Sustain. Environ. Res., 20(5), 275-280 (2010) (Formerly, J. Environ. Eng. Manage.)

275

REMOVAL OF ANTHRACENE CONTAMINATED SOIL USING SOYBEAN OIL Fung-Hwa Chi,* Min-Her Leu and Ruei-Chin Lee Department of Environmental Engineering Kun Shan University Tainan 710, Taiwan

Key Words: Contaminated soil, desorption, PAH (polycyclic aromatic hydrocarbon), remediation, soybean oil ABSTRACT This study investigated the feasibility of remediation of polycyclic aromatic hydrocarbon (PAH) contaminated soils using commercial soybean oil. It has been demonstrated that vegetable oil can be used as a nontoxic, inexpensive, and environmentally-friendly solvent to remove PAH from soils. The result showed that more than 72% PAH was removed from both low (50 mg kg-1) and high (200 mg kg-1) PAH contaminated soils in 24 h. If double extraction was conducted, the desorption efficiency increased by 16% as compared to a single extraction. For the factor of soil particle size, the desorption efficiency in a fine-grain (less than # 200 mesh, 0.074 mm) soil was 93%, which is higher than that in large-grain soil (#140-200 mesh, 0.074-0.1 mm). The feasibility of soybean oil regeneration showed that more than 90% PAH was removed by activated carbon. It was estimated that 1 g activated carbon can adsorbed up to 6 mg of PAH from contaminated oil. Based on the results shown above, the vegetable oil had a great capability to remove PAH from soils, and the contaminated oil can be regenerated by activated carbon and reused to reduce cost of remediation. INTRODUCTION Vegetable oil was used to remediate soils contaminated by hydrophobic compounds. One of its gradient is fat which has a high affinity for polycyclic aromtaic compounds (PAHs). Comparing to other flushing solvents, vegetable oil is biodegredable, will not cause the secondary pollution, and can be treated as nutrients to microbes. In this study, soybean oil and sunflower oil were tested to understand their desorption efficiency (DE), or extraction efficiency for anthracene contaminated soils. After reacting with contaminated soils, the soybean oil was regenerated by activated carbon (AC) for reuse. MATERAILS AND METHODS Soil obtained from the vicinity of campus was wind dried and sieved through mesh #140 (0.1 mm) and #200 (0.074 mm). The texture of the soil is shown in Table 1. Anthracene (CAS # 120-12-7) was chozen as the representive of PAHs in this study (also shown in Table 1). PAHs are generated when coal, oil, gas, garbage and other organic material burned imcompletly. *Corresponding author Email: [email protected]

The soybean and sunflower oil used in this study can be bought from any groceries. The fatty composition of soybean oil (Table 1) has saturated fatty acids 19% (C16 and C18), linoleic acid 52% (C18:2), oleic acid 21% (C18:1), alpha linolenic acid 6% (C18:3). AC is the product of carbon contained material through the process of carboniztion and bactivation. The one used in this study was derived from coconut shells with BET surface area 765 m2 g-1, porosity 0.27, and diameter of carbon pore 1.9 nm. Prior to HPLC, the sample was cleaned using clean-up column packed with 26 × 1.25 cm silica gel. The analytical column used was Thermo C18, 4.6 × 250 mm with the eluent of acetonitrile:water 90:10 (v:v), flow rate 1 mL min-1. HPLC was coupled with fluorescence detector at the excitation/emission wavelength 251/403 nm. Preparation of contaminated soils was done by dissolving 10 mg of anthracene in 10 mL of ether, and then adding 1 or 4 mL of it into a bottle with 20 g of soil. After capped for 24 h, the cap was opened and evaporated for 24 h in a hood. The final concentrtion of the soil is 50 or 200 mg kg-1. Vegetable oil 20 mL was added into 20 g of the prepared soil and sit for 24 h. The bottle was covered with foil to prevent the

Sustain. Environ. Res., 20(5), 275-280 (2010)

276

Table 1. Physical and chemical properties of soils and anthracene Soil texture

foc (%)

Pore volume (cm3 g-1)

mesh #140-200 < mesh #200

0.39 0.45

0.39 0.45

Soybean oil

Acid value

Relative density

0.15

0.92

log Kowa Anthracene (C14H10)

4.4

Sand 88 86

Saponification value 189-195

Texture (%) Slit Clay 8 4 6 8

Composition (%) Si K > 78 2.2 > 76.8 2.6

Al 19.6 19.5

Unsaturated fatty acids (%) C18:1 C18:2 C18:3 21 52 6

Water (%) 0.1

Swb -1

(μg L )

Melting point (°C)

Boiling point (°C)

70

216

340

Fe 0.003 2

Saturated fatty acids (%) C16 C18 10 9 Structural formula

a

Octanol-water partition coefficient, Kow Water solubilization, SW

b

Table 2. Experiment conditions Concentration of anthracene in soil (mg kg-1) 200 50 200 200 200 200 200 200

Soil size

Type of vegetable oil

Soil/oil ratio (m/v)

Reaction Time (h)

< #200 mesh

Soybean oil

1:1

24

#140-200 mesh

Soybean oil Soybean oil Sunflower oil Soybean oil Soybean oil Soybean oil Soybean oil

1:1 1:1 1:1 1:1 1:2 1:2 20 mL + 20 mL

24 24 24 12 24 48 24 + 24

Table 3. Regernation of vegetable oil by activated carbon Anthracene concentration (mg kg-1) 200

Activated Carbon (g) 0.5 1.0 1.5 2.0

Soybeane Oil (mL) 20

300-3000a 2.0 PAH concentrations: 300, 400, 500, 600, 800, 1200, 1400, 1600, 2100, 2400 and 3000 ppm.

20

Reaction Time (h) 24

24

a

photo-degradation. The quantity of oil used, the time of reaction, the type of oil, the size of soil, and single or double extraction were studied to obtain the optimized parameters. Based on the concentration found in contaminated sites [1], two concentrations were set at 50 to 200 mg kg-1 (Table 2). Two fractions of soil were tested, the first one has the size between mesh #140 and #200 (0.074-0.1 mm) with foc 0.39%, and the other has the size less than mesh #200 (0.074 mm) with foc 0.45%. Sunflower oil and soybean oil were adopted to investigate the DE in three time periods, 12, 24 and 48 h. Two ratios of soil to oil 1:1 and 1:2 (m/v) were studied in single and double extractions. The DE or extraction efficiency was calculated from the concentration of anthracene extracted in soybean oil. The amount of AC needed to regenerate vegeta-

ble oil and its adsorption capacity were investigated. There are two sets of experients (Table 3). First, different amount of AC (0.5-2.0 g) were placed in 200 mg L-1 anthracene contaminated vegetable oil shacking for 24 h. Second, the vegetable oil with 300-3000 mg L-1 anthracene contacted with 2 g of AC. After the first extraction of anthracene from soil, only 10 mL the oil were recycled, another 10 mL remained in the soil, i.e., 10 mL of fresh vegetable oil is needed to be added in the second run to keep the ratio of oil to soil 1:1 (m/v). The fraction from the first run was filtered with 0.45 μm filter paper to remove the particles before the second run. RESULTS AND DISCUSSION The solubility of anthracene in soybean oil in

Chi et al.: Anthracene Revmoval by Soybean Oil

this study is up to 3000 mg L-1, which is 60 times higher than that in water. When soybean oil was added into anthracene contaminated soil in the ratio 1:1 (v/v), it showed a DE of 61% which is lower than expectation. Since the diffusion was the predominant mechanism, it took longer for hydrophobic compound transfer from the soil phase to the oil phase. Also, it was obeserved that the color of soybean oil changed from yellow to brown, indicating that soil organic matter (SOM) dissolved into oil phase and which favored the anthracene dissolution. The DE of soybean oil reached 63% in a contaminated soil with 50 ppm anthracene. The efficiency is lower than that found in using sunflower oil which has more than 95% of DE in a contaminated soil with 147 ppm anthracene in a gas production plant [2]. As it has been known the moisture of soil affects the DE. There is a need to compare the values of DE without neglecting the water content of soil samples. The discrepance exists between different researches is probably due to the content of organic matter, concentration of PAH, moisture, and the chartcteristics of soil. Pannu et al. [3] showed that the DE changed from 50 to 92% when using peanut oil. In summary, vegetable oils show efficient desorption of PAH from contaminated soils; the experiment conditions need to be optimized to have the best DE. Contaminant concentrations, soil characteristics, flushing conditions, oil type and quantity all have essential impact on the desorption of PAH in soil. The concentration of anthracene found in contaminated site can reach up to several hundreds of ppm [1]. The DE using soybean oil in the soils with two common found anthracene concentrations, 50 and 200 mg kg-1, were examined with the ratio of soil to oil 1:1 (w/v). After reacting for 24 h, the DE reached 72% for both soils (Table 4), indicating that the mechanism was more partition like and reached equilibrium after 1 d. The effect of the concentration of contaminant on DE was studied by Pannu et al. [3], they used peanut oil (2.5% m/m) in anthracene contaminated soil of

277

50-100 ppm, the DE was above 90%. Same oil was applied in soil with anthrancene concentration of 1000 ppm, the DE droped to 16%. This is probablly due to the saturation of anthracene in peanut oil. Our previous study showed that the DE of sodium dodecyl sulfate in the soil containing 500 ppm anthracene were 20% after reacting for 24 h, and 30 and 33% when using TX-100 and Tween 80, respectively, in a 100 ppm anthracene contaminated soil. These results demonstrated that vegetable oil has much better DE for PAH desorption from soil. The DE of soybean oil reached 72% for 200 ppm anthracene contaminated soil, i.e., 1000 mL soybean oil has extracted anthracene 149 mg, which is still far below the tested concentration of 3000 mg L-1 (Table 4). Two different sizes of soil were used to investigate the DE by oil, it was found that the part #140-200 mesh (0.074-0.1 mm) soil with 0.39% organic carbon and 88% sand showed 74% of DE. The part of the particle size less than #200 mesh with higher organic carbon content 0.45% and sand 86% showed a DE of 93% (Table 4). It is probably due to that not only some of the anthrancene also dissolved into oil phase together with the orgnaic matter, but also the particles with smaller size have higher surface area. There is no doubt that the affinity of organic carbon with oil is higher than with water. The PAH sorbed or partitioning in SOM will dissolve together into oil phase. The fractions of the two soils less than sand has only a difference of 2%, but the part with particle size less than #200 mesh has higher soil orgnic carbon. It has been known that humic acid and fulvic acid exist in silt and clay [4]. Since the color changed from yellow to brown, it can be expected that a large amount of SOM is dissolved in oil phase. Other researches showed that the dissolving of humic acid into oil increased the solubility of pyrene by 3 folds [5,6]. The DE using soybean oil in 200 ppm anthracene contaminated soil is 74%, and 80% for sunflower oil (Table 4). Both oils contain a similar fat concentration, and the DE should not make too much difference.

Table 4. Desorption efficiencies of soybean and sunflower oil with various conditions Concentration or percentage of anthracene Percentage of anthracene in soybean oil (%) dissolved in oil Anthracene concentration in soila (mg kg-1) 50 36 mg L-1 72 200 149 mg L-1 74 Soil sizeb #140-200 mesh 74% 26 < #200 mesh 93% 7 Vegetable oilb Soybean oil 74% 26 Sunflower oil 80% 20 a Soil (20 g) was spiked with 50 and 200 mg kg-1 of anthracene and extracted with 20 mL of soybean oil for 24 h. b Same as a with 200 mg kg-1 of anthracene.

Sustain. Environ. Res., 20(5), 275-280 (2010)

278

Table 5. DE of soybean oil with two soil:oil ratios and two anthracene concentrations

-1

Anthracene conc. sorbed in soil (mg kg )

measured data 150

Ratio of soil to oil Anthracene conc. in soil (mg kg-1)

140

1:1 (m:v) 50

200

1:2 (m:v) 50

200

Anthracene 72 74 75 73 desorption (%) a Soil (20 g) was spiked with 50 and 200 mg kg-1 of anthracene and extracted with 40 mL of soybean oil for 24 h.

130

120

110

10

15

20

25

30

35

40

45

50

hours

Fig. 1. The adsorption of anthracene sorbed by soil.

For example, there is no apparent difference of DE for using sunflower oil and soybeen oil to remediate the soil in a manufactured gas plant [2]. When sunflower oil was applied to contaminated soils, the quantity of oil used depends on the concentrations of PAH [7]. For high concentration of contaminated soils, more oil is needed to reach high DE. The mass transfer or diffusion between anthracene and oil determins the time to reach equilibrium. The following equation can be used to express the time-depending phenomana [2,8,9]. Ct = Ce (1－e-kt)

(1)

-1

where Ct (mg L ) is the concentration at time t, Ce (mg L-1) is the concentration at equilibrium, k (h-1) is the mass-transfer constant. Two concentrations of contaminated soils were studied, the values of the anthracene concentrations at different time were measured. In the begining, the concentration increased quickly and reached the plateau after 12 h (Fig. 1) and reached the equilibrium after 24 h. The time reach equilibrium is affected by the hydrophobicity of contaminants and the texture of soil. High hydrophobic compunds such as pyrene need longer equilibrium time [10], and for sandy loam with anthracene, it needs only 3 h to reach equilibrium in peanut oil [3,15]. In the ratio of soil to oil 1:1 (m/v), the DE was 72% for 50 mg kg-1 anthracene contaminated soil after 24 h reaction (Table 5). If increasing the ratio to 1:2 (m/v), the DE was 75%. There was no apparent increase with incresing oil amount. When the anthraxcene concentration was increased to 200 mg kg-1, the

DE are 73 and 76% for the ratio of 1:1 and 1:2, respectively, indicating that the amount of 1:1 (m/v) was adequate for the amount of anthracene in the soil. Compared to other solvents and equipments used to extract anthracene from soils [11], a longer extraction time was needed, but it is compensated for the small quantity of oil and the cost. Single extraction and double extraction were conducted to compare the DE of soybean oil. At soil to oil raio 1:2 (m/v), 20 g soil with 40 mL oil, the DE were 73 and 75% for 24 and 48 h, respectively. If 40 mL of oil was separated into two parts equally, and the second part was added after 24 h after the first part was drained, the total reaction time was 48 h. The final DE increased to 84% (Table 6). It was found that the double extraction (consecutive desorpiton) has about 10% more DE than that of single extraction. The mechanism of diffusion is based on the concentration gradient between two phases. It is expected to have higher diffusion efficiency when changing solution to a clean one. Actually, the first 20 mL oil has shown 68% of DE, the second 20 mL oil increased another 16%. However, the 16% is more than 50% of the anthracene left in soil after the first extraction. A similar experiment was done by Pannu et al. [3], a double extraction in a weathered soil from a creosotecontaminated site using 5% peanut oil, followed by another 5% oil was better than a single 10% oil extraction, the DE increased about 8%. When 200 ppm anthracene-containing oil reacted with different quantity of AC (0.5-2.0 g), it was found that 68 and 92% of DEs reached when using 0.5 and 1.0 g of AC, respectively (Fig. 3). If more than 1.5 g AC was used, the removal efficiency was great than 95%. Others [2,12] also found that AC can efficiently remove the PAH in sunflower oil, and the removal efficiency reached 80-90% even at a high PAH concentration (5450 mg L-1). It was calculated from the isotherm experiments that the adsorption capacity of AC for anthracene is about 1.3-6 mg g-1. Fitting the experiment data to the Langumuir equation, the maxi-

Table 6. Comparison of DE in single and double extraction Ratio of Soil to Soybean oil 1:2 (m/v) 40 mL (24 h) 40 mL (48 h) 20 mL (24 h) + 20 mL (24 h) Anthracene conc. in oil (%) 73 75 68 16 a Soil (20 g) was spiked with 200 mg kg-1 of anthracene and extracted with 40 mL of soybean oil for

Chi et al.: Anthracene Revmoval by Soybean Oil

-1

Anthracene loading (mg g )

6

5

4

3 Experimental Freundlich isotherm Langmuir isotherm

2

1

0

10

20

30

40

50

60

70

80

-1

Equlibrium concentration of Anthracene (mg L )

Fig. 2. Two equations were used to depict experimental data.

279

200 mg kg-1 anthracene-contaminated soil did not decrease. The DE for the fresh soybean oil and the firsttime regenerated oil were 74 and 79%, respectively. The second-time and third-time regenerated soil have the DE of 79 and 85% (Fig. 4). The increasing of DE may due to the dissolved organic matter which enhanced the solubility of anthracene in the soybean oil. Compared to the reuse of surfactant for soil remediation, there is an apparent decrease in the concentration of the micelle after reacting with soils. The concentration of micelle needs to be adjusted to the designed value before using [13]. Thus, it is easier to use oil by adding the preset amount of oil. Furthermore, the amount of the soybean oil remained in the remediated soil will not result in the secondary pollution, but it can be ultilized as nutrients which favor the biodegradation of PAH [14-16]. CONCLUSIONS

100

Anthracene removal (%)

4 80 3 70 2 60

50 0.0

Anthracene removal Activated carbon adsorption 0.5

1.0

1

1.5

2.0

-1

5

90

Activated carbon adsorption (mg g )

6

0 2.5

Activated carbon dosage (g)

Fig. 3. Extraction efficiencies of AC from 200 ppm anthracene- containing soil.

mum content of adsorption is 6.3 mg g-1 (Fig. 2). Using 2.0 g of AC to adsorb anthracene in the range between 300-3000 mg L-1, the removal efficiencies all reached greater than 97%, and the capacity of sorbed anthracene increased from 3 to 29 mg g-1. It has been reported that the adsorption capacity of sunflower oil for anthracene was 48 mg g-1 [12]. The capacity of AC depends on the surface area, pore size distribution and surface chemistry. Both Langumuir and Freundlich equation could be used to describe the sorption data (Fig. 2). The mechanism of anthracene removal by AC is close to adsorption and not partitioning. The anthracene remaining in regenerated vegetable oil after contacted with 2 g of AC was 0.026 mg mL-1, and they were 0.058 and 0.077 mg mL-1 for the second and the third time regeneration, respectively. The possible reason to the increase of anthracene in oil is that more and more organic matter dissolved in the soybean oil, and it increased the apparent solubility of anthracene. The DEs of multiple-time regenerated oil for the

Vegetable oils provide another phase where hydrophobic compounds could partition into oil. Soybean oil and sunflower oil are commonly used in our daily life, they are environmental friendly, costeffective, and have high affinity for PAHs. The residual left in the pore of soil can be used as the nutrients to enhance biodegradation. It was found that the DEs for both 50 and 200 ppm anthracene-contaminated soils were above 72%. There was a higher DE for soil size less than mesh #200 (0.074 mm), this is probablly due to the dissolved SOM in oil and the higher BET surface area. Since the fat concentration in both soybean oil and sunflower is similar, it is expected to obtaine the similar DEs. The DE for both ratios of 1:1 and 1:2 (m/v) were similar. The ratio of 1:1 was sufficent. If the method of double extration was used, the DE incrased about 10% of the single extraction. AC efficiently removed the PAH in sunflower oil, and the DE reached 80-90% even at a high PAH concentration. In addition, it is interesting to note that the second or third time regenerated oil even has higher DE, possiblly due to more and more organic matter dissolved in oil and enhanced the solubility of anthracene in oil. ACKNOWLEDGMENT The authors thank the National Science Council of the Republic of China (Taiwan) for financially supporting this research under Contract No. NSC 972221-168-011. REFERENCES 1. Zemanek, M.G., S.J.T. Pollard, S.L. Kenefick and S.E. Hrudey, Multi-phase partitioning and cosolvent effects for PAHs in authentic petroleumand creosote-contaminated soils. Environ. Pollut.,

280

Sustain. Environ. Res., 20(5), 275-280 (2010)

98(2), 239-252 (1997). 2. Gong, Z., K. Alef, B.M. Wilke and P. Li, Dissolution and removal of PAHs from a contaminated soil using sunflower oil. Chemosphere, 58(3), 291-298 (2005). 3. Pannu, J.K., A. Singh, O.P. Ward, Vegetable oil as a contaminated soil remediation amendment: Application of peanut oil for extraction of polycyclic aromatic hydrocarbons from soil. Process Biochem., 39(10), 1211-1216 (2004). 4. Schnitzer, M. and H. Kodama, Interactions between organic and inorganic components in particle-size fractions separated from four soils. Soil. Sci. Soc. Am. J., 56(4), 1099-1105 (1992). 5. MacKay, A.A. and P.M. Gschwend, Enhanced concentrations of PAHs in groundwater at a coal tar site. Environ. Sci. Technol. 35(7), 1320-1328 (2001). 6. Hwang, S., N. Ramirez, T.J. Cutright and L.-K. Ju, The role of soil properties in pyrene sorption and desorption. Water Air Soil Pollut., 143(1-4), 65-80 (2003). 7. Gong, Z., K. Alef, B.M. Wilke, P. Li and Q. Zhou, Removal of polycyclic aromatic hydrocarbons from manufactured gas plant-contaminated soils using sunflower oil: Laboratory column experiments. Chemosphere, 62(5), 780-787 (2006). 8. Woolgar, P.J. and K.C. Jones, Studies on the dissolution of polycyclic aromatic hydrocarbons from contaminated materials using a novel dialysis tubing experimental method. Environ. Sci. Technol., 33(12), 2118-2126 (1999). 9. Gong, Z., K. Alef, B.M. Wilke and P. Li, Influence of soil moisture on sunflower oil extraction of polycyclic aromatic hydrocarbons from a manufactured gas plant soil. Sci. Total Environ., 343(1-3), 51-59 (2005). 10. Isosaari, P., T. Tuhkanen and T. Vartiainen, Use of olive oil for soil extraction and ultraviolet degradation of polychlorinated dibenzo-p-dioxins

11.

12.

13.

14.

15.

16.

and dibenzofurans. Environ. Sci. Technol., 35(6), 1259-1265 (2001). Wang, W., B. Meng, X. Lu, Y. Liu and S. Yao, Extraction of polycyclic aromatic hydrocarbons and organochlorine pesticides from soils: A comparison between Soxhlet extraction, microwave-assisted extraction and accelerated solvent extraction techniques. Anal. Chim. Acta, 602(2), 211-222 (2007). Gong, Z., K. Alef, B.M. Wilke and P. Li, Activated carbon adsorption of PAHs from vegetable oil used in soil remediation. J. Hazard. Mater., 143(1-2), 372-378 (2007). Ahn, C. K., M.W. Lee, D.S. Lee, S.H. Woo and J.M. Park, Mathematical evaluation of activated carbon adsorption for surfactant recovery in a soil washing process. J. Hazard. Mater., 160(1), 13-19 (2008). Gong, Z., K. Alef, B.M. Wilke, M. Mai and P. Li, Assessment of microbial respiratory activity of a manufactured gas plant soil after remediation using sunflower oil. J. Hazard. Mater., 124(1-3), 217-223 (2005). Pannu, J.K., A. Singh and O.P. Ward, Influence of peanut oil on microbial degradation of polycyclic aromatic hydrocarbons. Can. J. Microbiol., 49(8), 508-513 (2003). Pizzul, L., M.S.P. Castillo and J. Stenström, Effect of rapeseed oil on the degradation of polycyclic aromatic hydrocarbons in soil by Rhodococcus wratislaviensis. Int. Biodeter. Biodegr., 59(2), 111-118 (2007).

Discussions of this paper may appear in the discussion section of a future issue. All discussions should be submitted to the Editor-in-Chief within six months of publication. Manuscript Received: November 3, 2009 Revision Received: March 21, 2010 and Accepted: April 4, 2010