Evaluation of temperature effects on excess pore pressure in fine-grained soils

Evaluation of temperature and freeze–thaw effects on excess pore pressure
generation of fine-grained soils
Kenan Hazirbaba a,n
, Yu Zhang b
, J. Leroy Hulsey b
a
Department of Civil Engineering, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia
b
Department of Civil and Environmental Engineering, University of Alaska Fairbanks, AK, United States
a r t i c l e i n f o
Article history:
Received 1 April 2010
Received in revised form
8 September 2010
Accepted 13 September 2010
a b s t r a c t
The November 3, 2002 Denali-Alaska earthquake (Mw¼7.9) caused significant liquefaction associated
damage to various infrastructure built on fine-grained soils. The seismic response, liquefaction potential,
and excess pore pressure generation of soils in cold regions, especially those of fine-grained nature, have not
been studied thoroughly and therefore are not well-understood. This paper presents results from an
extensive laboratory study on the characteristics of excess pore pressure generation and liquefaction
potential of fine-grained soils. Laboratory-constituted soils specimens were tested in four categories: (1)
tests on specimens subjected to no thermal conditioning or freeze–thaw cycles; (2) tests on specimens
conditioned at 24, 5, 1, 0.5, and 0.2 1C; (3) tests on specimens subjected to 1–4 freeze–thaw cycles; and
(4) tests on specimens conditioned at near-freezing temperatures of 0.5 and 0.2 1C through different
freeze–thaw paths. Strain-controlled, undrained, cyclic triaxial tests were performed at shear strain levels of
0.005–0.8%. Specimens conditioned at different temperatures were found to generate significantly different
pore pressures with cyclic loading. The excess pore pressure generation at near or slightly below freezing
was found to change dramatically. A transitional change in the dynamic soil behavior, attributed to
unfrozen- or frozen-dominant pore water, was discovered. The threshold shear strain was also found to be
influenced by the temperature. Subjecting the soil specimens to 1, 2 and 4 freeze–thaw cycles caused a
reduction in excess pore pressure generation and slight change in the threshold shear strain. The
temperature conditioning path to reach the target temperature was found to be important on the
development of excess pore pressure at near and slightly below-freezing temperatures.
2010 Elsevier Ltd. All rights reserved.
1. Introduction
Saturated fine-grained, non-plastic to low-plasticity soil deposits
are commonly encountered in the seismically active Arctic region.
Moderate to strong shaking of such deposits can lead to significant
excess pore water pressure generation and ultimately to liquefaction
if the pore water is completely unfrozen. If the pore water is fully
frozen, no pore pressure is generated. As the ground temperature
nears the freezing point (i.e., 0 1C), the state of the pore water
becomes partially frozen. The response of the ground to earthquake
loading in this case is completely different from either the fully
frozen or the unfrozen case. The stiffness of the ground typically
increases with decreasing temperature [34]; however this increase
in stiffness may not translate into decreasing liquefaction potential.
This is because formation of ice layers within the ground can lead to
a decrease in permeability. Together with earthquake loading, this
may become a significant contributing factor to liquefaction of the
ground at near-freezing temperature. During the November 3, 2002,
Denali earthquake (Mw¼7.9), significant liquefaction damage was
observed at near-freezing ground temperatures [23,37]. In addition
to liquefaction damage, as much as 75 cm horizontal and 45 cm
vertical deformations were measured during the Denali earthquake
(AK DOT Report [1]). The main objectives of the present research
were: (i) to characterize the excess pore water pressure generation,
which is responsible for the occurrence of liquefaction, of fine-
grained soils of cold regions and investigate the effect of ground
temperature on the generation of excess pore pressure, and (ii) to
investigate the effect of freeze–thaw cycles (i.e., seasonal variation of
ground temperature) on excess pore pressure generation. Through a
systematic experimental program, which is outlined in detail in [40],
responses of laboratory-reconstituted soil specimens were studied.
The soil specimens were subjected to strain-controlled, undrained,
cyclic triaxial testing, and the excess pore water pressure generation
was measured directly at different levels of strain and for various
temperature conditions.
2. Review of previous studies
The liquefaction potential of fine-grained soils has been
extensively investigated by a number of researchers (Chang
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/soildyn
Soil Dynamics and Earthquake Engineering
0267-7261/$ - see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.soildyn.2010.09.006
n
Corresponding author.
E-mail address: k.hazirbaba@gmail.com (K. Hazirbaba).
Soil Dynamics and Earthquake Engineering 31 (2011) 372–384

et al. [13]; Singh [31]; Koester [24]; Polito and Martin [29];
Thevanayagam et al. [32]; Wijewickreme et al. [36]; Bray and Sancio
[11]; Boulanger and Idriss [10]). Most previous investigations on
cyclic resistance and liquefaction potential of fine-grained soils were
conducted for unfrozen ground conditions. Research efforts
concentrated specifically on how ground temperature or the
freezing–thawing process affects liquefaction potential are very
limited. Relevant previous research can be grouped under two main
categories: (1) studies that concentrated on how ground tempera-
ture and freeze–thaw affect physical and mechanical properties
of soil, and (2) studies conducted on cyclic resistance of partially
frozen or thawed soils. Each of these categories is reviewed in the
following section.
2.1. Effect of ground temperature and freeze–thaw process on
physical and mechanical properties of soil
Soil microfabric and physical and mechanical properties can be
affected by soil temperature [4]. It has been reported that the
freezing and thawing process substantially affects hydraulic
conductivity of compacted silts and clays [12,25,9,41,28].
Chamberlain and Gow [12] found that the hydraulic conductivity
increased with the increasing number of freeze–thaw cycles. They
attributed this increase of hydraulic conductivity to the formation of
ice lenses. When ice melts in frozen soil, the resulting cavities act as
channels to increase the permeability of the soil. Othman and
Benson [28] showed that hydraulic conductivity could increase by
one or two orders of magnitude after just one freeze–thaw cycle, but
there was no further increase in hydraulic conductivity after 3–5
freeze–thaw cycles. Greater changes in permeability also occurred
with faster rates of freezing.
Change of void ratio due to freeze–thaw processes was also
reported [25,33,30]. Viklander [33] and Qi et al. [30] reported a
critical void ratio for fine-grained soil above which some of the
engineering properties, such as pre-consolidation pressure, cohesion,
and density, remain unaffected with freeze–thaw cycles.
Eigenbrod et al. [16] studied static pore water pressure
development in a low-plasticity fine-grained soil due to freezing
and thawing. They found that the pore water pressure varied
cyclically during freezing. Formation of ice lenses during freezing
increased pore water pressure, and this decreased the effective
stress toward zero so as to speed up the formation of ice lenses. As
more and more ice lenses formed, further increase in the pore
water pressure promoted suction toward the growing ice lenses
adjacent to the unfrozen soil portion, and this appeared to cause
the pore water pressure to decrease. Then, decreasing pore water
pressure caused an increase of effective stress and thus,
compressed the soil. The compression of soil inversely increased
pore water pressure once more. Such cyclic variation of pore
water pressure during freezing continued until steady state was
reached. A sudden drop in pore water pressure was reported as
thawing began, and this was associated with a sudden drop in
specific volume as ice transformed to water. As the soil continued
to thaw, there was an increase in permeability and fissures,
which dissipated the thawed water. This caused an increase in
pore water pressure. From the above, it is clear that changes in
microstructure and physical properties during freeze–thaw
processes eventually cause changes in mechanical properties.
Changes in mechanical properties are strongly affected by
drainage conditions during freezing and thawing [2,3]. Alkire [2]
studied the freeze–thaw processes on silt by conducting a series
of triaxial tests for different drainage conditions during freezing
and thawing. His results showed that the drained frozen and the
undrained thawed condition produced the lowest post-thaw
shear strength, the highest water content, and the largest strain
at failure for a post-thaw state. Alkire and Morrison [3] conducted
repeated loading tests on a fine-grained soil subjected to a freeze–
thaw cycle and found softer stress–strain curves and lower
strength in comparison with samples not subjected to a freeze–
thaw cycle.
2.2. Cyclic resistance of partially frozen and thawed soils
The pioneering research studies on cyclic resistance and
liquefaction potential of partially frozen ground were conducted
after the 1964 Alaska earthquake (Mw¼9.2) mainly by Finn et al.
[17] and Vinson [34]. Using field studies and numerical analysis,
Finn et al. [17] and Finn and Yong [18] showed that the potential
for liquefaction of a saturated cohesionless soil layer when
interstratified by a frozen surficial layer and a perennially frozen
layer increases. This occurs because the frozen layers seal the
boundary of unfrozen saturated layers and prevent dissipation of the
pore pressure that is caused by seismic events. However, the studies
of Finn and Yong [18] and Finn et al. [17] did not specifically
consider the influence of the ground temperature or near-freezing
temperature, on liquefaction potential. Vinson [34] and Vinson et al.
[35] investigated the shear modulus and damping behavior of frozen
soils considering various parameters and found that variation of
temperature caused a drastic change in the dynamic properties.
Effects of freezing and thawing process on cyclic resistance
were studied indirectly by some researchers. This was done while
attempting to obtain high-quality undisturbed samples (by
freezing the ground) for liquefaction tests of clean sand and
sand-dominant soils. Goto [19], Yoshimi et al. [38] and Yoshimi
and Goto [39] investigated the effect of the uni-directional quick
freeze–thaw cycle on liquefaction resistance of clean sand and
sand with various fines ranging from 0% to 20%. They concluded
that a quick uni-directional freeze–thaw cycle did not affect the
liquefaction resistance of clean sand; however, the quick-freezing
method disturbed clean sand with fines, and a correction factor
was applied to describe the liquefaction resistance. Although
various aspects of the dynamic behavior of frozen and partially
frozen soils have been investigated to some extent, no research
has been reported on the effect of temperature and freeze–thaw
cycles on excess pore pressure generation. The need for this
research has arisen in the wake of November 3, 2002, Denali
earthquake. An extensive laboratory investigation to characterize
the excess pore water pressure generation of fine-grained soils,
which were found to cause significant damage during the Denali
earthquake, was conducted. The experimental program and
findings are discussed next.
3. Experimental program and procedures
Strain-controlled, cyclic triaxial, undrained tests were con-
ducted on a fine-grained soil to investigate the influence of
temperature and freeze–thaw on the excess pore water pressure
generation. Strain-controlled testing was adopted for this study
because the main mechanism for the occurrence of liquefaction
under seismic loading conditions is the generation of excess pore
water pressure, and the generation of excess pore pressure is
more directly related to the induced shear strains than stresses
[20]. Additionally, strain-controlled tests provide more realistic
pore pressure measurements than stress-controlled tests, and
generation of excess pore water pressure in strain-controlled tests
is less sensitive to factors such as relative density, soil fabric, and
previous loading [14]. Thus, a more fundamental approach to
study the seismic response and liquefaction potential of soils
would be to examine the pore water pressure generation
mechanism directly through strain-controlled cyclic tests.
K. Hazirbaba et al. / Soil Dynamics and Earthquake Engineering 31 (2011) 372–384 373

Four groups of tests were carried out on laboratory-recon-
stituted soil specimens: (1) the first group tests were performed
on specimens subjected to no thermal conditioning or freeze–
thaw cycles. These tests represent temperate ground conditions
and were run at room temperature. They also established a
baseline for comparison purposes, and the specimens in this
group are referred to as unconditioned specimens; (2) the second
group tests were performed on specimens conditioned at 24, 5, 1,
0.5, and 0.2 1C. This series focused on the effect of ground
temperature, especially at near-freezing state, on excess pore
pressure generation; (3) the third group of tests was conducted on
specimens subjected to various numbers of freeze–thaw cycles to
investigate how freeze–thaw cycles affect excess pore pressure
generation; and (4) the fourth group of tests was conducted on
specimens conditioned at near-freezing temperatures of 0.5 and
0.2 1C through different freeze–thaw paths. More details of the
experimental program can be found in [40].
3.1. Material tested
The soil used in this investigation was obtained from a site
near Mabel Creek Bridge and Slana River Bridge on the Tok Cutoff
Highway in Alaska, as shown in Fig. 1. An excavator was used to
dig down approximately 3 m deep in the bank of Mabel Creek.
Disturbed soil samples were collected for the purpose of
reconstituting soil specimens in the laboratory. The reason why
this specific site was selected for soil sampling is that extensive
liquefaction and associated damages were observed at and around
this location during the November 3, 2002, Denali Earthquake.
The maximum horizontal and vertical deformations in the area
were measured to be about 75 and 45 cm, respectively (AK DOT
Report [1]).
After the earthquake, Alaska Department of Transportation and
Public Facilities (AK DOTPF) initiated a repair program for the
damaged transportation infrastructure around the site. Through
this repair program a fairly extensive site investigation was
conducted. Compiled Standard Penetration Test (SPT) data from
four boreholes in the vicinity of the site are presented in Fig. 2.
The layer of concern, which consists mainly of silt, in these
profiles, is located below the top layers of fill at about a 6–12 m
depth below the ground surface. The SPT-values (N60) were
relatively low, ranging from 4 to 12. Due to the relatively shallow
water table, which is located at approximately 2.95–4.05 m below
the ground surface, the layer of concern was fully saturated.
A series of conventional geotechnical tests were conducted to
characterize the silt. Index properties including grain size
distribution (ASTM D422 [5]), specific gravity (ASTM D854 [8]),
and Atterberg limits (ASTM D4318 [6]) were determined. The
results from these tests are summarized in Table 1. The silt may
be classified as ML type soil according to Unified Soil Classification
System (USCS). Approximately 99% of this soil was measured to
pass through No. 200 sieve (74 mm). The mean grain size D50 was
found to be 0.007 mm. The coefficient of uniformity, Cu, was
measured as 7.50 and the coefficient of curvature, Cc, was found to
be 1.48. The grain size distribution is presented in Fig. 3.
3.2. UAF cyclic triaxial apparatus
MTS type triaxial equipment was employed to carry out the
cyclic loading testing program for this study. A schematic of the
testing system used is shown in Fig. 4. The testing equipment was
modified to closely simulate the in-situ freezing process of soils.
The top cap and bottom platen of the triaxial setup were designed
Fig. 1. Sampling location near Mabel Creek Bridge on the Tok Cutoff Highway,
Alaska.
Fig. 2. SPT-data from the vicinity of Mabel Creek Bridge.
(data compiled from reports obtained from AK DOT-public domain data)
Table 1
Soil index properties of the silt investigated.
USCS classification symbol ML
Specific gravity, Gs 2.78
D10 (mm) 0.0012
D50 (mm) 0.007
D60 (mm) 0.009
Coefficient of uniformity, Cu 7.50
Coefficient of curvature, Cc 1.48
Liquid limit (%) 34.5
Plastic limit (%) 29.2
Plasticity index 5.3
K. Hazirbaba et al. / Soil Dynamics and Earthquake Engineering 31 (2011) 372–384
374

to be connected to an external cooling unit, which allows for
controlling the temperature of the triaxial specimen. Additionally,
antifreeze coolant was used to further decrease the temperature
of the soil specimen. The antifreeze coolant was circulated
through a helical brass coil surrounding the triaxial specimen to
create a multi-dimensional freezing process. It is important to
note that the helical brass coil has a slightly larger diameter than
that of the specimen and thus is not in direct contact with the
specimen. To minimize heat loss and help maintain a constant
target temperature during the freezing process, the triaxial cell
was insulated completely with a Styrofoam box. To continuously
monitor and control the temperature inside the triaxial cell and
around the soil specimen, six (6) separate thermistors were
installed, as shown in Fig. 4. The displacement was measured with
a high precision LVDT, and the load was monitored using a load
cell. Details of the sensors used are presented in Table 2.
3.3. Specimen preparation, saturation, and consolidation
The specimens tested were 101.6 mm in diameter and
211.0 mm in height. They were reconstituted by the under-
compaction method [26] with initial water content that produced
50% saturation in the specimen. The moist soil was placed in
layers within the mold. To compensate for densiﬁcation during
compaction, the bottom layers were formed to have slightly
Fig. 3. Grain size distribution of the soil studied.
Fig. 4. Schematic of the testing system and components: (a) cyclic triaxial system and (b) installation of thermistors around the cylindrical triaxial specimen.

looser densities than the layers above them. Thus, approximately
uniform density distribution throughout the specimen could be
attained through the whole compaction process. From the field
investigation data the in-situ void ratio corresponding to 100 kPa
effective confining pressure was found to be e¼1.06. Thus, all
specimens were prepared at a target void ratio of 1.06. The slight
densification that typically occurred during saturation and
consolidation was accounted for by forming the specimens at an
initial void ratio of 1.08. To help expedite the saturation process,
the specimen was first percolated with carbon dioxide for about
1 h at a pressure difference (i.e., the head between top and bottom
of the specimen) of approximately 10 kPa. Then, the specimen
was flushed with de-aired water. The flushing of de-aired water at
a relatively low pressure difference of 15 kPa was completed in
about 24 h. Following flushing of the specimen, the triaxial cell
was filled with antifreeze liquid and the backpressure saturation
was initiated. Full saturation of the specimen was achieved
through application of backpressures of approximately 100 kPa.
After a minimum acceptable B-value of 0.95 was obtained, the
consolidation procedure was started. All of the specimens were
isotropically consolidated to the desired effective confining stress
of approximately 100 kPa. Some of the tests were conducted
at a slightly higher or lower effective confining stress than
100 kPa. This was due to the fluctuation in the air pressure source
over the long duration of testing (especially for specimens
conditioned at different temperatures). Nonetheless, constant
effective confining stress was maintained throughout each test.
Following completion of consolidation in about 24 h, the uncondi-
tioned specimens (specimens of the first group tests), which
were prepared for the study of moderate ground conditions (i.e.,
unfrozen), were directly subjected to strain-controlled, undrained,
Fig. 5. Thermistor and vertical displacement records at various conditioning temperatures: (a) at 24 1C; (b) at 5 1C; (c) at 1 1C; (d) at 0.5 1C; and (e) at 0.2 1C.
Table 2
Cooling unit and the sensors used in the triaxial system.
Setup or sensors Range Sensitivity
External cooling unit (1C) 50 to 40 70.02
Thermistor (1C) 80 to 75 70.1
Precision LVDT (mm) 70.50 70.001
Load cell (lb) 7500 71
Pore water pressure transducer (psi) 7100 70.20
Volume change transducer (ml) 80 71
376

cyclic triaxial tests. However, the other three group specimens
were subjected to thermal conditioning following consolidation
prior to execution of the cyclic triaxial test. The thermal
conditioning procedures are discussed in the following section.
3.4. Specimen thermal conditioning
3.4.1. Thermal conditioning at 24, 5, 1, 0.5, and 0.2 1C
After consolidation, each specimen was frozen multiaxially
under drained conditions at a confining effective pressure of
approximately 100 kPa. During the freezing process, chilled
coolant at 25 1C was circulated through the top cap and the
brass coiling around the specimen. The change in temperature
around the specimen was continuously monitored using thermis-
tors, as shown in Fig. 5. It took about 20–24 h to reach a stable
condition at about 10 1C, which was selected as a target freezing
temperature to ensure a fully frozen state. Once the frozen state
was achieved, the specimen was conditioned back to a target
temperature (e.g., 24, 5, 1, 0.5, and 0.2 1C). During thermal
conditioning, the change in the height of the specimen was also
monitored and recorded, as shown in Fig. 5. Conditioning was
complete when the target temperature was obtained and no
further change in height was observed. An acceptable target
temperature existed when the thermistor readings were stable.
3.4.2. Thermal conditioning through freeze–thaw cycles
The conditioning of the specimens in this series was achieved
through a similar procedure to the one outlined above. The only
difference was that after the fully frozen state was achieved at
about 10 1C, the thawing process was initiated by exposing the
triaxial cell to room temperature. In this series of tests, the soil
specimens were subjected to 1, 2, or 4 freeze–thaw cycles. Typical
thermistor records of specimens subjected to freeze–thaw cycles
are presented in Fig. 6. It should be noted that the record for 1
freeze–thaw cycle is identical to that of the specimen conditioned
at 24 1C in the first series of tests (Fig. 5a) and will not be repeated
in Fig. 6.
3.4.3. Thermal conditioning at 0.5 and 0.2 1C through three
different paths
The thermistor records of the specimens conditioned at near-
freezing (the term near-freezing is used here to indicate
temperatures slightly above and below 0 1C) temperatures of 0.5
and 0.2 1C consistently indicated larger permanent displace-
ment than those conditioned at higher temperatures as shown in
Fig. 5e and d. Although the amount of permanent displacement
due to thermal conditioning at 0.5 and 0.2 1C remained less than
3 mm (corresponding to less than 1.5% axial strain for triaxial
specimens), it raised a question as to whether allowing for longer
time than 48 h at the target conditioning temperature would
result in less displacement. To investigate this and also if different
thermal conditioning paths affect the generation of excess pore
pressure, the target near-freezing temperatures of 0.5 and 0.2 1C
were reached by three different paths as described in Table 3.
Thermal conditioning paths (TCP) 1 and 2 involved freezing
Fig. 6. Thermistor and vertical displacement records at various freeze–thaw
cycles: (a) 2 freeze–thaw cycles and (b) 4 freeze–thaw cycles (note that Fig. 5a is
also a record of 1 freeze–thaw cycle).
Table 3
Thermal conditioning paths at near-freezing temperatures.
Target
temperature
Thermal
conditioning
path
Procedure
0.5 1C TCP 1 1. Following consolidation, the soil specimen
conditioned at 10 1C for about 20–24 h
2. Conditioning temperature increased to
0.5 1C and remained at this value for 24 h
3. Temperature increased to the final target
value of 0.5 1C and conditioned for about 48 h
before cyclic testing was initiated
TCP 2 1. Following consolidation, the soil specimen
2. Conditioning temperature increased to final
target value of 0.5 1C and conditioned for 6–7
days before cyclic testing was initiated
TCP 3 Following consolidation, the temperature
decreased directly to the target value of
0.5 1C and conditioned for about 48 h before
cyclic testing was initiated
0.2 1C TCP 1 1. Following consolidation, the soil specimen
2. Conditioning temperature increased to
1.2 1C and remained at this value for 24 h
3. Temperature increased to the final target
value of 0.2 1C and conditioned for about
48 h before cyclic testing was initiated
TCP 2 1. Following consolidation, the soil specimen
2. Conditioning temperature increased to final
target value of 0.2 1C and conditioned for
6–7 days before cyclic testing was initiated
TCP 3 Following consolidation, the temperature
decreased directly to the target value of
0.2 1C and conditioned for about 48 h
before cyclic testing was initiated

followed by thawing to get to the target temperature, while TCP 3
reached the target temperature through cooling without freeze–
thaw.
Typical thermistor records from specimens at 0.2 and 0.5 1C
through different thermal conditioning paths are presented in
Fig. 7. TCP 1 for 0.2 and 0.5 1C is shown in Fig. 5d and e,
respectively. The displacement results in Figs. 5d, e and 7 showed
no consistent trend for TCP 1 and TCP 2, however TCP 3 always
resulted in little to no displacement.
3.5. Strain-controlled cyclic triaxial test
Following thermal conditioning, the triaxial test chamber
drainage valves were closed and a cyclic loading test was started.
ASTM D5311 [7] recommends a system capable of 0.1–2 Hz for
stress-controlled cyclic triaxial testing. Sinusoidal axial strain
amplitudes in this study were applied at 0.1 Hz frequency, and the
pore pressure response was monitored. Lower frequency loading
was preferred to ensure equilibration of pore water pressure
throughout the specimen and thus provide more accurate pore
pressure measurements [20]. The experiment was continued for
50 loading cycles or until an excess pore pressure ratio of 0.90 or
higher was obtained. The shear strain values can be estimated by
multiplying the axial strain values by a strain conversion factor
of 1.50, as recommended by Ishihara and Yoshimine [22] for
undrained conditions. A typical result from a strain-controlled
test is shown in Fig. 8. A soil specimen was subjected to a constant
shear strain amplitude of 0.3% (Fig. 8a). Excess pore pressure was
generated with increasing loading cycles (Fig. 8b). The sinusoidal
peak shear stress decreased with increasing loading cycles
(Fig. 8c). Shear stress–shear strain curves shown in Fig. 8d
reflect that the specimen became softer and softer with increasing
loading cycles.
Excess pore water pressure generation in cyclic testing follows
a pattern similar to that of loading (e.g., sinusoidal in this case),
and is typically defined by the excess pore water pressure ratio, ru,
which is given by the following equation:
ru ¼ Du=su3 ð1Þ
where Du is the excess pore pressure; and s0
3 the initial effective
confining pressure. Although excess pore pressure ratio can be
calculated at any time during a loading cycle, the value that is
commonly used is defined by taking into account the excess pore
pressure at the end of each loading cycle (i.e., residual excess pore
pressure). The values of excess pore pressure ratio throughout this
paper are evaluated based on cycle end measurements.
4. Results and discussion
4.1. Excess pore water pressure generation at various temperatures
The generation of pore water pressure in specimens tested at
room temperature without any thermal conditioning is shown
in Fig. 9(a) for various induced shear strain levels in terms of
excess pore pressure ratio, ru, versus number of loading cycles, N.
As indicated earlier, this case represents a temperate (i.e.,
unfrozen) ground condition. The induced shear strain level
ranged between 0.005% and 0.8%. At the smallest induced shear
strain level of 0.005%, no excess pore pressure was developed. For
the cyclic shear strain of 0.01%, very little excess pore pressure
was developed with excess pore pressure ratio was of 0.006 at
N¼10, and about 0.02 at the end of the test (N¼50). This finding
Fig. 7. Thermistor and vertical displacement records at near-freezing temperatures through various conditioning paths: (a) TCP 2 at 0.2 1C; (b) TCP 3 at 0.2 1C; (c) TCP 2
at 0.5 1C; and (d) TCP 3 at 0.5 1C (TCP: thermal condition path) (note that TCP 1 at 0.2 1C is presented in Fig. 5e and TCP 1 at 0.5 1C is shown in Fig. 5d).
378

is important in terms of defining the threshold shear strain level
(gt) [14], below which no volume change under drained condition
and little to no excess pore pressure under undrained condition
occurs during cyclic loading. Typical values of gt are 0.01–0.015%
for sands and 0.024–0.06% for cohesive soils [21]. The fact
that minimal excess pore pressure was developed at the 0.01%
induced shear strain in an unfrozen specimen indicates a
threshold shear strain level larger than 0.01% and closer to
0.02%. When considering the low-plasticity nature of the soil, a
threshold shear strain slightly larger than 0.01% is in agreement
with the findings of Hsu and Vucetic [21]. For g¼0.05%, the excess
pore pressure ratio at N¼10 is 0.12, with a maximum value of
0.23 obtained at N¼50. At higher levels of induced cyclic shear
strain, larger pore pressure values were obtained. For g¼0.8%, ru
was 0.25 for N¼1 and increased to about 0.93 at the end of the
test (N¼50).
Fig. 9(b) presents the pore pressure generation curve of
unconditioned specimens for N¼10 along with the upper and
lower bound curves proposed by Dobry [15] for clean sands at
relative densities ranging from 20% to 80% and under confining
effective stress of 25–200 kPa. The results from this study
consistently fall below the lower bound for clean sand indicating
less pore pressure generation than that of clean sand. At 0.3%
induced shear strain level, for example, clean sands are expected
to experience ru of 0.65–1.0 whereas at the same strain level, ru of
the silt under investigation appears to be less than 0.50. The lower
pore pressure generation of the silt at all strain levels indicate that
when subjected to the same level of shaking at temperate ground
conditions, the silt has lower liquefaction potential than clean
sands. The results from the unconditioned specimens are used as
the basis for comparison in the following sections.
The pore pressure histories from specimens conditioned at
various temperatures, are shown in Fig. 10. It is interesting to note
that at 0.1% shear strain level (Fig. 10a), the largest excess pore
pressure was recorded at 0.5 1C. The 24, 5, and 1 1C conditions all
generated similar pore pressures, which were lower than those of
the unconditioned case. Similar trends and patterns of excess
pore pressure generation were observed at 0.3% shear strain
level (Fig. 10b). The 0.2 1C condition resulted in development of
negative pore pressure, which implies a dilative behavior at 0.1%
shear strain level. Testing the 0.2 1C condition at 0.3% shear
strain was not possible due to the capacity of the load cell used.
However, to investigate the 0.2 1C condition further, a smaller
strain level of 0.03% was also tested and the results are presented
in Fig. 11. Very minimal excess pore pressure was generated
Fig. 8. Typical results from a strain-controlled cyclic triaxial test.
Fig. 9. Excess pore pressure generation of unconditioned specimens: (a) pore
pressure histories for various levels of shear strain and (b) pore pressure
generation curve at N¼10.

(ru o0.04) at 0.03% shear strain and unlike the response at 0.1%
shear strain, no negative pore pressure was recorded. Near or
below 0 1C, a portion of the pore water is still at unfrozen state
[27], and the development of negative pore pressure at 0.1% may
be attributed to the interaction between frozen and unfrozen pore
water as observed by Eigenbrod et al. [16] during static freeze–
thaw tests. Although the cyclic loading was initiated at the
equilibrium condition (i.e., steady freezing at 0.2 1C according to
thermistor records), a shear strain level of 0.1% appears to cause
redistribution of the unfrozen pore water within the specimen,
which promotes suction towards the growing ice lenses and thus
negative pore pressure develops. It may be argued that at smaller
induced shear strain levels, as observed in the case of g¼0.03%,
such redistribution of the unfrozen water does not occur and
therefore no negative excess pore pressure develops within the
specimen.
The influence of temperature on excess pore pressure generation
is shown in Fig. 12 in terms of ru versus g for N¼10. Upper and
lower bound pore pressure generation curves are proposed based on
the measurements from specimens conditioned at various
temperatures. These curves are useful in determining the variation
in pore pressure generation for a range of shear strain levels at a
specific number of loading cycles. The results in Fig. 12 indicate that
at all strain levels the excess pore pressure generation is significantly
influenced by the conditioning temperature. The data in Fig. 12
further suggest that the threshold shear strain level is highly
sensitive to the conditioning temperature; especially at near-
freezing state. The threshold shear strain at 0.5 1C is somewhat
similar to that of the unconditioned case (gtE0.02%) while much
larger threshold shear strain levels (gt40.1%) appear to exist when
the conditioning temperature is slightly below 0 1C. Fig. 13 presents
a summary of the experimental results in terms of excess pore
pressure versus conditioning temperature for two strain levels;
0.1% and 0.3%, and at various number of loading cycles. The results
indicate that subjecting the soil specimens to temperature
conditioning led to a significant drop in pore pressure generation
with the exception of conditioning at 0.5 1C, which consistently
generated high pore pressures. It is also evident from Fig. 13 that
there is relatively less variation in the amount of excess pore
pressure generation at 24, 5, and 1 1C conditions, and that the abrupt
change occurs at near-freezing at 0.5 1C.
4.2. Influence of freeze–thaw cycles on excess pore pressure
generation
In the second phase of this research, the influence of freeze–
thaw cycles on pore water pressure generation was investigated.
The excess pore pressure histories of specimens subjected to no
freeze–thaw (unconditioned), 1, 2, and 4 freeze–thaw cycles are
presented for the cyclic shear strain levels of 0.1%, 0.3%, and 0.8%
in Fig. 14. In general, the largest excess pore pressure was
obtained from the unconditioned specimens, which indicates that
subjecting the soil to freeze–thaw cycles caused reduction in the
generation of excess pore pressure. A reduction in the excess pore
Fig. 10. Excess pore pressure generation at various temperatures: (a) pore
pressure histories at g¼0.1% and (b) pore pressure histories at g¼0.3%.
Fig. 11. Excess pore pressure generation at 0.2 1C due to g¼0.03% and g¼0.1%.
Fig. 12. Influence of temperature on excess pore pressure generation at N¼10.
380

pressure generation describes increased cyclic resistance and
decreased liquefaction potential. It appears that the first freeze–
thaw cycle affects excess pore pressure generation most at lower
shear strain level of 0.1% through all loading cycles (Fig. 14a).
After 50 loading cycles, ru of the unconditioned case was 0.4
whereas the ru of the 1 freeze–thaw cycle case increased to only
0.25. A similar trend was recorded at 0.3% shear strain until N¼10
(Fig. 14b), beyond which the specimen subjected to 4 freeze–thaw
cycles appeared to transitionally generate the lowest excess pore
pressure. At the end of testing, ru values of 0.75 and 0.55 were
recorded for the unconditioned case and 4 freeze–thaw cycle case,
respectively. At 0.8% shear strain level (Fig. 14b), 2 and 4 freeze–
thaw cycles were more influential on excess pore pressure
generation with much lower ru values than those of the
unconditioned and 1 freeze–thaw cycle cases. However, all
specimens experienced significant excess pore pressure ranging
between ru of 0.93 (unconditioned case) and 0.78 (4 freeze–thaw
cycles case) after 50 loading cycles. The results in Fig. 14 also
show that the effect of freeze–thaw cycles becomes more
pronounced with increasing number of loading cycles.
Fig. 13. Excess pore pressure generation versus conditioning temperature: (a) trends at g¼0.1% and (b) trends g¼0.3%.
Fig. 14. Excess pore pressure generation after various freeze–thaw cycles: (a) pore pressure histories at g¼0.1%; (b) pore pressure histories at g¼0.3%; and (c) pore
pressure histories at g¼0.8%.

The influence of freeze–thaw cycles on pore pressure genera-
tion at N¼10 is presented in terms of ru versus g in Fig. 15. The
upper and lower bounds are formed by the responses of
unconditioned specimens and specimens subjected to one
freeze–thaw cycle, respectively. Based on the results in Fig. 15,
it can also be concluded that the threshold shear strain may be
expected to move slightly to the right (i.e., larger threshold shear
strain value) for the specimens subjected to freeze–thaw cycles.
4.3. Influence of thermal conditioning path at near-freezing
temperatures
The influence of different thermal conditioning paths on excess
pore pressure generation was investigated at 0.5 and 0.2 1C. The
thermal conditioning paths used were discussed earlier (see
Table 3 and Fig. 7). The testing results for 0.5 1C are presented in
Fig. 16. The pore pressure histories at g¼0.3% (Fig. 16a) show that
for all loading cycles TCP 1 resulted in the largest pore pressure
generation. The pore pressure generation in case of TCP 2 was
consistently less than that of TCP 1, while TCP 3 caused the least
pore pressure generation of the three thermal conditioning paths.
The difference in excess pore pressure ratio can be as much as
200%, especially in early loading cycles indicating the significance
of the thermal conditioning path on the pore pressure response.
The pore pressure generation versus cyclic shear strain for 0.5 1C,
as shown in Fig. 16b, indicates a consistent trend at various
shear strain levels and confirms the significant effect of thermal
conditioning path on excess pore pressure generation. The results
for the other near-freezing temperature of 0.2 1C are shown in
Fig. 17. TCP 1 and TCP 2 in this case displayed little to no excess
pore pressure generation during the first 10 loading cycles
(Fig. 17a). Past N¼10, negative pore pressure was recorded for
TCP 1. The negative pore pressure response at 0.2 1C was
discussed earlier (see Fig. 11) and attributed to the redistribution
of the unfrozen water content which promotes a tendency for the
existing ice lenses to grow, and thus create a suction resulting in
negative pore pressure. A similar response was recorded from TCP
2, however in case of TCP 2 it took more loading cycles (until
about N¼30) for the negative pore pressure to develop. The
response from TCP 3 is quite different with positive and relatively
high excess pore pressure generation at all loading cycles reaching
up to ru of 0.25 at N¼50. Despite a stable record of target
temperature (i.e., 0.2 1C) from all six thermistors employed
around the specimen, it can be argued that the unfrozen water
content in TCP 3 was much higher than that of TCP 1 or TCP 2, and
thus the behavior was dominated by the unfrozen water content,
which led to relatively significant pore pressure generation. TCP 1
and TCP 2 both involved freezing followed by thawing (or
warming) to the target temperature, and thus the ice lenses
within the specimen are dominant in case of either TCP 1 or TCP 2.
However, TCP 3 consisted of cooling the specimen down to the
target temperature of 0.2 1C and allowing to reach equilibrium
in 48 h. Formation of ice lenses, if any, throughout the specimen
in the case of TCP 3 will be very slow and therefore unfrozen
water content dominates within the first 48-h period after
reaching the target temperature of 0.2 1C. The results at
smaller strain levels confirm the findings at g¼0.1%. Fig. 17b
shows the excess pore pressure ratio for various strain levels for
Ttarget¼ 0.2 1C at N¼10 indicating TCP 3 as the only thermal
conditioning path to lead to a development of excess pore
pressure. Larger shear strain levels were avoided due to load
cell capacity limitation as discussed earlier.
5. Summary and conclusion
In this paper the influence of temperature and freeze–thaw
cycles on excess pore pressure generation of fine-grained soils
was evaluated. Laboratory-reconstituted soil specimens were
subjected to strain-controlled, undrained, cyclic triaxial testing.
The tests were performed at shear strain levels ranging between
0.005% and 0.8%. Four series of tests were conducted. They are:
Fig. 15. Influence of freeze–thaw on excess pore pressure generation at N¼10.
Fig. 16. Influence of thermal conditioning path on excess pore pressure generation
at 0.5 1C.
382

(i) tests on specimens subjected to no thermal conditioning or
freeze–thaw cycles; (ii) tests on specimens conditioned at 24, 5, 1,
0.5, and 0.2 1C; (iii) tests on specimens subjected to various
numbers of freeze–thaw cycles; and (iv) tests on specimens
conditioned at near-freezing temperatures of 0.5 and 0.2 1C
through different freeze–thaw paths. A triaxial setup was
modified to allow for temperature conditioning of the soil
specimens with multidirectional freezing. A series of thermistors
employed around the soil specimen provided continuous mon-
itoring of temperature.
The excess pore pressure generation of unconditioned speci-
mens was found to be lower than that of clean sand. This indicates
lower potential of liquefaction for the silt in comparison to clean
sands at the same level of shaking. However, strong enough
ground shaking (e.g., shaking that can induce shear strains beyond
0.3%) can lead to appreciable pore pressure generation and
eventual liquefaction of the silt. The testing of unconditioned
specimens also revealed a larger threshold shear strain level of
about 0.02% than the established value of 0.01% for clean sands.
This is in agreement with findings of Hsu and Vucetic [21] that
fine-grained materials display larger threshold shear strain than
clean sands.
Thermal conditioning of the specimens was found to be
influential on both excess pore pressure generation and the
threshold shear strain level. Specimens conditioned at 24, 5, and
1 1C displayed similar excess pore pressure generation, which was
significantly lower than that of unconditioned specimens.
However, as the temperature nears the freezing point of 0 1C,
there seems to be a transition in the cyclic behavior of the soil.
This transition was observed from the records at 0.5 and 0.2 1C.
Larger pore pressure generation than even the unconditioned case
was obtained at 0.5 1C while very little to negative pore pressure
was recorded at 0.2 1C. The abrupt response at near-freezing
temperatures may be attributed to the interaction between frozen
and unfrozen water content at those temperatures. Fig. 18
presents the change in unfrozen water content with respect to
temperature of a typical triaxial specimen of silt compacted at a
target void ratio of 1.06. The details of how the unfrozen water
content was measured are presented in [40]. The largest variation
in unfrozen water content occurs between 0.5 and 1.0 1C; the
initial 40% water content drops to less than 10% within this band.
Thus the dynamic response of the soil to cyclic straining at near
and slightly below-freezing temperatures is dictated by: (i) the
dominant phase (unfrozen or frozen) of the pore water, and (ii)
the level of induced shear strain. In case of unfrozen-dominant
pore water, the frozen portion appears to promote generation of
larger excess pore pressures than even completely unfrozen
condition. However, in case of frozen-dominant pore water, the
frozen portion (ice lenses) tends to grow when interacted with
unfrozen water, which causes development of negative excess
pore pressure. Such interaction occurs with the redistribution of
the unfrozen water, and this may not be initiated at small strains
as observed at the shear strain level of 0.03%.
An appreciable reduction in the excess pore generation was
found to occur due to freeze–thaw cycles. Specimens subjected to
1, 2, and 4 showed that the amount reduction in excess pore
pressure is dependent on the induced shear strain level and
number of loading cycles. More pronounced reduction was
recorded at large number of loading cycles. The threshold shear
strain was also found to slightly move to the right in the ru versus
g figure indicating larger values in the specimens subjected to
freeze–thaw cycles. The influence of complete freeze–thaw cycles
on the excess pore pressure generation is comparatively less than
that obtained at various conditioning temperatures. Fig. 19 plots
the excess pore pressure band for each case at N¼10. The
narrower inner band, based on freeze–thaw cycle tests, indicates
less variation of excess pore pressure generation.
Reaching the target conditioning temperature through differ-
ent thermal conditioning paths were found to cause quite
dissimilar developments of excess pore pressure at near and
below-freezing temperatures (i.e., 0.5 and 0.2 1C), which is also
attributed to the state of pore water and the interaction between
unfrozen and frozen water content. Reaching the target
Fig. 17. Influence of thermal conditioning path on excess pore pressure generation
at 0.2 1C.
Fig. 18. Change in unfrozen water content with temperature.

temperature by cooling the specimen down (from room tem-
perature) appeared to always create an unfrozen-dominant pore
structure with positive pore pressure generation.
Acknowledgments
Financial support for this research was provided by Alaska
Experimental Program to Stimulate Competitive Research Pro-
gram (EPSCOR)-NSF, Permafrost Technology Foundation, Institute
of Northern Engineering, Department of Civil and Environmental
Engineering, and Alaska University Transportation Center. This
support is gratefully acknowledged. Any opinions, findings and
conclusions or recommendations expressed in this document are
those of the authors and do not necessarily reflect the views of the
sponsors for this research. Professors Doug Goering and Yuri Shur
of University of Alaska Fairbanks, and Professor Joey Yang of
University of Alaska Anchorage are thanked for their comments
and support.
References
[1] AK DOT Report. Mable Creek Bridge #656, Tok Cutoff earthquake repairs,
2003.
[2] Alkire BD. Effect of variable-drainage freeze–thaw tests on post-thaw shear
strength. Transportation Research Record 1981;809:13–8.
[3] Alkire BD, Morrison JA. Comparative response of soils to freeze–thaw and
repeated loading. Special Report—United States Army, Cold Regions Research
and Engineering Laboratory, Hanover, New Hampshire, USA, 1983, p. 89–95.
[4] Andersland O, Anderson D. Geotechnical engineering for cold regions.
McGraw-Hill Companies, Incorporated; 1978.
[5] ASTM. Standard test method for particle-size analysis of soils, vol. D422.
West Conshohocken, PA: ASTM; 2002.
[6] ASTM. Standard test methods for liquid limit, plastic limit, and plasticity
index of soils, vol. D4318. West Conshohocken, PA: ASTM; 2005.
[7] ASTM. Standard test method for load controlled cyclic triaxial strength of soil,
vol. D5311. West Conshohocken, PA: ASTM; 2004.
[8] ASTM. Standard test methods for specific gravity of soil solids by water
pycnometer, vol. D854. West Conshohocken, PA: ASTM; 2000.
[9] Benson CH, Abichou TH, Olson MA, Bosscher PJ. Winter effects on hydraulic
conductivity of compacted clay. Journal of Geotechnical Engineering, ASCE
1995;121(1):69–79.
[10] Boulanger RW, Idriss IM. Liquefaction susceptibility criteria for silts and
clays. Journal of Geotechnical and Geoenvironmental Engineering, ASCE
2006;132(11):1413–26.
[11] Bray JD, Sancio RB. Assessment of the liquefaction susceptibility of fine-
grained soils. Journal of Geotechnical and Geoenvironmental Engineering,
ASCE 2006;132(9):1165–77.
[12] Chamberlain EJ, Gow AJ. Effect of freezing and thawing on the permeability
and structure of soils. Engineering Geology 1978;13(1–4):73–92.
[13] Chang NY, Yeh ST, Kaufman LP. Liquefaction potential of clean and silty
sands. In: Proceedings of the third international earthquake microzonation
conference, Seattle, USA, 1982, vol. 2, p. 1017–32.
[14] Dobry R, Ladd RS, Yokel FY, Chung RM, Powell D. Prediction of pore water
pressure buildup and liquefaction of sands during earthquakes by the cyclic
strain method. National Bureau of Standards, Building Science Series, vol.
168, 1982.
[15] Dobry R. Liquefaction of soils during earthquakes. Report No. CETS-EE-001,
National Research Council (NRC), Committee on Earthquake Engineering,
Washington, DC, 1985.
[16] Eigenbrod KD, Knutsson S, Sheng D. Pore-water pressures in freezing and
thawing fine-grained soils. Journal of Cold Regions Engineering, ASCE
1996;10(2):77–92.
[17] Finn WDL, Yong RN, Lee KW. Liquefaction of thawed layers in frozen soil.
Journal of the Geotechnical Engineering Division, ASCE 1978;104(10):
1243–55.
[18] Finn WDL, Yong RN. Seismic response of frozen ground. Journal of the
Geotechnical Engineering Division, ASCE 1978;104(10):1225–41.
[19] Goto S. Influence of a freeze and thaw cycle on liquefaction resistance of
sandy soils. Soils and Foundations 1993;33(4):148–58.
[20] Hazirbaba K, Rathje EM. Pore pressure generation of silty sands due to
induced cyclic shear strains. Journal of Geotechnical and Geoenvironmental
Engineering, ASCE 2009;135(12):1892–905.
[21] Hsu CC, Vucetic M. Threshold shear strain for cyclic pore-water pressure in
cohesive soils. Journal of Geotechnical and Geoenvironmental Engineering,
ASCE 2006;132(10):1325–35.
[22] Ishihara K, Yoshimine M. Evaluation of settlements in sand deposits following
liquefaction during earthquakes. Soils and Foundations 1992;32(1):
861–78.
[23] Kayen R, Thompson E, Minasian D, Moss RES, Collins B, Sitar N, et al.
Geotechnical reconnaissance of the 2002 Denali Fault, Alaska Earthquake.
Earthquake Spectra 2004;20(3):639–67.
[24] Koester JP. The Influence of fine type and content on cyclic strength. Ground
failures under seismic conditions, Geotechnical Special Publication no. 44,
ASCE, 1994, p. 330–45.
[25] Konrad JM. Effect of freeze–thaw cycles on the freezing characteristics of a
clayey silt at various overconsolidation ratios. Canadian Geotechnical Journal
1989;26(2):217–26.
[26] Ladd RS. Preparing test specimens using undercompaction. ASTM Geotech-
nical Testing Journal 1978;1(1):16–23.
[27] Neresova ZA, Tsytovich VA. Unfrozen water in frozen soils. NAS-NRC
Publishers; 1963. p. 230–4.
[28] Othman MA, Benson CH. Effect of freeze–thaw on the hydraulic conductivity
and morphology of compacted clay. Canadian Geotechnical Journal
1993;30(2):236–46.
[29] Polito CP, Martin JR. Effects of non-plastic fines on the liquefaction resistance
of sands. Journal of Geotechnical and Geoenvironmental Engineering, ASCE
2001;127(5):408–15.
[30] Qi J, Ma W, Song C. Influence of freeze–thaw on engineering properties of a
silty soil. Cold Regions Science and Technology 2008;53(3):397–404.
[31] Singh S. Liquefaction characteristics of silts. Geotechnical Special Publication
No. 44, ASCE, 1994; p. 105–16.
[32] Thevanayagam S, Shenthan T, Mohan S, Liang J. Undrained fragility of clean
sands, silty sands, and sandy silts. Journal of Geotechnical and Geoenviron-
mental Engineering, ASCE 2002;128(10):849–59.
[33] Viklander P. Permeability and volume changes in till due to cyclic freeze/
thaw. Canadian Geotechnical Journal 1998;35(3):471–7.
[34] Vinson TS. Parameter effects on dynamic properties of frozen soils. Journal
Geotechnical Engineering Division, ASCE 1978;104(10):1289–306.
[35] Vinson TS, Wilson CR, Bolander P. Dynamic properties of naturally frozen silt.
Washington, DC, USA, Fairbanks, AK, USA: National Academies Press; 1983.
p. 1315–20.
[36] Wijewickreme D, Sanin MV, Greenaway GR. Cyclic shear response of fine-
grained mine tailings. Canadian Geotechnical Journal 2005;42:1408–21.
[37] Yashinsky M, Eidinger J. Performance of lifelines during the November 3,
2002 Denali, Alaska Earthquake. Reconnaissance Survey Report of Technical
Council on Earthquake Engineering, ASCE 2003.
[38] Yoshimi Y, Tokimatsu K, Ohara J. In situ liquefaction resistance of clean sands
over a wide density range. Geotechnique 1994;44(3):479–94.
[39] Yoshimi Y, Goto S. Liquefaction resistance of silty sand based on in situ frozen
samples. Geotechnique 1996;46(1):153–6.
[40] Zhang Y. Impact of freeze–thaw on liquefaction potential and dynamic
properties of Mabel Creek silt. PhD dissertation, The University of Alaska
Fairbanks, 2010.
[41] Zimmie TF, La Plante C. Effect of freeze/thaw cycles on the permeability of a
fine-grained soil. In: Hazardous and industrial wastes—Proceedings of the
Mid-Atlantic industrial waste conference, Philadelphia, PA, USA. Lancaster,
PA, USA: Technomic Publishers Co. Inc.; 1990. p. 580–93.
Fig. 19. Comparison of temperature and freeze–thaw effects on excess pore
pressure generation.
384

Evaluation of temperature effects on excess pore pressure in fine-grained soils

Recommended

Recommended

More Related Content

Similar to Evaluation of temperature effects on excess pore pressure in fine-grained soils

Similar to Evaluation of temperature effects on excess pore pressure in fine-grained soils (20)

More from PinakRay2

More from PinakRay2 (7)

Recently uploaded

Recently uploaded (20)

Evaluation of temperature effects on excess pore pressure in fine-grained soils