Effect of specimen preparation on the swell index of bentonite-polymer GCLs
Introduction
Geosynthetic clay liners (GCLs) are factory-manufactured products that consist of a thin layer of natural or activated sodium bentonite (Na–B) sandwiched between two geotextiles or glued to a geomembrane (Koerner, 2012). GCLs are widely used as hydraulic barriers in waste containment applications in lieu of compacted clay liners due to ease of installation, increase in waste storage capacity (due to reduced thickness), and low hydraulic conductivity to water (<1.0 × 10−10 m/s) (Koerner, 2012). When incorporated between two geotextiles, the hydraulic conductivity of GCLs is primarily controlled by the Na–B (Petrov et al., 1997a; Shackelford et al., 2000; Jo et al., 2001; Kolstad et al., 2004; Lee and Shackelford, 2005; Setz et al., 2017; Chen et al., 2018; Rowe, 2020). The dominant mineral in the Na–B used in GCLs is montmorillonite and has Na⁺ as the predominant interlayer cation. When hydrated with water or dilute leachate, Na–B undergoes osmotic swelling due to “chemico-osmosis,” i.e., the flow of H2O from lower ion concentration (higher H2O chemical potential) outside the interlayer region of the montmorillonite particles to higher ion concentration (lower H2O chemical potential) inside the interlayer region of the montmorillonite particles, which results in the formation of a thick layer of ions and water molecules (so-called “diffuse double layer”) around the montmorillonite particles (McBride, 1994). This thick diffuse double layer causes separation of the naturally aggregated montmorillonite particles which reduces the intergranular pore spaces in the bentonite fabric resulting in low hydraulic conductivity (<1.0 × 10−10 m/s) (Mesri and Olson, 1971; Shackelford et al., 2000; Jo et al., 2001; Kolstad et al., 2004; Chen et al., 2018). Furthermore, the ability of Na–B to swell enables small defects like punctures that may occur during installation to self-heal (Sari and Chai, 2013; Li and Rowe, 2020).
However, aggressive leachates such as those that predominantly have high ionic strength (e.g., >300 mM), polyvalent cations, and/or extreme pH (2 > pH > 12) can inhibit osmotic swelling of Na–B due to the low concentration gradient between the permeant solution and the interlayer region of the montmorillonite particles, resulting in a thin diffuse double layer, large intergranular pore spaces, and consequently high hydraulic conductivity of Na–B GCL (>1.0 × 10−10 m/s) (Petrov and Rowe, 1997; Shackelford et al., 2000; Jo et al., 2001, 2004, 2005; Katsumi et al., 2001; Kolstad et al., 2004; Lee et al., 2005; Bradshaw and Benson, 2014; Bradshaw et al., 2016; Chen et al., 2018; Wang et al., 2019; Zainab and Tian, 2020). Because of this issue, significant research has focused on improving the chemical compatibility of Na–B GCLs to aggressive leachates by chemically modifying or blending bentonite with organic molecules and polymers, which are hypothesized to activate osmotic swelling, limit cation exchange or clog intergranular pores of the bentonite fabric, therefore yielding low hydraulic conductivity (Scalia et al., 2018a). For example, Onikata et al. (1996) developed multi-swellable bentonite (MSB) by mixing dry Na–B with propylene carbonate (PC) solution. Similarly, Di Emidio (2010) created HYPER clay by mixing dry Na–B with sodium carboxymethyl cellulose (Na-CMC) polymer solution. Scalia et al. (2014) also produced a bentonite-polymer composite (BPC) by polymerizing acrylic acid in a Na–B slurry to form sodium-polyacrylic acid (Na-PAA). Most manufactures sell GCLs containing a dry mixture of Na–B and proprietary polymer(s) (Salihoglu et al., 2016; Tian et al., 2016, 2017, 2019; Tian and Benson, 2017; Scalia et al., 2018a; Chen et al., 2019). These commercial GCLs are collectively referred to herein as B–P GCLs.
The hydraulic conductivity of B–P GCLs has been reported to be significantly lower than Na–B GCL, depending on the chemistry of the permeant solution and the quantity of polymer added to the bentonite (Salihoglu et al., 2016; Tian et al., 2016, 2017, 2019; Tian and Benson, 2017; Scalia et al., 2018a; Chen et al., 2019). For instance, Athanassopoulos et al. (2015) measured the hydraulic conductivity of a B–P GCL and Na–B GCL permeated with a bauxite liquor (with pH = 13 and ionic strength = 2.35 M) and a trona ash leachate (with pH = 11 and ionic strength = 1.05 M). The hydraulic conductivity of the B–P GCL to the bauxite liquor and trona ash leachate was less than 1.0 × 10−10 m/s, whereas the Na–B GCL had hydraulic conductivity greater than 1.0 × 10−8 m/s. Chen et al. (2019) also reported the hydraulic conductivity of B–P GCLs permeated with coal combustion product (CCP) leachates. The B–P GCLs (with polymer content > 1.9%) maintained low hydraulic conductivity (<1.0 × 10−10 m/s), whereas the Na–B GCLs showed high hydraulic conductivity (>1.0 × 10−8 m/s).
The hydraulic conductivity of GCLs typically is measured in the laboratory as per ASTM D5084 and ASTM D6766. Based on these standards, a hydraulic conductivity test is continued until hydraulic and chemical equilibrium are achieved in order to reflect the long-term hydraulic conductivity of GCLs, which may require the test durations lasting from a few months to years (Petrov and Rowe, 1997; Shackelford et al., 2000; Katsumi et al., 2001, 2007, 2008; Kolstad et al., 2004; Jo et al., 2005). The time required to reach chemical equilibrium can be even longer for B–P GCLs (Scalia et al., 2014; Razakamanantsoa and Djeran-Maigre, 2016; Tian et al., 2016; Donovan et al., 2017; Chen et al., 2019). Hence, quick and inexpensive index tests are often performed on the bentonite component of GCLs to assess indirectly the chemical and hydraulic compatibility of the GCL with the permeant liquid (Jo et al., 2001; Lee et al., 2005; Katsumi et al., 2008; Rosin-Paumier et al., 2010).
The swell index (SI) test (ASTM D5890) is a commonly used index test to assess the hydraulic performance and the chemical compatibility of GCLs to leachates. Two grams of bentonite are ground until 100% passes the No. (#) 100 mesh, and a minimum of 65% passes the #200 mesh U.S. standard sieve. Numerous studies have shown that a correlation exists between SI and hydraulic conductivity of Na–B (Jo et al., 2001, 2004; Kolstad et al., 2004; Olsta et al., 2004; Lee et al., 2005; Katsumi et al., 2008; Guyonnet et al., 2009; Chen et al., 2018). For Na–B GCLs, SI > 14 mL/2 g generally correlates to hydraulic conductivity <1.0 × 10−10 m/s, whereas SI < 14 mL/2 g generally correlates to hydraulic conductivity >1.0 × 10−10 m/s (Jo et al., 2001; Lee et al., 2005; Chen et al., 2018). However, the SI and hydraulic conductivity of B–P GCLs are not well correlated. For example, as shown in Fig. 1, B–P GCLs with SI < 10 mL/2 g can still maintain low hydraulic conductivity (<1.0 × 10−10 m/s) (Scalia et al., 2014, 2018a; Salihoglu et al., 2016; Shen et al., 2016; Tian et al., 2016, 2019; Chen et al., 2019). In contrast, the hydraulic conductivity of B–P GCLs can be greater than 1.0 × 10−10 m/s when SI of B–P exceeds 20 mL/2g. The poor correlation between hydraulic conductivity and SI for B–P GCLs may be an indication that the procedure and specimen preparation for the SI test are not appropriate for B–P GCLs. For example, during sieve analysis of a B–P GCL, Oren et al. (2018) observed some gel-like materials retained on the #200 sieve which were attributed to the polymer additive(s) in the bentonite.
As a result of the aforementioned considerations, the objective of this study is to investigate the influence of crushing and sieving on the swell index (SI) of B–P composites. Tests were conducted on seven B–P specimens prepared according to the ASTM standard method (D5890) and two other alternative methods. SI tests were performed using type II deionized (DI) water per ASTM D1193 and synthetic leachates. Hydraulic conductivity of B–P GCLs were performed with the synthetic municipal solid waste incineration (MSW-I) ash leachates. The swell indices of specimens prepared using the different methods were compared and correlated with measured hydraulic conductivity of GCLs manufactured using the B–P composites. Loss on ignition (LOI) tests were also conducted to quantify polymer content of the prepared specimens, in support of the explanations for the measured swell indices and observed relationship between SI and hydraulic conductivity.
Section snippets
Background
Various types of polymers have been used as amendments for bentonites proposed for containment of aggressive leachates (Di Emidio, 2010; Scalia et al., 2014, 2018a; Guler et al., 2018; Ozhan, 2018; Prongmanee et al., 2018; Salemi et al., 2018; Pandey et al., 2019; Prongmanee and Chai, 2019; Yu et al., 2019; Chai and Prongmanee, 2020). Polymers generally can be grouped into two main categories, depending on the basic structure of the polymer: linear polymers and crosslinked polymers. Linear
Bentonite-polymer composites
Seven B–P composites designated herein as B–P1, B–P2, B–P3, B–P4, B–P5, B–P6, and B–P7 were investigated in this study. All the B–P composites were supplied by the same manufacturer and were produced by dry mixing of granular Na–B with proprietary polymer(s). Though the specific polymer(s) used in the B–P composites was(were) not disclosed, the manufacturer specified that the bentonites used in B–P1, B–P2, B–P3, and B–P4 were blended with different linear polymer(s) (i.e., water-soluble
ASTM standard method
Results of the SI and LOI tests conducted on Na–B and B–P specimens prepared following the ASTM standard method (ASTM-P and ASTM-R) are summarized in Table 4. SI results given in Table 4 were performed using DI water. Images of the SI tests performed on the ASTM-P and ASTM-R specimens for B–P7 are shown in Fig. 3. The ASTM-P specimen (Fig. 3a), when hydrated, formed a grayish brown gel (see Fig. 3c), whereas the ASTM-R specimen (Fig. 3b) formed a clear gel (see Fig. 3d). To compare the SI
Summary and conclusions
This study was conducted to investigate how specimen preparation (crushing and sieving) affects the swell index (SI) of B–P composites and the relationship between SI and hydraulic conductivity of B–P GCLs. SI tests were conducted on seven B–P composites and the base sodium bentonite (Na–B) obtained from one GCL manufacturer using DI water and synthetic MSW-I ash leachate. The B–P composites contained granular Na–B dry blended with proprietary polymer. Based on information provided by the
Acknowledgments
This material is based upon work supported by the Environmental Research & Education Foundation (EREF), the Hinkley Center for Solid and Hazardous Waste Management (HCSHWM), and the Colloid Environmental Technologies Co. (CETCO). Any opinions, findings, and conclusions in this material are those of the authors and do not necessarily reflect the views of EREF, HCSHWM, and CETCO.
References (68)
- et al.
Barrier properties of a geosynthetic clay liner using polymerized sodium bentonite
Geotext. Geomembranes
(2020) - et al.
Hydraulic conductivity of geosynthetic clay liners with sodium bentonite to coal combustion product leachates
J. Geotech. Geoenviron. Eng.
(2018) - et al.
Hydraulic performance of anionic polymer-treated bentonite-granular soil mixtures
Appl. Clay Sci.
(2018) - et al.
Performance-based indicators for controlling geosynthetic clay liners in landfill applications
Geotext. Geomembranes
(2009) - et al.
Hydraulic conductivity of nonprehydrated geosynthetic clay liners permeated with inorganic solutions and waste leachates
Soils Found.
(2007) - et al.
Long-term barrier performance of modified bentonite materials against sodium and calcium permeant solutions
Geotext. Geomembranes
(2008) - et al.
Long term chemo-hydro-mechanical behavior of compacted soil bentonite polymer complex submitted to synthetic leachate
Waste Manag.
(2016) Geosynthetic clay liners: perceptions and misconceptions
Geotext. Geomembranes
(2020)- et al.
Self healing capacity of geosynthetic clay liners and influencing factors
Geotext. Geomembranes
(2013) - et al.
Comparison of prehydration and polymer adding effects on Na activated Ca-bentonite by free swell index test
Appl. Clay Sci.
(2017)