Hydraulic Conductivity of Polymer-Amended Sand-Bentonite Backfills Permeated with Lead Nitrate Solutions

Hydraulic conductivity of sand-bentonite (SB) backfills amended with polyanionic cellulose (PAC) to lead nitrate (Pb(NO3)2) solutions was evaluated experimentally in this study. PAC-amended sand-bentonite (PSB) backfills were synthesized by mixing sand-bentonite mixture with 0.3 to 1.2% dry PAC (by total dry mixture mass) and mixed with a certain weight of conventional bentonite (CB) slurry. -e rheology properties including the filtrate loss, viscosity, density, and pH testes of slurry with various bentonite dosages were measured to determine the reasonable CB dosage of slurry. -e slump tests on PSB backfills with various mass slurries were conducted to determine the corresponding water content of backfills with slump 125 ± 5mm. Under the applied pressure 100 kPa, the hydraulic conductivity to Pb(NO3)2 solutions (kc) of PSB backfills with various PAC contents was evaluated based on the modified filter press (MFP) tests, to ascertain the optimum PAC content of PSB backfills when permeated with Pb(NO3)2 solutions. Index properties, including the specific gravity (Gs) and liquid limit (wL) of PSB backfills, were measured after MFP tests. -e MFP tests for PSB backfills were then conducted under various applied pressures to obtain the relationship between void ratio (e) and hydraulic conductivity of backfills. Finally, the flexible-wall permeability test (FWP test) under osmotic pressure 100 kPa was conducted to verify the effectiveness of the MFP test. -e results indicate that slurry with 8% bentonite dosage is the reasonable choice in slurry wall construction. PSB has lower GS and higher wL compared to SB; increasing Pb concentration leads to GS of PSB increased and wL of PSB decreased. PSB with 0.6% PAC content is supposed as the optimum proportion of backfills when permeated with concentrated Pb(NO3)2 solution. PAC adsorbs large amount of bound water, which leads to higher water content (w) and e of PSB backfills, while lead ions (Pb) cause the diffuse double layer (DDL) of bentonite compressed and e of PSB backfills reduced.-e kc of PSB-0.6 remains lower than 10m/s and increases less than 10 times though the Pb concentration was up to 500mM, demonstrating that the hydraulic performance of backfills can be improved effectively in Pb(NO3)2 solution by the additive PAC.-e comparison results between k fromMFP tests and FWP tests show that the MFP test is an effective and easy evaluation of hydraulic conductivity of backfills.

e typical SB cutoff walls are constructed by first excavating a trench with 0.6 to 1.5 m in width while simultaneously filling the trench with bentonite-water slurry (typically 6-12% by dry weight of conventional bentonite) to maintain the trench stability.e sand is mixed with bentonite and then combined with bentonite-water slurry satisfied rheological properties to prepare homogeneous backfills with optimum slump (∼125 ± 25 mm).e backfills is then poured into the slurry trench, forming a barrier with low hydraulic conductivity, k (typically k ≤ 10 −9 m/s).
e potential for chemical incompatibility between backfills and the contaminated groundwater is a crucial consideration for vertical cutoff walls, which results in an increase in k. e k of the SB backfills increases obviously (typically greater than 10 times) when exposed to inorganic solution with ionic strength (SI) greater than 300 mM, resulting in the barriers performance destroyed [8][9][10].
Generally, polymer with lower molecular weight (10 3 -10 4 ) exhibits dispersibility, polymer with larger molecular weight (generally greater than 5 × 10 6 ) shows superior flocculability, and when the molecular weight of polymer is in the range of 10 4 to 5 × 10 6 , polymer is called super absorbent polymer (SAP) because of its excellent water absorbing capacity.As a result, PAC with molecular weight 1,730,000 (commercial products) is selected as the amended agent in this study.Polyanionic cellulose (PAC) is a kind of polymer, nontoxic and tasteless, which is a viscosity modifier and can improve bentonite-suspension viscosity.Additionally, PAC has been extensively used as thickener in petroleum drilling construction and emulsifier constitute in coating brushing [21,22].However, PAC-amended bentonite or sand-bentonite backfills used in hydraulic barrier applications has not been reported.
e objective of this study is to evaluate the potential use of SB backfills amended with polyanionic cellulose (PAC), and for cost effectiveness and low-carbon consideration, the PAC dosage for PAC-amended SB backfills (PSB) is optimized, which may be an alternative for conventional SB backfills to improve the hydraulic performance when exposed to contaminated solutions.In this study, powder PAC is directly mixed with dry bentonite and sand combined with bentonite slurry to prepare the PSB.e hydraulic conductivity to distilled water (k w ) and to Pb(NO 3 ) 2 solutions (k c ) of backfills are conducted by improved membrane filter press (MFP) tests.Index properties, including the specific gravity (G s ) and liquid limit (w L ) of PSB backfills, are measured after MFP tests.e optimum dosage of PAC (by dry mass) of PSB is obtained through the hydraulic performance of PSB with various dosages permeated to Pb(NO 3 ) 2 with concentration from 20 to 500 mM.e filter press tests for PSB backfills are then conducted under various applied pressures to obtain the relationship between void ratio (e) and hydraulic conductivity of backfills.Finally, the flexible-wall permeability test (FWP test) is conducted to verify the effectiveness of MFP test.

Constituent Soils and Polymer.
e backfills comprised powered conventional bentonites (CB), fine sand, and PAC.
e CB manufactured by Mu-Feng Co., Ltd was powdered natural calcium bentonite activated with Na 2 CO 3 treatment in order to obtain more superior swelling capacity and lower permeability.e physicochemical properties and mineralogical compositions of CB are summarized in Table 1.e X-ray diffraction phase identification is performed by China Petroleum and Chemical Corporation (Sinopec Corp.), Jiangsu Oilfield Branch.Based on the X-ray diffraction analysis, the dominant clay mineral of CB is found to be montmorillonite; the mineral constituent of CB is shown in Table 2. e properties of PAC provided by manufactures are shown in Table 3. e fine sand was chosen to model a typical in situ sandy aquifer, which was obtained from the floodplains of Yangzi River in China.e sand was washed by tap water and air dried and then screened with 1 mm sieve.e uniformity coefficient (C u ) is 3.66, and the curvature coefficient (C c ) is 0.76, indicating that the sand exhibits poor grain-size distribution.

Base Mixtures for Backfills.
e base mixtures used to prepare the backfills included a conventional sand-bentonite mixture and PAC-amended sand-bentonite mixtures.e sand-bentonite mixture comprised 90% air dried sand mixed with 10% CB. e dosage of PAC in the PAC-amended SB mixtures was controlled at 0.3, 0.6, 0.9, and 1.2%, and the bentonite content was also designed as 10%.
e PACamended SB mixtures were then made by mixing a certain mass of PAC with base mixture.e PAC-amended backfills with various PAC dosages were defined as PSB-i to present a base mixture for backfill with PAC content (PC) of i%, bentonite content (BC) of 10%, and sand content (SC) of (90−i)%.e designed proportions of each constituent for backfills are shown in Table 4.

Bentonite-Water Slurry.
e bentonite-water slurries (4 to 10% dry bentonite by mass) were prepared in order to evaluate the optimum bentonite content for using in slurry vertical cutoff wall construction.e slurries were mixed in a high-speed blender for 10 min and then sealed in a beaker and allowed to hydrate for 24 h.e slurries were stirred again for 5 min before measuring their rheological properties.e desired rheological properties of slurry to prepare SB backfills (Bohnhoff and Shackelford, 2013) include Marsh viscosity, density, filtrate loss volume, and pH values; the target viscosity is approximately 40 s as measured with a Marsh funnel viscometer in accordance with American Petroleum Institute (API) recommended practice 13B-1; the desired filtrate loss should be less than 25 mL as measured through the fluid loss test following the API 13B-1 and ASTM D5891; the pH values of slurries are in the range from 6.5 to 10.5 determined by a Model 720A pH meter (Cole-Parmer Instrument Company, Vernon Hills, IL), and the densities of slurries are in the range from 1.02 to 1.08 g/cm 3  as measured using a balance and a 1000 mL graduated cylinder.
e rheological property results of slurry of bentonite-water slurries are shown in Figure 1, and the results indicated that slurries with 8% and 10% bentonite content could satisfy the desired rheological properties for SB backfills slurry wall constructions.For economic and 2 Advances in Civil Engineering low-carbon costs consideration, the slurry with 8% bentonite content was selected as the reasonable test slurry.

Backfill Slump
Testing. e 8% bentonite-water slurry was thoroughly mixed with base mixture in various proportions to determine the amount of slurry, the mixing processes of which were carried out by using a paddle mixer, and the corresponding water content required to satisfy the workability of backfills with a measured slump of 125 ± 5 mm (ASTM C143).ree slump tests were performed for each specimen at any given water content to eliminate the variability in measured slumps, and the amount of added slurry was varied to provide a range of slump values and a corresponding range in the backfill-slurry water content (w bs ).e measured slumps of SB and PSB with various polymer dosages are shown in Figure 2.
e total bentonite content (BC total ) in the backfill was calculated using the following equations: where m ben is the mass of dry bentonite in the base mixture, m ben,s is the mass of dry bentonite from bentonite-water slurry, m total is the mass of backfill dry mass, c b is bentonite dosage of base mixture (i.e., 10% in this study), and c b,s is the bentonite contents of slurry (i.e., 8% in this study).e total bentonite contents of backfills with water contents of w bs were calculated, and the results are plotted in Table 5.
2.5.Permeate Liquids.Lead (Pb) was selected as a target contaminant as it was the most common contaminant found in subsurface soils and groundwater.To simulate Pb exposure, distilled water (DW) and lead nitrate (Pb(NO 3 ) 2 ) solutions were selected at various concentrations.e Pb(NO 3 ) 2 solutions were prepared by dissolving a certain mass of Pb(NO 3 ) 2 powder (chemical analytical reagent) in a known volume of distilled water to yield solutions with the designed concentrations (c d ) of Pb (20,50,100,200, and 500 mM).
ese concentrations are similar with the concentrations of heavy metal used in the previous studies [10,14,23].e Pb concentration (c m ) was measured by an atomic absorption spectrometer (ICE 3300, ermo Fisher Scientific Inc.).Triplicate samples are tested, and the average e density (ρ) of solutions was measured by a hydrometer 151H, and the absolute viscosity (μ) of solutions was tested by using a rotational Brook eld viscometer DV2T.e EC was tested through an electrical conductivity probe (150A conductivity meter, ermo Orion, Beverly, Massachusetts).e solution pH was conducted by a HORIBA D-54 pH meter as per ASTM D 4972.
e index property results of Pb(NO 3 ) 2 solutions are shown in Table 6.

Modi ed Filter Press Test.
e test process was originated with ASTM D5891 [24] and the API 131B, and hydraulic conductivity tests of back lls with API lter press apparatus (the inner diameter is 76.2 mm) are conducted.Firstly, the cell was upside down on the test platform, and then the petrolatum was lightly greased on sidewall to avoid sidewall leakage.e polyethylene (PE) backing ring, saturated porous stone, and lter paper were successively placed in the bottom of the cell.e aged saturated back lls with calculated mass was then poured into the cell, and it was compacted with an earth knife until it reached the designed thickness (i.e., 2.5 cm in this study).
e back lls were covered by another saturated lter paper and porous stone successively, and then the bottom cap of the cell was hermetically closed and turned over to keep it upright.Finally, the 120 mL test liquid was lled into the cell through inlet hole by a Pasteur pipette, and air pressure was applied on the top of the material through the inlet hole.
In order to obtain a relationship between hydraulic conductivity and stress, tests were performed under several di erent applied pressures ranging from 50 kPa to 400 kPa, which could be representative of stress conditions in site.
e pressures induced by liquid self-weight and soil selfweight were negligible compared to air pressure, which had very limited in uence on consolidation and e ective pressure of specimens.erefore, the liquid self-weight and soil self-weight could be ignored when the total applied pressure was calculated.
e typical termination criteria usually used during the test were (1) a steady hydraulic conductivity value for at least 4 consecutive measurements, (2) the hydraulic conductivity test should be performed for more than 24 h, (3) the ratio of pH, electric conductivity (EC), and the Pb concentrations between the e uents and in uents are within 1.0 ± 0.1 [25][26][27].At the end of the test, the e uent leachate volume and elapsed time were measured; the cell was disassembled immediately, and the properties of back lls (height, weight, and moisture content) were measured.
e height of the specimen was measured ve points on the surface with a  Advances in Civil Engineering digital slide caliper.e void ratio of backfills specimens was calculated from specific gravity, dry density, and water content of backfills specimens.e hydraulic conductivity of backfills was obtained from the following equations: where V is the leachate volume (m 3 ), t is the elapsed time (s), A is the cross section area of backfill specimens (m 2 ), i is the hydraulic gradient, H 0 is the headwater value, L is the length of backfills specimens, and c w is the gravity of water.
During the MFP test, the backfills were compressed by the applied pressure, P 0 (i.e., air pressure), and the average effective pressures (P e,ave ) of various backfills positions were different, which could be calculated from the following equations [27]: where x is the distance between the calculated position to the bottom of backfills specimen.
After the MFP test, the specific gravity (G S ) and liquid limits (w L ) of backfill specimens were evaluated following the ASTM standards.

Flexible-Wall Permeability Test.
e flexible-wall permeability test (FWP test) with the constant headwater method was conducted to verify the effectiveness of the MFP test.SB and PSB-0.6 backfills were used in comparison; the Pb(NO 3 ) 2 solutions with various concentrations (0-500 mM) were used as the permeant liquid.A rigid acrylic cylinder with inside diameter of 50 mm and height of 50 mm was placed around the flexible membrane to support the soft specimens during preparation of specimens, setting up into the testing cell, and permeation periods.After the specimens were prepared, they were placed into the vacuum saturation chamber for the purpose of specimen saturation, the air within cylinder was extracted for 2 h, then the influent valve of chamber was opened, and clean tap water was passed into the chamber under the air pressure until the liquid level higher than the specimens; the influent valve was then turned off, the influent tube moved out of the tap water surface, then influent valve was then turned on again for at least 4 h, and the specimens were saturated under the air pressure applied.After the specimen saturation was completed, the triaxial cell was assembled and a nominal confining pressure was applied.e average effective confining stress was maintained at 100 kPa, and the corresponding hydraulic gradient was 200.During the permeation period, the effectiveness stress was realized by setting cell pressure, bottom pressure, and top pressure as plotted in Table 7, respectively.e effluent bottle was connected with atmosphere, and hence the top pressure was 0 kPa.According to the ASTM D7100 standard [28], the permeation for each specimen was continued until at least four values of hydraulic conductivity are obtained over an interval of time and all of the following criteria are satisfied: (1) the ratio of outflow to inflow (Q out /Q in ) shall be between 0.75 and 1.25; (2) the hydraulic conductivity shall be considered steady if four or more consecutive hydraulic conductivity determinations fall within ±25% or better of the mean value for k ≥ 1 × 10 −10 m/s or within ±50% or better for k < 1 × 10 −10 m/s (3) at least 2 pore volumes of flow (PVF) shall occur through the specimen; (4) the ratio of pH, electric conductivity (EC), and Pb concentrations between the effluents and influents is within 1.0 ± 0.1.en, the permeated solution was replaced by Pb(NO 3 ) 2 solution with higher concentration, and permeation was conducted continually similar with the previous stage.

Index Properties of Backfills.
e specific gravity (G s ) results of both SB and PSB are shown in Figure 3(a).In  [29,30].
e results of liquid limits of both SB and PSB are presented in Figure 3(b).When exposed to DW, the w L of PSB increases with the dosage of PAC.With PAC content from 0 up to 1.2%, the w L of PSB increases from 34.5 to 60.5, which is almost twice of conventional back lls (i.e., SB).However, when exposed to Pb(NO 3 ) 2 solutions, the w L of both SB and PSB decreases with the increasing Pb concentration.With the Pb concentration increasing from 0 (DW) to 500 mM, the w L of SB back lls decreases from 34.5 to 19.1 and the w L of PSB-1.2 decreases from 60.5 to 22.5.e water absorbing capacity of PAC improves the w L of PSB, while the increasing predominance of lead ions (Pb) result from cation exchange with sodium ions (Na + ) and potassium ions (K + ) between bentonite interlayers, which causes stronger net interparticle forces and leads to a lower w L of bentonite [31,32].

Assessment of Chemical Equilibrium Conditions.
e chemical equilibrium is veri ed after the MFP test; the electrical conductivity (EC), pH, Pb concentrations of in uent solution (i.e., initial test solutions), and e uent leachate are measured as a supplemental criterion.Figure 4 shows the properties of e uent solution from SB and PSB specimens under 100 kPa overall stress at chemical equilibrium.e ratio of EC in e uent solution to EC in inuent solution (EC out /EC in ) is within 1.0 ± 0.1, the ratio of pH in e uent solution to pH in in uent solution (pH out / pH in ) is within 1.0 ± 0.1, and the Pb concentrations in e uent solution is within 1.0 ± 0.1 of the in uent solution (0.9 ≤ c out /c in ≤ 1.1).e results indicate that chemical equilibrium has been established in accordance with ASTM D7100.

e Optimum Polymer Dosage.
e k c of back lls permeated with Pb(NO 3 ) 2 solutions is shown in Figure 5, for a certain Pb concentration; the k c of back lls decreases with the increasing PAC content of back lls; and for a given PAC content, the k c of back lls increases with Pb concentration.Under applied pressure 100 kPa, the k c of PSB-0.6 was maintained lower than 10 −9 m/s, regardless the Pb concentration variation, which is below the limit designed value of hydraulic conductivity for typical sand-bentonite back lls in vertical cuto walls [1,33].e k c /k w of PSB-0.6 is lower than 10 permeated with concentrated Pb(NO 3 ) 2 solution (100-500 mM), indicating that the k c of PSB-0.6 has no obvious increase.
e results indicate that the additive PAC can improve the hydraulic performance and chemical compatibility of back lls when exposed to Pb(NO 3 ) 2 solutions.6 Advances in Civil Engineering Figure 6(a) presents the minimum PAC dosage of back lls (obtained from Figure 5) that satisfy requirements of hydraulic performance, when permeated with various Pb(NO 3 ) 2 solutions under 100 kPa.When permeated with concentrated Pb(NO 3 ) 2 solutions (100-500 mM), PSB with PAC dosage less than 0.6, the back lls exhibit k c > 10 −9 m/s and k c /k w > 10. erefore, the minimum PAC dosage of back lls that satisfy requirements of hydraulic performance is 0.6.Figure 6(b) shows the minimum PAC dosage of back lls exposed to 500 mM Pb(NO 3 ) 2 solutions under various pressures, which is a constant value 0.6, regardless the pressure variation; the result demonstrates that the minimum PAC dosage is independent on pressure.Based on the above analysis, the PAC content 0.6% is supposed as the optimum PAC dosage of back lls.In the following tests, the conventional back lls (SB) and PSB-0.6 were used in the MFP test under various pressures and the exible-wall permeability test.

Void Ratios of Back lls.
e correlation between void ratio (e) of SB and PSB-0.6 back lls under average e ective pressure is presented in Figure 7. Generally, PSB-0.6 back lls have greater e value than SB back lls, and the e values of both back lls decrease with increasing P e,ave and Pb concentration.
e results can be attributed to the following: the additive PAC in PSB-0.6 adsorbs more water and exhibits superior swell capacity, which also contributes to the higher water content of PSB-0.6, resulting in higher void ratio of PSB-0.6 compared to conventional back lls SB.Additionally, the lead ions (Pb) of liquids exchange the sodium ions (Na + ) and potassium ions (K + ) between bentonite interlayers, which causes the di use double layer (DDL) of bentonite compressed, and results in void ratio reduced.

Hydraulic Conductivities of Back lls.
e relationship between k c of back lls and the average e ective pressure (P e,ave ) is shown in Figure 8. e k c of PSB-0.6 is lower than k c of SB; the k c of both SB and PSB-0.6 back lls decreases with increasing P e,ave and increases with Pb concentration.
e results indicate PSB-0.6 has superior hydraulic performance compared to conventional back lls (SB), and Advances in Civil Engineering higher P e,ave results in smaller e value of back lls, which leads to lower k c for a certain Pb concentration.

Relationship between Hydraulic Conductivities and Pb
Concentrations. of SB is greater than that of PSB-0.6.When Pb concentration is equal to 20 mM, the k c of SB back lls exhibits greater than 10 −9 m/s and the k c /k w of SB greater than 10 under the applied pressure 100 kPa, while the k c of PSB-0.6 remains lower than 10 −9 m/s and the k c /k w of PSB-0.6 is less than 10 though the Pb concentration up to 500 mM.e results demonstrate that SB would have a signi cant increase once permeated with Pb concentration over 20 mM, whereas the PSB-0.6 could maintain low hydraulic conductivity (<10 −9 m/s) in the Pb concentration range from 0 to 500 mM.e hydraulic performance of back lls can be improved e ectively when exposed to Pb(NO 3 ) 2 solution by the additive PAC.
In order to investigate the impact of Pb concentrations on hydraulic conductivity of conventional back lls SB, the MFP tests on SB specimens permeated with Pb(NO 3 ) 2 solution with 0 to 50 mM (0, 5, 10, 15, 20, 30, 40, and 50 mM) were conducted, and the applied pressure was 100 kPa.
e k c and k c /k w of SB back lls exposed to Pb(NO 3 ) 2 solution with 0 to 50 mM are presented in Figure 10.When Pb concentration is lower than 5 mM, the k c of SB increases slightly and the k c /k w of SB is approximately equal to 1 time; with Pb concentration higher than 5 mM, the k c and k c /k w of SB exhibit obvious increase.When Pb concentration is equal to 15 mM, the k c of SB is 9 × 10 −9 m/s, which is almost equal to the limited value of hydraulic conductivity in hydraulic barriers (1 × 10 −9 m/s).With Pb concentration growing up to 45 mM, the k c /k w of SB is about 10 times, indicating that the k c of SB increases one order of magnitude and the hydraulic performance of SB back lls is signi cantly destroyed when exposed to Pb concentration higher than 45 mM. e reason for the increase of hydraulic conductivity of SB might be attributed to the di use double layer compression of bentonite; when exposed to divalent cations solutions (i.e., Pb 2+ in this study), divalent cations replace monovalent cations (i.e., Na + , K + ) originally in the exchange complex, thereby reducing or eliminating the osmotic swell of bentonite [17,32], resulting in ow paths more open and unimpeded and higher k c of SB.

e Validity of the Present Study.
e chemical equilibrium is evaluated during the exible-wall permeability (FWP) test.As shown in Figure 11, the ratio of out ow to in ow (Q out /Q in ) are within 1.0 ± 0.25 and the pore volumes of ow (PVF) are greater than 4 at each stage; the ratio of pH, electric conductivity (EC), and Pb concentrations between the e uents and in uents are within 1.0 ± 0.1.e results demonstrate that chemical equilibrium between Pb(NO 3 ) 2 solution and back lls has been established as per ASTM D7100 [28].
Hydraulic conductivity values of back lls from MFP tests (k MFP ) and exible-wall permeability tests (k FWP ) are plotted in Figure 12

Discussion
e results indicate that PAC-amended sand-bentonite back lls have superior hydraulic performance and chemical compatibility compared to conventional back lls, when exposed to Pb(NO 3 ) 2 solutions.e underlying mechanism of PAC-amended bentonite may include the following: (1) polymer intercalation between bentonite interlayers, which increases the space between platelets and activate osmotic swell [34].(2) Polymer adsorption to bentonite surface through exchangeable cation bridging [35].In general, the extent of adsorption depends on the molecular weight and number of hydrophilic functional groups [36]; PAC has a large amount of carboxyl and hydroxyl groups and an enough molecular weight (1,730,000), which can adsorb many bentonite surfaces with a single polymer molecule by exchangeable cation bridging.(3) Polymer chains remove the intergranular porosity and stitch the granules together [14].
Previous studies [14,37] indicate that a polymer can adsorb large amount of water molecules and then forms swellable hydrogels at the macroscale, which forms a threedimensional net structure between bentonite granules.e water molecules in the hydrogel polymer are immobile compared to free water in the pore space.erefore, the hydrogel polymer clogs larger pores that can conduct ow of water and solutes in back lls.
Moreover, the modi ed methods of bentonite may have an important impact on the chemical compatibility of polymeramended bentonites and back lls; previous studies have conducted several modi ed methods: (1) polymer is directly mixed with dry bentonite [15,27,38]; (2) polymer solution is blended with bentonite slurries [13,20]; and (3) monomer solution polymerization in bentonite slurry [14,34].e difference between several methods for PAC-amended bentonite and back lls should be investigated in the next study.
Additional research is warranted to investigate the adsorption property, membrane behavior, and microscopic mechanism of the PAC-amended sand-bentonite back lls under lead ion contaminated conditions.Data Availability e data used to support the ndings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no con icts of interest.

Figure 2 :Figure 1 :
Figure 2: e slump values of SB and PSB with various polymer dosages.

Figure 3 :
Figure 3: (a) e relationship between speci c gravity of back lls and Pb concentrations (number of replicates 3; coe cient of variation ≤ 0.01) and (b) the relationship between liquid limits of back lls and Pb concentrations (number of replicates 3; coe cient of variation ≤ 0.04).

Figure 5 :Figure 6 :Figure 4 :
Figure 5: e correlation of k c -polymer content of back lls permeated with Pb(NO 3 ) 2 solutions.

Figure 9 (
a) shows the relationship between k c of back lls and Pb concentration (c), and Figure 9(b) displays the correlation between k c /k w and c.Generally, the k c of SB and PSB-0.6 increases with Pb concentration, and k c . By comparison, k MFP values are 4 to 6 times greater compared to the k FWP value when permeated with DW.However, k MFP values are less than 4 times greater compared to the k FWP value when exposed to Pb(NO 3 ) 2Average effective pressure, P e,ave(kPa)

Figure 8 :
Figure 8: e relationship between hydraulic conductivity (k c ) of back lls and the average e ective pressure (P e,ave ).

Figure 7 :
Figure 7: e void ratio (e) of back lls and applied pressure.

AFigure 10 :
Figure 10: e hydraulic conductivity of (k c ) of SB back lls when exposed to Pb(NO 3 ) 2 solution with 0 to 50 mM.

Figure 9 :
Figure 9: e relationship between (a) hydraulic conductivity of (k c ) of back lls and Pb concentration (c), and (b) hydraulic conductivity ratio (k c /k w ) of back lls and c.

Figure 11 :Figure 12 :
Figure 11: e chemical equilibrium condition of the exible-wall permeability test.

Table 1 :
Properties of CB used in this study.

Table 2 :
e mineral fraction of CB based on X-ray diffraction analyses.

Table 3 :
Properties of polymer used in this study (data from the manufacturer).

Table 6 :
e index properties of the Pb(NO 3 ) 2 solutions.

Table 5 :
e initial index properties of backfills before modified filter press test.

Table 7 :
e pressure used in exible-wall permeability test.Hydraulic gradient, i Bottom pressure, P B (kPa) Top pressure, P T (kPa) Cell pressure, P C (kPa) E ectiveness stress, P e (kPa) It should be noted that the k MFP of SB are approximately equal to k FWP of SB when permeated with Pb(NO 3 ) 2 solutions.eseresults suggest that k MFP values and k FWP value of back lls are within the same order of magnitude, which demonstrate that the modi ed lter press test is an e ective and easy evaluation of hydraulic conductivity of back lls.