Remediation of Arsenic Contaminated Soil Using Phosphate and Colloidal Gas Aphron Suspensions Produced from Sapindus mukorossi

  • Soumyadeep Mukhopadhyay
  • Sumona Mukherjee
  • Mohd Ali Hashim
  • Bhaskar Sen Gupta
Article

Abstract

Phosphate and colloidal gas aphrons (CGAs) generated from saponin extracted from Sapindus mukorossi fruit, were evaluated for washing low levels of arsenic from an iron rich soil. Phosphate is one of the most commonly dispersed chemicals that increases arsenic mobility in soil due to their structural similarities, making it an important factor in arsenic removal process. Column washing experiments were performed with CGAs in down flow and up flow modes on soil of pH 5 and 6. Soapnut CGAs, when paired with phosphate removed up to 95 % arsenic while soapnut CGAs alone could only remove up to 70 % arsenic. The presence of phosphate improved efficiency of soapnut solution by up to 35 %. SEM image of washed soil revealed minor corrosion of soil surface while using phosphate with soapnut. Therefore, the addition of phosphates would have positive impact on soil washing using soapnut saponin.

Keywords

Arsenic Soapnut Sapindus mukorossi Soil remediation Colloidal gas aphrons 

USEPA has classified arsenic as a human carcinogen. Soil and water pollution through mining, smelting, coal burning, wood preservation and illegal waste dumping activities releases arsenic in ecosystem (Tokunaga and Hakuta 2002). In a recent study, soil of golf course was found to be contaminated with arsenic sourced from seaweed fertiliser (O’Neill et al. 2014). Usual screening concentration of arsenic in soil usually varies in the range of 10–16 mg Kg−1 for industrial soil (DOE Malaysia 2009). Arsenic and other contaminants can be removed from soil through a host of technologies including bioremediation and soil washing with surfactants and other chemicals (Jang et al. 2005; Majumder et al. 2013; Mukhopadhyay et al. 2015a).

CGAs have shown excellent potential as separation agents for contaminant removal from liquid and soil matrices (Hashim et al. 2012). Saponin, a natural surfactant obtained from the fruit pericarp of Sapindus mukorossi or soapnut has been used for removing organic and inorganic contaminants including arsenic from soil (Mukhopadhyay et al. 2013, 2015a; Roy et al. 1997; Song et al. 2008). Phosphates (PO43−) share chemical properties and structural similarities with arsenates (AsO43−) and have been frequently used for removing arsenic from soil. Alam et al. (2001) extracted up to 40 % arsenic from a yellow/brown forest soil by 0.9 M phosphate solution. Other researchers (Tokunaga and Hakuta 2002) also removed 80 %–99 % of the bound arsenic from contaminated soil by phosphate and phosphoric acids. Zeng et al. (2008) found that KH2PO4 at lower concentration was less corrosive than H3PO4. Hence, KH2PO4 was used in the present work in the range of 50–150 mM in combination with saponin extracted from soapnut to extract arsenic from soil.

In this work, soil As(V) concentration of up to 85 mg Kg−1 has been washed using CGAs generated from saponin extracted from soapnut pericarp mixed with phosphate (referred to as Ph + SN now on). The enhancement of arsenic removal performance by saponin CGAs in presence of phosphate has been investigated. The sustainability of the process has been assessed using SEM imagery to probe the soil corrosion. The mechanism of the soil washing process has also been described. Presence of large amount of phosphate in the terrestrial and aquatic environment originating from NPK fertilisers signify that an in situ soil washing process involving saponin and CGAs will need to consider phosphate’s effect on the process.

Materials and Methods

A composite soil sample was collected from the first layer aquifer in Hulu Langat, Malaysia. The soil characteristics are given in Table 1a. For this study, the soil was spiked with arsenic(V) salt (Na2HAsO4·7H2O) as described by Mukhopadhyay et al. (2015b). Natural saponin from the soapnut fruit pericarp was extracted with water following the methodology described by Roy et al. (1997). The aqueous soapnut extract contained up to 65 % saponin measured by UV–visible spectrophotometer (Roy et al. 1997). Based on preliminary experiments, 0.5 %, 0.75 %, 1 %, 1.25 % and 1.5 % (w/v) of soapnut extracts were combined with 50, 75, 100, 125 and 150 mM phosphate for generating solutions and CGAs. Solutions and CGAs of 20 mM SDS (Sodium dodecyl sulphate), a synthetic surfactant, were used for initial comparison, which ensured that the chosen surfactant concentrations exceeded their respective critical micelle concentration (CMC). Among the surfactants used, SDS is anionic while saponin is slightly anionic at natural pH of 4.5. Washing agents are characterized at selected concentrations as shown in Table 1b. The experiments were performed in a 15 cm long steel column with 5.5 cm internal diameter, discussed elsewhere (Mukhopadhyay et al. 2015b). About 300 g soil was packed in the column. The flow rate for the washing agent, other standard conditions and variables such as pH, flow modes, concentration of wash solutions are provided in Table 2. The packed column was first flooded with water in upflow mode at the rate of 5 mL min−1 to eliminate any air pocket. Arsenic desorption was then initiated by pumping surfactant solution or CGAs at constant flow rate of 10 mL min−1. The wash solutions using both soapnut and soapnut + phosphate were prepared at five different concentration levels, viz. low (L), medium low (LM), medium (M), medium high (MH) and high (H), the values have been provided in Table 2. Factorial experiments were performed in all combinations of variables mentioned in Table 2. CGAs were generated by a homogenizer and characterized by liquid drainage rates, half-life and the air holdup following Bhatia et al. (2005). FT-IR spectroscopy was used to identify the functional groups in the soapnut extract. Kinetics of arsenic desorption and corrosion of soil following soapnut and phosphate washings were investigated by the method reported by Mukhopadhyay et al. (2015b).
Table 1

(a) Characterization of natural soil sample; (b) arsenic speciation in spiked soil; (c) characterization of washing agents

Soil properties

Value

Method

(a) Characterization of spiked soil sample

 pH

4.45

USEPA SW-846 Method 9045D

 Specific gravity

2.63

ASTM D 854 – Water pycnometer method

 CEC (Meq)

5.2

Ammonium acetate method for acidic soil

 Organic matter content

0.15 %

Loss of weight on ignition (Storer 1984)

 Bulk density (gm cc−1)

1.50

 

 Total porosity (%)

39

 

 Total As (mg kg−1)

86.00

USEPA 3050B

 As (III) (mg kg−1)

2.70

 As (V) (mg kg−1)

83.30

 Total iron (mg kg−1)

3620

 Total silicon (mg kg−1)

~390,000

 Aluminium (mg kg−1)

2500

 Total manganese (mg kg−1)

190

 Magnesium (mg kg−1)

655

 Lead (mg kg−1)

20

 Zinc (mg kg−1)

15

Soil particle size distribution

 Sand (<50 μm)

92.06 %

Sandy soil according to USDA Soil Classification

 Silt (50–2 μm)

5.2 %

 Clay (>2 μm)

2.55 %

Empirical formula

Molecular weight

Concentrations used

CMC at 25°C

Surface tension (mN m−1)

pH

Viscosity (at 25°C) cP

(b) Characterization of washing agents

 H2O

18

 

71.2

7

0.89

 C52H84O21·2H2O

1081.24

0.5 %

0.1 %

41

4.64

1.10

1 %

 

40

4.48

1.22

1.5 %

 

39.5

4.38

1.31

 NaC12H25SO4

288.38

10 mM

8.2 mM

34

9.65

1.38

 

20 mM

 

32

10.05

1.43

 

30 mM

 

31

10.30

1.49

0.5 % + 50 mM

0.1 %

44.3

4.79

1.05

  

1 % + 100 mM

 

45

4.69

1.13

  

1.5 % + 150 mM

 

46.6

4.62

1.21

Table 2

Experimental conditions and variables

Standard conditions:

 300 gm soil in steel column

 Constant flow rate of 10 mL min−1

 Temperature = 25°C

 Pore volume = 80 mL

 No of pore volumes = 6

Variables:

 Washing agent: Soapnut, Soapnut + Phosphate

 Flow modes: upflow (UF), downflow (DF)

 pH: 5, 6

 Substrate: solution, CGAs

Concentration of wash solutions:

 Soapnut = L = SN 0.5 %, LM = SN 0.75 %, M = SN1 %,  MH = 1.25 %, H = SN 1.5 %

 Soapnut + Phosphate: L = SN 1 % + Ph 50 mM, LM = SN 1 % + Ph 75 mM, M = SN 1 % + Ph 100 mM, MH = SN 1 % + Ph 125 mM, H = SN 1 % + Ph 150 mM)

Results and Discussion

Stability of CGAs: Separation of CGAs from the liquid phase with respect to time can be measured from the movement of CGAs-liquid front and CGAs half-life (t1/2). Ph + SN CGAs were found to have the longest half-life among all extractants followed by soapnut and SDS. At low strength, i.e. 0.5 % soapnut- 50 mM phosphate, the CGAs had half-life of 900 s and at 1 % soapnut-100 mM phosphate concentration, it increased to 1200 s. Ph + SN CGAs had air hold-up of 35 %–39 % v/v and they were kept under constant stirring with magnetic stirrer at 1000 rpm to maintain homogeneity.

Arsenic removal performances: A column experiment for soil washing is influenced by a number of factors such as concentration of surfactant and physical state of surfactants, soil pH and flow modes of washing agents. Figure 1 summarises the data of cumulative arsenic removal in 6 pore volumes by solutions and CGAs under some selected experimental conditions. From the Fig. 1a, it can be seen that, water removed only up to 25 % arsenic at soil pH 5 in up flow mode signifying the necessity of adding other reagents. Solutions and CGAs generated from 1 % soapnut and 1 % soapnut + 100 mM phosphate removed up to 71 % more arsenic than water. Lastly, 20 mM SDS solutions and CGAs removed up to only 46 % arsenic. A higher pH 10 of 20 mM SDS did not favour arsenic solubilisation.
Fig. 1

a Comparison of arsenic removal performance by CGAs and solutions under different flow modes at pH5 and 6, b arsenic removal in up flow and down flow mode by solution and CGAs at pH 6 with 1 % soapnut, 1 % soapnut-100 mM phosphate and 20 mM SDS, c arsenic removal in different flow modes by Down flow and Up flow modes by CGAs and solutions with 1 % soapnut, 1 % soapnut-100 mM phosphate and 20 mM SDS at pH6

Arsenic removal by CGAs and solution of soapnut, Ph + SN and SDS at soil pH of 6 are illustrated in Fig. 1b. In down flow mode, 1 % soapnut solution removed more than 63 % of arsenic while CGAs removed only 41 % of arsenic. In up flow mode, CGAs removed 70 % arsenic while solution removed 76 % arsenic. For Ph + SN mixtures, CGAs performed better in up flow mode than solution, but fared slightly worse in down flow mode. In up flow mode, a highest arsenic removal of 93 % was achieved by Ph + SN CGAs where its solution showed 88.71 % arsenic removal. These results show that CGAs and solution had similar functional efficiencies. CGAs perform especially better in up flow modes than down flow mode due to higher buoyancy of the microbubbles on their introduction from the base of the column. Since CGAs entrap up to 35 % of air by volume, it is more economical than using surfactant solution of same volume. Addition of phosphate boosted the performance of soapnut CGAs from 70 % to 93 % in up flow mode and from 41 % to 78 % in down flow mode. The flow mode of wash solutions is another important factor in arsenic removal from soil column as shown in Fig. 1c. Two different wash modes, viz. down flow and up flow modes were used in this study. While in down flow mode, soapnut CGAs achieved up to 41.3 % arsenic removal, Ph + SN CGAs removed up to 78 % arsenic. The solutions, on the other hand, performed better under natural flow of gravity, the removal by soapnut and Ph + SN solutions being 63.63 % and 84.7 % respectively. It was observed that in down flow mode, tiny channels transmitting solutions and CGAs open up and close, by passing some parts of contaminated soil matrix. However, in up flow mode, this problem of flow channelling was not prevalent. In up flow mode, CGAs and solutions of soapnut removed 70 % and 76.21 % arsenic respectively and the corresponding values for Ph + SN mixture are 93 % and 88.7 %. Thus, up flow mode is found to perform better than down flow mode and phosphate addition definitely improved the performance of soapnut in both flow modes. Although Fig. 1c shows different flow modes at soil pH of 6, similar trends were observed for soil at pH 5. The concentration of washing solution and CGAs largely influenced the arsenic removal efficiency. In all cases, higher concentration solutions and CGAs remove more arsenic. Although the data for soil pH of 6 is shown, the trend is similar for soil of pH 5. Higher concentration solutions and CGAs contain more micelles compared to their lower concentration counter parts resulting in increased solubility of arsenic in the surfactant micelles. This observation is again supported by other research studies which reported that the number of micelles usually increase above CMC, resulting in enhanced mobility of arsenic, heavy metals and organic substances from soil (Mulligan 2005). Also, an increase in the arsenic removal was observed when only phosphate concentration was increased in 1 % soapnut solution. This is expected due to increase in the number of exchangeable phosphate molecules in the solution which can replace the arsenate from the soil (Alam et al. 2001; Zeng et al. 2008). Another notable feature that emerged from the experiments is that, the results were not influenced by variation of soil pH. In some cases, soil pH of 5 exhibited slightly higher removal of arsenic than soil pH of 6. However, the trends of arsenic removal under all the other conditions were similar for both soil pH values of 5 and 6. This signifies that soapnut can be used in conjunction with phosphate without altering the solution pH for treating slightly acidic soil.

The box-plot in Fig. 2 substantiates the data presented in Fig. 1 and portray the relative importance of different factors on arsenic removal. Figure 2a, c, e and g clearly show the advantage of up flow mode over down flow mode for both CGAs and solutions of soapnut and Ph + SN mixtures at five different concentrations. Figure 2b, d, f and h show the effect of soapnut and phosphate concentration on arsenic removal under up flow and down flow modes at soil pH values of 5 and 6. In Fig. 2b and d, the concentration of soapnut was increased in the order 0.5 %, 0.75 %, 1 %, 1.25 % and 1.5 %, respectively and the effect was found to be positive. For Fig. 2f, h soapnut concentration was kept constant at 1 % and phosphate concentration was increased in the order 50, 75, 100, 125 and 150 mM, respectively. The amount of arsenic removal increased with increasing phosphate concentration and the values were generally in higher range than the results shown in Fig. 2b, d which represent experiments for only soapnut as wash liquid. The solutions of soapnut and Ph + SN (Fig. 2b, f) experienced distinct rise in arsenic removal with increased concentration of phosphate. The variation with flow modes and soil pH were not significant. This is the due to the increased number of micelle in the soapnut and higher number of phosphate molecules that can replace arsenate in the solution. However, for CGAs (Fig. 2d, h), the improvement of arsenic removal had a broad variation under different conditions of flow modes and soil pH indicating that surfactant concentration affected CGAs performance less than the corresponding solution. In comparison, Fig. 2c, g show that flow modes have greater effect on CGAs performance than surfactant concentration.
Fig. 2

Variation in arsenic removal by solutions and CGAs at different saponin and phosphate concentrations under up flow and down flow modes at soil pHs of 5 and 6 (specific details are provided in the graphs). The Box-whisker plot represents maximum score, 75th percentile (Upper Quartile), Median, 25th percentile (Upper Quartile), Median, 25th percentile (Upper Quartile) and Minimum Score (DF: down flow; UF: up flow; pH 5 and pH 6 indicate pH of soil; for b, d: L = soapnut 0.5 %, LM = soapnut 0.75 %, M = soapnut 1 %, MH = 1.25 %, H = soapnut 1.5 %; for f, h: L = soapnut 1 % + phosphate 50 mM, LM = soapnut 1 % + phosphate 75 mM, M = soapnut 1 % + phosphate 100 mM, MH = soapnut 1 % + phosphate 125 mM, H = soapnut 1 % + phosphate 150 mM)

Mechanistic studies and soil corrosion: Desorption kinetic studies were performed using soapnut in combination with phosphate. The linear forms of the kinetic models e.g. Elovich equation, two-constant rate equation, parabolic diffusion equation and first order equation were explored. Elovich equation with highest R2 and lowest SE values of 0.969 and 0.017 describes desorption kinetics by Ph + SN solution appropriately. A decrease in “α” indicates a reduction in the desorption rate (Chien and Clayton 1980). Ph + SN has a higher value of α than soapnut, indicating its higher efficiency for desorbing arsenic.

Figure 3a demonstrates the FT-IR spectra of the influent and the effluent soapnut and Ph + SN solutions. Presence of carboxylic acid, phenolic OH and carbonyl groups in the soapnut extract supports similar findings reported earlier (Pradhan and Bhargava 2008). Carboxylic group gives saponin a mild anionic character at pH of 4.5. However, no shift of peaks in FT-IR spectra was detected for soapnut and Ph + SN solutions in absence or presence of arsenic, suggesting little chemical interactions. SEM image in Fig. 3b shows the surface morphology of the arsenic contaminated soil before and after treating with 1.5 % soapnut and 1.5 % soapnut + 150 mM phosphate solution at 1000× magnification. The SEM image of the soapnut washed soil reveals that much of the finer grains in the original soil are absent, exposing the underlying smooth surface. Ph + SN washing resulted in slight corrosion of the soil surface.
Fig. 3

a FT-IR spectra of influent and effluent wash solutions; b SEM micrographs of soil matrix before and after washing with soapnut and Ph + SN solution

Phosphate had a significant effect on soapnut solutions and CGAs for removal of low level arsenic residues from soil. Longer half-life, slower rate of liquid-bubble separation and higher air-holdup make Ph + SN CGAs highly suitable for soil washing. In general, CGAs comprising up to 35 % of its volume of air showed similar soil washing performance with solutions, indicating economic advantage. High arsenic removals up to 93 % were achieved by Ph + SN CGAs in up flow mode at soil pH of 6. Soapnut CGAs removed only 70 % arsenic in up flow mode at pH 6. This indicates a 33 % increase in efficiency of soapnut on phosphate addition. Phosphate replaced arsenate from soil due to its similarities in pK1, pK2, pK3 values and is found to be an advantage during competitive adsorption–desorption phenomenon (Mukhopadhyay et al. 2015a). SEM image of the washed soil reveals only minor corrosion of the soil particles in presence of Ph + SN washing agent. Therefore, Ph + SN solutions and CGAs are safe, efficient and environment friendly means of remediating arsenic-contaminated soil. The presence of phosphate in the natural soil is likely to have a synergistic effect on the performance of soapnut.

Notes

Acknowledgments

The authors acknowledge the funding provided by University of Malaya, Kuala Lumpur (Project UM-QUB6A-2011) for carrying out this research (Brookins 1986).

References

  1. Alam MGM, Tokunaga S, Maekawa T (2001) Extraction of arsenic in a synthetic arsenic-contaminated soil using phosphate. Chemosphere 43:1035–1041. doi:10.1016/s0045-6535(00)00205-8 CrossRefGoogle Scholar
  2. Bhatia D, Goel G, Bhimania SK, Bhaskarwar AN (2005) Characterization and drainage kinetics of colloidal gas aphrons. AlChE J 51:3048–3058CrossRefGoogle Scholar
  3. Brookins DG (1986) Geochemical behavior of antimony, arsenic, cadmium and thallium: Eh-pH diagrams for 25 C, 1-bar pressure. Chem Geol 54:271–278CrossRefGoogle Scholar
  4. Chien SH, Clayton WR (1980) Application of Elovich equation to the kinetics of phosphate release and sorption in soils. Soil Sci Soc Am J 44:265–268CrossRefGoogle Scholar
  5. DOE Malaysia (2009) Contaminated Land Management and Control Guidelines No. 1: Malaysian Recommended Site Screening Levels for Contaminated LandGoogle Scholar
  6. Hashim MA, Mukhopadhyay S, Sengupta B, Sahu JN (2012) Applications of colloidal gas aphrons for pollution remediation. J Chem Technol Biotechnol 87:305–324. doi:10.1002/jctb.3691 CrossRefGoogle Scholar
  7. Jang M, Hwang JS, Choi SI, Park JK (2005) Remediation of arsenic-contaminated soils and washing effluents. Chemosphere 60:344–354. doi:10.1016/j.chemosphere.2004.12.018 CrossRefGoogle Scholar
  8. Majumder A, Bhattacharyya K, Kole SC, Ghosh S (2013) Efficacy of indigenous soil microbes in arsenic mitigation from contaminated alluvial soil of India. Environ Sci Pollut Res 20:5645–5653. doi:10.1007/s11356-013-1560-x CrossRefGoogle Scholar
  9. Mukhopadhyay S, Hashim MA, Sahu JN, Yusoff I, Sen Gupta B (2013) Comparison of a plant based natural surfactant with SDS for washing of As(V) from Fe rich soil. J Environ Sci (China) 25:2247–2256. doi:10.1016/s1001-0742(12)60295-2 CrossRefGoogle Scholar
  10. Mukhopadhyay S, Hashim M, Allen M, Sen Gupta B (2015a) Arsenic removal from soil with high iron content using a natural surfactant and phosphate. Int J Environ Sci Technol 12:617–632CrossRefGoogle Scholar
  11. Mukhopadhyay S, Mukherjee S, Hashim MA, Sen Gupta B (2015b) Application of colloidal gas aphron suspensions produced from Sapindus mukorossi for arsenic removal from contaminated soil. Chemosphere 119:355–362. doi:10.1016/j.chemosphere.2014.06.087 CrossRefGoogle Scholar
  12. Mulligan CN (2005) Environmental applications for biosurfactants. Environ Pollut 133:183–198CrossRefGoogle Scholar
  13. O’Neill A, Gupta BS, Phillips DH (2014) Distribution of arsenic and risk assessment of activities on a golf course fertilised with arsenic-containing Ascophyllum nodosum seaweed. Sci Total Environ 482–483:252–259. doi:10.1016/j.scitotenv.2014.03.006 CrossRefGoogle Scholar
  14. Pradhan M, Bhargava P (2008) Defect and microstructural evolution during drying of soapnut-based alumina foams. J Eur Ceram Soc 28:3049–3057CrossRefGoogle Scholar
  15. Roy D, Kommalapati RR, Mandava S, Valsaraj KT, Constant WD (1997) Soil washing potential of a natural surfactant. Environ Sci Technol 31:670–675CrossRefGoogle Scholar
  16. Song S, Zhu L, Zhou W (2008) Simultaneous removal of phenanthrene and cadmium from contaminated soils by saponin, a plant-derived biosurfactant. Environ Pollut 156:1368–1370. doi:10.1016/j.envpol.2008.06.018 CrossRefGoogle Scholar
  17. Storer DA (1984) A simple high sample volume ashing procedure for determining soil organic matter. Commun Soil Sci Plant Anal 15:759–772CrossRefGoogle Scholar
  18. Tokunaga S, Hakuta T (2002) Acid washing and stabilization of an artificial arsenic-contaminated soil. Chemosphere 46:31–38. doi:10.1016/s0045-6535(01)00094-7 CrossRefGoogle Scholar
  19. Zeng M, Liao B, Lei M, Zhang Y, Zeng Q, Ouyang B (2008) Arsenic removal from contaminated soil using phosphoric acid and phosphate. J Environ Sci (China) 20:75–79. doi:10.1016/s1001-0742(08)60011-x CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  1. 1.Department of Chemical EngineeringUniversity of MalayaKuala LumpurMalaysia
  2. 2.Water Academy, Institute of Infrastructure and EnvironmentHeriot-Watt UniversityEdinburghScotland, UK

Personalised recommendations