Geotechnical properties of hazardous landfill leachate-contaminated sand

Many industrial activities dispose of their waste in the sand soil of the El-Khatatba region of Egypt. There is a need to reuse the site in building markets and shopping stores. Because it is difficult to obtain leachate from an open dumping site, we use hazardous landfill leachate (HLL) from an industrial waste landfill to identify changes in the geotechnical properties of soil. The HLL was obtained from Al-Nasreya Center, Alexandria, Egypt. This study, therefore, was directed to assess the effects of HLL on the geotechnical properties of the poor-graded sand (SP) soil. The natural sand was artificially mixed with HLL at different concentrations of 0%, 25%, 50%, 75%, and 100%. The results revealed that the permeability coefficient (k) of the natural sand decreased significantly from 1.28 × 10−2 to 0.8 × 10−2 cm/s when HLL concentration increased to 100%. Similarly, HLL contamination obviously reduced the angle of internal friction (φ) of sand. The optimum moisture content (OMC) of the natural sand decreased from 12 to 10%. In contrast, the maximum dry unit weight (MDD) of the natural sand increased slightly from 18.33 to 18.9 kN/m3 due to the increase of HLL concentrations from 0 to 100%. The change in the sand structure was interpreted by detailed analyses using scanning electron microscopy (SEM) and X-ray diffraction (XRD). Based on these findings, HLL was found to modify the geotechnical properties of the studied sand.


Introduction
The remarkable increase in population has recently let waste disposal and management to be a major concern confronting government (Venkatesan and Swaminathan 2009). For instance, Egypt generates 0.3 to 0.8 kg/day/capita of municipal solid waste (MSW) with an annual growth of 3.4% (Elfeki and Tkadlec 2015). Likewise, the annual production of industrial wastes accounts for approximately 6.2 million tons including 0.2 million tons of hazardous waste (Elfeki and Tkadlec 2015). In many developing countries including Egypt, the most traditional way to dispose of solid waste is open dumping in deserts or open areas. The Egyptian authorities have taken serious steps to close and reuse those open dumping sites. Recently, landfill construction is considered the most common way for final disposal of solid waste, thanks to its relatively low environmental threats. Yet, landfill constitutes a potential source of soil contamination because of the generation of highly toxic leachate (Meky et al. 2019). Consequently, the assessment of the landfill status (working or closed) is crucially important (Al-fares and Al-jarallah 2011). The well-designed landfills minimize the leakage of leachate; nevertheless; they cannot prevent it completely in the cases of occurrence in the vicinity of natural disasters and landfill operations (Ozcoban et al. 2006). Oztoprak and Pisirici (2011) estimated the leakage of leachate in a landfill with compacted clay liner to be ~ 30 mm/ year and geomembrane liners to be ~ 150 mm/year. In every country in the world, landfills are mostly unsuccessful since their sites have a limited time of usage (Narayana 2009). Thus, the need to assess the variation in the geotechnical, geo-environmental, and geochemical properties of those sites has become essential, as there is a strong need to reuse the polluted site due to the shortage of footprint for residential and industrial development areas. The reuse of polluted sites such as open dumping and unlined old landfills is becoming increasingly widespread in redeveloping the land where there are many projects constructed on old landfills such as golf courses, athletic fields, parks, roads, shopping centers, university buildings, and stadiums (Oztoprak and Pisirici 2011). From an engineering standpoint, the reuse of contaminated sand for road construction is possible; however, the contaminated site should be enclosed or capped to limit environmental pollution (Al-Sanad et al. 1995). Many studies have focused on the effects of municipal landfill leachate on the geotechnical properties of soil; however, there are a limited number of studies that discussed the influence of hazardous landfill leachate (HLL) on these properties (Khodary et al. 2018) which are extensively studied in this paper. Hazardous waste landfills that receive industrial wastes are mainly constructed in the desert where the sandy soil prevails. Therefore, it is necessary to examine the impact of hazardous waste leachate on the behavior of sandy soils, as well as understand their mechanisms. This issue has recently become more critical because open dumping or landfill sites are used to meet the increasing demand for roads/buildings caused by population growth.
The transport of dissolved pollutants in soil occurs due to advection, dispersion, or diffusion processes (Durães et al. 2018). Transport mechanisms in porous soil frequently cause changes in the pore structure, through precipitation and dissolution of biomass or minerals, and these changes are influenced by changes in the porosity and permeability of the material (Hommel et al. 2018). Therefore, soil contamination must be carefully considered during all construction stages to prevent new contamination occurrence and control the current one (Sridharan et al. 2010). Several properties of soil are subjected to change depending on the type of contaminants and the type of interactions that occur within the soil-water system (Sridharan et al. 2010).
Even though several studies (Jayasekera and Mohajerani 2001;Nayak et al. 2007;Francisca and Glatstein 2010;Li et al. 2013;Ören and Çağrı 2016;Khodary et al. 2020;Khodary et al. 2021) have previously examined the effects of landfill leachate on the geotechnical properties of clayey soil, slight attention has been paid to the influence of this leachate on the geotechnical properties of the sandy ones. For instance, Jayasekera and Mohajerani (2001) examined the effects of synthetic leachate based on the composition of typical MSW landfill leachate on the performance of compacted basaltic clay soil in different periods. The results revealed that the behavior of clay was affected significantly by clay-leachate interactions over time due to the alterations in mineralogical and physical properties of the soil. It was also observed that the plasticity index and free swelling reduced due to the reduction of the double layer thickness. Nayak et al. (2007) conducted the Standard Proctor test on a mixture of lateritic soil and leachate with different contents (5%, 10%, and 20%). The results demonstrated that maximum dry unit weight (MDD) reduced slightly from 15.5 to 14.3 kN/m 3 , whereas the optimum moisture content (OMC) increased from 19.52 to 26% with 20% leachate content due to the chemical reactions between the acidic leachate and soil. They found that the structure of the leachate-contaminated soil was changed to be aggregated, in which the effective pore space increased, leading to an increase in hydraulic conductivity. Sunil et al. (2009) as well carried out consolidated undrained on lateritic soil mixed with 0%, 5%, 10%, and 20% MSW leachate. The soil samples were molded at maximum dry unit weight obtained from Standard Proctor test. A slight increase in cohesion and a reduction in the friction angle were observed due to the increase of clay content in lateritic soil after the interaction with the leachate. Francisca and Glatstein (2010) evaluated the long-term hydraulic conductivity of the compacted silt-bentonite mixtures in the case of landfill leachate infiltration. The hydraulic conductivity of soil decreased obviously when the permeating liquid contained microorganisms. The reduction of the effective porosity minimized the hydraulic conductivity of the soil due to the pores clogging. Ören and Çağrı (2016)stated that the free swell of bentonite which was contaminated by landfill leachate was significantly reduced. Abdellah et al. (2020) indicated that the mixture of 10% bentonite, 20% calcareous sand, and 70% tuff can meet the requirements of landfill bottom liners design. A reduction in the pH and an increase in the EC of MSW leachate-contaminated mixtures were recorded. An increase in shear strength was observed as the calcareous sand content increased. Due to increasing the compressibility of leachate-contaminated mixture, the compressive strength and tensile strength increased, and saturated hydraulic conductivity decreased. Khodary et al. (2021) reported that the plasticity index, permeability, and shear strength parameters of leachatecontaminated clay significantly decreased.
In this study, an attempt has been made to explore the impact of HLL on the compaction characteristics, permeability, and shear strength of the sandy soil to assess the potential of reusing the contaminated site for construction purposes.

Materials
The natural sandy soil used in the present study was collected from an open excavation at a depth of 1.5 m below the natural ground level from El-Khatatba region (43 km north of the Egyptian capital Cairo). The grain size distribution curve is plotted in Fig. 1 (according to ASTM D 422), while the different properties of the sand are listed in Table 1. The sand is classified as poor-graded sand (SP) according to ASTM D2487.
To identify the changes in the geotechnical properties of soil, we use HLL from an industrial waste landfill because it is very difficult to obtain leachate from an open dumping site. The HLL was obtained from Al-Nasreya Center, Alexandria, Egypt, which is the only centralized hazardous industrial waste disposal plant in Egypt and has been established since 2004. The facility is considered the first and largest landfill of its type in Egypt and received limited types of industrial waste materials including dry inorganics, asbestos, ash from heavy fuel oils, contaminated soil, heavy metals, sludge from galvanic processes, insoluble metal salts, inks, dyes, lacquers, paint sludge, resins, polymers, and chemical containers. The landfill is still active and produces approximately 10 m 3 /day of leachate which is stored in a pond (Ramadan et al. 2005). Al-Nasreya Center involves a landfill cell with capacity of 40,000 m 3 and four solar evaporation ponds covering in total of 5000 m 2 . They were selected as means of drying liquid and sludge waste benefiting from the dry hot climate where the mean temperature is 30 °C and 19 °C in summer and winter, respectively. The average rainfall which occurs only in the winter is 200 mm per year.

Sample preparation
The sand specimens were dried in an oven at temperature of 105 °C and sieved through a 4.75-mm screen before testing.  The samples were mixed with HLL at concentrations of 0% (i.e., distilled water), 25% (i.e., one volume HLL to three distilled water), 50% (i.e., one volume HLL to one distilled water), 75% (i.e., three volume HLL to one distilled water), and 100% (i.e., raw HLL). All samples were artificially saturated with the HLL and were placed in moisture-tight bags for successive 7 days followed by air drying for 2 weeks. Afterward, the samples were dried in the oven at 40 °C for 7 days. A comprehensive laboratory testing was carried out on the natural and contaminated sand including permeability, standard compaction, and shear strength tests according to ASTM standards.

The electrical conductivity and pH of HLL
The HI 9033 conductivity meter was used to determine the EC of HLL, while the GLP21 pH meter calibrated with pH standard was used to determine the activity of the proton at 25 °C.

The total dissolved solids (TDS)
The HLL samples were filtered through glass-fiber filters (No. 47), which were then evaporated in the oven (103 to 105 °C) for 2 h. After cooling in a desiccator for 30 min, the TDS was calculated.

The total and soluble chemical oxygen demand (CODt and CODs)
The spectrophotometer was used to determine the CODt and CODs in the HLL sample using potassium dichromate and sulfuric acid as indicators.

Biochemical oxygen demand (BOD5)
A 250 mL of HLL sample was placed in a recipient sealed with a monometer in constant agitation in dark conditions at 20 °C. NaOH was added to absorb the carbon dioxide produced from the process. After 5 days, the BOD5 was calculated.

The soluble anions and cations
Sodium and potassium contents in HLL samples were determined by the flame photometric method, while calcium and magnesium contents were determined by titration using EDTA solution as an indicator. Bicarbonate content was determined by titration using phenolphthalein and methyl orange as indicators. The silver nitrate titration method was used to determine the chloride content. A spectrophotometer was used to determine the sulfate content.

The atomic absorption
An atomic absorption spectrophotometer (AAS) (Perki-nElmer Model Analyst 200) was employed to analyze the concentration of heavy metals in the HLL. Firstly, the HISWL samples were filtered through a 0.45-μm cellulose acetate membrane (Sigma-Aldrich, USA). The filtrate was then analyzed for residual heavy metals after being acidified with nitric acid (pH < 2) in order to avoid metal precipitation and adsorption.

The gas chromatography-mass spectrometry (GC/MS)
The presence of hydrocarbons in the HLL was checked using a gas chromatographic-mass spectrometer (GC/MS-Shimadzu-Qp2010 Ultra). The hydrocarbons in the HLL samples were extracted by using methyl tertiary-butyl ether (MTBE) as a solvent and injected into GC/MS with a ratio of 1:1, with 2-Bromo-1-chloropropane (20 ppm) used as an internal standard. The injection temperature was adjusted to 200 °C, and helium was used as the carrier gas with a flow rate of 1 ml/min. The oven temperature was kept at 35 °C during the first 7 min, and then the column temperature was programmed from 35 to 90 °C at a flow rate of 20 ml/min.

The specific gravity test
The G s of the soil was performed according to ASTM D 854-92 using a pycnometer.

The grain size test
According to ASTM D 422-63, a dry sieve analysis was used for the natural soil particle sizes retained on the No. 200 sieve.

Permeability test
The constant head test was conducted on the natural and contaminated samples to determine the permeability coefficients with dry unit weight equal to MDD according to ASTM D2434-68. The distilled water was used as a permeation fluid.

Shear test
Shear characteristics for the soil were determined using the direct shear box (60 × 60 mm) with a shear rate of 1 mm/ min at normal stresses of 0.5, 1, and 1.5 kg/cm 2 (49, 98.07, and 147.1 kN/m 2 ) on remolded samples at MDD and OMC (in each concentration) in accordance with ASTM D3080.

Compaction test
Compaction characteristics of the sandy soil were performed using the Standard Proctor test (ASTM D698-07 Method A) in order to determine the MDD and the OMC. A cylindrical mold with a 10.16 cm diameter and 11.64 cm height was used for this purpose.

Electrical conductivity and pH of sand
The EC of the natural and contaminated soil was determined by mixing the dried samples with distilled water in a ratio of 1 (solid)/1 (water), shaking this suspension periodically for 1 h, and measuring the EC of the suspension by a conductivity meter. The same process was done for the pH test by using a GLP21 pH meter.

The Brunauer, Emmett, and Teller (BET) test
The Brunauer, Emmett, and Teller (BET) test was employed in order to determine the total pore volume of natural and contaminated samples by using BET Sorb device. The measurement conditions were as follows: sample weight = 0.217 g, standard volume = 9.826 cm 3 , dead volume = 16.698 cm 3 , equilibrium time = 0.0 s, apparatus temperature = 0.0 °C, adsorption temperature = 77.0 K, saturated vapor pressure = 101.64 kPa, and adsorptive cross section area = 0.162 nm 2 .

The scanning electron microscope (SEM)
The scanning electron microscope (SEM) was performed to examine the surface morphology of the sand before and after the HLL contamination by using the JEOL model JSM 636O LA with a high-resolution range from 100 to 0.5 nm at 20 kV. The samples were prepared in an identical method employed in the XRD test and then were coated with a thin golden layer (5-50 nm). The microscopic images were magnified from × 1000 to × 30,000 for both natural and 100% HLL-contaminated sand.

The X-ray diffraction (XRD)
The X-ray diffraction (XRD) technique was used to identify the mineralogical composition of the natural and 100% HLL-contaminated samples. The samples were analyzed by Shimadzu Xlab 6100 equipment using parallel beam optics with Cu Ka radiation at 30 mA and 40 kV. The diffracted X-rays have been collected over a 2θ range of 10-80° using a sampling pitch of 0.02°. Other measurement conditions were as follows: divergence slit of 1.0°, scatter slit of 1.0°, receiving slit of 0.3 mm, continuous scan mode of 12.0 deg./ min, and preset time of 0.1 s.

Effect of HLL on the geotechnical properties of sand
The permeability coefficient (k) of the natural sand was affected by HLL contamination (Fig. 2). The k value declined slightly from 1.28 × 10 −2 to 1.2 × 10 −2 cm/s for the natural and 25% HLL-contaminated specimens, respectively. Nevertheless, a significant drop in k value of 0.94 × 10 −2 cm/s took place at increasing the HLL concentration up to 50%. The k levels were 0.845 × 10 −2 and 0.8 × 10 −2 cm/sec for the contaminated sandy soil with 75% and 100% HLL concentrations, respectively. The total pore volumes of natural and contaminated sand as determined by the BET test are presented in Fig. 2 where the total pore volume of sand soil decreased as the HLL concentration increased. The total pore volume of the natural sand decreased significantly from 1.24 × 10 −2 to 0.88 × 10 −2 cm 3 /g with the addition of 100% HLL concentration. Soil permeability depends on the shape and size of the particles, arrangement of the pores and soil particles, void ratio, and properties of the pore fluid (Michaels and Lin 1954). As reported earlier by Khodary et al. (2020), leachate containing suspended solids blocks the pores of the soil, resulting in a decrease in hydraulic conductivity. Moreover, the HLL used in the present study has alkaline characteristics (pH = 8.78) and affects the sandy Permeability coefficient Total pore volume Fig. 2 The permeability coefficients of natural and HLL-contaminated sand and their total pore volumes soil particles to be finer leading to a reduction in the pore space and permeability. The total pore volume of sand soil decreased when the HLL concentration increased resulting in a reduction of permeability. The results are in agreement with those reported by Eltarabily et al. (2015) who demonstrated that the alkaline solution caused a decrease in the permeability of the coarse sand. Additionally, the lubrication effect of hydrocarbon compounds in the HLL greatly affected the sand permeability leading to a deduction in its values. Singh et al. (2009) confirmed that the permeability of the natural sand decreased with oil contamination. The physicochemical interactions between the soil and pore fluids in granular soils are negligible, but the changes in the behavior of granular soils are attributed to the lubrication effect created by pore fluid viscosity. The relationship between normal and shear stresses of natural and HLL-contaminated sand are indicated in Fig. 3a. The angles of internal friction (φ) of the natural sand gradually decreased as the HLL concentrations increased, resulting R 2 of 0.87 (Fig. 3b). The φ of natural sand has dropped significantly reaching to 33.24° with supplying 25% HLL. However, the φ reduction was slight at increasing the HLL from 25 to 50%. The φ of the contaminated sand was 32.48° at 50% of HLL. Further increase in the concentration of HLL from 75 to 100% resulted in a marginal decrease in φ from 31.18 to 30.05°, respectively.
Mineral composition, grain size, particle shape, and roughness are the main factors that affect the geotechnical properties of soil (Ziȩba 2017). The shape and roughness of soil particles have a considerable influence on their internal friction angle, compressibility, and hydraulic conductivity (Singh et al. 2009). The particle shape determines the geometry and size of pore space which highlights the effective porosity as a decisive factor for the soil hydraulic conductivity (Robin et al. 2016). Shin and Santamarina (2013) reported that the friction angle of soil is mainly affected by particle roundness and size. In this study, the significant reduction in sand friction angle due to HLL contamination might be associated with the presence of the hydrocarbon components which have been detected by gas chromatography-mass spectrometer. The lubrication effect of hydrocarbon compounds alters the shear behavior of soil resulting in coating the sand particles with hydrocarbons. The reduction in internal friction due to hydrocarbon contamination was confirmed by many researchers (Cook et al. 1992;Puri 2000;Khamehchiyan et al. 2007;Siang et al. 2014;Nasehi et al. 2016). Li et al. (2013) reported that leachate could lead to an increase in the friction angle and a reduction in the cohesion of compacted clay. On the other hand, another possible rationalization for the reduction of sand internal friction is changing the geometry and size of the particle as explained in the scanning electronic microscope (SEM) analysis in the "Morphological and mineralogical changes of the sandy soil due to HLL contamination" section.
The MDD and OMC values of the natural and HLLcontaminated soil samples are presented in Fig. 4a, while the compaction curves are plotted in Fig. 4b. The MDD of the sandy soil slightly increased from 18.33 kN/m 3 (for the natural samples) to 18.48, 18.55, and 18.67 kN/m 3 at HLL concentrations of 25%, 50%, and 75%, respectively. However, the MDD has remained at the level of 18.9 kN/m 3 at the highest HLL concentration of 100%. This was not the case for the OMC where it decreased slightly from 12% (for the natural samples) to 11.5% at HLL concentrations of 25% and almost kept constant at a level of 11.5% at HLL concentrations of 50%. Furthermore, supplementation of 75% and 100% of HLL reduced the sand moisture content to 11% and 10%, respectively.
A rational explanation for the slight decrease in OMC and the marginal increase of MDD by HLL contamination may be due to the presence of hydrocarbons in the leachate. Reducing the friction between the sand particles results in the closer spacing of the grains and provides higher dry density. The lubricating effect caused by the presence of hydrocarbons in the HLL facilitated the compaction process and reduced the water amount needed to reach the maximum density. Some others (Al-Sanad et al. 1995;Puri 2000) have revealed that OMC decreased and MDD increased slightly by hydrocarbon contamination. These results are likely to be associated with the change in the nature of the pore fluid where the HLL contains high concentrations of suspended solids and salts which alter the soil-water characteristics.
The electrical conductivity (EC) of the natural and contaminated sandy soil is presented in Fig. 5. The EC of the natural sand was estimated to be 0.74 dS/m, which is lower than the EC of the HLL used in this study. Consequently, the blended sand samples are expected to provide EC higher than that of the natural sand. The EC values experienced a dramatic growth where they rocketed from 7.55 to 20.22 dS/m in the case of 25% and 100% HLL concentrations, respectively.

Morphological and mineralogical changes of the sandy soil due to HLL contamination
Typical images of the natural and 100% HLL-contaminated sand samples at different magnifications × 1000, × 5000, and × 30,000 are indicated in Fig. 6. The images have revealed significant micro-fabric differences between natural and 100% HLL-contaminated samples. From Fig. 6.1a, the samples of natural sand at × 1000 magnification presented irregular aggregates with a sub-angular to sub-rounded particle shape. In addition, they exhibited much more pores. Above the sand grains, there are fine grains of silt and clay particles. At × 5000 ( Fig. 6.1b), the sand particles showed irregular depressions, while at × 30,000 ( Fig. 6.1c), some halloysite tubes appeared. On the other hand, the textures of natural and contaminated sand were fundamentally different. The HLL caused the sand surface to be distorted and its particles to be finer. Moreover, the pores between aggregates are more common in the natural sand than in the contaminated ones as shown in Fig. 6.2a. As appeared in Fig. 6.2b, the edges of silt particles were attacked by HLL causing abrasion of their boundaries, while the particles were subdued with a layer of precipitated salts and the halloysite appeared to increase as shown in Fig. 6.2c.
The modification of the geotechnical properties of soil mainly takes place due to the physical and/or the physicochemical interactions between contaminant and soil (Khodary et al. 2018). Viscosity of the contaminant is highlighted as the most noteworthy factor in the physical interaction which is predominant in the granular soil (Ratnaweera and Meegoda 2006). From SEM images, the textures of  Due to its alkalinity characteristics, the HLL caused the sand particles to be distorted and the particles to be finer as well as led to dissolving the calcite in the contaminated soil. These features of the contaminated sand caused a decrease in friction angle, while soil permeability tended to decrease due to increasing the salt contents in the contaminated sand. The HLL is a complex liquid containing many components which can change the soil structure and lead to mineralogical changes. Li et al. (2013) found that quartz content in compacted clay decreased, while albite increased after leachate pollution.

Conclusions
To develop a full picture of the behavior of leachatecontaminated sand, a laboratory testing program was conducted to assess the effect of hazardous landfill leachate (HLL) contamination on the geotechnical properties of the Egyptian sand at different concentrations (0%, 25%, 50%, 75%, and 100%). Our investigation has concluded that HLL contamination led to decreasing the sand permeability where the permeability coefficient of 100% HLL-contaminated sand dropped from 1.28 × 10 −2 1a) 1 b) 1c) 2a) 2 b) 2c)

Fig. 6
Microscopic images for the 1) natural and 2) the 100% HLL-contaminated sand. The symbols a, b, and c represent magnification of × 1000, × 5000, and × 30,000, respectively to 0.8 × 10 −2 cm/s compared with the natural sand. Correspondingly, leachate contamination has some adverse effects on the shear strength of the sand. The angles of internal friction were sharply dropped by HLL contamination from 37.87° (for the natural sand) to 30.05° (for 100% HLL-contaminated samples). Such a result was mainly attributed to the hydrocarbons present in the HLL and changing the sand particles geometry caused by the alkaline HLL. The compaction test has revealed that the optimum moisture content decreased from 12 (in the natural sand) to 10% in the case of 100% HLL-contaminated samples. The maximum dry unit weight of sand was maximized to 18.9 kN/m 3 in the 100% HLL concentration. These findings can help geotechnical engineers to understand the behavior of such leachate-contaminated sand. It is predicted that the leachate-contaminated sand is prone to settle under loading and might experience a deterioration of its bearing capacity due to the reduction of its shear strength.