CHAPTER 4

4. SUBGRADE SOILS
Soil is the most important foundation and construction material for pavement structures.

Foundation material for all pavements as undisturbed in situ subgrade material or

transported and reworked embankment material.
Construction material for pavement structures either in its natural form (sand and
gravel) or in a processed form as stabilized layer.

Soil investigation is, thus, an integral part of the location, design and construction of
highways.

4.1.

Soil Surveys and Investigations

Soil survey for highway purposes involve the exploration of the soils along the highway
routes and the identification of suitable soils for use as subbase and fill materials. The results
of soil investigation provide pertinent information about soil and rock for a decision on one or
more of the following subjects:
i)
ii)
iii)
iv)
v)
vi)
vii)

Selection of roadway alignment;

Decision on the need for subgrade or embankment foundation treatment;
Investigation of slope stability in cuts and embankments;
Location and design of ditches and culverts;
Selection and design of the roadway pavement;
Location and evaluation of suitable borrow and construction materials; and
Design of foundations for bridges and other structures.

In selecting the alignment for a new highway the first step is normally to define a number of
conceivable corridors between the end termini of the road. The next step is to select the best
corridor for the proposed road and define within it one or more different alignments. These
alignments are compared, and a final selection is made for design purposes. The process
involves continuous searching and selecting, using increasingly more detailed knowledge of
subgrade soils at each decision-making stage.
Before a field investigation is carried out at the site, preliminary information regarding soil
condition can often be obtained from the following sources:
1.
2.

Geological and agricultural soil maps. These often indicate the types of soil or
geological formation that cover the area being investigated.
Aerial photographs. Terrain information visible on air photos can be used for
identification of most of the common bedrock types associated residual soils,
transported soils, and organic soils.

3.
4.

Satellite images. Satellite images are employed as a supplemental to air photos or

as a substitute for air photos for geological studies and soil investigations.
Area reconnaissance. Reconnaissance survey aids in securing broad understanding
of soil conditions and associated engineering problems that may be encountered.
The visual examination of vegetation cover, roadway cuts, and valleys in the area
can give clue. The depth to water level in adjacent wells may indicate the elevation
of the groundwater table.

Field investigations and sample collection for laboratory tests are commonly carried out by
the following four methods:

Geophysical methods (seismic or electrical). The seismic refraction method relies on

the principle that the velocity of sound in soils and rocks is different for different
materials. It is particularly useful in predicting the depth to bedrock. The electrical
resitiveity mainly depends on the content of clay minerals, moisture content, and type
and concentration of electrolyte in the soil-water. An increasing content of clay, water
or electrolyte causes decreasing the resistivity of soils.

Test pits or trenches: suitable for shallow depths only to sample soils and rocks, and
register soil profiles.

Hand augers: suitable for shallow depths only to obtain disturbed or mixed samples
of soils.

Boring test holes and sampling with drill rigs: the principal method for detailed soils
investigations.

4.1.1. Depth of Investigations

It is always required to plan and conduct soil surveys in a manner that examines subgrade
materials according to their suitability in load bearing layers within the zone of the design
depth as shown in Table 4-1 . The design depth is defined as the depth from the finished road
level to the depth that the load bearing strength of the soil no longer has an effect on the
pavements performance in relation to traffic loading. Properties of soil below the design
depth may indirectly affect pavement performance, but are generally unrelated to traffic
loading. The depth of test pits and borings should in no case be less than 1.5 m below the
proposed subgrade level unless rock material is encountered. The engineer in charge of
planning the investigations should make every effort to locate the test pits (along the
alignment as well as within the lateral extent of the anticipated excavation) in order to
optimise the representativity of the material excavated from the test pit. A preliminary vertical
alignment may be required at the time of the soil survey in order to ensure that soil samples
are actually taken at levels that fall within the design depth of the road.
For ordinary work, it is quite sufficient to go to a depth of about 3 m below the proposed
foundation level in areas of cut and 3 m below the existing ground in areas of fill.
Investigations in cuttings deeper than 3 m could be impractical and special equipment may
have to be required. If possible, postponement of sampling until the time of construction

should be considered under such conditions. Evaluation of subgrade strength in embankment

areas should be based on the best possible information about likely sources of fill materials
for use within the design depth.
Table 4-1.Design depth (Tanzania Pavement Design manual, 1999)
Road type

Design depth (m)

General requirements
Heavy traffic roads
(> 0.2 106 standard axles )

Paved trunk roads

Other roads

0.8
0.6

1.2
1.0

When required, investigations should be extended to below design depth to detect problems
that need special considerations such as presence of problem soils, unfavorable subgrade
conditions, and features associated with slope and embankment stability. If necessary, sub
surface investigation is carried out using field or in-place testing techniques. Standard
penetration test and piezometer installation are common methods of field or in-place testing
used to evaluate the strength of soils and measure water table depth respectively.
4.1.2. Testing Frequency
For highways pavements, testing sites are generally placed at 100 m to 5 km (generally 300
m) along the centerline of the proposed alignment. Midway between the center line stretches,
additional boreholes 30 m away from the centerline both in its left as well as right side. Table
4-2 shows the minimum average frequency for the purpose of sampling and material testing
which shall be increased according to site conditions. Indicator tests include Atterberg limits
and gradation of particles larger than 75m.
Table 4-2. Minimum materials testing frequency (Tanzania Pavement Design manual, 1999)
Road type
Paved trunk roads
Other paved roads
Gravel roads

4.2.

Indicator tests

Strength tests

Min 4 per km
Min 2 per km
Min 2per km

Min 2 per km
Min 1 per km
Min 1 per 2 km

Essential Laboratory Tests

In specifications for testing of soils reference is frequently made to the standard methods of
testing such as British Standards (BS), American Society for Testing and Materials (ASTM),
and American Association of State Highway and Transportation Officials (AASHTO). To a
large extent the same type of methods are used all over the world. However, when actually
performing tests it is of the utmost importance that the specified standards be followed
precisely, as small differences in the testing procedure may have a noticeable influence on the
test result. This section briefly describes the purpose and the general principle of the most
common tests used in highway engineering. Standard procedure of the tests can be obtained in
the BS, ASTM, AASHTO and other standards and manuals.

4.2.1. Particle Size Distribution Test

The distributions of particle sizes in soils is important in road engineering as the value of
many properties, such as internal friction, voids content, wear resistance and permeability,
and depend on the gradation.
4.2.2. Water Content
The engineering properties of a soil, e.g. strength and deformation characteristics, depend to a
very large degree on the amount of voids and water in the soil, i.e. the condition of the soil.
The water content is defined as the weight of water contained in a soil sample compared with
the oven-dry weight of the sample. It is customarily expressed as a percentage, although the
decimal fraction is used in most computations.
4.2.3. Atterberg Limits
The stiffness or consistency of the soil at any time depends on the state at which the soil is,
which in turn depends on the amount of water present in the soil. The water content levels at
which the soil changes from one state to the other are known as Atterberg limits. The
attreberg limits include the liquid limit (LL), the plastic limit (PL), and shrinkage limit (SL)
as illustrated in Figure 4-1, and they are determined by tests carried out on the fine soil
fraction passing the 0.42 mm sieve. They are important limits of engineering behavior
because they facilitate the comparison of the water content of the soil with those at which the
soil changes from one state to another. Attreberg limits are used in the classification of finegrained soils, since they correlate with the engineering behaviors of such soils. The range of
moisture content over which the soil is in the plastic state is the difference between the LL
and the PL (PI = LL-PL) and is known as the plasticity index (PI). Another index that is used
to reflect the properties of the natural soil is the liquid index (LI) and is defined as:

Figure 4-1. Consistency limits of soils

LI

wn PL
PI

where, wn is the natural moisture content of the soil.

4.2.4. Soil Compaction

When soil is to be used as embankment or subbase material in pavement construction, it is
essential that the material be placed in a uniform layers and compacted to a high density.
Increasing the soils density improves its strength, lowers its permeability, and reduces future
settlement. Compaction is achieved in the field by using hand-operated tampers, sheep-foot
rollers, rubber-tired rollers, or other types of rollers. The maximum density achieved as a
result of compaction with rollers and other types of compaction equipment is measured in the
field and compared with the maximum dry density of the soil previously determined in
laboratory tests. This is the most common method of quality control at construction sites.
The moisture-density relationship of soils was first studied by Procter, and the test sometimes
is known as Procter test. The relationship for practically all soils takes the form shown in
Figure 4-2. The dry density that can be obtained by compaction varies with the moisture
content, type of soil being compacted, and the compaction effort. It can be seen from this
relationship that for any compactive effort, the dry density of a soil will vary with its water
content. At low moisture content, the soil is dry and stiff to compress the particles. As water is
added the particles get lubricated by water film and is easy to compress the soil mass. Thus,
the density increases and the air content decreases as the water fills the voids and compacted.
The moisture content at which maximum dry density is obtained is known as optimum
moisture content (OMC). At moisture content higher than the OMC, the air and water in the
soil mass tend to keep particles apart and prevent compaction. The dry density at higher
moisture contents than OMC, thus, decreases and the total voids increase.

Figure 4-2. Typical moisture-density relationship for soils

The zero-air void curve shown in Figure 4-2 is the theoretical moisture-density curve for a
saturated soil and zero-air voids where the degree of saturation is 100%. This curve is usually
not attained in the field, since zero-air void cannot be attained. Points on the curve may be
calculated by

w Gs
1 wG s

where, d = dry density of soil,

w = density of water,
Gs = specific gravity of soil particles, and
w = moisture content of soil.
The distance between the zero-air void (ZAV) cure and the test moisture-density curve is an
indication of the amount of air voids remaining in the soil at different moisture contents. The
farther away a point on the moisture-density curve is from the ZAV curve, the more air voids
remain in the soil and the higher is the likehood of expansion or swelling if the soil is exposed
to water. Swelling of such soil can be reduced by compacting at higher moisture content.

a - Gravel-sand mixture
b - Well graded sand
c - Uniform sand
d - Mixed soil
e - Heavy clay

Figure 4-3. Compaction curves for different types soils

Soil type and gradation heavily affect the density that can be achieved by compaction.
Granular, well-graded soils generally have fairly high maximum densities at low optimum
moisture contents, while clay soils have lower densities. The edge-to-side bonds between clay
particles resist compactive efforts to force them into a denser structure. With granular soils,
the more well-graded soils have spaces between large particles that fill with smaller particles
when compacted, leading to a higher density than with uniform soils. Typical moisturedensity curves for different types of soils are shown in Figure 4-3. Note that a line joining the
peak points of the density curves would be approximately parallel to the ZAV curve. This is
due to the fact that most soils at their maximum density still contain about 2-3% air.
The OMC and the maximum dry density that can be attained on a given soil also depend on
the compactive effort used as shown in Figure 4-4. Compactive effort is a measure of the
6

mechanical energy imposed on the soil mass during compaction. Since Procters time the
loads imposed on pavementsnotably by airplaneshave increased tremendously, a revised
test using a much higher compactive effort is now often used.

Figure 4-4. Effect of compactive effort in dry density

Laboratory compaction test. The soil sample is first air dried and sieved usually through the
4.75-mm (No. 4) sieve, mixed thoroughly with water and then compacted in layers. The mass
of the compacted sample is measured, and a sample of it is taken to dry for the purpose of
determining water content. More water is then added to the soil, and it is compacted again.
This procedure is repeated until the density decreases. In the standard compaction test, the soil
is compacted in a 10 cm (4 in) diameter mold having a volume of 943.9 cm3 (1/30 ft3). The
soil is placed in three layers, each compacted with 25 blows of a 2.5 kg (5.5 1b) hammer
falling 30 cm (12 in). If coarser grains are to be included in the test, a larger mold 15 cm (6
in) in diameter with a volume of 2124 cm3 (0.075 ft3) is used. With these coarse materials,
the number of blows per layer is increased to 56 (from 25) to maintain the same compactive
effort. This compactive effort was originally chosen by Proctor to approximate the effort
provided by construction rollers. In the revised compaction test, known as modified Procter or
heavy compaction test, the compactive effort is provided by a 4.5 kg (10 lb) hammer falling
45 cm (18 in) on each of five layers in the same mold. The maximum dry density obtained in
the modified test is higher than that obtained in the standard test, and occurs at lower
optimum moisture content (see Figure 4-4).
Field density test. Since the compactibility of soils varies considerably, the construction
requirements for roads are usually specified as a percentage of the maximum dry density found in
a laboratory compaction test for each soil type encountered on the project. For example, a
project specification might require that the soil be compacted to 95% of the maximum dry
density found by the standard compaction test. Quality control of compaction on a
construction project involves conducting standard field compaction tests on each different soil
type after compaction, and comparing the results with the laboratory maximum density value
for the soil, to ascertain if the specifications have been met. If the maximum dry density from
the test was 2000 kg/m3 at an optimum water content of 11%, the required field density would
be 95% of 2000, or 1900 kg/m3. The moisture content of the soil should be as close as
possible to 11%, which reduces the required compactive effort (for example, number of
passes of the roller).

Field density tests are made using either destructive or nondestructive methods. In destructive
methods, the simplest is the core-cutter method. This method can be used only on cohesive
soils free from coarse-grained material. It involves driving a hollow metal cylinder, which has
a cutting edge, into the soil to remove an undisturbed sample on which dry density and
moisture content determinations can be made. The other commonly used method is in which a
sample of compacted material is dug out of a test hole in the soil layer being checked. The dry
mass of the soil removed and the volume of the hole are measured. The field dry density
equals the dry mass divided by the volume originally occupied by the sample. Two basic
methods are used to measure the volume of the hole: a balloon filled with a liquid and a sandcone apparatus. Using the balloon apparatus, the volume of the sample hole is found by
forcing a liquid-filled balloon into the test hole. The rubber membrane allows the fluid to fill
all the cavities in the test hole. The volume of fluid required to do this is read on a scale on the
apparatus. With the sand-cone apparatus, the volume of the test hole is obtained from the
mass of loose sand required to the fill the hole.
The quick and nondestructive method of measuring the in situ density and moisture content of
the compacted soil is the nuclear method. Using the nuclear equipment, the density is
obtained by measuring the scatter of gamma radiation by the soil particles since the amount of
rays is proportional to the bulk density of the soil. The moisture content is also obtained by
measuring the scatter of neutrons emitted in the soil due to the presence of hydrogen atoms.
The detector in the nuclear equipment measures the amount of rays and the neutrons that
passes through the soil, and thus the density and the moisture content can be calculated.
4.2.5. California Bearing Ratio (CBR) Test
Originally developed by the California Division of Highways, CBR test has been modified
and has been the most common strength test conducted on soils and other unbound road
making materials. In this test, laboratory specimens compacted in a standard CBR mold at the
OMC are tested to give a relative strength of the material for a pavement structure with
respect to crushed rock, which is considered an excellent coarse base material. A load is
applied by cylindrical metal plunger of 56 mm diameter to penetrate the specimen at a rate of
one mm per minute and readings of the applied load are taken at intervals of penetration of
0.25 mm up to a total penetration of not more than 7.5 mm. Depending upon the prevailing
climatic conditions of the site, CBR specimens are immersed in water for four days before the
test to obtain a saturation condition similar to what may occur in the field. During this period,
the sample is loaded with a surcharge load that simulates the estimates weight of pavement
layers over the material tested. Any swell due to soaking is also measured. Experience
suggests that the CBR test is of poor reproducibility, particularly with granular soils.
The CBR value is reported as a percentage of a standard value which is intended to represent
the value that would be obtained with compacted crushed stone. Typical test results are
illustrated in Figure 4-5. It sometimes happens that the plunger is still not perfectly bedded in
the specimen and, as a result of this and other factors, a load-penetration curve with a shape
similar to that of curve for Test 2 in Figure 4-5 may be obtained instead of the more normal
shaped curve illustrated by the curve for Test 1. When this happens the curve must be
corrected by drawing a tangent at the point of greatest slope and then transposing the axis of
load so that zero penetration is taken as the point where the tangent cuts the axis of
penetration. The corrected load-penetration curve is the tangent from the new origin to the
point of tangency and then the curve itself as illustrated in Figure 4-5.
8

Figure 4-5. Typical results from the CBR test

The CBR is then determined by reading off from the curve the load that causes a penetration
of 2.54 mm and dividing this value by the standard load (6.9 Mpa) required to produce the
same penetration in the standard crushed stone as

CBR

Unit load for 2.54 mm penetratio n in test specimen

100
Unit load for 2.54 mm penetratio n in sandard crushed roack

Similarly, the CBR at 5.08 mm penetration is obtained by dividing the load causing a
penetration of 5.08 mm with the load of 10.34 MPa required to produce the same penetration
in standard crushed stone. The two values are then compared and if the 2.54 mm value is
greater than the 5.08 value, it is the CBR of the material and used for design purposes. If on
the other hand the 5.08 mm value is larger, the test is entirely repeated on a fresh specimen. If
the new percentage valve at 5.08 mm penetration is still greater, then this is taken as the CBR
value.
Apparatus is also available to carry out in situ CBR tests in the field on exposed subgrades,
subbases, and bases. Such testes can be useful in investigating pavement failures and also in

examining existing roads in good condition. Accompanied by measurements of field densities

and moisture conditions, such testing provides a useful means of building up knowledge of
appropriate pavement design criteria for local soils under the locally prevailing climatic
conditions.
Design Subgrade CBR: The strength of subgrade soils is dependent on the type of soil,
density, and moisture content. The design CBR of the subgrade soil, therefore, should be
evaluated at the moisture content and density representative to the subgrade condition during
the service time of the pavement structure. For wet or moderate climatic zones and where the
ground water influences the subgrade moisture content, the CBR test is carried out after 4
days of soaking.
A road section for which a pavement design is undertaken should be subdivided into subgrade
areas where the subgrade CBR can be reasonably expected to be uniform, i.e. without
significant variations. Identification of sections deemed to have homogenous subgrade
conditions is carried out by desk studies on the basis of geology, pedology, drainage
conditions and topography, and considering soil categories which have fairly consistent
geotechnical characteristics (e.g. grading, plasticity, CBR). Usually, the number of soil
categories and the number of uniform subgrade areas will not exceed 4 or 5 for a given road
project. The design subgrade CBR for homogenous section is usually taken as the 90 %-ile
value of the CBR test results as shown in Figure 4-6.

Figure 4-6. Design CBR as the 90 % -ile value

4.2.6. Resilient Modulus Test
The resilient modulus, MR, is the elastic modulus obtained from repetitive load test that
simulates the actual pavement loading. MR is defined as the ratio of the repeated deviator
stress d to the recoverable axial strain r. It is well known that most paying materials are not
elastic but experience some permanent deformation after each load application. However, if
the load is small compared to the strength of the material and is repeated for a large numbers
of times, the deformation under each load repetition is nearly completely recoverable and
proportional to the load and can be considered as elastic.

10

Accumulated
plastic strain

Figure 4-7. Strains under repeated loads

Figure 4-7 shows the straining of a specimen under a repeated load test. At the initial stage of
load applications, there is considerable permanent deformation, as indicated by the plastic
strain. As the number of repetitions increases, the plastic strain due to each load repetitions
decreases. After 100 to 200 repetitions, the strain is practically all recoverable, as indicated by
r in the figure.
A triaxial device equipped for repetitive load condition is used to carryout resilient modulus
test. The test may be conducted on all types of unbounded pavement materials ranging from
cohesive to stabilized materials. The resilient modulus of asphalt mixtures can also be
determined by the repeated load application in tensile test.
4.2.7. Other Soil Tests
The soil tests presented above are only those most commonly used in highway engineering.
There are other tests likely to be used for soil surveys, design and control of construction
depending on the site conditions encountered and structures to be constructed. These include
field and laboratory tests such as those conducted to determine the shear and compression
strength, settlement, and permeability of soils. Procedures for these and those described above
are defined in BS, ASTM, AASHTO and other equivalent standards. Whenever any of these
tests are conducted, an appropriate standard procedure should always be employed.

4.3. Soil Classifications

The previous discussions have shown that soils are variable engineering materials. As a
means of obtaining general behavior, soils are systematically categorized on the basis of some
common characteristics obtained from visual description, during soil boring, and laboratory
tests. As a result very many soil classification systems are in use throughout the world in
different areas of study. In highway engineering, soils are classified by conducting relatively
simple tests on disturbed samples to serve as a means of identifying suitable materials and
predicting the probable behavior when used as subgrade or subbase material. The most
commonly used classification systems for highway purposes are the American Association of
State Highway and Transportation Officials (AASHTO) Classification System and the

11

Unified Soil Classification System (USCS). These classification systems only help engineers
to predict how the soil will behave if used as a subgrade or subbase material, however, the
information obtained should not be regarded as a substitute for the detailed investigation of
the soil properties.
4.3.1. AASHTO Classification System
The AASHTO Classification System is based on the Public Roads Classification System that
was developed from the results of extensive research conducted by the Bureau of Public
Roads, now known as the Federal Highway Administration of the United States. Several
revisions have been made to the system since it was first published. The system has been
described by AASHTO as a means for determining the relative quality of soils for use in
embankments, subgrades, subbases, and bases.
Table 4-3 AASHTO soil classification system

In this system of classification, soils are categorized into seven groups, A-1 through A-7, with
several subgroups, as shown in Table 4-3. The classification of a given soil is based on its
particle size distribution, LL, and PI. Soils are evaluated within each group by using an
empirical formula to determine the group index (G1) of the soils, given as
GI = (F - 35)[0.2 + 0.005(LL - 40)] + 0.01(F - 15)(PI - 10)
where, GI = group index
F = % of soil particles passing 0.075 mm (No. 200) sieve in whole number based on
material passing 75 mm (3 in.) sieve,
LL = liquid limit expressed in whole number, and
PI = plasticity index expressed in whole number.

12

The GI is determined to the nearest whole number. A value of zero should be recorded when a
negative value is obtained for the GI. Also, in determining the GI for A-2-6 and A-2-7
subgroups, the LL part is not used, that is, only the second term of the equation is used.
Classifying soils under the AASHTO system is finding the correct group for the particle size
distribution and atterberg limits of the soil from the classification Table 4-3. The group is then
designated using the GI value. Granular soils fall into classes A-1 to A-3. A-1 soils consist of
well-graded granular materials, A-2 soils contain significant amounts of silts and clays, and
A-3 soils are clean but poorly graded sands. A-4 soils cover non-plastic or moderately plastic
soils, and A-5 contains similar material to Group A-4 but exhibits high LL. A-6 soils are
typical plastic clays exhibiting high volume change between wet and dry states. Group A-7
covers plastic clays, having high values of LL and PI and show high volume change.
In general, according to the AASHTO system of classification, the suitability of a soil deposit
for use in highway construction can be summarized as follows.
1. Soils classified as A-1-a, A-1-b, A-2-4, A-2-5, and A-3 can be used satisfactorily as
subgrade or subbase material if properly drained. In addition, such soils must he
properly compacted and covered with an adequate thickness of pavement for the
surface load to be carried.
2. Materials classified as A-2-6, A-2-7, A-4, A-5, A-6, A-7-5, and A-7-6 will require a
layer of subbase material if used as subgrade. If these are to be used as embankment
materials, special attention must be given to the design of the embankment.
3. Generally, as the GI of a soil increases its value as subgrade material decreases. For
example, a soil with a GI of 0 (an indication of a good subgrade material) will be
better as a subgrade material than one with GI of 20 (an indication of a poor subgrade
material).
4.3.2. Unified Soil Classification System (USCS)
Originally developed by Casagrande during World War II for use in airfield construction,
USCS has been modified several times to obtain the current version. The fundamental
premise used in the USCS system is that, the engineering properties of any coarse-grained soil
depend on its particle size distribution, whereas those for a fine-grained soil depend on its
plasticity. Thus, the system classifies coarse-grained soils on the basis of grain size
characteristics and fine-grained soils according to plasticity characteristics.
In this system of classification, material that is retained in the 75 mm (3 in.) sieve is recorded,
but only that which passes is used for the classification of the sample. Soils are designated by
letter symbols with each letter having a particular meaning as defined as follows:
Coarse-grained soils. Soils with particles being retained on the No. 200 sieve are classified
as coarse-grained soils. The coarse-grained soils are subdivided into gravels (G) soils
having more than 50 percent of their particles larger than4.75 mm (i.e., retained on No. 4
sieve) and sands (S) those with more than 50 percent of their particles smaller than 4.75
mm (i.e., passed through No. 4 sieve). The gravels and sands are further divided into four
subgroup

13

each based on grain size distribution and the nature of the fine particles in them as well
graded (W), poorly graded (P), silty (M), or clayey (C). Gravels can be described as either
well-graded gravel (GW), poorly graded gravel (GP), silty gravel (GM), or clayey gravels
(GC), and sands can be described as well-graded sand (SW), poorly graded sand (SP), silty
sand (SM), or clayey sand (SC).
Table 4- 1. Unified Soil Classification System

14

A gravel or sandy soil is described as well graded or poorly graded, depending on the values
of two shape parameters known as the coefficient of uniformity, Cu, and the coefficient of
curvature, Cc given as
Cu

D60
D10

( D30 ) 2
Cc
D10 D60

and

where, D60 = grain diameter at 60% passing

D30 = grain diameter at 30% passing
D10 = grain diameter at 10% passing
Accordingly, gravels are described as well graded if Cu is above 4, and Cc is between 1 and 3.
Sands are also described as well graded if Cu is above 6, and Cc is between 1 and 3.
Moreover, coarse-grained soils with more than 12 percent fines (i.e., passes No. 200 sieve)
are classified as silty or clayey depending on their LL plots. Those soils with plots below the
A line (defined as below) or with a PI less than 4 are silty gravel (GM) or silty sand (SM),
and those with plots above the "A line with a PI greater than 7 are classified as clayey
gravels (GC) or clayey sands (SC).
Fine-grained soils. Soils with less than 50 percent of their particles retained on the No. 200
sieve are classified as fine-grained soils. The fine-grained soils are subdivided into clays (C)
or silt (M) based on a plasticity chart plotted PI versus LL of the soil from which a dividing
line known as the "A" line separates the more clayey materials from the silty materials. The
equation of the A line is

PI 0.73LL 20
Soils that fall below the A line are silty soils, whereas those with plots above the "A line
are clayey soils. Organic clays are an exception to this general rule since they plot below the
"A" line. Organic clays, however, generally behave similarly to soils of lower plasticity. The
organic, silty, and clayey soils are further divided into two groups, one having a relatively low
LL (L) and the other having a relatively high LL (H). The dividing line between high LL soils
and low LL soils is arbitrarily set at 50 percent. Fine-grained soils are, thus, further classified
as either silt with low plasticity (ML), silt with high plasticity (MH), clays with high plasticity
(CH), clays with low plasticity (CL), or organic with high plasticity (OH).

4.4. Stabilization
The term soil stabilisation may be defined as the alteration of the properties of an existing
soil either by blending (mixing) two or more materials and improving particle size

15

distribution or by the use of stabilizing additives to meet the specified engineering properties.
Quite often soils are stabilized for road construction in most parts of the world for the
following one or more objectives:
a.

Improve the strength (stability and bearing capacity) for subgrade, subbase, base,
low-cost road surfaces,

b.

Improve the volume stability to control swelling and shrinkage characteristics

caused by change in moisture,

c.

Improve durability to increase the resistance to erosion, weathering or traffic

usage, and

d.

Improve permeability to reduce permeability and hence the passage of water

through the stabilised soil.

Due to their mineralogical composition, soils may be rather complex materials. Stabilization
is therefore not a straightforward application of a given stabilizing agent; a number of aspects
should be taken into account in the selection of the proper stabilization technique. The factors
that should be considered include physical and chemical composition of the soil to be
stabilized, availability and economical feasibility of stabilising agents, ease of application, site
constraints, climate, curing time, and safety. Such factors should be taken into account in
order to select the proper type of stabilisation.
Basically four techniques of soil stabilization are commonly practiced in pavement
construction. These are: -

Mechanical stabilization,

Cement stabilization,

Lime stabilization, and

Bitumen stabilization.

Mechanical stabilization is a method by which a soil or gravel is mixed with the original soil
in order to improve the grading and mechanical characteristics of the soil. Other methods of
stabilization use additives such as cement, lime and bitumen to improve strength, workability
or waterproofing.

16

The principal cementing materials that may be used include Portland cement, lime, limeflyash mixtures and bitumen. Portland cement has been used with great success to improve
existing gravel roads, as well as to stabilize natural soils. It can be used for base courses and
subbases of all types. It can be used in granular soils, silty soils, and lean clays, but it cannot
be used in organic materials. Since soil cement shows strength gains over that of the natural
material, it is very often used for base-course construction. Another cementing agent, which is
often used, is hydrated lime. Lime increases soil strength primarily by pozzolanic action,
which is the formation of cementatious silicates and aluminates. This material is most
efficient when used in granular materials and lean clays; the quantity required for proper
hydration generally is relatively low. As will be seen in the coming sections, the suitability of
these methods depends on site constraints, materials, climate, and economic feasibility.

4.4.1. Mechanical Stabilization

As has been mention before, mechanical stabilization is an improvement of an available
material by blending it with one or more materials in order to improve the particle size
distribution and plasticity characteristics. Typical materials used for mechanical stabilization
include river deposited sand, natural gravel, silty sands, sand clays, silt clays, crushed run
quarry products and waste quarry products, volcanic cinders and scoria, poorly graded
laterites and peach sands, etc. Materials produced by blending have properties similar to
conventional unbounded materials and can be evaluated by ordinary methods.
The principal properties affecting the stability of compacted base or sub-base materials are
internal friction and cohesion. Internal friction is chiefly dependent on the characteristics of
the coarser soil particles, i.e. gravel, sand and silt sizes. The cohesion, shrinkage, swelling and
compressibility are mainly associated with the quantity and nature of the clay fraction as
indicated by plastic properties.
Preliminary mix design of mechanical stabilization is based on particle size distribution and
plastic properties. It is desirable also that strength tests (CBR, etc.) be carried out to verify
that the required improvement has been achieved. When unconventional materials are used,
more detailed testing and investigation will often be needed and may include the modification
of the accepted design or specification criteria.

17

Particle Size Distribution

While maximum frictional strength does not necessarily coincide with maximum density, the
achievement of a high density will generally provide a high frictional strength. Maximum
density, obtained with the closest packing and minimum voids, has been shown
experimentally to occur with a particle size distribution of the form:
P = 100 (d/D)n
where, P is the percentage passing the sieve with aperture d, D is the maximum particle size,
and n is the exponent. With most soils and other material, the value of the exponent 'n' for
maximum density is usually 0.45 to 0.50. However, with some materials, e.g. gravel-sandclays, high densities can be achieved with n values as low as 0.33. For materials with a
maximum size of 19 mm, the amount of fines passing the 75 m sieve will be 6 and 8 percent
for n values of 0.5 and 0.45 respectively. In certain cases higher percentages of material
passing the 75 m sieve may provide the best performance.
When the pavement design relies on a relatively low permeability in the pavement courses,
the materials used should be of uniform particle size of distribution within the limits
established by substituting values of 0.50 and 0.33 for n in the above equation. These limits
are sufficiently wide to allow for variations that will inevitably occur in field mixing.
However, if plant mixing is undertaken, more restrictive limits may be set. Where the value of
the exponent n is less than 0.33, the fines content of the material may be excessive. A high
fine content will result in reduced permeability and may lead to the development of pore
pressures and consequent instability during compaction or in service. Where n is greater
than 0.5, the material tends to be harsh, and may be prone to segregation and ravelling and
therefore more difficult to work.

18

Figure 4- 1. General properties of mechanically stabilised gradings

Liquid Limit and Plasticity Index
The limits listed in Table 4- 2 are generally accepted as satisfactory design criteria for
mechanical stabilized bases.
Table 4- 2. Limits for plastic properties of mechanical stabilised bases*
Type of Pavement
Sealed
Unsealed

Liquid Limit
(Percent)
Not more than 25
Not more than 35

Plasticity Index
(Percent)
Not more than 6
4 to 9

*Slightly higher values may be used for criteria under certain circumstances as follows:
(i)

For rainfall less than about 375 mm per year, some thought may be given to
increasing the upper limits of Plasticity Index to 10 for sealed pavements and
15 for unsealed pavements. Some relaxation of limits for well-drained
pavements may also be considered.

19

(ii)

When the percentage of soil binder is low, as a rough rule, the product of the
Plasticity Index and the percentage passing the 425 gm sieve (relative to the
whole material) should not exceed 200 for gravel to receive bituminous surface
treatment. In arid climates, consideration could be given to relax this product
to about 400, provided road formations are well drained.

(iii)

Slightly wider limits of Plasticity Index may prove satisfactory with some
ironstone gravels and limestone rubbles, if the soil binder has some natural
setting properties. This should not be taken as a general rule - each case should
be treated on its merits and caution should be exercised in dealing with new
and unfamiliar materials. In the case of major works it is advisable to construct
trial sections of pavement for evaluation at least two years before embarking
upon their large-scale use.

Strength Tests
Stabilized materials may be assessed by strength tests suitable for this purpose at the density
and moisture conditions prevailing in the pavement during the service life. It is important, in
the testing of a potential base material, to be able to predict its moisture condition. By this
means, the failure envelopes at moisture conditions bracketing the equilibrium moisture
conditions and at the required density anticipated in the proposed pavement may be derived.
The equilibrium moisture conditions to be expected in a pavement may be obtained by
examining existing roads constructed from materials similar to those being investigated and
assembling such information for future use.
One of the most commonly used strength tests is the laboratory CBR test. The values given in
Table 3.3 have generally been found to be applicable. A 4-day soaking of compacted
specimens before testing is generally used. Conditions adopted for the test may be altered in
respect of the degree of compaction and moisture content, to simulate the worst conditions
expected in service. In some circumstances, conditioning the specimens by soaking for 4 days
might be too conservative, and in other cases a period longer than 4 days might be more
appropriate for relatively impermeable materials. In this case the adoption of a minimum CBR
value different to those tabulated above should be considered.
Table 4- 3. California bearing ratio limits for mechanical stabilised materials

20

Pavement
High class, high traffic volume
Rural roads, wet areas
Rural roads, and areas

Minimum values of CBR

(Percent)
100
80
60

The selection of suitable criteria should take account of local experience, especially that
related to the performance of local materials. Design of stabilised mixtures involves
characterising the individual materials, proportioning them to fit the selected criteria, and
making up a trial mixture to check that the preferred proportions do provide the desired
qualities. In addition to adequate investigation and design, good construction and control
testing techniques are essential if a satisfactory road pavement is to result. This involves
careful proportioning and thorough mixing of the constituents materials to produce a uniform
unsegregated final product which can be compacted and finished in accordance with the
specification.

21