Modification of bentonite with black cotton soil and carboxyl methyl cellulose for the enhancement of hydraulic performance of geosynthetic clay liners

The expansive soils are widely present in all continents in different forms of soft clay minerals, such as smectite, bentonite, kaolinite, montmorillonite, beidellite, vermiculite, attapulgite, nontronite, and chlorite. In India, the black cotton soil (BCS) or vertisol (a type of montmorillonite) is available in plenty which exhibits high swell-shrink properties when exposed to moisture conditions (Oza & Gundaliya 2013; Ajit et al. 2022; Woldesenbet 2023). The versatility of expansive soils poses critical challenges to the infrastructural development activities happening on the ground and underground soils. Due to their inability to withstand the changes in environmental conditions such as an external pressure or infiltrating moisture, they tend to exhibit significant changes in their volume (swelling and shrinking) thereby causing the supporting infrastructure to lose strength and ultimately fail. Apart from the structural stability concerns, one key aspect of concern limiting the utility of these soft soils in the engineering context is in the form of geosynthetic linings which are provided for limiting the leachate from a waste containment facility constructed on the ground or underground locations (Scalia & Benson 2010). The failures of the liner system owing to the panel separation due to shrinkage, punctures, wrinkles, installation errors, liner design errors, etc. cause critical challenges both to the manufacturers as well as the operators (Rowe 2014; Rowe et al. 2017). Due to such conspicuous nature of the very physical structure, they require special attention to improve their physical and chemical behaviour when subjected to critical environmental conditions. Interestingly, this is not a new problem and the geotechnical engineers have attempted to address it only superficially, or at least tactfully for the sake of prevailing load and stability conditions.

Recent developments in this field have contributed a variety of innovative clay liners, geomembranes, and geosynthetic composites which are commercially available in various forms and textures (Garcia et al. 2007; Cheah et al. 2016; Jayalath et al. 2018; Gallage & Jayalath 2019; Weerasinghe et al. 2019, 2020). The composite geosynthetics are commonly available as geosynthetic clay liners (GCL), which consist of a layer of geotextiles or adhesive attached to a geomembrane and a core of low-permeability sodium bentonite clay with a woven and non-woven structure. The GCLs made of soft clays are found to offer resistance against the seeping moisture by virtue of internal expansion within the pore spaces resulting in a reduction in permeability (Pitanga et al. 2011; Viswanadham et al. 2012; Al-Taie & Pusch 2014; Rajesh & Viswanadham 2015). The bentonite clay used for the purpose of landfill lining has to have critical hydraulic properties to arrest the seepage within the landfill, in addition to their swell shrink and thermal features. Compared to calcium bentonite clay, sodium bentonite needs only minimum thickness (5–10 mm) and is less permeable (∼10⁻¹¹ m/s) (Acikel et al. 2018). The geomembranes made of high-density polyethylene or reinforced polypropylene are found to exhibit much lesser permeability (∼10⁻¹⁵ m/s) compared to the clay liners and are commonly available as flexible sheets for direct field implementation (Weerasinghe et al. 2021).

In addition to the structural support for the confining layers, the GCLs help control seepage and release of potentially harmful materials from the tailing ponds and mining containment areas into the environment (Xiong et al. 2009; Egloffstein et al. 2012; Mazzieri & Di Emidio 2015). However, the leachate characteristics from the landfills can significantly affect the hydraulic conductivity of various soils depending on their structure and qualities (Han et al. 2009; Xu et al. 2011; Ozcoban et al. 2013; Aldaeef & Rayhani 2014; Bradshaw & Benson 2014). The leachate can also impact the physico-chemical structure of the clay minerals present in the GCL causing changes in the hydraulic conductivity (Ozcoban et al. 2022). According to Han et al. (2009), the hydraulic conductivity of GCL for distilled water decreased from 3.5 × 10⁻⁷ to 5.35 × 10⁻⁹ cm/s. Ozcoban et al. (2013) reported that the hydraulic conductivity of the compressed clay dropped from 5.2 × 10⁻⁸ to 6.45 × 10⁻⁸ m/s when leachate water was used compared to distilled water. In a similar study, Ozcoban & Acarer (2022) supplemented three different nanomaterials, namely, iron oxide, aluminium oxide, and Oltu clay to kaolin at two different rates (1 and 5%) to study the permeability of the clay. The removal rates of nanomaterial-added kaolin samples were found to be higher than that of kaolin without nanomaterials.

It is understood that the hydration and self-healing characteristics of the GCL mainly depend on the initial placement condition and the ability to absorb moisture from the adjacent soil (Acikel et al. 2018). During the operations, the GCLs may or may not be sufficiently saturated to achieve the designed hydraulic conductivity of the cap barrier depending on the temperature and moisture conditions of the adjacent soil medium. This indicates that the water retention behaviour and unsaturated hydraulic conductivity of GCLs may need to be established for the proper understanding of hydraulic flow through GCLs (Rowe 2014; Rajesh & Khan 2018; Ghavam-Nasiri et al. 2019). In other words, it is imperative to ascertain whether the hydraulic performance of GCL can be enhanced by maintaining its optimum capacity throughout the barrier application (Daniel et al. 1997; Rowe et al. 1997; Xiong et al. 2009; Egloffstein et al. 2012; Mazzieri & Di Emidio 2015). The major factors affecting the hydraulic conductivity of the GCL are found to be overburden confining stress, shrinkage effect from temperature variation, chemical compatibility of bentonite to limit different permeants, wrinkle effect from the overlying geomembrane, and side slope shear effects (Fox et al. 1998; Weerasinghe et al. 2019). As evident from the recent research on the design of geosynthetics, the hydraulic performance of GCL has increased from about 1 × 10⁻⁷ to 1 × 10⁻¹² m/s by enhancing the structural and textural characteristics of the core clay layer (Petrov et al. 1997; Kendall & Buckley 2014; Weerasinghe et al. 2020). As stated before, even a marginal reduction in the hydraulic conductivity is considered an improvement in the hydraulic performance of the liner as it reduces the opportunity for dissolved chemicals to pass through the hydraulic barrier.

Based on the literature, it is inferred that the soft soils can be effectively modified to improve their engineering properties so as to make them less-problematic, and, in a larger sense, more-adaptive for the challenging geotechnical applications (Benson et al. 2005). The ionic nature of the hydrating (permeating) fluids is found to be influential in achieving the self-healing capacity and reduction in hydraulic conductivity in order to ensure the service life of the liners (Bouazza et al. 2006). For this reason, chemical stabilization has been practiced as an effective strategy for improving the engineering properties of expansive soils for the application of soil stabilization. However, the conventional approaches in employing cement and lime are becoming more objectionable based on environmental concerns. The use of geopolymers is found to be promising in this aspect as they utilize alkali solutions for making cementitious binders and they encourage the addition of industrial wastes rich in aluminosilicate compounds. Chemical polymers such as propylene carbonate (PC), sodium carbonate (SC), carboxymethyl cellulose (CMC), benzyl tri-ethyl-ammonium bromide (BTEAB), and hexa-decyl-tri-methyl-ammonium chloride (HDTMAC) are preferred as they can target on improving the performance of the bentonite core without affecting adjacent layers (Malusis & Di Emidio 2014; Scalia et al. 2014; Fehervari et al. 2016; Mazzieri et al. 2017). It is observed that the absorption with cationic polymers is very quick but irreversible and the hydraulic conductivity of the clay liner could not significantly improve as a result (Ashmawy et al. 2002; McRory & Ashmawy 2005). The PC-treated bentonite possesses strong chemical resistance but is found to be less effective with insufficient swelling when the permeating fluid is an active electrolyte leading to substantially increased hydraulic conductivity (Onikata et al. 1996; Mazzieri & Pasqualini 2008).

Notably, bentonite has demonstrated superior hydraulic performance as GCL material in a wide range of extreme solutions when treated with an anionic polymer due to the presence of carboxylic acids (COOH) or sulphonic acids (SO₃H) (De Camillis et al. 2017). The modification of the bentonite with sodium-based CMC revealed that the wet-dry cycle has a major impact on the self-healing and hydraulic characteristics of the GCL (De Camillis et al. 2017). In a similar study, Qiu & Yu (2007) reported that CMC-added natural bentonite has better water potential for both the wetting and drying paths, but the efficacy of the treatment is remarkably dose-dependent, hence requiring further optimization. From a recent study, Rajesh & Jain (2022) observed that the self-healing property of the desiccated-rehydrated specimens of the CMC-amended bentonite-based GCLs was enhanced under varying apparent degrees of saturation and suction pressure, indicating the influence of airflow characteristics on the GCL performance. Based on the literature reports on the experimental investigations on the performance of GCL at the laboratory scale, it is inferred that controlling the hydraulic properties of the clay liners without compromising the mechanical strength is very challenging. Though several modifications to the basic liner systems are attempted, the application of CMC-based GCL with typical Indian soil such as BCS and bentonite clay is not sufficiently attempted. Based on these aspects, the present study is formulated to analyse the hydraulic performance of bentonite-based GCL which is blended with CMC at different percentages of locally available BCS.

The raw materials for the preparation of the proposed GCL composite were identified and procured from the locality of the study area which is located in Udumalpet, Tamil Nadu, India (10.61943 N, 77.15159 E), where the soft clay deposits are abundantly present. The black cotton soils (BCS) present in the topsoil horizon were scraped with the help of a hand rake and collected in polyethylene bags before taking to the laboratory for preliminary analysis. The bentonite clay mineral (BCM) used in this study was obtained from M/s Swell Well Minechem Pvt Ltd, India, which is having 85% of montmorillonite mineral. The properties of the collected soft soil samples as per the standard testing procedures followed in the laboratory are listed in Table 1.

The identification and optimization of suitable ingredients are very crucial for preparing composite liners as the diversity in their individual and combinational applications makes it challenging to design them for a specific application. The CMC polymer is used in this study to improve the swelling characteristics of GCL in order to control the hydraulic properties. It is derived from cellulose, which is a natural polysaccharide found in plant cell walls and is produced by chemically modifying cellulose through the introduction of carboxymethyl groups (–CH₂–COOH) onto the cellulose backbone. This modification enhances the water-solubility and functional properties of the cellulose. We have considered an optimal dosage of 8% of CMC for the amended bentonite based on the limitations on plasticity index and swell pressure as reported in the literature (Jain 2017).

The laboratory-scale experiments for the GCL preparation involve a sequence of steps starting with the individual components and then mechanically joining them with sufficient endurance as well as flexibility. For the preparation of modified GCL specimens, a non-woven type of geotextile is selected for its low mass per unit area characteristics. The polypropylene non-woven geotextile having a thickness of 1.5 mm with a mass per unit area of 100 g/m² and a tensile strength of 8 kN/m is used as the cover for the geotextiles in the preparation of the modified GCL specimens.

The prepared GCL specimens were tested for their physical, microstructural, and hydraulic characteristics in order to assess the improvements in the swell-shrink properties due to the composite effects. The Swell Index test was conducted in accordance with ASTM D5890 (2019) on the raw bentonite, BCS, and BCM-BCS composite having 8% CMC (ASTM D5890). For this, the powder samples were taken after passing through the No. 200 sieve and oven drying at 105 °C for 24 h. The soil particles were added incrementally at the rate of 0.1 g until an accurate weight of 2 g and were added to a 100 mL graduated cylinder. The remaining portion of the cylinder was filled with deionized water and kept idle for 24 h for undisturbed swelling to occur. The free swell index value of each soil sample was determined using the procedure reported by Yang & Reddy (2018). The same procedure can be extended to test the specimens having polymers because of their lower density than the raw BCM samples (Billmeyer 2012).

As mentioned earlier, to test the effect of the permeating water quality on the hydraulic performance of the GCL, the pH of the solution was regularly monitored as per the standard procedure (Wilkinson et al. 2016). To measure the pH, an aqueous extract of the GCL specimens was prepared by adding 30 mg of the powder in 1 L of water, dispersing for 30 min at 200 rpm in a flocculator, followed by separating the top 10 mL from the mixture for the pH measurement using a table-top electronic pH meter (M/s U-Tech Pvt Ltd, India). In order to assess the microstructural features of the synthesized GCL specimens, non-destructive characterization tests such as X-ray diffraction analysis (XRD) and scanning electron microscopy (SEM) were performed. The sample preparation and administration for the characterization tests were carried out as per the procedure reported by Fan et al. (2017).

It is to be understood that the flow cell was uniformly filled up to saturation and sufficient care was exercised to avoid the possibility of air entrainment throughout the flow paths during the testing. The flow was initiated by using a mechanical compressor and the GCL kept in the flow cell was subjected to a uniform pressure which allows the flexible membrane to adjust its plasticity effect. The application of pressure was regulated through a valve-controlled mechanism based on the readings on the permeameter columns. The effluent was collected from the flow cell by virtue of the porous disk kept at the bottom and withdrawn with a separate flow line for further chemical analysis.

The presence of CMC is proclaimed to have benefited the characteristic swelling behaviour of the prepared GCL composite in at least two ways: firstly, it enabled a uniform and effective binding environment for the two types of clay minerals, and secondly, it demonstrated the direct impact of water absorption by the composite material. When the soil was fully saturated, the CMC was partially separated from the BCM composite specimens indicating that bentonite plays a major role in the swell behaviour. This is similar to that observation by Tian & Benson 2019, where the authors concluded that the swelling behaviour of the polymer-modified soil is controlled by the presence of montmorillonite-type minerals present in bentonite clay. It is also noted that the swell index started declining for higher addition of BCS (above 25%) which can be attributed to the fact that a substantial reduction in the available BCM content has limited the availability of pore space for efficient hydration to occur. The typical swelling behaviour exhibited by the GCL composite resembles the postulates of a strong parent material influence (PMI) which is generally anticipated when a complex composite is exposed to different environmental conditions (Balaganesh et al. 2022, 2023). Considering these views, the results from the present study can be taken as affirmative propositions towards developing the BCM-blended BCS-based GCL for geotechnical applications.

Further, closer observations at 10k× magnification revealed evidence of a progressive deformation in the crystalline structure of the clay minerals with many flocculated and curly features due to the increased interfacial adhesion between the CMC polymer and the clay phases. The presence of CMC has caused the necessary linkages between the clay particles thus contributing to the toughening mechanisms by absorbing and dissipating energy during physical deformation processes such as swelling. The addition of BCS has effectively contributed towards an increase in the specific surface area, which is evident from the edge-to-edge and face-to-face association of the clay particles. However, a higher amount of BCS (greater than 25%, which is found to be the optimum) has resulted in distorted microstructural features in the images at both magnifications. It is attributed to the fact that the CMC polymers are responsible for achieving an exfoliation effect in the GCL nanocomposites (where individual clay layers are dispersing in the polymer) compared to the intercalation effect (where the clay layers are separated by the polymer chains). It is conclusively evident that an optimum combination of blended clay layers in the polymer matrix is suitable for enhancing the barrier properties of the GCL composite material and preventing their early-stage degradation for effective field applications.

It was observed that BCS20 and BCS25 specimens have the lowest (or near-neutral) pH values corresponding to the fixed amount of CMC addition (8%). One specific reason for the higher range of pH is the presence of typical clay minerals in the composites making it as a net-negative (alkaline) material towards chemical equilibrium. Another aspect is to realize that the anionic polymer concentration is optimum for all the modified bentonite samples considered in the study. As per Tekin et al. (2010), cationic polymers have more adsorption capacity when the pH value is high due to the attraction of positive cations present in the polymer towards the negative charges present in the clay samples. However, since anionic polymer has been used in this study, the negative charges in the clay sample are not being attracted to the positive charges, causing the pH value to decrease. The influence of acids and bases on hydraulic conductivity generally occurs through the breakdown of clays as per Ozcoban et al. (2022). The test results show that the addition of BCS can decrease the adsorption capacity of the CMC-amended GCL composite thereby making it more resilient to the adverse chemical environment present in the soil systems. Wilkinson et al. (2016) and Linggi et al. (2021) also support the finding that soils with low pH values normally require anionic polymeric solutions to neutralize the pH of the soil. A similar result can be obtained when the polymeric composites are used for stabilizing the expansive soils for various geotechnical applications.

In this study, the hydraulic performance of bentonite blended with CMC and different percentages of locally available BCS was analysed. The variation of swell index, pH, and microstructural characteristics of the modified bentonite was studied. The swell index of modified bentonite increases with an increase in the percentage of BCS and attains a maximum for 25% addition of BCS. Further inclusion of BCS reduces the swell index since the bentonite content which plays a major role in the swelling behaviour is reduced. pH value decreases with the addition of BCS and attains a minimum for 20% addition of BCS. The stability of the bentonite decreases due to the presence of anionic CMC polymer solution and thus results in decrement of the pH. The microstructural properties from the SEM images denote that the pore void space decreases with the addition of BCS up to 25%, and thereafter the pore voids increase. The hydraulic conductivity of the modified GCL decreases up to 25% addition of BCS and increases with further addition of BCS. Based on the XRD, the Na content is found to be low in the modified bentonite for the addition of BCS beyond 25%. Consequently, this has increased the hydraulic conductivity of the modified GCL. From this study, it can be concluded that the optimum percentage of BCS and CMC for better hydraulic performance of GCL is 25 and 8%. Although the hydraulic conductivity of the modified GCL is higher compared to the stand-alone GCL with bentonite, nevertheless, this study has far-reaching applications in the disposal of solid wastes in landfills.

Funding received from Dr Mahalingam College of Engineering and Technology, Pollachi, Tamil Nadu, India under the in-house R&D scheme (Ref. No: In-House R&D/2022-23/01) is gratefully acknowledged.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.