Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

Geomechanics and Geoengineering: An International Journal Vol. 1, No. 2, June 2006, 151--162

Two- and three-dimensional bearing-capacity solutions for footings on two-layered clays

R. S. MERIFIELD* and V. Q. NGUYEN
University of Southern Queensland, Australia
(Received 23 November 2005; in final form 1 February 2006) In this paper, finite element analysis is used to predict the undrained bearing capacity of strip, square and circular footings resting on layered clays. The soil profile consists of two clay layers with different thicknesses and properties. The results are compared with previous solutions for strip footings on layered clays. The bearing-capacity behaviour is discussed and the bearing-capacity factors are given for various cases involving a range of layer thicknesses and properties of the two clay soil layers. Keywords: Deformation; Numerical modelling; Strip footing; Square footing; Circular footing; Layered soils; Bearing capacity; Bearing-capacity factor; Clays

1.

Introduction

The ability to predict the ultimate bearing capacity of a foundation is one of the most significant problems in foundation engineering, and as a consequence, extensive literature exists detailing both theoretical and experimental studies on this topic. Recent reviews of the major contributions to the subject can be found in the works of Tani and Craig (1995) and Poulos et al. (2001). The generally accepted method of estimating the bearing capacity of foundations is to assume that the soil below the foundation, along a critical plane of failure (slip path), is on the verge of failure and to calculate the bearing pressure applied by the foundation required to cause this failure condition. This bearing pressure is termed the ultimate bearing capacity qu. An early approximate solution to bearing capacity was defined by Terzaghi (1943) as general shear failure. The Terzaghi model is applicable to level strip footings placed on or near a level ground surface where the foundation depth is less than the footing width. The solutions are based on the Limit Equilibrium Method, and the ultimate bearing capacity is given as qu Qu 1 cNc qNq BN ; 2 A 1

For the simplest case of a strip footing on homogeneous clay (undrained conditions) without surface surcharge, equation (1) reduces to qu c u N c ; 2

where Nc is the dimensionless bearing-capacity factor (2 + p), an exact solution, first found by Prandtl 1920), and cu is undrained soil shear strength. Equation (2) can be rewritten to include the effect of footing shape as follows: qu cu Nc zcs : The empirical shape factor zcs is given by zcs 1    B Nq : L Nc 4 3

where c is the soil cohesion, g is the soil unit weight below the base of the footing, q is the soil surcharge at the base level of the footing, A is the cross-sectional area of the footing, and Nc, Nq, Ng are bearing-capacity factors.

*Corresponding author. Email: richard.merifield@usq.edu.au

The experimental data on which equation (4) is based are mostly due to Meyerhof (1951, 1963) and de Beer (1970), who tested both prototype and model foundations. For the case of circular and square footings on undrained clay where fu = 0, the factor Nq = 1, Nc = 2 + p, and therefore the shape factor zcs = 1.194. Recently, Salgado et al. (2004) investigated the bearing capacities of strip, square, circular, and rectangular foundations in clay using a rigorous finite-element limit analysis. The results of the analyses proposed rigorous, definitive values of the shape and depth factors for foundations in clays. They

Geomechanics and Geoengineering: An International Journal ISSN 1748-6025 print=ISSN 1748-6033 online 2006 Taylor & Francis http:==www.tandf.co.uk=journals DOI: 10.1080=17486020600632637

152

R. S. Merifield and V. Q. Nguyen

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

proposed that for 1

L /B

5 and 0

Df/B

B zcs 1 0:12 L r Df 0:17 square; circular; and rectangular footings; B 5 where Df is the depth of footing embedment. For a square or circular surface footing (Df = 0), the shape factor zcs = 1.12. Recently, Zhu and Michalowski (2005) estimated the shape factor zcs = 1.06 for square footings using the commercial finite element software FLAC3D (1997). Values of the shape factor can also be derived from the results of Wang and Carter (2000) and Michalowski (2001). While the ultimate bearing capacity of square and circular footings resting on a single layer of homogeneous undrained clay can be estimated using equation (3) along with equation (4) or (5), in reality, soil strength profiles beneath footings are not homogeneous but may increase or decrease with depth or consist of distinct layers having significantly different properties. While the effect of soil layering for continuous strip footings has been addressed by several researchers, notably Button (1953), Reddy and Srinivasan (1967), Brown and Meyerhof (1969), Chen (1975), Meyerhof and Hanna (1978), and Merifield et al. (1999), rigorous solutions to the problem of circular and square footings resting on layered clays do not appear to exist. Geotechnical engineers have addressed this issue by simply averaging layer strengths or adopting large safety factors to account for the uncertainty of soil layering. The purpose of this paper is to propose rigorous three-dimensional solutions for square and circular footings on two-layered clays using the displacement finite-element method. To do this, the commercial software package ABAQUS (2005) was used. In addition, two-dimensional analyses for the bearing capacity of strip footings have been performed for verification and comparison purposes. The results are an extension of the solution for strip footings on two-layer clays presented by Merifield et al. (1999).

The bearing-capacity solution to this problem will be a function of the two ratios H/B and cu1/cu2. Past research by Merifield et al. (1999) indicates that a reduction in bearing capacity for a strong-over-soft clay system may occur up to a depth ratio of H/B = 2 for strip footings. In this paper, solutions have been computed for problems where H/B ranges from 0.125 to 2, and cu1/cu2 varies from 0.2 to 5. This covers most problems of practical interest. Note that cu1/cu2 . 1 corresponds to the common case of a strong clay layer over a soft clay layer, while cu1/cu2 , 1 corresponds to the reverse. For the case of a layered soil profile it is convenient to rewrite equation (3) in the form
qu cu1 Nc

where cu1 is the undrained shear strength of the top layer, and Nc* is a modified bearing-capacity factor which is a function of both H/B, cu1/cu2 and the footing shape. The value of Nc* has been computed using the results from finite-element analyses for each ratio of H/B and cu1/cu2. For a homogeneous profile * with cu1 = cu2, Nc equals the Prandtl solution of (2 + p).

3.

Displacement finite-element modelling details

2.

Problem definition

The two- and three-dimensional bearing-capacity problem to be considered is illustrated in figure 1. A footing of width B or diameter D rests upon an upper layer of clay with undrained shear strength cu1 and thickness H. This is underlain by a clay layer of undrained shear strength cu2 and infinite depth. Symmetry has been exploited for the three-dimensional analyses, and only one-quarter of the problem domain has been modelled, as shown in figure 1. The soil was modelled as an isotropic elasto-perfectly plastic continuum with yielding described by the Mohr--Coulomb yield criterion. The elastic behaviour was described by a Poisson ratio n = 0.49, and a ratio of Youngs modulus to shear strength of E/cu = 100500, depending on whether the soil was soft or hard.

The displacement finite-element software ABAQUS was used for solving this problem. For strip, square, and circular footing problems, the ABAQUS model consisted of two parts, namely the footing and the soil. This is illustrated in figures 1(b) and 2. Typical meshes for the problem of strip and three-dimensional footings, along with the applied displacement boundary conditions, are shown in figure 2. For the strip footing case, the mesh is as shown in figure 2(a). The mesh consisted of six-node modified plane strain triangular elements which were found to provide the best solution convergence. For the three-dimensional case of square and circular footings, the 10-node modified quadratic tetrahedron element was adopted for similar reasons. These modified elements are specific to ABAQUS and have been constructed to reduce node locking and to have an unambiguous sign of the contact normal stress that is usually associated with second-order elements in contact analyses. The overall mesh dimensions were selected to ensure that the zones of plastic shearing and the observed displacement fields were contained within the model boundaries at all times. The undersides of all footings were modelled as perfectly rough by specifying a tied contact constraint at the footing/soil interface. To determine the collapse load of the footing, displacement defined analyses were performed where the footing was considered as being perfectly rigid. That is, a uniform vertical prescribed displacement was applied to all those nodes on the footing. The total displacement was applied over a number of substeps, and the nodal contact forces along the anchor were summed to compute the equivalent bearing capacity. A distinct advantage of using the displacement finite-element method is that it provides the complete load deformation response. This can provide insight into general footing

Two- and three-dimensional bearing-capacity solutions

153

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

Figure 1. Problem definition.

behaviour, particularly in regard to the development of the collapse mechanism and deformation serviceability issues. By observing the load-displacement response, a check can be made to ensure that the ultimate bearing capacity has been reached and that overall collapse has in fact occurred (i.e. the loaddisplacement plot reaches a plateau).

4.

Results and discussion

As an initial check on the three-dimensional finite-element model, the bearing-capacity factor for the homogeneous case Nc = N* c (cu1/cu2) for square and circular surface footings was calculated and compared with existing published numerical results. This comparison is summarized in table 1. As can be seen, the bearing-capacity factors from the current finite element study compare

well with that calculated using the widely adopted shape factor in equation (4) and the solutions of Salgado et al. (2004). The bearing-capacity factor for square footings (Nc = N* = 5.95) was found c to be around 2% below that for circular footings (Nc = N* = 6.05). c This observation is consistent with the findings of Salgado et al. as shown in table 1. The bearing-capacity factor for strip footings was found to be 5.24, around 2% above the classical Prandtl solution of (2 + p). The calculated shape factor zcs using equation (3) is therefore 1.14 and 1.15 for circular and square footings on a homogeneous clay profile, respectively. The bearing-capacity factors N* (equation (6)) for the nonc homogeneous case (cu1 6 cu2) for strip, circular, and square footings on two-layered clays are presented in table 2, and shown graphically in figures 3--6. Also shown in these figures are the bearing-capacity factors for strip footings by Merifield et al. (1999). Some general observations will be mentioned

154

R. S. Merifield and V. Q. Nguyen

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

Figure 2. Finite element modelling details: (a) strip footing; (b) square and circular footings.

Table 1. Values of bearing capacity factor N* c Salgado et al. (2004) Displacement FEM SQ
* Nc Compared with SQ (%) Compared with CI (%)

Square Equation (4) 6.14 3.19 1.49 Equation (5) 6.168 3.66 1.95 Lower bound 5.523 --7.18 -Upper bound 6.221 4.55 --

Circular Lower bound 5.856 ---3.21 Upper bound 6.227 -2.93

CI 6.05 -0

5.95 0 --

before discussing the results for strong-over-soft and soft-overstrong profiles separately in more detail. * The bearing-capacity factor Nc was found to be almost identical for square and circular footings for both the strongover-soft and soft-over-strong layer cases. In all cases, the bearing-capacity factor for circular footings was around 1--

2% above that of a square foot-ing. This is shown clearly in figures 3--6, where the results for square and circular footings are difficult to distinguish from each other. As discovered by Merifield et al. (1999), all of the analyses herein indicate that a complex relationship exists between the observed modes of shear failure and the ratios cu1/cu2 and

Two- and three-dimensional bearing-capacity solutions

155

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

* Table 2. Values of bearing-capacity factor Nc * Value of bearing capacity factor Nc

Upper bound (Merifield et al. 1999) H/B H / B = 0.125 cu1/cu2 0.20 0.25 0.40 0.50 0.80 1.00 1.25 1.50 2.00 2.50 4.00 5.00 0.20 0.25 0.40 0.50 0.80 1.00 1.25 1.50 2.00 2.50 4.00 5.00 0.20 0.25 0.40 0.50 0.80 1.00 1.25 1.50 2.00 2.50 4.00 5.00 0.20 0.25 0.40 0.50 0.80 1.00 1.25 1.50 2.00 2.50 4.00 5.00 0.20 0.25 0.40 0.50 0.80 1.00 1.25 1.50 2.00 2.50 4.00 5.00 0.20 0.25 0.40 0.50 0.80 Strip footing 8.55 8.55 8.55 8.55 6.36 5.32 4.52 3.93 3.09 2.61 1.82 1.55 6.52 6.52 6.52 6.52 6.25 5.32 4.6 4.08 3.34 2.88 2.12 1.85 5.84 5.84 5.84 5.84 5.84 5.32 4.78 4.28 3.65 3.2 2.42 2.13 5.49 5.49 5.49 5.49 5.49 5.32 4.94 4.48 3.89 3.47 2.74 2.44 5.36 5.36 5.36 5.36 5.36 5.32 5.2 4.94 4.37 4.01 3.28 2.98 5.3 5.3 5.3 5.3 5.3 Strip footing 8.61 8.61 8.60 8.40 6.21 5.24 4.45 3.83 3.03 2.53 1.75 1.48 6.34 6.34 6.34 6.33 5.95 5.24 4.62 4.05 3.31 2.84 2.06 1.71 5.78 5.78 5.78 5.76 5.76 5.24 4.77 4.27 3.58 3.13 2.37 2.07 5.46 5.46 5.46 5.46 5.46 5.24 4.92 4.49 3.86 3.43 2.67 2.36 5.24 5.24 5.24 5.24 5.24 5.24 5.17 4.87 4.35 3.96 3.22 2.90 5.27 5.27 5.27 5.27 5.27

Displacement finite-element method Square footing 7.96 7.96 7.96 7.96 6.73 5.95 5.11 4.53 3.73 3.21 2.33 2.00 6.35 6.35 6.35 6.35 6.27 5.95 5.45 5.03 4.39 3.92 3.04 2.70 6.00 6.00 6.00 6.00 6.00 5.95 5.74 5.46 4.96 4.53 3.67 3.32 5.96 5.96 5.96 5.96 5.96 5.95 5.94 5.82 5.46 5.08 4.22 3.89 5.95 5.95 5.95 5.95 5.95 5.95 5.98 5.98 5.96 5.91 5.24 4.94 5.93 5.93 5.93 5.93 5.93 Circular footing 7.95 7.95 7.95 7.89 6.85 6.05 5.27 4.66 3.85 3.32 2.41 2.07 6.36 6.36 6.36 6.36 6.34 6.05 5.59 5.17 4.51 4.02 3.13 2.78 6.10 6.10 6.10 6.10 6.10 6.05 5.88 5.60 5.10 4.68 3.80 3.44 6.04 6.04 6.04 6.04 6.04 6.05 6.02 5.90 5.58 5.23 4.39 4.03 6.03 6.03 6.03 6.03 6.03 6.05 6.05 6.05 6.05 6.03 5.47 5.15 6.03 6.03 6.03 6.03 6.03 (continued)

H / B = 0.25

H / B = 0.375

H / B = 0.5

H / B = 0.75

H / B = 1.0

156

R. S. Merifield and V. Q. Nguyen

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

Table 2. (Continued) 1.00 1.25 1.50 2.00 2.50 4.00 5.00 0.20 0.25 0.40 0.50 0.80 1.00 1.25 1.50 2.00 2.50 4.00 5.00 0.20 0.25 0.40 0.50 0.80 1.00 1.25 1.50 2.00 2.50 4.00 5.00 5.32 5.3 5.18 4.82 4.5 3.83 3.54 5.3 5.3 5.3 5.3 5.3 5.32 5.27 5.31 5.31 5.32 4.84 4.56 5.3 5.3 5.3 5.3 5.3 5.32 5.26 5.26 5.27 5.27 5.32 5.32 5.24 5.24 5.16 4.77 4.44 3.75 3.44 5.25 5.25 5.25 5.25 5.25 5.24 5.24 5.24 5.23 5.19 4.67 4.40 5.24 5.24 5.24 5.24 5.24 5.24 5.21 5.21 5.21 5.21 5.21 5.13 5.95 5.94 5.94 5.93 5.93 5.86 5.77 5.94 5.94 5.94 5.94 5.94 5.95 5.94 5.94 5.94 5.94 5.94 5.94 5.96 5.96 5.96 5.96 5.96 5.95 5.96 5.96 5.96 5.96 5.96 5.96 6.05 6.05 6.05 6.06 6.06 6.04 5.94 6.04 6.04 6.04 6.04 6.03 6.05 6.04 6.04 6.04 6.04 6.04 6.03 6.05 6.05 6.05 6.05 6.05 6.05 6.05 6.05 6.05 6.05 6.05 6.05

H / B = 1.5

H / B = 2.0

H/B. The modes of failures can best be described as being either general shear, partial punching shear, or full punching shear similar to that described by Merifield et al. (1999). Full punching shear (figure 7) is characterized by a vertical separation of the top layer, which then effectively acts as a rigid column of soil that punches through to the bottom layer. In this case, only a small amount of heaving is observed immediately adjacent to the footing, but significant yielding is observed below the upper layer. Conversely, only a small vertical separation of the top layer is evident for partial punching shear. General shear failure is as defined by many authors including Terzaghi (1943). In this mode of failure, we observe well-defined shear planes developing and extending to the surface, and bulging of the soil on both sides of the footing. When comparing the results for square and circular footings to those for strip footings, in general it was found that the * bearing-capacity factors Nc for square and circular footings were larger than those for strip footings. This is applicable for both the strong-over-soft and soft-over-strong profiles, and was to some extent expected. The only exception to this observation was for the soft-over-strong profile case where the upper layer is very thin compared with the footing width, namely when H/B 0.25. This is shown clearly in figure 3(a), where the above trend is reversed, and the bearing* capacity factors Nc for square and circular footings were smaller than those for strip footings. Further three-

dimensional and axi-symetric analyses with much finer finite-element meshes confirmed this observation. However, a further investigation is required to properly explain the mechanics of this failure mechanism and the lower-thanexpected collapse load.

4.1

Strip footings on layered clay cu1 6 cu2

For strip footings on layered clays, the finite-element bearingcapacity factors compare well with the upper bound solutions presented by Merifield et al. (1999). In general, the displacement finite-element solution presented here is within 2 % of Merifields solution (figures 3--6). For the strong-over-soft case where a moderately strong crust exists (1 cu1/cu2 2.5), failure is generally caused by partial punching shear. As the depth of the top crust approaches the footing width B, upward deformations within the bottom layer become restricted, causing an increase in the extent of plastic yielding. As the top crust becomes very strong compared with the bottom layer (cu1/ cu2 . 2.5), full punching shear through the top layer occurs. The very strong top layer then serves to greatly restrict both lateral and vertical movement of the soil contained in the soft layer below (figure 7). This results in the formation of a deep zone of plastic shearing within the bottom layer.

Two- and three-dimensional bearing-capacity solutions

157

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

11.0 10.0 9.0 8.0 7.0

* Nc

Soft over Strong

Strong over Soft

11.0
ABAQUS - Square footing ABAQUS - Circular footing ABAQUS - Strip footing Upper Bound, Merifield et al. 1999 - Strip footing

10.0 9.0 8.0 7.0

* N c 6.0

Soft over Strong

Strong over Soft

ABAQUS - Square footing ABAQUS - Circular footing ABAQUS - Strip footing Upper Bound, Merifield et al. 1999 - Strip footing

6.0 5.0 4.0 3.0 2.0 1.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0
cu1 cu 2
H = 0.125 B

5.0 4.0 3.0 2.0 1.0 0.0

H = 0375 B

3.5 4.0

4.5 5.0

0.0

0.5

1.0

1.5

2.0

2.5 cu1 cu 2

3.0

3.5

4.0

4.5

5.0

(a)

(a)
11.0

11.0 10.0 9.0 8.0 7.0

* N c 6.0

Soft over Strong

Strong over Soft

ABAQUS - Square footing ABAQUS - Circular footing ABAQUS - Strip footing Upper Bound, Merifield et al. 1999 - Strip footing

10.0 9.0 8.0 7.0

* N c 6.0

ABAQUS - Square footing

Soft over Strong

Strong over Soft

ABAQUS - Circular footing ABAQUS - Strip footing Upper Bound, Merifield et al. 1999 - Strip footing

5.0 4.0 3.0 2.0 1.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0
cu1 cu 2
H = 0.25 B

5.0 4.0 3.0 2.0 1.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 cu1 cu 2 3.0 3.5 4.0 4.5 5.0
H = 0.5 B

3.5 4.0

4.5 5.0

(b)
Figure 3. Bearing-capacity factor
* Nc

(b)
Figure 4. Bearing-capacity factor N* for strip, square, and circular footings. c

for strip, square, and circular footings.

The results indicate that a reduction in bearing capacity for a strong-over-soft clay system occurs up to a depth ratio of H/B & 1.5 -- 2. This lower limit is applicable for soil profiles where cu1/cu2 , 2.5, but for profiles that have a very strong top crust with cu1/cu2 . 2.5, punching failure through the top layer is likely to occur up to depth ratio of H/B = 2. For ratios of H/B . 2, failure is contained entirely within the top layer and is independent of the ratio cu1/cu2. These observations are consistent with those predicted by Merifield et al. (1999). For the soft-over-strong case, the finite-element results indicate that for ratios of H/B 0.5, the bearing capacity increases as the relative strength of the bottom layer rises. At a limiting ratio of cu1/cu2, no further increase in bearing capacity is achieved as the general shear failure mechanism becomes fully contained within the top layer.

For all values of H/B . 0.5, the solutions indicate that failure occurs entirely within the top layer, and the bearing capacity is independent of the strength of the bottom layer.

4.2

Square and circular footings on soft clay overlying strong clay cu1/cu2 , 1

* For thin upper layers where H/B 0.375, Nc increases as the ratio of cu1/cu2 decreases. That is, as the thin top layer becomes increasingly softer compared with the underlying strong layer, the contribution made by the underlying layer to the ultimate bearing-capacity increases. For very thin upper layers (H/B = 0.125), the increase in the bearing* capacity factor Nc above that for the homogeneous case is significant, as shown in figure 3(a). This can be explained

158

R. S. Merifield and V. Q. Nguyen

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

11.0 10.0 9.0 8.0 7.0

* N c 6.0

11.0
ABAQUS - Square footing

Soft over Strong

Strong over Soft

10.0 9.0 8.0 7.0

* N c 6.0

ABAQUS - Square footing

ABAQUS - Circular footing ABAQUS - Strip footing Upper Bound, Merifield et al. 1999 - Strip footing

Soft over Strong

Strong over Soft

ABAQUS - Circular footing ABAQUS - Strip footing Upper Bound, Merifield et al. 1999 - Strip footing

5.0 4.0 3.0 2.0 1.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0
cu1 cu 2
H = 0.75 B

5.0 4.0 3.0 2.0 1.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 cu1 cu 2 3.0 3.5
H = 1.5 B

3.5 4.0

4.5 5.0

4.0

4.5

5.0

(a)
11.0 10.0 9.0 8.0 7.0
* N c 6.0

(a)
11.0
ABAQUS - Square footing

Soft over Strong

Strong over Soft

10.0 9.0 8.0 7.0

* N c 6.0 5.0

ABAQUS - Circular footing ABAQUS - Strip footing Upper Bound, Merifield et al. 1999 - Strip footing

Soft over Strong

Strong over Soft

ABAQUS - Square footing ABAQUS - Circular footing ABAQUS - Strip footing Upper Bound, Merifield et al. 1999 - Strip footing

5.0 4.0 3.0 2.0 1.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 cu1 cu 2 3.0 3.5 4.0 4.5 5.0
H =1 B

4.0 3.0 2.0 1.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 cu1 cu 2 3.5 4.0 4.5 5.0
H =2 B

(b)
* Figure 5. Bearing-capacity factor Nc for strip, square, and circular footings.

(b)
* Figure 6. Bearing-capacity factor Nc for strip, square, and circular footings.

by observing the failure mechanisms, such as that shown in figure 8(a). The displacement vector diagram indicates that, although much of the failure is contained in the soft upper layer, yielding still occurs well into the strong underlying layer. For upper layer thicknesses where H/B > 0.375, the stronger underlying layer does not contribute greatly to the ultimate bearing capacity regardless of the ratio of cu1/cu2. As a consequence, no further increase in bearing capacity is achieved as the failure surface becomes fully contained within the top layer. The failure mechanisms shown in figures 9--11 confirm this. The bearing-capacity factor N* is equal to that found for the homogeneous case c and is shown as a horizontal line in figures 4--6. This observation is similar to that found by Merifield et al. (1999) for strip footings where, for H/B . 0.5, the limit

analysis solutions indicated that failure occurs entirely within the top layer, and the bearing capacity is independent of the strength of the bottom layer.

4.3

Square and circular footings on strong clay overlying soft clay cu1/cu2 . 1

In general, it was observed that for a given ratio of H/B , * 1, the bearing-capacity factor Nc was found to decrease in a nonlinear manner as the ratio of cu1/cu2 increases (figures 3--6). This indicates that the failure mechanism must penetrate into the underlying softer layer, which in turn has the effect of reducing the ultimate bearing capacity. This is confirmed by the displacement velocity plots shown in figure 8(b) and (c) and figure 9(b) and (c). In these figures,

Two- and three-dimensional bearing-capacity solutions

159

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

No heaving adjacent to footing

Strip footing, H B = 0.5 , cu1 cu 2 = 5

No heaving adjacent to footing

Strip footing, H B = 1, cu1 cu 2 = 5

Figure 7. Punching shear failure for strip footings.

large displacements and zones of yielding are seen to occur in the underlying soft layer. As previously mentioned, the ABAQUS finite-element results for strip footings and the limit analysis results of Merifield et al. (1999) indicate that a reduction in bearing capacity for a strong-over-soft clay system occurs up to a depth ratio of H/B 2. This limiting ratio of H/B is greater than that observed for circular and square footings (i.e. H/B , 1) as mentioned above. This behavioural difference is most * obvious in figure 5, where the bearing-capacity factor Nc for a strip footing is well below that of a circular or square footing on the same layered profile. Punching shear failure is a common mechanism observed in the case of strong-over-soft profiles. Full punching shear failure was generally observed when cu1/cu2 ! 2.5. For these cases, the top layer acts as rigid column of soil that pushes through the top layer into the underlying soft soil, and a significant zone of yielding is therefore apparent in the lower layer. Full punching shear failure is highlighted

in figures 8(c) and 9. Based on the observed displacement vectors and stress contours, some guidance is provided in figure 12 as a means of distinguishing the likely mode of failure for circular and square footings on a strong-oversoft profile.

4.4

Effect of footing roughness

The results of further analyses with a perfectly smooth footing/soil interface indicate that for a strong-over-soft clay profile, the soil/footing interface strength has little or no effect on the calculated bearing capacity. For strip footings, the reduction in the bearing capacity was ,4%, while for square and circular footings, the reduction was less than 1%. Similarly, for a soft-over-strong clay system where H/B . 0.5, the bearing capacity does not vary greatly with footing roughness (i.e. ,4% for all footing shapes). This agrees with the conclusions made by Merifield et al. (1999).

160

R. S. Merifield and V. Q. Nguyen

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

SOFT STRONG N c* = 6.35 SQ N c* = 6.36 CIRC

SOFT N c* = 5.96 SQ N c* = 6.04 CIRC STRONG

(a)

(a)

Circular & Square footing, cu1 cu 2 = 0.2 , H B = 0.25

Circular & Square footing, cu1 cu 2 = 0.2, H B = 0.5

STRONG SOFT N c* = 4.39 SQ N c* = 4.51 CIRC

STRONG N c* = 5.46 SQ SOFT N c* = 5.58 CIRC

(b)

(b)
Circular & Square footing, cu1 cu 2 = 2, H B = 0.25

Circular & Square footing, cu1 cu 2 = 2, H B = 0.5

STRONG SOFT N c* = 2.70 SQ N c* = 2.78 CIRC

STRONG N c* = 3.89 SQ SOFT N c* = 4.03 CIRC

(c)

Circular & Square footing, cu1 cu 2 = 5, H B = 0.25

Figure 8. Displacement vectors for square and circular footings on layered clay.

(c)

Circular & Square footing, cu1 cu 2 = 5, H B = 0.5

Figure 9. Displacement vectors for square and circular footings on layered clay.

For square, circular, and strip footings on a soft-over-strong clay system, where H/B 0.5, a perfectly smooth soil/footing interface serves to reduce the bearing capacity by up to 26%. The greatest reduction occurs when H/B = 0.125 and falls quickly to around 6% for H/B = 0.5.

5.

Conclusion

method has been investigated. The results obtained have been presented in terms of a modified bearing-capacity factor N* in c both graphical and tabular form to facilitate their use in solving practical design problems. The following conclusions can be made based on the finiteelement results:


The bearing capacity of strip, square, and circular footings on two-layered clays using the displacement finite-element

For homogeneous soil profiles, the displacement finite-element bearing capacity and shape factors for square and

Two- and three-dimensional bearing-capacity solutions

161

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

N c* = 5.94 SQ N c* = 6.04 CIRC SOFT STRONG

(a)

Circular & Square footing, cu1 cu 2 = 0.2, H B = 1.5

N c* = 5.94 SQ N c* = 6.03 CIRC STRONG SOFT

(b)

Circular & Square footing, cu1 cu 2 = 5, H B =1.5

N c* = 5.96 SQ N c* = 6.05 CIRC

(c)
Circular & Square footing, cu1 cu 2 = 0.2 5, H B = 2
Figure 10. Displacement vectors for square and circular footings on layered clay.

Figure 11. Displacement vectors for square and circular footings on layered clay.

circular footings compare well with previously reported finite-element and numerical limit analysis solutions. For strip footings on layered clays, the finite-element bearing-capacity factors compare well with the numerical upper bound limit analysis solutions presented by Merifield et al. (1999). In general, the displacement finite-element solutions presented here are within 2% of Merifields solution. For a soft-over-strong clay system where the ratio of H/B > 0.375, the stronger underlying layer does not contribute

greatly to the ultimate bearing capacity, regardless of the ratio of cu1/cu2. As a consequence, no further increase in bearing capacity is achieved as the failure surface becomes fully contained within the top layer. When H/B 0.375, the bearing-capacity factor N* increases as the ratio of cu1/cu2 c decreases. For very thin upper layers (H/B = 0.125), the increase in the bearing-capacity factor N* above that for the c homogeneous case is significant.

162

R. S. Merifield and V. Q. Nguyen

Downloaded By: [University of Newcastle] At: 00:17 25 January 2008

Figure 12. Division between full and partial punching shear.

For both square and circular footings, a reduction in bearing capacity for a strong-over-soft clay system occurs up to a depth ratio of H/B , 1. For depth ratios of H/B ! 1, failure is likely to be fully contained within the top layer, and the bearing-capacity factors for square (Nc = 5.95) and circular footings (Nc = 6.05) on a homogeneous profile can be used.

References
ABAQUS, Standard Users Manual: Version 6.5.4, 2005 (Hibbitt, Karlsson & Sorensen: Pawtucket, RI). Brown, J.D. and Meyerhof, G.G., Experimental study of bearing capacity in layered clays, in Proceedings of the 7th International Conference on Soil Mechanics and Foundation Engineering, 1969, Mexico, 2, pp. 45--51. Button, S.J., The bearing capacity of footings on a two--layer cohesive subsoil, in Proceedings of the 7th International Conference on Soil Mechanics and Foundation Engineering, Zurich, 1953, 1, pp. 332--335. Chen, W.F., Limit Analysis and Soil Plasticity, 1975 (Elsevier: Amsterdam). de Beer, E.E., Experimental determination of the shape factors and the bearing capacity factors of sand. Geotechnique, 1970, 20, 387--411.  FLAC3D, Fast Lagrangian Analysis of Continua in 3 Dimensions, Users Manual, 1997 (Itasca Consulting Group: Minneapolis, MN). Merifield, R.S., Sloan, S.W. & Yu, H.S., Rigorous solutions for the bearing capacity of two layered clay soils. Geotechnique, 1999, 49(4),  471--490.

Meyerhof, G.G., The ultimate bearing capacity of foundations. Geotechnique,  1951, 2, 301--332. Meyerhof, G.G., Some recent research on the bearing capacity of foundations. Can. Geotech. J., 1963, 1, 16--26. Meyerhof, G.G., and Hanna, A.M., Ultimate bearing capacity of foundations on layered soils under inclined load. Canadian Geotechnical Journal, 1978, 15(4), 565--572. Michalowski, R.L., Upper-bound load estimates on square and rectangular footings. Geotechnique, 2001, 51(9), 787--798.  Poulos, H.G., Carter, J.P. and Small, J.C., Foundations and retaining structures---research and practice. State of the art lecture, in 14th International Conference on Soil Mechanics and Geotechnical Engineering, Istanbul, 2001. Prandtl, L., Uber die Harte plastischer Korper. Nachr. Konigl. Ges. Wissensch. Gottingen Math.-Phys. Klasse, 1920, 74--85. Reddy, A.S. and Srinivasan, R.J., Bearing capacity of footings on layered clays. J. Soil Mech. Found. Div., 1967, 93(2), 83--99. Salgado, R., Lyamin, A.V., Sloan, S.W. & Yu, H.S., Two- and three-dimensional bearing capacity of foundations in clay. Geotechnique, 2004, 54(5),  297--306. Tani, K. & Craig, W.H., Bearing capacity of circular foundations on soft clay of strength increasing with depth. Soils Found., 1995, 35(4), 21--35. Terzaghi, K., Theoretical Soil Mechanics, 1943 (Wiley: New York). Wang, C.X. & Carter, J.P., Penetration of strip and circular footings into layered clays, in Advances in Theoretical Geomechanics---Proceedings of the John Booker Memorial Symposium, Sydney (Balkema, Rotterdam), 2000, pp. 193--210. Zhu, M. and Michalowski, R.L., Shape factors for limit loads on square and rectangular footings. J. Geotech. Geoenviron. Eng., 2005, 131(2), 223--231.