles to minimize
clogging or between geomembranes to control possible leakage from the membranes.
The arrangement of geonet is as shown in Figure 5.

Geomats and geowebs could be coarse woven textiles or joints obtained by

thermal heating, made of strips, filaments or extruded. They are generally flexible and
junctions of overlapping strips not firmly connected. Typical geomat is shown in Figure
6(a). There are also mattresses (Figure 6(b)) commonly used in erosion control as well
as strengthen soil. The double-layered fabrics infilled with sand-cement mattresses are
also used as canal liners to control erosion.

The form of geosynthetics required for various Geotechnical works depend on

the nature and life of the structures. The properties required for permanent structures are
likely to be different from these required for temporary structures Typical - properties of
some of the polymer geotextiles are presented in Table 1.

3. FUNCTIONS OF GEOSYNTHETICS

Geosynthetics have become popular due to their multifunctional capacity viz as

separator, reinforcement, drainage and Alteration, slope and bed protection (erosion
control). Giroud (1978) lists sixteen functions and in a given application, a geotextile
can perform more than one function. The functions and applications are listed in Table
2.

Drain

The geotextile is placed in a soil of low permeability through which water is

seeping slowly. Geotextile collects the water and permeates it towards outlet. Typical
chimney drain used in earth dam is shown in Figure 7(a).

138
 IMPROCE 2009

Water Proof Membrane

Geotextiles coated with impermeable material such as asphalt or plastic are used
to control seepage of contaminant and gas. An example is pond lining (Figure 7(b)).

Solid Filter

The geotextile is placed between fine grain soil, which has to be drained and
coarse grain soil which constitutes the drain. The geotextile permits water to drain but
prevents movement of fine particles. Typical trench drain is shown in Figure 7(c). A
bank protection is an example of two way dynamic flow (Figure 7(d)).
Table 1: Representative Properties of Geosynthetics
Tensile Extension at Apparent Unit
Geotextile Water flow
Strength, max. load opening size weight
Construction litre/m2/sec*
kN/m % mm g/m3
Conventional
3-25 20-60 0.01-0.35 25-150 70-350
Nonwovens Melt-
7-90 50-80 0.03-0.20 30-200 150-2000
bonded Needle-punched
Wovens Monofilament 20-80 9-35 25-2000 150-300
0.07-2.5 0.20-
Multifilament** Flat 40-800 9-30 20-80 250-1350
0.9 0.07-0.15
tape 8-70 10-25 5-20 90-250
Knitteds 2-5 300-600 0.2-1.2 60-2000 100-
Weft Warp 20-120 12-15 0.4-5 2000
Stitch-bonded 30-1000 8-30 0.07-0.5 30-80 250-1200
Special
Geogrids Cross-laid 25-200 3-20 50-300 NA 300-1200
strips Punched sheets 10-200 11-30 40-150 NA 200-1100
Geocomposites Strips 20-150# 20- 3-20 300-1200
50-300 NA
Bars Link structures 500 100- 0.5-20 NA 600-
NA NA NA NA
4000 3-20 4500
* normal to the plane of the geotextile with 10 cm constant head
** fibrillated tapes are included in this category
# measured in kN (not kN/m) NA not applicable
Table 2 Functions And Applications Of Geosynthetics
Application Primary Function Product
Soil Reinforcement Reinforcement Geotextiles/Geogrid
Vertical Walls Reinforcement Geotextiles/Geogrid
Embankments Reinforcement Geotextiles/Geogrid
Steep slopes
Stabilisation of sub-grade Reinforcement, separation and Geotextiles/Geogrid
Alteration
Railroad track bed stabilization, Drainage/Separation, Stress Geotextiles/Geocomposites
Asphalt over lay relieving, water proofing
Subsurface Drainage Filteration Geotextiles
Sedimentation Control/Silt fence Sedimentation Retention, Geotextiles
Filteration/Separation
Erosion Control Filter/Canal Filteration/Separation Seepage Geomattresses/Geotextiles
lining Control
Surface Erosion Control Turf Reinforcement Geomats
Subsurface Drainage Filteration/Fluid Prefabricated drainage
Transmission/Radial composites
Consolidation
Geomembrane Protection Protection/Cushion Geotextile

139
 IMPROCE 2009

Liquid Filter

The geotextile is placed across a flow of water carrying fine particles.

Geotextiles control the migration of particle and allows water to seep through it. The
best example is settling tank for ground water recharge (Figure 7(e)).

Support

The geotextile acts as a support when it is placed between waterproof membrane

and a material containing void spaces. The geotextile prevents the bursting of
membrane. over the voids. Example: Lining of cracked canal and pavement (Figure
7(f)).

Separator

Geotextile acts as a separator when it is placed between two soils which have a
tendency to mix with each other by external load. Further they also prevent the pumping
of mud under moving loads. Figure 7(g) shows the arrangement of Geotextile under a
ballast.

Surfacing

A geotextile is placed on the ground as shown in Figure 7(h) (Helicopter pad).

Its function is to provide a flat and clean surface for traffic.

Curtain

The geotextile is hung alongside on earth or rock mass. Geotextile prevents the
passage of light and falling of rocks. An example is a rockfall net (Figure 8(a)).

Membrane

Geotextile placed between two materials having different pressures contains the
materials having the highest pressure. An example is the use of a geotextiles in unpaved
road to prevent upheaval of soft subgrade soil between the wheels (Figure 8(b)).

Tie

The geotextile is attached to two earth, rock or concrete masses, which have a
tendency to move apart. An example is tie-back wall (Figure 8(c)).

Binder

The geotextile is placed directly over the surface of bank soil which has a
tendency to move apart when they are subjected to small mechanical actions. Its
function is to keep them together. An example is erosion prevention of a slope (Figure
8(d)).

140
 IMPROCE 2009

Reinforcement

The geotextile is placed on a soil which is not able to withstand the tensile loads
applied on it. Its function is to provide tensile strength through interface friction. Figure
8(e) shows the geotextile reinforced retaining wall.

Absorber

The geotextile is placed on a solid mass submitted to shocks and vibrations from
outside. Geotextile reduces the intensity of shock and vibration transmitted to it. Inter
position of a geotextile between railroad ties and a bridge back is an example (Figure
8(f)).

Crack Barrier

The geotextile is placed between two materials which have a tendency to exhibit
cracks. The function of geotextile is to prevent cracks from one material to the other.
Typical example is prevention of reflective cracking in road pavements (Figure 8(g)).

Bond
Geotextiles are placed between two materials which must not have differential
movement. Geotextile placed between them increases the interlocking. Figure 8(h)
shows the bonding of Geotextiles in pavement.
Lubricator

Geotextiles are also used to reduce the interlocking between the two materials.
An example is multilayer concrete - geotextile - geomembrane composite for a canal
lining where differential movement is expected (Figure 8 (i)).

4. STABILITY OF REINFORCED EMBANKMENTS ON SOFT SOILS

Various applications of Geotextiles in the field of Geotechnical Engineering

were illustrated in the earlier sections. Brief design methodology of embankments on
soft soil is dealt below.

The construction of embankments poses severe problem on soft soil deposits.

Several alternative techniques for improving the stability of embankments of soft soil
have evolved over several decades. Comparatively recent practice of constructing
reinforced embankments has provided an added dimension to researchers in the field of
Geotechnical Engineering.

A summary of the analytical procedures for the stability analysis of reinforced

embankments on soft soils developed by Kaniraj (1996) using limit equilibrium is
presented.

5. MODES OF FAILURE

Safety against five modes of failure, namely, bearing capacity, sliding,

spreading, foundation soil squeezing, and rotational failures should be considered in the
design of reinforced embankments on soft soils. The design consists of selecting an
141
 IMPROCE 2009

economical side slope and reinforcement which will make the embankment stable in
these modes of failure. The author (Kaniraj 1988) has explained a systematic approach
to this objective.

6. REINFORCED EMBANKMENT

Among the different modes of failure, the rotational failure is often the critical
mode, which governs the design of reinforced embankments on soft soils. The failure
surface is generally assumed to be an arc of a circle. The critical failure surface and the
maximum required reinforcement force, pmax to achieve the target factor of safety F, are
usually determined by limit equilibrium method. Total stress approach is used in the
analysis since the short term stability immediately after construction is more critical
than the long term stability.

The general case of reinforced embankment on soft soils has been considered.
The details are shown in Figure 9. The main features of this general case are:

1. The possibility of occurrence of tension crack in the embankment is considered. The

height of tension crack can vary from 0 to the full height of the embankment. The
tension crack is assumed to be dry.

2. The presence of stabilizing berm is considered. The berm is assumed to be made of

the same material as the embankment. It also has the same side slope as the
embankment.

3. The presence of an excavation outside the berm is considered. The effective weight
of the soil removed from the excavation is considered to be a destabilizing force
acting through the center of gravity of the excavation.

4. The reinforcement is a single layer reinforcement placed within the embankment at

a distance 'a'' above the ground surface. For reinforcements kept directly on the
ground, a = 0.

5. The direction of reinforcement force depends on the orientation of the reinforcement

at the point where the failure surface intersects the reinforcement. The assumption
that the reinforcement force acts in the horizontal direction (α= θ) is very common,
because it gives a conservative estimate of the maximum required reinforcement
force. The other directions that have been used are, tangential to the failure surface
(α = 0) (Binquet and Lee 1975, Quast 1983), and in the direction of the bisector to
the horizontal and tangential directions (α = θ/2). Five different directions of
reinforcement force (α = 0, θ/4, θ/2, 3θ/4 andθ ) are considered in the present
solutions.

6. The properties of the embankment and berm material are characterized by shear
strength parameters c and θ, and unit weight y.

7. The foundation soil properties are characterized by undrained strength Cu and θ =

0° The variation of Cu with depth can be considered in the analysis.

142
 IMPROCE 2009

7. PROCEDURE OF ROTATIONAL STABILITY ANALYSIS

The procedure of analysis has been summarized Kaniraj (1996). The following
is a restatement of the relevant portions from that paper.

As shown in figure 9, the failure plane is assumed to be a circular arc tangential

to a limiting tangent a depth D. The failure surface encloses the excavation and the berm
and terminates at the bottom of the tension crack.

The origin of the co-ordinate axes is at the intersection of the limiting tangent
and a vertical line passing through the toe E, of the embankment. X0 and Y0 are the co-
ordinates of the center of the slip circle.

The factor of safety of the reinforced embankment F is defined as

Total resisting moment , M r

F=
Total overturning moment , M 0 . . .(1)

Mr Consists of three components, namely

Mr = Mrf + Mre + Mrr . . .(2)

Mrf = moment due to resisting forces in the foundation soil along slip surface NMJ
Mre = moment due to resisting forces in the embankment along slip surface J11, and
Mrr = moment due to reinforcement force P.

M0 consists of four components, namely

Moe = Moe + Moe - Mob + Mox . . .(3)

Moe = moment due to the soil mass EG1I1J in the embankment

Moc = moment due to soil mass in G1GI I1
Mob = moment due to the soil mass ABCE in the berm
Mox = moment due to the soil mass in the excavation

The equation for Mrf is given by Low (1989). The expressions for the other
components of resisting and overturning moments are given by Kaniraj (1994).

From Eqn (1) the reinforcement force can be expressed as

M 0 F − M rf − M re
P=
La . . .(4)

La = moment arm of P about the center of the slip circle.

The value of La depends on a. The expressions for La for five different a are also
given. (Kaniraj 1996) Tentative recommendations for the selection of a was also given
in the paper.

143
 IMPROCE 2009

Equation (4) is differentiated with respect of X0 and Y0, separately and equated
to 0. The solutions to the two equations give the equations for the co-ordinates of the
center of the critical slip circle. On substituting these expressions for X0 and Y0 in Eqn
(4), the expressions for maximum required reinforcement force Pmax, is obtained. The
equations for X0, Y0 and Pmax are given in non dimensional form elsewhere. The overall
maximum required reinforcement force Pmax, can be determined by considering
different limiting tangents.

8. CONDITIONS TO BE SATISFIED

For the equations to give valid solutions, three conditions resulting from the
assumptions made in the analysis should be satisfied. These are:

a) The center of slip circle must lie at or above the bottom of the tension crack.
b) The entire berm and the excavation should lie within the failure plane.
c) The terminal point I1, of the failure plane should lie below the crest and not
below either of the two side slopes.

The expressions for these three conditions are given by Kaniraj (1994).

9. PROCEDURE OF DETERMINING Pmax

The following is a systematic procedure of determining Pmax

Step 1: choose a value of limiting tangent D such that D>0.5H.

Step 2: Select the value of a.
Step 3: For this limiting tangent determine the co-ordinates (X0, Y0) of the center
of the critical slip circle, Y0 is also the radius of the critical slip circle.
Step 4: check whether the conditions are satisfied. If not, the design should be
carried out using a trial-and-error slip circle analysis.
Step 5: Calculate the Pmax for the chosen limiting tangent.
Step 6: Choose another value of D. Repeat Steps 2 to 5.
Step 7: Plot a profile of D Versus Pmax. The overall maximum required
reinforcement force could be identified from this profile.

10. CONCLUSION

Various applications of geosynthetics in the field of ground improvement are

highlighted in this paper. Brief design procedure for embankments on soft clay using
limit equilibrium method has also been presented. The use of geosynthetics for soil
reinforcement was initially accepted with some reluctance but after 20 years these
materials are now accepted by construction engineers all over the world as economic
alternative for retaining walls and ground improvement. Polymer technology is
developing rapidly and now advanced forms of geosynthetics are being developed. The
challenge is to develop analytical tools which will permit their wide application.

144
 IMPROCE 2009

REFERENCES

1. Jones, CJFP; Geosynthetics used in reinforced walls and Landslide Protection,

International seminar and Technomeet on Environmental Geotechnology with
Geosynthetics, New Delhi 1996, pp. 3-30.

2. Giroud, LP. (1979), Behaviour Geotextiles, NATO, Advanced Study Institute,

Mechanics of Flexible Fibre Assemblies.

3. Rao, G.V. and Banerjee, P.K. (1999). Geosynthetics Recent developments, A short
course on Geosynthetics and reinforced soil structure, December 5-11, 1999, ITT
(Madras), Chennai, pp. 1.1 to 1.19.

4. Kaniraj, S.R. (1996). Stability of Reinforced Embankments on Soft Soils.

International Symposium and Technomeet on Environmental Geotechnology with
Geosynthetics, New Delhi, pp. 152-160.

5. Koerner, R.M (1986), Designing with Geosynthetics, Prentice Hall, Inc., New
Jersey.

Figure. l Scanning electron photomicrographs of (a) monofilament / tape woven

fabric (b) multiflament woven fabric ( C ) intersection detailing of a woven fabric
and (d) tape (Slit film) woven fabric

145
 IMPROCE 2009

Figure.2 Scanning electron photomicrographs of (a) Needle - Punched fabric (b)

Needle Punched fabric - close view and (c) Thermally - bonded fabric.

Figure.3 (a) Uniaxially oriented (mono - oriented) geo grid and

(b) Biaxially oriented (bi - oriented) geogrid (courtesy'. Texa Geosynthetics)

Fig.4 Composite strip and bar structures

146
 IMPROCE 2009

Fig.5 Typical view of a geonet

Fig.6 (a) Extruded geomat [Courtesy Tenax GeosyntheticsJ

Fig.6 (b) Double layered fabric with spacer yarn used in erosion control (courtesy :
Tenax Geosynthetics)

147
 IMPROCE 2009

Fig.7 (a) Drain (b) Water proof membrane (c) solid filter study flow (d) solid filter
dynamic flow (e) liquid filter support (g) separator (h) surfacing (Giroud J.P.1978)

Figure.9 Wide Crested reinforced embankment on soft soil

148
 IMPROCE 2009

IMPROVEMENT OF PROBLEMATIC SOILS BY THERMAL METHODS

Dr. V.K. Stalin *

1.0 INTRODUCTION

The thermal stabilization includes soil heating and freezing. Ground freezing has
two main functions: (1) to prevent groundwater seepage into excavations and (2) to
increase the shear strength of the soil and improve its structural capacity. Two basic
systems are usually followed in freezing: an open system where the refrigerant (liquid
nitrogen or carbon dioxide) is lost to the atmosphere after it has absorbed energy and
vaporized, or a closed-circuit hydraulic system using a conventional mechanical plant
and a circulating coolant. In either case, the groundwater is frozen and prevented from
entering the excavation, and the shear strength of the soil is increased as the ice acts as a
binding agent to the soil's particles. Ground freezing can be applied to a wide range of
soils. Soil heating is performed at the site using electrical methods(graphite electrodes
are heated to high temperature-vitrification) by pumping hot air blown into the
predrilled holes. In the case of shallow heating, heating electrodes are used such as
wood or any other non-toxic elements that can generate heat. In vitrification, the soil is
electrically melted at very high temperatures, typically in the range of 1,600 to 2,000
degrees Celsius. This is accomplished with graphite electrodes used to conduct
electricity through the soil. As the soil is melting, the flowing electricity is converted
into heat that moves outwards and melts new soil. The melted soil becomes electrically
conductive and forms a heat-transfer medium allowing the melt to move downward and
laterally through the soil. The inorganic portion of the soil typically breaks down into
major oxide groups, such as silica and alumina. Upon cooling, these groups form glass
and crystalline products with excellent environmental properties. The details of factors
influencing thermal methods and their advantages are discussed in the following
sections.

2.0 FACTORS INFLUENCING THERMAL STABILISATION

The effectiveness of thermal stabilization is influenced by the following issues:

(a) the degree of saturation of the soil, (b) the rate of groundwater movement, (c) the
creep potential of the frozen ground, (d) the post thawing behavior, (e) the heat transfer
in the melted soil and (f) the impact of heat on utilities and adjacent structures.

2.1 Degree of Saturation

Since the formed ice lens binds the soil's particles to increase strength, partially
saturated soils are weaker when frozen than fully saturated soils because less ice lenses
are formed. Water (or slurry) is sometimes added to certain soils above the groundwater
table to improve their frozen properties.

2.2 Groundwater Movement:

Moving groundwater presents a potential major problem for ground freezing.

When the groundwater velocity is less than 600 mm per day, the ground freezing
* Assistant Professor, Department of Civil Engineering, Anna University Chennai, Chennai - 25 149
 IMPROCE 2009

process is not affected. In faster-moving ground waters, such as those affected by tidal
or river fluctuations, the freezing process is affected, but can be dealt with in the design.
When the water velocity reaches 2 m per day, ground freezing becomes essentially
ineffective.

2.3 Creep of Frozen Ground:

When stressed for a long time, the frozen ground may experience deformations
and loss in the initial strength developed upon freezing. Clayey soils, for instance,
encountered on a shaft freezing project, exhibited a radial inward creep of the order of
38 mm per month under the load. Granular soils at similar depths showed negligible
deformations (Donohoe, 1993). One reason for the creep is that the water adsorbed on
the clay mineral's surface, which does not freeze at 0°C, freezes gradually as the
temperature is lowered substantially below 0°C.

2.4 Post - thawing Behavior:

Post thawing behavior of the frozen ground is still not well- understood, with
conflicting opinions voiced regarding the potential post thawing ground settlement.
Through a laboratory testing program performed on samples of the Boston Blue Clay,
before and after one cycle of freezing, Swan and Green (1998) illustrate signs of
disturbance of the initial structure of the soil due to micro cracks caused by the
formation of the ice lenses, accompanied by up to 60 percent reduction in the undrained
shear strength of the soil. The measured compressibility of the thawed out soil was also
higher than that of the untreated soil due to the disturbance caused by freezing. Wood
(1976) reports a 44% reduction in the undrained shear strength of the Leda clay after
one cycle of freezing, increasing to 66% after few cycles. Through scanning electron
microscopy, he determined that the loss of strength was due to the expansion of micro
fissures during the pore water freezing. The post-thawing soil behavior is influenced by
the freezing rate, surcharge loading and drainage characteristics of the treated soil.

2.5 Heat Transfer:

Transfer of heat through the melted soil is a key issue in vitrification. The
electric conductivity of the melted soil, and the rate in which the heat transfers through
it, affect the extent and rate of advancement of the soil's melt-down. These issues are
still in the experimental stage.

2.6 Impact on Utilities and Structures:

Because of the extreme temperature used, it is prudent that "no" objects made of
metals, polymers, geotextiles, etc are present at the site or in its-vicinity prior to the
application of the electric current. The intensity and distance of heat transfer in the soil
should be tested prior to the application of vitrification to protect against any potential
negative impact on adjacent facilities

150
 IMPROCE 2009

Figure1.Vitrification process (Soil heating through electrodes)

3.0 SOIL STABILIZATION BY HEATING

Thermal treatment or thermal stabilization is defined as an irreversible process,

which brings about an effective increase in the shear strength of soil. Thermal treatment
alters the physical and mechanical properties of clayey soils, in particular. Generally
speaking, the application of heat to the soil helps in accelerating the drainage of soil,
which leads to quicker consolidation hence resulting in increase in the strength of the
soil.
In comparison with other soil improvement methods, thermal stabilization
produces immediate results. Thermal treatment of clays brings about several changes in
their properties such as strength, cohesion, internal friction angle, resistance to abrasion
etc. However, a significant change in the properties of most soils is brought about only
when it is treated at temperatures close to fusion.

3.1 Applications of Ground Heating

Heating has been used since primitive times as a means of stabilizing soil and
improving soil properties. Thermally treated soils have been used for making trails for
access and bricks for dwelling. Aggregates produced by thermal treatment provide
durable and economic substitutes for gravel and crushed rock. These are then used for
pavement construction particularly in areas where construction materials have to be
exported at excessive costs.
Heating of clays with additives to form bricks for construction is widely
recognized. Heating is used for manufacturing Ferro-clay blocks (the soil is mixed with
2% alkaline ferric oxide and is burnt at temperatures from 500°C, for kaolinite, to
750°C, for montmorillonite clays. The mixes are molded and cured in air, achieving
strengths in excess of 10MPa within 1-2 weeks. Ferro-clay production has been likened
to natural laterization of soils and represents a combination of thermal and chemical
stabilization process. The savings in energy achieved by burning clay at less than
standard brick temperatures makes ferro-clay an attractive alternative building material.
Heating has also been used effectively for the treatment of buried wastes and is
referred to as “In-situ Vitrification” of the waste. First used in 1980, this waste
remediation technique was carried out using rectangular configuration of vertically
embedded graphite molybdenum electrodes to provide thermal energy to the soil.
Downward melting of soil was accomplished by resistive heating, initiated when an
electric potential was applied across a glass frit. After a melting period of around 2

151
 IMPROCE 2009

weeks, vitrification was completed to a depth of around 6m. The vitrified product
formed was chemically durable and highly resistant to leaching. However the use of
graphite rods for in-situ vitrification has been hindered by high operating cost and long
time for magma initiation.
Plasma technology is a relatively new advent that can provide that can provide
extremely high temperatures for long duration and is capable of melting any and all
earthly materials. It is used a high thermal treatment for in-situ applications involving
ground modification and geo-environment remediation. In this method plasma torches
are used to melt soil into magma in a quick and efficient manner. Within 5 minutes
exposure to a plasma flame positioned in a down hole configuration, the very high arc
temperatures of 4000 to 7000° C creates a pool of molten lava at the center of each soil
deposit, Subsequent cooling of the melt results in the formation of artificial igneous &
rock products and microcrystalline granites and basalts possessing very high strength.
4.0 METHODS OF HEATING FOR IN-SITU SOIL:

Heat can be generated by the burning of liquid, gaseous or solid fuel and is
transferred to the soil through surface contact or boreholes.
4.1 Ground surface heating
A pioneering effort in using heat treatment for road making was made by Irvine
in 1930 in Australia. He designed and built a wood fired furnace which traveled at the
rat of 2 to 10m/hr. on a previously prepared soil formation, producing a 50 to 200mm
thick baked layer of a firm non-plastic material out of very moisture sensitive clay.
After, thermal modification, the surface material was left in interlocking pieces of brick
varying in size from 100mm downward. Large particles were cracked and reduced to
pea gravel after compaction with roller. Bitumen or tar surfacing was applied while the
pavement was still warm, securing good penetration of binder into the cracks.

4.2 Heating through boreholes

Heating is brought about by:
a) Open type firing system; and
b) Closed type firing system
Heat treatment via boreholes has been reported in USSR, Rumania and in Japan.
In a closed type system, a burning unit is place over each borehole. In a project near
Fukuoka, a waste deposit was treated by heating 227 boreholes, 2 to 6m deep, and space
5m apart. The heat was maintained for 7 to 15 days burning an average total of 760L of
fuel oil. Observations indicated that a volume, within a radius of 1m, around the
borehole was effected, resulting in a fuel consumption of 60L/m3
A special version of the open type firing system was used to stabilize an
embankment at Kanazawa, Japan in the year 1971. Over a period of 10 days, 1200L of
fuel was burnt per hole by conventional oil burners until temperatures of around 300°C
was reached about 300mm from the center of the hole. The heat-treated soil showed
strength equivalent to 10 to 20 times the original value.

152
 IMPROCE 2009

5.0 SOIL STABILIZATION BY GROUND FREEZING

For over a century freezing has been used successfully to stabilize ground and to
control flow of ground water. First employed, by the mining industry in England in
1862, it has since then been widely adopted though out the world to stabilize earth (of
all types) of projects including deep excavations, mine and tunnel shafts and structural
underpinning during construction.
Modifying the ground through artificial freezing has great potential for solving
difficult foundation problems, because it promises very high gain in strength and
effective seepage control in a matter o hours or days.

Unlike most other methods of ground modification, which are restricted to

specific soil types, freezing is suitable for a wide range of soil or rock types. In fact, it is
suitable in any formation where freezing of pore water creates a bond between
individual particles and where the freezing process in not inhibited by significant water
flow.
However, the major draw back of the freezing process is that it has to be
maintained for some time in order to remain effective. Thawing of soil may make the
soil even weaker than before.
5.1 Principle of ground freezing
The principal of ground freezing is the use of refrigeration to remove heat form
the ground and convert in-situ pore water to ice. Pore water, which would have
otherwise led to a decrease in the shear strength of soil, now in the form of ice, becomes
a bonding agent fusing together adjacent particles of soil or blocks of rock to increase
the combined strength and make them impervious. Figure 2 shows the schematic
representation of ground freezing.

Figure 2. Alternative refrigeration methods for ground freezing

153
 IMPROCE 2009

5.2 Applications of freezing

Freezing of soils has a wide range of applications in solving different problems

related to soils. Freezing is especially effective in remedying unexpected subsurface
construction problems. Caving tunnel heading, torn sheet pile interlocks, leaking slurry
wall joints and excessive ground water flow are examples of construction problems
which can be controlled by the use of rapid freezing techniques. Fig 3 shows the
application areas of ground freezing.
Ground freezing may be used in any soil or rock formation regardless of its
structure, its grain size or its permeability. It is adaptable to any size, shape or depth of
the excavation. As a result soft as well as the boulder ground where other techniques of
stabilization may not prove effective favor freezing.
Soil and rock profiles, too heterogeneous to grout predictably, are easily frozen.
Freezing may be used successfully where pile driving or dewatering is difficult and
might lead to unwanted settlements.

Figure3. Application areas of ground freezing

a. Sinking tunnel b. Advancing tunnels in running ground

c. Providing direct underpinning d. Providing lateral restraint for excavation.

5.3 Advantages of Freezing

It has been noticed that ground freezing is readily accomplished where other
methods may be difficult. The ground freezing creates no smoke, vibration shock or
undue noise so work can proceed round the clock even in residential areas. It also does
not affect the existing water table and hence eliminates concern for adjacent water wells
or subsidence.
154
 IMPROCE 2009

As the earth itself is the principal structural element in ground freezing, very few
other materials are needed. Alternative construction techniques, especially for
excavation support require large amounts of timber, concrete and steel which are
becoming increasingly scarce and costly.

6.0 CASE STUDIES ON FREEZING

A better understanding of the role of freezing in changing soil properties can be

achieved by studying in brief three cases where ground freezing was used to stabilize
the soil through which tunnels were constructed.

Case 1: Temporary tunnel support for Georgia tunnel, 23m wide passing
beneath multiple set of road tracks, the fill below the road tracks was primarily
composed of loose micaceous silty-sand, having an average SPT of 5 blows/ft. The
unfrozen fill had a drained angle of internal friction of approximately 25° with no
cohesion. Moisture contents in fill varied from approximately 18%-36% with an
average of 23%. The average void ratio in the fill was approximately 0.9 main features
of the project are presented in Table1 (Jones and Brown,1978)

Table 1: Soil Parameters before and after freezing

Soil Parameters Before freezing After freezing

E(kN/m2) - -
Bulk density (kN/m3) 1,842 1,842
Poisson’s ratio 0.3 0.2
Friction angle 25° 0°
Cohesion (kN/m2) 0 1,101

Case 2 :Construction of New York Tunnel passing below mainline road track,
1.5m below the ground through cinder and sand fill having an SPT of 2 blows/ft. The
soil had average moisture content of 29% and void ration of 1.5. Main features of the
project are presented in Table 2 (Jones and Brown, 1978)

Table 2: Soil Parameters before and after freezing

Soil Parameters Before freezing After freezing

E(kN/m2) 9,576-2,3940 95,760-287,280
Bulk density (kN/m3) 1442 1442
Poisson’s ratio 0.35 0.35
Friction angle 30° 0°
Cohesion (kN/m2) 0 62

Case 3 : Construction of the Washington Tunnel through a subsurface consisting

of sand and gravel with varying clay content and silt contents, having an SPT of 2-50
blows/ft. The soil had moisture content of 34%. This soil was underlain by sand and a
silty gravel strata. Main features of the project are presented in Table 3 (Jones and
Brown, 1978)

155
 IMPROCE 2009

Table 3: Soil Parameters before and after freezing

Soil Parameters Before freezing After freezing

E(kN/m2) 28,728 95,760
Bulk density (kN/m3) 1857 1853
Poisson’s ratio 0.3 0.3
Friction angle 30° 0°
Cohesion (kN/m2) 0 30

On the basis of the above data, the effect of freezing on the soils has been
quantified as:

1. E value increases significantly. This indicates that the soil has stabilized due to
freezing. The increase in E with temperature decrease is in accordance with the
behavior of soils explained previously.

2. The bulk density shows no significant change. This indicates that freezing does not
significantly alter the bulk density of soils except in certain exceptional cases where
pressures melting of inter particular ice may result in compaction and denser
packing of soil particles.

3. The Poisson’s ratio also does not change significantly on freezing. However, it has
tendency to decrease as a result of freezing. The value of Poisson’s ratio is
influenced by the shear modulus of soil as well as E value of the soil.

4. The angle of internal friction decreases as a result of freezing. High value of friction
angle before freezing indicates that the shear strength of unfrozen soil is primarily
due to inter particular friction. However on freezing the cohesion effect due to the
presence of pore ice contributes to strength.

5. Freezing brings about a considerable increase in cohesion of the soil. Increase in

cohesion of soil on freezing is in accordance with the predicted soil behavior. The
shear strength of frozen soils is primarily due to this cohesive component. Increase
in cohesion is drastic which indicates that the strength of soils has increased
considerably on freezing.

All these factors put together explain quite clearly the effect of freezing in
bringing about soil stabilization.

8.0 SUMMARY

In thermal methods, heating of problematic soils permanently modifies the in-

situ soils and freezing is employed until the project is completed. Even though both
heating and freezing methods are relatively expensive, considering other advantages,
these techniques can be used when other methods of soil improvement are not viable.

156
 IMPROCE 2009

REFERENCES

1. Fang, H.Y. (1987), Introduction to Environmental Geotech

2. nology, CRC Press, Boca Raton, New York.

3. Phukan A. (1977), “Foundations in cold regions”, Foundation Engineering

handbook”, CBS Publishers, 735 – 749.

157
IMPROCE 2009

158
 IMPROCE 2009

FLY ASH TO CONTROL SWELL IN SOILS

P V. Sivapullaiah *

ABSTRACT: Swell in soils poses serious problems to foundation engineers. Several

studies focused to understand the phenomenon and to develop methods to deal with the
same. However, there are special instances where unexpected swell occurs. The nature
of swelling is different and usually occurs after a considerable delay period. Unlike the
conventional swell, this type of swelling that occurs in both swelling and non-swelling
soils which is due to presence of sulphate in lime and cement stabilized soils, due to
contamination soil with acid or alkali solutions etc. is typically irreversible. One of the
prominent methods to control is to use pozzolanic stabilizers such as fly ash is
presented. The presentation shows a few instances of effectiveness of fly ash.

1.0 INTRODUCTION

Soil volumetric changes may cause both unfavorable and favorable effects on
human activities. Unfavorable effects are the destruction of buildings, roads and
pipelines in uncropped soils, and the leaching of fertilizers and chemicals below the root
zone through desiccation cracks (by pass flow). In these soils horizontal cracks break
capillary flux of water. On the other hand, swelling clays can be used as liners to seal
landfills.

It is generally understood that while some soils swell, the others do not
swell. Swelling soils contain a high percentage of certain kinds of clay particles that are
capable of absorbing large quantities of water (Hansen and Smith, 2002). Soil volume
may expand as the clay becomes wet. The powerful force of expansion is capable of
exerting large pressures on foundations, slabs or other confining structures. Cracked
foundations, floors and basement walls are typical types of damage done by swelling
soils. Damage to the upper floors of the building can also occur when motion in the
structure is significant. Exposure to changed environmental loading during or after
development results in swelling. Many instances of significant deformations of
structures and equipments are reported in literature due to chemically induced swelling
of soils.

1.1 Swell in Soils

Swell process is essentially governed by moisture migration into the micro

structure. When the micro structural pore space is saturated, the matrix suction
responsible for moisture absorption mainly operates in the macro-structural pore space.
The exchangeable cations responsible for diffuse ion layer formation around the clay
platelets are encountered in the micro-structural pore space (Alonso and Gens, 1994).
Due to isomorphous substitution Exchangeable cations associate with the clay platelet
surfaces in the clay mineral lattice (Yong and Warkentin, 1975). In the dry state the
exchangeable cations are held close to the surface of clay platelets in the microstructure.
Moisture absorption in the macrostructural pore space disturbs the osmotic suction
equilibrium between the macrostructure and micro- structure components.
.Dissipation of osmotic suction gradient between the two levels occur by osmotic flow
* Professor, Department of Civil Engineering, Indian Institute of Science, Bangalore - 12 159
 IMPROCE 2009

of water molecules from the inter-aggregate (macrostructural pore space) to the intra-
aggregate pores (microstructural pore space), with the microstructure operating as
perfect semi-permeable membrane (osmotic efficiency = 1). Osmotic flow of water into
the micro-structure pore space expands the diffuse ion layers around the clay platelets.
Growth of diffuse ion layers mobilizes diffuse double layer (DDL) repulsion that is
responsible for the macroscopic swelling of clays (Bolt 1956, van Olphen, 1963).

1.2 Factors Affecting Swell in Soils

1.2.1 Mineralogy

Soils that swell when water is added are known as swelling or expansive soils.
Many of these clay minerals do occur as the dominant mineral in some areas and almost
all of them occur as minor constituents with other clay minerals. The mineralogy of
clays plays a very important role in the development of diffuse double layer near clay
particles. In case of expansive 2:1 minerals isomorphous substitution, pH dependent
charges and water adsorption are the causative agents for volume change. Water can
easily separate the layers and cause swelling. As a result of negative charges, cations are
absorbed on the surfaces of the expansive clay minerals (Fig.1). Diffuse double layer
theory can explain the extent of swelling in soils.

Fig. 1 Sketch of Smectite Mineral

In case of 1:1 minerals, no significant volume changes are observed due to

strong hydrogen bonding between their interlayers and very less negative charges on
clay particle surface and hence, no adsorbed ions on the mineral surface.. Soils
containing illite and mica, though contain negative charges, exhibit moderate
swelling due to potassium linkages between their layers. Soils containing
montmorillonite, vermiculite, and some mixed layer minerals behave as Expansive
soils. Discrete clay crystals may consist of alternating layers of two or more different
clay minerals, which are referred to as mixed-layer clays. Common combinations
contain the expandable clay mineral montmorillonite or beidellite, interlayered with
chlorite or with mica.
160
 IMPROCE 2009

The cation exchange capacity and the type of exchangeable ions play an
important role in development of diffuse double layer through their charge and ionic
size. The thickness of diffuse double layer decreases inversely with valency of
exchangeable cation, square root of the concentration and increases with the square root
of dielectric constant and temperature. Type and amount of different cations in a
clay-water-electrolyte system have a major influence on double layer interactions.
Changes in these interactions may change the physical and physico-chemical properties
under a given set of environmental conditions. The ease with which an ion can
replace another ion depends mainly on the valence, relative abundance of different ions
and ion size. Thus, these parameters along with specific surface area play a very
dominant role in controlling the volume change behaviour of 2: 1 minerals.

1.2.2 Pore fluid

Experiments conducted by the numerous researchers show that smectite

behaviour is greatly controlled by pore liquid composition. Macroscopic evidence of
these interactions is the change in volume, caused by ion diffusion under constant
external stresses (Barbour and Fredlund, 1989; Di Maio and Fenelli, 1994). Pore fluid
effects can be assessed by diffused double layer and fabric changes. Changes in external
loads produce effects which depend on pore fluid composition, such as electrolyte
concentration, dielectric constant, pH etc (Table

1. The effect of the pore fluid varies with the cation exchange capacity and the type of
exchangeable cations (Di Maio et al 2004; Mitchell, 1993). (Table1)

Table 1: Soil Behaviour Influenced with Environmental Factors (after Fang, 1997)

Environmental Double layer thickness Soil structure

parameter
Increase Decrease Deflocculated Flocculated
pH High Low Increase Decrease
Electrolytic Decrease Increase Decrease Increase
Concentration
Ion valence Decrease Increase Decrease Increase
Dielectric constant Increase Decrease Increase Decrease
Size of hydrated ion Large Small Increase Decrease
Anion adsorption Increase Decrease Increase Decrease

1.2.3 Fabric
The term ‘fabric’ is generally used to denote the physical arrangement of
individual particles in soils, while the term ‘structure’ is used to denote not only the
physical arrangement within soil particles but also their stability or integrity. The
particles may remain relatively independent of each other, but commonly, they are
found associated together in aggregates. The basic particle associations are illustrated in
detail by Mitchell (1993). The particle arrangements for small plate shape are:
Dispersed, Aggregated, Flocculated and Oriented. Dispersed particle arrangement arises
161
 IMPROCE 2009

in presence of high inter-particle repulsion, such as in a low salt content environment,

causing swelling. Face-to-face association, called aggregation, is realized in high salt
concentration environment. Addition of electrolytes causes edge-to- edge or edge-to-
face association. Oriented fabric arises primarily from consolidation effects.

1.2.3 Specific surface area

It is known that the platelets of clay minerals possess internal as well as external
specific surface area due to the presence of exposed oxygens or hydroxyls on faces
and/or other positively charged ions on edges. When water gets in touch with clay
particles the hydrogen bonding of water molecule to the clay mineral surface provides
the basic building blocks for the double layer water which, in reality, is the moving
force in soil expansion. Thus the swelling in soils increases with increase in specific
surface of soils.

1.3 NATURE OF SWELL

1.3.1 Modes for swelling

A two parallel plate model can describe the swelling behaviour of
montmorillonite sheets of clay. These two plates are subjected to repulsive and
attractive forces. The energy for attraction is caused by adhesive and van der Waals
forces, which depend on the orientation of the plates and pH of soil solution. When pH
decreases the charges at the edge of the plate becomes positive and edge bonding
develops, which is more like a plate-like arrangement than parallel arrangement of clay
particles. The repulsive forces between two parallel plates with negatively charged
surfaces consists of forces owing to double layer and forces owing to the differences in
electrolyte concentration between the plates and outer solution. Since the chemical
potential of interlayer is greater than outer solution, addition of water causes the plates
to drift apart till potential difference reaches equilibrium. The total potential energy is
the sum of attractive and repulsive potential energies, which is not necessarily a
monotonic function with distance.
1.3.2 Stages of swelling
Typical swelling of expansive soil is presented in Fig. 2.

Fig. 2 Typical Swell behaviour of expansive soils (after Sivapullaiah et al., 1996).
162
 IMPROCE 2009

The swelling of plate like clay particles can be described by three stages of
swelling. The first stage is when the initial distance between the particles is less than
two nanometers. During this stage, the swelling is opposed by the electrostatic attraction
between cations and negatively charged layers. Swelling beyond 2nm is possible
provided the hydration energy of cation is more than the energy of attraction. With the
addition of water the distance between the plates increases in discrete steps associated
with each molecular water layer formed between the sheets. The swelling continues to
the second stage if monovalent cations are present, however, if divalent and trivalent are
present, swelling ends at the first stage. In the beginning of the second stage, the
bonding of molecules to solid surface continues as swelling process continues. The
distance between the neighboring sheets rises smoothly up to tens of nm and edge to
face force are important for holding sheets together. In the third stage, the sheets are
totally separated and form an arrangement caused by edge to face and edge to edge
forces.

2.0 METHODS TO CONTROL SWELL IN SOILS

Conventional approaches to deal with this kind of swell in expansive soils are
familiar. Though many methods are there to control the swell in soils, none are more
permanent than pozzolanic stabilization of soils using lime, cement and fly ash. In this
presentation the use of fly ash to control the swell in soils is presented.

2.1 Types of Fly Ashes

Fly Ash: Around 75 - 85% of the ash produced from the combustion of the coal
is carried out of the furnace with the flue gasses and is extracted by electrostatic or
cyclonic precipitator. This is known as fly ash. Fly ash is composed of finer particles
and it can be easily blown off by air and hence the name, ‘fly ash’. ASTM C 618-05
defines the fly ash as a finely divided residue that resulted from the combustion of
ground or powdered coal. Fly ash consists of inorganic mineral constituents in the coal
and the organic matter, which is not fully burnt. Bottom Ash: The remaining portion of
the ash produced falls into the hopper placed at the bottom of the furnace. This ash is
called as bottom ash. The bottom ash is composed of primarily coarser and heavier
particles.

Pond Ash: The ash is disposed off using water in the form of slurry. The fly ash
slurry normally contains 70% - 80% water for transportation through pipes for storage
in ponds. Ash settled by sedimentation process is designated as pond ash.

2.2 Pozzolanic Reactivity

In almost all geotechnical applications pozzolanic property of ashes plays an

important role. It is the main property of the agent, which differentiates it from silt. The
term pozzolana is employed to designate a siliceous or a siliceous and aluminous
material which itself possesses no cementations value but in presence of water,
chemically react with calcium hydroxide to form compounds possessing cementatious
properties

Based on pozzolanic property, fly ash can be classified as self pozzolanic,

pozzolanic and non pozzolanic fly ashes. Self pozzolanic fly ashes or self hardening fly
163
 IMPROCE 2009

ashes are those fly ashes which produce cementatious compounds and thus harden with
time in presence of water, without external addition of lime. Pozzolanic fly ashes are
those, which produce cementations compounds on addition of lime. It is the lime
content which differentiates between self-pozzolanic and pozzolanic fly ashes. Non
pozzolanic fly ashes are those which don't produce sufficient cementatious compounds
even on addition of lime. Reactive fiy ashes are those, which reacts lime to give
significant amount strength. Non reactive fly ashes are those which don't give sufficient
strength even on addition of lime. Self pozzolanic and pozzolanic fly ashes are reactive
fly ashes whereas non pozzolanic fly ashes are nothing but non reactive fly ashes.

ASTM C 618-05 classifies fly ashes into Class F and Class C fly ashes. Class F
fly ashes shall be essentially pozzolanic fly ashes. Class C fly ashes shall be essentially
self pozzolanic fly ashes. Pozzolanic reactivity is influenced by - particle size
distribution, fineness, particle morphologies, mineralogical composition, chemical
composition.

Particle size distribution plays an important role in determining the activity of

fly ash. The particles in the size range of about 5 to 30 μm plays decisive role in
determining the activity of fly ash. Smooth spheres of sizes range from 2 to 25 μm are
considered to be more reactive. For low calcium fly ashes, the reactivity was found to
be directly proportional to the amount of particles smaller than 10 μm, and inversely
proportional to the amount of particles greater than 45 μm. However, sole evaluation of
fraction passing through a 45 μm sieve does not by itself appear to be fully appropriate.
Particle size distribution alone may not reflect the potential reactivity of a fly ash,
though the finer the fly ash the higher is the pozzolanic reactivity.

Fly ashes which contain mainly spherical particles, ranging between 2 to 25 μm

and the surface of glassy spheres being predominantly smooth without any deposition
but with only adherence, as highly reactive fly ashes. Fly ash, which contain mainly
spherical particles having size of 2 to 35 μm and surface deposition; and some sintering
are considered to have moderate reactivity. Fly ashes, which contain mainly irregular
shaped particles (being relatively coarser) and small amount of spherical particles, are
considered as poor reactive. But the rough textured and porous glass spheres possess
larger surface area for reactivity than smooth spheres and the particle size distribution
alone may not reflect the potential of reactivity of a fly ash.

The crystalline minerals - quartz, mullite, magnesite and hematite present in fly
ash are non reactive at ordinary temperature. The crystalline minerals like CS, C3A,
C4A3S etc. present in fly ash are highly reactive. High calcium fly ashes generally
contain minerals like CS, C3A, C4A3S. The mineral phases present fly ashes are rather
small and do not appear to be directly related to the pozzolanic properties of the fly
ashes. Significance of fly ash is only because of its source of active silica.

2.2.1 Reactive Silica

The active silica in fly ash is due to the presence of special micro structure of
quartz, having micro - amorphous fibrous silica structure along with some vitreous or
amorphous silica structure. Though ashes with higher contents of silica and alumina
may be thought to have greater pozzolanic potential, what is important is the amount of
silica and alumina, which is actually available for reaction with lime. Fly ash - water
164
 IMPROCE 2009

system is alkaline, only silica and alumina, which are soluble in alkaline environment
can take part in the pozzolanic reaction. The dissolved silica is able to form strength
developing calcium silicate hydrates with calcium oxide. Thus soluble silica contained
in fly ash is a characteristic criterion for the potential reactivity of the fly ash. A positive
correlation exists between soluble silica content and compressive strength. Now it is
clear that the pozzolanic reactivity mainly depends on reactive silica present in glass
phase. However they develop good strength with optimum lime content (Sivapullaiah et
al.1995).

The reactivity decreases with increasing Fe2O3 content in the glass phase. In
most fly ashes, most of the iron is present as unreactive hematite and magnetite and a
small amount is present in the glass phase, which has deleterious effect on its
pozzolanic activity. Carbon content in general appears to suppress the reactivity and rate
of strength gain.

2.3 Fly Ash Utilization

Beneficial properties of fly ashes based on the properties are:

1. Due to its lightweight it has been used for the production of lightweight aggregates.

2. Due to its resistance against variation in water content it has been used for clay
flooring and terracing.

3. Due to its adsorbing nature it has been used as a de-colouring agent in dye effluent.

4. Due to its pozzolanic nature it has been used for replacement of cement in concrete,
manufacture of bricks, manufacture of cement, for cement replacement in
manufacture of concrete blocks and entrapping waste in its matrix.

5. As a raw material for aluminium recovery as it contains significant amount

of aluminium.

6. Due to its fineness and pozzolanic nature it is used as an ingredient in grouting

applications.

7. As a fill material in several geotechnical works such as embankment, fill behind

retaining walls, reclamation fill, dams etc., because of its low unit weight, low
compressibility, faster rate of consolidation and pozzolanic property

8. As a sub-grade material in roads due to its pozzolanic property, good drainage and
resistance to variation in moisture content.

9. For stabilization of soils because of its pozzolanic property and resistance to

variation in water content.

10. To stabilize of hazardous wastes due to its pozzolanic nature.

11. Due to its high resistance, capacity to retain ions and pozzolanic nature, fly is being
used as liner for waste containment facilities.
165
 IMPROCE 2009

12. Due to its resistance to high temperature it is used for lighting mine fires, as a
medium for combustion control on refuse banks and as a refractory ingredient in the
manufacture of iron and steel.

13. Due to its high temperature resistance and poor electrical conductivity, used as an
insulating material for under ground high-tension wires.

14. Because of presence of cenospheres as a filler material in plastics and paints.

15. Because of its alkalinity in the fluid gas scrubbers.

16. Though the fly ash can find applications due to its beneficial properties, its use is
limited because of the lack of knowledge about the properties of fly ash and
environmental concerns.

2.4 Geotechnical Engineering Applications

Fly ash is extensively used for bulk applications in geotechnical engineering

They are : Fly Ash in Embankment and Sub-Base, Building Foundation using Fly Ash

Fly ash as Valley Fill

Fly ash in Dams

Fly ash as Interposing Layer Fly ash as Landfill Liner Stabilisation of Soils
Still only about 65% of ash generated is used. More uses are being explored.

3.0 SWELL IN SOILS WITH SULPHATE

Salts of sulphates or esters of sulphuric acid are formed by replacing one or both
of the hydrogen with a metal (e.g., sodium). Sulphates of sodium (thenardite,
NaSO4.10H2O), potassium (arcanite, K2SO4), calcium (gypsum or selenite,
CaSO4.2H2O) and magnesium (epsomite, MgSO4.7H2O) are commonly present in
the surface sediments, especially in regions of limited rainfall (Grim, 1968; Wild et
al., 1999). Most metal sulphates are readily soluble in water, but calcium and mercuric
sulphates are slightly soluble; while barium, lead, strontium, and mercurous sulphates
are insoluble. The release of sulphuric acid by oxidation of pyrites is shown in Fig. 3.

Sulphates present in pore water or in mixing water can interact with soil. They
can induce unexpected cation exchange reactions by removing higher valency ions from
ion exchange complex of clay due to formation of insoluble salts with them. Sulphates
can also alter pozzolanic reactions of cement and lime treated soil systems. The nature
of above reactions depends on the type and concentration of metal ions present, and the
availability of alumina and silica that can influence the engineering behaviour of soil
with time. It is important to understand the role of sulphate in soils in different
conditions.

166
 IMPROCE 2009

Fig. 3 Release of Sulphuric Acid due to Oxidation of Pyrite (Smith et al 1995).

3.1 Soil–Lime–Sulphate Reactions

Lime–soil reactions can be classified into short term and long term reactions.
Short term reactions include flocculation, lime migration, pH and cation
exchange reactions, carbonation, and the above reactions affect the physical properties
of the soil system such as Atterberg’s limits and particle size distribution. Whereas the
long term pozzolanic reactions include the formation of various new reaction products,
which aid in the growth of aggregates and affect the strength and compressibility of
clays.

There are several factors that influence the lime treated soil properties such as
cations, concentration of sulphates and clay minerals composition (available alumina
and silica). The anions of sulphates combine with the available calcium and alumina,
and form insoluble ettringite in the soil system. The formation of ettringite increases the
porosity and simultaneously decreases the free moisture content available during
ettringite nucleation and its subsequent growth.

Sherwood (1962) observed swelling and cracking in specimens of clay stabilized

with lime and immersed in sulphate solutions. Dal Hunter (1988) has reported induced
heave in lime-treated soil containing sulphate. Swell and shrinkage characteristics of
lime-treated sulphate soil have been extensively studied by Sivapullaiah et al (1993),
Sridharan et al. (1995, 1997), Puppala et al. (1998, 1999). They have shown that the
presence of sulphate increases the compressibility of lime-treated black cotton soil after
curing for prolonged periods. Kota et al (1996) reported that, the presence of sulphate
poses problem with calcium- based stabilisation. Most of these instances are associated
with the presence of sulphate, particularly in the presence of lime. The effect of sulphate
is due to the formation of ettringite and thaumasite in lime-treated soils containing
sulphate. Use of cement or double application of lime (McCallister, and Tidwell, 1994)
is the main alternatives to counteract the sulphate induced heave.

While the effects of sulphate in alkaline environment on the volume change

behaviour of soils are widely known, very little information is available regarding its
effects in acidic environment.

167
 IMPROCE 2009

4.0 SWELL IN SOILS DUE TO CONTAMINATION

An important pollutant that can have considerable impact on soils is high

strength alkali contamination. Hydroxides are released into the soil environment from
various industries

such as i) Paint and dyes ii) Paper and pulp industries iii) Cotton mills and iv)
Aluminium industries etc. High concentration of alkali solution used to extract alumina
from bauxite in the digestor can leak into foundation soils in alumina extraction
plants resulting in the upheaval of foundations footings and consequent distortion of
floor and roof beams (Sivapullaiah et al, 2004).

4.1 Heave due alkali Contamination

4.1.1 In swelling soil

The swell induced due to alkali contamination on two types of soils, viz.,
expansive soil containing montmorillonite mineral (BCSM) has been presented in Fig.
4. The soil specimens were compacted with water and inundated with sodium hydroxide
solutions of 0, 1, 2, 4 N concentration to study the swelling behaviour of soil at nominal
load. Samples are allowed to swell till equilibrium is reached.

Fig. 4 Variation of swell with time for BCSM inundated with different fluids

All the samples exhibited swell but the samples inundated with NaOH solutions
showed significantly high swell than that inundated with water. The total swell
decreased with the higher concentrations of sodium hydroxide solution. The swell is
about 10% with water, 20% in case of sample inundated with 1N NaOH solution, where
as the swell of 13% was observed in case 2N and 4N NaOH solution. All the samples
exhibited single stage of swelling. This can be explained as follows:

As the soil contains montmorillonite mineral its cation exchange capacity is high
and hence the negative charges on the clay surface increases due to which adsorption of
168
 IMPROCE 2009

water to form diffused double layer near the negative surface of clay mineral particles
increases. The swell in soil with water increased steeply for 1day (up to7%) and
continued very slowly for rest of the period. Thus the swell in soil with water is about
10%. The increase in the magnitude of swell to 20% with 1N NaOH solution is due to
increased negative charges on clay particles at higher pH. The magnitude of swell
decreased to 13% with increase in the concentration of alkali solution to 2N. This is due
to dominance of decrease in the thickness of diffuse double layer as influenced by the
electrolyte nature over the pH induced repulsive forces. The increase in concentration of
inundating fluid to 4N NaOH has no further effect on the swell. Consolidation
behaviour after full swell at seating load

After allowing the sample to swell completely which took unusually long
period, consolidation tests were conducted with load increment ratio of one. The void
ratio – pressure relationship for the different specimen are given in Fig. 5.

Fig. 5 Volume change for BCM inundated with different fluids

XRD of BCSM after interaction with 1N NaOH:

Soil underwent significant change, with a substantial decline in the intensity of

the montmorillonite peaks, which may be interpreted (Sivapullaiah et al., 2009) to be
sign for high swelling in the soil even with 1N solutions.

With 2N NaOH solution new peaks of fayalite - Fe2SiO4 and traces of

kirschsteinite - Ca3Fe2(SiO4)3 were formed due to long term soil alkali interaction.
Binding properties of kirschsteinite were highly improved in the presence of siliceous
substance.

With 4N NaOH solution new peaks of kirschsteinite - Ca3Fe2(SiO4)3 and traces

of fayalite - Fe2SiO4 are formed. The intensity of kirschsteinite has increased,
compared to 2N NaOH, which increases the binding property and hence the
compressibility of 4N alkali contaminated soil has reduced.

169
 IMPROCE 2009

4.1.2 In non expansive soil

Turer (2007) observed swelling of about 2.2% in non swelling kaolinite swell in
4N NaOH solution. Fig. 6 shows the variation of swell with time for both kaolinite and
red earth sample compacted with water and inundated with 1N NaOH solution
(Sivapullaiah and Manju, 2005). It can be seen that no swell has been observed initially
even after one-month time period in both red earth / kaolinite when inundated with 1N
NaOH solutions. But slowly swelling occurs after a time lag of about one-month.

The percent swell in red earth was higher at about 13% than in kaolinite, which
has swelled in 1N NaOH by about 4% only. Equilibrium has been reached after 71 days
at 6.25 kPa in both red earth and kaolinite. Percent swell is defined as the ratio of the
amount of

Fig. 6 Swelling of kaolinite and red earth with 1N NaOH solution

Oedometer Swell with 4N Alkali Solution

Fig. 7 shows the variation of swell with time for both kaolinite and red earth
sample compacted with water and inundated with 4N NaOH solutions.

Fig. 7 Swelling of kaolinite and red earth with 4N NaOH solution

170
 IMPROCE 2009

When red earth/kaolinite compacted with water is inundated with 4N NaOH

solutions, swelling commenced only after fifteen days. And equilibrium has been
reached after 50 days in red earth and after 68 days in kaolinite at 6.25 kPa. It was
observed that the increase in swelling with time is slow initially up to certain period of
time and increases rapidly and then reaches an asymptotic value. Thus, the curves have
an elongated ‘S’ shape.

The percent swell is 16.5% in red earth and 15% in kaolinite specimen
compacted with water and inundated with 4N NaOH solutions. Thus unlike the case of
1N solution, the effect of 4N solution is almost same for both red earth and kaolinite.
The slow rate of increase in swell in kaolinite might be due to the fact that all silica is
not exposed to react with alkali in case of kaolinite mineral unlike in the case of red
earth.

Consolidation behaviour after full swelling

It has been observed that considerable swelling occurs in soils after a

considerable time lag. Considerable change in consolidation behaviour has been
observed in the samples loaded after full swelling at nominal surcharge. Unlike the
consolidation tests, which undergo some compression during incremental loading,
almost negligible amount of compression took place with increase of effective pressure
up to 800 kPa in the inundating fluids i.e, 1N and 4N alkali solutions. This shows very
clearly that zeolites formed, after swelling, do not undergo compression like any
expansive soil containing smectite minerals. In the consolidation tests carried out as per
usual method, abnormal swelling during rebound have been observed with 4N NaOH
solutions, since the soil would have interacted with alkali, leading to the
formation of zeolites. Thus abnormal swelling occurred even at higher surcharge .The
times required for swelling to show at nominal surcharge and swelling during rebound
to occur are comparable.

4.2 Fly Ash to Control the Alkali induced Effects

The efficiency of different percentages of fly ash in controlling the swell

induced in the soil with 2N alkali solution, the behaviour of the soil specimen amended
with different amounts of fly ash compacted with water and inundated with 2N alkali
solution has been studied ( Sivapullaiah and Hari Prasad Reddy, 2009)..

4.2.1 Control on non-expansive Swell

The time – swell relationship for different percentages of fly ash mixed with
soil, compacted with water and inundated with 2N alkali solution is shown in Fig 8.

171
 IMPROCE 2009

Fig. 8. Effect of fly ash content on the swell of RE with 2N alkali solution

The high swell of about 16.5% observed for the soil inundated with 2N alkali
solution is considerably reduced when the fly ash is incorporated into the soil. It is clear
that the percentage reduction in the swell decreased drastically with inclusion of fly ash
in soil. The swell in soil is reduced to negligible levels (0.8%) with 10% and 25% fly
ash and not observed in soil with 50 % of fly ash. The reduction in swelling can occur
due to replacement of soil by fly ash as well as by cementation of soil particles by
pozzolanic fly ash. It is observed form figure that 10% of fly ash effectively controls
the swell in soil. This high reduction in swelling with 10% fly ash indicates that the
reduction in swelling is mainly due to cementation of soil particles by pozzolanic fly
ash and thus eliminating the swell due to mineralogical changes. Thus it can be
concluded that even low percentage of fly ash is sufficient in controlling the swell in red
earth which is mainly due to mineralogical changes.

Evidence form XRD Studies

The high second stage of swell in red earth with 2N alkali solution is due to the
formation of new mineral called sodalite which belongs to zeolite group.

Control on Compressibility

As there was no noticeable compression in soil with alkali solution, the effect of
fly ash to further reduce the compressibility soil in alkali solution is hardly seen.

4.2.2 Control on Expansive Soil

It is clear from the above discussion that the swelling which is mainly due to
mineralogical changes can be effectively controlled even with inclusion of a small
percent of fly ash in red earth soil. The manner in which fly ash controls the alkali
induced heave in BCS M, which is mainly due to pH effect, is studied in this section.

Control on Swell

Fig. 9 shows the effect of different percentages of fly ash (10%, 25% and 50%)
on swell behaviour of BCS M compacted with water and inundated with 2N NaOH
172
 IMPROCE 2009

solution. A swell of about 14 % which is observed for the soil inundated with 2N alkali
solution is considerably reduced, when the fly ash is incorporated into the soil. It is clear
that the percentage reduction in swell is very high even with small percentage of fly ash
which, however, is not improved with increase in percentages of fly ash in the soil. The
swell of soil is reduced to less than 0.5% with all percentages of fly ash. It is clear that
10% of fly ash controls the swell induced in soil even in case of 2N alkali solution. This
reduction in swell with 10% fly ash indicates that the reduction in swell is mainly due to
cementation of soil particles by pozzolanic fly ash and eliminating the swell due to
electrolyte effect.

Thus it is clear that addition of even 10% fly ash effectively eliminates the swell
in kaolinitic and montmorillonitic soils with alkali solution. However, the percentage
reduction in compressibility of BCS M can not exceed 70% even with higher amounts of
fly ash. It is clear that compressibility of soil has reduced with the addition of fly ash.
The percentage reduction of total compressibility with different amount of fly ash is
same and it is about 70%. Thus it is clear that fly ash addition can totally eliminate
swell in BCS M and can reduce compression in soil with 2N alkali solution. Thus the
effect of fly ash is about the same on this soil with 1N or 2N solution.

Fig. 9 Effect of fly ash content on the swell of BCS M with 2N alkali solution

Control on Compressibility

The e-log p curves for BCS M mixed with fly ash and inundated with 2N alkali
solution are shown in Fig. 10.

173
 IMPROCE 2009

Fig. 10 Effect of fly ash content on the void ratio – pressure relationships of BCS M
with 2N alkali solution

4.3 Swell due to Acid Contamination

As in the case of soils treated with alkali solutions, swell and subsequent
consolidation tests were conducted on expansive and non expansive soil compacted
with water and inundated with acid solutions varying from 1 N to 4N concentrations.

4.3.1 Swell in Swelling Soil

As seen from Fig. 11 the soil shows considerably higher swell when it is
saturated with 1N H2SO4 at nominal surcharge of 6.25 kPa.

Fig. 11 Swell Behaviour behaviour of expansive soil with sulphuric acid solutions

174
 IMPROCE 2009

This might be due to disassociation of hydrated hydrogen ion in sulphuric

acid which replaces the exchangeable cations in the interlayer of minerals initially with
hydrogen. However, the sample compacted and inundated with water shows a swelling
of about 2% reached in about 10 days. The moderate swelling in BC soil alone is due to
the presence of divalent or trivalent iron cation, though the soil is rich in
montmorillonite. The soil shows abnormally high swelling when the soil is saturated
with 4N H2SO4 solution in the consolidation cell under nominal surcharge. This
could be due to attack on the lattice structure of soil mineral by the dissociated
hydrogen ions.

Consolidation Behaviour after completion of swell

Consolidation behaviour of soil after completion swell in different

concentrations of acid solutions is presented in Fig. .Black cotton soil inundated with
1N sulphuric acid solutions, presented in Fig.12 exhibits higher compression than the
sample remoulded and inundated with water (Sivapullaiah et al., 2009). This supports
that the earlier swelling at nominal surcharge, due to cation exchange by hydrogen ions,
leads to subsequent higher compression with increase in surcharge. However the
sample remoulded with water and saturated with 4N solution shows lesser compression
than soil saturated with 1N acid solution.

Mineralogical Changes in soil

When sulphuric acid is diluted with water, its dissociation occurs in the following two
stages:

H2SO4 + H2O ↔ H3O+ + HSO4 -

HSO4- + H2O ↔ H3O+ + SO2-4

Fig. 12 Void ratio - Pressure relationship for Black Cotton Soils Treated with
Sulfuric acid Solutions
175
 IMPROCE 2009

Initially the exchangeable cations of the soil sample are substituted by released
H3O+ (hydronium). This is the reason for the release of exchangeable ions of soil on
interaction with sulphuric acid. In addition to this, due to the ionic potential (electric
charge of hydrated cation/smaller size) of hydronium it penetrates into the crystal of soil
structure, leading to the -2 displacement of structural ions. However, the anion, SO4 ,
remains mobile and acts as the counter-ion for the cations. Thus these reactions
cause considerable changes in the mineralogy of soil. The released H3O+ can cause
changes in cation exchange complex of the clay particles, which lead to mineralogical
changes. The effect of sulphuric acid on the swell behaviour of expansive and non
expansive soils is presented (Sivapullaiah et al., 2009).

4.3.2 Swell in non expansive soil

Figure 13 shows the percent swell of red earth compacted with water and
inundated with water, 1 and 4N H2SO4 solutions at nominal seating load. With water
as inundating fluid, only 0.36% of swell is observed.

The low swell is due to mineralogy, lower clay content and the lower cation
exchange capacity. It is very interesting that abnormal swell of 22% is observed in the
same soil when the inundating fluid is 1N sulphuric acid solution. This swell occurs
over a period of 275 days. The swell further increased with 4N sulphuric acid as
inundating fluid. An incredibly high swell of 55% is noticed when it reaches
equilibrium in about 230 days.

The induced swell by sulphuric acid can be explained as the Sulphuric

acid dissociates into hydronium and sulphate ion in water. Initially, the hydronium ion,
due to its smaller size and/or ionic potential, penetrates into the crystal of mineral
structure leading to the reduction in H-bonding between successive basic units of
kaolinite. Hydrogen ions can leach out iron present in the lattices of kaolinite and can
form a new mineral by association with dissociated anion, SO42-. Further from Fig. 13,
it can be observed that the delay in starting of swell process decreases with increase in
concentration of inundating fluid. The lag period of 160 days with 1N sulphuric acid is
reduced to 85 days with 4N sulphuric acid.

Fig. 13. Swell Behaviour of Acid Treated Red Earth

176
 IMPROCE 2009

Consolidation behaviour

The void ratio–pressure relationship for the soil remoulded with water and
inundated with water, 1N and 4N H2SO4 is compared in Fig. 14. It can be seen that
compression, with increase in the effective pressure, increases with concentration of
sulphuric acid.

Fig. 14 Void ratio - Pressure Relationships for Red Earth inundated with
Sulphuric

Acid Solutions

Mineralogical Changes in Red Earth

X-ray diffraction patterns of treated and untreated red earth shows the peaks at
7.14, 2.34 and 3.57 Å due to kaolinite mineral. After the completion of consolidation in
red earth inundated with acid solutions, formation of rozenite mineral (peaks at 4.47,
and 3.97 Å) has occurred, which is an iron sulphate hydrate. In addition, the peak of
corundum–aluminum oxide is also observed in the soils inundated with acid solutions.

4.4 Fly Ash to Control Acid induce Swell

4.4.1 On Expansive Soil

The variation in swelling with time for BC soil mixed with 20% fly ash and
inundated with sulphuric acid solutions of 1 and 4N is presented in Fig. 15. In both the
samples swelling is observed from the first day. The soil treated with 1N sulphuric acid
solution reached an equilibrium swell of 20% in about 20 days. This reported 20% swell
is very high compared with the swell of natural BC soil (~2.1%). This might be to due
to the formation of new minerals by soil fly ash - interaction under acidic solution
containing sulphate. The trend is further confirmed by very high swell in soil fly ash
sample treated with 4N sulphuric acid solution. An equilibrium swell of 55% after about
90 days is observed in this case. The swell that occurs in the soil without fly ash has not
reduced with the addition of fly ash in both 1 and 4N sulphuric acid solutions.
177
 IMPROCE 2009

Fig. 15 Swell Behaviour of BC Soil Mixed with Fly Ash and

Inundated with Sulphuric Acid Solutions

The void ratio--pressure relationships of BC soil mixed with fly ash and
inundated with acid solutions are given in Fig.16 . It can be seen that BC soil mixed
with fly ash and inundated with 4N sulphuric acid solution undergoes relatively less
compression compared to the soil inundated with 1N sulphuric acid solutions. The
lesser compression might be due to less swelling that has taken place in the BC soil
mixed with fly ash and inundated with sulphuric acid solution. During the formation
of calcium sulphate the soil might have swollen.

Fig. 16. Volume Change Behaviour of BC Soil Mixed with Fly Ash
Inundated with Acid Solutions

178
 IMPROCE 2009

4.4.2 On Non Expansive Soil

The variation in swelling with time for red earth mixed with 20% fly ash and
inundated with sulphuric acid solutions is presented in Fig. 17.

Fig. 17 Swell Behaviour of Red Earth with Fly Ash inundated with Sulphuric
Acid Solutions

The swell in soil admixed with 20% fly ash reduced considerably from 22 % to
2.5 % in 1N sulphuric acid solution. The low swell that occurred is completed
within 7 days. This indicates that upon the addition of fly ash the swelling is
controlled in red earth. Thus it can be concluded that the breakage of hydrogen bonding
between unit layers of kaolinite clay is probably counteracted by the cementation of the
units by pozzolanic cementation of the layers.

It is, however, observed that the sample inundated with 4N sulphuric acid
solution swells considerably even after amending with fly ash. The equilibrium swell is
marginally reduced from 55% to about 45%. The nature of swell in soil with and
without fly ash is almost the same. The effect of fly ash is very less probably due to
dissolution of cementitious compounds in high acidic solution.

The void ratio--pressure relationships of the red earth mixed with fly ash and
inundated with acid solutions are given in Fig. 18 .

179
 IMPROCE 2009

Fig. 18 Volume Change Behaviour of Red Earth with Fly Ash inundated with
Acid Solutions

The sample inundated with 4N sulphuric acid solution undergoes relatively less
compression compared to the soil inundated with 1N sulphuric acid solution. The lower
reduction in 4N solution may be due to continued slow effect of acid to dissociate the
kaolinite particles in the absence of sufficient cementation by fly ash particles. This
continued swelling, in fact, is exhibited as higher rebound during unloading and the soil
showed higher void ratio than during loading at very low effective pressures pertinent in
the system.

Mineralogical Changes Due to Acid Treatment in Red Earth Sample Mixed with
Fly Ash

The induced high percent of swell with 4N acid solution is analysed through the
XRD studies. The treated soil has shown the peaks at 3.49, 2.84, 2.20 and 1.64 Å
leading to the formation of calcium sulphate. Probably the calcium rich fly ash when it
reacts with sulphate, the mineral calcium sulphate might have formed, which in turn
imparts high swelling nature.

CONCLUSIONS

1. Heave can occur in swelling as well as non swelling soils when contaminated with
Acid or alkali contamination.
2. The amount of swell varies with the type of mineral present in the soil and
concentration of the acid/ alkali solutions. Second stage of swelling occurs in soils,
wherever mineralogical changes have occurred due soil contaminant interaction,
after considerable delay period.The conventional hyperbolic relationship for swell –
compression is not valid for the contaminated soil.
3. Distress occurs in structures when foundation is contaminated soil.
4. Pozzolanic stabilization can control the conventional as well as contaminated
induced swelling.
180
 IMPROCE 2009

REFERENCES
1. Alonso, E. E. and Gens. A., (1994). “On the mechanical behaviour of arid soils”,
Proc. First International Symposium on Engineering Characteristics of Arid Soils,
London,173- 205.
2. Barbour, S.L., Fredlund, D.G. (1989). “Mechanisms of Osmotic Flow and Volume
Change in Clay Soils”, Canadian Geotechnical Journal, 26, 551–562.
3. Bolt, G. (1956).“Physico-chemical analysis of the compressibility of pure clays”,
Geotechnique 6, 920-934.
4. Di Maio, C., Fenelli, G.B. (1994). “Residual Strength of Kaolin and Bentonite: The
Influence of their Constituent Pore Fluid”, Geotechnique, Vol. 2, 217–226.
5. Di Maio, C., Santoli, L., and Schiavone, P. (2004). “Volume Change Behaviour of
Clays: The Influence of Mineral Composition”, Pore Fluid Composition and Stress
State, Mechanics of Materials, Vol. 36, 435–451.
6. Hansen, J.M and Smit ,B. (2002). “Why clay swells?”J. Phys. Chem. B. 06, (49),
12664-67.
7. McCallister, L. D., and Tidwell, L.,1994, “Double Treatment to Minimize Sulfate-
Lime Induced Heave in Expansive Clays,” final report, U S M A , West Point
8. Mitchell, J.K. (1993). “Fundamentals of Soil Behavior”, 2nd edn. New York, Wiley.
9. Sivapullaiah, P. V. and Hari Prasad Reddy, P. (2009), “Fly Ash to Control Alkali
Induced Volume Changes In Soils” Ground Improvement, (In press).
10. Sivapullaiah, P. V. and Manju (2005), “Kaolinite –Alkali interaction and effects on
basic properties”, Geotech and Geol. Eng. Vol. 23, 601-614.
11. Sivapullaiah, P. V., Guru Prasad, B., Allam, M. M. (2009). “Effect of Sulfuric Acid
on Swelling Behaviour of an Expansive Soil”, Soil and Sediment Contamination,
Vol. 18 (2), 121.
12. Sivapullaiah, P. V., Sankara, G., and Allam, M. M.(2009).” Mineralogical Changes
and geotechnical properties of an expansive soil treated with caustic solution”,
Environmental Earth Sci., DOI 10.1007/s12665-009-0260-9.
13. Sivapullaiah, P.V., Sridharan, A. and Prashanth, J.P.(1995), “Optimization of lime
content of fly ash”, ASTM Jl. of Testing and Evaluation, Vol.23, 222-227.
14. Sivapullaiah, P.V., Sridharan, A., and Stalin, V.K. (1996), “Swelling Behaviour of
Soil- Bentonite Mixtures”, Canadian Geotechnical Journal, Vol.33/5, 808-814.
15. Smith B. Bush, R. and Sammut J. (1995), “Acid Sulfate Soil in the Noosa River
catchments,” Noosa Council Australia.
16. Turer, D.(2007). “Effects of Heavy Metal and Alkali Contamination on the
Swelling Properties of Kaolinite”, Environ.Geology, Vol. 52, 421-425.
17. Van Olphen, H. (1966). “An Introduction of clay colloid chemistry”, Interscience,
NY.
18. Yong, R. N. and Warkentin, B. P. (1975). “Soil properties and behaviour”, Elsevier,
NY.
181
IMPROCE 2009

182
 IMPROCE 2009

PERFORMANCE OF GROUND IMPROVEMENT FOR A

RESIDENTIAL PROJECT

Anirudhan I.V* and Radhakrishnan R**

ABSTRACT

The performance of a construction is to some extend subjective until the

construction is totally un-usable or deteriorated. It can be said that any behaviour not
conforming to the set standards is unsatisfactory as far as a construction is concerned.
A large scale ground improvement measure using vertical band drains and pre-loading
technique was adopted for a residential project in Kerala. The predictions and
performance during pre-loading were very satisfactory. The first two pilot constructions
experienced settlement to the tune of 80mm to 150mm whereas the anticipated
settlement was only less than 40mm. Though the settlement was uniform and the
structure did not experience any distress, the performance was ‘unsatisfactory’. This
paper discusses he reasons for such ’unsatisfactory’ performance and the remediation
adopted later.

1. INTRODUCTION

Anticipation in geotechnical engineering design is boundless. The capability of a

geotechnical designer to anticipate several possibilities makes the structure more and
more safe, and expensive too, with regard to the foundation performance. Several
factors unthought-of during design may come into force in real situation that would
upset the whole design and the anticipated performance. Unfortunately some of such
factors are really unpredictable and beyond the imagination of the geotechnical
designer.

After several discussions on various possibilities of foundation for G+1or G+2

residential bungalows in a medium size project in Kerala, ground improvement
technique involving installation of band drains and pre-loading was adopted. Probably
that was the first time such a massive ground improvement programme was suggested
and executed for a residential project comprising only G+1 bungalows. The programme
was executed with well defined architectural design since the pre-loading technique
should have predicted the load at the founding level fairly accurately to ensure the
satisfactory performance. The settlement at the original ground surface was monitored
using plate settlement markers and the recorded settlements were close to the
anticipated settlements. The performance of ground improvement programme was then
considered satisfactory.

Two bungalows out of the total 48 such constructions were completed and
monitored. One of the bungalows settled uniformly by about 80mm, while the other
experienced a uniform settlement of about 140mm. The anticipated settlement of the
constructions was, however, less than 40mm, a limit set by IS 1904 for such
constructions. The uniformity in construction schedule and relatively stiffer RCC strip
foundation provided for these bungalows resulted a uniform settlement and there was no
worry about the overall performance of the construction as such. The rate of settlement
* Geotechnical Solutions Pvt. Ltd, Chennai – 42, ** Bharat Geosystems Pvt, Ltd, Chennai – 41. 183
 IMPROCE 2009

was almost zero as soon as the constructions were over and a significant long term
settlement was also not expected. However, the performance of the foundations was far
less than satisfactory because of much larger settlement than the design settlement.

Review of the performance revealed that the finished ground level of this low
lying site was revised to a level higher by 0.80m (equivalent to about 1.52 t/m2) that
was not in the design of foundation. Anticipation of such revision in finished ground
level after a freeze on architectural design did not occur to the geotechnical designer.
On the other hand, the architect failed to understand the mechanism of the executed
ground improvement programme and found it unnecessary to inform such change to the
geotechnical engineer. The failure of geotechnical engineer lies in the fact that he did
not put the design terms very exclusively.

Future constructions were taken up with larger preload to compensate the

additional load from the extra fill required to reach the revised finished ground level.
Monitoring continued and confirmed that the settlements were less than 30mm.

2. THE PROJECT AND SUB-SOIL CONDITIONS

The project was construction of 48 independent bungalows with ground floor

and one floor as a township in a 45 acre land. Large common space as a water body,
wide roads and other amenities and relatively large backyard for each bungalow were
the main features of the project. The site was low lying, abutting a small river on two
sides of the land and coconut groves on two sides. The original ground level was very
low and a fill of 1.20m was already in place to secure the land from major flooding.
Additional 0.80m fill was suggested by the architects to keep the ground safe from any
flood. The sketch (Figure 1) on the right side show the project layout.

The subsoil profile comprised a fill of 1.20m thick murrum soil compacted to
medium density, a thick layer of clay of high compressibility, highly weathered and
disintegrated lateritic rock in the from of clayey sandy silt and weathered granite. The
average thickness of different soil layers are shown in Figure 2.

HOUSES NEIGHBORING
LAND

WATER BODY

DEVELOPED LAND

WATER WATER BODY

BODY

Figure 1: The project layout, Tellichery, Kerala

184
 IMPROCE 2009

The thickness of very soft clay varied between 4.50m and 7.50m. The profile in
the Figure 2 shows maximum thickness of the soft clay and corresponding elevations of
the remaining profile. The two pilot bungalows were built in an area where the
thickness of soft soil is large. The soft clay is highly plastic and compressible while the
clayey sandy silt found below is of medium plasticity and medium compressibility. The
shear strength of residual soil layers below 11.3m from original ground level improved
significantly and rapidly towards depth. The index and compressibility parameters
established by field and laboratory test on the soft soil are presented below.

Liquid limit, LL = 88% to 120%

Plastic limit, PL = 36% to 42%
Plasticity index, PI = 52% to 78%
Natural moisture content, NMC = 80% to 102%
Liquidity index, LI = 0.68 to 1.00
Bulk density, γb = 1.50 gm/cc
Undrained shear strength, cu = 0.11 kg/cm2
Pre consolidation pressure, Pc = 0.25 to 0.35kg/cm2
Coefficient of volume compressibility, Cv = 1.26m2 /year

Proposed Finished ground level 0.80m above

0.00
Red Earth Fill N=4
1.20

N =1, Su=0.16kg/cm2
Greyish very soft silty LL =120%, PI = 78%
clay NMC = 102%, LI = 0.80
cv=1.26m2/year
6.50
Black soft silty clay N =2 to 4, cu= 0.25kg/cm2
9.00 LL=95%, PI = 63%, LI = 0.77
Greyish loose dirty fine
sand N = 2 to 4
12.50
Loose Clayey silt N =9
14.20

Dense clayey sandy silt N = 30 to 40

18.00

Very dense clayey sandy

N=>100
silt
22.00
Weathered rock SPT Rebound

Figure 2: Subsoil profile of the project land

The coefficient of consolidation Cr in the radial direction (horizontal direction)

was assumed twice the coefficient of consolidation in the vertical direction.

The natural moisture content in residual clayey silt below the loose sand layer
varied between 17% to 38% while the liquid limit and plasticity index of this soil are
45% and 22%. This layer and the stiffer clayey silt below are expected to have low
compressibility.
185
 IMPROCE 2009

3. FOUNDATION ALTERNATIVES

Different types of foundation systems were reviewed for the projects and the
possible performance of each system was also reviewed in detail. Obviously, the first
option for any such construction would have been shallow footings or shallow strip
footings as the structures are lightly loaded. The existing fill of 1.20m was not
compacted well and there was further 0.80m fill to be placed to reach the finished
ground level. The total load of the fill was estimated as 3.8 t/m2 when the fill is
saturated. The average load from the proposed construction was roughly 3.0 t/m2,
together made a total load increment of 6.8 t/m2 over the soft clay with pre-
consolidation pressure not exceeding 3.5 t/m2. Large consolidation settlement was
hence expected even for a G+1 construction and significant differential settlement
between the building and the utility line as well as within the building itself was also
expected. The possible performance of such foundation is illustrated in Figure 3 shown
below.

Uneven Roads
Road Leaking
Construction over fill with Cracks
Drain shallow footings Drains
Final fill 0.80m

Initial fill 1.20m

Re-compacted using power roller OGL OGL

Soft clay Consolidated clay

(a) just after the construction (b) after one or two years
Figure 3: Performance of shallow foundation for the proposed development
There were issues related to the compaction of the fill and replacing the already
existing mururm fill to achieve the required density and shear strength for adopting
shallow foundation.

Naturally the second option was to consider deep foundation such as driven cast-
in-situ piles or bored cats-in-situ piles for supporting the buildings. The piles were to be
taken to a depth of 14m to 18m to realise adequate sage capacity that can cater the
negative drag from the soft soil also. The expected performance of such expensive
system is illustrated in Figure 4 below.
Probably a damaged frame or
a frame elevated from sunken
Portico over ground
Damaged
shallow footing
Portico

Road Frame structure over pile Uneven Roads

Drain Leaking
Drains

OGL OGL

Pile Foundation up to Rock Compressed clay causing

designed for very large negative uneven negative drag
drag

(a) just after the construction (b) after one or two years
Figure 4: Performance of deep foundation for the proposed development
186
 IMPROCE 2009

In this option the settlement of road and the utilities much larger than the
structure was expected. Even though the negative drag on the deep foundation could be
predicted to some accuracy, the deficiency in pile tip conditions in the bored cast-in-situ
piles and remoulding of very soft clay during driving of driven cast-in-situ or pre-cast
driven piles could have predicted the negative drag wrong that would result a failure
load on the foundation. The anticipation of such unfavourable conditions and
overloading would have lead to large pile sections and more number of piles escalating
the cost of the project significantly.

The third option that was finally adopted was pre-loading of the clay to a load
equal to the expected incremental load from the proposed construction so that a
significant portion of the expected settlement was pre-achieved. The effect of pre-
loading is presented in Figure 5.

Other areas Area for building

Area for building
Other areas

Preload fill Preload fill 5.5 t/m2

2.5 t/m 2 over To be removed up to finished ground
level after two months Preload for removal
initial fill
level
FGL FGL

OGL OGL

Soft clay
compressed by Soft clay compressed by 800 to
400 to 500mm 1000mm under pre-load
under pre-load Pre-consolidation improved by 2.5 t/m2

(a) Preload fill before consolidation (b) after pre-loading for 60days
Figure 5: Consolidation of soft soil under pre-load

The principle behind the pre-loading technique for pre-empting a very

significant part of the settlement is presented in the sketch given in Figure 6. The soil
being elasto-plastic, the advantage of only one seventh to one tenth of the virgin
compression as the re-compression will help in reducing the additional settlement when
the constr-uction load is applied after removing the pre-load. The complete pre-load is
not removed in normal conditions as the ground level is to be maintained ret-aining the
part of pre-load equal to the total settlement achieved during the pre-loading.

The soft clay being high plasticity type with very low coefficient of
consolidation, the process of settlement under pre-loading was expected to take more
than 3 years. The advantage of vertical drains to accelerate the consolidation process
was available. The conventional sand drains that required more construction time and
large amount of labour force was being replaced with pre-fabricated drains in those days
and then it was decided to adopt pre-fabricated band drains for accelerating the
consolidation settlement. The advantage of introduction of band drains in the pre-
loading scheme is illustrated in Figure 7.

187
 IMPROCE 2009
PRESENT PRE-CONSOLIDATION-
LOADING (LOADING HISTORY)

LOAD FROM PRE-LOAD +

PLASTIC YIELD
FILL UP TO FINISHED GL

TOTAL
SETTLEMENT
NET
SETTLEMENT

SETTLEMENT
REMOVAL OF PRE-LOAD
EQUAL TO LOAD FROM STRUCTURE

PROCESS OF CONSOLIDATION
(EXPELLING WATER FROM VOIDS)

Figure 6: Design principle of the preload

Figure 8: Installation of vertical drains

Use of such method of ground stabilisation is well documented (Hansbo-1979,

Mitchell & Katti-1981, Choa-1985, Ramaswamy, et.al-1996 and M.W. Bo, et.al –
2003). Prefabricated vertical drains (band drains) of the type Mebra Drain which is
100mm wide, 3mm thick in rolls of 300m long were used for the project. They were
installed to full length of the clay layer by crane mounted drain stitches rig. During the
installation, the drain is protected within a steel mandrel with a dispensable anchor shoe
at the bottom. When the drain penetrates the soil to the required depth, the mandrel is
withdrawn leaving the drain and anchor shoe within the soil (Figure 8).

4. GROUND IMPROVEMENT FOR THE PROJECT SITE

Average depth of vertical drain treatment was approximately 8.0m. In order to

achieve 95% degree of consolidation within a period of 60 days, the drains were
installed in a triangular grid pattern at 0.90m c/c within the building plan area. The
height of pre-load was decided as 3.0m in addition to the existing fill of 1.20m to 1.50m
thick, so as to produce a load intensity of 0.50 to 0.55 kg/cm2 (5.0 to 5.5 t/m2). The load
distribution form shallow foundation of the Ground + one building was expected to
produce approx 0.17 to 0.20 kg/cm2 incremental load in the soft clay. The pre-
consolidation pressure of the soft clay is 0.40 to 0.45 kg/cm2 more than the original
effective overburden pressure, ensuring no further consolidation of the soft clay once
the pre-load is removed and the building constructed.

188
 IMPROCE 2009

(a) without vertical drains (b) With vertical drains (c) Pre-loading scheme with vertical drains
Figure 7 Principle of faster radial drainage and pre-loading scheme

PRELOAD-SETTLEMENT PLOT
NO OF DAYS
0 10 20 30 40 50 60 70 80 90 100 110 120
0
Preload full height
100 Preload full height

200 SP1-Villa # 42
CUM SETTLEMENT MM

SP2- BLDG B
300

400

500

600

700
Preload removed
800

900

1000

Figure 9: Time- settlement plot of two plate settlement markers SP-1 and SP-2 at villa # 41and 42

The effect of preloading was confirmed by time-settlement measurements using plate

settlement markers at different locations. The estimated total settlement under the pre-
load was 0.90m to 1.20m. The observed settlements were 780mm to 850mm, close to
90% of the total estimated settlement. The time settlement curves also suggested 90 %
consolidation under the prescribed pre-loading period of 60 days. The settlement
observations made at two locations using surface plat settlement marker are presented in
Figure 9.

5. CONSTRUCTION AND PERFORMANCE OF TWO BUNGALOWS

Construction of two bungalows was taken up immediately and a systematic

measurement of settlement under every stage of construction was carried out.

189
 IMPROCE 2009

SETTLEMENT VILLA # 42
NO OF DAYS
0 20 40 60 80 100 120 140 160 180 200
0

20

40 REAR
First floor roof

CUM SETTLEMENT MM
completed FRONT
60

80

100 Roof tiling completed

120

140

160

180

200

Figure 10: Settlement observation of Villa # 42

The construction of both the buildings was fairly uniform and took about 6
months. Regular measurements of settlement at different corners of the buildings were
made. The band drains functioned very effectively during the construction and the
settlements were very rapid. The settlements measured after completing the construction
were 145mm and 80mm for the two buildings. The Figure 17 illustrates the settlement
of one villa. Settlement stabilised within fifteen days from the completion of the
construction. Uniform construction schedule helped in achieving fairly uniform
settlement without any signifi-cant differential settlement.

Fortunately, the settlements were uniform because continuous RCC strip footing
foundation was provided for these buildings, whereas the initial design was based on
RR masonry strip footing. Even though the structure did not experience any distress
because of uniform settlement, the excessive settlement far beyond the expected
settlement, was considered as a ‘failure’ in the design of ground improvement
programme. The reasons for excess settlement was to be ascertained so that the
necessary changes in the design or improvement could be carried out.

6. DEFECT AND RECTIFICATION IN GROUND IMPROVEMENT

SCHEME

6.1 Changes in the Design

Initially when the ground improvement programme was finalised, the finished
ground level was decided as 0.80m above the existing ground level after 1.20m filling.
Since the expected settlement under pre-load was about 800mm, equivalent height of
the preload was to be retained to maintain the finished ground level. Thus the effective
preload is 2.00m to 2.20m. The preloading of the pilot construction area was completed
and the construction of two villas were to be taken up.

The owner and architect meanwhile decided to raise the ground level by another
0.80m for better appearance and also to avoid any possibility of flooding of the
premises. But this was informed to the consultant and the ground improvement agency
only after the removal of the pre-load and after starting the foundation excavation. The
190
 IMPROCE 2009

amount of preload removed was only 1.40m retaining 1.60m soil required to maintain
the revised finished ground level. Since this amounted to deficiency the preload
required for compensating the incremental load form the constriction, the geotechnical
consultant advised further pre-loading to compensate the deficiency arose because of
revision in the finished ground level.

However, the owner decided to compact the soil at founding level and carried
out plate load test on the compacted soil to confirm the safe bearing capacity. The plate
load test using a 450mm x 450mm plate naturally showed less settlement since the test
was done on the compacted fill. Satisfied with the small settlements measured in the
plate load test, the owner proceeded without further pre-loading. However, the structural
consultant insisted upon adopting continuous RCC strip footing as the foundation for
these two villas.

EXTRA LOAD FROM THE

REVISED FGL

SETTLEMENT
EXPECTED

ACTUAL
SETTLEMENT

Figure 11: Deficiency in pre-loading

6.2 Consequence of Deficiency in Preload Intensity

As suggested earlier, thus the effective preloading suffered a deficiency of about

2
1.5 t/m equal to 0.80m thick fill (Figure 11).

A detailed settlement analysis revealed that the probable settlement due to this
extra loading is 80 to 105mm. More settlement for one villa could have been the result
of compression of initial portion of the preload fill that remained below the founding
level.

6. FURTHER CONSTRUCTIONS

Even though the owner did not fully agree with the consultants, he adopted
increased preload intensity for the remaining sections of the project. The settlements
observed during preloading increased to 900 to 1000mm and the completed bungalows
recorded settlements less than 30mm.

7. CONCLUSIONS

Ground improvement programmes like pre-loading with settlement accelerators

require good planning with respect to site development. All the possible changes in the
development programme shall be thoroughly investigated and the improvement shall
191
 IMPROCE 2009

take care of the worst case. Co-ordination between the foundation consultant, the
architect, the structural engineer and the owner is a must in such programmes.

A thorough and systematic monitoring of improvement in terms of settlements

versus the load and time is needed.

Adoption of expensive RCC strip raft foundation for the area with under-
improvement helped in preventing large distress in the structure. However, adoption of
similar measures for all the villas could have affected the economy of the project.

REFERENCES

1. Bo M.W., Chu J., Low, B.K. and Choa V., (2003), Soil Improvement, Thomson,

2. Choa, V., (1985) ‘Preloading and Vertical Drains, 3rd Int. Geotechnical Seminar,
Soil Improvement Methods, Singapore

3. Hansbo, S., (1979) ‘Consolidation of Clay by Band-shaped Prefabricated Drains,

Ground Engineering

4. Mitchell, J.K. and Katti, R.K., (1981) ‘Soil Improvement – State of the Art Report,
10th Int. Conf. on Soil Mechanics and Foundation Engg, Stockholm

5. Radhakrishnan R and Anirudhan I.V. (2003), ‘Ground improvement with pre-

fabricated V drains and pre-load – A case study, Proc. Symposium on Advances in
Geotechnical Engineering, SAGE 2003, IIT Kanpur, pp 426-430

6. Ramaswamy, S.D & Radhakrishnan R. (1996), ‘Prefabricated Vertical Drains for

deep Subsoil Improvement’, Indian Geotechnical Conf. Madras

192
IMPROCE 2009

193
IMPROCE 2009

194
IMPROCE 2009

195
IMPROCE 2009

196
IMPROCE 2009

197
IMPROCE 2009

198
IMPROCE 2009

199
IMPROCE 2009

200
IMPROCE 2009

201
IMPROCE 2009

202
IMPROCE 2009

203
IMPROCE 2009

204
IMPROCE 2009

205
IMPROCE 2009

206
IMPROCE 2009

207
IMPROCE 2009

208
IMPROCE 2009

209
IMPROCE 2009

210
IMPROCE 2009

211
IMPROCE 2009

212
IMPROCE 2009

213
IMPROCE 2009

214