See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/267034554

Pile failure during seismic liquefaction – theory and practice

Conference Paper · April 2004

DOI: 10.1201/9781439833452.ch42

CITATIONS READS
17 1,127

2 authors:

Subhamoy Bhattacharya Malcolm D. Bolton

University of Surrey University of Cambridge
330 PUBLICATIONS 7,554 CITATIONS 248 PUBLICATIONS 15,515 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Subhamoy Bhattacharya on 18 October 2014.

The user has requested enhancement of the downloaded file.

 PILE FAILURE DURING SEISMIC LIQUEFACTION:
THEORY AND PRACTICE

by

Subhamoy Bhattacharya and Malcolm Bolton

1
Fugro Limited (U.K)
2
University of Cambridge (U.K)

International Conference on
"Cyclic Behaviour of Soils and Liquefaction Phenomena"
Bochum, Germany, 31 March – 02 April 2004

organized by

Ruhr University Bochum

Faculty of Civil Engineering
Institute of Soil Mechanics and Foundation Engineering

PUBLISHER: AA BALKEMA, the Netherlands.

Proc. of the International Conference on “Cyclic behaviour of soils and liquefaction phenomena”, Bochum, Germany, 31st Mar-2nd Apr 2004, A.A.Balkema
Publisher.

Pile failure during seismic liquefaction – theory and practice

S. Bhattacharya
Fugro Limited (U.K.)
M.D. Bolton
University of Cambridge (U.K.)

ABSTRACT: Collapse of structures resting on piled foundations is still observed after strong earthquakes de-
spite the fact that a large factor of safety is employed in their design. It has been identified that the assumed
failure mechanism underlying the current design methods cannot explain some of the observed seismic pile
failures. In this paper a new theory of pile failure in liquefiable soils is described and is compared with the
current design methods. First, the theory of pile failure based on buckling instability is described introducing
the concept of effective length of pile in liquefiable zone. The main postulate of this theory is that if piles are
too slender, they require lateral support from the surrounding soil if they are to avoid buckling instability.
This lateral support can fall to near zero due to seismic liquefaction and a slender pile may buckle. Detailed
centrifuge testing, in-depth study of field case records and analytical studies form the basis of this theory.
Next, the design methods of Eurocode 8, JRA (1996) and NEHRP (2000) are examined with respect to this
theory of pile failure. It has been shown that the current codes of practice for pile design omit considerations
necessary to avoid buckling of fully embedded piles in liquefiable soils. These codes should be modified to
address buckling. It is proposed that the slenderness ratio of pile (i.e. the effective length of pile/ minimum
radius of gyration) in the liquefiable soils be kept below 50 i.e. piles should have length to diameter ratio of
about 12 in the likely liquefiable zone to avoid instability failure.

INTRODUCTION quakes are very rapid events and as much of the

damage to piles occurs beneath the ground, it is hard
1.1 Collapse of pile-supported structures during
to ascertain the failure mechanism unless deep exca-
earthquakes
vations are carried out.
Pile-supported structures founded on or passing
through liquefiable soils still collapse during earth-
quakes. Figure 1 shows photographs of two such
cases.

Figure 2: Pile failure observed during the excavation of the

NHK building during the 1964 Niigata earthquake, Hamada
(1992a).
Figure 1: (a): Piled tanks after 1995 Kobe earthquake (Japan),
photo courtesy NISEE; (b): Piled Kandla Port Tower after
2001 Bhuj earthquake. Figure 2 shows the damage pattern of the piles re-
vealed during the excavation of the N.H.K building.
In each of the above examples of failure, the super- This building was damaged during the 1964 Niigata
structure remained undamaged, but the structure as a earthquake. As can be seen from the photograph, the
whole tilted or moved laterally. It is most likely that piles failed by formation of plastic hinges. Similar
the piles beneath the foundation suffered structural observations of plastic hinge formations in piles
failure by formation of plastic hinges. As earth- were reported by Tokimatsu et al (1997) in a 3-
storied building after the 1995 Kobe earthquake, by
Hamada (1992) in the NFCH building after the 1964 Bhattacharya (2003), Bhattacharya et al. (2002,
Niigata earthquake and by Fukuoka (1966) in the 2003a, 2003b), Bhattacharya and Bolton (2004a,
well known Showa Bridge failure. 2004b) have shown the inconsistency of the current
understanding with some of the observed seismic
pile failure at liquefiable sites. They note that the lo-
1.2 Pile-supported structures still collapse during
cation of a plastic hinge due to bending is expected
earthquakes: What is missing in the design
to occur at the interface of liquefiable and non-
method?
liquefiable layer as this section experiences the high-
Structural failure by the formation of plastic hinges est bending moment. It is often seen, however, that
in piles passing through liquefiable soils has been the hinge forms within the top third of the pile.
observed in many of the recent strong earthquakes.
This suggests that the bending moments or shear Bhattacharya and Bolton (2004a) also points out that
forces that are experienced by the piles exceed those it is a common observation in seismic bridge failure
predicted by design methods (or codes of practice). that piers collapse while abutments remain stable. In
All current design codes apparently provide a high a bridge design, the number of piles required to sup-
margin of safety (using partial safety factors on load port an abutment is governed by lateral load due to
and material stress which increase the overall safety the fact that the abutment, as well as carrying the
factor), yet occurrences of pile failure due to lique- dead load of the deck, has to retain earth and fills of
faction are abundant. This implies that the actual the approach roads to the bridge. On the other hand,
moments or shear forces experienced by the pile are the bridge piers (intermediate supports) predomi-
many times those predicted. It may be concluded nantly support the axial load of the deck. The lateral
that design methods may not be consistent with the load acting on the pier during an earthquake is pri-
physical mechanisms that govern the failure. In marily the inertial force. The lateral capacity of a
other words, something is missing. pile is typically 10 to 20% of the axial load capacity.
Thus, for a multiple-span bridge having similar span
The current method of pile design under earthquake lengths, the number of piles supporting an abutment
loading is based on a bending mechanism where in- will be more that that of a pier. It is worthwhile to
ertia and slope movement (lateral spreading of soil) note that in these examples only the bridge piers col-
induces bending moments in the pile. Figure 3 sche- lapsed while the abutments remained stable. This
matically describes the mechanism. In other words, hints that the failure of bridge pier foundations may
the current understanding of pile failure simply be influenced by axial load. In contrast, the current
treats piles as beam element. design methods only concentrate on lateral loads.

The purpose of this paper is to:

1. Highlight the importance of axial load that the
pile continues to carry during its entire design
period, needless to say also during seismic lique-
faction. This axial load can be quite close to the
buckling load of the pile during seismic liquefac-
tion due to the diminishing soil support from the
surrounding soil owing to liquefaction. The ef-
fect of axial load is currently overlooked in the
current design methods.
2. Describe a new theory of pile failure taking into
consideration the axial load, which is generally
consistent with the observations of seismic pile
failure.
Figure 3: Current understanding of pile failure, Finn and Tha- 3. Compare the new theory of pile failure with the
varaj (2001). design methods.
Bhattacharya (2003) shows through an example that
the overall safety factor against plastic yielding of a 2 THEORY OF PILE FAILURE
typical concrete pile considering the bending mecha-
nism may range between 4 and 8. This is due to the A new theory of pile failure is described in this sec-
multiplication of partial safety factors on load (1.5), tion. This theory of pile failure is based on the buck-
material (1.5 for concrete), fully plastic strength fac- ling instability of slender columns, the structural na-
tor (ZP/ZE = 1.67 for circular section) and practical ture of piles and “Critical State” soil mechanics.
factors such as minimum reinforcements or mini- This theory has been formulated based on a study of
mum number of bars. fifteen case histories of pile foundation performance
and verified using dynamic centrifuge tests. Analyti- monly known as the “critical load” of the frame or
cal studies also support this theory of pile failure. A buckling load (Pcr). The critical load can be esti-
hypothesis of pile-soil interaction is also proposed in mated based on the concept of “Euler’s effective
this theory. Details of this theory can be seen in length of an equivalent pin-ended strut (Leff)”.
Bhattacharya (2003). This section describes the ba-
sics of this theory.

2.1 Structural nature of piles

From a structural perspective, axially loaded piles
are long slender columns with lateral support pro-
vided by the surrounding soil. If unsupported, these
columns will fail in buckling instability and not due Figure 5: A row of slender columns representing a row of
to crushing of the pile material. Figure 4 shows the piles.
length and diameter of tubular piles used in different
projects around the world after Bond (1989). Figure 7 shows the concept of effective length of
pile adopted from column stability theory to normal-
ise the different boundary conditions of pile tip and
pile head.

Figure 4: Length and diameter of tubular piles, Bond (1989)

This figure shows that piles normally have ratios of

length to diameter of 25 to 100. A parameter rmin Figure 6: A group of slender columns representing a pile
(minimum radius of gyration) given by Equation 1 is group.
introduced to represent piles of any shape (square,
tubular or circular). This parameter is used by struc-
tural engineer to study buckling of slender columns.
Pile head free
to translate but
I fixed in
rmin =
Pile head
(1) unrestrained Leff= L(1)
0
direction

A
Leff= 2L0
Buckling zone/
where I is the moment of inertia; A is the area of the Liquefiable layer= L0

pile section.
Leff= 2L0

Figures 5 and 6 show the failure pattern of structures Euler’s buckling of

equivalent pinned strut
resting on slender columns. The length to diameter Euler’s buckling of
equivalent pinned strut
ratio shown in the example is 93. Figure 5 may rep-
resent a row of piles in the absence of soil as in the
well-known failure of Showa Bridge, see Fukuoka
(1966). Figure 6 may represent a piled raft or a pile Figure 7: Concept of effective length of pile.
group in the absence of soil. Thus in the absence of
soil, we would expect a pile-supported structure to The Euler’s elastic critical load (Pcr) is calculated
fail in a similar pattern. The failures shown in Fig- from the well known buckling formula given by
ures 5 and 6 are due to axial load alone. Equation 2.
π 2 EI
The static axial load at which a frame supported on Pcr = (2)
L2 eff
slender columns becomes laterally unstable is com-
The above failure is due to the effects of axial load creases inversely with the square of its length fol-
alone. In the presence of static lateral loads, the criti- lowing Euler’s formula (Equation 2). Figure 9 shows
cal load would come down. In presence of dynamic a typical plot for the variation of allowable load (P)
loads, i.e. if the entire set up shown in Figure 5 and and buckling load (Pcr) of a pile (if unsupported)
6 is shaken, the frame would become unstable at a against length of the pile. The pile in the above ex-
much lower load. ample has a diameter of 300mm (typical pile dimen-
sion in 1964 Japan) and is passing through a typical
Stability analysis of elastic columns (see Ti- liquefied soil. The allowable load (P) is estimated
moshenko and Gere, 1961) shows that the lateral de- based on conventional procedures with no allowance
flections caused by lateral loads gets amplified in the for liquefaction.
presence of axial loads. If (δ0) is the deflection due 2000 Buckling load (Pcr) if unsupported
to lateral loads alone, the final deflection (δ) gets
amplified by the presence of axial loads. The term
(δ/δ0) can be termed as the “Buckling amplification 1500
Allowable load (P)

Load (kN)
factor (BAF)” given by equation 3
1000
1
B. A.F = (3) (2)
⎛ P ⎞
⎜⎜1 − ⎟⎟ 500

⎝ Pcr ⎠
0
This form of expression, sketched in Figure 8, can 0 5 10 15 20
be used with good accuracy (less than 2% error) for Pile length (m)
all beam-columns having (P/Pcr) less than 0.6, Ti-
moshenko and Gere (1961). Beyond the ratio of 0.6 Figure 9: Allowable load and buckling load of a typical pile (in
unsupported).
the induced plastic strains cause a deterioration of
elastic bending stiffness leading to a reduction in the
Structural engineers generally demand a factor of
critical buckling load Pcr, and to premature collapse.
safety of at least 3 against linear elastic buckling to
allow for eccentricities, imperfections and reduction
21 of stiffness due to yielding. Thus from Figure 9, if
Buckling Amplification Factor

19
17
unsupported over a length of 10 metres or more,
15 these columns will fail due to buckling instability
13 and not due to crushing of the material. During
11 earthquake-induced liquefaction, the soil surround-
9
ing the pile loses its effective confining stress and
7
5
can no longer offer sufficient lateral support. The
3 pile may now act as an unsupported column prone to
1 axial instability. This instability may cause it to
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 buckle sideways in the direction of least elastic
(P/Pcr) bending stiffness under the action of axial load. In
this case the pile may push the soil and it may not be
Figure 8: Buckling amplification factor and normalised axial necessary to invoke lateral spreading of the soil to
load. cause a pile to collapse. Dynamic centrifuge model-
ling has been carried out to investigate this failure
The theory of pile failure is based on an idealisation mechanism and is described in the next section.
that a pile-supported structure such as a bridge or a
building is a frame supported on slender columns 2.3 Centrifuge tests to verify the buckling instability
such as Figure 5 and 6 with support from liquefiable of piles
soil. The part of the pile in liquefiable soil is the un- Dynamic centrifuge tests were carried out by Bhat-
supported zone during seismic liquefaction. Each of tacharya (2003), Bhattacharya et al (2002, 2003) to
these structures has a critical load i.e. the minimum verify if fully embedded end-bearing piles passing
axial load at which the frame becomes unstable. through saturated, loose to medium dense sands and
resting on hard layers buckle under the action of ax-
ial load alone if the surrounding soil liquefies in an
2.2 Allowable load and buckling load of a pile
earthquake. This would verify the hypothesis of pile
Generally, as the length of the pile increases, the al- failure by buckling instability. During earthquakes,
lowable load on the pile increases primarily due to the predominant loads acting on a pile are axial, in-
the additional shaft friction, but the buckling load (if ertial and those due to slope movement. The tests
the pile were laterally unsupported by soil) de- were designed in level ground to avoid the effects of
lateral spreading. Twelve piles were tested in a se-
ries of four centrifuge tests including some which
decoupled the effects of inertia and axial load. Fig-
ure 10 summarises the performance of the piles in
the test showing the normalised axial load (P/Pcr). In
the figure, P denotes the applied axial load on the
pile and Pcr represents the elastic critical load of the
pile treated as a column neglecting any support from
the soil. It can be observed that piles having a P/Pcr
ratio less than 0.5 did not fail and in each case a se-
ries of earthquakes had been fired.

Did not fail Failed

1.04 1.25
1.48 P/Pcr
0.97 1.01
0.75 0.96
0.5
0.22 0.25 0.35

Figure 11: Pile buckling and Euler’s classical buckling

Axial load + Inertia Axial load + Inertia Only axial load

2.4 Pile-soil interaction during buckling
Sufficient information has been obtained from the
centrifuge models to propose a hypothesis of pile-
Figure 10: Summary of the pile performance in the centrifuge soil interaction during a buckling event. The pile be-
tests. gins to buckle when the front of zero effective stress
reaches a critical depth HC given by Equation 4 ob-
On the other hand, piles having P/Pcr ratio greater tained from Equation 2, taking Leff = 2HC for a pile
than 0.75 failed during an earthquake either under with no restraint at the head.
the action of axial load or combined axial load and
inertia. This result is consistent with Figure 8 where π 2 EI
HC = (4)
it can be observed that for a P/Pcr ratio of 0.75, the 4P
transverse deflection due to lateral loads will be am-
This buckling instability will cause the pile to shear
plified at least 4 times giving rise to an additional P-
the soil adjacent to it, which will start offering tem-
δ moment which eventually leads to plastic collapse.
porary resistance. Figure 12 schematically shows the
pile-soil interaction during the post-buckling period.
Figure 11(a) shows one photograph of pile failure.
The same pile was tested in similar conditions, but
The soil element in front of the buckling pile,
in the absence of soil and the mode shape is shown
marked A in Figure 12, will be subjected to mono-
in Figure 11(b). For Euler’s buckling, it is obvious
tonic shearing in addition to the cyclic shearing due
that the hinge should form at the bottom for a canti-
to the earthquake. The “V/k” ratio (i.e. the ratio of
lever strut (Figures 5), but for the pile buckling, it
velocity of the pile to the permeability of the soil)
was observed that the hinge formed within the top
which was of the order of 100’s, suggests that the
half of the pile. For pile buckling, it was also noted
event is best looked upon as undrained. The resis-
that curvature decreased with depth. Curvature being
tance to the buckling pile is due to this “undrained
related to bending moment, the test result suggests
strength of the soil”, which is the strength when
that that lower part of the liquefied soil zone offered
sheared at constant volume. It should be obvious
resistance to the buckling pile. It is concluded by
from the definition that the stress path must follow
Bhattacharya et al (2002) and Bhattacharya (2003)
the Critical State line.
that liquefied soil cannot prevent the initiation of
buckling in an initially straight pile, but that some
In the q-p’ plot shown in Figure 13, a soil element
secondary support then becomes available.
during the pre-buckling period, will start from some
point in the q-p’ plot, shown by X, and generate
Thus pile buckling is in some way different from
positive pore pressure due to the earthquake shaking.
Euler’s classical buckling. In the case of Euler’s
The stress path will progress towards the origin until
classical buckling, the resistance to the buckling
it hits the “Phase Transformation” line.
strut is air. On the other hand for the buckling of
piles, the resistance is due to the liquefied soil.
 Direction of Figure 12) will be sheared more than the bottom soil
buckling
in the near field (marked 3 in Figure 12).

The imposition of undrained monotonic shear strains

(pile pushing the soil) in loose to medium dense
δi sand at low effective stresses will lead to an attempt
(1 ) to dilate. The event being at constant volume will
suppress this potential dilatancy by a negative in-
crement of pore pressure in the locally sheared soil.
(2) This negative increment of pore pressure creates an
increase in effective stress, which temporarily pro-
vides support to the buckling pile.
(3)

The pore fluid pressure in the sheared zone can drop

D ire ction
of s ha king to a maximum of -100kPa, which would correspond
to the greatest possible effective stress increment
during shearing. Beyond this value, vapour bubbles
(C )
tends to nucleate. On the stress path the soil element
(A )
(B ) in front of the pile moves from Y to Z, Figure 15.
This local reduction of pore pressure would induce a
transient flow into the sheared soil from the neigh-
Plan of the ESB B ox bouring “liquefied but not monotonically sheared
soil” shown schematically in Figure 14. The lateral
Figure 12: Pile-soil interaction during buckling
resistance of the liquefied soil would then decrease.
The stress path will run up and down like a butterfly Initial
wing passing through or near the origin. At the same position of Position of pile at a
time, the pile length will progressively be unsup- the pile particular instant
ported by the soil grains, in top down fashion, until of time (t)
the critical depth is reached. As the critical depth is
reached the pile starts to buckle.
Transient
Due to monotonic flow
undrained shearing as
the buckling pile
pushes the soil
Upper limit of
q shearing due to
cavitation Soil sheared Soil having
to its critical σv’=0
100 kPa state
+ uhy
Figure 14: Transient flow

X p’ It is the upper part of a liquefiable sand layer that

Y
remains longest in a state of zero effective stress due
Due to earthquake
shearing (Cyclic) to upward hydraulic gradients. It is also the upper
part of the pile which displaces most, and which can
fully soften to zero shear strength the supporting soil
adjacent to it. Ultimately, therefore, the upper part of
the soil can be properly described as liquefied, in the
Z true sense of the word in science or common lan-
guage, according to the necessary conditions laid
Figure 13: Stress path of a soil element close to the pile, where down by Schofield (1981). For the lower part of the
uhy = hydrostatic pressure. pile, the resistance will increase as the pile shears
the “initially liquefied soil” but not for a sufficient
The behaviour of a soil element in front of the pile duration. This is due to the development of negative
(marked A in Figure 12) can be described as similar pore pressure in the sheared soil which will induce
to “Triaxial Extension” while the soil behind the pile transient flow from the neighbouring “liquefied but
(marked B in Figure 12) can be described as similar not monotonically sheared” soil. The buckling pile
to “Triaxial Compression”. Due to the mode shape will also suffer increasing loss of bending stiffness
of the pile, the top-soil in the near field (marked 1 in
due to plastic yielding, so the restraint necessary to “short” columns. Columns having a slenderness ratio
hold it in equilibrium will also increase. below 50 are expected to fail in crushing whereas
This imbalance between increasing bending moment those above 50 are expected to fail in buckling in-
created by displacement of pile cap, deteriorating stability.
bending stiffness of the pile and the reducing differ-
ential soil support along its length, creates a shallow
plastic hinge which then leads to the dynamic col- 3 METHODS OF PILE DESIGN IN CODES OF
lapse of the structure. PRACTICE
3.1 Japanese Road Association Code
2.5 Study of field case records
The mechanism of pile failure in JRA (1996) is
Fifteen reported cases of pile foundation perform- shown in Figure 16. The code advises practising en-
ance during earthquake-induced liquefaction were gineers to design piles against bending failure as-
studied and analysed as listed in Table 1. Six of the suming that the non-liquefied crust offers passive
piled foundations were found to survive while the earth pressure to the pile and the liquefied soil offers
others suffered severe damage. 30% of the total overburden pressure. Other codes
such as the USA code (NEHRP 2000) and Eurocode
Table 1. Case Histories studied 8, part 5 (1998) also focus on the bending strength
of the pile.
ID in Case History and Reference
fig 15
A Hokuriku Building, Hamada (1992)
B Landing Bridge, Berrill et el (2001)
C 14 storey building, Tokimatsu et al (1996)
D Hanshin expressway pier, Ishihara (1997)
E LPG tank 101, Ishihara (1997)
F Kobe Shimim hospital, Soga (1997)
G N.H.K building, Hamada (1992)
H NFCH building, Hamada (1992)
I Yachiyo Bridge, Hamada (1992)
J Gaiko Ware House, Hamada (1992)
K 4 storey fire house, Tokimatsu et al (1996)
L 3 storied (Kobe Univ) Tokimatsu et al (1996)
M Elevated port liner railway, Soga (1997)
N LPG tank –106,107, Ishihara (1997)
O Showa bridge, Hamada (1992). Figure 16: JRA (1996) code of practice

3.2 Eurocode 8
The Eurocode advises designers to design piles
against bending due to inertia and kinematic forces
arising from the deformation of the surrounding soil.
It goes on saying:
“Piles shall be designed to remain elastic. When
this is not feasible, the sections of the potential plas-
tic hinging must be designed according to the rules
of Part 1-3 of Eurocode 8”.
Eurocode 8 (Part 5) says
“Potential plastic hingeing shall be assumed for:
• a region of 2d from the pile cap
• a region of ± 2d from any interface between
two layers with markedly different shear
Figure 15: Plot of effective length (Leff) and minimum radius of stiffness (ratio of shear moduli > 6)
gyration (rmin). where d denotes the pile diameter. Such region shall
be ductile, using proper confining reinforcements ”.
Figure 15 presents the effective length and the mini-
mum radius of gyration for the case histories stud-
ied. A line representing Leff/rmin = 50 is shown in the 4 CONCLUSIONS
plot which distinguishes the piles of poor perform-
ance from the piles of good performance. This line is Buckling is a possible failure mode of piled founda-
of some significance in structural engineering, as it tions in areas of seismic liquefaction. Influences
is often used to distinguish between “long” and such as lateral loading due to slope movement, iner-
tia effects due to early shaking or out-of-line Bhattacharya,S. Madabhushi, S.P.G and Bolton,M.D 2003a
straightness, cause lateral deflections, which are se- “Pile instability during earthquake liquefaction”, Proc of
the 16th ASCE Engineering Mechanics Conference (EM
verely amplified if the axial load is permitted to ap- 2003), Paper no–404, University of Washington, Seattle
proach the buckling load. These lateral loads are, 16-18th July 2003.
however, secondary to the basic requirements that Bhattacharya, S., Madabhushi, S.P.G, Bolton,M.D, S.K.Haigh
piles in liquefiable soils must be checked against and Soga, K (2003b) “A reconsideration of the safety of
Euler’s buckling. the piled bridge foundations in liquefiable soils”. Technical
report (TR 328) of Cambridge University.
(http://www-civ.eng.cam.ac.uk/geotech_new/publications/TR/TR328.pdf)
4.1 Comparison between theory and practice Bhattacharya, S. 2003. Pile instability during earthquake lique-
faction. PhD thesis: University of Cambridge, U.K.
All current design methods, such as JRA(1996), Bhattacharya, S. and Bolton, M.D. 2004a. A fundamental
NEHRP (2000) or Eurocode 8 (1998) focus on the omission in seismic pile design leading to collapse, Proc.
bending strength of the pile and overlook considera- 11th Int. Conf. on Soil dynamics and Earthquake Engineer-
ing. Berkeley, California, January 7-9, 2004.
tions necessary to avoid buckling in the event of soil Bhattacharya, S. and Bolton, M.D. 2004b. Errors in design
liquefaction. leading to pile failure during seismic liquefaction, Proc. 5th
In design, beam bending and column buckling are Int. Conf. on Case Histories in Geotechnical Engineering.
approached in two different ways. Piles have erro- Eds (Shamsher Prakash) New York, April 13-17, 2004.
neously been designed as cantilever beams. Bond, A.J. 1989 “Behaviour of displacement piles in over-
consolidated clay” PhD thesis, Imperial College (U.K).
Bending is a stable mechanism as long as the pile is Eurocode 8 (Part 5): Design provisions for earthquake resis-
elastic, i.e. if the lateral load is withdrawn, the pile tance of structures, foundations, retaining structures and
comes back to its initial configuration. This failure geotechnical aspects”, European Committee for Standardi-
mode depends on the bending strength (moment for zation, Brussels.
first yield, MY; or plastic moment capacity, MP) of Finn W.D.L and Thavaraj, T (2001)/. “Deep foundations in li-
the member under consideration. quefiable soils: Centrifuge tests and method of analysis”.
Proc. of the 4th Int. Conf. on recent advances in geotechni-
On the other hand, buckling is an unstable mecha- cal earthquake engineering and soil dynamics. San Diego,
nism. It is sudden and occurs when the elastic criti- California, March 26-31, 2001.
cal load is reached. It is the most destructive mode Fukuoka, M (1966) “Damage to Civil Engineering Structures”,
of failure and depends on the geometrical properties Soils and Foundations, Volume-6, No-2, pp 45-52.
of the member, i.e. slenderness ratio, and not on the Hamada,M (1992)“Large ground deformations and their ef-
fects on lifelines: 1964 Niigata earthquake, 1983 Nihonkai-
yield strength of the material. Chubu earthquake Case Studies of liquefaction and lifelines
For example, steel pipe piles with identical length performance during past earthquake,”. Technical Report
and diameter but having different yield strength [fy NCEER-92-0001, Volume-1.
of 200MPa, 500MPa, 1000MPa] will buckle at al- Ishihara,K (1997). Geotechnical aspects of the 1995 Kobe
most the same axial load but can resist different earthquake, Proc. of ICSMFE, Hamburg, pp 2047-2073.
JRA (1996), Specification for Highway Bridges, Part V, Seis-
amounts of bending. Bending failure may be mic Design, Japanese Road Association.
avoided by increasing the yield strength of the mate- NEHRP 2000 (National Earthquake Hazard Reduction Pro-
rial, i.e. by using high-grade concrete or additional gram). Commentary for Federal Emergency Management
reinforcements, but it may not suffice to avoid buck- Agency (FEMA, USA 369). Seismic regulation for new
ling. To avoid buckling, there should be a minimum buildings.
pile diameter depending on the depth of the liquefi- NISEE: National Information Services for Earthquake Engi-
neering, University of California, Berkeley.
able soil. Soga,K. “Geotechnical aspects of Kobe earthquake”, Chapter 8
It is proposed in this paper that piles may be better of EEFIT report on Kobe earthquake, Institution of Struc-
looked upon as “columns carrying lateral loads” tural Engineers, UK.
rather than “axially loaded beams”. While designing Schofield, A.N (1981) Dynamic and earthquake geotechnical
piles in liquefiable soil, the part of the pile in lique- centrifuge modelling, Proc. of the Int. Conf on recent ad-
vances in geotechnical earthquake engineering and soil
fiable soil is to be assumed as unsupported. dynamics.
Timoshenko, S.P and Gere,J.M Theory of elastic stability,
McGraw Hill book company, New York, 1961.
5 REFERENCES Tokimatsu, K. Mizuno, H and Kakurai, M (1996).“Building
Damage associated with Geotechnical problems”. Special
Berrill,J.B., Christensen, S.A, Keenan, R. P., Okada, W. and issue of Soils and Foundations, pp 219-234.
Pettinga, J.R., 2001. “Case Studies of Lateral Spreading Tokimatsu, K., Oh-oka Hirishi, Satake, K., Shamato Y., Asaka
Forces on a Piled Foundation”,Geotechnique 51, No. 6, pp Y (1997); Failure and deformation modes of piles due to
501-517. liquefaction-induced lateral spreading in the 1995 Hyogo-
Bhattacharya, S, Madabhushi, S.P.G and Bolton, M.D 2002 ken-Nambu earthquake., Journal Struct. Eng. AIJ (Japan)
“An alternative mechanism of pile failure in liquefiable de- No 495, pp 95-100.
posits during earthquakes”, Geotechnique (in press). Avail-
able as Technical report (TR 324) of Cambridge University.
(http://www-civ.eng.cam.ac.uk/geotech_new/publications/TR/TR324.pdf)

View publication stats