Geotechnical Testing Journal, Vol. 28, No. 1 Paper ID GTJ12520 Available online at: www.astm.org Jae-Myung Lee 1 and Charles D. Shackelford 2 Solution Retention Capacity as an Alternative to the Swell Index Test for Sodium Bentonite ABSTRACT: A new test, referred to as the solution retention capacity (SRC), is proposed as a potentially less tedious and quicker alternative to the swell index (SI) test (ASTM D 5890) for bentonite. The SRC represents the amount of a 50-mL solution retained in one gram of oven-dried bentonite after centrifugation at 2750 g, and the SRC values are reported in units of mL/g. The SRC for two sodium bentonites from geosynthetic clay liners measured using deionized water and solutions containing calcium chloride (CaCl 2 ) at concentrations ranging from 5 mM to 500 mM is compared with the SI for both bentonites and the same test liquids. In general, both SI and SRC decrease with increasing CaCl 2 concentration due to a decrease in the thickness of adsorbed layer of the bentonite particles. However, the SI is greater than the SRC, with SI falling in the range of 7.5 mL/2 g ≤ SI ≤ 30 mL/2 g and SRC falling in the range of 1.7 mL/g ≤ SRC ≤ 7.2 mL/g. The difference in the magnitudes of SI relative to SRC is attributed, in part, to the differences in the units (i.e., mL/2 g for SI versus mL/g for SRC) and the accelerations (i.e., 1 g in SI versus 2750 g in SRC) for the two procedures. Also, the SI includes the volume of the solid bentonite, whereas the SRC does not. KEYWORDS: bentonite, geosynthetic clay liner, index properties, swell index Introduction In the presence of water, sodium montmorillonite can swell to as much as 20 times its own volume (Norrish 1954). This high swelling potential of montmorillonite is attributed to adsorption of hydrated cations and water molecules, resulting in strong repul- sive forces and interlayer expansion in the presence of electrolyte solutions (Norrish 1954; Mitchell 1993; Shackelford et al. 2000). The primary factors contributing to the high swelling potential of montmorillonite relative to other clay minerals, such as illite and kaolinite, include a relatively high specific surface (i.e., 100– 800 m 2 /g) and a relatively high net negative charge as reflected by relatively high cation exchange capacities that typically range from 80 meq/100 g to 150 meq/100 g (Grim 1953; Norrish and Quirk 1954; Mitchell 1993). Since montmorillonite also is the primary mineral in most bentonites (typically 60–90 %), bentonites gener- ally exhibit the same swelling characteristic as montmorillonite. In fact, the low hydraulic conductivity values to water typically mea- sured for sodium bentonites (i.e., 10 −8 –10 −9 cm/s) are attributed to the relatively high swelling potential of the bentonite in the presence of water (Mitchell 1993; Shackelford et al. 2000). Two methods historically have been used to determine the swelling behavior of bentonite from geosynthetic clay liners (GCLs), i.e., Geosynthetic Research Institute (GRI) GCL-1, Free Swell of the Clay Component of Geosynthetic Clay Liners (e.g., Narejo and Memon 1995; Lin and Benson 2000), and ASTM D 5890, Standard Test Method for Swell Index of Clay Mineral Component of Geosynthetic Clay Liners, (e.g., Didier and Comeaga 1997; Ruhl and Daniel 1997; Shackelford et al. 2000; Egloffstein 2001; Jo et al. 2001; Ashmawy et al. 2002; Shan and Lai 2002). Received January 16, 2004; accepted for publication June 22, 2004; published January 2005. 1 Post-Doctoral Research Assistant and former Graduate Research Assistant, Department of Civil Engineering, Colorado State University, Fort Collins, CO 80523. 2 Professor, Department of Civil Engineering, Colorado Sate University, Fort Collins, CO 80523, shackel@engr.colostate.edu. The primary differences between these two tests are the method of hydration (i.e., 0.7-kPa confinement for GRI GCL-1 versus self- weight sedimentation for ASTM D 5890) and the longer equilib- rium time required for GRI GCL-1 (i.e., >400 h versus <24 h) relative to that required for ASTM D 5890 (Lin and Benson 2000; Shackelford et al. 2000; Jo et al. 2001). However, only the test spec- ified in ASTM D 5890 (i.e., swell index test) is currently accepted as a standard by the industry (Shackelford et al. 2000). In general, the quality of bentonites is indicated macroscopically by the magnitude of the swelling capacity in the presence of water (i.e., the greater the montmorillonite content, the greater the swelling capacity). As a result, swell index tests have been used by the industry as a prescreening method for qualitatively assess- ing the quality of bentonites. In addition, several studies have shown that the swelling behavior of bentonite is directly correlated with the hydraulic conductivity of bentonite-based GCLs (Egloffstein 1995; Didier and Comeaga 1997; Ruhl and Daniel 1997; Shackelford et al. 2000; Egloffstein 2001; Jo et al. 2001; Kashir and Yanful 2001; Ashmawy et al. 2002; Shan and Lai 2002). For example, Shackelford et al. (2000) and Jo et al. (2001) cor- relate the results of swell index tests on sodium bentonite taken from a GCL using solutions with various concentrations of sev- eral inorganic compounds (e.g., NaCl, CaCl 2 , and LaCl 3 ) with the hydraulic conductivity of the GCL permeated with the same test solutions. Their results show that a decrease in swell index of the bentonite with increasing electrolyte concentration and/or cation valence is correlated with an increase in hydraulic conduc- tivity of the GCL with increasing electrolyte concentration and/or cation valence. The observed effects are consistent with changes in the thickness of adsorbed layer such that an increase in cationic valence and/or electrolyte concentration reduces the swell index of sodium bentonite. Thus, the results of swell index tests may provide a qualitative indication of the expected hydraulic behav- ior of bentonite-based materials (e.g., GCLs), thereby providing a relatively simple, rapid, and inexpensive method for assessing the compatibility between the bentonite and the permeant liquid (Didier and Comeaga 1997; Ruhl and Daniel 1997; Shackelford et al. 2000; Copyright © 2005 by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. 61