0% found this document useful (0 votes)
60 views38 pages

11 - Pore Water Pressure, Capillary Water

Uploaded by

Xiang Yu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
60 views38 pages

11 - Pore Water Pressure, Capillary Water

Uploaded by

Xiang Yu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 38

Source: GEOTECHNICAL ENGINEERING

11 Pore Water Pressure,


Capillary Water,
and Frost Action

11.1 PUTTING IT TOGETHER

11.1.1 Of Soil and Water


The preceding chapters dealt with the nature of soil minerals and their
relationships to water; this chapter emphasizes water pressures, both positive
and negative, that either tend to push soil grains apart, reducing their contact
friction, or pull them together, increasing friction. The first is called positive pore
water pressure; the second, negative pore water pressure or suction. The reference
pressure for both is atmospheric pressure, which tends to be equalized in soils
above a groundwater table that are more or less permeable to air.

11.1.2 Positive Pore Water Pressure


Because positive pore water pressure acts to push soil grains apart and reduce
their intergranular friction, it can be a key component of shearing strength.
Positive pore water pressure plays a key role in landslides, which therefore usually
occur after heavy rains when the groundwater table is high.

Positive pore water pressure also is subject to dynamic influences from


compression, which creates a gradient that moves water out of the soil and
allows it to compress. A gradient also exists across soil zones through which
seepage occurs because of the viscous resistance to flow. These factors are
discussed in detail in later chapters that deal with settlement of foundations and
seepage rates through and under dams and levees, and into open excavations.

11.1.3 Negative Pore Water Pressure


Soil suction is what holds a child’s sand castle together, until waves or tides
produce a feeling of sad inevitability as negative pore pressure is relieved and the
208 Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 209

castle collapses. The sand must be moist in order to be able to build a sand castle
in the first place. The transitory nature of the part of soil strength that depends on
moisture content gives it the name ‘‘apparent cohesion.’’

Negative pore water pressure also sets up a pressure gradient that acts to suck in
more water, which if available is responsible for frost heave. As this is closely
related to capillarity it is discussed in this chapter, along with the depth of freezing
and permafrost.

11.2 EFFECTIVE STRESS

11.2.1 Equilibrium Pore Water Pressure


A scuba diver knows from personal experience that water pressure increases with
depth under water. The same increase exists in saturated soils, since voids are
interconnected and transmit pressure. In a static situation, the pore water pressure
is simply calculated from
u ¼ w h ð11:1Þ
where u is pore water pressure,  w is the unit weight of water, and h is the depth
below a groundwater table.

11.2.2 Equation for Effective Stress


One of the most important insights by Terzaghi was to recognize that, because
water pressure on soil grains tends to push the grains apart, it can simply be
subtracted from total stress to obtain what he defined as effective stress.
This actually is a simplification because it assumes that water pressure acts
across the entire cross-section of soil instead of only on the area represented by the
voids. However, true grain contact areas are very small and make up only a small
fraction of the entire cross-section. This assumption greatly simplifies calculations
and, as it tends to overpredict the influence from positive pore water pressure, is
on the safe side. By considering soil stress on a total area basis, one can write:

0 ¼   u ð11:2Þ
0
where  is the effective stress,  is total stress, and u is the pore water
pressure. This is the Terzaghi equation for effective stress, and even though
it is deceptively simple, it is one of the most important relationships in soil
mechanics.

Example 11.1
The unit weight of a saturated soil is 125 lb/ft3 (19.6 kN/m3 or 2.0 Mg/m3). What are the
total and effective stresses at a depth of 3.05 ft (1 m)?

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

210 Geotechnical Engineering

Answer: Total stress is  ¼ 3.05  125 ¼ 381 lb/ft2 (1  19.6 ¼ 19.6 kN/m2 or kPa). The pore
water pressure is 3.05  62.4 ¼ 190 lb/ft2 (1  9.81 kPa). Then

 0 ¼ 381  190 ¼ 191 lb=ft2 ð19:6  9:81 ¼ 9:79 kPaÞ

In the cgs system  ¼ 1  2.0 ¼ 2.0 Mg/m2, u ¼ 1  1.0, and  0 ¼ 2  1.0 ¼ 1.0 Mg/m2.

It will be noted that the effective stress for a submerged soil having this unit weight is
almost exactly one-half of the total stress, which is a useful approximation.

Example 11.2
Repeat the above calculation for effective stress using the submerged unit weight of
the soil.

Answer: As discussed in the preceding chapter, submerged unit weight at saturation


equals total unit weight minus the unit weight of water:  sub ¼ 125  62.4 ¼ 62.6 lb/ft3
(19.6  9.81 ¼ 9.79 kPa). Then

 0 ¼ 3:05  62:6 ¼ 191 lb=ft2 ð1  9:79 ¼ 9:79 kPaÞ

The two answers are the same, and spreadsheet calculations sometimes are simplified by
using submerged unit weight for the part of the soil that is below a groundwater table to
calculate effective stress.

11.2.3 Does Pore Water Pressure Affect Shearing Stress?


Pore pressure acts normal or perpendicular to grain surfaces and therefore
subtracts from compressive stress. How does pore water pressure influence
shearing stress? Under static conditions water has no shearing resistance and is
not influenced by pressure within the water. It is only by exerting pressure, which
tends to separate the soil grains, that pore water pressure affects the shearing
strength of soil.

11.2.4 Effective Stress at the Bottom of a Lake or Ocean


Let us assume that soil is submerged under 1000 m of water; does eq. (11.2)
still hold true? It might intuitively be assumed that the weight of the water would
push the soil grains together. On the other hand it can be argued that the
high water pressure will push them apart. Actually, both occur and balance
out. That is, pressure from the water column adds both to the total stress, , and
adds equally to the pore water pressure, u, in eq. (11.2), so effective stress,  0 ,
remains the same. This explanation for low densities of sea bottom soils was
first recognized by a geologist, Lyell, and later quantified by Terzaghi in the
effective stress equation.

The same observation can be made by applying Archimedes’ principle: as a


buoyant force equals the weight of water displaced, it is not affected by the weight
of water underneath or on top of a submerged object.

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 211

11.2.5 Excess Pore Water Pressure and Shear Strength


Pushing soil grains together, as occurs under a foundation load, reduces the pore
space and causes a corresponding increase in the pore water pressure in a
saturated soil, until the water can escape and the system returns to equilibrium.
The temporarily increased pore pressure affects intergranular contact pressures
and friction the same as does static pore water pressure, and eq. (11.2) still applies.

The response of pore water pressure to external loading is a critical relationship in


geotechnical engineering. It explains why rapid loading of saturated clay soil, as
under the wheel of a car, can result in the car being stuck in the mud, whereas if
the same load is applied after the soil dries out there is no problem.

The rate of loading is particularly important for large structures such as earth
dams, so the pore water pressure is monitored using observation wells or piezo-
meters. If the pore water pressure is too high for the structure to be safe,
construction is stopped until there has been sufficient time for part of the excess
pore water pressure to drain out. The heavy stone castles of Europe survived
because they were built over time spans of hundreds of years instead of being
plunked down in a matter of months.

Settlement occurs as excess pore pressure declines and more load is transferred
from the water to the soil grains, causing the soil to compress. This time-related
compression is called consolidation. Consolidation of soil takes time, which
depends in part on the permeability of the soil. The time function was analyzed
by Karl Terzaghi and is presented in a later chapter.

Structures built on saturated sand are of less concern because sand has a
sufficiently high conductivity for water that it can drain out as a load is applied.

Excess pore water pressure also develops during formation of a sedimentary


deposit as more load is slowly applied at the surface. Very high pore water
pressures have been measured in deep oil wells, still left over from sedimentation
and enhanced by expansion of the pore water as a result of the increase in
temperature with depth, called a ‘‘geothermal gradient.’’

11.2.6 Is an Artesian Condition the Same as Excess Pore


Water Pressure?
This can be argued, because there is no question that an artesian condition
represents an excess of water pressure that can be relieved by drainage. However,
the cause is not compression of the soil voids, but water moving downward from
an elevated position through a confined porous soil layer such as a sand or gravel,
much as if the water were moving in a pipe. The source normally is a hill where the
groundwater table naturally is higher. An artesian condition therefore does
not qualify as excess pore pressure, because the pressure is in equilibrium with
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

212 Geotechnical Engineering

remotely connected groundwater. Artesian pressure is analogous to pressure in


a water faucet that is remotely connected to a water tower.

11.3 NEGATIVE PORE WATER PRESSURE

11.3.1 Capillary Rise


A simple demonstration of negative pore water pressure is by analogy to water
retained by capillary forces in a small clean glass tube, as shown in Fig. 11.1.
Generally, the smaller the diameter of the tube, the higher the water will rise.
A similar situation exists with soils; the finer the soil and the smaller the void
diameter, the higher the capillary rise. As atmospheric pressure at the phreatic
surface is the base for measurement, pressure in the capillary is negative and
decreases with height above that surface, as shown in the figure.

11.3.2 Surface Tension


Capillary rise is attributed to surface tension of the water, which is the result of
unbalanced attractive forces at the interface between water and air. This is
illustrated in Fig. 11.2, where the low attractive force between water and air
molecules results in higher dipolar attractions in the water surface.

Surface tension exists at any interface between solid, liquid, and gas, but its effects
are most obvious at liquid-gas interfaces, where surface tension causes the surface
of a liquid to behave as if it were covered with a tightly stretched membrane. The
surface sometimes is considered a separate phase called a ‘‘contractile skin.’’
A membrane analogy is an appropriate and useful concept in helping to explain

Figure 11.1
Capillary rise and
negative pore
pressure.

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 213

Figure 11.2
Unbalanced
molecular
attractions in a
surface.

surface-tension phenomena, such as support for water insects that skate across
and depress the water surface.

11.3.3 Surface Tension and Capillary Rise


In a capillary tube, surface tension forces attach to the glass and hold the water
up, but the weight of the water causes the surface to sag downward in the middle,
somewhat like a blanket that is being held at the edges. This creates a visible
meniscus. The height of capillary rise usually is measured to the bottom of the
meniscus, even though that is not strictly correct. The walls of the capillary must
be clean and free of any hydrophobic material such as oil that would prevent the
adsorption of water molecules to the glass.

If a fluid does not bond to glass, the  angle in Fig. 11.1 exceeds 908 and
capillarity pushes downward so that the meniscus is reversed. A reversed meniscus
may be observed at the top of a mercury column in a thermometer.

11.3.4 Oil on Troubled Waters


Oil is hydrophobic, or water-hating, and being less dense than water spreads over
the surface in a thin film that diffracts light into rainbow-like colors. An oil film
therefore creates two surface-tension interfaces, air-to-oil and oil-to-water. The
net increase in total surface tension acts to smooth wave action and give meaning
to the expression to ‘‘spread oil on troubled waters.’’ Plutarch’s explanation was
that oil makes water more slippery so that wind slips on by, but he had no
knowledge of surface tension. However, this misconception lives on in abortive
experiments to develop an ‘‘airplane grease.’’

11.3.5 Surface Tension and Surface Energy


Surface tension is a force per unit length, F/L. Multiplying the numerator and
denominator by L gives (FL) divided by (F/L2), which is work or energy per unit
area. Surface tension therefore is synonymous with surface energy, the energy
required to generate a surface. Reducing surface tension therefore reduces surface
energy.
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

214 Geotechnical Engineering

11.3.6 Soaping Surface Tension


The manipulation of surface tension is at the heart of the soap and detergent
industries, which use organic compounds whose purpose is to lower the surface
tension between the unclean and that which is to be cleaned. Although the
amount of energy required to make new surfaces is reduced, energy still is
required, whether by use of a stirring mechanism in a washer or pounding clothes
on a rock.

Organic molecules in detergents are somewhat hydrophobic or water-hating so


they concentrate in the water surface. There they interrupt and weaken water-to-
water molecular attractions, making it easier to generate suds or foam. Only a tiny
amount of organic chemical can produce this effect, so foam is an early indication
of organic pollutants in a lake or river.

11.3.7 Measuring Surface Tension


A simple device that illustrates measurement of surface tension is shown in
Fig. 11.3, where the pulling force is 2LT, where L is the length of the expanding
surface and T is surface tension in units of force per unit length. The force is 2
because two surfaces are involved.

Another way to measure surface tension is from the height of rise in a capillary
tube, as in Fig. 11.1, if the inside diameter of the tube is known. The height of rise
is obtained by equating the surface tension at the circumference with the weight of
the water column in the tube:

2rT cos  ¼ r2 hg ð11:3Þ

where r is the radius of a circular capillary tube in units that are consistent with
units of T,
T is the surface tension of the liquid, which can be in millinewtons per
meter,
 is the contact angle between the meniscus surface and the wall of the tube,
h is the height of rise of the liquid in meters,
 is the density of the liquid, in grams per cubic meter, and
g is the acceleration of gravity (9.81 m/s2).

Figure 11.3
Measuring surface
tension.

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 215

Solution of eq. (11.3) for h yields


2T
h¼ cos  ð11:4Þ
rg
With perfect wetting,  ¼ 08 so cos  ¼ 1.0. Dimensions in the above equation will
be seen to be consistent since by definition 1 N ¼ 1 kg-m/s2.

Example 11.3
Predict the height of rise of water with surface tension 72 mN/m, in a clean capillary tube
0.1 mm in diameter.

Answer: From eq. (11.4),


2 x 72 mN=m
h¼ cos 0
0:5ð10Þ m  ð10Þ6 g=m3  9:81 m=s2
4

¼ 0:294 m ¼ 294 mm
In solving an equation such as this it is essential that units be included to avoid decimal
error.

According to eq. (11.2) the height of capillary rise is inverse to capillary diameter,
so as an approximation
0:03
de ¼ ð11:5Þ
h
where de is defined as a capillary equivalent pore diameter in mm, and h is in
meters.

Example 11.4
What is the capillary equivalent pore diameter in a soil where the height of capillary
saturation above the groundwater table is 2 m?

Answer: Soil pores obviously are very irregular and do not have a uniform diameter.
According to eq. (11.5) the equivalent diameter is de ¼ 0.03/2 ¼ 0.015 mm or 15 mm, which is
silt-size.

11.3.8 The Vadose Zone


Because the height of the zone of capillary saturation is somewhat erratic, being
dictated by the maximum pore size, this zone also is called the closed capillary
fringe. Additional capillary water is held above that zone in an open capillary
fringe that is held by smaller pores and can extend to the ground surface. Open
and closed capillary fringes are illustrated in Fig. 11.4. The unsaturated zone of
soil including the open capillary fringe is called the vadose zone (from the Latin
vadum for shallow water, as in ‘‘wade’’).

Negative pore pressure in the closed capillary fringe is readily calculated from the
distance above the groundwater table, but is less easily found in the open zone

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

216 Geotechnical Engineering

Figure 11.4
Zones of capillary
water and the
vadose zone.

because part of the capillary cross-section is occupied by air. A correction


therefore must be made to the effective stress equation to account for this effect,
as discussed later in this chapter.

Example 11.5
Soil at a depth of 3.05 ft (1 m) is in a zone of capillary saturation that extends 2.02 ft
(0.67 m) above the groundwater table. The unit weight of the soil is 125 lb/ft3 (19.6 kN/m3).
(a) What are the total and effective stresses at this depth? (b) Is the soil stronger or weaker
than that at the level of the groundwater table?

Answer: Total stress is 3.05  125 ¼ 381 lb/ft2 (1  19.6 ¼ 19.6 kPa). Pore water pressure is
2.02  62.4 ¼ 126 lb/ft2 (0.67  9.81 ¼ 6.6 kPa).

The effective stress is  0 ¼   u ¼ 381  (126) ¼ 507 lb/ft2 (19.6 {6.6} ¼ 10.0 kPa).
Because of the higher normal stress and intergranular friction the soil is stronger in the
capillary zone.

11.4 POSITIVE AND NEGATIVE PRESSURES FROM SURFACE TENSION

11.4.1 Pressure Inside a Soap Bubble


Calculating the pressure inside a soap bubble may seem a bit whimsical, but even
whimsy can have purpose. Furthermore the concept and calculation are so
simple, and the result is sure to enliven a flaccid conversation. Does the pressure
required to make a bubble increase, decrease, or stay the same as the bubble grows
larger?

From Fig. 11.5 the internal pressure over a diametric area equals 2 times the
surface tension around the circumference because both an inner and an outer
surface are involved:

r2 p¼ 2T  2r ð11:6Þ

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 217

Figure 11.5
Force balance
inside a soap
bubble.

where r is the radius of the bubble, p is the pressure inside the bubble, and T is
surface tension at a film-air interface. Hence
T
p¼ ð11:7Þ
r
The internal pressure therefore decreases as the bubble grows larger. Soap lowers
the surface tension and surface energy.

11.4.2 Capillary Rise and Surface Tension


The influence of surface tension on the height of rise in a cylindrical capillary can
be solved by assuming that the half-bubble shown in Fig. 11.5 is rotated 908 to
represent the meniscus in the tube. In this case only one surface is involved and u
can represent negative pressure in water immediately under the capillary surface:
2T
u¼ ð11:8Þ
r
This may be equated to the weight of the column of suspended water:
2T
u¼ ¼ r2 h
h
from which
2T
h¼ ð11:9Þ
r2 
where h is the height of capillary rise, T is surface tension, r is the radius of the
capillary, and  is the density of the fluid. The smaller the capillary, the higher the
height of capillary rise.

11.4.3 Negative Pressure in Capillary Water Linking


Two Soil Particles
Capillary water in unsaturated soil can take a doughnut or annular shape around
particle contacts, as illustrated in Fig. 11.6. This problem is somewhat more
complicated because the water-air interface has two radii, r1 and r2, one of which
is convex and the other concave. The concave radius acts to reduce pressure in the
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

218 Geotechnical Engineering

Figure 11.6
Annular ring of water linking two soil particles. The main attractive force is not from surface tension
but from the negative pore pressure created by the surface tension.

water, whereas the convex radius tends to increase it. As the concave radius is
smaller, the net effect is a decrease in pressure that pulls the grains together.

The equation for negative pressure in this case uses the Laplace function to define
the surface:
 
1 1
u ¼ T þ ð11:10Þ
r1 r2

in which r1 and r2 are radii of curvature of a warped surface where it intersects two
orthogonal principal planes. The expression in the parentheses is called the total
curvature of the surface. A negative radius creates a positive pore pressure, so in
Fig.11.5, r1, the internal radius of the ring of water, increases pressure in the water
and therefore is negative.

The negative pore pressure in capillary water therefore equals the surface tension
of the water times the total curvature of the meniscus or air-water interface.
This relationship is important because, as the moisture content changes, both
radii of curvature also change.

The influence of total curvature is illustrated in Fig. 11.7, where it will be seen
that, if the moisture content increases such that the outer surface of the annular
ring of connecting water becomes straight-sided (Fig. 11.7(a)), r1 is infinite so its
reciprocal makes no negative contribution to pore water pressure, which therefore
is positive.

When a soil is dried (Fig. 11.7(b)), the smaller radius is positive so pore pressure
becomes negative. Opposing this tendency is the cross-sectional area of the ring
of water as the soil dries out (Fig. 11.7(c)), which eventually will result in a net
decrease in attraction between the two particles. Finally, r1 can become so small
that the negative pressure equals the vapor pressure of water, causing the soil to
spontaneously dry out. This does not occur in clay because its adsorptive
properties permit a high negative pressure. When sand dries it crumbles whereas
a clay keeps getting stronger.
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 219

Figure 11.7
Influence of drying
on capillary forces
in soils: more
suction but over
a smaller area.

This relationship was studied in the early 1920s by Haines and Fisher, two soil
scientists at the U.S. Bureau of Soils. A relationship for the negative pressure
inside the water is:
Tð3r1  2aÞ
u¼ ð11:11Þ
r21
where r1 is as shown in Fig. 11.6 and a is the radius of the spheres. If a is unity,
r1 ¼ 0.67 for u ¼ 0, which is between (a) and (b) in Fig. 11.7. The compressive force
pulling the particles together is the sum of the surface tension and negative
pressure times the cross-sectional area of the water ring, which can be shown to be
(Kirkham and Powers, 1972):
f ¼ Tð2a  r1 Þ ð11:12Þ
The total force per unit area is the sum of all such forces between particles. If N is
the number of particles per unit area, with cubic packing, N ¼ 1/4a2. Then

F ¼ N  f ¼ f=4a2
 
1 r1
F ¼ T  ð11:13Þ
2a 4a2

This formula indicates that the smaller the particle size, the larger is the cohesive
force per unit area resulting from capillary tension. The maximum force will be
when r1 is very small, that is, as the soil becomes drier so long as the moisture
films remain intact.

11.5 SOIL MECHANICS OF UNSATURATED SOILS

11.5.1 A Shift in Emphasis


Much of the emphasis in geotechnical engineering has been on saturated soils, as
this normally is the most critical condition when strength is lowest and unit weight
is at a maximum, or is reduced to a minimum by submergence. However, many
soils are not and may never become saturated. An important example is soil that
has been mechanically compacted for support of a road or foundation, because
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

220 Geotechnical Engineering

compaction is halted when most but not all of the air has been squeezed out.
A compacted soil therefore is not a saturated soil, as some air remains trapped
even if the soil is submerged.

Unsaturated conditions dominate in soils such as dune sand and coarse loessial
silt that are so permeable that they do not support a groundwater table. Surficial
zones in most soils are not saturated, and an unsaturated zone can extend very
deep in deserts and semiarid areas where there is little rainfall to replenish the
groundwater supply.

As previously mentioned, a correction is required to the Terzaghi effective stress


(eq. (11.2)) in unsaturated soils because pore water does not occupy the complete
cross-sectional area of the soil pores. A simplified version of a correction
originally proposed by Bishop is as follows (Fredlund and Rahardjo, 1993):
 0 ¼   u ð11:14Þ
where  is a number between zero and one, and the other variables are as defined
in eq. (11.2).

When a soil is saturated,  ¼ 1, and eq. (11.14) reduces to eq. (11.2). When a soil is
completely dry,  ¼ 0 and total stress equals effective stress, as obviously there is
no influence from pore water pressure if there is no pore water. However, the
relationship to percent saturation is not linear, and has been found to depend on
soil type and even on the direction of the applied stress according to the void
orientation and particulate packing arrangement.

The increase in cohesive strength on drying was shown by a classic experiment by


a co-worker of Atterberg, whose contributions are presented in the next chapter.
In 1914 in Germany, Johansson dried soil briquettes and measured the force
necessary to split them with a wedge. His results are shown in Fig. 11.8, where
strength steadily increases down to a moisture content of 14 percent at point A,
where there is a break in the curve. At that point the soil suddenly became lighter
in color, indicating the entrance of air into the pores and supporting the
introduction of a  factor as in eq. (11.14).

The complete Bishop equation includes pore air pressure, which under most field
conditions is negligible. However, this inclusion is necessary for laboratory testing,
where, in order to apply an external pressure to soil specimens, they are sealed in a
rubber membrane such that included air cannot escape and also is compressed.

11.5.2 Pressure, Head, and Potential


In fluid mechanics pressure is one component of total head, which is the sum of
pressure head, elevation head, and velocity head. Because fluid flow in soils is
relatively slow, velocity head is negligible.

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 221

Figure 11.8
Change in
penetration
resistance of soil
from the presence
of air, from data of
Johansson.

Pressure is a driving force analogous to voltage in an electrical system, and


therefore is a potential. A capillary driving force was defined in the early 1900s
as a positive quantity called ‘‘capillary potential,’’ but as capillary pressures
are negative relative to atmospheric pressure, the preferred term now is matric
potential.

Osmotic potential refers to the attractive forces from salts and by clays, which can
be many negative atmospheres.

Total suction is the sum of matric and osmotic potentials.

Matric and osmotic potential are used in the algebraic sense, that is, the potential
increases when it becomes less negative. Moisture flow is toward drier soil,
other factors being the same, because the potential is low, that is, more
negative. In a wet soil the matric potential is high, and in a saturated soil or a
soil below the water table the matric potential is zero, this being the highest
possible value.

In thermodynamic terms, potential represents the free energy of soil water, or the
amount of work required to remove a unit mass of water.

11.5.3 Units of Potential


Potential may be expressed in any units that are appropriate for indicating
pressure, such as kilopascals, kilonewtons per square meter, bars, or pounds
per square inch. Also, it may be convenient in certain cases to express the

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

222 Geotechnical Engineering

potential in terms of the height of a column of liquid such as meters or feet of


water. Conversions are:
1 bar ¼ 100 kPa ¼ 100 kN=m2 ¼ 14:5 lb=in:2 ¼ 1:02 kg=cm2
¼ 750 torr ¼ 750 mm Hg ¼ 29:5 in:Hg ¼ 10:2 m H2 O ¼ 33:5 ft H 2 O
¼ 10 dyne=cm2 ¼ 0:987 atmosphere
Another term, pF, is no longer widely used, and is the logarithm of the height
of capillary water in centimeters. The p is analogous to pH and F signifies free
energy.

Example 11.6
The zone of capillary saturation in a soil under equilibrium conditions extends 1 m above
the water table. What is the indicated matric potential in meters of water?

Answer: 1 m.

11.5.4 Movement of Capillary Water


When capillary potential is in balance with gravitational potential, the soil
moisture is in a state of static equilibrium so there is no upward or downward
flow. Ordinarily, the matric potential is more negative in dry soils than in wet
ones. However, this is not necessarily true for soils that are texturally different, as
moisture may flow from a relatively dry sandy soil into a wetter clay if that is
dictated by a potential gradient. In other words, the criterion for unsaturated
moisture flow from one soil region to another is the relative potential of the two
regions, and not their respective moisture contents.

In addition, moisture in a field soil seldom reaches a state of equilibrium because


of the relatively slow rate at which capillary water moves in the soil compared
with changing weather conditions. When a capillary tube is inserted into a vessel
of water, the liquid rises in the tube almost immediately, but in the soil,
considerable time is required for a similar movement to take place. Consequently,
when rain falls on the surface, that portion of the precipitation that soaks into
the soil causes a wave of wetness that migrates slowly downward. Before an
equilibrium distribution of moisture is established, however, evaporation may dry
soil near the surface so that the matric potential is reduced and moisture starts to
move back up. Alternating rains and periods of evaporation, plus the effects of
transpiration and variations in barometric pressure and soil temperature, keep
capillary water in motion.

11.5.5 Moisture Contents Stabilized by an Impervious Cover


By preventing normal evaporation and transpiration, a building or pavement is
conducive to establishing equilibrium with the water table below, and as a result,
capillary moisture tends to accumulate in a pavement subgrade after the pavement
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 223

is laid. Measurements indicate that such accumulation occurs slowly and that 3 to
5 years may elapse before the maximum accumulation is reached. The pavement
or building also serves as an insulator that reduces the range and rate of change
of temperature in the underlying soil. Even subgrades under pavements in arid
regions have been known to become very wet, and the soil may lose a substan-
tial part of its bearing capacity by accumulation of capillary moisture from a
water table below. These factors point to the relevance of capillary characteristics
of soils under buildings and pavements.

11.5.6 Capillary Siphoning


A siphon normally is thought of as a tube filled with fluid and connecting a higher
with a lower fluid level so that the weight of the fluid in the lower part of the
siphon causes it to flow out and draw more fluid in at the top. If a host of tiny
capillaries is substituted for the tube, they also can act as a siphon, but the action
is relatively slow.

Capillary siphoning can cause an unexpected lowering of the water level behind an
earth dam if there is a continuous soil layer running over the top of the impervious
core. The core material therefore is carried to the ground surface even though it is
substantially higher than the highest water level behind the dam.

11.5.7 Effect of Temperature


The surface tension of water is inversely related to temperature, so cooling
increases the attraction of soil for capillary water. Thus, cooling during fall and
winter tends to draw more moisture into soil subgrades under pavements, whereas
summer heating reduces the soil affinity for water. Emergency water supplies in
deserts may be collected by excavating a shallow hole and covering it with a
plastic film so that water moving upward toward the cooler night temperatures
condenses on the underside of the film. The film is depressed by a weight in the
middle to make a drip point into a cup placed in the hole.

11.5.8 Ground Freezing


A sudden and drastic influence from temperature occurs when water freezes,
as the surface tension (surface energy) of ice is very high and the vapor pres-
sure very low. Freezing temperatures therefore attract both capillary water and
water vapor that migrate to a freezing zone to form ice crystals and lenses.
These generally are oriented horizontally and lift the overlying soil due to
frost heave. Water is replenished by upward movement of water from the
groundwater table.

The amount of frost heave also depends on the capillary conductivity and
matric potential of the soil, which in turn relate to its gradation. This relationship
is discussed in the next chapter, which is on the engineering classification
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

224 Geotechnical Engineering

of soils. The most troubling soils for frost heave are those that combine per-
meability and capillary potential, which are silts containing small amounts
of clay.

11.5.9 Effect of Dissolved Salts and Chemicals


An increase in the amount of dissolved salts in the soil water slightly increases
its surface tension and thereby lowers the matric potential of the soil, but the
effect is relatively minor. On the other hand a chemical additive that changes
the wetting angle, such as a bituminous compound, can halt capillarity.
In road embankments this also can be achieved by incorporating a layer of
impermeable plastic, but that has the disadvantage of also preventing infiltration
of rainwater.

11.5.10 Effect of Clay Mineralogy


Because of their high affinity for water and the associated diffuse double layer of
ions and water molecules, active clay minerals such as smectite can profoundly
influence matric potential. As the water content is decreased by drying, adsorptive
forces increase, and the last water adsorbed is held by an osmotic potential of
thousands of bars.

11.6 SORPTION CURVES

11.6.1 A Simple Model that Illustrates Capillarity in Soil


A simple and direct way to determine the equilibrium water content in soil in the
capillary fringe or vadose zone is to fill a tube with soil and suspend the bottom in
water. After a sufficient amount of time has elapsed the soil can be extruded,
cut into sections, and the moisture content of each section measured and, if
appropriate, its strength determined.

A typical result is shown in Fig. 11.9. The height above the free water surface
represents the matric potential. The curves for till and loess are based on more
precise measurements of matric potential, discussed below. As can be seen, soil
layers tend to reach equilibrium moisture contents that are characteristic of each
layer, and this system can model soil layers under a pavement. These experiments
also demonstrate the difference in capillary conductivity of the two soils, since in
the experiment shown in the figure it took approximately three times as long for
the wetting front to rise in the glacial till as in the loess. Note also that the loess
has a substantially higher capacity for capillary water, even though it contains
less clay than the till.

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 225

Figure 11.9
Moisture contents
sucked up by
different soil layers
in a tube with the
bottom in water to
simulate a
groundwater table.
Dashed lines are
from adsorption
measurements for
the two soils.

11.6.2 Hysteresis
If a sorption experiment such as that outlined above is performed from the
wet side down instead of the dry side up, different curves are obtained because
of the lag effect or hysteresis. A simple model for capillary hysteresis is
called the ‘‘ink-bottle effect’’ and is illustrated in Fig. 11.10. In the left part
of the figure, water rising into a capillary stops where the capillary is
enlarged at the ‘‘bottle.’’ However, if the bottle already is filled, as shown
at the right, capillary attraction allows it to retain water. The ink-bottle there-
fore is reluctant to take in water, but one that is full also is reluctant to
give it up.

The ink-bottle analogy also applies to expansive clay mineral particles as edges of
particles pinch together during drying so that the interior tends to retain water,
and then act to slow down re-entry of water during re-wetting.

11.6.3 Adsorption and Desorption Curves


Capillary hysteresis also becomes apparent by comparing the sorption curve
obtained during draining, called a desorption curve, with that obtained during
wetting, called an adsorption curve. Examples are shown in Fig. 11.11, where it
also will be seen that upon repeating the experiment the second cycle does not
exactly track the first. This is reasonable since the first adsorption-desorption

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

226 Geotechnical Engineering

Figure 11.10
Ink-bottle
hypothesis to
explain capillary
hysteresis.

Figure 11.11
Adsorption and
desorption curves
for a loessial soil
in the Tama soil
series, after
Richards (1941).

cycle can be expected to leave some of the ‘‘ink-bottles’’ full of water so that the
second cycle tends to retain higher moisture contents. Another factor may be
modification of the voids themselves, particularly as this soil contains expansive
clay mineral.

As a practical matter, water that is drawn up into soil underneath a pavement as


the result of a temperature gradient or freezing will tend to be retained according
to the desorption instead of the adsorption part of the cycle, resulting in higher
retained moisture contents.

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 227

11.7 MEASURING MATRIC POTENTIAL

11.7.1 Filter Paper Method


A simple procedure that is used in soil science and demonstrates the princi-
ples might be called the ‘‘limp potato chip’’ method, except that instead of
potato chips it uses filter paper. A standardized filter paper is selected and must
be calibrated by wetting and then re-weighed after equilibrating at different values
of suction. These are provided by glass desiccators containing selected salt
solutions.

A test is performed by placing three layers of oven-dried filter paper in contact


with about 200 g of soil in a container that is sealed and allowed to equilibrate.
The outer layers of paper protect the middle layer from becoming soiled, and it is
the middle layer that is used in the determination. The filter paper is removed
and weighed, then oven-dried and re-weighed. Then by reference to the filter
paper calibration curve the matric potential is obtained at that particular soil
moisture content. The soil also is weighed, oven-dried, and re-weighed to obtain
its moisture content.

If the filter paper contacts the soil, the result represents combined matric suction
and osmotic suction, the latter related to the dissolved salts and clay mineral
surface activity.

11.7.2 Principle of a Tensiometer


A more rapid method for measuring matric potential of a soil either in
the laboratory or in the field is with a tensiometer. The essential features are
shown in Fig. 11.12. A porous ceramic cup is sealed to a glass or plastic tube
that is connected to a vacuum gauge. At the top of the tube is a cap with an
O-ring seal.

The tensiometer must first be prepared by soaking the ceramic cup in water that
has been recently de-aired by boiling. Then a vacuum is pulled at the top to draw
water up into the tube and de-air the gauge. The top is sealed off and the system
tested by allowing water to evaporate from the surface of the ceramic, which
should create a negative pressure that can be read on the gauge. The cup then is
re-immersed in de-aired water and the gauge pressure should return to zero in one
or two minutes. If it does not, the de-airing procedure is repeated and the gauge
may have to be re-zeroed.

The saturated tensiometer cup then is imbedded into close contact with the soil
that is to be measured. Unsaturated soil sucks water through the porous wall
of the cup to create a negative pressure reading on the gauge. The end point,

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

228 Geotechnical Engineering

or the point of zero potential, will always be at the saturation moisture content
of the soil. Several instruments based on this principle are described by Fredlund
and Rahardjo (1993).

A limitation of a tensiometer is that tension cannot exceed 1 atmosphere because


water vapor bubbles will form in the porous cup so that the instrument is no
longer operative. Nevertheless, this device offers many possibilities for studying
soil-moisture changes within its range of applicability.

11.7.3 Zero-Point Translation Devices


In order to avoid cavitation of water in a tensiometer, the entire system can be put
under pressure. The most common method for doing this is to place saturated soil
specimens on a porous ceramic plate that is saturated with water and has
sufficient capillary retention that it does not allow entry of air (Fig. 11.13). Then
air pressure is applied to push water out of the soil and into the plate. Equilibrium
will be reached at a particular differential pressure that represents the matric

Figure 11.12
Schematic
diagram of a soil
tensiometer to
measure capillary
suction or matric
potential.

Figure 11.13
Pressure-plate
apparatus for
measuring
negative pore
water
pressures41
atmosphere.

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 229

potential. The system then is depressurized and the moisture content of the soil
determined. By performing the test with different values of differential pressure,
one obtains a desorption curve that gives moisture-tension relationships in the
approximate range 1 to 15 bars.

A modification of this system by Hilf of the U.S. Bureau of Reclamation uses a


probe with a ceramic tip that is inserted into a soil specimen that is under external
air pressure. In order to minimize water transfer, a null-type measuring system is
used and connected to a vacuum gauge. As water is sucked out of the ceramic tip
the gauge becomes negative, so to prevent cavitation the external air pressure is
increased to a null point where water does not move into or out of the soil
specimen. That pressure is the matric potential.

11.8 FROST HEAVE

11.8.1 Where to Look


There probably is no more dramatic illustration of the effects of movement of
capillary water than frost heave, where the ground surface may spontaneously rise
many inches or tens of millimeters in a single winter season. Frost heave normally
is limited to northern climates where freezing occurs, but also can be a serious
problem under cold-storage lockers, where installations often become unusable in
a matter of one or two decades.

It once was assumed that the frost damage was caused by expansion of water
upon freezing, but this 9 percent volume change is not sufficient to explain heaves
measured in fractions of a meter. A close examination of the soil involved in frost
heave will reveal a series of layers of pure ice separating soil layers, as shown in
Fig. 11.14. The cumulative thickness of the ice layers represents the total amount
of frost heave, which can lift and crack building floors and foundations and
pavements, and prevent doors from opening.

11.8.2 Spring Breakup


Frost heave is only part of the problem because when the ice layers and lenses
melt, the soil becomes overly saturated with water and can turn into mud. In
unpaved roads such areas are called ‘‘frost boils.’’ They also occur under paved
roads where they become evident when a truck breaks through. It is for this
reason that a road may be embargoed for heavy loads, until the ground has
thawed and the excess water has drained out.

Thawing from the ground surface downward aggravates an already serious


problem by preventing excess water from draining downward. Thawing that is
hastened by rain can speed drying by allowing drainage downward.

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

230 Geotechnical Engineering

Figure 11.14
Two cores
showing ice
lenses that lifted
a seven-story
building in
Canada. (From
Penner and
Crawford, 1983,
National Research
Council of
Canada.)

11.8.3 Three Requirements for Frost Heave


Obviously frost heave requires cold temperatures. A second requirement is the
availability of unfrozen water, usually from a shallow groundwater table. A third
requirement is for the soil to be sufficiently permeable to transmit a damaging
amount of water during a single winter season.

Frost heave results from a potential gradient that draws water into the freezing
zone. If the freezing front remains stationary, water will continue to be drawn up
and freeze to make a layer of relatively pure ice. If the weather turns colder so that
freezing penetrates downward, it encapsulates the ice layer and prevents further
growth. Slow freezing is conducive to frost heave.

Soil and water conditions required for frost heave are as follows:
 The soil must have sufficient capillary tension to draw water into the freezing
zone, which means that sands do not exhibit very much frost heave.
 The soil must be sufficiently permeable that a significant amount of water can
be transferred during a winter season, which means that dense clays do not
exhibit very much frost heave.
 There must be an available supply of water that normally will come from a
groundwater table.
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 231

11.8.4 Control of Frost Heave


Frost heave can be reduced or prevented by interfering with any one of the
required conditions. For example, freezing of foundation soils is prevented by
placing the foundation below the maximum depth of annual freezing or by
keeping a building heated during winter. Pavement can be laid on top of an
insulating layer such as a few inches of StyrofoamÕ .

Because so many different factors contribute to frost heave it is not always


predictable, and a road that suffers damage one year may not be damaged the
next. The reason may be lowering of the groundwater table, or perhaps rapid
freezing that does not give sufficient time for moisture to move to a rapidly
progressing freezing front. Snow is an insulator that in theory can either
contribute to frost heave by slowing freezing, or prevent frost heave by preventing
ground freezing, depending on weather conditions.

11.8.5 Diary of a Frost Heave


As the frost line, or line separating frozen from unfrozen soil, penetrates
downward, water in the soil pores freezes, which from the standpoint of matric
potential dries the soil out. Liquid water therefore moves to the freezing front.
Some water may also move as vapor, which creates a loose array of ice crystals
inside voids called ‘‘frazzle ice,’’ but this does not make a major contribution to
heave.

As water reaching the frost line becomes supercooled, freezing is triggered by


ice crystals already in the soil so that the crystals grow. The tenacity with
which the molecules are held causes a rejection of foreign matter, so the ice is
relatively clear. Heave-producing ice layers therefore grow along the frost line as
long as they are fed from below by capillary water. The frost line and ice layers
normally are parallel to the ground surface but can be inclined or vertical behind
a wall.

As the weather becomes colder and freezing proceeds faster than capillary
conductivity can supply water to the freezing front, freezing penetrates deeper and
encapsulates the ice layer in frozen soil. If freezing slows or pauses, conditions are
reinstated for production of a new ice layer. This cycle can continue and build
layer upon layer of ice, as shown in Fig. 11.14. The total frost heave therefore is
the sum of all ice layer thicknesses (Fig. 11.15).

With the onset of warmer weather, thawing for the most part takes place
from the ground surface down, creating temporary boggy conditions until
thawing progresses all of the way through the frozen soil layer so that excess
water from thawing of the ice layers can drain downward. In areas of frost
heave, roads often are protected during seasonal thawing by embargoing heavy
truck traffic.
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

232 Geotechnical Engineering

Figure 11.15
Frost heave
causing muddy
conditions when
the ice melts.

11.9 DEPTH OF FREEZING

11.9.1 Locating Foundations below the Depth of Freezing


The most obvious way to prevent damage from frost heave is to found structures
below the maximum depth of freezing. A map showing maximum depths of
annual freezing that may be expected in the U.S. is shown in Fig. 11.16.

Freezing depths are determined not only by air temperatures but also by thermal
properties of soils and their cover. For example, a saturated soil will not freeze as
deeply under a given set of conditions as an unsaturated soil because of the heat
transfer required to freeze water. Also, soil under a thick layer of vegetation or
under a snow blanket will not freeze as deeply as if the soil were exposed to the
elements.

11.9.2 Weather Records


An evaluation of freezing conditions can be obtained from weather records.
Maximum and minimum daily temperatures are averaged and subtracted from the
nominal freezing temperature of water to obtain degree-days below freezing. The
same general procedure is used for furnace and air conditioner design, only the
reference temperature is the comfort zone for people and/or equipment. For
example, if the temperature range for a particular day is from 58C to 158C,
the average is 108C, or 10 degree-days below freezing.

11.9.3 The Stefan Equation


Freezing and thawing depths during a single winter season can be used to estimate
the extent of freeze-thaw cycling, which is an important factor affecting
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 233

Figure 11.16
Maximum frost
depths for design
purposes.

deterioration of pavement materials. The Stefan equation is a simplification as it is


based on the heat capacity of water in the soil and therefore overpredicts freezing
depths in dry soils. Also, it does not take into account additional water drawn into
the soil by frost heave, which can have a substantial effect on the rate of thawing.
The equation is
rffiffiffiffiffiffiffi
Lt F
Z ¼ Ct ð11:15Þ
L
where Z ¼ depth of freezing, in meters (ft).
Ct ¼ a constant. In SI, Ct ¼ 13.1; in the English system, 6.93.
kt ¼ thermal conductivity coefficient of frozen soil, W/m  K (watts per meter-
Kelvin) or Btu/ft2-hr-8F per ft.
F ¼ degree days below 08C or 328F.
L ¼ latent heat of fusion of water in soil, kJ/m3 (kilojoules per cubic meter) or
Btu/ft3.
L for water is 333.7 kJ/or 142.4 Btu/lb. L for soil water therefore is

L ¼ 340 wd or
ð11:16Þ
L ¼ 1:43 wd

where w is the percent moisture and  d is the dry unit weight of the soil, in kN/m3
or lb/ft3, respectively.
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

234 Geotechnical Engineering

Table 11.1 kt ku
Thermal conductivity Dry density (frozen soil) (unfrozen soil)
Moisture
(from charts of 3 3 content, %
kN/m lb/ft W/mK Btu/ft2-hr/8 F/ft W/mK Btu/ft2-hr/8F/ft
Kersten, 1952)
Sandy soils:
20.4 130 5 2.3 1.3 2.6 1.5
10 4.2 2.4 3.1 1.8
18.9 120 5 1.7 1.0 2.0 1.2
15 4.0 2.3 2.8 1.6
Soils with 450% silt and clay:
18.9 120 10 1.7 1.0 1.6 0.9
17.3 110 10 1.3 0.8 1.3 0.7
18 2.1 1.2 1.7 1.0
15.7 100 10 1.0 0.6 1.0 0.6
20 1.8 1.0 1.4 0.8
14.1 90 10 0.8 0.5 0.8 0.5
30 2.1 1.2 1.3 0.7
Selected thermial conversion factors:
1 Btu ¼ 1.055 kJ
2
1 Btu/ft -hr-8F/ft ¼ 1.730 W/mK (watts per meter-kelvin)
1 cal/cmsK ¼ 418.4 W/mK
1 Btu/ft2 ¼ 37.26 kJ/m3 (kilojoules per cubic meter)
1 cal/cm3 ¼ 4187 kJ/m3
1 Btu/lb ¼ 2.326 kJ/kg ¼ 22.81 kJ/kN
1 cal/g ¼ 4.1868 kJ/kg
1 W (watt) ¼ 1 J/s
1J ¼1 Nm

Some representative values for kt are given in Table 11.1. Calculations for depth
of thawing use ku, which is lower than kt because the heat conductivity of liquid
water is less than that of ice.

Example 11.7
The average temperature for a 10-day period is 58C. Calculate the depth of freezing in a
sandy soil with  d ¼ 18.9 kN/m3 (120 lb/ft3) and a moisture content of 12%.
Answer: Interpolating in Table l1.1 gives kt ¼ 3.3 W/mK. L ¼ 340  12  18.9 ¼ 77,000 kJ/
m3. From eq. (11.15),
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
13:1 3:3  10  5
Z¼ ¼ 0:6 m ð24 in:Þ
77,000

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 235

(b) How many degree-days above 08C will be required for complete thawing, assuming no
change in moisture content and no direct heat from the sun?

Answer: From Table 11.1, ku ¼ 2.6 W/mK, so


sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
13:1 2:6  F
0:6 ¼ , or F ¼ 62 degree-days C:
77,000
Because of the difference between kt and ku, in this case it will require 62 degree-
days of thawing to compensate for 50 degree-days of freezing. As previously
mentioned, the calculations do not take into account additional water drawn into
the soil by frost heave, which mainly will affect the rate of thawing.

An application of the Stefan equation is illustrated in Fig. 11.17 and shows several
periods when thawed soil was on top of frozen soil that would prevent downward
drainage. Also, by simply counting the thawing and refreezing periods at different
depths it will be found that the number of freeze-thaw cycles is a maximum at the
ground surface and decreases with depth, which is an important consideration for
pavement design in freezing climates.

Question: In Fig. 11.17, during what months might it be expected that most of the
frost heave would occur?
Answer: November and December. Why?

Figure 11.17
Freezing depths
calculated using
the Stefan
equation indicate
8 cycles in the
road base course
during one winter
season. (After
K. P. George.)

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

236 Geotechnical Engineering

11.9.4 Insulating Against Freezing


The most obvious control measure against frost heave is to prevent freezing.
Foam plastic, usually high-density polystyrene called ‘‘Geofoam,’’ has a thermal
conductivity of about 0.026 W/Km, which is about one-hundredth that of soil.
However, after absorbing water the thermal effectiveness is estimated to be
reduced to about 30 to 40 times that of soil. In the 1980s a 51 mm (2 in.) thick
layer was successfully used to prevent frost heave on I-70 in Colorado where the
freezing depth was 1.8 m (6 ft).

The Stefan equation is not applicable to predict foam thickness because L for the
foam depends on a variable and unknown content of water. However, as a rough
guide the required thickness can be taken as 1/30 to 1/40 times the normal depth
of freezing. In the example cited above the thickness would have been 45 to 60 mm
(1.8 to 2.4 in.), which closely agrees with the case history.

Adding salt or calcium chloride to soil also can reduce or prevent frost heave by
lowering the freezing point, but the effect may last only one or two seasons until
the salts leach out and become pollutants.

11.9.5 Drainage
The most common methods used to control or reduce frost heave in roadbeds is to
raise the grade by scooping out ditches, and provide a drainable subgrade soil.
As a general guide, the distance from the bottom of a pavement to the highest
probable elevation of the water table should not be less than 2 m (6 ft), and a
larger distance is desirable if it can be obtained at a reasonable cost. Increasing the
distance to a water table decreases the potential gradient and the rate of upward
movement of capillary water. Another advantage of a high-crowned road grade
is that snow blows off.

11.9.6 Granular Subbase


Another approach is to replace frost-susceptible soil with a coarse-grained soil
that has a low capillary attraction for water. Ideally the subbase should extend to
the maximum freezing depth, but as a practical matter this usually is not feasible.
Therefore a granular subbase often is combined with drainage by ditches and/or
tile lines.

11.9.7 Cutoff Blanket


A third approach is to replace upper subgrade soil with a nonexpansive clay
having low permeability. This has the disadvantage that clay tends to be weak.
An alternative is the use of an impermeable geomembrane. In either case it is
important to also control and minimize infiltration of surface water by providing
surface drainage.
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 237

11.10 PERMAFROST

11.10.1 Overview
Permanently frozen ground, termed ‘‘permafrost,’’ underlies nearly one-fifth
of the total land area of the world. Permafrost is most widespread in the
northern hemisphere in the treeless high Arctic, referred to as tundra, but also
can extend south of the tree line and as far south as the 50th parallel. The
southern boundary is irregular, and patches of permafrost occur sporadically
beyond the boundary. As may be expected, extensive permafrost occurs in
Antarctica.

Permafrost occurs where the mean ground surface temperature is below 08C
(328F), which in turn is influenced not only by the air temperature but also by
ground cover, topography, and local climatic and soil conditions. The depth to
which permafrost penetrates is limited by the geothermal gradient, as shown in
Fig. 11.18. The thickness of permafrost therefore varies from a few meters to
several hundred meters, in general being thicker along the coast of the Arctic
Ocean, and diminishing in thickness farther south.

Figure 11.18
Permafrost and
the geothermal
gradient. The
active zone
undergoes annual
changes in
temperature, and
the permafrost
thickness is limited
by the geothermal
gradient.

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

238 Geotechnical Engineering

11.10.2 Permafrost Active Zone


Seasonal changes in temperature from winter to summer cause thermal energy
waves that slowly penetrate and are damped out with depth as the heat flux from
each wave moves upward as well as downward. The result is a damped sinusoidal
curve for ground temperatures at any given time, shown by the curved line
between d and e in Fig. 11.18. These lines present an envelope of minimum and
maximum ground temperatures, and extend downward into the permanently
frozen zone.

The line for the maximum temperature defines the maximum depth of thawing,
which is called the active zone. Ice lenses generally are abundant below the active
zone and cause major ground subsidence if conditions change so that melting
proceeds downward. According to the theory for frost heave the lenses may be
‘‘paleo-ice’’ formed during initial freezing.

11.10.3 Permafrost Landforms


Thermal cycling in frozen soil below the active zone causes contraction tensile
cracks that fill with water trickling down from the thawed active zone. The water
immediately freezes, creating permanently frozen vertical wedges of ice. The
wedges then grow incrementally during subsequent thermal cyling, pushing the
ground up between adjacent wedges. The cracks often follow a hexagonal pattern
similar to a honeycomb, and create a characteristic feature of permafrost areas
called polygonal ground, shown in Fig. 11.19. If soil immediately next to the
wedges is pushed up, the result is ‘‘low-centered’’ polygons, or if the entire soil
mass between wedges is pushed up, the result is ‘‘high-centered’’ polygons.

Figure 11.19
Summer on the
tundra. Water is
trapped on the
surface of Arctic
tundra by
permafrost.
Vertical ice
wedges have
pushed up soil to
form low-centered
polygons that
pond water and
can cause partial
melting of the
permafrost to form
large lakes.

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 239

Polygons may follow linear soil boundaries such as abandoned lakes or river
channels, in which case they are rectangular.

11.10.4 Drainage in Permafrost Areas


Permanently frozen ground does not allow infiltration of surface water, so
permafrost areas are poorly drained and marshy during melt seasons, even though
the annual precipitation is that of a desert. During thaw seasons the tops of ice
wedges melt down, and low-centered polygons fill with meltwater. Small streams
tend to follow ice wedges, and where wedges meet they melt out pockets that may
be meters deep and appear like beads—hence the name ‘‘beaded streams.’’

11.10.5 Artesian Pressure from Confined Drainage


Water seeping downslope through the active zone becomes confined when the top
of the soil freezes during the onset of winter. This can lead to some extraordinary
consequences. A pingo is a large mound, of the order of 30–45 m (100–150 ft) in
diameter and 8–9 m (25–30 ft) high, where a frozen soil layer has been pushed up
by hydrostatic pressure and contains a core of ice. A palsa is similar to a pingo,
but consists mainly of peat soil and occurs most frequently near the southern
fringe of the discontinuous permafrost.

An abandoned building can delay ground freezing so confined downslope seepage


finds an outlet inside the building, which then fills with water so that ice cascades
out of the windows.

11.10.6 Thawing of Permafrost


Permafrost is sensitive to changes in the thermal regime and, as shown in
Fig. 11.20, will thaw underneath a heated building unless steps are taken to
prevent it. Destruction of ground cover such as by clearing of trees promotes
thawing, and tilting of the trees as ice wedges melt away creates an effect known
as ‘‘drunken forest.’’ Several years are required after clearing for ground
temperatures to warm up sufficiently for agriculture.

Lakes dot the high Arctic and change the soil thermal regime sufficiently to thaw
the upper part of the permafrost. This causes subsidence as ice lenses and wedges
melt, so lakes are perpetuated. Large areas of the Arctic coastal plain of North
America therefore present a patchwork of oblong thaw lakes that are oriented in
the direction of summer prevailing winds. As the lakes migrate and find outlets
they drain and the cycle starts over again.

Irregular pock-marked topography resulting from thawing of permafrost and


associated ice masses is called ‘‘thermokarst,’’ analogous with limestone sink
areas.
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

240 Geotechnical Engineering

Figure 11.20
Settlement from
thawing of
permafrost by a
heated building,
Big Delta, Alaska.

11.10.7 Premafrost and Engineering


The most direct approach to prevent problems from melting of permafrost is to
keep it frozen, but this is becoming more difficult because of the sensitivity of
northern climates to change, with the result that these areas are the first to feel
major impacts from global warming. The moisture content at which a soil
becomes liquid is called the ‘‘liquid limit,’’ and when permafrost melts, the soil
moisture content usually exceeds the liquid limit and is trapped until it can drain
away or evaporate.

Seasonal thawing can be prevented by supporting structures on piles extending


into the permafrost layer and allowing winter air to circulate underneath the
structures in the winter. Another method uses ducts in soil under a structure in
order to force-circulate cold winter air.

Oil pipelines must be protected from loss of support by thawing permafrost in


order to prevent environmentally disastrous oil spills. The scheme shown in
Fig. 11.21 was used for the Trans-Alaska pipeline, and employs a passive heat
transfer mechanism with a vaporizing refrigerant. During winter when the soil is
warmer than the air, refrigerant in the soil zone evaporates, taking heat with it,
then rises and condenses in the colder ends of the tubes, releasing heat to the air.
The refrigerant then flows down by gravity to replenish the supply at the bottom.
During summer the cycle is broken because the refrigerant remains at the bottom
of the tubes. A similar, less efficient, but less costly system uses convection, as air
inside a pipe is warmed by the soil and moves up and out. The air is replenished by
cold winter air from the ground surface and conducted downward through a
center pipe.

Roads can pose a difficult problem in permafrost areas because, if they are built
up for better drainage, permafrost can rise in the road embankment and create a
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 241

Figure 11.21
Schematic diagram
showing method
for preservation of
permafrost under
the Trans-Alaska
pipeline.

barrier to water seeping through the active zone, leading to ponding on the uphill
side of the road. Culverts should be large enough to allow flow of air and thawing
during summer months.

Tundra of the high Arctic is marshy and supports a variety of mosses, lichens,
flowers, and shrubs that are part of the eco-chain, are easily damaged, and are
very slow to recover. Most land travel therefore is during winter when the tundra
is frozen.

11.10.8 Cold-Storage Plants


Cold-storage facilities simulate Arctic conditions and gradually create a perma-
frost bulb and frost heave. An air space or auxiliary subfloor heating system
separated from the cold room by insulation can avoid such eventualities.

Shutting down a frost-heaved cold-storage facility can create another set of


problems as thawing of ice lenses leads to soft ‘‘frost boil’’ soil conditions under
the floor and footings. It is not unusual to find older facilities for sale.

11.10.9 Insulation to Prevent Thawing or Freezing


If a mean temperature is above or below freezing, insulation alone will not prevent
ground thawing or freezing, respectively, but will only delay it. Insulation can be
effective during a single cycle, for example to prevent frost heave during a single
winter season, but will not solve a permafrost problem if the overall direction of
heat flow is not changed. A method that is successful for relatively small buildings
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

242 Geotechnical Engineering

is to place the structure on a raised gravel pad that has a relatively high thermal
conductivity so that it cools from the outside. This and other arrangements are
discussed by Phukan (1991).

11.11 SUMMARY

Positive pore water pressure weakens soil by reducing friction between soil grains.
Pore pressure therefore is subtracted from total stress to obtain effective stress.

Capillary action draws water up into a fine-grained soil, creating a capillary


fringe. Soil in the capillary fringe may be saturated, so saturation is not a criterion
for locating a groundwater table. A groundwater table is measured in an open
boring where there is no capillary effect. Pore water pressures in a capillary fringe
are negative relative to atmospheric pressure.

Under static conditions, pore water pressure increases linearly with depth below a
groundwater table, whereas soil weight-induced stresses increase with depth below
the ground surface.

Compressing saturated soil increases its pore water pressure. The increase over
static groundwater conditions is excess pore water pressure, which like static pore
water pressure acts to decrease effective stress and friction.

Capillary action is a lead actor in frost heave.

Permafrost conditions require special considerations and will become more


prominent as mineral resources are developed in Arctic areas.

Problems
11.1. Why is the term ‘‘matric potential’’ now preferred over ‘‘capillary
potential’’?
11.2. A soil unit weight above the groundwater table averages 115 lb/ft3 and
below the groundwater table averages 112 lb/ft3. The water table is 6 ft deep
below the ground surface. Prepare a plot of total and effective stresses to
a depth of 20 ft.
11.3. Distinguish between artesian water pressure and excess pore water
pressure.
11.4. (a) Calculate the height of rise of pure water in a clean glass tube whose
inside diameter is 0.15 mm, using T ¼ 7.75 mN/m. (b) What is the height of
rise if the tube is sufficiently dirty to produce a wetting angle of 158?
11.5. If the outside diameter of the tube in Problem 11.4 is 0.35 mm, what is the
compressive stress in the walls of the tube in part (a) of that problem?
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 243

What parallel can be drawn to compressive stress in the capillary zone


of a soil?
11.6. Two soils have the sorption characteristics shown in the accompanying
table. Draw the sorption curves with moisture content along the x-axis.

Moisture content (%)


Water, m Soil A Soil B

0.10 47 32
0.316 33 17
1.0 22 12
3.16 17 9
10.0 13 7
31.6 10 5

11.7. Assume that two masses of soils A and B in Problem 11.6 are at the same
elevation above a water table and are in intimate contact with each other.
If soil A contains 17% moisture and B contains 12%, will capillary water
flow from A to B or from B to A, or will it remain static?
11.8. A subgrade soil under an impervious pavement has the sorption
characteristics of soil A in Problem 11.6. What will be the accumulated
moisture content of the soil 0.3 m (1 ft) below the bottom of the pavement,
if the water table is constant at an elevation 1.5 m (5 ft) below the bottom
of the pavement?
11.9. Repeat Problem 11.8, using soil B in Problem 11.6.
11.10. What is meant by the zone of capillary saturation? Is the pressure in the
soil water within this zone greater than, less than, or equal to atmospheric
pressure?
11.11. In a certain locality the water table rises steadily from October to April and
then falls steadily from May through September. Will the zone of capillary
saturation extend a greater or lesser distance above the water table in July
than in March?
11.12. The water in a capillary tube extends 0.9 m (3 ft) above a free water surface.
Determine the pressure in the water in this capillary tube: (a) at the free
water surface; (b) 0.3 m (1 ft) above the free water surface; (c) 0.6 m (2 ft)
above it; (d) just below the meniscus; (e) just above the meniscus.
11.13. Explain to a 6-year-old why sand castles built at the water margin on a
beach crumble when the tide comes in. Do they also crumble when they dry
out?
11.14. What is the best time to drive a four-wheeler on a sandy beach, when the
tide is coming in or going out? Why?
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

244 Geotechnical Engineering

11.15. Soil immediately underneath a pavement is 6 ft above the groundwater


table. If the sorption curve for this soil is as shown in Fig. 11.11, what will
be the probable moisture content? Which is more appropriate, the
desorption or adsorption curve? Why?
11.16. At what stage of frost action in the subgrade may it be necessary to
impose a traffic embargo on a road or runway? How long should the
embargo last?
11.17. Under what conditions will rain dry out a soil?
11.18. Name three basic conditions that must exist before ice lenses will grow in a
soil subgrade.
11.19. Describe three principles that may be employed in pavement design to
inhibit the growth of heave-producing ice lenses. Explain how each method
accomplishes its purpose.
11.20. Should ice lenses in soil behind a retaining wall be oriented with their thin
dimensions vertical or horizontal? Explain.
11.21. Explain how freezing an expansive clay can cause it to shrink.
11.22. What minimum foundation depths are recommended to prevent frost
damage in the following cities: Chicago, New York, Denver, Atlanta,
Minneapolis, Los Angeles, Seattle, Phoenix?
11.23. Maximum and minimum recorded daily air temperatures over a 10-day
period are as follows. Predict the depth of freezing: (a) in a clay with a unit
weight of 14.9 kN/m3 (95 lb/ft3) and a moisture content of 20%; (b) in a
still lake.

Day 1 2 3 4 5 6 7 8 9 10
Min (8C) 0 14 9 6 20 16 16 10 8 10
Max (8C) 12 2 1 3 12 8 6 2 4 10

11.24. Define and give reasons for permafrost, active zone, oriented lake, pingo.
11.25. What dictates the bottom elevation of permafrost?
11.26. Explain the development and significance of patterned ground.
11.27. A four-story apartment building located on permafrost is sinking into the
ground. Is there any way to prevent further damage to this building?
11.28. Why, when permafrost thaws, does the ground usually settle, and why does
it do so unevenly?
11.29. Suggest two methods to maintain permafrost under a heated building.
How might these methods be modified to also apply to foundations for a
cold-storage warehouse?

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Pore Water Pressure, Capillary Water, and Frost Action

Pore Water Pressure, Capillary Water, and Frost Action 245

References and Further Reading


Andersland, O. B., and Anderson, D. M. (1978). Geotechnical Engineering for Cold Regions.
McGraw-Hill, New York.
Baver, L. D., Gardner, W. H., and Gardner, W. R. (1972). Soil Physics, 4th ed. John Wiley
& Sons, New York.
Fredlund, D. G., and Rahardjo, H. (1993). Soil Mechanics for Unsaturated Soils. John
Wiley & Sons, New York.
Jumikis, A. R. (1962). Soil Mechanics. Van Nostrand, Princeton, N.J.
Jumikis, A. R. (1966). Thermal Soil Mechanics. Rutgers University Press, New
Brunswick, N.J.
Jury, W. A., and Horton, R. (2004). Soil Physics, 6th ed. John Wiley & Sons, New York.
Kersten, M. (1952) ‘‘Thermal Properties of Soils.’’ Special Report No. 2, Highway
Research Board, 161–166.
Kirkham, D., and Powers, W. L. (1972). Advanced Soil Physics. Wiley-Interscience,
New York.
Penner, E., and Crawford, C. B. (1983). ‘‘Frost Action and Foundations.’’ National
Research Council, Canada, DBR Paper No. 1090.
Phukan, A. (1991). ‘‘Foundations in Cold Regions.’’ In H.-Y. Fang, ed., Foundation
Engineering Handbook, 2nd ed. Van Nostrand Reinhold, New York.
Richards, L. A. (1941). ‘‘Uptake and Retention of Water by Soil as Determined by
Distance to a Water Table.’’ J. Am. Soc. Agron. 33, 778–786.
Soil Science Society of America (1996). Glossary of Soil Science Terms. SSSA,
Madison, Wis.
Spangler, M. G. (1952). ‘‘Distribution of Capillary Moisture at Equilibrium in Stratified
Soil.’’ Special Report No. 2, Highway Research Board.
Van Olphen, H. (1991). An Introduction to Clay Colloid Chemistry. Krieger Publ. Co.,
Melbourne, Fla. (reprint edition).

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)


Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.

You might also like