Modified bentonites for soil-bentonite cutoff wall applications with hard mix water

Abstract

The influence of mix water hardness and bentonite type on the quality of bentonite slurry and hydraulic performance of soil-bentonite backfill mixtures was investigated through slurry quality testing and flexible-wall hydraulic conductivity (k) testing of soil-bentonite backfill specimens. A conventional sodium bentonite (NG) and two types of modified bentonite, HYPER clay (HC) and multiswellable bentonite (MSB), were used to prepare the slurries and backfills. Slurries were prepared with mix waters containing CaCl2 concentrations (CM) ranging from 0.5 to 25 mM (hardness, HM = 50–2500 mg/L as CaCO3), and backfills with similar total bentonite contents (5.5–5.9%) were prepared by combining dry sand-bentonite mixtures with the slurries in a benchtop mixer. Backfill specimens were permeated using CaCl2 solutions with concentrations, CP, ranging from 0.5 to 50 mM. The results of the slurry testing showed that the impact of the mix water on slurry quality parameters (Marsh viscosity, filtrate loss, and bleed) was dependent upon CM, bentonite type, and bentonite content. For CM ≤ 5 mM (HM ≤ 500 mg/L), slurry containing 5% NG exhibited adequate viscosity, filtrate loss, and bleed. However, NG contents of 7% and 12% were required to obtain adequate viscosity, filtrate loss, and bleed for CM = 10 mM (HM = 1000 mg/L) and 25 mM (HM = 2500 mg/L), respectively, whereas adequate slurry properties were obtained for CM = 10 mM using lower percentages (5–6%) of HC or MSB. In terms of backfill k, the specimens containing HC amended with polyanionic cellulose appeared to be the most resilient against hard mix water. The k of backfill specimens containing NG generally exceeded 10−9 m/s when prepared using slurry with CM = 10 mM and exceeded 10−8 m/s when prepared using slurry with CM = 25 mM. In contrast, HC backfill specimens prepared using slurry with CM = 10 or 25 mM exhibited k ranging from 4 × 10−11 to 5 × 10−10 m/s. Overall, CM had greater influence on backfill k relative to CP for the range of CP considered in this study.