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

net/publication/290586632

Interaction between vertical and lateral loads on the response of piles in soft
clays

Article · January 2005

CITATIONS READS

2 391

3 authors, including:

Karthigeyan . S Rajagopal Karpurapu

Anna University, Chennai Indian Institute of Technology Madras
35 PUBLICATIONS   377 CITATIONS    127 PUBLICATIONS   3,705 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Stabilization of Expansive Soil Bed by Using Micropiles View project

Liquefaction Hazard Mapping of Chandigarh Region Based on Site Characterization View project

All content following this page was uploaded by Karthigeyan . S on 14 July 2016.

The user has requested enhancement of the downloaded file.

 Interaction between vertical and lateral loads on the response of piles in soft clays
L'interaction entre les chargements verticaux et latéraux sur la réponse de tas dans les argiles Douces

S.Karthigeyan & VVGST.Ramakrishna

Central Building Research Institute, Roorkee – 247 667 (India)
K.Rajagopal
Indian Institute of Technology, Madras, Chennai – 600 036 (India)

ABSTRACT
The paper presents the results of three-dime2nsional finite element analysis of a single pile installed in homogeneous soft clay under
individual lateral load and combined vertical and lateral loading. The interaction between vertical and lateral loads on piles has been
analyzed using the finite element program GEOFEM-3D. In this analysis, the pile is treated as linear elastic and the behaviour of soil
has been idealized using non-associated elastic perfectly plastic based Drucker-Prager constitutive model. The numerical results pre-
sented in this paper highlight the effect of vertical load interaction on the lateral load deflection response of pile with reference to
some of the important parameters viz. method of loading, pile head fixity conditions and L/B ratio.
RÉSUMÉ
Le papier présente les résultats d'analyse d'élément finie à trois dimensions d'un tas seul installé dans l'argile douce homogène sous le
chargement latéral individuel et le chargement combiné, vertical et latéral. L'interaction entre les chargements verticaux et latéraux sur
les tas a été analysée l'utilisation du programme d'élément fini GEOFEM-3D. Dans cette analyse, le tas est traité comme linéaire élas-
tique et le comportement de sol a été idéalisé non-associé d'utilisation le Drucker-Prager basé, parfaitement en plastique et élastique
modèle constituant. Le pré-sented numérique de résultats dans ce papier souligne l'effet d'interaction de chargement verticale sur la
réponse de déflexion de chargement latérale de tas avec la référence à certains des paramètres importants à savoir. la méthode de
chargement, entasser les conditions de fixité de tête et la proportion de L/B.

1 INTRODUCTION the model, pile and soil continuum are divided into number of
nodes and elements. Fig. 1 shows the schematic 3-d finite ele-
Pile foundations are the preferred alternatives for supporting off- ment mesh discretization. Solid 20-noded isoparametric brick
shore structures constructed in soft marine clays. These struc- elements have been used to represent the pile and the soil con-
tures are subjected to large wave induced lateral loads and verti- tinuum. The interface between the pile-soil has been modeled
cal loads due to the weight of the structure and the loads on the using 16-node joint elements of zero thickness. All the nodes on
platform. Besides, due to high compressibility and low shear the lateral boundaries are restrained from moving in the normal
strength nature of the soft clays, large deformations might occur. direction to the surface representing rigid, smooth lateral
It is well understood from the reported works that the response boundaries. The nodes on the bottom surface are restrained in all
is different in case of combined axial and lateral loading. Ana- the three directions representing rough, rigid bottom surface.
lytical investigations (Davisson, 1965; Ramasamy, 1974; The finite element meshes consisted of approximately 7,000
Goryunov, 1975) suggest an increase in lateral deflection nodes and 1,450 20-node isoparametric brick elements. The pile
whereas the experimental investigations (Bartolomey, 1977; has been considered as linear elastic material and the homoge-
Sarochan & Bykov, 1976; Karasev et. al. 1977), suggest a re- neous soil has been treated as elastic-perfectly plastic materials
duction in lateral deflection due to the presence of vertical load.
The possible reasons for this contradiction are to be addressed
based on vigorous analyses by numerical methods. Pile
In view of the above stated issues, this investigation on the
interaction between vertical and lateral loads on the response of
piles in homogeneous soft clay has been taken up using the fi-
nite element program GEOFEM-3D. This paper presents the re-
sults of three-dimensional finite element analyses of piles in
homogeneous soft clay under different loading conditions such
as individual lateral and combined vertical and lateral loading.
Since the lateral response of piles is more critical and interesting Z

for the design engineers, as piles are not often structurally de-
signed to sustain significant lateral loads, the analysis mainly
focuses on the effect of vertical load on the lateral load response
of piles in soft clay with reference to key parameters influencing
the behavior. Y X

2 PRESENT 3-D FINITE ELEMENT MODEL

In the present analysis, the finite element program GEOFEM3D

developed by the authors was used to study the interaction be- Figure 1. Schematic 3-d finite element mesh
tween pile and soil under combined axial and lateral loading. In

1997
 based on Drucker-Prager constitutive model with non-associated 600
flow rule. The analysis scheme, stages of analysis, constitutive
model, incremental finite element procedure adopted etc. have 500
been described in more detail by Rajagopal et al. (2003). vertical load

lateral load on pile (kN)

400 0%Vult.
3 PARAMETRIC STUDIES 20%Vult.
300
40%Vult.
Since, allowable lateral deflection governs the design criteria in
most of the cases, the results are presented and discussed in this 200
60%Vult.
paper in terms of load vs. deflection response of piles under pure 80%Vult.
lateral loads and combined vertical and lateral loads. Accord-
ingly, series of three-dimensional finite element analyses have 100 90%Vult.
been performed by using GEOFEM3D program for single pile 100%Vult.
in homogeneous soft clay and subjected to pure lateral load as 0
well as combined vertical and lateral loads. For the sake of
analysis, the pile has been assumed to be 1200 � 1200mm 0 25 50 75 100 125 150 175
square pile with a length of 10 m and made of M25 grade of lateral deflection (mm)
Figure 3. Lateral load-deflection relationship of free-headed pile for
concrete with deformation modulus of 25,000 MPa and Pois-
VPL case.
son’s ratio of 0.15. The homogeneous soft clay soil is assumed
to have cohesive strength (cu) of 40 kPa; Young’s modulus of
15,000 kPa and Poisson’s ratio of 0.49. The analyses were con- vertical load (20%Vult to 100%Vult) on the lateral load response
ducted for both free head and fixed head conditions. The pile of pile. It is seen from Fig. 2 that the influence of vertical load
head was allowed to rotate during lateral load application for the for SAVL case is rather limited i.e. almost nil at small lateral
free head case while it is not allowed to rotate for the fixed head load level and it is more prominent at higher lateral load level.
case by preventing the vertical deformation of pile head nodes. On the other hand, in VPL case, as seen from Fig. 3, that the in-
The numerical results obtained from this investigation with ref- fluence of vertical load is significant at all load levels.
erence to various parameters are presented as follows. Also, it is commonly observed in both the cases that the lat-
eral capacities for a specified deflection decrease as vertical load
increases. Since the influence of vertical load on the lateral re-
3.1 Influence of method of loading sponse of pile is more significant in VPL case, the results related
to VPL case only are discussed here after. The VPL case is more
In the present analysis, the combined loading on pile is applied practically relevant as the piles in harbour structures are subject
in two different ways, simultaneously applied vertical and lateral to permanent vertical loads and are subjected to lateral loads
loads (SAVL) and vertical load prior to lateral load (VPL). In only during ship berthing operations.
the first case (SAVL), both vertical and lateral loads are applied
simultaneously in incremental form in each load step. In the 3.2 Influence of vertical load
second case (VPL), the loads are applied in two stages. In the
first stage, only the vertical loads were applied in number of From the above Fig. 3, the normalized lateral load capacities
load steps and then in the second stage, the lateral loads were (H*, i.e. H*=H/cuB2, where ‘H’ is the lateral load capacity of
applied in small increments while keeping the vertical loads pile) have been estimated for different levels of deflections and
constant. The ultimate vertical load (Vult.) of single pile has been are plotted against the vertical load variation in Fig. 4. From the
evaluated a priori by numerical analysis. For the combined ver- figure, it is clear that the influence of vertical load on the lateral
tical and lateral load analysis, the lateral load analyses have been load capacities is nominal at small vertical loads (0%Vult. to
performed with a vertical load equal to 0Vult., 20%Vult., 20%Vult.) and then considerable decrease is noticed for vertical
40%Vult., 60%Vult., 80%Vult 100%Vult. The analysis in lateral loads from 20%Vult. to 80%Vult followed by an exponential de-
direction was performed by applying equal lateral displacements crease in lateral load capacities at greater vertical loads. This has
to pile head and the reaction forces were monitored to calculate been commonly observed at all deflection levels.
the lateral force acting on the pile.
Figs. 2 and 3 show the lateral load vs. deflection response for
12
the pile under SAVL and VPL cases respectively. In these fig-
ures, the curves indicated by 0%Vult represents the case of pure
lateral load and the rest of the curves indicate the influence of 10
specified deflections
600
8 0.004B

500 0.008B
vertical load 6
H*

0.012B
lateral load on pile (kN)

400 0%Vult. 0.016B

4
20%Vult. 0.025B
300
40%Vult. 0.04B
2
200 60%Vult. 0.08B
80%Vult. 0
100 100%Vult.
0 30 60 90 120 150 180
vertical load in % of Vult.
0
Figure 4. Normalized lateral load capacities at different al-
0 25 50 75 100 125 150 175 lowable lateral deformations
lateral deflection (mm)

Figure 2. Lateral load-deflection relationship of free-headed pile

for SAVL case.
1998
 lateral soil stress (kPa) 15
-80 -60 -40 -20 0 20 40
Free head pile
12
0 0.004B

vertical load 0.008B

9
2 0.016B
0%Vult.

H*
20%Vult.
4 6 Fixed head pile
80%Vult.
0.004B
depth (m)

6 3 0.008B
0.016B
8 0

10 0 30 60 90 120 150 180

vertical load in % of Vult.

12 Figure 7. Normalized lateral load capacity under influence of vertical

loads with reference to free and fixed head piles
Figure 5. Variation of lateral soil stresses in front of free-headed pile

For the sake of better understanding the reduction of lateral load ence of vertical loads at smaller L/B ratios up to say 10-20 and
capacity, the variation of lateral soil stresses in front of free- beyond these values of L/B influence is rather limited. For L/B
head pile for vertical load levels of 0%Vult, 20%Vult and 80%Vult ratios more than 30, the influence of vertical load is almost nil.
and corresponding to a specified deflection level of 0.008B from Similar behavior has been observed for other vertical loads also.
the VPL cases are presented in Fig. 5. It is to be noted from the Thus, it can be inferred that the long flexible piles can be de-
figure that the lateral soil stresses in front of pile are almost signed for pure lateral loads only irrespective of presence of any
same for the case of 0%Vult and 20%Vult, and reduced under the vertical load. Anagnostopoulos and Georgiadis (1993) based on
influence of greater vertical loads (80%Vult), thus indicating that their experimental analysis of long aluminium piles (L/B = 50)
the lateral soil stress along the depth also reduce with increase in in soft clay also concluded that the influence of vertical load on
vertical load. This substantiates the phenomenon observed for lateral response is limited.
load deflection response curves as discussed in previous section
through Figs. 2 and 3.
3.4 Influence of Pile head fixity

3.3 Influence of L/B ratio From the lateral load deflection curves, the normalized lateral
load capacities (H*=H/cuB2) have been assessed for different de-
The results presented so far are based on an assumption that the flections of both free headed and fixed headed piles and are plot-
pile is rigid. However, in order to decide whether a pile would ted against vertical load variation as shown in Fig.7. From the
behave rigid or flexible, it is essential to study the influence of figure, it is commonly observed in both the cases that lateral
L/B and to arrive at limiting values of L/B on the influence of load capacities decrease under the influence of vertical load. In
vertical loads. Figure 6 shows the influence of L/B ratio on the order to understand the influence of vertical load on the pile
normalized lateral load capacities for a specified deflection level head fixity, the lateral load capacities estimated for different de-
of 0.004B. The combined load shown in the figure refers to flections are compiled in Table 1. The percentage reduction in
normalized lateral load capacities for a specified deflection level capacity over pure lateral load case is also given in the same ta-
of 0.004B. The combined load shown in the figure refers to a ble. From the table, it is clear that the influence of vertical load
particular vertical load (60%Vult). It is noted from the curves is almost within the range of –3 to +10 for both free head and
that the lateral load capacity of piles decrease under the influ- fixed head conditions for small vertical loads (upto 60%Vult).
However, with the increase in the vertical load, the reduction in
3.0
capacities is more or rather the influence is observed to be more
in case of free headed piles than fixed headed piles. It is well-
recognized fact that the capacities of piles under pure lateral
2.5
loads are more in case of fixed head piles than free head piles
due to restrained head condition. Besides, from the present ob-
2.0 servations, it can be inferred that even under the influence of
vertical loads, the head restraint is playing a vital role in increas-
H*

ing the lateral load capacities at different allowable lateral dis-

1.5 placement limits.
pure lateral load
1.0 combined loading 4 CONCLUSIONS

The numerical model based on three-dimensional finite element

0.5 analysis described herein investigated the effect of vertical load
interaction on the lateral load-deflection response of piles in soft
5 10 15 20 25 30 35 40 45 50 55 clay with reference to some of the key parameters, viz., the ver-
L/B ratio tical load level, sequence of loading, L/B ratio, pile head fixity
Figure 6. Lateral load capacities of piles under the influence of L/B
conditions. The main conclusions that could be drawn from this
investigation are as follows:

1999
 Table 1. Influence of vertical load on the normalized lateral load capacities (H*) of free and fixed-headed pile

Verti- Lateral load ca- % reduction Lateral load ca- % reduction over % reduction over
Lateral load capac-
cal pacity at 0.004B over pure lat- pacity at pure lateral load pure lateral load
ity at 0.016B de-
load deflection eral load case 0.008B deflec- case (0.008B) flection case (0.016B)
in % (0.004B) tion
of
Vult.
Free Fixed Free Fixed Free Fixed Free Fixed Free Fixed Free Fixed
head head head head head head head head head head head head

0 2.07 4.13 0 0 4.1 7.78 0 0 6.84 12.98 0 0

20 2.1 4.27 -1 -3 4.2 8.02 -2 -3 6.9 13.28 -1 -2

40 2.01 4.20 +3 -2 4.0 7.91 +3 -2 6.5 13.09 +5 -1

60 1.91 4.06 +8 +2 3.8 7.15 +8 +8 6.2 12.45 +10 +5

80 1.75 3.87 +15 +6 3.5 6.56 +15 +16 5.5 11.73 +20 +10

100 0.94 3.47 +55 +16 1.8 5.64 +56 +28 2.95 10.9 +56 +16

The vertical load has significant influence on the lateral re-

sponse of piles. When both vertical and lateral loads are simul- REFERENCES
taneously applied (SAVL case), the influence of vertical load on
Anagnostopoulos, C. & Georgiadis, M. 1993. Interaction of axial and
the lateral response of pile is felt only at higher vertical load lateral pile responses, Journal of Geotechnical Engineering, ASCE,
levels, whereas when the vertical load is applied prior to lateral Vol.119, No.4, pp.793-798.
loads (VPL case), the influence is noticed at all the load levels. Bartolomey, A.A. 1977. Experimental analysis of pile groups under lat-
The lateral capacities decrease and deflections increase under eral loads, Proceedings of the Special Session 10 of the Ninth Inter-
the presence of vertical loads in general and it is more prominent national Conference on Soil Mechanics and Foundation Engineer-
at higher vertical load levels. This could be mainly attributed to ing, Tokyo, pp. 187-188.
the reason that in case of soft clays under vertical loads, the soil Davisson, M.T. & Robinson, K.E. 1965. Bending and buckling of par-
gets initially compressed and when the lateral load is applied, tially embedded piles, Sixth International Conference on Soil Me-
the corresponding net passive resistance will be less than the ex- chanics and Foundation Engineering, Montreal, Vol.2, pp. 243-246.
Gandhi, S.R., Suresh, P.K. and Raju, V.S. 1988. Lateral Load Tests on
pected passive resistance. The initial compression increases with Large diameter Board Pile and Analysis, Proceedings of the Indian
respect to the amount of vertical load. This has also been illus- Geotechnical Conference (IGC-88), Vol. No.1, Allahabad, Decem-
trated through the lateral stress variation plots along the depth ber 1988, pp.373-377.
for low and high fractions of vertical load (Figure 5). Goryunov, B.F. 1975. Discussion on Analysis of Piles Subjected to the
The influence of vertical loads on the lateral response is felt Combined Action of Vertical and Horizontal Loads, Journal of Soil
only upto a certain limiting value of L/B (15 to 30) for a speci- Mechanics and Foundation. Engineering., Vol.10, No.1 pp.10.
fied lateral deflection and beyond this values, the piles can be Jain, N.K. Ranjan, G. and Ramasamy, G. 1987. Effect of Vertical Load
designed as flexible piles under pure lateral loads irrespective of on Flexural Behaviour of Piles, Geotechnical Engineering, Vol. 18,
the presence of vertical loads. The influence of the vertical load pp.185-204.
Karasev, O.V., Talanov, G.P., Benda, S.F. 1977. Investigation of the
is significant in the L/B range of up to 15. As most of the piles work of single cast piles under different load combinations, Soil
installed on land fit in to this category, it is very important to Mechanics and Foundation Engineering (Translated from Russian),
consider this aspect for onshore piles. On the other hand, the No.6, pp. 173-177.
piles installed in offshore structures tend to be very long for Rajagopal, K. 1998. Users Manual for the finite element program
which the interaction effects can be expected to be minimum. GEOFEM-3D, Department of Civil Engineering, Indian Institute of
Fixed head piles offer more resistance to lateral loads in gen- Technology Madras, Chennai.
eral. The current results show that this is true even in the pres- Rajagopal, K., Ramakrishna, VVGST., Karthigeyan, S. and Gandhi,
ence of vertical loads. S.R. 2003. 3-D finite element analysis of piles under combined axial
and lateral loads. Proceedings of the Indian Geotechnical Confer-
ence (IGC-2003), Roorkee, pp. 83-87.
Ramasamy, G. 1974. Flexural behaviour of axially and laterally loaded
ACKNOWLEDGEMENT individual piles and group of piles, Thesis submitted to Indian insti-
tute of Science for the award of Ph.D degree, Bangalore.
The results presented in this paper form a part of Ph.D. pro- Sarochan, E.A. & Bykov, V.I. 1976. Performance of groups of cast in
gramme of the first author being carried out jointly at the insti- place piles subjected to horizontal loading, Soil Mechanics and
tutes of IIT, Madras and Central Building Research Institute Foundation Engineering, (Translated from Russian) Vol.13, No.3.,
CBRI), Roorkee. The authors are thankful to the Director, pp. 157-161.
CBRI, Roorkee, for his encouragement and interest in this pro-
ject. The authors are also grateful to Prof. S.R. Gandhi, IIT Ma-
dras for many useful discussions.

2000

View publication stats