Ocean Engineering 119 (2016) 4757

Contents lists available at ScienceDirect

Ocean Engineering
journal homepage: www.elsevier.com/locate/oceaneng

Undrained bearing capacity of ring foundations on two-layered clays

Joon Kyu Lee a, Sangseom Jeong b,n, Julie Q. Shang c
a
Department of Civil Engineering, University of Seoul, 163 Seoulsiripdae-ro, Dongdaemun-gu, Seoul 02504, Republic of Korea
b
Department of Civil and Environmental Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of Korea
c
Department of Civil and Environmental Engineering, Western University, London, Ontario, Canada N6A 5B9

art ic l e i nf o a b s t r a c t

Article history: This paper presents the results of a numerical investigation into the undrained vertical bearing capacity
Received 2 October 2015 of rough ring foundations resting on two-layered clays of both homogeneous and linearly increasing
Received in revised form shear strength proles. Small displacement nite element predictions are compared with the available
11 March 2016
empirical, analytical and numerical solutions, and expressed in the familiar form of bearing capacity
Accepted 20 April 2016
factors reecting the coupling effects of the dimensionless parameters related to foundation internal
opening, relative top layer thickness, strength difference between two layers, and strength non-homo-
Keywords: geneity. The depth beyond which the shear strength of the bottom layer does not affect the bearing
Ring foundation capacity, dened here as critical depth ratio, is identied. The failure mechanisms of ring foundations are
Bearing capacity
also discussed in terms of the displacement pattern.
Finite element method
& 2016 Elsevier Ltd. All rights reserved.
Layered soil
Clay
Offshore engineering

1. Introduction of ring foundations subjected to eccentric loads.

In geoengineering practice, ring foundations are often utilized
as bases for offshore and onshore axisymmetrical structures such
as wind turbines, annular platforms, transmission towers, water
tanks and silos because the foundations with internal opening are
more economical than mat foundations (Bowles, 1997). The in-
teraction of such foundations with supporting medium is affected
by footing geometry as well as soil and loading conditions. A clear
understanding of bearing responses for the ring foundations is
indispensable to a successful design.
Several studies have been performed to predict the bearing ca-
pacity behavior of ring foundations. This has been done through using
the plastic stress eld approach constructed by the method of char-
acteristics (Kumar and Ghosh, 2005), limit equilibrium theory (Kar-
aulov, 2005, 2006), nite difference method (Zhao and Wang, 2008;
Benmebarek et al., 2012a) and nite element method (Choobbasti
et al., 2010; Lee et al., 2016). The laboratory test results of model ring
plates in sand were reported by Ohri et al. (1997). On the other hand,
some attempts have been made to analyze the geotechnical stability
of ring foundations on reinforced soil. This problem was studied ex-
perimentally and numerically by Boushehrian and Hataf (2003) who
explored the effects of number and spacing of geogrid reinforcement
layers. El Sawwaf and Nazir (2012) investigated the bearing capacity
n
Corresponding author.
E-mail address: soj9081@yonsei.ac.kr (Jeong).
Seabed sediments and natural soils usually comprise discrete
layers of different thicknesses and properties, namely soil prole
beneath foundations is heterogeneous. The stability for multi-
layered soils is particularly more important for structures with
large shallow foundations, in which the foundation loads extend to
great depth below the surface. Especially, the potential for un-
expected punching shear failure of offshore foundations may occur
during installation in layered clays (Poulos 1988; Kim et al., 2015).
The bearing capacity problems in a simple two-layered clay sys-
tem have been extensively studied by numerous researchers:
Button (1953) determined upper bound solutions for surface strip
foundations by assuming a simple circular rupture surface. On the
basis of the limit equilibrium method, Reddy and Srinivasan (1967,
1971) calculated bearing capacity factors for spread foundations
resting on anisotropic clays. Semi-empirical solutions for various
shapes of foundations (i.e., strip, rectangular, square and circular
foundations) were proposed by Brown and Meyerhof (1969), Vesic
(1973) and Meyerhof and Hanna (1978), and primarily based on
experimental results. The kinematic approaches of limit analysis
for strip foundations were developed by Florkiewicz (1989), Mi-
chalowski (2002) and Huang and Qin (2009), although they pos-
tulated slightly different failure mechanisms of two-layered sys-
tems. Merield et al. (1999) employed the upper and lower bound
theorems of limit analysis in conjunction with nite element (FE)
to evaluate the plasticity solutions for strip foundations. Goss and
Grifths (2001) presented the bearing capacity calculations of strip

http://dx.doi.org/10.1016/j.oceaneng.2016.04.019
0029-8018/& 2016 Elsevier Ltd. All rights reserved.
48 J.K. Lee et al. / Ocean Engineering 119 (2016) 4757
foundations from the small displacement FE analysis, followed by
the extensive works of Zhu and Michalowski (2005) and Merield
and Nguyen (2006) for circular, square and rectangular founda-
tions. The large deformation FE analyses, simulating the con-
tinuously penetration process of strip, square and circular foun-
dations into strong over weak clays, were suggested by Wang and
Carter (2002) and Yu et al. (2011). Boulbibane and Ponter (2005)
provided the collapse loads for strip foundations by means of the
linear matching method. Kuo et al. (2009) assessed the feasibility
of applying the articial neural network (ANN) technique for es-
timating the bearing capacity of strip foundations. More recently,
the inuences of embedment and strength non-homogeneity on
the bearing response of strip foundations were addressed by
Bandini and Pham (2011) and Benmebarek et al. (2012b), respec-
tively. Nevertheless, it can be pointed out that almost all of these
studies are limited to solid foundations without any holes on two-
layered clays, although the methods of analysis varied.
The objective of this study is to present calculations of un-
drained vertical bearing capacity factors Nc* for ring rough foun-
dations on two-layered clays, accounting for external-to-internal-
radius ratio, relative top layer thickness, strength ratio of the two
clays, and strength non-homogeneity. A numerical solution is es-
tablished using a small displacement FE method. The values of Nc*
obtained in this study are compared with existing solutions from
literature and the FE displacement patterns at collapse for the ring
foundations are examined.
2. Problem statement
Fig. 1 illustrates the geometry and parameters of the bearing
capacity problem considered. A rigid ring foundation is placed on a
two-layered clay medium with horizontal ground surface. The ring
foundation is denoted as external and internal radii R0 and Ri,
respectively. In this study, ve external-to-internal- radius ratios
(Ri/R0 0, 0.25, 0.33, 0.5, 0.75) were taken into account, which
covers most problems of practical interest (Benmebarek et al.,
2012a). The foundation roughness is idealized as perfectly rough,
which does not allow any relative movement at all along the soil-
foundation interface. A similar assumption is used in the literature
(Houlsby and Martin, 2003; Edwards et al., 2005; Merield and
Nguyen, 2006; Yu et al., 2011) for undrained bearing capacity of
shallow foundations. The top layer of clay with thickness H and
non-uniform undrained shear strength su (z)su0  t kz is under-
lain by a clay layer of (nominally) innite depth and non-uniform
undrained shear strength su (z) su0  b k(z  H); where su0  t and
Fig. 1. Ring foundation on two-layered clays.
su0  b are the undrained shear strength at foundation level and
topbottom layer interface, respectively. Note that the stability of a
two-layer system in which the underlying layer is innite is often
encountered in engineering practice, especially in offshore foun-
dation and highway designs. k is the rate of increase of shear
strength su with depth z, i.e., k dsu/dz. For a foundation of dia-
meter D ( 2R0), it is convenient to quantify the degree of non-
homogeneity beneath the foundation in terms of the dimension-
less ratio kD/su0-t, accounting for values ranging from 0 (uniform)
to 4 (Houlsby and Martin, 2003). It is apparent that the effect of
strength non-homogeneity is especially signicant for large
foundations on weak clays. The strength difference between the
two distinct layers is characterized by the ratio of the top layer
strength to the bottom layer strength su0-t/su0-b. The strength ratio
of the two clays varies from 0.25 to 5, indicating that su0  t
/su0  b o1 corresponds to the cases of a soft clay over a stiff clay
layer, whereas su0  t/su0  b 41 corresponds to the reverse. To ob-
tain solutions to a range of the two-layered clay problem, the re-
lative thickness of the top layer dened as the ratio of the top layer
thickness to the footing diameter H/D is introduced, encompassing
values of H/D varying from 0.125 to 1 (Yu et al., 2011). The specic
cases of single layer (H/D innite) with uniform and non-uniform
strength are also considered. The soil strength proles for para-
metric study is visible in Fig. 2. For a two-layered clay system, the
undrained bearing capacity factor Nc* of ring foundations can then
be expressed as
Q R kD su0 t H
Nc*= =f i , , ,
(R 02R i2 ) su0 t R 0 su0 t su0 b D (1)
where Q is the vertical load given by
R0
Q = 2 q Ri
rdr
(2)
where q is the unit bearing pressure acting on the footing and r is
the radial distance of a point from the axis of symmetry.
3. Finite element analysis
Small displacement nite element analyses were carried out
with the commercially available software PLAXIS version 2012
(Brinkgreve et al., 2012). The nite element model consisted of two
parts: the soil and the foundation. Six-node triangular elements
were used to represent soil, while foundation was composed by six-
node triangular plate elements. Due to the symmetry in geometry
and loading congurations, only half of the problem domain was
 J.K. Lee et al. / Ocean Engineering 119 (2016) 4757 49

(a) su0-t (b) su0-t (c) su0-t

su su su

H H
su0-b
su0-b

z z z

(d) su0-t (e) su0-t (f) su0-t

su su su

H 1 H 1
k k
su0-b
su0-b

1
1 1
k
k k

z z z
Fig. 2. Undrained shear strength proles considered (modied from Benmebarek et al. (2012a)).

discretized. An axisymmetric stress state was imposed to simulate R0 (= D/2) 6D

the ring foundations with a vertical line of symmetry at the center
of the foundation. To avoid boundary effect, both the radial and
vertical boundaries were placed 6D away from the edges of the
foundation, with radially constrained nodes at the sides and fully
constrained nodes at the base. The boundary conditions of the soil
domain are comparable with that adopted by Gourvenec et al.
(2006) to reach an acceptable accuracy for small displacement nite
element analyses. An iterative study to mesh renement was un-
dertaken to enable accurate calculations of foundation bearing ca-
pacity. The total number of elements ranged from 6815 to 6982,
depending on problem geometries. Fig. 3 shows a FE mesh for the 6D
case of Ri/R0 0 and H/D0.125, su0 t/su0 b 5. The mesh consists
of 6281 elements and the size of elements close to the foundation is
approximately 0.05 m.
The soil was modeled as a linear elastic perfectly plastic Tresca
material according to MohrCoulomb yield criterion with an as-
sociated ow rule. Poisson's ratio of s 0.495 and friction and
dilation angles of 0 were prescribed to simulate the un-
drained soil response. Bulk unit weight and soil stiffness Eu/su
were set to be 0 (weightless soil) and 500, regardless of depth.
The undrained shear strength su was assumed to be 10 kPa. It is
well known that the undrained bearing capacity of surface
7D
foundation is independent on the unit weight and stiffness of
Fig. 3. Typical nite element mesh.
50 J.K. Lee et al. / Ocean Engineering 119 (2016) 4757

soils (Yun and Bransby, 2007; Shaiu et al., 2011). Isotropic initial Table 2
stress was adopted, with the coefcient of lateral earth pressure Undrained bearing capacity factor Nc* for ring foundations in two-layered clays
(kD/su0-t 1).
K0 of unity.
The foundation was taken to be a non-porous linear elastic H/D su0-t/su0-b Ri/R0
material with 1 m thickness and Young's modulus for concrete Ec
30 GPa. The external radius was kept constant at 4 m and the 0.00 0.25 0.33 0.50 0.75

internal radius was varied. The foundations were initially situated 0.125 0.25 7.91 6.80 6.53 6.07 5.63
horizontally on the soil, and the interface between the foundation 0.5 7.91 6.80 6.53 6.07 5.63
and soil was full bonded, preventing slip or separation occurring 0.8 7.31 6.72 6.51 6.06 5.63
under tension, as would be proper for concrete foundations under 1 6.62 6.27 6.12 5.89 5.63
2 4.68 4.69 4.63 4.60 5.33
short-term loading. A uniform vertical load was applied to the rigid 3 3.88 3.96 3.95 3.97 4.76
foundation, resulting in uniform vertical motion of the entire 4 3.42 3.54 3.55 3.61 4.41
foundation. The failure of foundations is clearly indicated by a 5 3.12 3.25 3.28 3.36 4.16
plastic plateau in the load-displacement response. 0.25 0.25 7.03 6.54 6.36 6.04 5.63
0.5 7.03 6.54 6.36 6.04 5.63
0.8 7.00 6.54 6.36 6.04 5.63
1 6.68 6.50 6.36 6.04 5.63
2 5.42 5.69 5.77 6.01 5.63
4. Verication
3 4.78 5.09 5.25 5.62 5.63
4 4.37 4.66 4.88 5.31 5.63
The results from nite element analyses are summarized in 5 4.09 4.38 4.61 5.06 5.63
Tables 1-4, in which the bearing capacity factor Nc*, expressed in 0.5 0.25 6.99 6.54 6.36 6.04 5.63
0.5 6.99 6.54 6.36 6.04 5.63
Eq. (1), is a function of strength homogeneity kD/su0  t for the
0.8 6.99 6.54 6.36 6.04 5.63
1 6.99 6.54 6.36 6.04 5.63
2 6.83 6.54 6.36 6.04 5.63
3 6.50 6.54 6.36 6.04 5.63
Table 1
4 6.20 6.56 6.36 6.04 5.63
Undrained bearing capacity factor Nc* for ring foundations in two-layered clays
5 5.96 6.39 6.36 6.04 5.63
(kD/su0-t 0).
0.75 0.25 6.99 6.54 6.36 6.04 5.63
0.5 6.99 6.54 6.36 6.04 5.63
H/D su0-t/su0-b Ri/R0
0.8 6.99 6.54 6.36 6.04 5.63
1 6.99 6.54 6.36 6.04 5.63
0.00 0.25 0.33 0.50 0.75
2 6.99 6.54 6.36 6.04 5.63
3 6.99 6.54 6.36 6.04 5.63
0.125 0.25 7.44 6.39 6.18 5.71 5.40
4 6.99 6.54 6.36 6.04 5.63
0.5 7.32 6.38 6.17 5.71 5.40
5 6.99 6.54 6.36 6.04 5.63
0.8 6.84 6.37 6.16 5.71 5.40
1 0.25 6.99 6.54 6.36 6.04 5.63
1 6.08 5.93 5.81 5.62 5.40
0.5 6.99 6.54 6.36 6.04 5.63
2 3.84 4.06 4.11 4.20 4.98
0.8 6.99 6.54 6.36 6.04 5.63
3 2.88 3.09 3.23 3.44 4.29
1 6.99 6.54 6.36 6.04 5.63
4 2.36 2.53 2.67 2.98 3.86
2 6.99 6.54 6.36 6.04 5.63
5 2.04 2.18 2.30 2.66 3.58
3 6.99 6.54 6.36 6.04 5.63
0.25 0.25 6.28 5.93 5.81 5.62 5.40
4 6.99 6.54 6.36 6.04 5.63
0.5 6.28 5.93 5.81 5.62 5.40
5 6.99 6.54 6.36 6.04 5.63
0.8 6.28 5.93 5.81 5.62 5.40

1 6.08 5.93 5.81 5.62 5.40
Note: Ri/R0 0: circular foundation.
2 4.52 4.80 5.04 5.38 5.40
3 3.60 3.86 4.07 4.72 5.40
4 3.08 3.30 3.48 4.09 5.40
5 2.72 2.93 3.08 3.66 5.40 Table 3
0.5 0.25 6.08 5.93 5.81 5.61 5.40 Undrained bearing capacity factor Nc* for ring foundations in two-layered clays
0.5 6.08 5.93 5.81 5.61 5.40 (kD/su0-t 2).
0.8 6.08 5.93 5.81 5.61 5.40

1 6.08 5.93 5.81 5.61 5.40 H/D su0-t/su0-b Ri/R0
2 5.60 5.93 5.81 5.61 5.40
3 4.84 5.18 5.46 5.61 5.40 0.00 0.25 0.33 0.50 0.75
4 4.32 4.61 4.88 5.61 5.40
5 3.96 4.20 4.44 5.61 5.40 0.125 0.25 8.28 7.18 6.90 6.42 5.84
0.75 0.25 6.08 5.92 5.81 5.61 5.40 0.5 8.28 7.18 6.90 6.42 5.84
0.5 6.08 5.92 5.81 5.61 5.40 0.8 7.64 7.03 6.82 6.41 5.84
0.8 6.08 5.92 5.81 5.61 5.40 1 7.04 6.59 6.42 6.16 5.84

1 6.08 5.92 5.81 5.61 5.40 2 5.28 5.14 5.03 4.94 5.64
2 6.08 5.92 5.81 5.61 5.40 3 4.52 4.49 4.40 4.37 5.18
3 5.89 5.92 5.81 5.61 5.40 4 4.12 4.11 4.06 4.04 4.86
4 5.45 5.68 5.81 5.61 5.40 5 3.80 3.86 3.82 3.82 4.64
5 4.99 5.35 5.65 5.61 5.40 0.25 0.25 7.64 7.04 6.82 6.41 5.84
1 0.25 6.08 5.92 5.81 5.61 5.40 0.5 7.64 7.04 6.82 6.41 5.84
0.5 6.08 5.92 5.81 5.61 5.40 0.8 7.52 7.03 6.82 6.41 5.84
0.8 6.08 5.92 5.81 5.61 5.40 1 7.20 7.02 6.82 6.41 5.84

1 6.08 5.92 5.81 5.61 5.40 2 6.12 6.32 6.38 6.41 5.84
2 6.08 5.92 5.81 5.61 5.40 3 5.60 5.80 5.97 6.27 5.84
3 6.08 5.92 5.81 5.61 5.40 4 5.24 5.57 5.70 6.04 5.84
4 6.04 5.92 5.81 5.61 5.40 5 5.00 5.33 5.47 5.87 5.84
5 6.00 5.92 5.81 5.61 5.40 0.5 0.25 7.64 7.03 6.82 6.40 5.84
0.5 7.64 7.03 6.82 6.40 5.84

Note: Ri/R0 0: circular foundation; su0  t/su0  b 1: single homogeneous layer 0.8 7.64 7.03 6.82 6.40 5.84
corresponding to the soil proles illustrated in Fig. 2(a).
 J.K. Lee et al. / Ocean Engineering 119 (2016) 4757 51

Table 3 (continued ) discrete values of top layer thickness H/D, strength ratio of the two
clays su0  t/su0-b and external-to-internal-radius ratio Ri/R0. A total
H/D su0-t/su0-b Ri/R0
of 800 combinations for the four dimensionless parameters are
0.00 0.25 0.33 0.50 0.75 given in the tables.
Fig. 4 compares the bearing capacity factors Nc* of a surface rough
1 7.64 7.03 6.82 6.40 5.84
circular foundation (Ri/R0 0) on single homogeneous clay (H/D1
2 7.64 7.03 6.82 6.40 5.84
3 7.64 7.03 6.82 6.40 5.84 and kD/su0 t 0) from the FE analysis with existing solutions for
4 7.48 7.03 6.82 6.40 5.84 equivalent conditions. Theoretical exact solutions of 6.05 using Tresca
5 7.32 7.03 6.82 6.40 5.84
material were determined by many researchers (Meyerhof, 1951;
0.75 0.25 7.64 7.03 6.82 6.40 5.84
0.5 7.64 7.03 6.82 6.40 5.84 Shield, 1955; Cox, 1961). Semi-empirical solutions were reported by
0.8 7.64 7.03 6.82 6.40 5.84
1 7.64 7.03 6.82 6.40 5.84
2 7.64 7.03 6.82 6.40 5.84
3 7.64 7.03 6.82 6.40 5.84
4 7.64 7.03 6.82 6.40 5.84
5 7.64 7.03 6.82 6.40 5.84
1 0.25 7.64 7.03 6.82 6.40 5.84
0.5 7.64 7.03 6.82 6.40 5.84
0.8 7.64 7.03 6.82 6.40 5.84
1 7.64 7.03 6.82 6.40 5.84
2 7.64 7.03 6.82 6.40 5.84
3 7.64 7.03 6.82 6.40 5.84
4 7.64 7.03 6.82 6.40 5.84
5 7.64 7.03 6.82 6.40 5.84

Note: Ri/R0 0: circular foundation.

Fig. 4. Comparison of bearing capacity factors for surface rough circular foundation
Table 4
on single homogeneous clay (H/D 1 and kD/su0  t 0): EX exact solution;
Undrained bearing capacity factor Nc* for ring foundations in two-layered clays
SE semi-empirical solution; MC method of characteristics; LB lower bound
(kD/su0-t 4).
solution; UB upper bound solution; FE nite element method.

H/D su0-t/su0-b Ri/R0

0.00 0.25 0.33 0.50 0.75

0.125 0.25 9.17 7.88 7.58 7.03 6.22

0.5 9.16 7.87 7.58 7.03 6.22
0.8 8.30 7.60 7.37 6.97 6.22
1 7.72 7.18 6.98 6.69 6.22
2 6.25 5.90 5.74 5.58 6.19
3 5.59 5.33 5.19 5.06 5.86
4 5.19 5.00 4.88 4.78 5.63
5 4.93 4.78 4.67 4.58 5.46
0.25 0.25 8.78 7.84 7.56 7.03 6.22
0.5 8.78 7.84 7.56 7.03 6.22
0.8 8.51 7.83 7.56 7.03 6.22
1 8.24 7.83 7.57 7.03 6.22
2 7.37 7.44 7.45 7.03 6.22
3 6.94 7.12 7.17 7.03 6.22
4 6.68 6.92 6.99 7.03 6.22
5 6.49 6.77 6.85 7.03 6.22
0.5 0.25 8.82 7.84 7.57 7.03 6.22
0.5 8.82 7.84 7.57 7.03 6.22
0.8 8.82 7.84 7.57 7.03 6.22
1 8.82 7.84 7.57 7.03 6.22
2 8.82 7.84 7.57 7.03 6.22
3 8.82 7.84 7.57 7.03 6.22
4 8.82 7.84 7.57 7.03 6.22
5 8.80 7.84 7.57 7.03 6.22
0.75 0.25 8.82 7.84 7.57 7.03 6.22
0.5 8.82 7.84 7.57 7.03 6.22
0.8 8.82 7.84 7.57 7.03 6.22
1 8.82 7.84 7.57 7.03 6.22
2 8.82 7.84 7.57 7.03 6.22
3 8.82 7.84 7.57 7.03 6.22
4 8.82 7.84 7.57 7.03 6.22
5 8.82 7.84 7.57 7.03 6.22
1 0.25 8.82 7.84 7.57 7.03 6.22
0.5 8.82 7.84 7.57 7.03 6.22
0.8 8.82 7.84 7.57 7.03 6.22
1 8.82 7.84 7.57 7.03 6.22
2 8.82 7.84 7.57 7.03 6.22
3 8.82 7.84 7.57 7.03 6.22
4 8.82 7.84 7.57 7.03 6.22
5 8.82 7.84 7.57 7.03 6.22
Fig. 5. Comparison of Nc* values for circular foundations resting on two-layered
Note: Ri/R0 0: circular foundation.
clays (kD/su0  t 0).
52 J.K. Lee et al. / Ocean Engineering 119 (2016) 4757

Table 5 Skempton (1951), Brinch Hansen (1970) and Vesic (1973). In these
Comparison of obtained Nc* values of circular foundation on single non- cases, the Nc* value is underpredicted by 0.8% by Skempton, and
homogeneous claya with those available from literature.
overpredicted by 2%, compared to the exact solution. The method of
kD/su0-t H/D HW TC HU GR This study characteristics was used for the bearing capacity calculation, as for-
mulated by Tani and Craig (1995) and Martin (2001) for application to
1 Innite 6.946 NA NA 6.77 6.98 undrained problems. The stress eld solution of Martin is identical to
2 Innite 7.632 7.70 8.15b 7.45 7.64
the exact solution, although the results obtained by Tani and Craig are
4 Innite 8.734 8.77 9.62b NA 8.78
higher by 4.8%. The Nc* value of 6.08 predicted by the present analysis
Note: NA not available; HW Houlsby and Worth (1983) method of char- is very close to the nite element solution of Edwards et al. (2005), but
acteristics; TC Tani and Craig (1995) lower bound method; HU Hu et al. comparable to the FE results by Gourvenec et al. (2006) and Taiebat
(1999) upper bound method; GR Gourvenec and Randolph (2003) nite
element code ABAQUS.
and Carter (2010). The current value stays well in the range between
a
Single nonhomogeneous layers corresponding to the soil proles illu-
lower and upper bound solutions given by Salgado et al. (2004) with
strated in Fig. 2(d). the usage of the nite element limit analysis.
b
Estimated from plots. Fig. 5 provides the obtained Nc* values of a surface circular

Fig. 6. Variation of Nc* values of ring foundations with strength ratio of two clays su0  t/su0  b (kD/su0  t 0).
 J.K. Lee et al. / Ocean Engineering 119 (2016) 4757 53

footing (Ri/R0 0) on two-layered clays (kD/su0  t 0), together

with other available empirical and numerical solutions for same
foundation and soil conditions. The gure shows the variation of
Nc* values with su0  t/su0  b for the cases of H/D 0.25 and 0.5.
Based on small-scale tests of rough plates, Brown and Meyerhof
(1969) proposed that the bearing capacity factor for two-layered
clays is approximated as
H s
Nc*=3. 0 +6. 05 u0 b 6. 05
D su0 t (3)

It is seen from Fig. 5 that the predictions from Eq. (3) are overly
conservative within the range of clays strength ratio (su0  t/su0  b
0.1255) for which they were dened, underpredicting them
signicantly for higher values of su0  t/su0  b. Vesic (1973) sug-
gested that the Nc* values for the case of su0  t/su0  b o 1 and H/
D r0.25, which were derived by interpolation between rigorous
solution for relevant problems. As indicated in Fig. 5(a), the Vesic
solutions are slightly lower than the present results. Wang and
Carter (2002) provided small displacement FE results for the case
of su0  t/su0  b o1 and H/D 0.5 and 1. As observed in Fig. 5(b), the
FE solutions are found to be generally conservative. However, this
underprediction is not surprising because the bearing capacity
factors were computed from the contribution of smooth base of
the foundation, so they are not directly comparable. Merield and
Nguyen (2006) also reported the FE solutions for wide ranges of H/
D ( 0.125 to 2) and su0-t/su0-b (0.2 to 5), which agree well with
the obtained values of Nc*, hence conrming the reasonable ac-
curacy of current nite element analysis.
For single non-homogeneous clay (H/D 1 and kD/su0  t0), a
comparison of the values of Nc* obtained in this study was made
with the analyses of Houlsby and Worth (1983) using the method
of characteristics, Tani and Craig (1995) using the lower bound
method, Hu et al. (1999) using upper bound method, and Gour-
venec and Randolph (2003) using the ABAQUS software. A com-
parative representation of all these values is given in Table 5. The
present results compare most favorably with the solution of
Houlsby and Worth (1983).

5. Results and discussion

The variation of the bearing capacity factors for different

combinations of su0  t/su0  b and Ri/R0 is shown in Fig. 6 for the
case of kD/su0  t 0 and H/D 0.125 to 1. For the soft layer over-
lying stiff layer (su0  t/su0  b o1), the Nc* values for thin top layers
(e.g., H/D r0.25) increase with decreasing the strength ratio
su0-t/su0-b until it reaches an asymptotic value. That is, when the
thin top soil becomes increasingly weaker than the underlying
strong soil, the contribution by the bottom soil to the bearing
capacity increases. This can be supported by the collapse me-
chanism illustrated in Fig. 7(a). The nite element displacement
diagram indicates that although a zone of plastic failure develops
Fig. 7. FE displacement contours for a ring foundation (Ri/R0 0.25) resting on soft-
within the soft top layer, yielding is still generated in the stiff
over-stiff clays (su0  t/su0  b 0.5): (a) H/D 0.125; (b) H/D 0.5; (c) H/D 1.
underlying layer. This observation is similar to that reported by
Merield and Nguyen (2006) for circular foundations on two-
to the bearing capacity of the thin strong top soil reduces as the
layered clays (su0  t/su0  b o1) with H/Do0.375. The inuence of
underlying soil is increasingly weaker. This can be conrmed by
su0  t/su0  b on Nc* is found to be predominant for the circular
the FE displacement pattern depicted in Fig. 8(a)(c), indicating
foundation (Ri/R0 0) and becomes less sensible for ring founda-
that the collapse mode is characterized by the punching shear
tions with greater values of Ri/R0. It is also noticed that for all H/D,
the ring foundation has lower value of Nc* than the circular failure: the top layer acts as a rigid column of soil that punches
foundation, and the Nc* value of ring foundations decreases con- through into the underlying softer layer, which in turn causes an
tinuously with increasing the value of Ri/R0. In contrast, for the stiff increase in the zone of yielding. The rate of decrease of Nc* with
layer overlying soft layer (su0  t/su0  b 41), the values of Nc* for su0  t/su0  b is found to be more extensive for smaller values of Ri
thin top layers (e.g., H/D r0.75) decrease with an increase in the /R0, which reects the change of failure mechanism from local
strength ratio su0  t/su0  b. It implies that the relative contribution shear (a small well-dened wedge underneath the foundation is
54 J.K. Lee et al. / Ocean Engineering 119 (2016) 4757
Fig. 8. FE displacement contours for ring foundations resting on stiff-over-soft clays (su0  t/su0  b 5 and H/D 0.25): (a) Ri/R0 0; (b) Ri/R0 0.25; (c) Ri/R0 0.5; (d) Ri/R0
0.75.
formed but the slip surface fade into soil mass beyond the edges of
the foundation) to the punching shear, as shown in Fig. 8. It is also
noted that for the case of H/Dr 0.25, the Nc* values of the circular
foundation for a given su0  t/su0  b are always lower than those of
the ring foundations. For cases with H/D Z0.5, however, an op-
posite tendency is also observed, depending on the strength ratio
su0  t/su0  b.
The obtained values of the bearing capacity factors with H/D
and Ri/R0 are plotted in Fig. 9 for the case of kD/su0  t 0 and su0  t
/su0  b 0.25 (soft-over-stiff clay), 1 (homogeneous clay) and 5
(stiff-over-soft clay). From Fig. 9(a) and (c), it is clear that at a
given value of Ri/R0, the behavioral difference with respect to the
top layer thickness H/D is sensitive to the relative strength of two
distinct layers su0  t/su0  b: the results for soft-over-stiff clay
(Fig. 9(a)) reveal that for smaller values of Ri/R0, the decrease of
Nc* with H/D is particularly more appreciable for H/D r 0.25, al-
though the variation in Nc* is insignicant for H/D 4 0.25, irre-
spective of Ri/R0. This nding is in conrmation with the dis-
placement mechanisms shown in Fig. 7: for thicker soft top lay-
ers, failure is likely to be fully contained within the top layer,
indicating that the overall bearing capacity is mainly governed by
the top soil. A similar tendency is also observed for other cases of
su0  t/su0  b o 1, which is not included in Fig. 9. The results for
stiff-over-soft clay (Fig. 9(c)) demonstrates that the value of Nc*
exhibits an increasing tendency as the top layer thickness in-
creases, and this overall trend is also applicable for other cases of
su0  t/su0  b 4 1. On the contrary, for homogeneous clay (Fig. 9(b)),
it is reasonable to expect that there is no effect of H/D on Nc*
values for all Ri/R0.
In Fig. 9, there exists the value of H/D at which the bearing
capacity factor reaches a limiting value. This top layer thickness is
referred to as critical depth ratio Hcr/D, beyond which the strength
of the bottom layer does not inuence the bearing capacity. For
example, the critical depth ratio are 0.5 for su0-t/su0-b 5 and Ri
/R0 0.5. From a practical point of view, the estimation of the
critical depth ratio is important since locating the bottom layer
beyond Hcr/D will not lead to an increase or decrease in foundation
capacity. The critical depth ratio of ring foundations is illustrated
in Fig. 10 as a function of su0  t/su0  b and Ri/R0 for a strength ratio
kD/su0-t 0. At a given value of su0  t/su0  b, the value of Hcr/D de-
creases with an increase in Ri/R0. It is also mentioned that for the
stiff-over-soft clay, the variation of Hcr/D with su0  t/su0  b is pro-
nounced for small values of Ri/R0.
The variation of the bearing capacity factors with different
values of kD/su0  t and Ri/R0 is presented in Fig. 11(a) for the single
non-homogeneous layer (H/D 1). As expected, there is a clear
trend in each case of increasing the value of Nc* as the strength
homogeneity increases. The bearing capacity factors can be de-
scribed as a quadratic function of kD/su0-t. Expressed in terms of
the bearing capacity ratio ( Nc*/Nc*(k 0)):
Nc* kD kD 2
=1 + c1 +c2
*
Nc (k = 0) su0 t su0 t (4)
where c1 and c2 are the coefcients shown in Fig. 11(b) for the
ratios of Ri/R0. As stated in Eq. (4), it is obvious that the value of Nc*
/Nc*(k 0) for homogeneous clay drops towards unity with de-
creasing the value of kD/su0  t. It is worth noting that these coef-
cients are accurate for interpolation only, and should not be used
for extrapolation beyond kD/su0  t 4. The relations of c1 and c2
 J.K. Lee et al. / Ocean Engineering 119 (2016) 4757 55

Fig. 10. Critical depth ratio of ring foundations for kD/su0  t 0.

Fig. 9. Variation of Nc* values of ring foundations with top layer thickness H/D
(kD/su0  t 0).

Fig. 11. Variation of Nc* and Nc*/Nc*(k 0) values of ring foundations with kD/su0  t for
single nonhomogeneous layer (H/D 1).

with Ri/R0 have the form of (R2 0.990 and R2 0.994, respec-
tively) Hence the bearing capacity ratio for the single non-homo-
geneous layer (H/D 1) can be expressed as
R
c1= 0. 1374 i +0. 1472
R0 (5)

R
c2=0. 0110 i 0. 0096
R0 (6)
56 J.K. Lee et al. / Ocean Engineering 119 (2016) 4757

Fig. 12. Variation of Nc* values of ring foundations with kD/su0  t for two-layered clays.

Nc* R kD 6. Conclusions
=1 + 0. 1374 i +0. 1472
Nc*(k = 0) R0 su0 t
The undrained vertical bearing response of rough ring foun-
R kD 2
+ 0. 0110 i 0. 0096 dations resting on two-layered clays has been examined using
R0 su0 t (7) small displacement nite element analyses. A comprehensive
series of undrained bearing capacity factors Nc* are presented for
Meanwhile, the effect of kD/su0 t on the bearing capacity factors clays of both non-homogeneous and linearly increasing shear
of ring foundations on two-layered clays with su0  t/su0 b 0.25, strength proles. The numerical solutions obtained in this study
1 and 5 is explored in Fig. 12 for the extreme cases of H/D0.125 match very well with existing solutions. The following conclusions
and 1. It is evident that an increase in kD/su0  t causes an increase in can be drawn from the present results:
the value of Nc*, which is more substantial for smaller values of Ri
/R0. The bearing capacity factor can be described by the quadratic 1) Regardless of H/D, the value of Nc* of ring foundations is in general
function of kD/su0 t, as applied to the result of Fig. 11. found to increase with decreasing the shear strength ratio su0 t
 J.K. Lee et al. / Ocean Engineering 119 (2016) 4757
/su0 b, which is more prominent for greater values of Ri/R0.
2) For soft-over-stiff clays, the reduction of Nc* value with H/D is
shown for H/D o0.25, which is particularly signicant for small
values of Ri/R0. For stiff-over-soft clays, however, Nc* value has
an increasing tendency as H/D increases: it is especially more
marked for small values of Ri/R0.
3) Provided that the depth of the bottom layer of clay is sufciently
large, the shear strength of this layer will not inuence the
foundation baring capacity. This depth, called as the critical
depth ratio Hcr/D is in the range of 0.10 and 1.05, depending on
the values of Ri/R0 and su0  t/su0  b.
4) The bearing capacity factor Nc* continuously increases with in-
creasing the degree of strength non-homogeneity, which is
more noteworthy at smaller values of Ri/R0. Quadratic function
provides good approximation to the bearing capacity factors for
ring foundations on clays with homogeneous and linearly in-
creasing shear strength proles.
Acknowledgments
The authors acknowledge support in this research for the Na-
tional Research Foundation of Korea (NRF) (Grant no. 2011-
0030040).
References
Bandini, P., Pham, H.V., 2011. Bearing capacity of embedded strip footings in two-
layered clay soils. In: Han, Jie, Alzamora, Daniel E. (Eds.), Geo-Frontiers: Ad-
vances in Geotechnical Engineering, pp. 332341.
Benmebarek, S., Benmoussa, S., Belounar, L., Benmebarek, N., 2012a. Bearing ca-
pacity of shallow foundation on two clay layers by numerical approach. Geo-
tech. Geol. Eng. 30, 907923.
Benmebarek, S., Remadna, M.S., Benmebarek, N., Belounar, L., 2012b. Numerical
evaluation of the bearing capacity factor N of ring footings. Comput. Geotech.
44, 132138.
Boulbibane, M., Ponter, A.R.S., 2005. Limit loads for multilayered half-space using
the linear matching method. Comput. Geotech. 32, 535544.
Boushehrian, J.H., Hataf, N., 2003. Experimental and numerical investigation of the
bearing capacity of model circular and ring footings on reinforced sand. Geo-
text. Geomembranes 21, 241256.
Bowles, J.E., 1997. Foundation Analysis and Design. McGraw-Hill, New York.
Brinch Hansen, J., 1970. A revised and Extended Formula for Bearing Capacity.
Danish Geotechnical Institute, Copenhagen, p. 21, Bulletin No. 28.
Brinkgreve, R.B.J., Engin, E., Swolfs, W.M., 2012. Plaxis User's Manual. Plaxis BV,
Netherlands.
Brown, J.D., Meyerhof, G.G. 1969. Experimental study of bearing capacity in layered
clays. In: Proceedings of the 7th International Conference on Soil Mechanics
and Found Engineering. Mexico City, vol. 2, pp. 4551.
Button, S.J. 1953. The bearing capacity of footings on two-layer cohesive subsoil. In:
Proceedings of the 3rd International Conference on Soil Mechanics and Found
Engineering. Zurich, vol. 1, pp. 332335.
Choobbasti, A.J., Naja, A., Pirzadeh, S., Farrokhzad, F., Zahmatkesh, A., 2010. Nu-
merical evaluation of bearing capacity and settlement of ring footing: case
study of Kazeroon cooling towers. Int. J. Res. Rev. Appl. Sci. 4, 263271.
Cox, A.D., 1961. Axially-symmetric plastic deformation on soils II. Indentation of
ponderable soils. Int. J. Mech. Sci. 4, 371380.
Edwards, D.H., Zdravkovic, L., Potts, D.M., 2005. Depth factors for undrained bearing
capacity. Geotechnique 55, 755758.
El Sawwaf, M., Nazir, A., 2012. Behavior of eccentrically loaded small-scale ring
footings resting on reinforced layered soil. J. Geotech. Geoenviron. Eng. 138,
376384.
Florkiewicz, A., 1989. Upper bound to bearing capacity of layered soils. Can. Geo-
tech. J. 26, 730736.
Goss, C.M., Grifths, D.V., 2001. Rigorous plasticity solutions for the bearing capa-
city of two-layered clays. Geotechnique 51, 179183.
Gourvenec, S., Randolph, M., 2003. Effect of strength non-homogeneity on the
57
shape of failure envelopes for combined loading of strip and circular founda-
tions on clay. Geotechnique 53, 575586.
Gourvenec, S., Randolph, M., Kingsnorth, O., 2006. Undrained bearing capacity of
square and rectangular footings. J. Geotech. Geoenviron. Eng. 6, 147157.
Houlsby, G.T., Martin, C.M., 2003. Undrained bearing capacity factors for conical
footings on clay. Geotechnique 53, 513520.
Houlsby, G.T., Worth, C.P. 1983. Calculation of stresses on shallow penetrometers
and footings. In: Proceedings of the International Union of Theoretical and
Applied Mechanics (IUTAM)/International Union of Geodesy and Geophysics
(IUGG) Symposium on Seabed Mechanics, Newcastle-upon-Tyne, pp. 107122.
Hu, Y., Randolph, M.F., Watson, P.G., 1999. Bearing response of skirted foundation
on nonhomogeneous soil. J. Geotech. Geoenviron. Eng. 125, 924935.
Huang, M., Qin, H., 2009. Upper-bound multi-rigid-block solutions for bearing ca-
pacity of two-layered soils. Comput. Geotech. 36, 525529.
Karaulov, A.M., 2005. Static solution of the liming-pressure problem for ring
foundations on soil beds. Soil. Mech. Found. Eng. 42, 189194.
Karaulov, A.M., 2006. Experimental and theoretical research on the bearing capa-
city of ring foundation beds. Soil. Mech. Found. Eng. 43, 3740.
Kim, Y.H., Hossain, M.S., Wang, D., Randolph, M.F., 2015. Numerical investigation of
dynamic installation of torpedo anchors in clay. Ocean Eng. 108, 820832.
Kumar, J., Ghosh, P., 2005. Bearing capacity factor N for ring footings using the
method of characteristics. Can. Geotech. J. 42, 14741484.
Kuo, Y.L., Jaksa, M.B., Lyamin, A.V., Kaggwa, W.S., 2009. ANN-based model for
predicting the bearing capacity of strip footing on multi-layered cohesive soil.
Comput. Geotech. 36, 503516.
Lee, J.K., Jeong, S., Lee, S., 2016. Undrained bearing capacity factors for ring footings
in heterogeneous soil. Comput. Geotech. 75, 103111.
Martin, C.M., 2001. Vertical bearing capacity of skirted circular foundations on
Tresca soil. In: Proceedings of International Conference Soils Mechanics and
Geotechnical Engineering (ICSMGE), Istanbul, Part 1, pp. 743746.
Meyerhof, G.G., 1951. The ultimate bearing capacity of foundations. Geotechnique 2,
301332.
Meyerhof, G.G., Hanna, A.M., 1978. Ultimate bearing capacity of foundations on
layered soils under inclined load. Can. Geotech. J. 15, 565572.
Merield, R.S., Nguyen, V.Q., 2006. Two- and three-dimensional bearing-capacity
solutions for footings on two-layered clays. Geomech. Geoengin. Int. J. 1,
151162.
Merield, R.S., Sloan, S.W., Yu, H.S., 1999. Rigorous plasticity solutions for the
bearing capacity of two-layered clays. Geotechnique 49, 471490.
Michalowski, R.L., 2002. Collapse loads over two-layer clay foundations soils. Soils
Found. 42, 17.
Ohri, M., Purhit, D., Dubey, M., 1997. Behavior of ring footing on dune sand over-
laying dense sand. In: Proceedings of International Conference on Civil En-
gineering, pp. 2224.
Poulos, H.G., 1988. Marine Geotechnics. Unwin Hyman, London.
Reddy, A.S., Srinivasan, R.J., 1967. Bearing capacity of footings on layered clays. J.
Soil. Mech. Found. Div. 93, 8399.
Reddy, A.S., Srinivasan, R.J., 1971. Bearing capacity of footings on clays. Soils Found.
11, 5164.
Salgado, R., Lyamin, A.V., Sloan, S.W., Yu, H.S., 2004. Two- and three-dimensional
bearing capacity of foundations in clay. Geotechnique 54, 297306.
Shaiu, J.S., Merield, R.S., Lyamin, A.V., Sloan, S.W., 2011. Undrained stability of
footings on slopes. J. Geotech. Geoenviron. Eng. 11, 38913900.
Shield, R.T., 1955. On the plastic ow of metals under conditions of axial symmetry.
Proc. R. Soc. Ser. A 233, 267287.
Skempton, A.W. 1951. The bearing capacity of clays. In: Proceedings of Building
Research Congress, London, Part 1, pp. 180189.
Taiebat, H.A., Carter, J.P., 2010. A failure surface for circular footings on cohesive
soils. Geotechnique 60, 265273.
Tani, K., Craig, W.H., 1995. Bearing capacity of circular foundations on soft clay of
strength increasing with depth. Soils Found. 35, 2135.
Vesic, A.S., 1973. Analysis of ultimate loads of shallow foundations. J. Soil Mech.
Found. Div. 99, 4573.
Wang, C.X., Carter, J.P., 2002. Deep penetration of strip and circular footings into
layered clays. Int. J. Geomech. 2, 205232.
Yu, L., Liu, J., Kong, X., Hu, Y., 2011. Three-dimensional large deformation FE analysis
of square footings in two-layered clays. J. Geotech. Geoenviron. Eng. 137, 5258.
Yun, G., Bransby, M.F., 2007. The horizontal-moment capacity of embedded foun-
dations in undrained soil. Can. Geotech. J. 44, 409424.
Zhao, L., Wang, J.H., 2008. Vertical bearing capacity for ring footings. Comput.
Geotech. 35, 292304.
Zhu, M., Michalowski, R.L., 2005. Bearing capacity of rectangular footings on two-
layer clay. In: Proceedings of 16th International Conference on Soil Mechanics
and Geotechnical Engineering. Geotechnology in Harmony with the Global
Environment, vol. 15, pp. 9971000.