Design of Foundations On Sensitive Champlain Clay Subjected To Cyclic Loading
Design of Foundations On Sensitive Champlain Clay Subjected To Cyclic Loading
Design of Foundations On Sensitive Champlain Clay Subjected To Cyclic Loading
Abstract: Sensitive clay subjected to cyclic loading may experience gradual loss of its shear strength, which may lead to liquefaction.
Foundations built on this clay would suffer extensive settlement and significant loss of bearing capacity or perhaps catastrophic failure.
This paper presents an experimental investigation on sensitive 共Champlain兲 clay obtained from the city of Rigaud, Quebec 共Canada兲.
Consolidation tests, static and cyclic undrained and drained triaxial tests were performed on representative samples of this clay. The
objective of this investigation was to examine the influence of the physical and mechanical parameters, which govern the shear strength
of sensitive clay subjected to cyclic loading. Based on the results of the present investigation and those available in the literature, it can
be reported herein that the undrained response is the most critical for these foundations; furthermore, the preconsolidation pressure is
considered as an important parameter in establishing the shear strength of sensitive clay. A design procedure is developed to determine the
safe zone for the undrained and drained responses, within which a combination of the cyclic deviator stress and the number of cycles for
a given soil/loading/site conditions can achieve a quasielastic resilient state without reaching failure. The proposed design procedure is
applicable to all regions around the world, where sensitive clays can be found. Furthermore, this procedure can be adopted to examine the
conditions of existing foundations built on sensitive clay at any time during its lifespan.
DOI: 10.1061/共ASCE兲1090-0241共2008兲134:7共929兲
CE Database subject headings: Clays; Stress; Pore water pressure; Drainage; Foundation design; Cyclic loads.
Introduction reported that for overconsolidated highly sensitive clay, the pore
pressure generated during cyclic loading is independent of the
Cyclic loading on foundations built on sensitive clay produces strain rate, while the peak shear strength decreases due to the
remolding action that facilitates the dissolution of the bonding increase of the strain rate. Liang and Ma 共1992兲 proposed a model
agents between particles, which may lead to slippage, loss of to predict the degradation of the undrained shear strength, stiff-
shear strength, and often liquefaction. ness, total strain, and the pore pressure in sensitive clay during
Previous research on sensitive clay focused on developing em- cyclic loading. Lefebvre and Pfendler 共1996兲 reported that for
pirical formulas to relate the number of cycles and the level highly sensitive clay under no initial static shear stress, the shear
strength of the clay decreases rapidly upon increasing the number
of sensitivity to the stress-strain characteristics of the clay 共San-
of cycles. This rate, however, decreases due to an increase of the
grey et al. 1969; Ansal and Erken 1989兲. Eekelen and Potts
initial static shear stress. Miller et al. 共2000兲 proposed an empiri-
共1978兲 have introduced a fatigue parameter for predicting the
cal relationship between the cyclic shear stress ratio and the de-
shear strength of clay at the end of a given loading cycle. Iwasaki
gree of saturation of the clay.
et al. 共1978兲 introduced the concept of “quasielastic resilient”
Based on these studies, it can be reported herein that serious
state for soil under drained, cyclic stress-controlled conditions.
efforts were made to model the complex behavior of sensitive
The field and laboratory data reported by Chagnon et al. 共1979兲 clay subjected to cycling loading. Nevertheless, these models
provides valuable information to this complex behavior. Seed and have failed to recognize the preconsolidation pressure as a major
Idriss 共1982兲 reported that for sensitive clay under drained condi- parameter in governing this behavior and further as the link be-
tions, the higher the frequency of loading, the closer the response tween the physical and mechanical parameters of sensitive clay.
resembles the undrained condition. Lefebvre and LeBoeuf 共1987兲 Physical parameters include the natural water content 共wc兲,
liquid limit 共LL兲, and plastic limit 共PL兲, which can be combined
1
Professor, Dept. of Building, Civil, and Environmental Engineering, in terms of the liquidity index 共IL兲 given in Eq. 共1兲
Concordia Univ., 1455 de Maisonneuve Blvd., W. Montreal, Quebec,
Canada, H3G 1M8. wc − PL
2 IL = 共1兲
Research Associate, Dept. of Building, Civil, and Environmental Ip
Engineering, Concordia Univ., 1455 de Maisonneuve Blvd., W. Montreal,
Quebec, Canada, H3G 1M8. where I p = plasticity index 共LL− PL兲.
Note. Discussion open until December 1, 2008. Separate discussions The mechanical parameters include: The sensitivity number
must be submitted for individual papers. To extend the closing date by 共St兲; number of cycles 共N兲; cyclic deviator stress 共qcyc兲; pore
one month, a written request must be filed with the ASCE Managing water pressure 共u兲; axial strain 共兲; preconsolidation pressure
Editor. The manuscript for this paper was submitted for review and pos- 共c兲; and confining pressure 共3兲.
sible publication on September 11, 2003; approved on August 14, 2007. Bjerrum 共1954兲 proposed a coefficient k, which relates the
This paper is part of the Journal of Geotechnical and Geoenvironmental liquidity index IL to the sensitivity number St as given in Eq. 共2兲
Engineering, Vol. 134, No. 7, July 1, 2008. ©ASCE, ISSN 1090-0241/
2008/7-929–937/$25.00. St = exp共kIL兲 共2兲
Downloaded 16 Jul 2009 to 147.83.51.2. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
Table 1. Summary of the Physical Characteristics of the Clay Table 2. Summary of Unconfined Compression Test Results
Property Value Cu Cur Sensitivity
Test 共undisturbed sample兲 共remolded sample兲 number
Classification Highly plastic 共CH兲
number kPa kPa St
Water content 44–51%
Liquid limit 共LL兲 69% 1 5.482 0.581 9.435
Plastic limit 共PL兲 25% 2 4.937 0.683 7.228
Liquidity index 共IL兲 0.43–0.59 3 5.571 0.849 6.561
Plasticity index 共I P兲 44%
Unit weight 共␥兲 17 kN/ m
Specific gravity 共Gs兲 2.74 Cu
St = 共3兲
Cur
where Cu = undrained shear strength of the undisturbed sample;
and Cur = undrained shear strength of the remolded sample. The
This paper introduces the preconsolidation pressure as a gov- deduced values of St were in the range of 6.5 to 9.5; hence, the
erning parameter for the behavior of sensitive clay; furthermore, clay was identified as having medium to high sensitivity similar
it presents a design procedure for foundation on sensitive clay to Chagnon et al. 共1979兲, which confirms the results of this
subjected to cyclic loading. investigation.
Conventional consolidation tests were performed on these
samples. The deduced value of preconsolidation pressure 共c兲
Experimental Investigation was close to the value of the overburden pressure 共o兲 at the
depth of 4 m; accordingly, the clay is classified as normally con-
Samples of sensitive clay were collected from the City of Rigaud, solidated.
Quebec, Canada, which represents Champlain clay deposits in Static undrained triaxial compression tests were performed on
eastern Canada and the United States. Quigley 共1980兲 described samples having a cross-sectional area of 9 cm2 and length of
these deposits as postglacial marine clay with varying properties. 7.5 cm. In order to maintain the samples in the triaxial mold and
Samples were extracted from a depth of 4 m, as blocks measuring further to reach full saturation, the samples were first consolidated
300 mm⫻ 300 mm⫻ 300 mm. The index properties of this clay under a confining pressure of 207 kPa. The deviator stress 共q
were determined in the laboratory and given in Table 1. = 1 − 3兲 at failure was found to be in the range of 88 to 98 kPa.
In order to evaluate the sensitivity number 共St兲 of the clay Cyclic undrained and 共partially兲 drained triaxial tests were per-
tested, unconfined compression tests were conducted on undis- formed on these samples. Prior to cycling, the samples were
turbed and remolded samples having the same void ratio. Re- consolidated under a confining pressure, equivalent to the precon-
molded samples were kneaded, and then carefully placed in a solidation pressure c 共207 kPa兲. In this investigation, samples
slurry consolidometer apparatus, avoiding entrapping air. After were subjected to cyclic deviator stress equivalent to 33, 35, or
removal from the mold, the mass and dimensions of the specimen 67% of the deduced deviator stress at failure from the static tri-
were determined. Fig. 1 presents stress-strain curves for these axial compression test 共88 kPa兲. The cyclic deviator stress was
tests and Table 2 presents a summary of these test results. The applied at a rate of four cycles per h. This rate represents the
sensitivity number 共St兲 was determined as follows: frequency that wind and wave action impose on tall structures;
Downloaded 16 Jul 2009 to 147.83.51.2. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
Fig. 2. Cylic triaxial test results—pore-water pressure 共Du兲 kPa versus number of cycles 共N兲
furthermore, this rate allows enough time to record stresses, clay and subjected to cyclic loading. However, these foundations
strains, volume changes, and pore water pressure at regular inter- may also reach failure at a relatively higher number of cycles
vals during testing. Procter and Khaffaf 共1984兲 reported that for during the drained period.
undrained load-controlled tests, the cyclic stress ratio is indepen- Table 3 presents a summary of the results obtained in the
dent of the load frequency. In this investigation, the failure point present experimental investigation. It can be noted from this Table
was defined at the cycle when the pore water pressure was sig- that:
nificantly increased accompanied by a rapid increase of the axial 1. Sample UT-01 failed at a relatively low number of cycles due
strain 共 % 兲. The initial pore water pressure, confining pressure, to the high cyclic deviator stress ratio 共67%兲.
and the rate of load cycling remained unchanged during testing. 2. Sample UT-02 failed at a relatively higher number of cycles,
In this investigation, all measuring devices were calibrated as compared to test UT-01, due to the low cyclic deviator
regularly and each test was repeated at least twice to ensure re- stress ratio 共35%兲.
producibility of test data. 3. Sample UT-03 did not reach failure due to the fact that, in
this test, the pore water pressure built up rapidly up to a
maximum value at the seventh cycle, beyond which the
Results sample attained a quasielastic resilient state without reaching
failure. This can be attributed to the relatively low cyclic
In general, the samples tested under drained condition required deviator stress ratio 共33%兲 used in this test as compared to
relatively higher number of cycles 共N兲 and higher stress deviator tests 共UT-01兲 and 共UT-02兲.
ratio to reach failure, as compared to the samples tested under 4. Sample DT-04 failed at a significantly higher number of
undrained condition. Furthermore, for both undrained and drained cycles mainly due to the relatively low pore water pressure
tests, the higher the cyclic deviator stresses 共qcyc兲, the fewer generated during this response. Thus, samples under the
cycles 共N兲 needed to reach failure. For relatively lower values of drained condition 共DT-04兲, required a much higher number
the cyclic deviator stress 共qcyc兲, the clay may reach a quasielastic of cycles to reach failure as compared to those under und-
resilient state without reaching failure. rained condition 共UT-01兲.
Fig. 2 presents the pore water pressure 共⌬u兲 versus the number 5. Sample DT-05 did not reach failure. Comparing the results of
of cycles 共N兲 for both the undrained 共UT兲 and drained 共DT兲 tests. tests DT-04 and DT-05, it can be noted that sample DT-05
It can be noted from this figure that for undrained tests, the pore was subjected to a relatively low cyclic deviator stress ratio.
water pressure 共⌬u兲 was rapidly increasing, while for drained 6. Comparing the results of tests UT-02 and DT-05, it can be
tests, the pore water pressure was relatively low, due to the dis-
sipation, which takes place as a result of the drained response.
The high pore water pressure 共⌬u兲 observed during the undrained Table 3. Cyclic Triaxial Compression Test Results
response was the prime reason for the sample to reach failure at a Initial degree Cyclic Axial Total
relatively low number of load cycles. Accordingly, it can be stated Test of saturation stress ratio strain number of
that during the undrained period, the shear strength of sensitive number 共S兲% 共qcyc / q兲% 共兲% cycles 共N兲 Notes
clay is significantly reduced, and catastrophic failure of the struc- UT-01 90.13 67 30 63 Failed
ture may take place. While during the drained 共or partially UT-02 89.13 35 32 121 Failed
drained兲 response, the consolidation of the clay that occurs in-
UT-03 86.01 33 10 101 Not failed
creases the shear strength, which may offset in part or in full the
DT-04 87.36 67 48 454 Failed
loss due to load cycling. Thus, the undrained period is regarded as
DT-05 97.06 35 13 179 Not failed
the critical period in the lifespan of foundations built on sensitive
Downloaded 16 Jul 2009 to 147.83.51.2. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
Fig. 3. Stress path 共s-qcyc兲 for test 共UT-01兲
noted that while both samples were tested under the same cycles. Furthermore, test UT-02 underwent a large axial strain
cyclic deviator stress ratio, sample DT-05 showed higher re- compared to UT-03. This is due to a possible threshold/critical
sistance to failure at 179 cycles, although the initial degree of cyclic stress ratio, which appears to have taken place within the
saturation was relatively higher than UT-02. This emphasizes range of 33 to 35%. In addition, test 共UT-03兲 showed a relatively
the importance of the drainage condition compared to the small accumulated strain of 9% 共recoverable plus unrecoverable兲,
initial degree of saturation. which reached a value of 10% after 25 cycles, while test 共UT-01兲
Fig. 3 presents the results of the undrained test UT-01 in the form showed a significant increase in the accumulated strain 共about
of effective stress-path 共 − qcyc兲. It can be noted from this figure 38% after 10 cycles and 45% at 63 cycles兲, as the sample had
that the effective stress path is shifting towards the failure enve- already failed.
lope 共qs / s兲 due to an increase in the number of cycles. In this Figs. 5 and 6 present typical hysteric loop patterns for the tests
analysis, the failure envelope was determined from the results of UT-01 and DT-04, respectively, in the form of the cyclic deviator
static triaxial undrained tests. These curves were further used to stress 共qcyc兲 versus the strain 共 % 兲. While both tests were per-
establish the cycle at which failure of the sample was reached. In formed at the same cyclic deviator stress ratio 共67%兲, sample
the case of samples that did not reach failure, effective stress path UT-01 reached failure at relatively less accumulated strain and
did not touch the failure envelope. number of cycles compared to test DT-04.
Fig. 4 presents the results of the undrained tests UT-01, UT-02, Fig. 7 presents the factor of safety 共F兲 versus the number of
and UT-03 in the form of the axial strains 共 % 兲 versus the num- cycles for tests UT-01, UT-02, and DT-04. In this analysis, the
ber of cycles 共N兲. It can be noted from Fig. 4 that in general, the factor of safety is defined as the ratio of the shear strength at a
incremental strain decreases due to an increase of the number of given cycle over the shear strength at failure 共Hyodo et al. 1994兲.
Fig. 4. Cyclic triaxial test results—strain 共e兲 versus number of cycles 共N兲
Downloaded 16 Jul 2009 to 147.83.51.2. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
Fig. 5. Cyclic undrained triaxial test results—cyclic deviator stress 共qcyc兲 versus strain 共e兲 共UT-01兲
Fig. 6. Cyclic drained triaxial test results—cyclic deviator stress 共qcyc兲 versus strain 共e兲 共DT-04兲
Fig. 7. Number of cycles 共N兲 versus factor of safety 共F兲 关after Hyodo et al. 共1994兲兴
Downloaded 16 Jul 2009 to 147.83.51.2. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
Fig. 8. Liquidity index 共IL兲 versus sensitivity number 共St兲
It can be noted from Fig. 7 that the predicted number of cycles at value of the preconsolidation pressure 共c兲, there is a unique re-
failure 共F = 1兲 according to Hyodo et al. compared well with the lationship between the liquidity index 共IL兲 and sensitivity number
number of cycles deduced at failure from the present experimen- 共St兲 and that the preconsolidation pressure 共c兲 is a governing
tal investigation 共Table 3兲. parameter linking the physical and mechanical parameters of sen-
Fig. 8 presents the liquidity index 共IL兲 versus the sensitivity sitive clay. It should be reported herein that a similar trend was
number 共St兲, deduced from the present experimental investigation obtained for other values of the coefficient k.
as well as the data available in the literature for values of the
coefficient k 关Eq. 共2兲兴 equal to 1, 2, 3, and 4. It can be noted from
Fig. 8 that the sensitivity number increases due to an increase of Design Procedure
the liquidity index up to an ultimate value, beyond which the
sensitivity number will continue to increase, while the soil will be Considering the complexity of the behavior of sensitive clay, es-
in the liquid state. Furthermore, for a given value of St, the con- pecially, when subjected to cyclic loading, any attempt to develop
stant k decreases due to an increase of IL. A higher value of the mathematical models or empirical formulas may not incorporate
coefficient k indicates that the clay will be highly sensitive at all the governing parameters pertinent to field condition. Further-
lower values of the liquidity index. more, while this paper is focused on the physical and the me-
The scatters of the test data shown in Fig. 8 can be explained chanical parameters of the clay, it is believed that the chemical
by the fact that these data were produced for a wide range of the and environmental characterization of the site may influence the
preconsolidation pressure 共c兲, which is believed to be one of the results obtained by these methods. Accordingly, and in order to
important parameters governing the behavior of sensitive clay. In overcome these anticipated problems, an experimental design
order to qualify this statement, the data given in Fig. 8 were procedure is developed to achieve the range of a safe design for
filtered for the various ranges of the preconsolidation pressure foundations built on sensitive clay for a given site condition. The
共c兲 for the case of k = 2 共Fig. 9兲. It can be noted that by narrow- proposed design procedure will lead to determine the allowable
ing the range of the preconsolidation pressure 共c兲, a better agree- load on a foundation on sensitive clay during the undrained and
ment between the liquidity index 共IL兲 and the sensitivity number drained responses for the given site and loading conditions. The
共St兲 was achieved. Thus, it can be clearly stated that, for a given proposed design procedure is described below:
Fig. 9. Undrained shear strength versus liquidity index for different values of the preconsolidation pressure 共k = 2兲
Downloaded 16 Jul 2009 to 147.83.51.2. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
Fig. 10. Schematic sketch—safe zones for foundation on sensitive clay
1. Collect undisturbed representative samples from the pro- structure system, cost, and function of the project under
posed site. Every effort should be made to preserve the struc- consideration.
ture and the water content of these samples. 11. In the case of unfavorable conditions, efforts should be made
2. Determine the water content, liquid limit, and plastic limit of to reduce the duration of the undrained response, which rep-
the samples collected. Calculate the liquidity index 共IL兲 of resents the critical period in the lifespan of the foundation,
the clay using Eq. 共1兲. using soil improvement techniques, and/or to reduce the
3. Conduct unconfined compression tests on undisturbed and number of cycles during the undrained response 共if possible兲,
remolded samples. Determine the sensitivity number 共St兲 and/or to reduce the cyclic deviator stress.
using Eq. 共3兲. 12. This procedure may also be followed to examine the condi-
4. Conduct consolidation tests to determine the preconsolida-
tion for existing foundations built on sensitive clay. This can
tion pressure 共c兲 of this clay. Estimate the duration of the
be achieved by locating the point representing the existing
undrained period for the given soil/structure conditions using
soil/foundation conditions on a chart similar to Fig. 10. A
Terzaghi’s method.
point located on the safe zone will indicate a stable founda-
5. Conduct consolidated undrained static triaxial tests on these
samples. Determine the static deviator stress at failure. tion 共quasielastic resilient state兲; otherwise, the foundation
6. Conduct at least three consolidated undrained cyclic triaxial could be on the verge of collapse.
tests on these samples. In these tests, samples are consoli- The procedure described above was used to develop the safe
dated under the confining pressure 共3兲 values, equivalent to zones for the Champlain clay tested in this investigation. Fig. 11
the field preconsolidation pressure 共c兲, which was deter- presents the resulting safe/unsafe zones for both the undrained
mined in step No. 4 above. In each test, assign a value to the and drained conditions together with the test results. Due to the
cyclic stress ratio 共qcyc / q兲 within the range of 0.2 to 0.8, and limited test results for the drained condition, the envelope was
then report the number of cycles needed for that sample to estimated for demonstration purposes.
reach failure. In these tests, duplicate the expected frequency
of load on the foundations.
7. The result of each test will produce a point on the failure Conclusions
envelope 共Fig. 10兲. This failure envelope will identify the
boundary of the safe/unsafe zones for foundations during the
Based on the results of the present experimental investigation
undrained response.
and the data available in the literature, the following can be
8. Repeat step No. 6 but for consolidated drained cyclic test.
concluded:
The results of these tests will produce points on the failure
envelope that identify the limit of the safe/unsafe zones for 1. Cyclic loading on sensitive clay causes remolding action that
foundations during the drained response 共Fig. 10兲. facilitates the dissolution of the bonding agents between par-
9. Fig. 10 can be used as design charts for foundation design on ticles by the pore water. This leads to the destruction of soil
the clay obtained from the region where samples were ob- structure and further, the loss of shear strength and perhaps
tained. Specifically, the designer will be able to determine the liquefaction of the soil. Accordingly, foundations built on
allowable cyclic deviator stress 共qcyc兲 on a foundation, which this clay would suffer extensive settlement and significant
corresponds to the number of cycles expected on this foun- loss of bearing capacity, or perhaps catastrophic failure.
dation during the undrained or the drained responses. 2. The parameters governing the behavior of sensitive clay can
10. A reasonable factor of safety should be applied to determine be categorized as physical and mechanical parameters. Fur-
the allowable cyclic deviator stress 共qcyc兲, depending on the thermore, the drainage condition during load cycle 共und-
Downloaded 16 Jul 2009 to 147.83.51.2. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
Fig. 11. Deduced safe zones for Champlain clay
rained or drained兲 is considered as a major parameter in personalized for the site, where sensitive clay will be ex-
examining this behavior. tracted, and accordingly it will duplicate the chemical and
3. The physical parameters include: Natural water content 共w兲, the environmental conditions of the site. Furthermore, it
liquid limit 共LL兲, and plastic limit 共PL兲, or the liquidity index treats undrained and drained conditions separately, and
共IL兲. The mechanical parameters include the sensitivity num- evaluates the physical and the mechanical properties of the
ber 共St兲, number of cycles 共N兲, cyclic deviator stress 共qcyc兲, soils. The proposed procedure is mainly experimental, simple
preconsolidation pressure 共c兲, and confining pressure 共3兲. to follow, and utilizes classic soil mechanics laboratories.
4. For a given sensitivity number 共St兲, the shear strength of the Although this design procedure was developed and validated
clay decreases due to an increase of the number of cycles for Champlain clay, it can be equally developed for any re-
共N兲. This rate increases due to an increase of the sensitivity gion around the word where sensitive clay can be found.
number 共St兲 and/or cyclic deviator stress 共qcyc兲 and/or liquid- 10. The proposed design procedure identifies the range of the
ity index 共IL兲. i.e., the higher the St, qcyc, and IL, the lower the cyclic deviator stress ratio 共qcyc兲 for a given number of cycles
number of load cycles 共N兲 needed to reach failure. Further- 共N兲 to achieve a quasielastic resilient state without reaching
more, for relatively lower values of the cyclic deviator stress failure. A safe zone is defined for each of the undrained and
ratio 共qcyc兲, the clay may reach a quasielastic resilient state drained conditions.
without failing. 11. The proposed design procedure could also be followed to
5. The preconsolidation pressure 共c兲 is considered a major examine whether an existing foundation built on sensitive
governing parameter in the case of foundations on sensitive clay is within the safe zone or on the verge of failure.
clay, and is regarded as the link between the physical and
mechanical parameters that govern the shear strength of sen-
sitive clay subjected to cyclic loading.
Acknowledgments
6. The sensitivity number of the clay 共St兲 increases due to the
increase of the liquidity index 共IL兲 up to a limit, beyond
The financial support from the Natural Science and Engineering
which, the clay will liquefy. The relationship between the
Research Council of Canada 共NSERC兲 and Concordia University
liquidity index 共IL兲 and the sensitivity number 共St兲 for sensi-
are acknowledged.
tive clay is unique for a given value of the preconsolidation
pressure 共c兲.
7. During the undrained period, the shear strength of the sensi-
tive clay is significantly reduced, due to the rapid increase of Notation
the pore water pressure, which takes place during this period.
Accordingly, samples reached failure at a relatively low The following symbols are used in this paper:
number of load cycles. Thus, the undrained period is re- Cu ⫽ undrained undisturbed shear strength;
garded as the critical period in the lifespan of foundations Cur ⫽ undrained remolded shear strength;
built on sensitive clay and subjected to cyclic loading. F ⫽ factor of safety;
8. During the drained response 共or partially drained兲, the con- G ⫽ specific gravity;
solidation of the clay, which takes place during this period, IL ⫽ liquidity index;
will be accompanied by dissipation of the pore water pres- I p ⫽ plasticity index;
sure and accordingly, the increase of the shear strength. This k ⫽ constant, describing the variation of sensitivity
increase may offset 共in part or in full兲 the reduction in the number;
shear strength caused by load cycling. Nevertheless, during LL ⫽ liquid limit;
this period, foundations may also reach failure at a relatively N ⫽ number of cycles;
higher number of cycles 共N兲 in a relatively short period. PL ⫽ plastic limit;
9. A design procedure for foundations on sensitive clay and q ⫽ deviator stress 共1 − 3兲;
subjected to cyclic loading is presented. The procedure is qcyc ⫽ cyclic deviator stress;
Downloaded 16 Jul 2009 to 147.83.51.2. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
qs ⫽ static deviator stress; Hyodo, M., Yamamoto, Y., and Sugiyama, M. 共1994兲 “Undrained cyclic
S ⫽ degree of saturation; shear behavior of normally consolidated clay subjected to initial static
St ⫽ sensitivity number; shear stress.” Soils Found., 34共4兲, 1–11.
u ⫽ pore water pressure; Iwasaki, T., Tatsuoka, F., and Takagi, Y. 共1978兲. “Shear modulus of sands
under cyclic torsional shear loading.” Soils Found., 18共1兲, 39–50.
wc ⫽ natural water content;
Lefebvre, G., and LeBoeuf, D. 共1987兲. “Rate effects and cyclic loading of
⌬u ⫽ change in pore water pressure; sensitive clays.” J. Geotech. Engrg., 113共5兲, 476–489.
⫽ axial strain; Lefebvre, G., and Pfendler, P. 共1996兲. “Strain rate and pre-shear effects in
⫽ effective stress; cyclic resistance of soft clay.” J. Geotech. Engrg., 122共1兲, 21–26.
c ⫽ preconsolidation pressure; Liang, R. Y., and Ma, F. 共1992兲. “Anisotropic plasticity model for und-
o ⫽ overburden pressure; rained cyclic behavior of clays. I: Theory.” J. Geotech. Engrg.,
1 ⫽ major principal stress; and 118共2兲, 229–245.
3 ⫽ confining pressure. Miller, G. A., Teh, S. Y., Li, D., and Zaman, M. M. 共2000兲. “Cyclic shear
strength of soft railroad subgrade.” J. Geotech. Geoenviron. Eng.,
126共2兲, 139–147.
Procter, D. C., and Khaffaf, J. H. 共1984兲. “Weakening of undrained satu-
References rated clays under cyclic loading.” J. Geotech. Engrg., 110共10兲, 1431–
1445.
Ansal, M., and Erken, A. 共1989兲. “Undrained behavior of clay under Quigley, R. M. 共1980兲. “Geology, mineralogy and geochemistry of Ca-
cyclic shear stresses.” J. Geotech. Engrg., 115共7兲, 968–983. nadian soft soils: Geotechnical perspective.” Can. Geotech. J., 17,
Bjerrum, L. 共1954兲. “Geotechnical properties of Norwegian marine 261–285.
clays.” Geotechnique, 4共2兲, 49–69. Sangrey, D. A., Henckel, D. J., and Esrig, M. L. 共1969兲. “The effective
Chagnon, J. Y., Lebuis, J., Allard, J. D., and Robert, J. M. 共1979兲. “Sen- stress response of a saturated clay soil to repeated loading.” Can.
sitive clays, unstable slopes, corrective works and slides in the Que- Geotech. J., 6共3兲, 241–252.
bec and Shawinigan area.” Geological Association of Canada. Seed, H. B., and Idriss, I. M. 共1982兲. “Ground motions and soil liquefac-
Eekelen, H. A. M., and Potts, D. M. 共1978兲. “The behavior of Drammen tion during earthquakes.” Earthquake Engineering Research Institute,
clay under cyclic loading.” Geotechnique, 28共2兲, 173–196. Oakland, Calif., Monograph Series.
Downloaded 16 Jul 2009 to 147.83.51.2. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright