Journal of Geotechnical and Geoenvironmental Engineering

The paper presents the results from an experimental program carried out at Trinity College Dublin, in which instrumented model piles were jacked into loose dry sand in a large testing chamber. A number of pile installations were carried out to study the effects of in situ stress, diameter, and wall thickness on the behavior of open-ended piles in sand. These indicated that plug stiffness and capacity may be expressed as simple functions of the cone penetration test end resistance and the incremental filling ratio prior to loading. The magnitude and distribution of shear stresses measured on the inner wall are shown to be compatible with existing experimental data and can be related directly to the stress level, interface friction angle, and dilation of the sand at the pile wall. The data are shown to facilitate a better understanding of the factors controlling plug resistance.

Experimental investigations on model pile groups of configuration 1 × 1, 2 × 1, 3 × 1, 2 × 2, and 3 × 2 for embedment length-to-diameter ratios L/d = 12 and 38, spacing from 3 to 6 pile diameter, and pile friction angles δ = 20° and 31°, subjected to lateral loads, were conducted in dry Ennore sand obtained from Chennai, India. The load-displacement response, ultimate resistance, and group efficiency with spacing and number of piles in a group have been qualitatively and quantitatively investigated. Analytical methods have been proposed to predict the ultimate lateral capacity of single pile and pile groups. The proposed methods account for pile friction angle, embedment length-to-diameter ratio, the spacing of piles in a group, pile group configuration, and soil properties. These methods are capable of predicting the lateral capacity of piles and pile groups reasonably well as noted and substantiated by the comparison with the experimental results of the writers and other researchers.

In the literature, various “failure criteria” or methods of estimating the failure load in pile loading tests have been proposed. The criteria, based on varying assumptions, were intended for different methods of pile testing and were verified on tests of a variety of pile types and sizes. Most of the criteria were not developed for slow maintained loading tests of large-diameter (greater than 0.6 m) and long bored piles. Piles of this kind have considerable resistance, and it is often impractical to reach failure load as defined by the various criteria. In this paper, a total of 38 large-diameter bored piles (drilled shafts) that were tested, ranging from 0.6 to 1.8 m in diameter, varying from 12 to 66 m in depth, and founded in weathered geomaterials (rocks and saprolites), are critically reviewed and studied. Among them, a selection of seven pile load tests is examined in detail by using different existing failure criteria and specifications. The tests were chosen for their high degree of mobilization of pile capacity and the availability of reliable load-movement relationships. Specific aspects of pile behavior, such as the mobilization of toe resistance and shaft shortening, are also investigated using 31 loading tests to develop a new failure load criterion. The writers were heavily involved with the construction, testing, and analysis of 15 of the 38 piles. From the results of the study, a new nonsubjective, semiempirical method is proposed for estimating the approximate interpreted failure loads for piles founded in weathered geomaterials. The method is based on a moderately conservative estimation of the movement required to mobilize toe resistance and incorporates observations of shaft shortening from pile loading tests. Generally, the new method may allow more effective and consistent designs for large-diameter bored piles in weathered geomaterials.

Development of urban cities in hilly terrain often involves the construction of high-rise buildings supported by large diameter piles on steep cut slopes. Under lateral loads, the piles may induce slope failure, particularly at shallow depths. To minimize the transfer of lateral load from the buildings to the shallow depths of the slope, an annulus of compressible material, referred to as sleeving, is usually constructed between the piles and the adjacent soil. However, the influence of the sleeving on the pile performance in a sloping ground is not fully studied and understood. To investigate the influence, a 3D numerical analysis of sleeved and unsleeved piles on a cut slope is described in this paper. The influences of relative soil stiffness on the response of sleeved piles are also examined. The load transfer from the laterally loaded sleeved pile to the sloping ground is primarily through a shear load transfer mechanism in the vertical plane. Under small lateral loads, the sleeving can lead to a significant reduction in subgrade reaction on the sleeved pile segment and may considerably increase the pile deflection and bending moments. Under large lateral loads, the influence of the sleeving on pile performance appears to diminish because of the widespread plastic zones developed around the pile.

Leakage through composite landfill liners having various characteristics was analyzed using existing analytical and numerical models developed for the study. Three-dimensional numerical models were used to analyze leakage through circular defects and two-dimensional numerical models were used to analyze leakage from defective seams. Leakage rates predicted with the numerical models were compared to leakage rates predicted using existing equations and analytical models currently being used. These comparisons show that existing equations and analytical models all have limitations and no universal equation or method is available for predicting leakage rates. To overcome some of the deficiencies in the existing equations and models, new equations were developed based on results from the numerical models. Recommendations are made for using the new equations, existing equations, and analytical models to predict leakage rates in thick composite liners having a geomembrane overlaying a compacted soil liner and thin composite liners having a geomembrane overlaying a geosynthetic clay liner.

This paper describes an analytical model of filtration for granular media, based on the mechanics of particle migration under hydraulic loads. A new equation to predict the probability of particle movement through a 3D network model of the filter voids has been developed. Void constriction sizes are determined based on the particle-size distribution and relative density of the filter. An important new development is the differentiation between particles that form part of the filter structure and fine particles that are loose within the filter voids, or coarse particles that are enmeshed in a matrix of fines. The rate of particle erosion and transport is governed by the consideration of mass and momentum conservation. The model describes the time-dependent change of flow rate and base and filter particle-size distribution, porosity, and permeability. The model has application in the design of granular filters for noncohesive uniform, well-, and broadly graded base and filter materials.

The accurate flow measurement of each individual phase is important in unsaturated flow through rock joints, where both air and water phases flow together. An increase in the quantity of one fluid phase decreases the relative permeability of the other phase. The relative permeability is important in numerical models to analyze the risk of ground-water inundation and uncontrollable gas flows in underground excavations in jointed rock. A new apparatus, the High Pressure Two-Phase Triaxial Apparatus (HPTPTA), has been designed for examining the strength and coefficient of permeability characteristics of fractured and intact rocks under two-phase flows. In single-phase triaxial equipment, the rock specimen is subjected to a single fluid flow (either water, oil, or gas) through the fractures. In the HPTPTA, two fluids (e.g., water + air, water + oil, and oil + air) can be forced to flow through the specimen, and the flow rates of the fluids can be measured independently. The scope of tests that can be carried out in this apparatus is wide, including the evaluation of (1) stress-strain behavior subject to internal fluid flow; (2) relative permeability of each fluid phase under different degrees of saturation; and (3) the associated volume change of the specimen. In this paper, the design concepts of the HPTPTA and the results based on the testing of fractured rock specimens are discussed. The laboratory results are compared to a simplified mathematical model developed by the writers. Based on the laboratory results, it is shown that the well-known Darcy's law can be modified for estimating the two-phase flow rates using the relative permeability concept.

A study of interface shear strengths between smooth and textured high density polyethylene (HDPE) geomembranes (GMs) and a woven/nonwoven needle-punched geosynthetic clay liner (GCL) is presented. Tests were performed using a large direct shear machine capable of measuring peak and large displacement (200 mm) shear strengths. The failure surface was located at the GM/GCL interface for all tests conducted, corresponding to a normal stress range of 1–486 kPa. Small positive pore pressures were measured for all interfaces at peak shear strength. Thus, the practice of preparing failure envelopes using total normal stress, instead of effective normal stress, appears to be conservative. Interface shear strengths for textured GMs placed against the nonwoven side of the GCL were higher than those corresponding to the woven side. By comparison, differences in peak shear strength for laminated and coextruded GM interfaces were relatively less. Limited tests showed that peak and large displacement shear strengths were independent of displacement rate and dependent on the shear direction of the GM. The quantity of extruded bentonite at the interfaces generally increased with normal stress and was less for nonwoven geotextile interfaces than for woven geotextile interfaces. Implications of the findings to the testing of GM/GCL interfaces and the characterization of GM/GCL interface shear strength are discussed.