Repository logo

Membrane behavior, diffusion, and compatibility of a polymerized bentonite for containment barrier applications




Bohnhoff, Gretchen L., author
Shackelford, Charles D., advisor
Benson, Craig H., committee member
Borch, Thomas, committee member
Sale, Thomas C., committee member

Journal Title

Journal ISSN

Volume Title


Conventional (untreated or unmodified) bentonites are commonly used in hydraulic containment barriers to contain liquid flow and contaminant transport, because of the ability of bentonite to swell and achieve low hydraulic conductivity to water, substantial membrane behavior, and low solute diffusion coefficients. However, conventional bentonites also have been shown to be affected adversely by environmental conditions that promote multivalent-for-monovalent cation exchange. In this study, the membrane behavior and diffusive properties of a polyacrylic acid modified bentonite referred to as a bentonite polymer nanocomposite, or BPN, were determined through the simultaneous measurement of membrane efficiency coefficients, ω, and solute diffusion coefficients, D*, during combined multi-stage membrane and diffusion tests using either potassium chloride (KCl) with concentrations ranging from 4.7 mM to 54 mM or calcium chloride (CaCl2) with concentrations ranging from 5 mM to 20 mM. The BPN exhibited substantial membrane behavior when exposed to KCl with values of ω that were higher than those previously reported for conventional (unmodified) bentonite under similar testing conditions. For example, the ω value measured in this study for a BPN specimen contained within a rigid-wall cell and based on circulation of 20 mM KCl was 0.43, whereas that previously reported for a GCL specimen containing a conventional bentonite under similar testing conditions except at a lower porosity (0.74 vs. 0.92) was only 0.30. Also, in contrast to previously reported results for conventional bentonite, the membrane behavior of the BPN was sustained when exposed to 5 mM CaCl2, and values of ω for the BPN were higher than those previously reported for conventional and other modified bentonites. For example, the value of ω for the BPN tested in a rigid-wall cell with 5 mM CaCl2 was 0.95, whereas the ω values for an anionic polymer modified bentonite, known as Hyper clay, and a GCL were 0.13 and 0, respectively. However, exposure of specimens of the BPN to 10 mM CaCl2 for a test conducted in a rigid-wall cell and 20 mM CaCl2 for a test conducted in a flexible-wall cell did ultimately result in complete destruction of the membrane behavior. The destruction of the membrane behavior of the specimen in the rigid-wall test was attributed to short-circuiting along the side-walls of the rigid cell after shrinkage of the BPN specimen, whereas the destruction of the membrane behavior of the specimen in the flexible-wall test correlated with the time to reach steady-state diffusion of calcium (Ca2+). Similar to a previous study involving a conventional bentonite, the diffusive properties of the BPN also were shown to correlate well with the membrane behavior of the BPN, such that that the diffusive solute mass flux decreased as the membrane efficiency of the BPN increased. However, in contrast to previous test results, the steady-state values of D* for K+ and Ca2+ were not only not equal to but also lower than the D* value for Cl- at steady state, although the differences between the D* for K+ or Ca2+ versus that for Cl- diminished with increasing source concentration of KCl or CaCl2, respectively. This inequality between salt cation and salt anion D* values at steady state was attributed to the complicating existence of significant excess Na+ that was initially present within the specimen of BPN prior to testing and contributed to satisfying the requirement for electroneutrality, a contribution that diminished with time as the Na+ diffused out of the specimen. Finally, the use of BPN in soil-bentonite (SB) backfills of vertical cutoff walls was investigated. The hydraulic conductivity, k, to tap water, the consolidation behavior, and the chemical compatibility (Δk) based on permeation with CaCl2 solutions of SB backfills amended with BPN were evaluated and compared with those for a backfill comprised of a conventional bentonite. Although the backfills containing BPN were more sensitive to stress conditions than the backfill containing conventional bentonite, the overall hydraulic performance of a backfill containing 5 % dry BPN was better than that of the backfill containing 5 % dry conventional bentonite by approximately two orders of magnitude in terms of k. Overall, the BPN exhibited improved membrane and diffusion properties relative to conventional and other modified bentonites previously tested under similar conditions. However, the improved membrane behavior of the BPN was ultimately destroyed upon exposure to 10 mM CaCl2 in a rigid-wall cell and 20 mM CaCl2 in a flexible-wall cell. Also, despite an overall lower k of the sand-BPN backfills relative to a backfill comprised of the same sand but a conventional bentonite upon permeation with a 50 mM CaCl2 solution, the chemical resistance of the sand-BPN backfills in terms of changes in k was not any better than that for the sand-conventional bentonite backfill. Thus, the beneficial behavior of the BPN was not unlimited nor without issues, such that any perceived benefit of polymerized bentonites must first be properly characterized on a case-by-case basis prior to use.


2012 Fall.
Includes bibliographical references.

Rights Access


geosynthetic clay liner
clay membranes
polymer modified bentonites


Associated Publications