Analyses of a dynamic lateral load test conducted on a full‐sized pipe pile supporting a rigid cap‐mass superstructure are described. Three algorithms representing different categories of mathematical models that simulate pile‐soil interaction were employed in the analyses. These algorithms represented: (1) A hysteretic subgrade reaction model capable of modeling gap formation; (2) a fully coupled subgrade reaction (boundary element) model that accounts for nonlinear behavior; and (3) an approximate (plane strain) linear model. Input parameters were varied for each of the models in order to develop an appreciation for appropriate inputs. All of the models were found to yield reasonable, but not exact, correspondence with frequency response functions at the location of the cap‐mass measured during sweep loading of the capmass. Inputs necessary to obtain such correspondence were found that were predicted upon field measurements or upon published criteria. System damping was predicted to be too low with all models, with the exception of one set of inputs for Model 1, based on comparisons of the widths of the predicted and measured response functions.

Linear regression and related statistical techniques are used to analyze hydraulic conductivity, (k) data, and test termination criteria are suggested. These criteria include that the slope of k versus pore volumes does not differ significantly from zero and that a minimum number of permeant pore volumes pass through the clay to provide saturation. To simplify the presentation, specific characteristics of the clay/chemical combination, such as cation exchange capacity, are not a focus of this presentation. The use of clay liners at hazardous waste disposal sites has prompted an interest in the k of clays exposed to chemicals, k is commonly determined in the laboratory using fixed or flexible wall permeameters and consolidometers, but past and current procedures reflect a lack of consistency in termination criteria for such tests. Fixed wall permeameters are used to study the effects of calcium sulfate and two simulated waste chemicals on three field clays. Results indicate the usefulness of the analysis in defining equilibrium conditions in laboratory testing.

Studies of the pore size characteristics and permeability of compacted sandy soils are described. Samples of mixtures of Ottawa sand and kaolinitic clay were prepared by kneading compaction and pluvial compaction was used to prepare samples of a clean Ottawa sand. The pore size distributions (PSD) of the samples were determined by mercury intrusion porosimetry, and the permeability of each sample was measured by a falling head test with back pressure. A pore size density function was derived from the PSD data. For the sand specimens, this function showed a singly modal characteristic on a log diameter scale while those of the sand‐clay mixtures showed bimodal characteristics. The influence of varying water content and compactive effort on the fabrics of the soils studied could be readily characterized by changes in the resulting pore size distributions and density functions. A probability‐based permeability model which relates the pore size density function to the permeability of compacted soils was verified experimentally.

Grouted sand is considered as a two‐phase particulate composite in which both the cohesive and adhesive properties of the grout and the particle‐to‐particle interaction of the sand are considered to contribute toward the overall macro‐response (strength, failure strain, stiffness, and failure modes) under tensile, shear, and compressive loadings. Adhesion tests were used to evaluate the interface bond under tensile, shear, and mixed mode loadings, and inclusion models were used to study failure mechanisms and how they affect the stress‐strain relationship. The bond strength can be represented by a Mohr‐Coulomb failure criterion, and the failure of grout and grouted sand can be approximated by a Drucker‐Prager criterion. A theory has been proposed to explain the behavior of grouted sand, and tensile strength and stiffness models were developed to predict the properties of the grouted sand from the properties of the constituents.

An analytical method has been proposed to predict the ultimate uplift capacity of piles embedded in sand. The method takes into consideration the length, diameter, and surface characteristics of piles and the soil properties. Typical charts for evaluation of average skin friction values for piles are presented through figures in dimensionless quantities. Critical depth of embedment beyond which the average skin friction attains a constant value depends not only on the relative density of sand but significantly on pile surface characteristics. Reasonable agreement has been observed between the measured and predicted values of uplift capacities. Among the available methods the proposed analysis predicts more realistic values of uplift capacities.