Recent Developments of Soil Mechanics and Geotechnics in Theory and Practice

The deposition of excavated natural soils, e.g. in open pit mines or during land reclamation, produces in general lumpy soils with multimodal pore sizes. The mechanical behaviour of such soils is determined by interaction between the firm to stiff lumps and the soft to liquid soil material in the macrovoids between the lumps. The initial skeleton composed of lumps resembles a coarse grained soil. However, contrary to mineral grains, the macrograins (lumps) are not incompressible and time-independent. Due to rising overburden during the subsequent deposition of the excavated soil, a time-dependent deformation of the lumps takes place, being accompanied by a softening of the lumps’ surface. The space between the lumps, originally occupied by air, becomes gradually filled with that soft material. A constitutive description of the lumpy soils should take into account all the mentioned effects. In this paper, main ideas and fundamentals for the constitutive modelling of the relevant phases of lumpy soils are outlined. With help of a homogenization method, a material model suitable for practical applications can be obtained.

A simple, yet versatile yield surface in the stress space is combined with a hypoplastic equation to simulate the influence of recent deformation history on the mechanical behaviour of sand. This yield surface is used to describe the intensity of anelastic flow. In the model, the state is fully described by the current stress, the void ratio and a novel back stress-like tensor. This new state variable determines the shape and size of the yield surface and accounts for recent deformation/stress history. The direction of the anelastic flow upon shearing is obtained from a generalization of the Taylor’s dilatancy rule. As a distinctive feature of the model, this dilatancy is able to reproduce the strong contractancy upon reversals observed in experiments without the need of additional state variables. The model corrects some known shortcomings of previous hypoplastic models like overshooting and the excessive accumulation of stress/strain upon strain/stress cycles of small amplitude (ratcheting). Laboratory tests are simulated to show the capabilities of the model to reproduce the soil behaviour under monotonic and cyclic loading conditions after different deformation histories.

This paper examines the influence of dilatant processes that can occur at a discontinuity of finite dimensions located at a compressed elastic geological interface, due to relative shear movement in the plane of the discontinuity. The dilatant phenomena will result in displacements in a direction normal to the shear movement and these displacements will be influenced by the elasticity of the geological materials and the compression at the interface. The dilatant displacements will also create a zone where there is loss of contact at the unilaterally constrained interface. The resulting problem is examined by appeal to results of the mathematical theory of elasticity. The influence of dilatant processes on the development of shear at the elastically constrained interface is examined by considering the procedures proposed by D.W. Taylor to analyze dilatant phenomena. The mathematical developments illustrate the combined influence of elastic constraints and interface compression on the amplification of the shear stress generated at the finite region. In the absence of dilatancy, the developments reduce to the classical result involving only Coulomb friction.

This work provides a new evaluation method of the dilatancy for cohesive soils from monotonic and cyclic undrained triaxial tests. Herein it is implemented for experiments performed on Kaolin. Leastwise for this soft soil the dilatancy turns out to be a function of the stress ratio \(\eta \) and the void ratio e along with the intrinsic material parameters. Furthermore, an OCR-definition, which includes the influence of both the stress ratio and void ratio such that \(d = f(\mathrm{OCR},C)\) with C being a set of inherent parameters is proposed. In addition based on the experimental observations it is suggested that there is an overconsolidation ratio \(\text {OCR}_{ci}\) at which the soft soil behaviour changes from contractant in case of \(\text {OCR}<\text {OCR}_{ci}\) to dilatant (the material can both contract and dilate depending on \(\eta \)) in case of \(\text {OCR}>\text {OCR}_{ci}\) with the PTL lying below the CSL in this case. Finally, a constitutive relation describing the behaviour of soft soils including the dilatancy and viscosity is proposed. Some simulations of monotonic as well as cyclic tests are shown to prove the accurate performance of the model.

In saturated soils the difference between the acceleration of soil skeleton and water is usually disregarded. In this paper it is reintroduced in a simplified manner into the balance equation and into the conservation of mass. This leads to a symmetric element matrix. Also the inertial interaction force associated with this difference are taken into account. It turns out, however, that numerical errors due to the standard Galerkin spatial discretization and due to the usual trapezoidal rule of Newmark time integration scheme blur the physically motivated extension. The conventional FEM is insufficient to calculate boundary value problems of dynamic nature even if displacements is better interpolated than pressure. Stabilization techniques and discrete Galerkin formulation seem to be indispensable to judge about effects of the proposed extension.

A number of geotechnical engineering problems involve large deformations in the soil. While small deformation geotechnical problems can be adequately analyzed by means of conventional Lagrangian FEM, such an approach exhibits considerable shortcomings when the soil undergoes significant deformation. Hence methods have been developed which overcome these shortcomings. Among the methods using a computational mesh, the most promising approaches include the ALE methods: Coupled Eulerian-Lagrangian (CEL) method, the Simplified or Single-Material Arbitrary Lagrangian-Eulerian (SALE) method, and the Multi-Material Arbitrary Lagrangian-Eulerian (MMALE) method. In this contribution the possible advantages and limitations of ALE in comparison to other numerical approaches are presented. The performance of ALE methods is evaluated by means of two application examples. First the vibratory driven installation of open-ended tubular steel piles in sand is modeled and local large deformation due to initial pile imperfections and soil heterogeneity is analyzed. Second the sand column collapse problem is simulated using different modeling approaches and soil models as well. Simulation results assess the feasibility of ALE methods in geotechnical large deformation problems. It can be concluded that the ALE method could be considered as a promising framework for solving complex large deformation problems in geotechnical engineering.

The design of foundations for offshore wind turbines is still a challenging engineering task. Design rules are not always settled in national standards and especially tasks like the design against cyclic loading remain a field where results of scientific research are immediately implemented in practical design approaches. The PISA project demonstrated, that even the static design of monopiles can be improved by validating new design approaches and implementing them in design tools. This paper compares different approaches for the static and cyclic design of foundations for offshore wind turbines, including new approaches like PISA, SOLDYN and TANDEM, as far as available. Focus is laid on the validation process of a cyclic approach, based on the high cycle accumulation (HCA) model. The main results of this validation on monopiles and shallow foundations in different scales are discussed. The HCA model showed to be very well suited to simulate and predict the cyclic soil structure interaction processes of foundations for offshore wind turbines. Effects like the accumulation of deformations or the redistribution of stresses and internal forces induced by a cyclic loading on monopiles or shallow foundations are well reproduced by the HCA model. This is demonstrated based on a comparison of results from model tests and in situ tests with numerical predictions. Furthermore, the results of these studies on the HCA model are set in relation to the approaches from other research groups. The outcome of the PISA project for the static design of monopiles is shown and reviewed based on the own results of numerical simulations.

The paper summarizes a long-term experimental research program on the behaviour of granular soils under high-cyclic loading, that means a loading with many cycles (\(N \ge 10^3\)) and relatively small strain amplitudes (\(\varepsilon ^{\text {ampl}}\le 10^{-3}\)). Numerous drained cyclic triaxial and hollow cylinder triaxial tests have been performed, most of them on Karlsruhe fine sand. The effects of the stress or strain amplitude, average stress, relative density, grain size distribution curve, particle shape and a content of non-plastic fines or shell fragments on the cumulative deformations are discussed. The important role of initial fabric (sample preparation method) is demonstrated, while the geometry and dimensions of the samples seem to be of secondary importance. The effect of multiple changes of the polarization (i.e. direction) of the cycles was found less important than previously thought. The bundling of a cyclic loading with continuously changing amplitude into bundles of cycles each with a constant amplitude is demonstrated to be conservative. An interesting effect has been observed in tests with bundles of cycles interrupted by monotonic loading phases: The monotonic loading can lead to a loss of the cyclic preloading memory, and thus to an increase of the rate of strain accumulation in the next bundle of cycles. Results from tests with 1D, 2D, 3D and 4D stress and strain paths confirm the amplitude definition for multi-dimensional strain loops incorporated into a high-cycle accumulation (HCA) model. The description of the strain accumulation rates by the HCA model is discussed based on the results from the various test series.

This paper describes a new screening method for evaluating the liquefaction potential of sand deposits with varying percentages of fines. The method is based on measurements of shear wave velocity (V_s), and is developed from a comprehensive experimental program comprising small-strain shear wave testing and large-strain undrained shear tests for sand samples with different quantities of non-plastic fines. A novel point of the method is the unified characterization of shear wave velocity for both clean sand and silty sand through a state parameter that properly combines the effects of void ratio and confining stress in a sound theoretical context. As modern technology has made it more convenient and reliable to measure the shear wave velocity both in the laboratory and in the field, and since the state parameter is a rational index for characterizing various aspects of soil behavior, the proposed method is promising in a wide range of geotechnical applications.

Experimental results have shown that waves propagating in nonhomogeneous and composite materials exhibit dispersive behavior even in cases where they are characterized by isotropic effective properties. Concrete is such a material and at both its fresh and hardened state wave dispersion is observed. The dispersion becomes more pronounced when microdefects like microcracks and air voids appear in the main body of the concrete. The goal of the present work is threefold: first to examine numerically the contribution of various defects to the wave dispersion in concrete, second to show through numerical experiments that circular voids embedded in a concrete matrix is a good approximation for simulating wave dispersion in damaged concrete and third to capture the dispersive behavior of a longitudinal plane wave propagating in damaged concrete with the aid of two theories, namely the multiple wave scattering model of Waterman and Truell and the generalized elastic theory of a strain gradient elastic material. Numerical, experimental and theoretical results are compared and discussed throughout the paper.

The stiffness of soils in compression increases with increasing pressure. This property makes the strain-stress relation nonlinear and strongly influences the propagation of compression waves: a smooth wave front steepens and turns into a shock front. This paper discusses some theoretical questions related to the formation and propagation of longitudinal shock waves in soil within the pressure range typical of geotechnical engineering problems (up to a few megapascals). In particular, the topics discussed in the paper are the critical distance for dry and saturated soils (the distance covered by a smooth front before it becomes discontinuous), the jump conditions on a discontinuity, and smooth viscous shocks.

The use of former opencast mines as lakes has a long tradition in the Rhenish lignite mining area, as does agricultural and forest rehabilitation. There are 57 lakes, that have been created from the final mine voids of a multitude of small-scale opencast mines, varying in size between 0.2 ha and 100 ha. Added to this are lakes that were planned specifically for reasons of landscape preservation in the context of a recultivation scheme (known as landscape lakes). All lakes are used today for a variety of different purposes, i.e. nature conservation, recreation and aquatic sports.

With the operation concentrated on a small number of large-scale opencast mines on the one hand and complete backfilling of opencast mines in the central part of the mining area with boxcut-masses from newer opencast mines (mainly Hambach) on the other hand, a concept was developed for the entire mining area which provides for the creation of three large opencast mine lakes in the active opencast mines of Inden, Garzweiler and Hambach. The planned lake sizes vary between 12 km² and 40 km² at depths of between 180 m and 330 m.

In the creation scheme for the opencast mine lake slopes with a general inclination of 1:5 (below the future wave braking zone) by using main mine equipment are planned. The area where the final water levels will later connected with the groundwater level is to be turned into a wave zone with a minimum width of 100 m and a compensation inclination of between 1:20 and 1:25. Since the lakes do not have a reservoir function and are not situated higher than the surrounding terrain either, the requirements of DIN 19700 (‘Dam plants’) are not relevant. Instead, geotechnical dimensioning of the slopes is carried out in compliance with the Guideline for Stability Analyses (RfS) issued by the Arnsberg regional government as NRW’s competent mining authority.

The filling scheme for the planned opencast mine lakes provides for one-time filling with external water from local streams and rivers. It is flanked by a purposeful continuation of opencast mine dewatering, preventing seepage pressures from the slopes in the direction to the lake.

The stability-related boundary conditions were already taken into account when the slope design was decided on, with seismic impacts also being considered in accordance with the requirements of the RfS. This was done by means of a quasi-static approach according to GOLDSCHEIDER that has been further developed and takes account of the acceleration effects of seismic impacts on the mass of the pore water as well.

A current field of investigation in line with an RfS requirement is the verification of the safety of lake slopes constructed by backfilling against possible liquefaction effects resulting from seismic impacts. A corresponding verification procedure is at present being developed in conjunction with Prof. Triantafyllidis at the Institute of Soil and Rock Mechanics (IBF) of the Karlsruhe Institute of Technology (KIT). It consists of a combination of field investigations, technical-centre and laboratory tests as well as numerical calculations. By comparing the dependency, derived from these tests, of the cyclic stress ratio with shear wave velocity and cone-penetration-test tip pressure with cyclic impacts determined by means of numerical calculations, we will in future be able to establish safety against liquefaction effects with sufficient reliability.

Earthquake stability assessments of large opencast mine slopes (dimensions exceeding several hundred meters) are complex and non-linear problems, often addressed using pseudo-static approaches that neglect material-induced failures and the role of pore-fluids. In this study, a numerical approach is used to understand the dynamic response of saturated and unsaturated soils. Since the requirements for such simulations are often not yet met by commercial software packages, user-defined finite elements and user-defined material models have been implemented in Abaqus/Standard. To account for the large depth of the finite element model, a scaling procedure of the system of equations is proposed to purge the influence of initial stress. The constitutive model parameters are calibrated based on laboratory tests and by back calculation of downhole measurements. Large-scale fully coupled finite element simulations are performed to study the response of a flooded opencast mine under earthquake loading. The present work illustrates the importance of the pore-fluids treated as independent phases in the context of seismic analysis of slopes. The simulations show strong wave diffraction effects for inhomogeneous dump structures, resulting in smaller displacements in near-surface areas of the slope. Furthermore, it was found that large areas of the dump show a (significant) temporary decrease of effective mean pressure.

This paper describes a new methodology for predicting and controlling ground movements for a tunnel that traverses an interface between two clay layers of contrasting shear strength and stiffness. We use non-linear finite element analyses to investigate effects of face pressure on tunnel face stability and steady state ground surface deformations. This leads to a series of design charts linking soil properties (undrained shear strength and stiffness), stratigraphy (layer interface), tunnel cover depth and mechanized control parameters (face and grout pressure). The methodology has been validated for a recent case study where EPB tunnels traverse an interface between stiff clay (Old Alluvium) and soft, Marine clays. The proposed methodology successfully uses the measured face pressures to predict ground movements for these mixed face conditions. Further generalization of the method is now needed to represent mixed face conditions with contrasting permeability.

The paper deals with the numerical modelling of the blast-induced deformation of two neighbouring shallow tunnels and the surrounding soil. The deformation is caused by an explosion inside one of the tunnels. The explosion is simulated by a short-term pressure load of moderate amplitude (8 MPa) applied to the tunnel lining. The lining of both tunnels is circular with an inner diameter of 9.6 m and consists of concrete segments (tubbings) assumed to be linearly elastic. The tunnels are located at a depth of 17 m in fully saturated soil. The effective stresses in the soil are described by a hypoplasticity model. The modelling incorporates pore water cavitation at zero absolute pore pressure. The dynamic problem is solved in a two-dimensional plane-strain formulation with the finite-element program Abaqus/Standard. The transient deformation of the tunnel lining and the soil is analysed in detail. In particular, the solution reveals the emergence of large cavitation zones in the soil during the dynamic deformation.

Borehole breakouts, as well as breakouts in tunnels and shafts, are a common occurrence, especially under high in situ stresses or stress states with high deviatoric component. Though they can pose a risk to stability, often they are of use, especially in deep boreholes, as they can help to determine to a certain extent the primary in situ stress. Observations have shown that while their depth evolves, their width remains constant. Currently the width only is used in conjunction with the Kirsch analytical solution to establish a linear relationship between the two in plane principal primary stress components. The stress state cannot be fully determined since one equation is available (failure criterion) for two unknowns. A recently proposed numerical tool based on conformal mapping is used in this work to simulate the formation of shear breakouts and investigate the feasibility of the determination of both principal primary in situ stress components, by making use of both the depth and the width of the breakout. Concluding, recommendations are provided for the use of the proposed methodology and limitations of its applicability are discussed.

The Rosenheimer Basin in the alpine region near the German-Austrian border is a deep ancient glacial lake, which has been filled with fine-grained sediments over the course of the past 10.000 to 100.000 years. Due to the rapid economic growth in the region over the last few decades, the high population density and the increasing utilisation of infrastructure in the region a significant demand for development exists, particularly for structures with large loads. The design and construction of such structures represents a significant challenge for engineers due to the sensitive soft, fine-grained lacustrine sediments. This article focusses on the conception and design of a cable-stayed bridge pylon in Rosenheim. The foundation consists of bored piles connected to a pile cap, with additional displacement piles and vertical drains for soil improvement. As the stiffness and strength of the sensitive lacustrine sediments are strongly reduced due to the disturbances caused by pile installation, displacement piles in combination with vertical drains around the piles are prescribed in order to re-consolidate the soil (“soil healing”) and increase the shaft friction after the pile installation. In order to take into account the effect of the soil disturbance and the healing effect of the soil improvements on the foundation behaviour realistically, high quality pile loading tests were carried out. Based on the results of laboratory and field tests, and the simulation of the single pile loading tests using a visco-hypoplasticity constitutive model, a 3D Finite-Element Model was developed and calibrated to predict the time-dependent behaviour of the mixed foundation. The numerical prediction shows that the serviceability requirements of the foundation can be fulfilled and the scheduled underpinning of the superstructure to compensate the foundation settlements will likely not be required prior to 50 years of operation. An extensive monitoring program will be implemented during the construction to validate and, if necessary, to adjust the numerical model to realistically predict the long-term deformation behaviour of the bridge foundations.

The development of diaphragm wall technology in Europe began at the beginning of the 1950s with the work of Christian Veder and Hans Lorenz. The advantages of this special civil engineering method quickly became apparent in the construction of inner-city excavation pits and infra-structural measures in the form of low-deformation, water-impermeable reinforced concrete walls, as well as in the use of sealing walls in earth dam and landfill construction.

In the beginning, the focus was on the investigation of the supporting effect of bentonite suspension in the open slot, but soon the technical developments of the excavation tools followed. After setbacks in the quality of the walls, the pioneers of the method tried to identify the sources of error and to avoid them by consistently adhering to the self-imposed specifications. To this day, the process is a speciality in terms of planning and execution and requires a great deal of experience on the part of those involved. The project-specific quality assurance of diaphragm wall construction sites is a decisive aspect for the safe and economical execution of the work.

The stability and geometric nonlinearities of slender structures are a major topic in structural design. While this topic is most relevant in the field of Structural Engineering, e.g. for slender steel or concrete structures, only few applications take the role of soil-structure-interaction explicitly into account. The focus of this paper is placed on the impact of soil support and its modelling for the buckling analysis based on examples both for pile foundations and for railway track stability. The general interaction between steel design and the geotechnical input is addressed.

Three methods in engineering practice are mainly implemented to investigate the behavior of excavation walls. In the majority of the cases beam models with classical supports seem to be sufficient.

For modelling more accurate the deformation behavior of the foot of the wall it may be worth to improve the prediction of the deformation behavior using a subgrade reaction model. In cases of more complex pit geometries and soil conditions a finite element analysis may be more appropriate.

All these three methods of calculation are shortly described in the paper and compared with each other.

With the consideration of bound theorems an attempt is made in the paper to discuss the safety issues. All the presented three methods can be used to calculate the wall deformations. It is demonstrated that even a finite element analysis has limitations in cases of deformation predictions induced due to geotechnical installation processes in the vicinity of the wall. Data from field records may be used to estimate the order of magnitude of wall deformations due to installation processes. As an example of using field data in the wall deformation prediction and FEM in a recent research project the vibroinstallation of uplift piles near to the wall has been used.

The numerical results show quite satisfactorily that the new developed model may serve as a basis for the prediction of wall deformations due to some installation processes.

From 1998 to 2008 Prof. Triantafyllidis was head of the chair of foundation engineering and soil mechanics of the Ruhr-Universität Bochum. During that time, he initiated a number of research projects and the topics of these initiatives came along with the scientific work of Prof. Triantafyllidis for the next decades. But also in Bochum these research activities created a base for further developments. Within this contribution a number of the research initiatives of Prof. Triantafyllidis are recalled and two examples are given, how these works are influencing the developments in Bochum up to know.