Effect of biopolymers on permeability of sand-bentonite mixtures

M.S. Biju, D.N. Arnepalli

Department of Civil Engineering, Indian Institute of Technology Madras, Chennai, 600036, India

Keywords:Geotechnical centrifuge Landfill liner Synthetic leachate Xanthan gum Guar gum

ABSTRACT Application of biopolymer-modified geomaterials in waste disposal practices is gaining wide acceptance due to their superior tensile characteristics and improved crack resistance. Permeability is an important design parameter which determines the suitability of a material as a liner for construction of engineered landfills. Given this, the permeability characteristics of sand-bentonite mixtures amended with biopolymers was studied using a modified-falling head permeability apparatus under an accelerated gravity environment. Both distilled water and synthetic leachate were utilized as permeant liquid to assess the role of biopolymer amendment on the permeation behavior of sand-bentonite mixtures. Experimental results indicate that addition of biopolymers causes aggregation of the clay platelets, which in turn enhances the permeation behavior of the biopolymer-modified sand-bentonite mixtures.These mixtures meet the regulatory requirement of the liner.

1. Introduction

The effectiveness of sand-bentonite mixtures(SBMs)as barriers in the construction of waste disposal facilities is a proven fact(Sobti and Singh,2017).Permeability is the critical property that determines the suitability of SBM as a liner.The presence of cracks in a barrier having sufficiently low permeability may fail to perform its intended purpose. A cover liner of an engineered landfill is subjected to cracking due to differential settlements,desiccation,and distress in the slope.For this, various remedial methods have been proposed to mitigate crack initiation and propagation in compacted clay barriers. Researchers have investigated the effects of various soil amendment techniques involving cement and fly ash, and inclusion of materials like polypropylene, glass, softwood pulp and polyester fibers to the soil matrix, for enhancing the liner performance (e.g. Adaska,1985;Maher and Ho,1994;Chaduvula et al.,2017;Thyagaraj and Soujanya,2017).Considering the adverse environmental impacts of traditional additives,the feasibility of biopolymer application for improving the engineering performance of the geomaterials is also reported (e.g.Biju and Arnepalli, 2019). Favorable modifications of strength, dispersivity, erosion resistance, permeability, and swell-shrink of soils,upon amendment of these materials by biopolymers,have been reported in the literature (Khachatoorian et al., 2003; Bouazza et al.,2009; Chang et al.,2016;Acharya et al., 2017; Ayeldeen et al.,2017;Cabalar et al., 2017; Latifi et al., 2017; Dehghan et al., 2018;Ikeagwuani and Nwonu, 2019). In general, stabilization of the geomaterials using a variety of additives has resulted in improved crack resistance and tensile strength behaviors of geomaterials; however,these additives have posed a detrimental effect in terms of increased permeability(Maher and Ho,1994;Abdi et al.,2008;Mukherjee and Mishra, 2017; Thyagaraj and Soujanya, 2017). Though many studies have been made on the application of biopolymers for stabilization of geomaterials,assessing their influences on the permeability behavior of SBMs is rarely reported.

The majority of the regulatory authorities suggest that the coefficient of permeability of a landfill liner shall be less than 1×10-9m/s.Permeability measurement of such low permeable materials using a flexible wall permeameter requires a long testing duration(Anderson et al., 2015). Further, the presence of microorganisms in permeant liquid and porous media may induce clogging, which in turn yields erroneous results when testing periods are long.On the contrary,the coefficient of permeability of SBMs obtained indirectly from consolidation data yields low values compared to that obtained as per ASTM D5084-16a (2016) using a flexible wall permeameter (Gueddouda et al., 2016). In this regard, researchers have conducted falling-head permeability test in an accelerated gravity environment using a geotechnical centrifuge to reduce the test duration(Singh and Gupta,2000;Singh and Kuriyan,2002;Taylor,2005;Timms et al.,2016).Over the recent years, few studies have adopted complex data acquisition systems to record pressure variation throughout the test for enhancing the accuracy in permeability measurement (Tonder and Jacobsz,2017).

On the contrary, the conventional falling-head test setup involving a drop in water level at the inlet boundary, and by maintaining a constant water level at the outlet boundary, yields result with moderate accuracy (Singh and Gupta, 2000). However, these methods either suffer from inaccuracies due to the evaporation loss and/or demand expensive data acquisition systems. Further, quantification of variation in the head difference due to evaporation is impractical.The error associated with these scenarios can result in a significant variation in the measured coefficient of permeability.For instance, if the permeability test on a soil sample having 50 mm in diameter and 10 mm in thickness in a conventional falling-head setup with inlet tube diameter of 6 mm provides a head drop from 180 mm to 130 mm in 3 h,the coefficient of permeability would be 4.34×10-9m/s.If 2 mL of water was evaporated from the inlet tube during the test,which is possible during the permeability test in an accelerated gravity environment, instead of the final inlet head of 130 mm, the observed head would be 78 mm resulting in a coefficient of permeability of 1.12×10-8m/s.Given this,the development of a permeability setup that can overcome this lacuna is mandatory.In addition, when a geomaterial under consideration is to be evaluated for its suitability as a landfill liner, involving leachate as permeant liquid is inevitable. Considering the practical difficulty,researchers have resorted to simulated or synthetic leachate instead of actual leachate collected from an engineered landfill(Rowe et al.,2002; Badv and Omidi, 2007; Bradshaw and Benson, 2014; Mishra et al.,2015;Leme and Miguel,2018).

The present study aimed to investigate the permeability behavior of biopolymer-modified SBMs from the perspective of landfill liner applications.A simple permeability apparatus that can be used in the geotechnical centrifuge, without any complex data acquisition system, was developed. Also, the permeability of the chosen biopolymer-modified SBMs was assessed as per ASTM D5084-16a(2016), using the flexible wall permeameter, to validate the results obtained from the proposed permeability apparatus.Further,the role of confinement in terms of changing volume of the sample by allowing and restraining its void ratio was assessed.

2. Materials and experimental investigations

Fig.1. X-ray diffractogram of (a) fine-sand and (b) bentonite.

The specimens used for this study were prepared by mixing guar gum and xanthan gum at different dry weight fractions with SBMs. Sodium bentonite of high purity supplied by the Versatile Specialty Mine Chem,Bhuj,India,was used for preparing various SBMs having different bentonite fractions in them. The X-ray diffractogram and other basic characteristics of selected bentonite reported in Fig. 1a and Table 1 confirm that the dominant mineral is montmorillonite.The chemical composition of the bentonite was assessed by employing X-ray fluorescence spectrometer,and the elemental composition in its oxide form is listed in Table 2.

Srikanth(2017)studied the effect of the grain size distribution of sand in SBM on its overall engineering performance and reported that fine-sand provides smaller coefficient of permeability compared to coarse-sand. Given this observation, Indian standard fine-sand(grade-III)was chosen for this study,and its physical and mineralogical characteristics are shown in Fig.1b and Table 1.

Table 1Physical properties of materials used in the study.

Table 2Chemical composition (%) of bentonite.

The fine-sand was mixed thoroughly with bentonite in a dry condition at various proportions by weight (25%, 45%, 65% and 85%). The geotechnical characteristics of the SBMs were determined as per relevant standards (ASTM D698-12, 2012; ASTM D4318-17, 2017), and the obtained results are reported in Table 3.

Anionic biopolymer xanthan gum and non-ionic biopolymer guar gum supplied by Sisco Research Laboratories, Bombay, India, were used as the soil stabilizers. As per the World Health Organization(WHO),the selected biopolymers are considered to be nontoxic,and hence, no dosage limits were prescribed for ingestion (WHO,1975,1987). Guar gum is derived from the seeds of Cyamopsis tetragonoloba,and the hydroxyl groups in its polymer chains help in forming hydrogen bonds.Xanthan gum,a branched polysaccharide produced by Xanthomonas bacteria, is commercially available since 1960.Xanthan gum has cellobiose as the monomer and the polymer side chains consisting of a trisaccharide were composed of D-mannose(β-1,4), D-glucuronic acid (β-1,2) and D-mannose, which is attached to alternate glucose residues in the backbone by α-1,3 linkages (Petri,2015). Xanthan gum is a highly stable polysaccharide that is not easily degraded by the microorganisms presented in the soil(Cadmus et al.,1982).The viscosities of 2%guar gum and 3%xanthan gum solutions are more than 700 Pa s, and hence wet mixing of biopolymer beyond these concentrations with the soil becomes tedious(Chudzikowski,1971;Chen et al.,2013).As the high viscosity brings down the workability, the maximum quantities of guar gum and xanthan gum to be added were restricted to 2%and 3%by weight of soil,respectively.

Biopolymers can be mixed with geomaterials using either wetmixing or dry-mixing technique. In the wet-mixing procedure,initially, a solution of a biopolymer is prepared by mixing a known quantity of biopolymer with water,and then the solution is uniformly mixed with soil. In dry-mixing, soil and powdered biopolymer are mixed thoroughly in dry condition,and subsequently,water is added to it(Chang et al.,2015).When a biopolymer is added to water in the wet-mixing approach, the configuration of the biopolymer changes from coiled to an extended form (Theng, 2012). On the contrary,during dry-mixing, the coiled biopolymers in the constrained void spaces cannot access as many soil-solid surfaces as in the wet-mixing even after addition of water.Owing to the advantages of wet-mixing,the biopolymers in different fractions(0.5%,1%,1.5%,2%and 0.5%,1%,2%,3%for guar gum and xanthan gum,respectively)and SBMs werewet-mixed, to assess the engineering properties of the biopolymermodified SBMs.

Table 3Geotechnical characteristics of sand-bentonite mixtures.

Based on the results of the standard Proctor compaction test,the weight of SBM and quantity of water required for preparing specimens having a known compaction state were calculated.A known quantity of biopolymer was added to the predetermined volume of deionized water and mixed thoroughly in an orbital shaker for 48 h,to ensure the proper dispersion of biopolymer in the suspension.The suspension thus prepared was mixed with SBM and allowed it to mature by storing it in an air-tight polyethylene bag.

Deionized water and strong synthetic leachate were selected for conducting the permeability experiments. Bradshaw and Benson(2014) analyzed data of more than 2000 landfill leachate samples and suggested the composition of real and synthetic leachates.Their analysis was primarily based on the ionic strength and relative abundance of monovalent and multivalent cations (RMD)presented in the leachate.Given this observation,the present study has considered synthetic leachate having the chemical composition as suggested by Bradshaw and Benson (2014), and the details are listed in Table 4.

For performing a flexible wall permeability test, the premixed SBM was placed in a mold and compacted to obtain a cylindrical specimen of 38 mm in diameter and 25 mm in height representing its optimum compaction state. The permeability of the specimen was measured following ASTM D5084-16a (2016). The initial portions of the curves of time versus volumes of the influent and effluent were found to exhibit slightly different slopes due to the progressive saturation of the sample.Hence,the segment in which the curves of time versus influent and effluent volumes have shown identical response is used for determining the coefficient of permeability of SBMs.

Since the flexible wall permeameter demands long testing periods to arrive at the coefficient of permeability,a modified-falling head permeameter setup is fabricated, as shown in Fig. 2, for assessing the permeability of SBMs using a geotechnical centrifuge.The permeability apparatus consists of a 50 mm diameter Perspex tube fixed on a rigid platform,which allows the water to move out through an opening connected to a 10 mm diameter graduated tube. The soil sample is kept at the bottom of the Perspex tube,which,in turn,is sandwiched between porous stones.To avoid the change in volume of the specimen due to swelling, the movement of the upper porous disc is restrained using a spacer rod.Water was filled in the 10 mm diameter graduated tube to impose a known hydraulic head, and the entire setup was subjected to centrifugation, allowing permeant liquid to flow through the soil in an upward direction. This arrangement avoids the preferential flow of permeant liquid through an unnoticeable annular space between the sample and the inner circumference of the Perspex tube as well as through the micro-cracks which might have formed during the sample preparation. The percolate passing through the specimen was allowed to accumulate on its top(i.e.within the large Perspextube of 50 mm in diameter).Before measurement of the coefficient of permeability of the sample,a minimum of three pore volumes of permeant liquid was allowed to pass through it for ensuring its complete saturation.

Table 4Chemical properties of permeant liquid.

Fig. 2. Schematic and pictorial view of proposed falling-head permeameter.

While constructing the compacted clay liners as a part of the cover liner system of an uncontrolled waste dumps or as an engineered intermediate landfill cover,these liners are not subjected to surcharge loads.To simulate these field scenarios,the permeability tests were conducted by allowing the change in volume of the specimen in an upward direction.The permeant liquid was allowed to flow through the sample until the volume change ceases before measuring its permeability.For simulating the 500 mm thick liner in the field, the specimen of 10 mm in thickness was subjected to accelerated gravity of 50 times the acceleration of gravity on the surface of the earth (Taylor, 2005). A similar procedure was followed for simulating the permeation of leachate through the liner by considering the synthetic leachate as the permeant liquid.

Since the hydraulic boundary conditions(i.e.water levels in the proposed setup) at the inlet and outlet of the specimen are continuously varying during the falling-head test, the formulation of a mathematical expression for determining the coefficient of permeability is essential and the derivation of the same is explained, as shown in Fig. 3.

Let the cross-sectional area of graduated inlet tube be a,and the cross-sectional area of specimen(i.e.large Perspex cylinder)be A.In a given test duration dt,the water level was reduced by dh1in the graduated tube and increased by dh2in the large Perspex cylinder,causing a head difference to reduce from h1to h2,as shown in Fig.3.

When the soil sample is completely saturated, the decrease in the volume of water in the smaller tube should be equal to an increase in the volume of water in the larger cylinder as given below:

The head difference caused due to flow in time dt is

Substituting Eq. (1) into Eq. (2) results in the following relationship:

Fig. 3. Variation of the hydraulic head during the falling head permeability test. dh1 and dh2 are the water level variations in graduated inlet and outlet tubes in the permeameter, respectively; and h1 and h2 are the initial and final head differences,respectively.

In general, the governing equation for flow through porous media can be expressed as

where p is the pressure,x is the length of the porous media,k is the intrinsic coefficient of permeability, μ is the dynamic viscosity of permeant liquid,vfis the velocity of flow,ρ is the fluid density,and β is the Forchheimer coefficient(Bear,1972;Hellstr?m and Lundstr?m,2006; Sobieski and Trykozko, 2014). When the seepage velocity is low,and the corresponding Reynolds number,Re,is less than unity,Eq. (4) reduces to Darcy’s law. Incidentally, the seepage velocity through the geomaterials considered in this study is found to be very small, and hence, Darcy law can be used to represent the flow of permeant liquid in these materials. According to Darcy law, the volume of water flowing through the sample in time dt is

where Asis the cross-sectional area of soil specimen,L is the length of soil specimen, and i is the hydraulic gradient.

Substituting Eq. (3) into Eq. (5)yields

If the flow through the soil sample from time t1to t2causes a head difference to fall from h1to h2, it can be represented as

Integrating and rearranging the terms results in the expression for the coefficient of permeability as

In each permeability test, the angular velocity in terms of rotations per minute (RPM) was calculated to obtain imposed gravitational field(i.e.Ng value,where N is a multiplying factor,and g is the acceleration due to gravity),and the permeameters containing soil specimens were centrifuged for a definite period.The weight of permeameter with the sample and permeant liquid, and the fluid level in the inlet and outlet tubes were noted before and after the test. The cross-sectional areas, test duration, and angular velocity were recorded. For each sample, experiments were conducted in duplicate and the average value of the coefficient of permeability was reported.

Evaporation from the inlet and outlet occurs with the permeation during the centrifugation. To quantify the water level drop due to evaporation,a setup having four different diameter tubes is fabricated,as shown in Fig.4 and subjected to centrifugation along with permeameters. The loss of water level in the tubes was determined and it was found that the evaporation is proportional to Dn,where D is the diameter of the tube and n is a constant.Hence,mass loss,wt,from the graduated tube of diameter d in time Δt can be written as follows:

where keis the proportionality constant.

Similarly, mass loss, wc, due to evaporation from the large Perspex tube of diameter D can be written as

If the total mass loss which can be directly measured is w, the constant of proportionality can be determined as

Fig. 4. Test setup to study the variation in evaporation rate with tube diameter.

The level drop from tube and containers can be determined as wdn／[πd2(dn+Dn)]and wDn／[πD2(dn+Dn)],respectively.The final head difference h′2after incorporating the evaporation correction becomes

The coefficient of permeability determined using the centrifuge permeability test at Ng condition will be N times that of the permeability at the 1g condition.Hence,coefficient of permeability from centrifuge permeability test can be written as

To understand the primary reason behind the change in permeability of biopolymer-modified SBMs with sand and biopolymer contents,the morphological study was conducted with the help of a scanning electron microscope (SEM). The fabric variation of bentonite with the addition of biopolymers was imaged and investigated.

3. Results and discussion

The results of evaporation from the tubes having different diameters are shown in Fig. 5. The analysis of evaporation data provides the n value used in Eqs. (9)-(12) as 1.8, which matches with the value reported in the literature (Hisatake et al.,1993).

Fig. 6 shows the results of the permeability test conducted on SBMs having different sand contents. The results from the flexible wall permeability tests under the constant-head and centrifuge permeability tests in 50g,100g,and 150g conditions under variablehead corroborate each other, demonstrating the efficacy of the proposed test methodology.The trend shows a sudden variation in the permeability only when the sand content exceeds 65% in the SBMs.Hence,the value 65%can be considered as a threshold,which causes a variation in the soil structure. The coefficient of permeability of pure bentonite is moderately less than that of all SBMs considered in this study.Based on these results,pure bentonite and SBM with 65%sand content(SBM65)were considered for assessing the influence of biopolymer amendment on their permeability characteristics.

Fig. 5. Variation in evaporation rate with the diameter of tubes.

Fig. 6. Variations in the coefficient of permeability of sand-bentonite mixtures with sand content.

Fig.7 shows the variations of coefficient of permeability of SBM65and pure bentonite with the biopolymer content. The difference in permeability when volume change is restricted (confined) and allowed(unconfined)is also depicted.Fig.8 shows the variations of the coefficient of permeability when the permeant liquid is synthetic leachate.It can be observed from Figs.7 and 8 that the coefficient of permeability of geomaterials considered in this study is greater when tested by allowing the volume change of the sample. For comparison, both flexible wall and confined falling-head tests are required to perform; however, the unconfined falling-head tests represent the field scenario of cover liners, and hence results from these tests shall be used in the design of liners. An increase in biopolymer content causes an increased permeability, which is contradicting with the permeation behavior of biopolymer-modified noncohesive and nonexpansive soils reported in the literature. Results from the tests using synthetic leachate also indicate higher permeability values. Though the coefficient of permeability of biopolymer-modified SBMs is higher than that of untreated SBMs,the coefficients of permeability of modified-SBMs are less than threshold of 1 ×10-9m/s,recommended by various regulatory authorities (CJJ113, 2007; CPHEEO, 2016). Hence, biopolymer modification can be recognized as an effective stabilization strategy,considering its superior performance in terms of crack resistance and strength,as highlighted in the literature(Biju and Arnepalli,2019).

3.1. Variation of permeability with sand content

It can be observed from Fig. 7 that the permeability of SBM increases with an increase in sand content.In addition to the formation of preferable paths through the sand-bentonite interface,the change in the fabric also plays a crucial role in the permeability increase.The geomaterial initially in cohesive structure moves towards the coarsegrained skeletal structure with an increase in sand content.Previous studies have concluded that geomaterial in cohesive structure performs better as hydraulic barriers (Kenney et al., 1992; Proia et al.,2016). On the contrary, for the geomaterials to be in coarse-grained structure, the sand skeletal void volume, esv, should be higher than the volume occupied by bentonite.With an increase in sand content,esvincreases and bentonite volume decreases.When the sand content reaches 85%,the fabric changes from fine-grained to coarse-grained skeletal structure, providing easier paths for permeant liquid to move through it. This results in a considerable increase in the coefficient of permeability. Though the volume of hydrated bentonite volume is sufficient to fill all the voids corresponding to esv,nonuniform distribution of bentonite and sand may create preferential paths,as shown in Fig.9.The SBM having a sand content up to 65% yields clay matrix structure even when the mixtures are airdried. This is due to the fact that the volume of bentonite at its shrinkage limit is sufficient to fill the sand skeletal void volume.Consequently, the increase in sand content beyond this threshold leads to an increase in coefficient of permeability,as depicted in Fig.6.

3.2. Variation of permeability with biopolymer content

Figs.7 and 8 show an increase in permeability with addition of biopolymers. The flow of permeant liquid through a compacted specimen will be through inter-and intra-aggregate flow paths,as shown in Fig. 10, and the inter-aggregate flow paths will be contributing much to the coefficient of permeability (Olsen,1962;Scholes et al., 2007; Bayesteh and Mirghasemi, 2015). Water molecules firmly attached around the clay particles inside aggregate will create negligibly small effective flow paths allowing flow only through the inter-aggregate paths.Thus the size and formation of aggregates or packets in clays considerably influence permeability (Benson and Daniel,1990).

Fig.7. Variations in the coefficient of permeability of bentonite and sand-bentonite mixture determined using deionized water as permeant liquid,amended with(a)xanthan gum and (b) guar gum.

Fig. 8. Variations in the coefficient of permeability of bentonite and sand-bentonite mixture determined using synthetic leachate as permeant liquid, amended with (a) xanthan gum and (b) guar gum.

Fig. 9. Schematic indicating the formation of preferential flow paths (a) from uniform distribution of sand and bentonite and (b) from nonuniform distribution of sand and bentonite.

Fig.10. Flow paths in compacted clay specimen through (a) inter-aggregate pores and (b) intra-aggregate pores.

Fig.11. Scanning electron microscope images of biopolymer-modified bentonite.

The addition of biopolymers causes aggregation of clay particles by binding them together, forming higher sized aggregates(Nugent,2011). When a biopolymer is added to the soil, it forms a thin coating over the aggregates through secondary bonds. The polar components in the biopolymer may participate in the interaction with the charged surfaces of the soil, which in turn enhances the adhesion. As a result, the ionic, hydrogen or van der Waals bonds lead to enwrapment of the soil aggregates and formation of larger aggregates. The SEM images at 68 times of magnification shown in Fig.11 helps in explaining how the aggregate size changes with an increase in biopolymer content. Smaller aggregate size and pore size can be observed in the SEM image of untreated bentonite.In the next image with 0.5%biopolymer,these aggregates combine together to form a much larger one causing larger voids. The aggregate size and void size progress with increases in both guar gum and xanthan gum addition,as indicated in Fig. 11 using double-headed arrows. The increased aggregation leads to larger inter-aggregate flow paths resulting in higher permeability.

The observed increase in permeability with addition of biopolymers is well within one order of magnitude, as presented in Figs.7 and 8.If the sole reason for the change in permeability is the change in aggregate size, the variation should have been higher than the observed value.Liu et al.(2017)described the mechanisms of soil stabilization by polymers as a collective process of filling the void volume, enwrapping, and physicochemical interaction. The hydrogel formed with the biopolymers existing in the pore spaces will cause a reduction in the flow.The interactions between the soil and biopolymer are predominantly physical in nature. Majority of the physisorption interactions exhibit reversibility; however, the probability of desorption of the polymer from the soil is restrained by the octopus effect of the biopolymer (Theng, 2012). Even after the flow of many pore volumes of water or leachate through the soil, the octopus effect holds the biopolymers in position. Thus a combined effect of the widening of the flow paths,filling a portion of the voids and the octopus effect controls the coefficient of permeability.

3.3. Role of confinement

The flow through the selected material under confined conditions is less than that under unconfinements.When the material is used as an intermediate cover liner, no significant surcharge load acts on it to arrest the volume change.This scenario was simulated by performing the tests under unconfined conditions. The permeability test under an accelerated gravity environment conducted to simulate a field liner of 500 mm thick provided higher permeability,as seen in Figs.7 and 8.As the flow occurs through the interaggregate voids,it tries to push away the aggregates,creating larger flow channels; the flow will be through these enlarged channels providing higher permeability.However,in the field,the possibility of such a high potential head,which can cause the creation of easier paths through pushing away the particles, is rare. Nevertheless,being on the conservative side, it is advisable to use the permeability values obtained from the unrestrained test.

3.4. Role of leachate

The lower coefficient of permeability of SBM obtained while tested with deionized water is mainly attributed to the reduction in the size of the effective flow channels, which in turn is influenced by the diffused double-layer of montmorillonite mineral presented in it (Mesri and Olson,1971). The increase in cation concentration(leachate as a permeant liquid) results in contraction of diffuseddouble layer and causes broader flow paths, which in turn alters the coefficient of permeability.For instance,the k value of SBM65is changed from 10-11m/s to 10-10m/s,when the permeant liquid is synthetic leachate instead of deionized water. Since the concentration of cations in synthetic leachate is significant, the permeability increases when tested under both confined and unconfined conditions.As the flow of leachate through an intermediate liner is evident,the permeability values obtained using SL are advisable for design of such liners.

4. Conclusions

This study analyzed the effect of the biopolymer amendment on permeability characteristics of SBMs.To overcome the limitation of the longer duration of test associated with flexible wall permeameter,a modified-falling head permeameter apparatus,which considers the evaporation aspect, is presented. The proposed apparatus was used to assess the variation in permeability of biopolymer-modified SBMs by allowing and restraining volume change of the specimen,under an accelerated gravity environment.The permeability of a compacted material subjected to volume change is higher compared to that of restrained condition. The variation in permeability when the permeant liquid changed from deionized water to leachate was also presented.The results inferred that SBMs with sand content up to 65%can be used as liner material for the construction of intermediate compacted clay liners.

Further,it was established that the biopolymer amendment had induced aggregation of clay platelets, which in turn created wider effective flow paths.The change of permeant liquid from deionized water to leachate resulted in the contraction of the diffused doublelayer on the clay surface. Flow-through the soil in an unrestrained condition allowed the aggregates to move apart by enabling the permeant liquid to pass through the increased pore spaces. The synergetic influence of the factors mentioned above is found to be responsible for an increase in the permeability of biopolymermodified SBMs. However, even in the most critical condition,where the flow of synthetic leachate through SBM with 65% sand content under an unrestrained condition, the coefficient of permeability of the modified-SBM was found to be less than the minimum threshold advocated by regulatory guidelines (i.e.≤1×10-11m/s).Based on this observation,it can be concluded that the proposed environmental-friendly stabilization technique can be accepted to improve the overall engineering properties of geomaterials.

Declaration of competing interest

The authors wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome.