Construction Effect on Load Transfer along Bored Piles

Ming-Fang Chang, P.E., M.ASCE,1 and Hong Zhu2

Abstract: The load transfer behavior along bored piles is affected by details of pile construction particularly those imposing stress and
moisture changes to the surrounding soils. An investigation involving moisture migration tests, in situ horizontal stress measurements, and
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

borehole shear and pressuremeter tests shows clear effects of construction that lead to subsequent changes in soil properties. The
construction of bored piles in Singapore and the region often involves casting of concrete either in unsupported ‘‘dry’’ boreholes or in
‘‘wet’’ boreholes filled with water. It is necessary to differentiate these two extreme construction conditions in bored pile design. Based on
triaxial compression and pressuremeter tests on the residual soil of the Jurong Formation in Singapore, the variation of soil modulus with
shear strain can be described by a hyperbolic function. A procedure is recommended for assessing the combined effect of stress relief and
soaking on soil modulus by introducing a modulus reduction factor. Modulus degradation curves from pressuremeter tests with the
borehole conditions properly simulated are found capable of producing load transfer curves that are comparable to those deduced in the
field.
DOI: 10.1061/共ASCE兲1090-0241共2004兲130:4共426兲
CE Database subject headings: Load transfer; Bored piles; In situ tests; Degradation; Shaft resistance, piles.

Introduction holes may remain unsupported and stay ‘‘dry.’’ In other cases
involving high water table and pervious strata, the borehole may
Large diameter bored piles, also known as drilled shafts, are com- become ‘‘wet’’ because of inflow of water and stabilization of
monly used for the support of heavy loads in stiff to hard clay boreholes by means of steel casing or drilling fluids such as ben-
worldwide and in intermediate geomaterials such as residual soils tonite slurry is necessary. Very seldom is drilling fluid introduced
and weathered rocks in tropical regions. Traditional methods of during excavation to assist the drilling process from the begin-
design based on static formulas and soil parameters derived from ning.
laboratory tests, or unit shaft and base resistance estimated from Casting of concrete is carried out by direct pouring when the
results of in situ tests, such as the penetration resistance, N, from borehole is ‘‘dry’’ or by the tremie method when the borehole is
the standard penetration test, have not been very successful. An ‘‘wet.’’ Usually, high slump concrete mixes are used, and/or suit-
improved design method is possible by making use of the load– able additives added, to avoid problems that may affect pile in-
settlement relationship predicted from site-specific load transfer
tegrity.
curves using the load–transfer analysis 共e.g., Reese 1978; Reese
One unique practice among the piling contractors in Singapore
and O’Neill 1988; Chang and Goh 1989; Chang and Broms
and Malaysia is to pour water into the borehole after inflow of
1991兲. This method of design relies on careful evaluation of
water is identified, usually close to the bottom of the borehole or
load–transfer curves from site investigation data and a proper
after the completion of excavation, and the borehole becomes
account of construction details.
In Singapore, Malaysia, and other countries in the region, the ‘‘wet.’’ In this case, the borehole could be soaked in water before
construction of bored piles in residual soils and other intermediate the arrival of ready-mix concrete for a varied period of time. The
geomaterials 共O’Neill and Reese 1999兲 typically involves exca- effect of ‘‘wet’’ construction may not always be critical, particu-
vation of a borehole and casting of concrete in the hole. Because larly if proper drilling slurry is used. Experience from borehole
intermediate geomaterials are generally stiff to hard and cohesive, excavation using mineral slurry 共or polymer fluids in recent years兲
a rotary rig equipped with a short-flight auger is usually employed in the United States has indicated that, in clayey soils, the load
in excavation without the introduction of drilling fluid. In most transfer developed along the shafts are comparable to those de-
cases where the ground is stable and the water table is low, bore- veloped in shafts constructed in the ‘‘dry’’ 共Reese and Touma
1972兲. Nevertheless, soaking of borehole under water after
1
Associate Professor, School of Civil & Environmental Engineering, completion or near completion of excavation that is pertaining to
Nanyang Technological Univ., Block N1, Nanyang Ave., Singapore the local practice represents an extreme condition that will lead to
639798. E-mail: [email protected] a lower bound to the load transfer expected in various ‘‘wet’’
2
Geotechnical Engineer, Land Transport Authority, 1 Hampshire Rd., construction methods and deserves a review and careful investi-
Singapore 219428. E-mail: hongគ[email protected] gation.
Note. Discussion open until September 1, 2004. Separate discussions Yong 共1979兲, by studying model piles 共38 mm in diameter兲
must be submitted for individual papers. To extend the closing date by formed in kaolin, reported a 10% reduction of unit shaft resis-
one month, a written request must be filed with the ASCE Managing
tance for a 12 h delay and 20% for a 27 h delay, compared with
Editor. The manuscript for this paper was submitted for review and pos-
sible publication on October 7, 2002; approved on August 26, 2003. This that with insignificant delay before concreting.
paper is part of the Journal of Geotechnical and Geoenvironmental Balakrishnan et al. 共1999兲 reported results of a study on a large
Engineering, Vol. 130, No. 4, April 1, 2004. ©ASCE, ISSN 1090-0241/ number of instrumented bored piles, some constructed by ‘‘dry’’
2004/4-426 – 437/$18.00. excavation and others by ‘‘wet’’ excavation. These piles were

426 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

 installed in the residual soils and weathered rocks of the sedimen- crease in the surrounding soil will lead to deterioration in soil
tary Kenney Hill Formation of Malaysia, which is geologically stiffness and soil strength. Further migration of moisture and
related to the Jurong Formation of Singapore. The ground consists stress changes will occur during and right after casting from the
predominantly of sandy or clayey silt with N ranging from 11 to action of wet concrete and subsequent pozzalanic effect cum hy-
over 100 blows/0.3 m. Two distinctive groups of t – z data were dration of cement during curing. These physical changes associ-
found among these bored piles excavated using short flight augers ated with the various phases of construction will affect the de-
and concreted in ‘‘dry’’ and ‘‘wet’’ boreholes. Balakrishnan et al. tailed load transfer mechanism and the behavior of bored piles
共1999兲 found that the borehole condition appeared to have a sig- and deserve a brief review.
nificant effect on the shape of the load transfer curve, but not the
limit unit shaft resistance f su . A closer examination of the data for Changes from Boring
two specific sites where both ‘‘dry’’ and ‘‘wet’’ excavation bore- From borehole excavation, a relatively narrow zone of soil sur-
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

holes were present, however, reveals a rather consistent reduction rounding the borehole will undergo remolding. The size of the
of the f su /N ratio of typically 10– 60% from wetting. remolded zone varies with the drilling technique and the soil
Yu 共2000兲 load tested eight numbers of miniature bored piles properties. Lord 共1989兲 found a remolded zone about 20–30 mm
on the Nanyang Technological University 共NTU兲 in Singapore. thick along bored piles in slightly to highly weathered chalk.
The site located on the top of hill cut is underlain by stiff, brown- Marsland and Randolph 共1977兲 observed a drop of E s /s u
ish silty sand 关liquid limit 共LL兲⫽32; plastic index 共PI兲⫽6兴 with a value (E s ⫽equivalent linear elastic modulus; s u ⫽undrained shear
fines content 共⫺75 ␮m兲 of 40% and a natural moisture content w 0 strength兲 from 500 a short time after excavation to 100–200 after
of 19.0% in the upper 0.7–1.0 m and yellowish and brownish a delay of over 8 h at the bottom of pile shafts in London Clay.
silty clay 共LL⫽34; PI⫽15兲 with a fines content of 60% and w 0 of O’Neill 共2001兲 reported that the effect of the stress relief could
17.5% immediately blow the top layer. These piles, with a diam- extend to 2–3 borehole radii away from a drilled shaft in Beau-
eter of 0.10 m and lengths ranging from 1.0 to 2.0 m, were in- mont Clay, and the shear wave velocities in the direction perpen-
stalled by using a hand-operated power auger. A number of piles dicular to the pile shaft at the concrete–soil interface were only
were cast by direct pouring of cement mortar 共water: sand⫽1:3兲 about 70% of those in the free field approximately 3 h after the
with a water/cement (w/c) ratio of 0.5 after completion of drill- borehole was open.
ing in ‘‘dry’’ holes, and others in ‘‘wet’’ holes soaked under water
for a specific period of time. From pullout tests, Yu 共2000兲 found
that f su decreased with the soaking time for pile sections embed- Changes from Casting
ded in both soil layers. The f su value decreased by 5–10% after Lings et al. 共1994兲 reported that when the borehole was filled
soaking for 0.5–2 h, and by as much as 20% after 24 h. with concrete with a high w/c ratio, the concrete pressure would
To study the influence of different construction practices on follow the hydrostatic line at shallow depths, but increase at a
pile behavior, a systematic investigation of effect of construction reduced rate at greater depths in diaphragm wall construction. The
details, in particular those related to stress changes associated pressure distribution against a circular borehole in pile construc-
with delay in construction and wetting from soaking of boreholes tion, however, could be different due to differences in geometry
in water, on the load transfer along bored piles is important. In- and arching effects. Indeed, O’Neill and Hassan 共1994兲 reported
vestigations by means of in situ soil tests, laboratory moisture that the total lateral concrete pressure exerting on the borehole
migration and shear tests, and monitoring of stress changes during face at the time of concrete placement, in the case of drilled
a bored pile construction will be useful. Of great importance will shafts, was probably nearly equal to the normal radial interface
be a close simulation of two typical extreme construction condi- stress at the time of loading.
tions involving ‘‘dry’’ and ‘‘wet’’ boreholes. Results of these in- Yong 共1979兲 observed a change in moisture content in kaolin
vestigations should help to verify if the influence of construction of up to 60 mm from the pile–soil interface, with the maximum
on load transfer that have been previously observed is directly increase measured a few days after concreting. Chuang and Reese
relevant for bored piles in intermediate geomaterials or in weath- 共1969兲, based on tests on remolded Beaumont clay, found that w 0
ered rock Formations in the tropics. in the soil was important, but the w/c ratio was a key factor in
The focus of this investigation is on construction effect, par- water migration. Clayton and Milititsky’s 共1983兲 study in re-
ticularly the combination of stress and moisture changes from molded London clay showed that the moisture content in the vi-
delayed construction and wetting from direct contact of borehole cinity of concrete depended largely on the w/c ratio.
with water, on the load transfer characteristics along bored piles. Beech and Kulhawy 共1987兲 observed the dependency of mois-
ture migration on w 0 relative to the LL and the plastic limit 共PL兲,
as well as the plasticity index 共PI兲 of the soil. The magnitude of
Stress and Moisture Changes in Surrounding Soils the water content change increased as PI increased and when w 0
is close to PL.
The load transfer along a pile depends directly on the prevailing In the field, Meyerhof and Murdock 共1953兲 observed that, in
horizontal effective stress, the soil shear characteristics and the London clay 共liquidity index LI⬇0.04兲, the water content was
roughness of the pile-soil interface. The interface roughness, 2–7% higher than w 0 at the shaft face of a bored pile and gradu-
which depends on the construction machinery and techniques, ally decreased over a distance of 51 mm. However, O’Neill and
usually does not vary significantly from one pile to another on a Reese 共1970兲 observed no definite trend in the vicinity of two
given site 共O’Neill 2001兲. The effective horizontal stress and the shafts in Beaumont clay.
soil shear characteristics are controlled by the in situ stress con-
dition in the ground and the subsequent changes from pile con-
Changes from Curing
struction details, such as time taken for boring and duration of
borehole opening, and the presence of water in the borehole prior The concrete pressure exerted on the borehole wall will decrease
to concreting. If the borehole is soaked in water, moisture in- as the concrete starts to set, resulting in the redistribution of

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004 / 427

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

 stresses around the bored piles. The magnitude of the horizontal
stress at the final stage will depend on the interaction between
concrete and the soil in contact. In addition, temperature change
due to heat of hydration and pozzalanic action may also have an
effect.
Field measurements on a 4.5 m long diaphragm wall con-
structed in a stiff soil by Uff 共1969兲 showed a pressure reduction
of about 30% 8 weeks after concreting. Uriel and Oteo 共1977兲
observed in the construction of a 3.4 m long wall panel an aver-
age reduction of 20% 24 h after casting. Lings et al. 共1994兲 re-
ported an average reduction of 25% on an 8.5 m long diaphragm
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

wall in stiff to very stiff Gault clay 10 days after concreting.

It should be noted that the stress at the pile–soil interface can
decrease or increase depending on the details of concrete mix, the
suction potential in the surrounding soil, the temperature of the
ambient soil, and the rate of the placement of concrete. The ob-
served pressure often varies and an independent verification by
direct investigation adjacent to a pile shaft will be useful.

Laboratory Investigation of Moisture Migration

A series of tests were conducted in the laboratory to study the
effect of w i in the soil and the w/c ratio of the cement mortar on
the magnitude and pattern of moisture migration in the residual
soil of the Jurong Formation. The soil samples were taken from a
bored pile construction site on NTU campus in Singapore. The
soil 共LL⫽36.3, PL⫽19.5; w 0 ⫽18.9%), with a fines content of
61.4%, is classified as silty clay.
The soil was first air dried and sieved through U.S. No. 4 共2
mm opening兲 to remove the gravel size particles. A 100 mm high
sample was prepared by manually compacting the soil layer by
layer using a hand-held Standard Proctor hammer into a 200 mm
long plastic pipe with an internal diameter of 95 mm. The sample Fig. 1. Influence of initial moisture content and water/cement ratio
were compacted until the specified height was reached and the on moisture distribution of compacted residual soil samples
resulting bulk density was 2.1 Mg/m3. Mortar with a designated
w/c ratio was then poured on the top of the sample. The two ends
of the pipe were immediately wrapped with cling film and alumi- within which a significant change in w was observed is seen to
num foil and dipped in wax. The specimen was weighed and extend to around 20–30 mm from the interface, about half of that
stored in a humidity controlled room. After 7 days, sampling was observed by Yong 共1979兲 in the more pervious Kaolin and by
then carried out using a 38 mm diameter ‘‘mini’’ sampler. The Meyerhof and Murdock 共1953兲 in the often fissured London clay.
sample was subsequently sliced and the water content of each Fig. 2 shows the average change in water content within first
slice evaluated. 20 mm from mortar plotted as a function of LI of the soil. It
Samples with three different combinations of w/c ratios of 0.5, shows clearly that the demand of soil for water is stronger as LI
0.6, and 0.7 and w i values of 14.1, 17.0, and 22.2% were prepared becomes smaller, and consequently more water is absorbed by the
in the tests. The corresponding degrees of saturation of the com- soil, similar to that observed by Beech and Kulhawy 共1987兲. Note
pacted soil samples were 82.8, 92.4, and near 100%, respectively. that the final moisture distribution in the soil depends not only on
The suction pressures based on the average soil water character- the suction or demand of the soil for water, but also on that of the
istic curve for the residual soil under investigation 共Leong et al.
2002兲 were estimated at 60 kPa, 10 kPa, and close to 0, respec-
tively.
Fig. 1 shows the distribution of water content, w, in soil
samples in contact with mortar of different w/c ratios. A variation
of w with the distance from the soil–mortar interface is clearly
observed for all tests with different w i and w/c ratios. Except for
the combination of w/c of 0.5 and w i of 22.2% for which there
was a consistent decrease in w throughout the soil sample, an
increase in w is generally observed. The higher the w/c ratio, the
more abundant is the water supply and the greater is the amount
of increase in w. Similarly, the lower the w i , the greater is the
suction and the greater is the increase in w.
The maximum observed increase in w is typically 3% at the
soil/mortar interface, similar to that reported by Clayton and Mili-
Fig. 2. Influence of liquidity index on moisture migration
titsky 共1983兲 for a mortar with w/c⫽0.3. The zone of influence

428 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 3. Penetration response profiles in the compacted fill: 共a兲 from dilatometer test and 共b兲 from cone penetration test

mortar. The suction balance between soil and mortar may change was taken. The measured ( p 0 ) i was 180 kPa from DMT-1, similar
with time, leading to further changes in w near the interface and to that in DMT-3; but it was at a much higher value of 340 kPa
consequently the interface shear characteristics. from DMT-2, possibly due to the heterogeneous nature of the
compacted fill. The measured ( p 0 ) i corresponded to an earth pres-
sure coefficient K of around 2 at DMT-1 level and 1.2 at DMT-3
Field Observation of Changes in Horizontal Stress level, withstanding the disturbance induced to the ground in the
vicinity of the dilatometer blade.
Field monitoring of changes in horizontal stress in the ground
adjacent to a pile bore during the construction of a bored pile
共diameter⫽0.8 m, length⫽30 m兲 was made at a piling site on the
NTU campus. The upper part of the ground consisted of over 8 m
of compacted residual soil fill with its water content similar to w 0
in the Jurong Formation that immediately underlies the com-
pacted fill.
Prior to excavation, one DMT, using a standard dilatometer
blade, and one CPT, using a mechanical cone, were carried out.
The DMT results are presented as profiles of p 0 共membrane lift-
off pressure兲 and p 1 共pressure that corresponds to a membrane
displacement of 1.0 mm at the center兲, as shown in Fig. 3. The
compacted soil can be classified as sandy silt according to Mar-
chetti 共1980兲 based on the material index, I d , calculated from p 0
and p 1 pressures. The cone penetration resistance, q c , from the
CPT, also shown in Fig. 3, was found to fluctuate between 5 and
10 MPa. The friction ratio is typically between 3 and 5%, also
indicating that the soil is rather silty.
Three flat dilatometers, arranged in a row on one side of the
pile, were installed at 0.5, 0.9, and 1.7 m away from the borehole
wall 5 days prior to pile construction and left in place. Their
respective depths of embedment were 4.5 m 共DMT-1兲, 6.0 m
共DMT-2兲, and 8.0 m 共DMT-3兲 within the compacted fill.
During excavation by augering, some localized collapses of
the borehole wall were observed. In an attempt by the piling
contractor to prevent further collapse of the borehole wall, the
borehole was filled with water one hour after completion of ex-
cavation. Casting of concrete (w/c⫽0.55) was subsequently con-
ducted by the tremie method after over 4 h of delay. Dilatometer
p 0 pressures were taken at regular time intervals during the pile
construction.
Fig. 4 shows the observed variations of p 0 pressure, normal-
ized by the initial p 0 pressure, (p 0 ) i , prior to excavation, with
Fig. 4. Measured normalized lateral stress during pile construction
time from DMT-1 and DMT-2. For unknown reasons, DMT-3,
from 共a兲 dilatometer-1 and 共b兲 dilatometer-2
registered a (p 0 ) i of 190 kPa, did not work after the initial reading

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004 / 429

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 5. Soil stratification and general soil properties at Nanyang Technological University test site

The change in p 0 pressure clearly shows that borehole exca- A borehole shear test apparatus 共Handy 1981兲 was used for a
vation led to a significant reduction in horizontal stress in the soil rapid, qualitative assessment of the friction angle and the cohe-
around the borehole. The measured decreases in horizontal stress sion intercept of soils. In the BST, a set of shear plates with two
at the end of excavation were about 23 and 15% of ( p 0 ) i at 0.5 sharp 30° half-wedge teeth at 25.4 mm apart, providing an
and 0.9 m away from the excavation face, respectively. These equivalent shear surface of 645 mm2, was adopted to ensure suf-
reductions increased to 78 and 18% after the borehole was left ficient penetration in the stiff residual soils 共Yu 2000兲. The test
open for 1 h, prior to pouring of water. Immediately after the involved applying a normal stress against the borehole wall and
borehole was filled with water, there was a partial recovery in p 0 measuring the maximum pulling force that is required to fully
and the reductions dropped to 60 and 15%, respectively. Interest- mobilize resistance along the shear head during pullout. The mea-
ingly, there was a further recovery in the p 0 pressure after the surements were repeated at the same depth for typically a mini-
borehole was soaked for over 3 h possibly due to moisture mum of three normal stresses at three different orientations ap-
changes in the soil adjacent to the membrane, and the reductions proximately 120° apart. Two series of BSTs, one in freshly
were maintained at 55 and 11%, respectively, just prior to casting. prepared boreholes and the other in boreholes drilled and soaked
The measured p 0 pressures from DMT-1 and DMT-2 at the for specified periods of time, were conducted. Both the time for
completion of casting by tremie using 175 mm high slump con- consolidation of soil after normal stress application and that for
crete recovered to 67 and 94% of their respective original values. the pull out were so selected that the test could be assumed
The horizontal stress in the compacted soil near the pile shaft did drained. Typically, the consolidation time was about 5 min, and
not return completely to the original values even 19 days after the pulling rate was 1 mm/min.
construction. The final equilibrium horizontal stress at 0.5 and 0.9 A single cell Japanese 共OYO兲 type pressuremeter was used in
m from the excavation face were 80 and 94% of ( p 0 ) i , respec- the PMT. The probe, made of rubber membrane, was lowered
tively. The common practice of assuming K⫽K 0 共the coefficient down to the test depth in a borehole carefully prepared using the
of earth pressure at rest兲 for bored and cast-in-place piles cannot rotary wash boring technique. The membrane was then expanded
be verified in this case. Neither is the similar extent of horizontal by means of pressurized gas and the corresponding radial dis-
stress reduction after casting in diaphragm wall construction as placement was measured using two callipers linked to a displace-
reported by others observed here. ment transducer. Through step-by-step applications of pressure,
an expansion curve showing the relationship between the applied
pressure and the radius of the borehole was obtained. Two com-
Field Investigation of Soaking Effect on Soil parative series of PMTs were conducted in two 70 mm in diam-
Characteristics eter boreholes. In the first series, the expansion test was con-
ducted immediately after the borehole reached a predetermined
An investigation was carried out at the NTU test site to see the depth, as it is done in practice, and was denoted as the ‘‘un-
effect of soaking of boreholes on the properties of the residual soaked’’ test. In the second series, the probe was kept in the
soil in Jurong Formation. Fig. 5 shows soil stratification and the borehole for a certain period of time after installation, with the
profiles of selected soil properties at the site. The groundwater borehole soaked in water before the expansion of membrane, and
table was found at around 19 m below the ground surface and the was denoted as the ‘‘soaked’’ test.
soil above the water table was generally unsaturated with w 0 Table 1 summarizes the shear strength parameters, c ⬘ and ␾⬘,
⬍PL. Two types of in situ tests, the borehole shear test 共BST兲 and deduced from BSTs for the two distinctive soils, namely the
the prebored pressuremeter test 共PMT兲, were used in the investi- sandy silt and the silty clay, present at the test site. Both soils are
gation. cohesive and with LI⬍0. With few exceptions, the decrease in c ⬘

430 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

 Table 1. Summary of Shear Strength Parameters from Borehole Shear Tests
Cohesion c ⬘ 共kPa兲 Friction angle ␾⬘ 共deg兲
Soaking time 共h兲 Soaking time 共h兲
Depth
Soil type 共m兲 0 1 2 3 4 5 0 1 2 3 4 5
Sandy silt 3.5 12.8 8.8 — — 6.2 — 36.2 34.3 — — 31.4 —
8.0 14.9 — 5.6 — — — 29.8a — 34.4 — — —
12.0 10.6 5.4 2.0 — — — 24.0a 42.8 45.0 — — —
13.5 13.9 — 4.6 — 2.8 — 36.7 — 35.0 — 38.6 —
Silty clay 6.0 15.3 — 8.6 — — — 28.8 — — 26.7 — —
7.0 22.8 — 15.7 — — — 35.6 — 34.4 — — —
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

10.0 26.4 — 9.6 — 6.8 — 42.0 — 36.0 — 35.9 —

16.5 14.0 — 8.7 — — — 37.2 — 30.0 — — —
19.0 21.0 — — — — 12.4 39.6 — — — — 31.0
a
Exceptional values.

and ␾⬘ due to soaking is evident for both soils, particularly for c ⬘ . cent soils from the combined effect of stress relief and moisture
The reduction in c ⬘ after 1– 4 h of soaking is 30– 80% for the migration in the soil within a distance approximately equal the
sandy silt and 30–70% for the silty clay. The reduction in ␾⬘ length of the probe, which is 0.6 m, as a result of soaking.
angle varies from negligible to around 20% for both groups of
soils. Notwithstanding the fact that a translation of these changes
into the change in shaft resistance is not that direct, the soaking Effect of Stress Relief and Soaking on Modulus
effect is evident. Degradation
Fig. 6 shows typical expansion curves for unsoaked and
soaked conditions from PMTs at a depth of 6.0 m. The pres- The stress–strain behavior of natural soils is highly nonlinear and
suremeter shear modulus, G p , was calculated from the slope of the elastic modulus generally decreases with the increase in shear
the initial linear portion based on cavity expansion theory 共e.g., strain. Extensive studies have shown that this degradation of soil
Mair and Wood 1987兲. The limit pressure p L , the pressure re- shear modulus, G, with shear strain, ⑀ s , significantly influences
quired to double the cavity volume, was obtained from the expan- the performance of a foundation system, especially in stiff soils
sion curve or by extrapolation. The net limit pressure, defined as 共e.g., Jardine et al. 1986兲.
p L* ⫽(p L – ␴ v 0 ), where ␴ v 0 is the in situ total vertical stress, can The modulus degradation of a residual soil can be expressed as
be readily calculated. a hyperbolic function as follows 共e.g., Zhu and Chang 2002兲

冉 冊
Figs. 7共a and b兲 show, respectively, the G p 共typically at 1.5% G G ⑀s g
cavity strain兲 and the p L* profiles from PMTs in both ‘‘unsoaked’’ ⫽1⫺ f (1)
and ‘‘soaked’’ boreholes. The soaking process has clearly affected G max G max ⑀ r
G p and p L* and the effect is more significant in the sandy silt than where f and g⫽fitting parameters and ␶ max and G max⫽maximum
in the silty clay. Within the soaking duration of 2–5 h, the reduc- shear stress and maximum shear modulus, repectively; and
tion in G p ranges from 25 to 60% for the sandy silt and 17–50% ⑀ r ⫽reference shear strain defined as ⑀ r ⫽␶ max /Gmax .
for the silty clay. The reduction in p L* ranges from 20 to 60% for In the construction of bored piles, stress relief from excavation
the sandy silt and 20–50% for the silty clay. Note that the obser- followed by soaking of boreholes will affect the shear modulus of
vation based on the PMT represents an average response of adja- the surrounding soils. Both the triaxial compression test with

Fig. 6. Comparison of pressuremeter expansion curves for unsoaked and soaked conditions

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004 / 431

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 7. Profiles from unsoaked and soaked pressuremeter tests: 共a兲 Shear modulus and 共b兲 net limit pressure

small strain measurements in the laboratory and the pressuremeter described by Eq. 共1兲, with values of G max obtained from the ul-
test that incorporated unloading–reloading in the field are useful trasonic impulse velocity measurements, and the appropriate fit-
for investigating such effects in the study. ting parameters f and g. The parameter f appears unaffected by
A series of consolidated undrained 共CU兲 triaxial compression soaking and is at 0.98 for both types of soils. However, the pa-
tests were carried out on both ‘‘natural’’ and ‘‘presaturated’’ speci- rameter g for the ‘‘dry’’ condition is two times of that in the
mens of the silty clay and the sandy silt from the NTU test site ‘‘wet’’ condition for the silty clay, and the corresponding ratio of
共Fig. 5兲. ‘‘Undisturbed’’ samples obtained using a Mazier sampler g is 2.5 for the more pervious sandy silt.
were used. For the ‘‘presaturated’’ tests, the samples were first The PMT is an ideal alternative for investigating the combined
subjected to saturation under a selected back pressure and a con- effect of stress relief and soaking of borehole in water on the soils
fining pressure that produced an effective confinement equal to adjacent to a bored pile, as the relevant borehole conditions can
the initial suction in the sample. The initial suction measured be simulated in a practical manner and the corresponding changes
using the null-type axis-translation technique 共e.g., Fredlund and in shear modulus can be readily assessed. For this application,
Rahardjo 1993兲 were typically 30–50 kPa for the samples. Both unload–reload cycles will need to be incorporated in the expan-
‘‘natural’’ and ‘‘presaturated’’ samples were firstly isotropically sion test so that the reloading curve, which is believed to be
consolidated to an effective pressure, ␴ ⬘c , equivalent to the mean representative of the in situ response of the soil, can be used to
in situ effective stress and subsequently unloaded to a mean ef- derive the modulus degradation curve.
fective stress that equals to 0.5␴ ⬘c to simulate the stress relief due Two series of comparative PMTs were carried out in both the
to a delayed borehole excavation. Subsequently, the sample was silty clay and the sandy silt in two parallel boreholes, one soaked
sheared under undrained condition and the mean effective stress and the other unsoaked, at the NTU test site. Fig. 9 shows the
was maintained at a constant value to simulate the expansion of a PMT modulus degradation data corresponding to both unsoaked
cavity as that in the pressuremeter test. Local displacement or and soaked conditions in the silty clay at a depth of 10 m and in
strain measurements were carried out during shearing using sub- the sandy silt at a depth of 13.5 m, as derived by Zhu and Chang
mersible LVDTs. 共2002兲. Soaking has clearly affected the modulus degradation
It is interesting that the initial water contents, w i , in the ‘‘natu- curves for both soil groups, similar to that observed in the triaxial
ral’’ and the ‘‘presaturated’’ specimens prior to shear were 20.3 compression tests. Fig. 9 also shows the best fitting curves and the
and 22.5%, respectively, for the silty clay and 12.9 and 16.2%, appropriate fitting parameters, with values of G max estimated by
respectively, for the sand silt. This difference of 2– 4% in w i assuming G max /Gur⫽2.5, where G ur is the unload–reload modu-
between the two types of specimens is similar to that observed lus 共Zhu 2000兲. From the best fits, the f parameter is found to be
near the soil–mortar interface in the moisture migration tests dis- at a constant value of 0.98, the same as that observed in the
cussed earlier. triaxial CU tests. The g parameters from the PMTs, however, are
Fig. 8 shows the variation of the shear modulus, G, normalized consistently lower, at 40– 60% of the triaxial values. It is believed
by G max with the shear strain, ⑀ s , for the two soils subjected to differences in the simulated stress relief, mass of soil tested, and
different saturation conditions. The reduction in soil modulus is progressive failure are probably responsible for the observed dif-
significant, particularly when ⑀ s ⭐0.1%. Fig. 8 also shows how ference between the degradation curves from the two types of
the modulus degradation data are fitted with hyperbolic curves as tests. Nevertheless, the ratios of g between the ‘‘dry’’ and the

432 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 8. Shear modulus degradation behavior from triaxial compression tests in: 共a兲 Silty clay and 共b兲 sandy silt

‘‘wet’’ boreholes, which is 2.5 for the silty clay and 2.3 for the used to evaluate the combined effect of stress relief and soaking
sandy silt, are similar to those from the triaxial tests. For evalu- on modulus degradation.
ating the combined effect of stress relief and soaking of borehole Fig. 10 shows the variation of f G with stress level for different
on pile behavior, the PMT is preferred due to its closer simulation ratios of g 2 /g 1 . A single value of f 1 and f 2 of 0.98 that appeared
of the excavation condition in pile construction, even though to be a characteristic that is typical of the cohesive residual soil of
fewer and often less consistent data points are expected when the Jurong Formation, was assumed. The modulus reduction fac-
compared to the CU triaxial test. tor f G is seen to increase as the g 2 /g 1 ratio increases. The reduc-
It should be noted, by proper selection of different sets of tion factor f G which is relatively independent of the stress ratio is
fitting parameters of ( f 2 ,g 2 ) and ( f 1 ,g 1 ), respectively, Eq. 共1兲 typically between 0.2 and 0.7.
can match ‘‘natural’’ and ‘‘presaturated’’ test data from both the Note that from triaxial compression tests, the g 2 /g 1 value is
triaxial compression tests and the pressuremeter tests well, as il- around 2.0 for the silty clay, implying that saturation would lead
lustrated in Figs. 8 and 9. It appears that the difference in satura- to a modulus reduction of about 40%. Similarly, the correspond-
ing g 2 /g 1 value of 2.5 for the sandy silt implies that the modulus
tion condition leads to a significant reduction in parameter g that
reduction would be around 50%, which is slightly larger that ob-
controls the shape of the modulus degradation, but not in param-
served in the sitly clay. Similar extents of reduction in shear
eter f that signifies the magnitude of degradation in the Jurong
modulus were also observed in the PMTs based on the observed
Formation.
g 2 /g 1 ratios of between 2.3 and 2.5 for two different soils, as well
From Eq. 共1兲, a factor reflecting modulus reduction can be
as from the profile of G p presented earlier in Fig. 7, although the
deduced as follows: various G values may correspond to different levels of shear
1⫺ f 1 冉 冊
␶
␶ max
g1
strain.

冉 冊
f G ⫽1⫺ (2)
␶ g2
Effect of Stress Relief and Soaking on Load
1⫺ f 2
␶ max Transfer Curves
where f G ⫽factor of modulus reduction; and g 1 and g 2 ⫽g param- Zhu and Chang 共2002兲 proposed a procedure using modulus deg-
eters corresponding to ‘‘presaturated’’ and ‘‘natural’’ borehole radation curves as obtained from PMTs for evaluating load trans-
conditions, respectively. This modulus reduction factor can be fer (t – z) curves along piles, on the basis of Randolph and

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004 / 433

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 9. Modulus degradation in different test conditions from pressuremeter tests in: 共a兲 silty clay; 共b兲 sandy silt

Wroth’s 共1978兲 framework for pile analysis. The procedure was

adopted in this study for investigating the effect of stress relief
and soaking on load transfer along bored piles. The key param-
eters required in the analysis are the pile slenderness ratio L/d
共L⫽length; d⫽diameter兲, the mid-level to pile base modulus ratio
or the inhomogeneity factor ␳, and the degradation curve fitting
parameters f and g.
Field observed load transfer data from a large number of bored
piles, instrumented with strain gauges at different levels, con-
structed in the Kenny Hill Formation in Malaysia report by Bal-
akrishnan et al. 共1999兲 provide a source of comparison for veri-
fying the usefulness of the modulus reduction factor. The bored
piles, five constructed in ‘‘dry’’ excavation and another five in
‘‘wet’’ excavation for which either water or bentonite was added
during drilling, had similar dimensions: L⫽15– 40 m; d
⫽600– 750 mm, and L⫽10– 40 m; d⫽600– 1,200 mm, respec-
tively. Fitting parameters similar to those obtained from PMTs
Fig. 10. Variation of modulus reduction factor with stress level
with borehole conditions simulating significant delay in concret-

434 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 11. Comparison of measured and predicted t – z curves for bored piles in Kenny Hill Formation

ing in the Jurong Formation, which is geologically linked to the propriate modulus reduction factor f G as obtained from Fig. 10
Malaysian rock Formation, were used in the analysis. for a typical g 2 /g 1 value of 2.5 can be applied so that the com-
Fig. 11 shows a direct comparison of the predicted t – z curves bined effect of stress relief and borehole soaking is properly re-
with the field load transfer curves reported by Balakrishnan et al. flected on the load transfer curves. A typical reduction of 20% can
共1999兲. Note that t – z curves in both ‘‘wet’’ and ‘‘dry’’ conditions be applied to the limiting or the critical values of unit shaft resis-
have been normalized by the same z c that corresponds to the tance for bored piles for which the boreholes might be soaked for
‘‘wet’’ condition, where z c is the critical shaft displacement be- over 2 h or more during construction based on results from
yond which the shaft friction mobilizes or increases at a drasti- present study on the residual soils of the Jurong Formation in
cally reduced rate. The range of t – z curves for ‘‘wet’’ construc- Singapore.
tion calculated based on f of 0.98 and g of 0.1 for significant delay
and 0.2 for moderate delay is seen to compare well with the field
t – z data for bored piles actually constructed in ‘‘wet’’ holes. The Conclusions
predicted t – z curves for ‘‘dry’’ construction, which was gener-
ated by assuming that the g factor for ‘‘dry’’ condition was 2.5 An investigation has been carried out on the NTU campus that is
times that for ‘‘wet’’ condition, also matches reasonably well with covered by the residual soil of the sedimentary Jurong Formation
the measured t – z data for piles actually constructed in ‘‘dry’’ in Singapore to study the construction effects, particularly those
holes. It is clear that using the fitting parameters f and g as ob- lead to stress and moisture changes, on load transfer along bored
tained from PMTs, there is a good match between the predicted piles. The main conclusions that can be drawn are as follows:
and the field t – z curves for the bored piles. 1. Laboratory moisture migration observations on compacted
residual soil specimens in contact with cement mortar con-
firmed the dependency of moisture changes from wetting on
Practical Applications the initial moisture content, the plasticity of the soil and the
water/cement ratio in the mortar as observed by others.
For general design applications, the modulus degradation fitting 2. Direct measurements of horizontal stress changes in the
parameters f and g can be deduced from the pressuremeter test ground during the construction of a bored pile in compacted
that incorporates unloading–reloading cycles in a freshly pre- residual soil indicated a drastic reduction of horizontal stress
pared borehole for bored piles that are expected to be constructed after borehole excavation and a gradual recovery of the hori-
in ‘‘dry’’ condition without significant delay. Modulus fitting pa- zontal stress during and after concreting to a level of around
rameter g from triaxial compression tests on samples reconsoli- 80 and 94% of the corresponding initial horizontal stresses,
dated to in-situ mean effective stress should be reduced by half respectively, at 0.5 and 0.9 m away from the borehole wall.
共based on Figs. 8 and 9兲, if PMT results are not available. Zhu and 3. Borehole shear tests and pressuremeter tests carried out in
Chang’s 共2002兲 procedure can be used for the prediction of nor- both ‘‘dry’’ and ‘‘wet’’ boreholes indicated a maximum re-
malized t – z curves. The limiting or critical shaft resistance, f su , duction in measured friction angle of 20% and cohesion in-
and the critical shaft displacement, z c , that are required in the tercept of 80% after soaking of boreholes in water for 1– 4 h.
design can be estimated from correlations presented by Chang The observed reduction in the pressuremeter modulus typi-
and Broms 共1991兲 and Zhu and Chang 共2002兲. A rational pile cally ranged from 17 to 60% and in the net limit pressure
design can then be carried out based on the load transfer method from 20 to 60% after soaking of boreholes for 2–5 h, in
using the predicted t – z curves. comparison with tests in unsoaked boreholes.
For bored piles that need to be constructed in the ‘‘wet’’ con- 4. The variation in stiffness with shear strain as observed in
dition and a significant delay before concreting is likely, an ap- both the consolidated undrained triaxial compression tests

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004 / 435

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

 and the pressuremeter tests which simulated both ‘‘dry’’ and su ⫽ undrained shear strength;
‘‘wet’’ borehole conditions follows a hyperbolic function, al- w/c ⫽ water/cement ratio;
though the fitting parameters are different between the two wi ⫽ initial moisture content;
types of tests possibly due to differences in the simulated w0 ⫽ natural moisture content;
stress relief, mass of soil tested, and progressive failure. z ⫽ shaft displacement;
5. With a proper simulation of the borehole conditions, the zc ⫽ critical shaft displacement;
pressuremeter test which incorporates unload–reload cycles ⑀r ⫽ reference shear strain;
can yield representative modulus degradation curves that are ⑀s ⫽ shear strain;
suitable for the prediction of load transfer curves for bored ␳ ⫽ inhomegeneity factor;
piles constructed in ‘‘dry’’ or ‘‘wet’’ boreholes. The same ␴ ⬘c ⫽ isotropic consolidation pressure in triaxial test;
approach should be applicable to bored piles that are con- ␴ v0 ⫽ in situ vertical total stress;
␶ ⫽
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

structed in other specific borehole conditions such as when shear stress;

bentonite slurry or polymer fluids are used in the stabiliza- ␶ max ⫽ maximum shear stress; and
tion. ␾⬘ ⫽ effective friction angle.
6. A procedure has been developed to account for combined
effect of stress relief and soaking of borehole in water in the
References
prediction of relevant load transfer curves by introducing a
modulus reduction factor. The recommended procedure is Balakrishnan, E. G., Balasubramaniam, A. S., and Noppadol, P.-W.
capable of capturing the influence of two extreme borehole 共1999兲. ‘‘Load deformation analysis of bored piles in residual weath-
conditions of ‘‘dry’’ and ‘‘wet’’ on the load transfer for bored ered formation.’’ J. Geotech. Geoenviron. Eng., 125共2兲, 122–131.
piles constructed in the intermediate geomaterials in Sin- Beech, J. F., and Kulhawy, F. H. 共1987兲. ‘‘Experimental study of the
gapore and Malaysia, and possibly other similar geological undrained uplift behavior of drilled shaft foundation.’’ Rep. No. EL-
formations. 5323, Electric Power Research Institute, Palo Alto, Calif.
Chang, M. F., and Broms, B. B. 共1991兲. ‘‘Design of bored piles in re-
sidual soils based on field performance data.’’ Can. Geotech. J., 28共2兲,
200–209.
Acknowledgments Chang, M. F., and Goh, A. T. C. 共1989兲. ‘‘Design of bored piles consid-
ering load transfer.’’ Geotech. Eng., 20共1兲, 1–18.
The study described herein forms the main part of an applied Chuang, J. W., and Reese, L. C. 共1969兲. ‘‘Studies of shearing resistance
research project funded by the Nanyang Technological University between cement mortar and soil.’’ Research Rep. Prepared for Center
for Highway Research, Univ. of Texas, Austin, Tex.
共NTU兲, Singapore. The writers wish to thank Indeco Consultants
Clayton, C. R. I., and Milititsky, J. 共1983兲. ‘‘Installation effects and the
Pte Ltd and Jackson Piling Pte Ltd. for assistance in the investi- performance of bored piles in stiff clay.’’ Ground Eng., 16共2兲, 17–22.
gative work carried out on the NTU campus. The writers would Fredlund, D. G., and Rahardjo, H. 共1993兲. Soil mechanics for unsaturated
also like to extend their sincere thanks to Mr. Yu Jin for his soils, Wiley, New York.
contributions in certain aspects of the investigations reported in Handy, R. L. 共1981兲. Instructions of borehole shear test, Handy Geotech-
the paper. nical Instrument Co. Inc., Madrid, Iowa.
Jardine, R. J., Potts, D. M., Fourier, A. B., and Burland, J. B. 共1986兲.
‘‘Studies of the influence of nonlinear stress–strain characteristics in
soil–structure interaction.’’ Geotechnique, 36共3兲, 377–396.
Notation Leong, E. C., Rahardjo, H., and Tang, S. K. 共2002兲. ‘‘Characterization
and engineering properties of Singapore residual soils.’’ Proc., Work-
The following notations are used in this paper: shop on Characterization and Engineering Properties of Natural
c ⬘ ⫽ effective cohesion intercept; Soils, Singapore, 1279–1304.
d ⫽ diameter of shaft; Lings, M. L., Ng, C. W. W., and Nash, D. F. T. 共1994兲. ‘‘The lateral
pressure of wet concrete in diaphragm wall panels cast under bento-
E s ⫽ elastic modulus of soil;
nite.’’ Proc., Inst. Civ. Eng., Geotech. Eng., 107, 163–172.
f, g ⫽ fitting parameters for modified hyperbola; Lord, J. A. 共1989兲. ‘‘Design and construction in chalk.’’ Proc., Int. Chalk
f G ⫽ modulus reduction factor; Symposium, Brighton, U.K., 301–325.
f s ⫽ unit friction; Mair, R. J., and Wood, D. M. 共1987兲. Pressuremeter testing: Methods and
f su ⫽ ultimate unit shaft resistance; interpretation, Butterworths, London.
G ⫽ shear modulus of soil; Marchetti, S. 共1980兲. ‘‘In situ tests by flat dilatometer.’’ J. Geotech. Eng.,
G max ⫽ maximum shear modulus of soil; 106共3兲, 299–321.
Marsland, A., and Randolph, M. F. 共1977兲. ‘‘Comparison of the results
G p ⫽ pressuremeter shear modulus;
from pressrememeter tests and large in-situ plate tests in London
I d ⫽ material index from dilatometer test; clay.’’ Geotechnique, 27共2兲, 217–243.
K ⫽ coefficient of earth pressure; Meyerhof, G. G., and Murdock, L. J. 共1953兲. ‘‘An investigation of the
K 0 ⫽ coefficient of earth pressure at rest; bearing capacity of some bored and driven piles in London clay.’’
L ⫽ pile length; Geotechnique, 3共2兲, 267–282.
N ⫽ standard penetration resistance 共blow/0.3 m兲; O’Neill, M. W. 共2001兲. ‘‘Side resistance in piles and drilled shafts.’’ J.
p L ⫽ limit pressure from pressuremeter test; Geotech. Geoenviron. Eng., 127共1兲, 1–16.
O’Neill, M. W., and Hassan, K. M. 共1994兲. ‘‘Drilled shafts: effects of
p L* ⫽ net limit pressure from pressuremeter test;
construction on performance and design criteria.’’ Proc., Int. Conf. on
p 0 ⫽ lift-off pressure in dilatometer test; Design and Construction of Deep Foundation, U.S. Federal Highway
p 1 ⫽ pressure at 1 mm displacement of membrane at Administration, Washington, D.C., Vol. 1, 137–187.
center in dilatometer test; O’Neill, M. W., and Reese, L. C. 共1970兲. ‘‘Behavior of axially loaded
q c ⫽ cone penetration resistance; drilled shafts in Beaumont clay.’’ Research Rep. No. 89-8, Center for

436 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004

J. Geotech. Geoenviron. Eng. 2004.130:426-437.

 Highway Research, Univ. of Texas, Austin, Tex. at Seaforth, Liverpool.’’ Proc., Conf. on In Situ Investigations in Soils
O’Neill, M. W., and Reese, L. C. 共1999兲. ‘‘Drilled shafts: construction and Rocks, London, 229–239.
procedures and design methods.’’ Report No. FHWA-IF-99-025, Fed- Uriel, S., and Oteo, C. S. 共1977兲. ‘‘Stress and strain besides a circular
eral Highway Administration, Washington D.C. trench wall.’’ Proc., 9th Conf. on Soil Mechanics and Foundation
Randolph, M. F., and Wroth, C. P. 共1978兲. ‘‘Analysis of deformation of Engineering, Vol. 1, ISSMFE, Tokyo, 781–788.
vertically loaded piles.’’ J. Geotech. Eng., 104共12兲, 1465–1488. Yong, K. Y. 共1979兲. ‘‘A laboratory study of the shaft resistance of bored
Reese, L. C. 共1978兲. ‘‘Design and construction of drilled shafts.’’ J. Geo- piles.’’ PhD thesis, Univ. of Sheffield, Sheffield, U.K.
tech. Eng., 104共1兲, 91–116. Yu, J. 共2000兲. ‘‘Effect of construction on axial load transfer along bored
Reese, L. C. and O’Neill, M. W. 共1988兲. ‘‘Drilled shafts: Construction piles in residual soils.’’ M. Eng. thesis, Nanyang Technological Univ.,
procedures and design methods.’’ Rep. No. FHWA-HI-88-042, Federal Singapore.
Highway Administration, Washington, D.C. Zhu, H. 共2000兲. ‘‘Evaluation of load transfer behavior of bored piles in
Reese, L. C., and Touma, F. T. 共1972兲. ‘‘Load tests of instrumented drilled residual soil incorporating construction effect.’’ PhD thesis, Nanyang
shafts constructed by the slurry displacement method.’’ Research Rep. Technological Univ., Singapore.
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 05/15/15. Copyright ASCE. For personal use only; all rights reserved.

Contract 108, Center for Highway Research, The Univ. of Texas, Zhu, H., and Chang, M. F. 共2002兲. ‘‘Load transfer curves along bored
Austin, Tex. piles considering modulus degradation.’’ J. Geotech. Geoenviron.
Uff, J. F. 共1969兲. ‘‘In situ measurements of earth pressure for a quay wall Eng., 128共9兲, 764 –774.

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / APRIL 2004 / 437

J. Geotech. Geoenviron. Eng. 2004.130:426-437.