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

net/publication/262454159

Undrained Bearing Capacity of Shallow Foundations on Structured Soils

Conference Paper · August 2005

DOI: 10.1201/NOE0415390637.ch46

CITATIONS READS

4 2,038

2 authors:

D. S. Liyanapathirana John P. Carter

Western Sydney University University of Newcastle
103 PUBLICATIONS   1,140 CITATIONS    439 PUBLICATIONS   6,800 CITATIONS   

SEE PROFILE SEE PROFILE

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

Mechanics of soft soils View project

Sydney Zoo Permeable carpark View project

All content following this page was uploaded by John P. Carter on 16 July 2014.

The user has requested enhancement of the downloaded file.

Undrained bearing capacity of shallow foundations on structured soils
D.S. Liyanapathirana
University of Wollongong, NSW, Australia
J.P. Carter
University of Sydney, NSW, Australia

ABSTRACT: This paper examines the undrained bearing capacity of shallow circular foundations on struc-
tured soil deposits. Guidelines are given to identify the importance of structural features of the soil when as-
sessing its bearing resistance. Results obtained using a finite element model have been compared with those
from existing bearing capacity formulae based largely on plasticity theory. A new bearing capacity equation
has been proposed incorporating critical state soil parameters and additional parameters that quantify the ef-
fects of soil structure on its mechanical behaviour.

1 INTRODUCTION
Many structures are founded on shallow footings
which bear directly on either natural or man-made
soils. Usually these soils in situ have a structure and
behave differently from the same material in a re-
constituted state (e.g., Burland, 1990; Cuccovillo
and Coop, 1999). At a fundamental level, there have
been useful advances in formulating constitutive
models incorporating the influence of soil structure,
such as those proposed by Wheeler (1997), Rouainia
and Muir Wood (2000), Kavvadas and Amorosi
(2000), Liu and Carter (2002) and Carter and Liu
(2005). However, our understanding of the influ-
ence of soil structure on the bearing capacity of shal-
low foundations is still relatively modest.
One of the important features of the mechanical
behaviour of structured soils is the occurrence of a
destructuring phase as these soils are loaded. Dur-
ing this phase, the structure of the soil may be com-
pletely or partially lost and only a small change in
stress state may cause very large strains. Conse-
quently, significant errors in prediction of founda-
tion behaviour can arise if the influence of soil struc-
ture is not incorporated into these predictions.
Well-established formulae for determination of
the undrained bearing capacity of shallow circular
foundations have been proposed by Terzaghi and
Peck (1967), Salençon and Matar (1982), Kusakabe
et al. (1986) and Tani and Craig (1995). These
equations take into account the foundation shape,
size, depth of embedment and variation of soil prop-
erties with depth, but they do not directly take into
account the influence of soil structure on the bearing
resistance.
Geotechnical engineers have long experience in the
use of a factor of safety in design, and generally they
have been successful in designing shallow founda-
tions on natural and man made soils using this ap-
proach. With this method, the effects of soil struc-
ture on its bearing behaviour are generally ignored.
However, for special cases, such as offshore
structures, where there is often less experience,
incorporating the complex behaviour of structured
soils directly in bearing capacity predictions may be
very important (Leroueil, 2002). Therefore, the
main objective of this paper is to examine the sig-
nificance of soil structure on the undrained bearing
capacity of shallow foundations resting on the sur-
face of structured soil deposits.
A series of numerical simulations has been car-
ried out to investigate the influence of soil structure
on the load-displacement response of shallow foun-
dations. These simulations have been carried out by
incorporating the Structured Cam Clay (SCC) model
(Liu and Carter, 2002; Carter and Liu, 2005), which
is an extension of the widely used Modified Cam
Clay model developed originally by Roscoe and
Burland (1968), into the finite element program
AFENA (Carter and Balaam, 1995) developed at the
University of Sydney. SCC is a relatively simple
elastoplastic model, which is fully defined by rela-
tively few parameters, each of which has a clear
physical meaning and can be conveniently identified
by standard soil mechanics tests.
Based on these numerical simulations, guidelines
are provided to identify when the structural features
of the soil would become important in assessing the
undrained (short-term) bearing capacity of shallow
circular foundations. The factor, Nc, used in the
classical bearing capacity theory has been improved
by incorporating directly the structural features of
the soil to quantify the undrained bearing capacity of
shallow foundations on structured soils.
2 STRUCTURED CAM CLAY MODEL
The Structured Cam Clay model uses six parameters
to define the soil structure in addition to the usual
parameters used to define destructured, reconstituted
soil behaviour in the Modified Cam Clay model
(Roscoe and Burland, 1968). The additional
parameters are b, p co ′ , ω, γ, a and c. The
destructuring index, b, quantifies the rate of de-
structuring with increasing mean stress and
′ defines the size of the initial yield surface. The
p co
two model parameters ω and γ define respectively
the influence of soil structure on the plastic potential
of the soil and the effect of shearing on de-
structuring. The latter is directly proportional to the
value of γ. The two parameters a and c are used to
define the additional voids ratio, ∆ei, sustained by
the soil, which is given by:
b
 p′ 
∆ei = a co  + c (1)
 p ′s 
where p ′s is the size of the current yield surface.
For most natural soils, two of these parameters,
p c′o and ∆ei, are linked, as explained in Section 4.
The yield surface of the SCC model in p′ - q
space is elliptical and passes through the origin,
similar to the Modified Cam Clay model, but non-
associated plastic flow is assumed. In what follows,
all properties of the destructured reconstituted soil
are denoted by the superscript *.
3 FINITE ELEMENT MODEL
The finite element mesh used for the analysis con-
sisted of fifteen-noded cubic strain triangles, and a
16-point Gaussian quadrature was performed. Ac-
cording to Sloan and Randolph (1982), this element
is capable of accurate computations in the fully plas-
tic range during undrained loading in problems
which involve axial symmetry. In all analyses pre-
sented in this paper, it is assumed that the contact
between the footing and the soil is rough.
4 EFFECT OF SOIL STRUCTURE ON BEARING
CAPACITY
To study the effect of soil structure on the bearing
capacity, a parametric study was carried out by vary-
ing the footing diameter and the model parameters
which govern the structural features of the soil for
typical examples of both stiff and soft clays. The
values of other parameters were kept constant and
they are the same as those given in Table 1.
For reconstituted soils loaded under both drained
and undrained conditions, it was found that for a
unique value of the combined parameter γ ′B / p co ′* ,
the non-dimensional bearing pressure q av / pco ′*
plotted against the non-dimensional footing
settlement δ/B was almost the same, irrespective of
the individual values of each variable. Here, qav is
the average applied footing pressure, δ is the footing
settlement, B is the footing diameter and γ′ is the
effective unit weight of the soil.
For a particular soil, the size of the yield surface
′ , is always greater than the
of the structured soil, p co
size of the yield surface of the reconstituted soil,
′* , due to the additional voids ratio sustained by
p co
the soil structure, ∆ei. The variation of p co′ depends
on λ*, κ* and ∆ei, as illustrated in Figure 1, and is
given by:
 ∆e 
′ = pco
p co ′* exp * i *  (2)
λ −κ 
Therefore, for a particular soil the degree of
structure with respect to the reconstituted soil can be
defined by using either p co ′* or ∆e i (λ* − κ * ) . It
′ / p co
was found that the influence of γ, ω and the
destructuring index, b, do not have a significant
influence on the undrained bearing capacity. Thus,
in the parametric study the influence of soil structure
has been studied by varying only ∆ei (λ* − κ * ) .
gradient = κ ∗
p' co1
p' co2 p' co3
∆ e i1
e ∆ e i2
∆ e i3
∗
gradient = λ
c
*
p' co ln(p' )
Figure 1. Variation of p′co with ∆ei for a structured soil
4.1 Bearing capacity of stiff structured clay
Figure 2 illustrates the influence of the degree of soil
structure on the undrained bearing response of a 2 m
diameter circular footing. Clearly, the bearing ca-
pacity reaches an ultimate value during undrained
loading, in contrast to the approximate bilinear load-
displacement response observed during drained
loading of similar materials (Liyanapathirana et al.,
 2003a, 2003b). This indicates that for these cases state and structural parameters of the soil, as shown
the soil beneath the footing should fail in the general below:
shear failure mode. n
q av  ∆e i  γ ′B 
Table 1. Properties of stiff and soft clay = exp
 1 .3  (3)
Property Stiff clay Soft clay
′*
p co  λ* − κ *  p co
′* 
λ* 0.161 0.22
κ* 0.033 0.022 −0.25
M* 1.0 1.3  ∆e i 
where n = 0.23 
* 
ν* 0.2 0.4 *
 λ −κ 
e*cs 2.75 2.86
b 4.0 4.0 The new equation takes into account the influence of
γ 0.5 0.5 soil structure on the bearing capacity.
ω 1.0 1.0
γ′ (kN/m3) 8.19 7.19 50
λ = gradient of the normal compression line in e – ln (p′)
*

space, 40
κ* = gradient of the unloading and reloading line in e –
ln (p’) space, 30

co
*
M* = gradient of the CSL in q - p′ space,

q av /p'
ν* = Poisson’s ratio, 20
e*cs = void ratio at p′ = 1 kPa on the CSL in e - ln (p′) space,
10

With an increase in the degree of soil structure,

0
the bearing capacity increases significantly. For ex-
ample, when ∆ei (λ* − κ * ) increases from 1.07 to
0 5 10 15
*
γ 'B / p' co

2.16, there is a four-fold increase in the bearing ca- ∆ ei/(l*-k*)

D ei/(λ -κ *)
*
= 0.78 ∆Dei/(l*-k*)
ei/(λ -κ *)
*
= 1.07
pacity of the footing. If the undrained bearing ca- * *
∆ ei/(l*-k*)
D ei/(λ -κ *) = 1.75 ∆Dei/(l*-k*)
ei/(λ -κ *) = 2.16
pacity is calculated based on the reconstituted soil *
properties, neglecting the structural properties of the ∆ ei/(l*-k*)
D ei/(λ -κ *) = 2.56
soil, the predicted bearing capacity can be a very Figure 3. Variation of qav /p′*co with ∆ei/(λ*-κ*) for a 2 m di-
low value compared with the true theoretical load ameter circular footing on stiff clay
carrying capacity of the footing.
4.2 Influence of degree of soil structure on
300 undrained shear strength
*
∆ e i /(λ - κ * )=2.56
250
2.16 In the determination of the short term or undrained
200 bearing capacity, the undrained shear strength of the
q av (kPa)

soil, Su, is an important parameter. However, the

150 1.75 undrained shear strength is not an input parameter of
100 the Structured Cam Clay model. For the Modified
1.07
Cam Clay model, which simulates the constitutive
50 0.78
0.0
behaviour of reconstituted soil, the undrained
0
strength can be derived from the following equation
0 1 2 3 4 5 6
(e.g., Zdravkoviċ and Potts, 2003):
δ/Β %
Su = g (θ )cos(θ )
OCR
6
(
1 + 2 K oNC 1 + B 2 × )( )
Figure 2. Variation of qav for different values of ∆ei/(λ -κ ) for* *

a 2 m diameter circular footing on stiff clay κ* (4)

 (
2 1 + 2 K oOC  λ* )
Figure 3 shows the non-dimensional ultimate

( NC
) 2 
 1 + 2 K o OCR 1 + B  ( )
bearing capacity, qu pco′* , obtained from the finite
element analysis for different values of degree of ′
where K oNC = 1 − sin φ′ , K oOC = K oNC OCR sin φ and
soil structure, ∆ei (λ* − κ * ) , and γ ′B / p co
′* for stiff
clay. For the range of γ ′B / pco ′ considered in the
*
B=
(
3 1 − K oNC )
parametric study, it can be seen that the degree of ( )(
g − 30o 1 + 2 K oNC ).
soil structure has a significant influence on the bear-
ing capacity of the foundation. Based on the above θ is the Lode angle, OCR is the overconsolidation
observation, an equation for the bearing capacity ratio, and K oNC and K oOC are the coefficients of earth
calculation can be formulated based on the critical pressure at rest for a normally consolidated and an
overconsolidated soil, respectively. If we assume a
circular yield surface in the deviatoric plane, g(θ) ever, according to Figure 5, in addition to kB/Suo,
can be replaced by M*/√3. qu/Suo depends on the individual values of Suo or in
For a structured soil ∆ei (λ* − κ * ) increases with other words on the OCR of the soil. To sufficient
the degree of soil structure and p′co is related to ∆ei accuracy this effect can be incorporated into the
according to Equation (1). The OCR of the soil is bearing capacity calculation as follows:
also related to soil structure and according to n
Equation (4), Su increases with OCR. Therefore, the qu  S    kB 
= N c s c = 3.25 ln uo  + 13.5  
 

 (6)
influence of the degree of soil structure can be S uo   pa    S uo 
incorporated into the undrained shear strength of the
−0.2
soil via equation (4). S 
where n = 0.14 uo 
 and pa is atmospheric pressure
 pa 
4.3 Bearing capacity of stiff clay based on (101.32 kPa). If a value of Suo is available for the in-
undrained shear strength tact undisturbed soil, i.e., the peak strength of the
Figure 4 shows the variation of Su with depth in the structured soil, then Equation (6) can be used to ob-
structured soil deposit for different values of tain the ultimate bearing capacity. This may be
∆ei (λ* − κ * ) and the corresponding OCR. In each more convenient than using the alternative, Equation
case it is assumed that OCR is independent of depth. (3), which requires direct knowledge of the struc-
The ultimate bearing capacity of a surface foun- tural parameters of the soil.
dation, qu, is normally expressed in the form: 16
qu = N c sc Su o (5) 14
12
where sc is the shape factor, Suo is the undrained 10
q u /S uo

shear strength of the soil at the ground surface, and 8

Nc is the bearing capacity factor. 6
4
0
( )
ei λ* − κ* = 0.78
OCR = 1.5, ∆Dei/(l*-k*) 2
( )
ei λ* − κ* = 1.07
OCR = 2.0, ∆Dei/(l*-k*) 0
2 OCR = 4.0, Dei/(l*-k*)) = 1.75
( * *
∆ei λ − κ 0 2 4 6 8
e (λ − κ ) = 2.16
* *
OCR = 6.0, ∆Dei/(l*-k*)
i
kB/S uo
e (λ − κ ) = 2.56
* *
OCR = 9.0, ∆Dei/(l*-k*)
i OCR = 1.5 OCR = 2
4
Depth (m)

OCR = 4 OCR = 6
OCR = 9 Kusakabe et al. (1986)
Salencon and Matar (1982) Tani and Craig (1995)
6 Terzaghi and Peck (1967)

8 Figure 5. Comparison of bearing capacity obtained from the

10
0 50 100 150 200
Undrained Shear Strength, S u (kPa)
Figure 4. Undrained shear strength profiles for stiff clay
Based on plasticity theory, several authors have
recommended values for Nc and sc (or their product)
as functions of kB/Suo, where k is the gradient of Su
over the depth of the soil deposit, e.g., Davis and
Booker (1973), Salençon and Martar (1982), Kusa-
kabe et al., (1986) and Tani and Craig (1995).
In Figure 5 the bearing capacity obtained from se-
lected plasticity solutions and the finite element
analysis of the structured soil are compared. Ac-
cording to Terzaghi and Peck (1967), the value of Nc
is 5.14 for a soil of uniform strength. Three other
recommended values are also plotted in Figure 5. In
these solutions, qu/Suo increases with kB/Suo. How-
SCC model with solutions based on plasticity theory
250 4.4 Bearing capacity of soft clay
Usually soft clay deposits have a crust above the
ground water table and are close to normally
consolidated below the water table. In this section
the influence of a surface crust is studied for a
shallow, 2 m diameter circular foundation, and it is
assumed that the water table is 2 m below the
ground surface. The undrained shear strength
profiles for the four cases considered are shown in
Figure 6. In the case of OCR = 1, there is no surface
crust. In the other three cases, the OCR values at the
ground surface are 3, 6 and 9, respectively, and each
has a polynomial variation of strength above the
water table. Below the water table however, it is
assumed OCR = 1 for all four cases. Usually in soft
clays, this surface crust provides the necessary
strength to carry surface loading (e.g., Zdravkoviċ
and Potts, 2003). In each case, ∆ei was allowed to
vary through the crust and values were calculated ever, if the OCRo at the surface is less than or equal
using Equation (1), assuming p′*co = 10 kPa. Below to about 3, the bearing capacity calculated using the
the surface crust ∆ei was assumed to have constant Terzaghi and Peck equation matches reasonably well
value of 0.1. with the finite element results.
Figure 7 shows the variation of average bearing
pressure with footing displacement for the four cases 100
given in Figure 6. Clearly, the presence of the sur- OCRo = 9.0
face crust significantly increases the bearing capac- 80
ity of the footing. When the OCR at the surface of
the crust is 6, the bearing capacity has nearly dou- 60 OCRo = 6.0

q u (kPa)
bled compared to the bearing capacity obtained
without any crust. 40 OCRo = 3.0
Unlike the case of stiff clay, for these soft clays,
the bearing capacity does not reach an ultimate value
20
but continues to increase slowly with increasing OCR = 1.0, No crust
footing displacement.
For soft clay deposits, it is not possible to com- 0
pute qu using the methods proposed by Salençon and 0 1 2 3 4 5 6
Matar (1982), Kusakabe et al. (1986) and Tani and δ/Β %
Craig (1995), because in all these methods the bear- Figure 7. Variation of bearing capacity with and without sur-
ing capacity is a function of kB/Suo. For these soft face crust (B = 2.0 m)
clays, the distribution of Su is not a simple linear
increase with depth, as assumed in the derivation of
200
those methods. Therefore, the finite element results
have been compared only with the predictions of Finite element
Terzaghi and Peck (1967) method. 150
Terzaghi and Peck (Suo)
(S uo )
q u (kPa)

100 Terzaghi and Peck (Suave)

(S uave )
0

50
2

0
4
De pth (m)

OCR = 1.0 0 2 4 6 8 10
OCRo = 3.0 OCR
6 OCR0 = 6.0 Figure 8. Bearing capacity for different OCR values at the sur-
OCRo = 9.0
face of the soft clay deposit

8
10
0 10 20 30
Undrained Shear Strength, S u (kPa)
Figure 6. Undrained shear strength profile for soft clay
Figure 8 shows the bearing capacity obtained
from the finite element analysis when the footing
displacement is 10% of the footing diameter, and
from the Terzaghi and Peck (1967) method. In the
Terzaghi and Peck method (Equation 5), the bearing
capacity has been calculated using the undrained
shear strength at the ground surface and the average
undrained shear strength over the depth interval
equivalent to the diameter of the footing. It is clear
that where there is a crust with OCRo greater than
about 3, the bearing capacity equation proposed by
Terzaghi and Peck (1967) significantly over predicts
the load carrying capacity of the foundation. How-
5 FAILURE MECHANISMS
40
Figures 9 (a) and (b) show, respectively, the incre-
mental soil displacements for stiff and soft clay de-
posits when the cumulative displacement of the foot-
ing is 7.5% of the footing diameter. The footing
considered has a diameter of 2 m. The stiff clay de-
posit considered has an OCR of 6 throughout the
soil deposit and the soft clay deposit considered has
an OCR of 6 at the ground surface.
It can be seen that in stiff and soft clays, the fail-
ure mechanisms are not the same. In stiff clay, soil
beneath the centre of the footing moves predomi-
nantly in the vertical direction, but towards the outer
edge of the footing soil movement is predominantly
in the radial direction and the soil heaves around the
footing, similar to a general shear failure. The load
displacement response given in Figure 2 also con-
firms this, as the bearing pressure increases with
footing penetration and reaches an ultimate value.
 In soft clay, the soil flow pattern shown in Figure
9 (b) is different to that observed for stiff clay. Soil
flow beneath the footing is predominantly in the ver-
tical direction confined largely to a zone beneath the
footing. Although some vectors are at an angle to
the vertical, they do not show a flow pattern in a ra-
dial shearing zone. According to the load displace-
ment response shown in Figure 7, no visible collapse
is observed and a continuous increase in the vertical
load is needed to maintain the footing movement in
the downward direction. Therefore, it can be con-
cluded that the deformation of the structured soft
clay beneath the footing occurs predominantly as lo-
cal shear failure under undrained conditions.
Figure 9 (a). Incremental displacements at δ/B = 7.5% in stiff
clay (OCR = 6)
Figure 9 (b). Incremental displacements at δ/B = 7.5% in soft
clay (OCRo = 6)
6 CONCLUSIONS
Bearing capacity of circular footings resting on
structured soil has been studied using a finite ele-
ment model based on the Structured Cam Clay soil
model. Both stiff and soft structured clays have
been considered. The parametric study carried out
for stiff clay shows that the influence of the soil
structure significantly increases the mobilised bear-
ing resistance of surface footings. An equation
based on the critical state soil parameters has been
proposed to predict the undrained bearing capacity
of shallow foundations on stiff clay.
View publication stats
Bearing capacity obtained from the model for
stiff clay has been compared with published solu-
tions. A modified bearing capacity factor, Nc (or
strictly Ncsc), has been introduced, taking into ac-
count the complex behaviour of structured soil. Fi-
nally it is concluded that the failure of footings on
stiff structured clays occurs mostly as a result of
general shear failure but in soft structured clays,
failure occurs as a local shear failure.
REFERENCES
Burland, J.B. (1990). On the compressibility and shear strength of
natural soils. Géotechnique, 40(3), 329-378.
Carter, J.P. and Balaam, N.P. (1995). AFENA User Manual, Ver-
sion 6, University of Sydney.
Carter, J.P. and Liu, M.D. (2005). Review of the structured cam
clay model, Geo-Frontiers, Austin, Texas, Jan 24-25, 2005.
Cuccovillo, T. and Coop M.R. (1999). On the mechanics of struc-
tured sands. Géotechnique, 49(6), 741-760.
Davis, E.H. and Booker, J.R. (1973). The effect of increasing
strength with depth on the bearing capacity of clays.
Géotechnique, 23(4), 551-562.
Kavvadas, M. and Amorosi, A. (2000). A constitutive model for
structured soils. Géotechnique, 50(3), 263-273.
Kusakabe, O., Suzuki, H. and Nakase, A. (1986). An upper bound
calculation on bearing capacity of a circular footing on a non-
homogeneous clay, Soils and Foundations, 26(3), 143-148.
Leroueil S. (2002). Well known aspects of soil behaviour so often
neglected. Annual workshop of the Centre for Offshore Foun-
dation Systems, University of Western Australia, 11-12 July
2002.
Liu, M.D. and Carter, J.P. (2002). Structured Cam Clay model.
Canadian Geotechnical Journal, 39(5), 1313-1332.
Liyanapathirana, D.S., Liu, M.D., Airey, D.W. and Carter, J.P.
(2003a). Predicting the behaviour of foundations on structured
soils. XIIIth European Conference on Soil Mechanics and Geo-
technical Engineering, Czech Republic, 2, 255-260.
Liyanapathirana, D.S., Carter, J.P. and Airey, D.W. (2003b). Bear-
ing response of shallow foundations on structured soils. In
Proceedings, International Conference on Foundations: Inno-
vations, Observations, Design and Practice, Dundee, 521-530.
Roscoe, K.H. and Burland, J.B. (1968). On the generalised stress-
strain behaviour of wet clay. Engineering Plasticity, Cam-
bridge University Press, Cambridge, 535-609.
Rouainia, M. and Muir Wood, D. (2000). A kinematic hardening
model for natural clays with loss of structure. Geotechnique,
50(2), 153-164.
Salençon, J. and Matar, M. (1982). Bearing capacity of axially
symmetrical shallow foundations. J. Mechanique Theorique at
Appliquee, 1(2), 237-267.
Sloan, S.W. and Randolph, M.F. (1982). Numerical prediction of
collapse collapse loads using finite element methods, Interna-
tional Journal of Numerical and Analytical methods in Ge-
omechanics, 6, 47-76.
Tani, K. and Craig, W.H. (1995). Bearing capacity of circular
foundations on soft clay of strength increasing with depth,
Soils and Foundations, 35(4), 21-35.
Terzaghi, K. and Peck, R.B. (1967). Soil Mechanics in Engineer-
ing Practice, John Wiley and Sons, NY.
Wheeler, S.J. (1997). A rotational hardening elasto-plastic model
for clays. 14th International Conference on Soil Mechanics
and Foundation Engineering, 1, 431-434.
Zdravkoviċ, L. and Potts, D.M. (2003). Numerical study of the ef-
fect of preloading on undrained bearing capacity, International
Journal of Geomechanics, ASCE, 3(1), 1-10.