PART I
Physical Properties of Soils
The subject matter of Part I is divided into four chapters.
The first deals with the procedures commonly used to
discriminate among different soils or among different
states of the same soil. The second deals with the methods
and program of soil exploration. The third is concerned
with the hydraulic and mechanical properties of soils
and with the experimental methods used to determine
numerical values representative of these properties. The
fourth chapter deals with the physical processes involved
in the drainage of soils.
1
CHAPTER 1
Index Properties of Soils
ARTICLE 1 PRACTICAL IMPORTANCE OF
INDEX PROPERTIES
In geotechnical engineering, more than in any other field
of civil engineering, success depends on practical experi-
ence. The design of ordinary soil-supporting or soil-sup-
ported structures is necessarily based on simple empirical
rules, but these rules can be used safely only by the
engineer who has a background of experience. Large
projects involving unusual features may call for extensive
application of scientific methods to design, but the pro-
gram for the required investigations cannot be laid out
wisely, nor can the results be interpreted intelligently,
unless the engineer in charge of design possesses a large
amount of experience.
Since personal experience is necessarily somewhat lim-
ited, the engineer is compelled to rely at least to some
extent on the records of the experiences of others. If
these records contain adequate descriptions of the soil
conditions, they constitute a storehouse of valuable infor-
mation. Otherwise, they may be misleading. Conse-
quently, one of the foremost aims in attempts to reduce
the hazards in dealing with soils has been to find simple
methods for discriminating among the different kinds of
soil in a given category. The properties on which the
distinctions are based are known as index properties, and
the tests required to determine the index properties are
classification tests.
The nature of any soil can be altered by appropriate
manipulation. Vibrations, for example, can transform a
loose sand into a dense one. Hence, the behavior of a
soil in the field depends not only on the significant proper-
ties of the individual constituents of the soil mass, but
also on those properties that are due to the arrangement
of the particles within the mass. Accordingly, it is conve-
nient to divide index properties into two classes: soil
grain properties and soil aggregate properties. The prin-
cipal soil grain properties are the size and shape of the
grains and, in clay soils, the mineralogical character of
the smallest grains. The most significant aggregate prop-
erty of cohesionless soils is the relative density, whereas
that of cohesive soils is the consistency.
The discussion of soil grain and aggregate properties
is preceded by a description of the principal types of soil,
and it is followed by a condensed review of the minimum
requirements for adequate soil descriptions to be incorpo-
rated in the records of field observations.
ARTICLE 2 PRINCIPAL TYPES OF SOILS
The materials that constitute the earths crust are rather
arbitrarily divided by the civil engineer into the two cate-
gories, soil and rock. Soil is a natural aggregate of mineral
grains that can be separated by such gentle mechanical
means as agitation in water. Rock, on the other hand, is
a natural aggregate of minerals connected by strong and
permanent cohesive forces. Since the terms strong and
permanent are subject to different interpretations, the
boundary between soil and rock is necessarily an arbitrary
one. As a matter of fact, there are many natural aggregates
of mineral particles that are difficult to classify either as
soil or as rock. In this text, however, the term soil will
be applied only to materials that unquestionably satisfy
the preceding definition.
Although the terminology described in the preceding
paragraph is generally understood by civil engineers, it
is not in universal use. To the geologist, for example, the
term rock implies all the material that constitutes the
earths crust, regardless of the degree to which the mineral
particles are bound together, whereas the term soil is
applied only to that portion of the earths crust that is
capable of supporting vegetation. Therefore, the civil
engineer who makes use of information prepared by
workers in other fields must understand the sense in which
the terms soil and rock are used.
On the basis of the origin of their constituents, soils
can be divided into two large groups, those that consist
chiefly of the results of chemical and physical rock weath-
ering, and those that are chiefly of organic origin. If the
products of rock weathering are still located at the place
3
4 INDEX PROPERTIES OF SOILS
where they originated, they constitute a residual soil.
Otherwise they constitute a transported soil, regardless
of the agent that performed the transportation.
Residual soils that have developed in semiarid or tem-
perate climates are usually stiff and stable and do not
extend to great depth. However, particularly in warm
humid climates where the time of exposure has been long,
residual soils may extend to depths of hundreds of meters.
They may be strong and stable, but they may also consist
of highly compressible materials surrounding blocks of
less weathered rock (Article 47.12). Under these circum-
stances they may give rise to difficulties with foundations
and other types of construction. Many deposits of trans-
ported soils are soft and loose to a depth of more than a
hundred meters and may also lead to serious problems.
Soils of organic origin are formed chiefly in situ, either
by the growth and subsequent decay of plants such as
peat mosses or by the accumulation of fragments of the
inorganic skeletons or shells of organisms. Hence a soil
of organic origin can be either organic or inorganic. The
term organic soil ordinarily refers to a transported soil
consisting of the products of rock weathering with a more
or less conspicuous admixture of decayed vegetable
matter.
The soil conditions at the site of a proposed structure
are commonly explored by means of test borings or test
shafts. The inspector on the job examines samples of the
soil as they are obtained, classifies them in accordance
with local usage, and prepares a boring log or shaft record
containing the name of each soil and the elevation of its
boundaries. The name of the soil is modified by adjectives
indicating the stiffness, color, and other attributes. At a
later date the record may be supplemented by an abstract
of the results of tests made on the samples in the
laboratory.
The following list of soil types includes the names
commonly used for field classification.
Sand and gravel are cohesionless aggregates of
rounded subangular or angular fragments of more or less
unaltered rocks or minerals. Particles with a size up to 2
mm are referred to as sand, and those with a size from
2 mm to 200 mm as gravel. Fragments with a diameter
of more than 200 mm are known as boulders.
Hardpan is a soil that has an exceptionally great resis-
tance to the penetration of drilling tools. Most hardpans
are extremely dense, well-graded, and somewhat cohesive
aggregates of mineral particles.
Inorganic silt is a fine-grained soil with little or no
plasticity. The least plastic varieties generally consist of
more or less equidimensional grains of quartz and are
sometimes called rock flour; whereas the most plastic
types contain an appreciable percentage of flake-shaped
particles and are referred to as plastic silt. Because of its
smooth texture, inorganic silt is often mistaken for clay,
but it may be readily distinguished from clay without
laboratory testing. If shaken in the palm of the hand, a
pat of saturated inorganic silt expels enough water to
make its surface appear glossy. If the pat is bent between
the fingers, its surface again becomes dull. This procedure
is known as the shaking test. After the pat has dried, it
is brittle and dust can be detached by rubbing it with the
finger. Silt is relatively impervious, but if it is in a loose
state it may rise into a drill hole or shaft like a thick
viscous fluid. The most unstable soils of this category are
known locally under different names, such as bulls liver.
Organic silt is a fine-grained more or less plastic soil
with an admixture of finely divided particles of organic
matter. Shells and visible fragments of partly decayed
vegetable matter may also be present. The soil ranges in
color from light to very dark gray, and it is likely to
contain a considerable quantity of H2S, COz, and various
other gaseous products of the decay of organic matter
which give it a characteristic odor. The permeability of
organic silt is very low and its compressibility very high.
Clay is an aggregate of microscopic and submicro-
scopic particles derived from the chemical decomposition
of rock constituents. It is plastic within a moderate to
wide range of water content. Dry specimens are very
hard, and no powder can be detached by rubbing the
surface of dried pats with the fingers. The permeability
of clay is extremely low. The term gumbo is applied,
particularly in the western United States, to clays that are
distinguished in the plastic state by a soapy or waxy
appearance and by great toughness. At higher water con-
tents they are conspicuously sticky.
Organic clay is a clay that owes some of its significant
physical properties to the presence of finely divided
organic matter. When saturated, organic clay is likely to
be very compressible, but when dry its strength is very
high. It is usually dark gray or black and it may have a
conspicuous odor.
Peat is a somewhat fibrous aggregate of macroscopic
and microscopic fragments of decayed vegetable matter.
Its color ranges between light brown and black. Peat is
so compressible that it is almost always unsuitable for
supporting foundations. Various techniques have been
developed for carrying earth embankments across peat
deposits without the risk of breaking into the ground, but
the settlement of these embankments is likely to be large
and to continue at a decreasing rate for many years.
If a soil is made up of a combination of two different
soil types, the predominant ingredient is expressed as a
noun, and the less prominent ingredient as a modifying
adjective. For example, silty sand indicates a soil that is
predominantly sand but contains a small amount of silt.
A sandy clay is a soil that exhibits the properties of a
clay but contains an appreciable amount of sand.
The aggregate properties of sand and gravel are
described qualitatively by the terms loose, medium, and
dense, whereas those of clays are described by hard, stifJ;

medium, and soft.These terms are usually evaluated by
the boring foreman or inspector on the basis of several
factors, including the relative ease or difficulty of advanc-
ing the drilling and sampling tools and the consistency
of the samples. However, since this method of evaluation
may lead to a very erroneous conception of the general
character of the soil deposit, the qualitative descriptions
should be supplemented by quantitative information
whenever the mechanical properties are likely to have an
important influence on design. The quantitative informa-
tion is commonly obtained by means of laboratory tests
on relatively undisturbed samples (Article 11.3), or by
suitable in situ tests (Articles 11.4 and 11.5).
A record of the color of the different strata encountered
in adjacent borings reduces the risk of errors in correlating
the boring logs. Color may also be an indication of a real
difference in the character of the soil. For example, if the
top layer of a submerged clay stratum is yellowish or
brown and stiffer than the underlying clay, it was probably
exposed temporarily to desiccation combined with weath-
ering. Terms such as mottled, marbled, spotted, or speck-
led are used when different colors occur in the same
stratum of soil. Dark or drab colors are commonly associ-
ated with organic soils.
Under certain geological conditions soils form that are
characterized by one or more striking or unusual features
such as a root-hole structure or a conspicuous and regular
stratification. Because of these features, such soils can
easily be recognized in the field and, consequently, they
have been given special names by which they are com-
monly known. The following paragraphs contain defini-
tions and descriptions of some of these materials.
Till is an unstratified glacial deposit of clay, silt, sand,
gravel, and boulders.
Tuff is a fine-grained water- or wind-laid aggregate
of very small mineral or rock fragments ejected from
volcanoes during explosions.
Loess is a uniform, cohesive, wind-blown sediment,
and is commonly light brown. The size of most of the
particles ranges between the narrow limits of 0.01 and
0.05 mm. The cohesion is due to the presence of a binder
that may be predominantly calcareous or clayey. Because
of the universal presence of continuous vertical root holes,
the permeability in vertical direction is usually much
greater than in horizontal directions; moreover, the mate-
rial has the ability to stand on nearly vertical slopes. True
loess deposits have never been saturated. On saturation
the bond between particles is weakened and the surface
of the deposit may settle.
Mod@ed loess is a loess that has lost its typical charac-
teristics by secondary processes, including temporary
immersion, erosion and subsequent deposition, chemical
changes involving the destruction of the bond between
the particles, or chemical decomposition of the more per-
ishable constituents such as feldspar. Thorough chemical
ARTICLE 2 PRINCIPAL TYPES OF SOILS 5
decomposition produces loess loam, characterized by
greater plasticity than other forms of modified loess.
Diatomaceous earth (kieselguhr) is a deposit of fine,
generally white, saliceous powder composed chiefly or
wholly of the remains of diatoms. The term diatom applies
to a group of microscopic unicellular marine or fresh-
water algae characterized by silicified cell walls.
Lake marl or boglime is a white fine-grained powdery
calcareous deposit precipitated by plants in ponds. It is
commonly associated with beds of peat.
Marl is a rather loosely used term for various fairly
stiff or very stiff marine calcareous clays of greenish color.
Shale is a clastic sedimentary rock mainly composed of
silt-size and clay-size particles. Most shales are laminated
and display fissility; the rock has a tendency to split along
relatively smooth and flat surfaces parallel to the bedding.
When fissility is completely absent, the clastic sedimen-
tary deposit is called mudstone or clay rock. Depending
on clay mineralogy, void ratio, and degree of diagenetic
bonding or weathering, compressive strength of shales
may range from less than 2.5 MPa to more than 100 MPa.
Adobe is a term applied in the southwestern United
States and other semiarid regions to a great variety of
light-colored soils ranging from sandy silts to very plas-
tic clays.
Caliche refers to layers of soil in which the grains are
cemented together by carbonates deposited as a result of
evaporation. These layers commonly occur at a depth of
several meters below the surface, and their thickness may
range up to a few meters. A semiarid climate is necessary
for their formation.
Varved clay consists of alternating layers of medium
gray inorganic silt and darker silty clay. The thickness of
the layers rarely exceeds 10 mm, but occasionally very
much thicker varves are encountered. The constituents
were transported into freshwater lakes by melt water at
the close of the Ice Age. Varved clays are likely to com-
bine the undesirable properties of both silts and soft clays.
Bentonite is a clay with a high content of montmorillon-
ite (Article 4).Most bentonites were formed by chemical
alteration of volcanic ash. In contact with water, dried
bentonite swells more than other dried clays, and saturated
bentonite shrinks more on drying.
Each term used in the field classification of soils
includes a great variety of different materials. Further-
more, the choice of terms relating to stiffness and density
depends to a considerable extent on the person who exam-
ines the soil. Consequently, the field classification of soils
is always more or less uncertain and inaccurate. More
specific information can be obtained only by physical
tests that furnish numerical values representative of the
properties of the soil.
The methods of soil exploration, including boring and
sampling, and the procedures for determining average
6 INDEX PROPERTIES OF SOILS
numerical values for the soil properties are discussed in
Chapter 2.
ARTICLE 3 SIZE AND SHAPE OF SOIL
PARTICLES
The size of the particles that constitute soils may vary
from that of boulders to that of large molecules.
Grains larger than approximately 0.06 mm can be
inspected with the naked eye or by means of a hand lens.
They constitute the very coarse and coarse fractions of
the soils.
Grains ranging in size from about 0.06 mm to 2 p (1
p = 1 micron = 0.001 mm) can be examined only under
the microscope. They represent the fine fraction.
Grains smaller than 2 p constitute the veryfinefraction
(clay size fraction, CF).Grains having a size between 2
p and about 0.1 p can be differentiated under the micro-
scope, but their shape cannot be discerned. The shape of
grains smaller than about 1 p can be determined by means
of an electron microscope. Their molecular structure can
be investigated by means of X-ray analysis.
The process of separating a soil aggregate into frac-
tions, each consisting of grains within a different size
range, is known as mechanical analysis. By means of
mechanical analysis, it has been found that most natural
soils contain grains representative of two or more soil
fractions. The general character of mixed-grained soils
is determined almost entirely by the character of the
smallest soil constituents. In this respect soils are some-
what similar to concrete. The properties of concrete are
determined primarily by the cement, whereas the aggre-
gate, which constitutes most of the concrete, is inert. The
aggregate, or the inert portion of a mixed-grained soil,
comprises about 80 or 90% of the total dry weight. The
decisive or active portion constitutes the remainder.
Very coarse fractions, for example gravel, consist of
rock fragments each composed of one or more minerals.
The fragments may be angular, subangular, rounded, or
flat. They may be fresh, or they may show signs of
considerable weathering. They may be resistant or
crumbly.
Coarse fractions, exemplified by sand, are made up of
grains usually composed chiefly of quartz. The individual
grains may be angular, subangular, or rounded. Some
sands contain a fairly high percentage of mica flakes that
make them very elastic or springy.
In the fine and very fine fractions, any one grain usually
consists of only one mineral. The particles may be angu-
lar, flake-shaped, or tubular. Rounded particles, however,
are conspicuously absent. Exceptionally, the fine fraction
contains a high percentage of porous fossils, such as
diatoms or radiolaria, that produce abnormal mechanical
properties. In general, the percentage of flaky constituents
in a given soil increases with decreasing grain size of the
soil fraction.
If the size of most of the grains in an aggregate of soil
particles is within the limits given for any one of the soil
fractions, the aggregate is called a uniform soil. Uniform
very coarse or coarse soils are common, but uniform very
fine or colloidal soils are very seldom encountered. All
clays contain fine, very fine, and colloidal constituents,
and some clays contain even coarse particles. The finest
grain-size fractions of clays consist principally of flake-
shaped particles.
The widespread prevalence of flake-shaped particles
in the very fine fractions of natural soils is a consequence
of the geological processes of soil formation. Most soils
orginate in the chemical weathering of rocks. The rocks
themselves consist partly of chemically very stable and
partly of less stable minerals. Chemical weathering trans-
forms the less stable minerals into a friable mass of very
small particles of secondary minerals that commonly have
a scale-like or flaky crystal form, whereas the stable
minerals remain practically unaltered. Thus the process
of chemical weathering reduces the rock to an aggregate
consisting of fragments of unaltered or almost unaltered
minerals embedded in a matrix composed chiefly of dis-
crete scaly particles. During subsequent transportation
by running water the aggregate is broken up, and the
constituents are subjected to impact and grinding. The
purely mechanical process of grinding does not break up
the hard equidimensional grains of unaltered minerals
into fragments smaller than about 10 p (0.01 mm). On the
other hand, the friable flake-shaped particles of secondary
minerals, although initially very small, are readily ground
and broken into still smaller particles. Hence, the very
fine fractions of natural soils consist principally of flake-
shaped particles of secondary minerals.
ARTICLE 4 PROPERTIES OF VERY FINE
SOIL FRACTIONS
4.1 Mineralogical Composition
The most important grain property of fine-grained soil
materials is the mineralogical composition. If the soil
particles are smaller than about 0.002 mm, the influence
of the force of gravity on each particle is insignificant
compared with that of the electrical forces acting at the
surface of the particle. A material in which the influence
of the surface charges is predominant is said to be in
the colloidal state. The colloidal particles of soil consist
primarily of clay minerals that were derived from rock
minerals by weathering, but that have crystal structures
differing from those of the parent minerals.
All the clay minerals are crystalline hydrous alumino-
silicates having a lattice structure in which the atoms are
arranged in several layers, similar to the pages of a book.
 ARTICLE 4 PROPERTIES OF VERY FINE SOIL FRACTIONS 7
The arrangement and the chemical composition of these
layers determine the type of clay mineral.
The basic building blocks of the clay minerals are
the silica tetrahedron and the alumina octahedron. These
blocks form tetrahedral and octahedral layers (Fig. 4.1),
different combinations of which produce a unit sheet of
the various types of clays.
A single particle of clay may consist of many sheets
or films piled one on another. Because each sheet or film
has a definite thickness but is not limited in dimensions
at right angles to its thickness, clay particles are likely
to exhibit flat or curved terraced surfaces. The surfaces
carry residual negative electrical charges, but the broken
edges may carry either positive or negative charges in
accordance with the environment.
In problems of interest to the civil engineer, clay parti-
cles are always in contact with water. The interactions
among the clay particles, the water, and the various mate-
rials dissolved in the water are primarily responsible for
the properties of the soil consisting of the particles.
The characteristics of the principal clay minerals are
Figure 4.2 Photomicrograph of kaolinite.
described in the following paragraphs.

4.2 Characteristics of Principal Clay Minerals
Kaolinite is one of the most common clay minerals in
sedimentary and residual soils (Grim 1968, Swindale
1975). A unit sheet of kaolinite, which is approximately
0.7 nm (nm = m) thick, is composed of one alumi-
num octahedral layer and one silicon tetrahedral layer,
joined together by shared oxygens. A typical particle of
kaolinite consists of a stack of sheets forming a stiff
hexagonal plate with flat-faced edges. It is about 100 nm
in thickness with a breadthhhickness of about 5 to 10,
and a specific surface of 5 to 15 m2/g. The scanning
electron microscope (SEM) photomicrograph in Fig. 4.2
shows a range of particle sizes and shapes including
terraced surfaces where packets of 0.7-nm sheets
terminate.
Hulloysite is one of the most common minerals in
residual soils, particularly those derived from volcanic
parent material. It is a member of the kaolin subgroup
of clay minerals. A unit sheet of hydrated halloysite,
including one molecular layer of water, is approximately
la (6)
0 Oxygen 0 Hydroxyl
SiJicon Aluminum, Magnesium,
Iton
Figure 4.1 (a) Tetrahedral layer. (b)Octahedral layer. (After
Grim 1968)
1 nm thick. A typical particle has the shape of a hollow
tube or prism with outside and inside diameters of 70
and 40 nm, respectively, and is 300 to 500 nm long. The
intersheet water in the hydrated halloysite is removed
irreversibly starting at 60" to 75C.
Allophane, a major constituent of young residual soils
formed from volcanic ash, is an amorphous hydrous alu-
minosilicate commonly associated with halloysite (Grim
1968, Fieldes and Claridge 1975, Wesley 1973). In some
residual soils, the transition to halloysite from allophane,
which forms at early stages of weathering, is not well
defined. Allophane consists of very loosely packed chains
of silica tetrahedra and alumina octahedra, cross-linked
at a relatively small number of points. In the natural state,
it exists as microaggregates of extremely fine particles
of the order of several nanometers and specific surface
areas of 250 to 800 m2/g. The allophane aggregates or
clusters are relatively incompressible. They are some-
times cemented by iron or aluminum oxides, and they
enclose a large amount of water. Since the aggregates
are susceptible to structural breakdown upon mechanical
manipulation, as by heavy equipment, a deposit of allo-
phane may change from a granular material to a plastic
sticky mass that cannot be handled easily. The natural
aggregates do not readily lose water; however, when water
is removed and the fabric shrinks, the process cannot be
reversed. The resulting soil, with silt- and sand-sized
particles that are quite hard, is practically nonplastic.
Illite is the most common clay mineral in stiff clays
and shales as well as in postglacial marine and lacustrine
soft clay and silt deposits (Grim 1968, Radoslovich 1975,
Reichenbach and Rich 1975). It is often present, some-
8 INDEX PROPERTIES OF SOILS
times interstratified with other sheet silicates, in sedimen-
tary and residual soils, except in residual soils derived
from amorphous volcanic material. Illite is also referred
to as fine-grained mica and weathered mica.
The crystal structure of illite is similar to that of musco-
vite mica in the macroscopic form. However, in micro-
scopic illite particles the stacking of the sheets is not so
regular as in well-crystallizedmicas, and weathering may
remove intersheet K+ from the edges of the plates. The
resulting illite particles, with terraced surfaces where one
or more unit sheets terminate, have frayed and tattered
edges, are flexible and elastic, are 10 to 30 nm in thick-
ness, have a breadthhhickness of 15 to 30, and have a
specific surface of 80 to 100 m2/g. A SEM photomicro-
graph of illite particles is shown in Fig. 4.3.
Chlorite is a clay mineral commonly associated with
micas and illite, but is usually a minor component (Grim
1968, Bailey 1975). A unit sheet of chlorite, which is
about 1.4 nm thick, consists of one biotite mica sheet in
which all octahedral sites are occupied by magnesium and
one brucite sheet, an octahedral layer in which magnesium
atoms are in octahedral coordination with hydroxyls. The
biotite mica and brucite sheets are strongly bonded
together. The unit sheets are stacked and are connected
to each other by hydrogen bonding of surface oxygens
of the tetrahedral layer of mica and surface hydroxyls of
brucite. The size and platyness of chlorite particles are
similar to those of illite. However, in contrast to illite,
in which the unit sheets are bonded by potassium, the
hydrogen bonding between chlorite sheets results in pseu-
dohexagonal or euhedral-shaped platelets that are flexible
but inelastic.
Figure 4.3 Photomicrograph of illite.
Montmorillonite, the most common member of a group
of clay minerals known as smectites, is the dominant clay
mineral in some clays and shales and in some residual
soils derived from volcanic ash (Grim 1968, Mering
1975). Relatively pure seams of montmorillonite are
found in some deposits, Wyoming bentonite being the
best-known example. A unit sheet of montmorillonite is
similar to that of the micas. In montmorillonite,octahedral
A1 is partially replaced by Mg atoms. Each isomorphous
replacement (Article 4.3) produces a unit negative charge
at the location of the substituted atom, which is balanced
by exchangeable cations, such as Ca and Na+ situated
at the exterior of the sheets. In a packet of montmorillonite
in the anhydrous state, where as many as 10 unit sheets
are in contact, the stacking of the sheets is disordered in
the sense that the hexagonal cavities of the adjacent sur-
faces of two neighboring sheets are not matched face-to-
face. In a hydrous environment,water molecules penetrate
between the sheets and separate them by 1 nm (Le., four
molecular layers of water).
In sodium montmorillonite, the exchange capacity is
satisfied by Na cations. If the electrolyte concentration
in water is less than 0.3 N, further separation of unit
sheets takes place to more than 3 nm by difluse double-
layer repulsion (Article 4.4). The resulting particles of
sodium montmorillonite in water are thin films 1 nm
thick, with breadth-to-thickness ratios in excess of 100
and specific surfaces as much as 800 m2/g. In calcium
montmorillonite, the electrostatic .attraction of the Ca cat-
ions links the successive sheets together and prevents
separation beyond 1 nm. The resulting domains of 8 to
10 unit sheets, which are separated from each other by
up to four molecular layers of water, experience minimal
swelling. Among clay minerals, sodium montmorillonite
has the smallest and most filmy particles, as is shown by
the SEM photomicrograph in Fig. 4.4.
Atrupulgite is a fibrous clay mineral composed of silica
chains linked together by oxygens along their longitudinal
edges to form single laths or bundles of laths (Grim 1968,
Henin and Caillkre 1975). Water molecules fill the inter-
stices between the chains. The particles of attapulgite are
relatively rigid, 5 to 10 nm thick, 10 to 20 nm wide, and
0.1 to 1 pm in length, as shown in the SEM, Fig. 4.5.
The specific surface of the bundles is about 150 m2/g,
but it can be much higher for dispersed single laths.
Attapulgite, frequently associated with carbonate rocks,
is not a very common clay mineral in soil deposits. How-
ever, when present, it results in unusual physical proper-
ties such as a very high plastic limit, a very high plasticity
index, and high frictional resistance (Articles 7.2 and
19.2).
Mixed-layer Clay Minerals
Particles in some soils are composed of regularly or ran-
domly interstratified unit sheets of two or more types
 ARTICLE 4 PROPERTIES OF VERY FINE SOIL FRACTIONS 9
the physical properties of interstratified clay minerals are
not well defined, inasmuch as a wide range of combina-
tions is possible. Mixed-layer clay mineral particles pos-
sess properties generally representative of the
constituent minerals.
4.3 Role of Isomorphic Substitution
Isomorphic substitution is the replacement of a cation in
the mineral structure by another cation of lower electrova-
lence. The difference in the valences leads to a negative
charge, and the difference in size of the cations produces
a distortion of the mineral structure. Both effects tend to
decrease the resistance of a mineral structure to chemical
and mechanical weathering. Quartz is a space-lattice sili-
cate composed of silica tetrahedrons, (Si04)-4 linked
together by primary valence bonds to form a three-dimen-
sional network with the formula Si02.There is no isomor-
phous substitution in quartz, and each silica tetrahedron
is firmly and equally braced in all directions. As a result,
quartz has no planes of weakness and is very hard and
Figure 4.4 Photomicrograph of montmorillonite.
highly resistant to mechanical and chemical weathering.
Quartz is not only the most common mineral in sand-
and silt-sized particles of soils, but quartz or amorphous
silica is frequently present in colloidal (1 to 100 nm) and
molecular (<1 nm) dimensions (Mitchell 1975). In some
young soils of volcanic origin, clay-sized particles may
consist essentially of silica, an appreciable proportion of
which is amorphous. The boundary between quartz and
amorphous silica, however, is not distinct, inasmuch as
physical processes such as grinding may render crystal-
line quartz amorphous.
In feldspar, some of the silicon atoms are replaced by
aluminum. This results in a negative charge and in distor-
tion of the crystal structure, because A1 atoms are larger
than Si atoms. The negative charge is balanced by taking
in cations such as K', Na', and Cat* in orthoclase, albite,
and anorthite feldspars, respectively. The distortion of the
lattice and the inclusion of the cations cause cleavage
planes that reduce the resistance of feldspars to mechani-
cal and chemical weathering. For these reasons, feldspars
are not so common as quartz in the sand-, silt-, and clay-
sized fractions of soils, even though feldspars are the most
common constituent of the earth's crust.
Figure 4.5 Photomicrograph of attapulgite. Common micas such as muscovite and biotite are often
present in the silt- and sand-sized fractions of soils. In a
unit sheet of mica, which is 1 nm thick, two tetrahedral
of clay minerals (MacEwan and Ruiz-Ami1 1975). For layers are linked together with one octahedral layer. In
example, a marine or lacustrine clay deposit may include muscovite, only two of every three octahedral sites are
regularly alternating montmorillonite-illite particles, and occupied by aluminum cations, whereas in biotite all sites
a volcanic ash residual soil may contain randomly inter- are occupied by magnesium. In well-crystallized micas
stratified halloysite and hydrated halloysite. These mixed- one fourth of the tetrahedral Si+4are replaced by A1+3.
layer clay minerals often represent an intermediate phase The resulting negative charge in common micas is bal-
in the transformation, by weathering or through a diage- anced by intersheet potassiums. In a face-to-face stacking
netic process, of one mineral to another mineral, such as of sheets to form mica plates, the hexagonal holes on
that of mica to montmorillonite or vice versa. However, opposing tetrahedral surfaces are matched to enclose the
10 INDEX PROPERTIES OF SOILS
intersheet potassium. Perfect cleavage occurs along the
K-plane, however, as the intersheet bond strength of
( K O , Z ) - is
~ ~only 1/12 of the bond strength of (Si04)-4
inside the sheets.
There is very little isomorphous substitution in kaolin-
ite, the particles of which are formed by hydrogen bonding
of unit sheets. The edge charge is important, because the
particles are thick and the edge surface area is significant.
The edge charge, which results from broken bonds and
exposed oxygens at the edges of the particles, may be
positive or negative in accordance with the pH of the
water. When the pH is low (acidic condition), excess H+
ions associate with the oxygens at the edge of the particles,
and give the edge a net positive charge. When the pH is
high, H' ions disassociate from the edge, which then
becomes negatively charged. In mica and illite a major
part of the negative charge resulting from isomorphous
substitution is neutralized by K atoms which also connect
the successive unit sheets to each other to form particles.
4.4 Cation Exchange and Adsorbed Water
Cations that neutralize the net negative charge on the
surface of soil particles in water are readily exchangeable
with other cations. The exchange reaction depends mainly
on the relative concentrations of cations in the water and
also on the electrovalence of the cations. Cation exchange
capacity, measured in milliequivalents of cations per gram
of soil particles, is a measure of the net negative charge on
the soil particles, resulting from isomorphous substitution
and broken bonds at the boundaries. The values of the
cation exchange capacity for the principal clay minerals
are indicated in Table 4.1. Montmorillonite has a rela-
tively large exchange capacity because its particles may
consist of single unit sheets. Very fine particles of other
minerals such as mica and quartz also carry a net negative
charge in water as a result of broken bonds at the bound-
aries. However, even in the range of small particle sizes
in which nonclay minerals occur in soils, the exchange
capacity is relatively small.
Table 4.1 Cation Exchange Capacity of Principal
Clay Minerals
Cation Exchange
Mineral Capacity (meq/g)
Kaolinite 0.03-0.1
Illite 0.2-0.3
Chlorite 0.2-0.3
Attapulgite 0.2-0.35
Hydrated halloysite 0.4-0.5
Montmorillonite 0.8-1.2
The highly polar water molecule has the ability to form
strong bonds with the surface of soil particles, as well as
with the exchangeable cations that surround it. The strong
short-range adsorption forces hold one to four molecular
layers of water at the surface of the soil particles. This
water is said to be adsorbed. It is of most importance if
the particles are very small, such as films of sodium
montmorillonite 1 nm thick, and is insignificant if they
are large, such as 200-km grains of quartz sand.
If a soil particle is surrounded by water, the exchange-
able cations are not attached to it. Its negative electrical
charge tends to attract the cations, but the cations diffuse
toward the lower cation concentration away from the
particle. Therefore, the soil particle is surrounded by a
domain known as an electric double layer (van Olphen
1977, Mitchell 1976). The inner layer of the double layer
is the negative charge on the surface of the soil particle.
The outer layer is the excess of cations and deficiency
of anions with respect to the concentration in the free
water not influenced by the force field of the particle. The
cation concentration has a finite value near the surface of
the particle and decreases exponentially with distance to
the concentration of the cations in the free pore water.
The thickness of double-layer water, which is determined
by the valence of the exchangeable cations and by the
electrolyte concentration in free pore water, can exceed
50 nm. Thick double-layer water develops with exchange-
able cations of low electrovalence such as Na+, and in
free pore water of low electrolyte concentration, as in
freshwater rivers and lakes. On the other hand, exchange-
able cations of high valence such as Caf2and high electro-
lyte concentration as in the marine environment tend to
depress the thickness of the double-layer water. Thus, in
general, a soil particle is covered by a 1-nm layer of
adsorbed water, surrounded by 1 to > 50 nm of double-
layer water, enveloped in turn by free water. Double-
layer water is most significant in sodium montmorillonite
because of its very small and filmy particles.
When soil particles approach each other, as during
deposition or consolidation, a number of long-range and
short-range interparticle forces influence their geometri-
cal arrangement and interaction. Repulsion develops
when two soil particles approach each other and their
double layers come in contact. The particles will remain
dispersed unless they are pushed together by an external
force, such as the weight of overburden, equal to the
repulsive force. For example, a pressure of 500 kPa is
required to bring two face-to-face-oriented sodium mont-
morillonite particles, in 0.01 N electrolyte concentration,
to within 4 nm of each other. An increase in electrolyte
concentration depresses double-layer thickness and
allows closer approach of the particles, which aggregate
into pocs. The flocs of clay-mineral particles may settle
simultaneously, without segregation, with other fine parti-
cles and with silt-sized particles of quartz, mica, and
 ARTICLE 4 PROPERTIES OF VERY FINE SOIL FRACTIONS 11

feldspars. On the other hand, in a freshwater lacustrine

environment flocculation is absent, and segregated varved
clay and silt deposits develop.
Adsorbed water layers must be removed for soil parti-
cles to approach within less than 1 nm of each other.
Enormous external pressures are required to squeeze them
out. For example, a pressure of 125 MPa is required to
bring two parallel plates of montmorillonite or mica to
within 0.5 nm of each other, and a pressure of 400 MPa
is required to remove the last molecular layer of water.
Such magnitudes of contact pressure are likely to develop
only at mineral-to-mineral contact points of sand-sized
and larger particles, where in general the force per particle
can be large and the contact area very small. This is
illustrated in Figs. 4.6through 4.9, which show an assem-
blage of sand grains, close-ups of grain-to-grain contacts,
and a magnified smooth fracture surface, respectively. It
is apparent that the area of contact between grains is
extremely small and that even small forces between grains
would produce very high pressures. At the other extreme,
it is unlikely that geologic pressures due to overburden
on clays and shales or typical construction loads could Figure 4*7 PhotomicrographOf contact between grains of
quartz sand.
squeeze adsorbed water from between plates of montmo-
rillonite in which the force per particle is small and the
contact area, as compared with the size of the particles,
is very large. Edge-to-face contact in sodium montmoril-
lonite is also unlikely, because the highly flexible films
cannot be pushed together. In kaolinite, short-range edge-
to-face contact between particles is possible, because the
particles are relatively large. As a result of their small
breadth-to-thickness ratio, they are rather stiff and thus

Figure 4.8 Photomicrograph of contact between grains of

dune sand.

are capable of transmitting forces parallel to the plates.

The particle size and shape of illite suggest a behavior
intermediate between those of kaolinite and montmoril-
lonite, but closer to the former.
Mineral-to-mineral contact is established after the last
Figure 4.6 Photomicrograph of Lake Michigan beach sand molecular layer of water is desorbed and the very short
particles. range repulsion of approaching mineral surfaces is over-
12 INDEX PROPERTIES OF SOILS

Canada, suggest a possible cementing action by molecular

Figure 4.9 Magnification of smooth fracture surface of
quartz.
come. This type of contact is required for developing
primary valence bonding between the particle surfaces
or, alternatively, for engaging microscopic roughness of
the contact surfaces, such as protruding lattice points or
regions. It is also required for generating sliding fric-
tional resistance.
Although cementing between mineral grains is a signif-
icant factor in rocks such as sandstone and even in residual
soils at the early stages of weathering, the nature and
extent of microscopic cementing in soils is not well estab-
lished. The existence of rather stable aggregates of clay
minerals in most clay and shale deposits, and the brittle
behavior of soils such as certain soft clays from eastern
coatings of carbonates or silicates on the surface of soil
particles that are connected to each other. For example,
such an interparticle bonding may be responsible for the
rather high water content of the soft clays from eastern
Canada as compared with their relatively inactive mineral
composition. However, most soft and stiff clay deposits
as well as granular soils are uncemented.
4.5 Fabric
The term fabric describes the geometrical arrangement
of soil particles with respect to each other. During sedi-
mentation of very fine-grained soil particles in quiet fresh
water, the elementary mineral units tend to settle individu-
ally. The resulting sediment has a random geometrical
arrangement in which the mineral particles have no pro-
nounced preferred orientation. As overburden accumu-
lates, the platy or elongated particles tend to rotate into
more horizontal positions, whereupon the sediment
becomes more anisotropic but retains its random microfa-
bric. If the supply of sediment is seasonal, the coarser
sediments, which settle more quickly than the finer ones,
soon cease to accumulate during the periods of low inflow,
and only the finer particles continue to settle out. The
result is a segregated fabric, of which a varved clay is
an example. On the other hand, if the sediment accumu-
lates in salt water, flocculation occurs, and all mineral
types with elementary particles from fine clay-size to
coarse silt-size settle simultaneously to form a random
unsegregated deposit.
Most fine-grained natural soil deposits consist of a
mixture of mineral types that often include sheet silicates
as well as space lattice silicates and some carbonates.
This is illustrated by the mineralogical composition of
several soft-clay deposits listed in Table 4.2. A typical
soft-clay deposit includes, in addition to illite and chlorite
and some montmorillonite, substantial amounts of quartz

Table 4.2 Mineralogical Composition of Soft Clays

Total mineral content (%) Clay mineral content (%)

Clay Plagio- K- Montmor-

Clay Mineral Quartz clase feldspar Amphibole Calcite Dolomite Illite Chlorite illonite Other

Boston Blue 30 35 24 8 3 0 0 74 21 5 0
St. Hilaire 32 23 21 0 8 9 8 55 39 6 0
Berthierville 21 37 25 0 17 0 0 53 34 13 0
La Grande 11 29 25 7 0 10 17 70 30 0 0
Vasby 38 33 16 13 0 0 0 67 14 9 10
Pisa* 73 14 5 0 0 6 0 60 16 24+ 0

* Also contains 2% pyrite.

Mixed-layer: 17% illite/montmorillonite and 7% chlorite/montmorillonite.
 ARTICLE 4 PROPERTIES OF VERY FINE SOIL FRACTIONS 13

Figure 4.10 Fabric of undisturbed Boston blue clay. Figure 4.11 Fabric of undisturbed St. Hilaire clay.

and feldspar and some carbonates. Some of the very fine

minerals may have been eroded from shales and were
probably already in an aggregated form when they
reached the sedimentation basin.
The SEM photomicrographs of natural soft-clay depos-
its in Figs. 4.10 through 4.15 generally suggest more
or less flocculated- or aggregated-random fabrics. The
mineralogical composition of these clays is shown in
Table 4.2 and their index properties are presented in Table
4.3. Boston Blue Clay is a glacial outwash deposited
about 14,000 years ago in coastal brackish water. The
sample described in Fig. 4.10 and Tables 4.2 and 4.3 was
a firm to stiff clay preconsolidated by desiccation. St.
Hilaire clay and Berthierville clay are postglacial clays
deposited about 10,000years ago in the inland Champlain
Sea. La Grande clay is a postglacial clay deposited 6,000
to 8,000 years ago in the marine environment of the
Tyrrell Sea. Vasby clay is a recent postglacial marine clay
from the eastern central coast of Sweden. Pancone clay

Table 4.3 Index Properties of Soft Clays

W O W1 WP CF a,:
No. Clay (%) (%I (%) (-2p,m%) (kPa) up:, Degrees

1 Boston Blue 27-30 34 17 40 155 3.20-3.50 30

2 St. Hilaire 6 1-68 55 23 77 83 1.40-1.57 26
3 Berthierville 57-63 46 24 36 39 1.30-1.40 27
4 La Grande 55-58 64 26 52 83 1.75-2.00 28
5 Vasby 94- 103 121 40 67 28 I .20-1.34 23
6 Pisa-Pancone 54-65 86 35 72 124 1.57-2.02 23
7 Mexico City 421-574 500 150 27 58 1.40-1.60 45

w, = natural water content (Article 6).

w I= liquid limit (Article 7.2).
wp = plastic limit (Article 7.2).
CF = clay fraction (Article 8.2).
ai, = effective vertical overburden pressure (Article 15.3)
up = preconsolidation pressure (Article 16.4).
+ = effective-stress friction angle (Article 19.2).
14 INDEX PROPERTIES OF SOILS
Figure 4.12 Fabric of undisturbed Berthierville clay.
Figure 4.13 Fabric of undisturbed LaGrande clay.
from Pisa was deposited 7,000 years ago in a marine
environment. The salt concentration in the pore water of
these brackish or marine deposits has been substantially
reduced, through either leaching or diffusion, following
crustal rebound that elevated the clay deposits above the
sea level after withdrawal of glacial ice (Bjerrum 1954,
Kenney 1964).
Figure 4.14 Fabric of undisturbed Vasby clay.
Figure 4.15 Fabric of undisturbed Pancone clay.
The fabric of the postglacial soft clays from eastern
Canada is very open, and their natural water content is
high in comparison with their mineralogical composition
and liquid limit. A slow rate of deposition accompanied
by simultaneous development of interparticle bonds by
carbonates and possibly silicates apparently is responsible
for the special fabrics (Quigley 1980). When the salt
content of these clays was subsequently reduced by leach-

ARTICLE 4 PROPERTIES OF VERY FINE SOIL FRACTIONS
Figure 4.16 Fabric of undisturbed Mexico City clay.
ing or diffusion leading to a tendency for particle disper-
sion, the clays acquired the characteristics to lose most
of their strength and to liquefy when subjected to remold-
ing at constant water content (Bjemm 1954).
Not all soft-clay deposits have a mineralogical compo-
sition similar to those in Table 4.2. Mexico City clay is
a dramatic example of soft clay derived from volcanic
material. It consists of about 5 to 10% sand-sized concre-
tionary particles composed of calcium carbonate; 55 to
65% silt-sized siliceous diatoms; 20 to 30% clay-sized
particles ( 10% interlayered montmorillonite, with
exchangeable cations that are mostly sodium, and the
remainder biogenic or volcanogenic silica); and 5 to 10%
organic material. Thus, Mexico City clay is largely com-
posed of microfossils that mainly are siliceous skeletons
and skeletal fragments of diatoms. Apparently the large
quantities of silica released by the volcanic ash as it
underwent weathering initiated a great bloom of diatoms
in the Pleistocene lake waters of the valley of Mexico.
The undisturbed fabric of Mexico City clay is shown in
Fig. 4.16, and photomicrographs of the silt-sized fraction
are shown in Figs. 4.17 and 4.18. Figure 4.16 shows that
Mexico City clay has an open flocculated-random fabric.
The highly poriferous skeletal fragments are mainly
responsible for such unusual properties as very high plas-
tic and liquid limits, a very high friction angle, and great
loss of strength upon manipulation. Similar high values
of plastic limit and friction angle are displayed by volca-
nic ash residual soils of Japan, which contain allophane
and halloysite.
Next Page
15
Figure 4.17 Silt-size fraction of Mexico City clay showing
poriferous whole geometric forms and fragments of diatoms.
In contrast to the random fabric or structure of most
natural clays, clays that have been sheared by tectonic
activity, by sliding, or even by manipulation in the labora-
tory or by construction activities, are likely to lose their
flocculated, random structure. They are then said to have
a dispersed, highly oriented fabric. In this state they are
likely to have properties quite different from those of
clays with a flocculated or aggregated random fabric.
4.6 Organic Soils
Organic substances in soil range from macroscopic
incompletely decomposed plant and animal residues to
microscopic dark-colored humus. Humus includes prod-
ucts of advanced decomposition of organic residues, prod-
ucts of microbial resynthesis, precipitates of dissolved
organic compounds, and organic molecules in solution
(Gieseking 1975a). Organic substances are composed
mainly of carbon, oxygen, and hydrogen. However, dif-
ferent organic parent materials, various aerobic and anaer-
obic conditions of degradation, and different degrees of
humification produce organic substances with a wide
range of molecular structure and particle morphology. A
highly poriferous and flexible cellular structure is the
most important characteristic of organic coarse particles,
which are either fibrous or granular. Organic fine sub-
stances, usually smaller than 100 km, consist of irregu-
larly shaped organic skeletons such as cell fragments and
tissue parts, as well as of globular organic precipitates
smaller than 1 km, and of 3- to 9-nm organic polymolec-
ules. Organic fine substances are negatively charged and