Full factorial design on screening experiments for biosurfactant enhanced remediation of hydrophobic substances in soil L. Timma1, K. Sams2, S. Valtere3, J. Vilgerts4 and D. Blumberga5, 1,3,4,5Institute of Energy Systems and Environment, Riga Technical University, Riga, Latvia, 2“AURAVIA LATVIA” Ltd., Riga, Latvia
Abstract—this paper evaluates the cleaning efficiency of the glycolipid based anionic biosurfactant produced by the local company. Before use in the practical application of oil removal or other hydrophobic substances from soil, the knowledge on the behavior of the biosurfactants across different systems is necessary. As the process variables the temperature of environment, contact time with the dilution of biosurfactant and the concentration of the biosurfactant in a washing solution is be modeled by the application of full factorial design. As the response cleaning efficiency was obtained experimentally by a set of the laboratory tests. The screening design is employed for the evaluation of the interactions between the response variable and the process variables. The cleaning efficiency showed various results depending on the initial values of the variables. At the upper limit of the variables (+ 35 ºC for the temperature of environment, 15 minutes contact time with the dilution of the biosurfactant and 0.3 wt% concentration of the biosurfactant in a washing solution) the cleaning efficiency was 99.32 %. The results showed that all variable had significant effects on the cleaning efficiency at the 95 % confidence level. Index Terms—biosurfactant, cleaning efficiency, design of experiment, full factorial design, hazardous waste, soil washing.
I.
INTRODUCTION
Wide application, improper disposal and accidents connected to the oil product and other hazardous waste results in long-lasting contamination of the soil and subsurface environments. The contamination inevitably will affect ecosystems and human health [1]. Although considerable clean-up work is being conducted in situ and on site, the state of development among different technologies varies widely from the fundamental research to commercial products and processes. Those technologies use the physical, chemical and biological principles to remove and/or eliminate the contaminants from soil.
Manuscript received April 15, 2013. Biosurfactant enhanced remediation of hydrophobic substances in soil. L. Timma is with the Institute of Energy Systems and Environment, Riga Technical University, Riga, Latvia, LV-1010 (e-mail: lelde.timma@ rtu.lv). K. Sams is now with the “AURAVIA LATVIA” Ltd., Riga, Latvia, LV-1004 (e-mail:
[email protected]). S. Valtere is with the Institute of Energy Systems and Environment, Riga Technical University, Riga, Latvia, LV-1010 (e-mail:
[email protected]). J. Vilgerts is with the Institute of Energy Systems and Environment, Riga Technical University, Riga, Latvia, LV-1010 (e-mail:
[email protected]). D. Blumberga is with the Institute of Energy Systems and Environment, Riga Technical University, Riga, Latvia, LV-1010 (e-mail:
[email protected]).
One of the feasible ways to clean the contamination from the soil is bioremediation. The utilization of plants or microorganisms to convert pollutants to the carbon dioxide and water often is time consuming and is not cost effective for large quantities of the soil or high concentration of pollutants [2]. In the case of highly toxic substances (bacterocides) bioremediation is entirely impossible. One of the alternative methods is soil washing, which can be done faster and can be applied for large amounts of pollutant. The penetration of oil or other waste products in the soil depends on biological, chemical and physical factors in the soil, therefore involves complex mechanisms [3]. The hydrophobic nature of the oil products limits their availability for microorganisms, thus lowering the effectiveness of bioremediation processes [4]. In order to increase the rate of desorption, solubilization and to facilitate the bioavailability of hydrophobic pollutants, surfactants are used. Studies show that the synthetic surfactants tend to be toxic and resistant to biodegradation [5], [6]. As the result studies on the environmentally sound type of surfactants – biosurfactants increased exponentially, between 2000 and 2010 [7]. The biosurfactants can be composed biologically by fermentation, by the or cell growth of microorganisms and by bioconversion carried out via enzymes synthesized by bacteria or yeast (microbial cultivation) or from substances such as sugars, oils, alkanes, etc. (chemical synthesis) [8-11]. Bio-based surfactants are obtained also by transesterification of fats or derived from plants or animal fat via saponification [12]. The biosurfactants synthesized by microorganisms can be grouped into 6 major classes based on the producing microorganism: glycolipids, phospholipids, polysaccharide–lipid complexes, lipoproteins–lipopetides, hydroxylated and cross-linked fatty acids, and the complete cell surface [13]. In comparison with the synthetic surfactants the biosurfactants have better surface activity, lower toxicity, can bind heavy metals, have higher biodegradability and biological activity, are produced from renewable resources and can be reused by regeneration [14-17]. Other advantages of the biosurfactants is that there is no need to remove the biosurfactants from effluents during soil washing, as their release will not damage on environment [5], [9], [18]. There are a lot of studies on the wide range of application of the biosurfactants for a joint application of the biosurfactants and chemical-biological treatment [19], on the utilization of waste frying oils for the biosurfactant production [20], for bioremediation in a tropical climate by the application of a novel bacterial strain found indigenous for
the local environment [21], [22] on the identification of novel microorganisms within the petroleum contaminated sites [23-25]. In a former military base in Latvia remediation was succeed only by the way of a combination of low temperature desorption, steam stripping, membrane filtration, advanced oxidation [26], [27] and bioremediation [28]. During a site recovery partly destroyed acute toxic missile fuel and xylidine salts were isolated from the environment. The properties of the surfactants can be estimated by their net charge, structure and hydrophilic-lipophilic-balance value (HLB) [7]. Various types of characterization methods have been reviewed by [29], [30]. Apart from the chemical and physical characterization the performance indicators of the product from the biosurfactant includes: haptic properties of surfactant, foaming abilities, odor and color [7]. The cleaning efficiency of the surfactants’ solution depends on the parameters of the surfactant, surrounding environment and the properties of the pollutant. Therefore the study on the interaction between the various factors affecting the cleaning efficiency of the surfactants in classical ways is often expensive, time-consuming and performed under simplified conditions (for example, soil consisting of a single fraction). Besides some interactions have a non-linear nature [31], [32]. Under given conditions the application of statistical data analysis tools presents an effective way for the optimization of process parameters [31]. One of the first steps in the design of experiment is a screening design. The screening design is used before optimization and robustness tests are performed. Within the screening design interaction polynomials are used employing a factorial design [33]. The method of the factorial design requires few runs per investigated parameter, allows identifying influential process parameters without time consuming and costly tests. Also a regression model obtained presents the nature of the interconnection between process variables at high confidence level. The model can be upgraded to form composite designs. The results can be presented in 2- and 3-dimensional surfaces, thus the factorial design has great practical value at the early stages of a project [33]. The references on the application of factorial screening design in the field of the biosurfactants can be found in the area of process optimization for biodegradation of ethyl benzene and xylene compounds in mixtures [34], for efficiency increase in the removal of total petroleum hydrocarbons [31] and diesel [35], for the location of the most suitable temperature and surfactant concentration [36] and for the efficiency of enzyme recovery [37]. The bioremediation technology becomes even more sustainable in the case when the raw materials for the technological processes are not only renewable in the origin but also locally available and produced. This paper evaluates the cleaning efficiency of the glycolipid based anionic biosurfactant produced by the local company. Nevertheless before use in the practical application of oil removal or other hydrophobic substances from soil, the knowledge on the behavior of the biosurfactants across different systems is necessary. As the process variables the temperature of environment, contact time with the dilution of biosurfactant and the concentration of the biosurfactant in a washing solution is be modeled by the application of full
factorial design. As the response cleaning efficiency was obtained experimentally by a set of the laboratory tests. The screening design is employed for the evaluation of the interactions between the response variable and the process variables.
II. MATERIALS AND METHODS Within the research the interactions between 3 independent variables: the temperature of environment (°C), contact time with the dilution of the biosurfactant (minutes) and the concentration of the biosurfactant in a washing solution (% by weight (wt%)) was analyzed. As the response (dependent variable) cleaning efficiency (%) was obtained experimentally by a set of the laboratory tests. A. Full factorial design of experiments For this research a factorial design for experimental data was chosen, because the design allows to determinate the factors with the highest impact on a process. A full factorial design of 2k+k runs, where k is the number of variables, was selected for the screening design. The full factorial screening design involves runs at every possible combination at the upper and lower defined limit for each variable, see Table I. To the matrix of design additional centerpoints (the set of 3 experiments at the same conditions) were added, see Table I. TABLE I: THE DESIGN MATRIX OF THE EXPERIMENT Parameter
Upper
Lower
limit
limit
Centerpoint
Temperature, ºC
+ 35
+ 30
+ 32
Contact time, min.
15
5
10
Biosurfactant concentration, wt%
0.3
0.1
0.2
The upper and lower values of the variables have been chosen to mimic actual soil washing facility. As pointed out by the Reference [32] a time-efficient treatment test should take a couple of minutes rather than hours. In total 11 experiments were made (23 for the variables and 3 at the centerpoint). As the output of the factorial screening tests the key interactions between all variables are given. The resolution of the experimental data is set to 3-factor and higher order interactions between the variables. A cubic interaction between the variables has been set as the initial model for the process factors. The validity of the model is verified by the analysis of the values in the ANOVA table. The predication limit is set to 95.0 %. The interactive statistical data analysis tool STATGRAPHICS Centurion 16.1.15 was used to construct the model for the factorial screening design. B. Preparation of samples The experimental oily stain was composed from rapeseed oil (94 wt%), carnauba wax (5.5 wt%) and Sudan red 7B (0.5 wt%). The stain mixture was prepared by warming up rapeseed oil up to 100 °C then carnauba wax was added to the mixture. After the wax melted completely Sudan red 7B (dye content 95 % by ALDRICH Chemistry) was added in small quantities and mixed till complete dissolution. The total
volume of the stain mixture needed for all experiments (including calibration) was prepared at once. After the prepared stain mixture was cooled down to 50 °C, the mixture was applied on the stainless steel plates (40×20×3 mm) with a fine brush. The stainless steel plates have been weighed before the mixture was applied on them. The stain mixture was applied on the plates in several layers. In order to facilitate solidification of the stain mixture on the plates, a time interval was ensured between applications of each following layer of the mixture. The prepared plates were held in an exsiccator at room temperature approximately for 30 minutes until complete cooling and after weighted. C. Biosurfactant used The blend of a glycolipid based anionic biosurfactant with the commercial name AURA-PURE ST (produced by the company “AURAVIA LATVIA” Ltd., Latvia) was used in this research. A biosurfactant was purchased as a 5 wt% solution with the pH value of 10. The critical micelle concentration (CMC) was 0.1 % (a surface tension 27 mN/m) and the hydrophilic-lipophilic balance (HLB) was 10. D. Washing method After been weighted, each stainless steel plate was placed on the stand made from a stainless steel wire. The stand with a wire was used in order to avoid the mechanical damage of the stain (a contact between the stain surface and the walls of the beaker) while washing of the samples is performed. Than each plate was immersed in a shallow beaker (500 mL) filled with a 200 mL biosurfactant dilution. The biosurfactant dilution has been previously heated in a water bath (BWT-U, BIOSAN Ltd., Latvia) till a constant temperature (accordingly to the values for the “temperature” in the Table I) was reached. Immediately after a plate has been immersed the mechanical stirring (100 rpm) was applied using an overhead mixer equipped with a 2 folded propeller stirrers (MM-1000, BIOSAN Ltd., Latvia). The mechanical stirring time was altered for each test sample accordingly to the values for the “contact time” given in the Table I. After the washing plates were removed from the beaker and dried at room temperature. E. Analytical method Quantitative data about the cleaning efficiency was obtained through the interpretation of measurements done with the UV spectrophotometer (UNICAM HEλIOS γ). Before the experiments a measuring range and calibration curve for the UV spectrophotometer has been obtained. The calibration was performed for the concentration of the stain mixture from 0 to 0.15 wt%. After the washing with the surfactant and drying the plates were placed in a shallow form beaker (150 mL) and rinsed with 25 mL of 99 % chloroform (Sigma-Aldrich) for 5 minutes until the complete dissolution of the stain mixture was achieved. The amount (concentration) of the stain mixture in the chloroform solution was evaluated by the colorimetric method. The absorbance intensities at 415 nm were compared with the predefined calibration curve. The basic principles of for the experimental methodology are borrowed from the Reference [38].
III. RESULTS AND DISCUSSION A. Statistical validity of the model The results of data analysis show that the cubical interactions between the parameters have not been observed at the 95.0 % confidence level. Therefore the cubical interaction model can be simplified to the quadratic. The general form of the quadratic model is given by y 0 1x1 2 x2 11x12 22 x22 12 x1x2 ...
(1)
Where y is the predicted response, β0 is a constant, β1, β2 are linear coefficients, β11, β22 are second order coefficients, β12 is an interaction coefficient, x1, x2 are variables and ε is a constant representing the noise [33]. Estimates of the P-value for the quadratic model are > 0.05 therefore the model can describe experimental data at the 95.0 % confidence level. The same results about quadratic relationships between the process variables have been reported for the removal efficiency of the total hydrocarbons from the soil by a nonionic surfactant [31]. The R-Squared statistic indicates that the model as fitted explains 99.0845 % of the variability in the cleaning efficiency. The adjusted R-squared statistic, which is more suitable for comparing models with the different number of independent variables, is 96.9484 %. It is important to examine the Durbin-Watson statistic, since it can indicate the dependence between successive observations (a drift over the experimental data). Since the Durbin-Watson statistic is 2.53323 at P=0.5652, the serial autocorrelation in the residuals is not observed at the 5 % significance level. B. Pareto analysis The Pareto chart shows effects of each variable in the absolute values for thePareto cleaning efficiency, see Fig.1. Chart for Cleaning efficiency, % Contact time (B) Temperature (A) AB AA Surfactant conc.(C) AC BC
+ -
0
4
8
Effect
12
16
20
Fig. 1. The Pareto chart (an unstandardized form) for the cleaning efficiency.
The color of the bars depicts either an effect of the variable on the cleaning efficiency is a positive or negative. Fig.1 shows that the most influential variables are the temperature of environment and contact time with the dilution of the biosurfactant. The standardized Pareto chart gives the summary for the variables in order of the absolute magnitude. The effect of each variable is converted to the t-statistic by dividing the value of the variable by its standard error, see Fig.2.
Standardized Pareto Chart for Cleaning efficiency, % + -
Contact time (B) Temperature (A) AB AA Surfactant conc.(C) AC BC
biosurfactant, wt%. The revised model in the form of the standardized Pareto chart is given in Fig.4. Standardized Pareto Chart for Cleaning efficiency, %
Contact time (B)
+ -
Temperature (A)
0
2
4 6 8 Standardized effect
10
12
AB AA
Fig. 2. The standardized Pareto chart for the cleaning efficiency.
Surfactant conc.(C)
C. Half-normal probability plot Another test to validate the effects of the variables (excluding the impact of the degrees of freedom) is the half-normal probability plot for the variables, see Fig.3. An advantage of the probability plot is in the arrangement of the effects from the variables on the “noise” line (red line in the Fig.3). The more distant a variable is from the “noise” line, the greater effect on the cleaning efficiency with higher accuracy can be obtained.
0
90 85 80 30.0 35.0 5.0 15.0 0.1 0.3 Temperature, °C Contact time, min Surfactant conc., % Fig.5. The main effects plot for the cleaning efficiency.
Temperature (A) AB
0
AC BC 2
AA Surfactant conc. (C)
4 6 8 10 12 Standardized effects Fig.3. The half-normal probability plot for the cleaning efficiency.
The Fig.3 confirms the conclusions of Fig.1 and Fig.2 that 3 significant factors are present in the model. The antagonistic and synergistic interfaces between the variables have been revealed statistically by the Pareto plot and half-normal probability plot analysis. The results reviled that interaction between the concentration of the biosurfactant and other parameters (AC and BC in Fig.3) is below the 5 % significance level. D. Revised design of the model From the Pareto chart and half-normal probability plot for variables it is evident that the structure of the model can be simplified. The terms BC and AC are not statistically significant and are excluded from the further model. For the final version of the model following empirical equation for the cleaning efficiency is obtained: cl 1013.94 57.6467t 20.3275 18.487c 0.74825t 2 0.5659t
(2)
Where ηcl is the cleaning efficiency, %, t is temperature, °C, τ is contact time, minutes and c is concentration of the
The main effects plot shows how each of the analyzed responses changes within the boundaries of the stated upper and lower limit values. Fig. 5 shows that the increase in the absolute values for all variables stimulated the raise in the cleaning efficiency. Nevertheless, a steeper gradient for the temperature of environment and contact time with the dilution of the biosurfactant then for the concentration of the biosurfactant in a washing solution was observed. The main interaction within the model is observed between temperature and contact time, so the interaction plot is given in Fig.6. Interaction Plot for Cleaning efficiency, % 100
Cleaning efficiency, %
Standard deviations
1,6
0
15
95
Contact time (B)
0,4
12
100
2
0,8
6 9 Standardized effect
E. Main effects and interaction plots The empirical eq. (2) can be visually depicted by the main Main Effects Plot for Cleaning efficiency, % effect plot for the cleaning efficiency, see Fig.5.
Half-Normal Plot for Cleaning efficiency, %
1,2
3
Fig. 4. The standardized Pareto chart for the cleaning efficiency for the revised model.
Cleaning efficiency, %
The blue line drawn represents the statistically significant variables at the significance level of 5 %. As can be seen from Fig.2, increasingly significant is contact time and temperature as well as the interaction of the both variables. Below the 5 % significance level is the concentration of biosurfactant and all effects involving interaction with this variable.
90
Contact time, min=15.0 Contact time, min=15.0 Contact time, min=5.0
80 70
Contact time, min=5.0
60 30.0 35.0 Temperature, °C Fig.6. The interaction plot for the cleaning efficiency.
The interface between the temperature of environment and contact time with the dilution of the biosurfactant has non-linear nature, see Fig.6. The greater impact on the cleaning efficiency at the short contact times has the increase in temperature. The opposite also is true, at longer contact times, the influence of the temperature level losses
importance. The positive effect of elevated temperature on the soil washing has been reviewed also by the Reference [39]. F. Response plots At the upper limit of the variables (temperature of 35 ºC, 15 minutes contact time and 0.3 wt% concentration of biosurfactant) the cleaning efficiency obtained experimentally was 99.32 %. In the 3-dimensional space the empirical eq. (2) is shown in Fig.7.
Surfactant conc., %
Cleaning efficiency, % 80,0 85,0 90,0 95,0 100,0
0,3 0,26 0,22 0,18 0,14 0,1 30
13 15 9 11 31 7 32 33 34 35 5 Contact time, min Temperature, °C Fig.7. The response plot for the cleaning efficiency in a 3-dimensional space.
The validation of the tests leads to corroboration of several practical observations. Temperature and contact time plays a major role for the cleaning efficiency. The concentration of detergent solution also could be a relevant parameter, but the impact of the concentration usually is quite complicated. As the concentration of the surfactant increases, the monomers aggregate to form micelles. The concentration at which the micelles begin to form at the first time is known as the critical micelle concentration (CMC). The CMC concentration corresponds to the point where the surfactant shows the lowest surface tension for the first time too. Generally, the increase in temperature decreases the CMC of some non-ionic surfactants, but increases solubility of ionic surfactants. Many physical properties used to characterize surfactants depend on the CMC such as; emulsion formation, oil solubilization, foaming and detergency, interfacial and surface tensions. These properties may be used to assess the suitability of surfactant for the environmental bioremediation such as soil washing.
IV. CONCLUSION Instead of usually used the gravimetric assessment of the cleaning efficiency, a photo colorimetric method is recommended in this paper. The experimental tests showed high reliability for the assessment of degreasing and therefore are especially suited for exploration and optimization of different surfactants and their mixes. The experimental test is easy to set up, highly sensitive, and can be adapted to solve the problems encountered by formulators of detergent cleaners for better degreasing properties. These experiments suggest that utilizing surfactants to increase the solubility of dense organic pollutants can be an effective and relatively inexpensive way of ex situ remediation of contaminated soils and aquifers.
The screening design was employed for the evaluation of the interactions between the response variable and the process variables. The cleaning efficiency showed various results depending on the initial values of the variables. At the upper limit of the variables (+ 35 ºC for the temperature of environment, 15 minutes contact time with the dilution of the biosurfactant and 0.3wt% concentration of the biosurfactant in a washing solution) the cleaning efficiency was 99.32 %. The results showed that all variable had significant effects on the cleaning efficiency at the 95 % confidence level. The antagonistic and synergistic interfaces between the variables have been revealed statistically through the full Estimated Response Surface Mesh factorial design and by the application of the Pareto plot and half-normal probability plot analysis. The results reviled that the interactions between the concentration of the biosurfactant in a washing solution and the other parameters is below the 5 % significance level and can be removed from the regression model. The increase in the values for the all variables stimulated increases the cleaning efficiency. Nevertheless, the main effects plot shows a steeper gradient for the temperature of environment and contact time with the dilution of the biosurfactant then for the concentration of the biosurfactant in a washing solution. The main interaction within the model is observed between the temperature of environment and contact time with the dilution of the biosurfactant. The interface has a non-linear nature. The greater impact on the cleaning efficiency at the short contact times has the increase in temperature. The opposite also is true, at longer contact times, the influence of the temperature level losses importance. The next phase of the experiments will be the remediation of the polluted soil with the glycolipid based anionic biosurfactants.
REFERENCES [1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
S. Paria, “Surfactant-enhanced remediation of organic contaminated soil and water,” Advances in Colloid and Interface Science, vol. 138, pp. 24-58, 2008. C. C. Lai, Y. C. Huang, Y. H. Wei, and J. S. Chang, “Biosurfactant-enhanced removal of total petroleum hydrocarbons from contaminated soil,” Journal of Hazardous Materials, vol. 167, pp. 609-614, Jan 2009. M. S. Kuyukina, I. B. Ivshina, S. O. Makarov, L. V. Litvinenko, C. J. Cunningham, and J. C. Philp, “Effect of biosurfactants on crude oil desorption and mobilization in a soil system,” Environment International, vol. 31, pp. 155-161, Feb 2005. S. S. Cameotra, and R. Makkar, “Biosurfactant-enhanced bioremediation of hydrophobic pollutants,” Pure Applied Chemistry, vol. 82, pp. 97-116, Jan 2010. D. P. Sachdev, and S. S. Cameotra, “Biosurfactants in agriculture,” Applied Microbiology and Biotechnology, vol. 97, pp. 1005-1016, Feb 2013. R. Marchant, and I. M. Banat, “Microbial biosurfactants: challenges and opportunities for future exploitation,” Trends in Biotechnology, vol. 30, pp. 558-565, Nov 2012. M. M. Müller, J. H. Kügler, M. Henkel, M. Gerlitzki, B. Hörmann, M. Pöhnlein, C. Syldatk, and R. Hausmann, “Rhamnolipids—Next generation surfactants?” Journal of Biotechnology, vol. 162, pp. 366-380, June 2012. C. N. Mulligan, R. N. Yong, and B. F. Gibbs, “Surfactant-enhanced remediation of contaminated soil: a review,” Engineering Geology, vol. 60, pp. 371-380, 2001.
[9]
[10]
[11]
[12]
[13] [14]
[15]
[16]
[17] [18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
N. Christofi, and I. B. Ivshina, “A review: Microbial surfactants and their use in field studies of soil remediation,” Journal of Applied Microbiology, vol. 93, pp. 915-929, 2002. S. Rupp, “Next-generation bioproduction systems: Cell-free conversion concepts for industrial biotechnology,” Engineering in Life Sciences, vol. 13, pp. 19-25, Sep 2012. M. Deleu, and M. Paquot, “From renewable vegetables resources to microorganisms: new trends in surfactants,” Comptes Rendus Chimie, vol. 7, pp. 641-646, June-July 2004. M. Henkel, M. M. Müller, J. H. Kügler, R. B. Lovaglio, J. Contiero, C. Syldatk, and R. Hausmann, “Rhamnolipids as biosurfactants from renewable resources: Concepts for next-generation rhamnolipid production,” Process Biochemistry, vol. 47, pp. 1207-1219, April 2012. G. Bognolo, “Biosurfactants as emulsifying agents for hydrocarbons,” Colloid Surf. A. Physico. Chem. Eng., 1999. T. Pekdemir, M. Copur, and K. Urum, “Emulsification of crude oil-water system using biosurfactant,” Process Safety and Environmental Protection, vol. 83, pp. 38-46, 2005. I. N. A. Van Bogaert, K. Saerens, C. De Muynck, D. Develter, W. Soetaert, and E. J. Vandamme, “Microbial production and application of sophorolipids,” Applied Microbiology and Biotechnology, vol. 76, pp. 23-34, Aug 2007. P. Vatsa, L. Sanchez, C. Clement, F. Baillieul, and S. Dorey, “Rhamnolipid biosurfactants as new players in animal and plant defense against microbes,” International Journal of Molecular Sciences, vol. 11, pp. 5095-5108, Dec 2010. E. Ron, and E. Rosenberg, “Natural roles of biosurfactants,” Environmental Microbiology, vol. 3, pp. 229-236, Apr 2001. K. Urum, and T. Pekdemir, “Evaluation of biosurfactants for crude oil contaminated soil washing,” Chemosphere, vol. 57, pp. 1139–1150, July 2004. M. Viisimaa, O. Karpenko, V. Novikov, M. Trapido, and A. Goi, “Influence of biosurfactant on combined chemical–biological treatment of PCB-contaminated soil,” Chemical Engineering Journal, vol. 220, pp. 352-359, Jan 2013. S. D. George, and K. Jayachandran, “Production and characterization of rhamnolipid biosurfactant from waste frying coconut oil using a novel Pseudomonas aeruginosa,” Journal of Applied Microbiology, vol. 114, pp. 373-383, Jan 2013. R. Das, and B. N. Tiwary, “Isolation of a novel strain of Planomicrobium chinense from diesel contaminated soil of tropical environment,” Journal of Basic Microbiology, Article in Press, Jan 2013. M. S. Rizi, A. A. Sepahi, and W. S. Tabatabaee, “Crude oil biodegradation by a soil indigenous bacillus sp. isolated from lavan island,” Bioremediation Journal, vol. 16, pp. 218-224, Nov 2012. G. Granzotto, P. R. Franco Marcelino, A. De Melo Barbosa, E. P. Rodrigues, M. I. Rezende, and A. L. M. De Oliveira, “Culturable bacterial pool from aged petroleum-contaminated soil: Identification of oil-eating Bacillus strains,” Annals of Microbiology, vol. 62, pp. 1681-1690, Dec 2012. M. Chandran, and M. A. Prabharkaran, “A study on isolation, extraction, purification and characterization of surfactin by Bacillus subtilis from oil contaminated soils,” BioTechnology: An Indian Journal, vol. 10, pp. 312-316, 2012. A. Saimmai, J. Kaewrueng, and S. Maneerat, “Used lubricating oil degradation and biosurfactant production by SC-9 consortia obtained from oil-contaminated soil,” Annals of Microbiology, vol. 62, pp. 1757-1767, 2012. H. Whittaker, D. Cooper, R. Whittaker, J. Avotins, V. Kokars, J. Malers, J. Millers, S. Valtere, Dz. Zarina, “Remediation Practice at a Former Missile Base in Latvia,” Environmental Contamination and Remediation Practice at Former and Present Military Bases, Series 2: Environmental Security, Kluwer Academic Publishers, vol. 48, pp. 267-276, 1997. J. Avotins, V. Kokars, J. Malers, S. Valtere, H. Whittaker, D. Cooper, R. Whittaker, “Contamination and environmental remediation of former missile sites near Liepaja,” Critical Reviews in Analytical Chemistry, vol. 28(2), pp. 97-100, 1998. E. Petrovskis, J. Avotins, S. Valtere, H. Whittaker, “Landfarming of Xylidine-Impacted Soil at a Former Soviet Military Base,” Proc. of the Fourth International Symposium In Situ and On-Site Bioremediation, New Orleans, p. 10, 1997. V. Walter, V. Syldatk, and R. Hausmann, “Screening Concepts for the Isolation of Biosurfactant Producing Microorganisms,” Biosurfactants Advances in Experimental Medicine and Biology, vol. 672, pp. 1-13, 2010.
[30] S. K. Satpute, A. G. Banpukar, P. K. Dhekephalkar, I. M. Banat, and B. A. Chopade, “Methods for investigating biosurfactants and bioemulsifiers: a review,” Review in Biotechnology, vol. 30, pp. 127-144, June 2010. [31] A. Kalali, T. Ebadi, A. Rabbani, and S. S. Moghaddam, “Response surface methodology approach to the optimization of oil hydrocarbon polluted soil remediation using enhanced soil washing,” Int. J. Environ. Sci. Tech, vol. 8, pp. 389-400, Spring 2011. [32] A. Uhmann, and T. J. Aspray, “Potential benefit of surfactants in a hydrocarbon contaminated soil washing process: Fluorescence spectroscopy based assessment,” Journal of Hazardous Materials, vol. 219-220, pp. 141-147, April 2012. [33] L. Eriksson, E. Johansson, N. Kettaned-Wold, C. Wikstrom, and S. Wold, Design of Experiments, Principles and Applications, 3rd revised and enlarged ed. Umea, Sweden: Umetrics, 2008, ch 3, pp. 53-100. [34] E. R. Rene, M. S. Jo, S. H. Kim, and H. S. Park, “Statistical analysis of main and interaction effects during the removal of BTEX mixtures in batch conditions using wastewater treatment plant sludge microbes,” Int. J. Environ. Sci. Tech., vol. 4, pp. 177-182, 2007. [35] L. Molina-Barahona, R. Rodrıguez-Vázquez, M. Hernández-Velasco, C. Vega-Jarquın, O. Zapata-Pérez, A. Mendoza-Cantú, A. Albores, “Diesel removal from contaminated soils by biostimulation and supplementation with crop residues,” Applied Soil Ecology, vol. 27, pp. 165-175, 2004. [36] D. Goswami, R. Sen, J. K. Basu, and S. De, “Surfactant enhanced ricinoleic acid production using Candida rugosa lipase,” Bioresource Technology, vol. 101, pp. 6-13, Jan 2010. [37] F. A. Hasmann, A. Pessoa, and I. C. Roberto, “beta-Xylosidase recovery by reversed micelles. Use of cationic surfactant,” Applied Biochemistry and Biotechnology, vol. 84-86, pp. 1101-1111, 2000. [38] A. Fitzner, and U. Aßmus, “Recommendation for the Quality Assessment of the Product Performance of All-Purpose Cleaners” Degreasing efficiency according to internal test closed to IKW guideline, Journal SÖFW, vol. 131, pp. 54-66, Sep 2005. [39] K. Elgh-Dalgren, Z. Arwidsson, A. Camdzija, R. Sjöberg, V. Ribé, S. Waara, B. Allard, T. von Kronhelm, P. A. W. van Hees, “Laboratory and pilot scale soil washing of PAH and arsenic from a wood preservation site: Changes in concentration and toxicity,” Journal of Hazardous Materials, vol. 172, pp. 1033-1040, July 2009.
