Clay Minerals

and
Soil Structure

1
 Contents-

Clay Minerals  Soil Fabric - Natural Soil

 Identification of Clay Minerals  Soil Fabric - Clay Soils

 Specific Surface (Ss)  Soil Fabrics - Granular Soils

 Interaction of Water and Clay Minerals  Loess

 Interaction of Clay Particles  Suggested Homework

 Soil Structure and Fabric  References

2
1. Clay Minerals

3
1.1 Origin of Clay Minerals
“The contact of rocks and water produces clays, either at or near the surface of the
earth” (from Velde, 1995).

Rock +Water  Clay

For example,
The CO2 gas can dissolve in water and form carbonic acid, which will become
hydrogen ions H+ and bicarbonate ions, and make water slightly acidic.
CO2+H2O  H2CO3 H+ +HCO3-
The acidic water will react with the rock surfaces and tend to dissolve the K ion
and silica from the feldspar. Finally, the feldspar is transformed into kaolinite.

Feldspar + hydrogen ions + water  clay (kaolinite) + cations, dissolved

silica
2KAlSi3O8+2H+ +H2O  Al2Si2O5(OH)4 + 2K+ +4SiO2
4
 The alternation of feldspar into kaolinite is very
common in the decomposed granite.

 The clay minerals are common in the filling materials of

joints and faults (fault gouge, seam) in the rock mass.

Weak plane!

5
1.2 Basic Structural Units
Clay minerals are made of two distinct structural units.

Aluminium or
Magnesium

Silicon tetrahedron Aluminium Octahedron

6
Tetrahedral Sheet
Several tetrahedrons joined together form a
tetrahedral sheet.

tetrahedron

hexagonal
hole

7
Tetrahedral & Octahedral Sheets

For simplicity, silica tetrahedral sheet by:

Si or Si

and alumina octahedral sheet by:

Al

8
Basic Unit-Summary

Mitchell, 1993

9
1.3 Different Clay Minerals

Different combinations of and octahedral

sheets form different clay minerals:
Clay Mineral (e.g., kaolinite, halloysite):

10
 Minerals-Kaolinite Basal spacing is 7.2 Å

layer

Typically 70-100 layers

• Si4Al4O10(OH)8. Platy shape

17 m • The bonding between layers are van der Waals

( Mitchell, 1993) forces and hydrogen bonds (strong bonding).
• There is no interlayer swelling
• Width: 0.1~ 4m, Thickness: 0.05~2 m
11
Minerals-Halloysite

• Si4Al4O10(OH)8·4H2O 2 m (Mitchell, 1993)

• A single layer of water between unit layers.

• The basal spacing is 10.1 Å for hydrated halloysite and 7.2 Å for dehydrated halloysite.
• If the temperature is over 50 °C or the relative humidity is lower than 50%, the hydrated
halloysite will lose its interlayer water (Irfan, 1966). Note that this process is irreversible and will
affect the results of soil classifications (GSD and Atterberg limits) and compaction tests.
• There is no interlayer swelling.
• Tubular shape while it is hydrated. Å=Angstrom=10-10m
12
Kaolinite

 used in paints, paper and in pottery and pharmaceutical industries

 (OH)8Al4Si4O10

Halloysite

 kaolinite family; hydrated and tubular structure

 (OH)8Al4Si4O10.4H2O

13
Clay Mineral (e.g., montmorillonite, illite)

14
Minerals-Montmorillonite
• Si8Al4O20(OH)4·nH2O (Theoretical
unsubstituted). Film-like shape.
• There is extensive isomorphous
substitution for silicon and aluminum
by other cations, which results in
n·H2O+cations charge deficiencies of clay particles.
0.96 nm • n·H2O and cations exist between unit
layers, and the basal spacing is from
9.6 Å to  (after swelling).
• The interlayer bonding is by van der
Waals forces and by cations which
balance charge deficiencies (weak
bonding).
• There exists interlayer swelling,
which is very important to
engineering practice (expansive
clay).
• Width: 1 or 2 m, Thickness: 10
Å~1/100 width
5 m (Holtz and Kovacs, 1981) 15
 Montmorillonite

 A highly reactive (expansive) clay

swells on contact with water
 (OH)4Al4Si8O20.nH2O

high affinity to water

Bentonite
 montmorillonite family

 used as drilling mud, in slurry trench walls, stopping leaks

16
Minerals-Illite (mica-like minerals)
• Si8(Al,Mg, Fe)4~6O20(OH)4·(K,H2O)2. Flaky
shape.
• The basic structure is very similar to the mica,
so it is sometimes referred to as hydrous mica.
Illite is the chief constituent in many shales.
potassium K
• Some of the Si4+ in the tetrahedral sheet are
3+ 3+
0.96 nm replaced by the Al , and some of the Al in
the octahedral sheet are substituted by the Mg2+
or Fe3+. Those are the origins of charge
deficiencies.
• The charge deficiency is balanced by the
potassium ion between layers. Note that the
potassium atom can exactly fit into the
hexagonal hole in the tetrahedral sheet and
form a strong interlayer bonding.
• The basal spacing is fixed at 10 Å in the
presence of polar liquids (no interlayer
swelling).
7.5 m (Mitchell, 1993)
• Width: 0.1~ several m, Thickness: ~ 30 Å 17
Minerals-Vermiculite (micalike minerals)

Illite Vermiculite Mitchell, 1993

• The octahedral sheet is brucite.

• The basal spacing is from 10 Å to 14 Å.
• It contains exchangeable cations such as Ca2+ and Mg2+ and two layers
of water within interlayers.
• It can be an excellent insulation material after dehydrated. 18
1.4 Synthesis Mitchell, 1993

Noncrystall
ine clay -
allophane

19
 Minerals-Chlorite

Gibbsite The basal spacing

or is fixed at 14 Å.
brucite

20
 Chain Structure Clay Minerals
•They have lathlike or
Attapulgite
threadlike morphologies.
chain structure (no sheets);
needle-like appearance
•The particle diameters are
from 50 to 100 Å and the
length is up to 4 to 5 m.
•Attapulgite is useful as a
drilling mud in saline
environment due to its high
4.7 m stability.
(Mitchell, 1993)
21
 Mixed Layer Clays

• Different types of clay minerals have similar structures

(tetrahedral and octahedral sheets) so that
interstratification of layers of different clay minerals can
be observed.

• In general, the mixed layer clays are composed of

interstratification of expanded water-bearing layers
and non-water-bearing layers. Montmorillonite-illite
is most common, and chlorite-vermiculite and chlorite-
montmorillonite are often found.

(Mitchell, 1993)
22
Noncrystalline Clay Materials

Allophane
Allophane is X-ray amorphous (Formless) and has no
definite composition or shape. It is composed of
hollow, irregular spherical particles with diameters of
3.5 to 5.0 nm.

23
2. Identification of Clay Minerals

24
2.1 X-ray diffraction

Mitchell, 1993

• The distance of atomic planes d can be determined based on the Bragg’s

equation.
BC+CD = n, n = 2d·sin, d = n/2 sin
where n is an integer and  is the wavelength.
• Different clay minerals have various basal spacing (atomic planes). For
example, the basing spacing of kaolinite is 7.2 Å.
25
2.2 Differential Thermal Analysis (DTA)

• Differential thermal analysis

(DTA) consists of simultaneously
heating a test sample and a
thermally inert substance at
constant rate (usually about 10
ºC/min) to over 1000 ºC and
continuously measuring differences
in temperature and the inert
material T.

For example:
• Endothermic (take up heat) or
Quartz changes from the  to  form at 573 ºC exothermic (liberate heat)
and an endothermic peak can be observed. reactions can take place at different
heating temperatures. The mineral
T types can be characterized based
on those signatures shown in the
left figure.
(from Mitchell, 1993)
Temperature (100 ºC) 26
 If the phase transition of the
If the sample is
sample occurs,
thermally inert,
T

T
Crystallize

Melt

Time t Time t

Endothermic reactions take Exothermic reactions liberate

up heat from surroundings and heat to surroundings and
therefore the temperature T therefore the temperature T
decreases. increases.

T= the temperature of the sample – the temperature of the thermally inert substance.
27
2.3 Other Methods
i) Using Plasticity chart

28
 ii) Scanning Electron Microscope

 common technique to see clay particles

 qualitative

plate-like structure

29
iii) Specific Surface (Ss)
Definition:
Specific surface  surface / volume
Specific surface  surface / mass Commonly used

Surface related force Surface related forces:

van der Waals forces, capillary
Gravational force forces, etc.

Example: 111cm cube,   2.65g / cm 3

6 1cm 2 4
Ss   2.3  10  m 2
/g
1cm  2.65 g / cm
3 3 Ss is inversely
proportional to the
111m cube,   2.65g / cm 3
particle size
6 1m 2
Ss   2.3  m 2
/g
1m  2.65 g / cm
3 3

Note: smaller the grain, higher the specific surface 30

Typical Values
Montmorillonite
50-120 m2/gm (external surface)
700-840 m2/gm (including the interlayer surface)

Interlayer surface

65-100 m2/gm

10-20 m2/gm

31
iv) Cation exchange capacity (cec)
( Isomorphous Substitution)

• substitution of Si4+ and Al3+ by other lower valence (Mg2+) cations

• results in charge imbalance (net negative)

positively charged edges

+ +
+ _ _ _ _+ + _
+ _ _ _ __ negatively charged faces
+ _ _ _
_ _ _ _
_ _ _
_ _ _

Clay Particle with Net negative Charge

32
 v) Potassium determination
Well-organized 10Å illite layers contain 9% ~ 10 % K2O.

vi) Thermogravimetric analysis

It is based on changes in weight caused by loss of water or
CO2 or gain in oxygen.

Sometimes, identification of clay minerals not only based on one method.

33
4. Interaction of Water
and
Clay Minerals

34
 4.1 Origins of Charge Deficiencies

1. Imperfections in the crystal lattice -Isomorphous

substitution.
• The cations in the octahedral or tetrahedral sheet can be
replaced by different kinds of cations without change in
crystal structure (similar physical size of cations).
For example,
Al3+ in place of Si4+ (Tetrahedral sheet)
Mg2+ instead of Al3+(Octahedral sheet)
unbalanced charges (charge deficiencies)
• This is the main source of charge deficiencies for
montmorillonite.
• Only minor isomorphous substitution takes place in kaolinite.
35
2. Imperfections in the crystal lattice-The broken edge

The broken edge can be positively or negatively charged.

36
 3. Proton equilibriaum (pH-dependent charges)

M  OH  H   M  OH 2 Protonation
M  OH  OH   M  O   H 2O Deprotonation

H
M O +
H Kaolinite particles are positively charged on
M O H their edges when in a low pH environment, but
M: metal
negatively charged in a high pH (basic)
M O-
environment.

37
 4.2 “Charged” Clay Particles
• External or interlayer surfaces
- or + are negatively charged in general.

• The edges can be positively or

negatively charged.

• Different cations balance charge

deficiencies.

Cation

- or +
Kaolinite and negative gold sol Dry condition
(van Olphen, 1991)

39
4.3 Polar Water Molecules

Structure Polar molecule O(-)

H(+) H(+)
Hydrogen bond Salts in aqueous solution

hydration
40
4.5 Clay-Water Interaction

1. Hydrogen bond

Kaolinite

H
O O
H H

Clay Surfaces
Adsorbed layers Free water
3 monolayers Bulk water

Oxygen Hydroxyl
O OH
The water molecule “locked” in the adsorbed
layers has different properties compared to
that of the bulk water due to the strong
attraction from the surface.
H
O
H 41
 2. Ion hydration
Na+ crystal radius: 0.095 nm
radius of hydrated ion: 0.358 nm

cation

Clay
layers

The cations are fully hydrated,

Dry condition The water molecules which results in repulsive forces
(Interlayer) wedge into the interlayer and expanding clay layers
after adding water (hydration energy). 42
3. Osmotic pressure

A B

From Oxtoby et
al., 1994

The concentration of cations is higher in the interlayers (A) compared with that
in the solution (B) due to negatively charged surfaces. Because of this
concentration difference, water molecules tend to diffuse toward the interlayer
in an attempt to equalize concentration.
43
Relative sizes of adsorbed water layers on sodium
montmorillonite and sodium kaolinite

Holtz and Kovacs, 1981

44
5. Interaction of Clay Particles (or Layers)

Interlayer
Particle
Interparticle

Layer

45
5.1 Diffuse Double Layer

- - Cations
-

- - Exponential decay

Concentration
+ + -
x Distance x
Clay particle with
negatively charged
surface Anions

46
47
48
 5.2 Interaction Forces

Net force between clay particles (or interlayers)

DLVO = van der Waals attraction +
forces Double layer repulsion (overlapping of the double layer)+
Coulombian attraction (between the positive edge and negative face)

49
5.3 Thickness of Double Layer

1/ 2
  0   kT 
K   
2 2 
 2n 0 e  
Thickness of  0 : Permittivity in vacuum
double layer K
 : Re lative permittivity
k : Boltzman cons tan t
K repulsion force  T : Temperature
n 0 : Cation concentrat ion
e : Electron ch arg e
n0 K repulsion force   : Valence
v K repulsion force 
 decreases with increasing
T K  repulsion force (?)
temperature
50
5.4 Interaction of Clay Particles

Dispersed fabric Flocculated fabric

Edge-to-face (EF): positively
charged edges and negatively
charged surfaces (more common)
Edge-to-edge (EE)
The net interparticle force
between surfaces is repulsive

Aggregated fabric

Face-to-Face (FF) Shifted FF

51
5.5 Atterberg Limit of Clay Minerals
Na-montmorillonite
•Thicker double layer
•LL=710
Ca-montmorillonite
•Thinner double layer
•LL=510
The thickness of double
layer increases with
decreasing cation valence.

Lambe and Whitman, 1979

52
5.6 Cation Replaceability

• Different types and quantities of cations are adsorbed to balance charge

deficiencies in clay particles.
• The types of adsorbed cations depend on the depositional environment. For
example, sodium and magnesium are dominant cations in marine clays
since they are common in sea water. In general, calcium and magnesium
are the predominant cations.
• The adsorbed cations are exchangeable (replaceable). For example,

Na Na Na Na Ca Ca
+4CaCl2  +8NaCl
Na Na Na Na Ca Ca

(Lambe and Whitman, 1979)

53
 5.6 Cation Replaceability (Cont.)
The ease of cation replacement Non-hydrated Hydrated
depends on the Cations radius (Å) radius (Å)
(1) Valence (primarily) Li+ 0.68 3.8
 Higher valence cations can replace
cations of lower valence. Na+ 0.95 3.6
(2) Ion size K+ 1.33 3.3
 Cations with larger non-hydrated Cs+ 1.69 3.3
radii or smaller hydrated radii have
greater replacement power. Be2+ 0.31 4.6
According to rules (1) and (2), the Mg2+ 0.65 4.3
general order of replacement is
Li+<Na+<K+<Rb+<Cs+<Mg2+<Ca2+<Ba2+ Ca2+ 0.99 4.1
<Cu2+<Al3+<Fe3+<Th4+ Ba2+ 1.35
(3) Relative amount Al3+ 0.5 4.8
 High concentration of
Na+ can displace Al3+. Fe3+ 0.6
(Data compiled from
Israelachvili, 1991)
54
5.7 Cation Replaceability (Cont.)

•Hard water softener

Hard water contains soluble calcium and magnesium salts such as
Ca(HCO3)2 and Mg(HCO3)2. The hardness can be removed by exchanging
Ca2+ and Mg2+ with sodium ions Na+. For example,

Na2Z(s) (Zeolite) + Ca2+(aq)  CaZ(s)+2 Na+(aq)

As the ion-exchange capacity of Zeolite is saturated, the capacity can be

regained by passing through a concentrated solution of NaCl.

55
5.7 Cation Exchange Capacity (cec)

• The quantity of exchangeable cations is termed the cation

exchangeable capacity (cec) and is usually expressed as
milliequivalents (meq) per 100 gram of dry clay ( from Mitchell, 1993).

• One equivalent = 6.021023 electron charges or 96500 Coulombs, which

is 1 Faraday.

56
5.8 Swelling Potential
Practically speaking, the three ingredients generally necessary for
potentially damaging swelling to occur are (1) presence of
montmorillonite in the soil, (2) the natural water content must be around
the PL, and (3) there must be a source of water for the potentially
swelling clay (Gromko, 1974, from Holtz and Kovacs, 1981)

U.S. Bureau of
Reclamation Holtz and Kovacs, 1981

57
 5.9 Engineering Applications
Lime treatment for the swelling clay
• The swelling clay such as Na-montmorillonite beneath the foundation is
potentially harmful to the light structure. Adding lime (CaO) into such
soil can effectively reduce the swelling potential due to Ca2+ displacing
Na+, and can increase the strength by dehydration of soils and
cementation.
Drilling mud The swelling clays can
Soil particle form a so-called “filter
cake” and enable soil Pressure
layers to become profile of
relatively impermeable. slurry

Earth
Bentonite or pressure+
Polymer ground water
pressure
Trench
Montmorillonite is the dominant clay mineral in bentonite Xanthakos, 1991
58
5.9 Engineering Applications (Cont.)

Xanthakos, 1991 59
5.9 Engineering Applications (Cont.)

Xanthakos, 1991 60
5.9 Engineering Applications
Dispersion agents (drilling mud; hydrometer analysis)
• Sodium hexa-metaphosphate (NaPO3) and sodium silicate (Na2SiO3) are used as
the dispersion agent in the hydrometer analysis. How does this dispersion agent
work?
• Three hypotheses:
(1) Edge-charge reversal
The anions adsorption onto the edge of the clay particle may neutralize the positive edge-
charge or further reverse the edge-charge from positive to negative. The edge-charge
reversal can form a negative double layer on the edge surfaces to break down flocculated
structure, and assist in forming a dispersed structure.
(2) Ion exchange
The sodium cations can replace the divalent cations existing in the clay particles such as
Ca2+ and Mg2+. The decrease of cation valence can increase the thickness of the double
layer and interparticle repulsion, which can assist in forming a dispersed structure.
(3) pH
The higher pH may make the edge-charge tend to be negative, which can break down the
flocculated structure and assist in forming a dispersed structure. The adding of dispersing
agent such as sodium carbonate may slightly increase the pH.

61
6. Soil Structure and Fabric

62
6. Soil Structure and Fabric
• The structure of a soil is taken to mean both the geometric
arrangement of the particles or mineral grains as well as the
interparticle forces which may act between them.

• Soil fabric refers only to the geometric arrangement of particles (from

Holtz and Kovacs, 1981).
*Fabric and structure are used interchangeably sometimes.

• The interparticle forces (or surface forces) are relatively important

for fine-grained soils at low confinement (low state of stress).
• “Although the behavior of a coarse-grained soil can often be
related to particle size distribution, the behavior of a fined-
grained soil usually depends much more on geological history
and structure than on particle size” (from Lambe and Whitman, 1979).
63
7. Soil Fabric-Natural Soil
(fine-grained soils)

64
7.1 Microfabric Features in Natural Soils

1. Elementary particle arrangements, which consist of

single forms of particle interaction at the level of
individual clay, silt, or sand particles or interaction
between small groups of clay platelets or clothed silt
and sand particles.
2. Particle assemblages, which are units of particle
organization having definable physical boundaries and
a specific mechanical function. Particle assemblages
consist of one or more forms of elementary particle
arrangements or smaller particle assemblages.
3. Pore spaces within and between elementary particles
arrangements and particle assemblages.
(from Holtz and Kovacs, 1981)

65
 7.1 Elementary Particles Collins and McGown, 1974
Individual clay
platelet interaction

Individual silt or sand

particle interaction

Clay platelet group

interaction

Clothed silt or sand

particle interaction Particle discernible
Collins and McGown, 1974
(from Holtz and Kovacs, 1981) 66
7.2 Particle Assemblages Collins and McGown, 1974

Collins and McGown, 1974

(from Holtz and Kovacs, 1981)

67
7.3 Pore Space Types

Collins and McGown, 1974 (from Mitchell, 1993) 68

8. Soil Fabric-Clay Soils

69
8.1 Terminology
Face (F)
Edge (E)
Clay Particle

Dispersed: No face-to-face association of clay particles

Aggregated: Face-to-face association (FF) of several clay particles.

Flocculated: Edge-to-Edge (EE) or edge-to-face (EF) association

Deflocculated: No association between aggregates

(from Mitchell, 1993)
70
 8.2 Particle Associations

Dispersed and deflocculated Aggregated but deflocculated

Edge-to-face flocculated Edge-to-edge flocculated

but dispersed but dispersed

Edge-to-face flocculated Edge-to-edge flocculated Edge-to-face and edge to

and aggregated and aggregated edge flocculated and
aggregated

71
 8.3 Summary

Flocculated fabric Dispersed fabric

Edge-to-face (EF): positively
charged edges and negatively
charged surfaces (more common)
Edge-to-edge (EE)
The net interparticle force
Aggregated fabric between surfaces is repulsive

Face-to-Face (FF) Shifted

Face-to-Face (FF)

72
 8.4 Fabric of Natural Clay Soils
“The individual clay particles seem to always be aggregated or
flocculated together in submicroscopic fabric units called

domains. Domains then in turn group together to form

clusters, which are large enough to be seen with a visible light
microscope. Clusters group together to form peds and even
groups of peds. Peds can be seen without a microscope, and they
and other macrostructural features such as joints and fissures
constitute the macrofabric system” (from Holtz and Kovacs, 1981).

Domain  Cluster Ped 73

 8.4 Fabric of Natural Clay Soils
Domains and 1.Domain
clusters with 2.Cluster
micropores
3.Ped
4.Silt grain
5.Micropore
6.Macropore

(from Holtz and Kovacs, 1981) Enlargement

74
8.4 Fabric of Natural Clay Soils (cont.)
• Macrostructure, including the stratigraphy of fine-grained soil deposits, has
an important influence on soil behavior in engineering practice. Joints, fissures,
silt and sand seams, root holes, verves, and other defects often control the
engineering behavior of the entire soils mass.
• The microstructure reflects the depositional history and environment of
the deposit, its weathering history (both chemical and physical), and stress
(From Holtz and Kovacs, 1981)
history.
Clay
particle

Water

75
9. Soil Fabrics-Granular Soils

76
 9.1 Packing
Loose packing Dense packing

Honeycombed fabric
•Meta-stable structure
•Loose fabric
•Liquefaction
•Sand boil

Holtz and Kovacs, 1981 77

9.1 Packing (Cont.)-Sand Boil

Loose sand

Kramer, 1996

78
 9.1 Packing (Cont.)

“Contrary to popular belief, it is not possible to drown in quicksand,

unless you really work at it, because the density of quicksand is
much greater than that of water. Since you can almost float in water,
you should easily be able to float in quicksand “(from Holtz and Kovacs, 1981).

79
9.2 Load Transfer
Loading

The black particles carry most

of load. The remaining particles
prevent the buckling of the load-
carrying chains (From Santamarina et al., 2001).

80
 9.3 The Relative Density (Dr)
The relative density Dr is used to characterize the density
of natural granular soil.

e max  e
Dr  100%
e max  e min
 d max  d   d min
  100%
d  d max   d min

(Lambe and Whitman, 1979)

The relative density of a natural soil deposit very strongly affects its
engineering behavior. Consequently, it is important to conduct laboratory tests
on samples of the sand at the same relative density as in the field ( from Holtz and
Kovacs, 1981). (compaction)
81
 Derivation
e max  e
Dr  100%
e max  e min
 d max    d min
  d 100%
d  d max   d min

82
 9.3 The Relative Density (Dr) (Cont.)
“The relative density (or void ratio) alone is
not sufficient to characterize the engineering
properties of granular soils” (Holtz and Kovacs, 1981) .
Two soils with the same relative density (or
void ratio) may contain very different pore
sizes. That is, the pore size distribution
probably is a better parameter to correlate
with the engineering properties (Santamarina et al.,

2001) .

2 : 1
Holtz and Kovacs, 1981

83
 10.Loess
 Loess is a type of aeolian soils, and the particles are predominantly silt-
size. The soil structure is mainly stabilized by (1) the capillary force and (2)
light cementation arising from the salt and fines (e.g. clay) precipitation
around the contacts (Holtz and Kovacs, 1981; Santamarina, 2001).

Capillary force Cementation

 After loess is submerged, collapse of the soil structure occurs due to loss of
suction and cementation
Why?
Capillary force cementation The interaction
between water
and salts and
clay
84
1. Read Chapter 4 (Holtz)
2. Problem 4-1, 4-3, 4-4, 4-5, 4-6, 4-8( plus all
exercises)

85
11. References
Holtz, R.D. and Kovacs, W.D. (1981). An Introduction to Geotechnical
Engineering, Prentice Hall. (Chapter 4)
Mitchell, J.K. (1993). Fundamentals of Soil Behavior, 2nd edition, John Wiley &
Sons (Chapter 3).

86