A study of the effect of reinforcement on the (cumulative) plastic settlement, ultimate bearing capacity, and resilient rebound of a repeatedly loaded plane‐strain footing on granular soil overlying different compressible bases is presented. Results were obtained using several different granular soils. The repeated load was returned to zero at the end of each cycle. This is characteristic of a rail‐guided vehicle loading on a track support. The results demonstrated the importance that both base compressibility and correct location of reinforcement have on the observed plastic and rebound deformations and on ultimate bearing capacity. Incorrect location of reinforcement demonstrated no beneficial effect. The plastic settlements were evaluated using a hyperbolic equation. The rebound readings were evaluated using Steinbrenner's analysis. The practical significance to gantry cranes and railway tracks is briefly discussed.

A series of large‐scale loading tests of square and rectangular footings on a naturally deposited dense scoria (a granular material produced by volcanic activities) was conducted in a pneumatic caisson. Standard penetration tests (SPTs) showed that the blow counts required for a 30‐cm penetration (TV‐value) were more than 70. This paper presents the method of the loading tests, characteristics of load‐settlement relations and failure mechanisms observed. The test results showed that bearing‐capacity factors decreased with increasing footing size. A set of stress characteristic equations incorporating stress dependency of angle of shearing resistance was solved numerically by use of the finite difference method to interpret the experimental data. Parameters required for the numerical computations were determined from the triaxial compression test data on an undisturbed sample. Good agreement was obtained between the numerical results and the experimental observations. A close examination of the failure mechanisms revealed that visible shear bands were developed because of particle crushing during shearing under high pressure.

The limit analysis of stability problems in geomechanics is formulated as a pair of primal‐dual linear programs that encode, respectively, the kinematic and static limit theorems in a discrete version. The failure surface can take any arbitrary shape. The soil domain is divided, into rigid elements connected by interfacing Mohr‐Coulomb layers. For an assumed finite element mesh, the solution of either linear program identifies the critical collapse mechanism among all the possible failure mechanisms contained within the mesh, and gives the associated values of both the static and kinematic variables as well as the critical load parameter. This solution is both kinematically and statically admissible for the discretized system; for the continuum, it is an upper‐bound solution. The proposed method is able to deal with external forces acting on a soil mass with varying pore‐water pressure, and inhomogeneous materials having both cohesion and internal friction. An illustrative example is presented; this kinematic formulation accurately gives the upper‐bound solution.

The limit analysis for stability problems in geomechanics can be formulated as a pair of primal‐dual linear programs. This formulation provides a solution that is both kinematically and statically admissible for a discretized soil mass; for the continuum, it gives an upper‐bound solution. This paper investigates the accuracy and the scope of applicability of this kinematic formulation. Several numerical examples of bearing capacity and slope stability problems are presented, including problems of a cohesive and frictional soil mass with inhomogeneous properties and varying pore‐water pressure. In bearing‐capacity analysis, the bearing pressure is considered to be “activating,” while the surcharge and self‐weight are considered to be “fixed.” In evaluating the safety factor for slope stability analysis, two different approaches are used: for frictional materials, the actual loading is considered to be “fixed,” and a fictitious horizontal loading is introduced as “activating”; for purely cohesive materials, the vertical loading is treated as “activating.” The computed results agree very well with analytical values and other numerical results. A comparison of this kinematic method with standard limit equilibrium methods is also presented.

The postliquefaction shear strength of sands, called the undrained critical strength or $s_u$(critical), is evaluated in terms of the critical strength ratio, $s_u\left(\mathrm{critical}\right)/\acute{\rho }{}_{\mathrm{vo}}$ This allows postliquefaction stability analyses to incorporate the variation of $s_u$(critical) with effective vertical stress instead of using a single value as proposed elsewhere. Comparison of back‐calculated critical strength ratios and the cyclic stress ratios triggering liquefaction suggests that drainage occurs in most cases during the postliquefaction flow of liquefied sands. To evaluate the stability of an existing slope that is predicted to liquefy during a dynamic event, the original slope configuration and the constant volume $s_u$(critical) must be used instead of the back‐calculated partially drained $s_u$(critical) values. The proposed procedure for estimating the constant volume $s_u$(critical), using the results of field and/or laboratory tests, shows that the critical strength ratio is approximately one‐half the yield strength ratio at the triggering of liquefaction for an earthquake magnitude of 7.5. It has been shown that the yield strength ratio at the triggering of liquefaction for a magnitude of 7.5 can be estimated by 0.011 times the equivalent clean sand blow count. Therefore, the critical strength ratio is 0.0055 times the equivalent clean sand blow count.

A 3‐D slope‐stability‐analysis method, explicitly satisfying all limiting‐equilibrium equations, is presented. To make the problem statically determinate, the variational approach, rather than intuition in 3‐D, is used. In this approach, the normal stress over the user's specified slip surface, which satisfies global limiting equilibrium and produces the minimum factor of safety for the critical surface, is sought and obtained mathematically. A simple numerical scheme to attain the safety factor is detailed and all the algorithms needed to assemble a computer program are given. Upon discretization of the surface by n rectangular elements, n simultaneous linear equations are solved for n discrete normal stresses. It is also necessary to solve three nonlinear equations for the factor of safety, $F_s\text{,}$ and the coordinate of the center of rotation $x_c{z}_c\text{.}$ It was observed that for a selected potential slip surface there may be many possible combinations of $(n+2)$ roots, all giving essentially the same $F_s\text{.}$ The presented example problem demonstrates the importance of 3‐D back‐analysis if one is to configure in‐situ soil strength.

Creep and stress relaxation are two processes where phenomenological models are most useful in predicting soil behavior. However, they are often treated as two separate phenomena although physical considerations suggest that one process could be responsible for bringing about the other. This paper aims at unifying existing phenomenological models for creep and stress relaxation. A unique stress‐strain‐time function for cohesive soils is postulated and used to construct a rate‐constitutive equation. The delayed model is generalized in terms of the first two stress invariants to describe the volumetric and deviatoric responses of soils, herein interpreted as induced by stress relaxation. For the volumetric model, the inverted secondary compression law is used; for the deviatoric model, the inverted Singh‐Mitchell creep equation is assumed. To further generalize the model, the characteristic time scales are replaced by so‐called age variables computed from the stress distances of a state point to its volumetric and deviatoric images. The performance of the composite model is then assessed by comparing the model predictions with reported creep and stress relaxation behavior of bay muds under isotropic and triaxial stress conditions.

The influence of anchor stiffness, inclination, and vertical spacing on the seismic response of multianchored retaining walls is investigated using the finite element method. Anchor stiffness was varied over several orders of magnitude. Anchor inclination is varied from a horizontal position to a steep, but reasonable, inclination. The vertical anchor spacing is varied from a single‐anchor wall to a four‐anchor wall. All walls are subjected to three levels of input excitation with single, half‐sine, displacement pulses with two different wavelengths. An increase in anchor stiffness results in a decrease in maximum and permanent displacements. A stiffer anchor will work over a larger range of input acceleration than a more flexible anchor. An increase in anchor inclination slightly reduces the amount of displacement occurring along the wall. Inclined anchors will experience slightly higher loads than horizontal anchors. A single‐anchor wall experiences more displacement below the anchor level and less displacement above the anchor than a two‐ or four‐anchor wall. Additional rows of anchors cause the wall to experience more of a rotational movement into the backfill, and a decrease in the center‐span displacement suffered by a single‐anchor wall.

A finite element model is used to evaluate seismic behavior of static and seismic tiedback retaining wall designs. Two static designs are tested. The first design assumes active earth pressure resulting in a single‐anchor design. The second static design uses a rectangular apparent‐pressure diagram, resulting in a two‐anchor wall. These two designs are then modified using the Mononobe‐Okabe (M‐O) method to account for seismic loading. Results illustrate that static earth pressure design assumptions critically impact seismic response. The earth‐pressure magnitude and distribution assumed for static conditions are as important to seismic behavior as application of the M‐O seismic design. Not only does the two‐anchor static design suffer less displacement than the single‐anchor static design; it also exhibits less displacement than the single‐anchor seismic design. The two‐anchor seismic design suffers less displacement than the two‐anchor static design, only at accelerations greater than 0.2g. The effect of an increased depth of embedment beyond the M‐O calculated value provides no benefit in limiting displacements above the dredgeline. Use of a seismic preload results in significantly decreased displacements compared to the same design using a preload based on static conditions.

Foundation walls have been formed by assembling lightweight sandwich panels without a crane, then pouring lightweight concrete fill between the panels and surrounding excavation. The wet fill exerts low pressure on the panels because it weighs less than 50 lb/cu ft and is poured in two lifts. The first lift is hardened with an accelerator before the second lift is poured. The hardened fill stiffens the panels against subsequent earth pressure. The method is useful for soils cohesive enough to support a vertical cut until the fill is placed. Construction is illustrated using mesh‐reinforced mortar‐faced polystyrene foam‐core sandwich panels and backfill containing fly ash and recycled polystyrene foam aggregate. Design examples use laboratory test values and illustrate the calculation of minimum‐cost thicknesses of sandwich panel and fill for required bending stiffness.

Electro‐osmosis tests are conducted on saturated kaolinite specimens loaded with phenol at a level of 500 μg/g of dry kaolinite. This amount was below the phenol adsorption capacity of this mineral. The flow behavior and the electrochemistry (voltage, current, resistance, pH gradients, and conductivity) in these experiments are presented. Electroosmotic flow with open electrodes is time‐dependent. Acid front generated at the anode sweeps across the specimen neutralizing the base at the cathode. The adsorbed phenol was removed by 85%‐95% by the process. Phenol concentration profile in the effluent did not display any retardation due to desorption. The energy expenditure for removal was of the order of 18–39 $\mathrm{kWh}/m^3$ of soil processed. The study displays that it may be feasible to remove low‐concentration organic contaminants from cohesive soils by application of a direct current.