Estimation of specific surface areas of coarse- grained materials with grain-size curves represented by two-parameter lognormal distributions M. Mbonimpa A Department of Applied Sciences, Université du Québec en Abitibi-Témiscamingue (UQAT), Rouyn-Noranda, Québec, Canada M. Aubertin 1,A , R.P. Chapuis 1,A , A. Maqsoud 2,A , B. Bussière 2,A 1 Department of Civil, Geological and Mining Engineering, École Polytechnique de Montréal, Montreal, Québec, Canada. 2 Department of Applied Sciences, Université du Québec en Abitibi-Témiscamingue (UQAT), Rouyn-Noranda, Québec, Canada A Industrial NSERC Polytechnique-UQAT Chair, Environment and Mine Wastes Management. ABSTRACT The specific surface area (SSA) is an important characteristic of soils and other particulate media. The value of SSA can, for instance, be used to assess the hydraulic conductivity and moisture retention curve. Various methods have been proposed to evaluate the SSA of granular materials from the grain size curves. In this paper, a new approach is presented to estimate SSA for S-shaped grain-size distributions that can be represented by two-parameter lognormal distributions, using the equivalent mean diameter DH. Calculated values from this method are compared with those obtained from existing analytical equations that rely on grain size curve parameters. The introduction of the proposed relationship in the modified Kovács (MK) model developed to predict the water retention curve is also discussed in a preliminary manner. RÉSUMÉ La surface spécifique (SS) est une caractéristique importante des sols et autres matériaux particulaires. La valeur de SS peut par exemple être utilisée pour estimer la conductivité hydraulique et la courbe de rétention d’eau. Diverses méthodes ont été proposées pour évaluaer la SS des matériaux granulaires à partir des courbes granulométriques. Dans cet article, une nouvelle approche est présentée pour estimer la SS de matériaux dont les courbes granulométriques en forme de S peuvent être représentées par des distributions log-normales à deux paramètres, en utilisant le concept du diamètre moyen équivalent DH . Les valeurs calculées ici sont comparées à celles obtenues avec d’autres équations analytiques existantes basées sur les paramètres de la courbe granulométrique. L’introduction de la relation proposée dans le modèle de Kovács modifié (MK) développé pour prédire la courbe de rétention d’eau est analysée de façon préliminaire. 1 INTRODUCTION Many physical, mechanical, and chemical properties of soils and other particulate materials are related to surface phenomena that occur at the interface between a fluid (liquid or gas) and the solid grains. Some of these properties have been correlated to the specific surface area (SSA) of the solid phase, which is assumed to correspond to the interstitial surface area of the voids in the porous media. In hydrogeology and geotechnique for instance, the SSA is sometimes used to predict the hydraulic conductivity and moisture retention curve of soils and similar materials such as tailings (e.g. Chapuis and Montour, 1992; Aubertin et al. 1996, 1998; Mbonimpa et al. 2002; Chapuis et Aubertin, 2003). The value of the specific surface area can be related to parameters As, Ms, Vs, and Vt which represent the total surface area of particles, and their mass, volume of solids and total volume, respectively. Three distinct specific surface areas can be defined: a solid mass-based value Sm=As/Ms [L 2 /M], a solid particle volume-based value Ss=As/Vs [L 2 /L 3 ], and a total volume-based value Sv=As/Vt [L 2 /L 3 ]. These three SSA expressions are interrelated in the following manner: ( ) s s s v m  n 1 S  S S ⋅ - = = [1] where ρs [M/L 3 ] is the density of solid grain and n is total porosity of the medium. In this paper, the SSA value will be defined from the mass-based (Sm) expression. The value of Sm can be directly measured, using various methods having different ranges of applicability (e.g., Lowell and Shields 1984; Igwe 1991; Arnepalli et al. 2008). Methods based on physisorption isotherms, such as the well known BET method, are particularly useful but these results must be interpreted with great care. In fact, none of the existing methods provide absolute values of the SSA. As these measurement techniques require the 1607 GeoHalifax2009/GéoHalifax2009