Abstract:Void ratio has traditionally, been used as a state variable for predicting the liquefaction behaviour of soils under the Critical State Soil Mechanics framework. Recent publications show that void ratio may not be a good parameter for characterizing sand with fines. An alternative state variable referred to as equivalent granular void ratio has been proposed to resolve this problem. To calculate this alternative state variable, a bparameter is needed. This b-parameter represents the fraction of fines that actively participate in the force structure of the solid skeleton. However, predicting the "b" value is problematic. Most, if not all, of the b-values reported were determined by casespecific back-analysis, that is, the b-value was selected so that a single correlation between equivalent granular void ratio and the measured steady state strength (or cyclic resistance) could be achieved. This paper examines the factors affecting the "b" value based on published work on binary packing. This leads to a simple semi-empirical equation for predicting the value of "b" based on fines size and fines content. Published data appears to be in support of the proposed equation. A series of experiments were conducted on a specially designed sand-fines type in order to provide additional validation of the proposed equation and to reinforce the use of equivalent granular void ratio in a more generalized context. KEY WORDS:Void ratio, state variable, Steady State, Critical State, liquefaction One of the possible reasons that lead to the lack of a single Steady State Line may be due to the use of void ratio as a state variable. This is because of the non-active participation of fines in the force structure of a sand-fines mixture. To resolve this problem, a new state variable referred to as intergranular void ratio was proposed by considering the non-active fines as voids. It has been suggested that the intergranular void ratio should be used instead of void ratio for low fines content (Georgiannou et al. 1990;Georgiannou et al. 1991; Hight and Georgiannou 1995; Kuerbis et al. 1988;Ovando-Shelley and Pérez 1997, Thevanayagam 1998). However, intergranular void ratio is not applicable at higher fines content (Ni et al. 2006;Ni et al. 2004; Thevanayagam 2000;Yang et al. 2006a;2006b). Thevanayagam (2000) introduced the concept of equivalent granular void ratio for higher fines content and this approach requires an additional parameter "b" that presents the fraction of fines participating in the force structure of the solid skeleton. However, the prediction of b-value is a problematic and controversial issue (Ni et al. 2004; Thevanayagam 2001 Literature Review Intergranular void ratioMitchell (1976) pointed out the non-active role of fines in a granular phase structure and used such concept to determine the amount of non-active clay content. Kenney (1977) found that a clayey sand with a combined fines and water content of less than 40% to 50% by volume had a residual strength close to that of the host sand with a void sp...

This paper presents modelling of the consolidation of foundation soil under a wide embankment constructed over soft soil. An elastic–viscoplastic (EVP) constitutive model is used to represent the foundation soil for the coupled finite element analysis (FEA). A unit-cell analysis is carried out to capture the maximum settlement and the development of excess pore-water pressure with time below the centreline of the embankment for a long period (9 years). A new function for capturing the varying nature of the creep or secondary compression coefficient is proposed and used in association with the EVP model. The input material parameters for this study were determined from extensive laboratory experiments except for the equivalent horizontal permeability, which was systematically estimated by using vertical permeability data obtained from one-dimensional consolidation tests and by back-analysing the first 12 months of field settlement data. Comparisons are made among the predictions obtained adopting an elastoplastic modified Cam clay model and the EVP model with constant and varying creep coefficients for the foundation soil and the corresponding field data. The predictions with the EVP model are found to be better than those with the elastoplastic model and the use of a varying creep coefficient for the EVP model seems to further improve its predicting ability.

This paper presents the long-term performance of a wide geogrid-reinforced road embankment constructed on soft clay improved with prefabricated vertical drains (PVDs) at a freeway extension site 150 km north of Sydney in Australia. The foundation soil and the embankment were instrumented and monitored for about 400 days for excess pore-water pressure, earth pressure, and reinforcement tension, and for 9 years for displacement profiles. The embankment was constructed in stages and surcharged in an attempt to reduce post-construction settlement. As the embankment width was wide relative to the thickness of the soft clay, the settlement near the centre was modelled by a unit cell analysis. The equivalent horizontal permeability was determined by back analysis of the central zone using the first 12 months of settlement data. All other soil parameters were determined from the laboratory and field testing. The predicted pore-water pressure response over the first 400 days showed reasonable agreement with measured values. The same analysis was then continued to predict settlement over a period of 9 years. The predicted settlement was, however, smaller than the measured value at the centre region of the embankment.Résumé : Cet article présente la performance à long terme d'un large remblai routier armé de géogrid construit sur de l'argile molle améliorée avec des drains verticaux préfabriqués (« PVDs ») sur le site d'une extension d'autoroute à 150 km au nord de Sydney en Australie. Le sol de fondation et le remblai ont été instrumentés et des mesures ont été prises durant une longue période, d'environ 400 jours pour l'excédent de la pression interstitielle, pour la pression des terres et pour la traction dans l'armature, et durant 9 ans pour les profils de déplacement. Le remblai a été construit en stages et surchargé pour tenter de réduire le tassement postconstruction. Comme la largeur du remblai était grande par rapport à l'épaisseur de l'argile molle, le tassement près du centre a été modélisé par une analyse de cellule unique. La perméabilité horizontale équivalente a été déterminée par une analyse à rebours de la zone centrale en utilisant les premiers 12 mois de données de tassement. Tous les autres paramètres de sol ont été déterminés par des essais sur le terrain et en laboratoire. La réaction prédite de la pression interstitielle durant les 400 premiers jours a montré une concordance raisonnable avec les valeurs mesurées. La même analyse a alors été poursuivie pour prédire le tassement durant une période de 9 ans. Cependant, le tassement prédit était plus petit que la valeur mesurée dans la région centrale du remblai.