Physical modelling of cohesive soil inherent variability: consolidation problem

Reza Jamshidi Chenari,Malahat Ghoreishi,Ali Taleb,Meghdad Payan

doi:10.1186/s40703-018-0094-y

Abstract

The main goal of this study is to substantiate the heterogeneous soil modelling in the laboratory and to examine the effect of spatial variability on the soil consolidation properties. For this purpose, several heterogeneous soil physical models are constructed using nine homogeneous soil clusters, which are prepared by mixing variable proportions of kaolin and bentonite at a specific water content equal to the liquid limit of the mixtures. The index properties of different homogeneous kaolin–bentonite clusters are utilized to model the spatial variability of the heterogeneous soil models using the random field generation approach. The physical model contains different discrete cells where the variability is controlled by the specific realizations of a random field. The number of these realizations in the heterogeneous model is augmented by varying the liquid limit of the mixtures. The constructed physical models are then subjected to several one-dimensional consolidation stages using the oedometer apparatus and the influence of the heterogeneity of the soil sample on its hydraulic parameters is evaluated. The results show that the higher variability involved in the soil strata, expressed through the larger coefficient of variation and smaller correlation distance, leads to the increase in the hydraulic conductivity and coefficient of consolidation of the heterogeneous soil physical model. It is also revealed that the experimental modelling of soil spatial variability can shed light on the numerical simulation of random field theory presented in several previous studies.

Highlights

Soil, resulting from several physical, chemical, geological and geo-environmental phenomena, is a complicated structural material which is highly heterogeneous in nature in terms of mechanical properties
The samples were prepared by mixing dry powders of kaolin, bentonite and sufficient amount of colorant, passing sieve No 200 and the initial water content of the mixture was considered equal to its respective liquid limit
Considering the preliminary properties of the kaolin–bentonite mixtures and adopting variable specific realizations of their liquid limits, different discrete cells in the heterogeneous physical model were associated with the particular random soil properties

Summary

Introduction

Soil, resulting from several physical, chemical, geological and geo-environmental phenomena, is a complicated structural material which is highly heterogeneous in nature in terms of mechanical properties. Implementing well-established numerical modellings along with the incorporation of the random field theory and Monte Carlo simulations, innumerable studies have been performed throughout the literature regarding the introduction of uncertainty involved in the estimation of soil properties into the variety of geotechnical problems, including bearing capacity and settlement of shallow and deep foundations, slope stability and liquefaction hazards [2, 4, 6,7,8, 11,12,13,14,15,16, 18–20, 22] Despite these comprehensive studies on the numerical simulation of soil spatial variability and due to the large amount of cost and time required to construct full-scale physical models, there exist very few studies in the literature on the experimental verification of the numerical models, inspired by the seminal contributions of Garzón et al [9], Ghoreishi [10] and Taleb [21]. This novel approach was later employed by Ghoreishi [10] and Taleb [21] in separate studies to reconstitute the heterogeneous physical models for the characterization of their properties in direct shear apparatus

Objectives

Methods

Findings

Conclusion