Louisiana State University LSU Digital Commons
LSU Historical Dissertations and Theses Graduate School
1995 Remediation of Contaminated Soils Using a Plant Based Surfactant. Raghava Rao Kommalapati Louisiana State University and Agricultural & Mechanical College
Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses
Recommended Citation Kommalapati, Raghava Rao, "Remediation of Contaminated Soils Using a Plant Based Surfactant." (1995). LSU Historical Dissertations and Theses. 6115. https://digitalcommons.lsu.edu/gradschool_disstheses/6115
This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zceb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600 REMEDIATION OF CONTAMINATED SOILS USING A PLANT BASED SURFACTANT
A Dissertation
Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy
in
The Department of Civil and Environmental Engineering
by Raghava Rao Kommalapati B.Tech., Nagarjuna University, 1988 M.Tech., Kakatiya University, 1990 M .S., Louisiana State University, 1994 December 1995 UMI Number: 9618305
UMI Microform 9618305 Copyright 1996, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 This dissertation is dedicated to my parents, brother, two sisters and their families whose love, support, understanding, and encouragement inspired me to travel that additional mile to achieve a tomorrow that today was thought impossible. This dissertation is also dedicated to all the teachers who taught me everything and influenced my life. Lastly it is dedicated to all my friends and colleagues who helped me in overcoming the problems by their constant encouragement and support. ACKNOWLEDGEMENTS
I am indebted greatly to my advisor, Dr. Dipak Roy, for his inspiring
guidance, encouragement, and financial support during the entire period of my stay at
Louisiana State University. I would like to take this opportunity to extend my sincere
appreciation to Dr. D. D. Adrian, who always had time to listen to my problems and
help me. I thank sincerely Drs. Constant and Valsaraj for their valuable suggestions
and critical reviews of the dissertation and the manuscripts. Thanks are also due to
my other committee members Dr. Cruise and Dr. Lawson. Special thanks to Janet
Labatut in Civil and Environmental Engineering.
I would like to thank my long time colleagues and good friends in the department Dr. Pradeep Chaphalkar, Dean Muirhead and Andrew Jackson. Thanks are due to other fellow graduate students and friends who graduated previously. Special thanks are due to my dear friend Asha who was always there with me when I needed help, particularly during the critical stages of my work. Thanks are also due to my long time friends Ravindra, Teja, Yatindra, and Naresh.
Most importantly, I would like to thank my parents, Rangamma and Peraiah
Kommalapati, for believing in me and my abilities. They sacrificed their pleasure for my well being and higher education. I would also like to thank my two sisters and their families for their support and encouragement. I will fail in my duty if I don’t thank my brother Rama Koteswara Rao Kommalapati without whose inspiration, I would not be here today. He has been a constant source of support and motivation all through my life. TABLE OF CONTENTS
ACKNOWLEDGEMENTS ...... iii
LIST OF TABLES ...... vii
LIST OF FIGURES...... viii
ABSTRACT ...... xii
CHAPTER
1 INTRODUCTION ...... 1
2 LITERATURE R E V IE W ...... 7 2.1 General ...... 7 2.2 Surfactants in Remediation ...... 8 2.3 Colloidal Gas Aphron (CGA) Suspensions ...... 13 2.4 Natural Surfactant (Ritha) ...... 17 2.5 Bioenhancement in the Presence of Surfactant ...... 21 2.6 Test Organic Com pound ...... 23 2.7 Scope of the Present W o rk ...... 24
3 OBJECTIVES...... 26
4 MATERIALS AND METHODS ...... 27 4.1 Materials ...... 27 4.1.1 Sapindus mukorossi (Ritha) ...... 27 4.1.2 Sodium Dodecyl Sulfate ...... 27 4.1.3 Hexachlorobenzene ...... 29 4.1.4 S o i l ...... 29 4.1.5 Basal Salt M e d ia ...... 29 4.1.6 Heterotrophic M ed ia ...... 32 4.2 Experimental Methods ...... 32 4.2.1 Glassware ...... 32 4.2.2 Extraction of Fruit Pericarp into Different Solvents .... 32 4.2.3 Standard Method for Preparation of Natural Surfactant Solutions ...... 33 4.2.4 Surface Tension Measurements ...... 34 4.2.5 Viscosity Measurements ...... 34 4.2.6 Solubility of Hexachlorobenzene in Natural Surfactant Solutions ...... 35
iv 4.2.7 Total Organic C a rb o n ...... 35 4.2.S Chemical Oxygen Demand ...... 36 4.2.9 Organic N itrogen ...... 36 4.2.10 p H ...... 36 4.2.11 Absorbance Spectrum for Natural Surfactant Solution . . 37 4.2.12 Generation of Colloidal Gas A phrons ...... 37 4.2.13 Size Distribution of CGA Suspensions ...... 39 4.2.14 Stability and Quality of CGA Suspensions ...... 40 4.2.15 Soil Contamination ...... 40 4.2.16 Desorption of Hexachlorobenzene from S o il ...... 41 4.2.170ne Dimensional Soil Flushing Experim ents ...... 41 4.2.18Bioenhancement Studies Under Aerobic Conditions .... 43 4.2.19 Bioenhancement Studies Under Anaerobic Conditions . . 46 4.2.20Extraction of Hexachlorobenzene from Aqueous Surfactant Solutions ...... 48 4.2.21 Extraction of Hexachlorobenzene from Soil ...... 48 4.2.22Analysis of Hexachlorobenzene on Gas Chromatograph . 49 4.2.23 Statistical Analysis of the D a ta ...... 49
5 RESULTS AND DISCUSSION...... 51 5.1 Development of Method for Hexachlorobenzene A nalysis ...... 51 5.2 Preparation and Characterization of Natural Surfactant ...... 54 5.2.1 Natural Surfactant Solutions ...... 54 5.2.2 Chemical Oxygen Demand(COD) ...... 61 5.2.3 Total Organic Carbon (TO C ) ...... 62 5.2.4 Nitrogen and Phosphorous ...... 62 5.2.5 pH ...... 62 5.2.6 Empirical Form ula ...... 64 5.2.7 Quantification of Natural Surfactant Using UV Absorbance ...... 66 5.2.8 Effect of Sterilization ...... 70 5.2.9 Critical Micelle Concentration ...... 72 5.3 Generation and Characterization of Colloidal Gas Aphron (CGA) Suspensions ...... 76 5.3.1 Size Distribution Param eters ...... 77 5.3.2 Size Distribution ...... 78 5.3.3 Effect of Natural Surfactant Concentration on the CGA Size Distribution ...... 82 5.3.4 Effect of Electrolytes ...... 86 5.3.5 Comparison with CGAs Generated from Commercial Surfactants...... 88 5.3.6 Stability of CGA Suspensions ...... 90
v 5.4 Solubility of Hexachlorobenzene in Surfactant Solutions ...... 96 5.4.1 HCB Solubility in Natural Surfactant Solutions ...... 96 5.4.2 Effect of Natural Surfactant Sterilization on the Solubility of H C B ...... 102 5.4.3 HCB Solubility in SDS Solutions ...... 103 5.4.4 Comparison Between Natural Surfactant Solutions and Commercial Surfactant Solutions ...... 107 5.5 Batch Desorption Studies ...... 110 5.5.1 Desorption Studies with Natural Surfactant Solutions . . . I l l 5.5.2 Desorption Studies with SDS Solutions ...... 115 5.5.3 Comparison Between Natural Surfactant and SDS Solutions ...... 119 5.6 Application of Natural Surfactant Solutions to Soil Flushing .... 119 5.6.1Flushing of Soils Contaminated to Different Levels of HCB ...... 120 5.6.2 Comparison Between CGA Suspensions and Natural Surfactant Solutions ...... 127 5.6.3 Effect of Surfactant Concentration ...... 129 5.6.4 Pressure Build-up Across the Soil C olum ns ...... 130 5.6.5 Effect of Alternating the Flushing Media with Water . . . 133 5.7 Bioenhancement in the Presence of Natural Surfactant Solutions . . 139 5.7.1 Aerobic Bioenhancement Studies ...... 140 5.7.1.1 Effect of Hexachlorobenzene ...... 142 5.7.1.2 Effect of Nutrients ...... 145 5.7.1.3Effect of Natural Surfactant Concentration .... 148 5.7.2 Anaerobic Bioenhancement Studies ...... 150 5.7.2.1 Effect of Hexachlorobenzene ...... 153 5.7.2.2 Effect of Nutrients ...... 155 5.7.2.3Effect of Natural Surfactant Concentration .... 157
6 CONCLUSIONS ...... 161
7 RECOMMENDATIONS ...... 167
REFERENCES...... 169
APPENDIX ...... 185
VITA ...... 187
vi LIST OF TABLES
Table 4.1: Properties of the Surfactant, Sodium Dodecyl Sulfate ...... 28
Table 4.2: Physical Properties of Hexachlorobenzene (Montgomery 1990) ...... 30
Table 4.3: Physical and Chemical Characteristics of the S o i l ...... 31
Table 4.4. Treatments Used in the Aerobic Bioenhancement Studies 45
Table 4.5. Treatments Used in the Anaerobic Bioenhancement Studies . . . 47
Table 5.1: Recovery of Hexachlorobenzene from Natural Surfactant Solutions ...... 53
Table 5.2: Extraction of Fruit Pericarps of Soap Berry into Different Solvents: Residue Remaining and the Residue E xtracted 56
Table 5.3. Comparison of the Size Distribution of CGAs Generated with Natural Surfactant and Commercial Surfactants (Chaphalkar, 1994) 91
Table 5.4: Comparison of Continuous and Alternate Flushing ...... 136
vii LIST OF FIGURES
Figure 2.1 : Structure of (a) Air Bubble, (b) Soap Bubble and (c) Colloidal Gas A phron ...... 14
Figure 2.2: Photograph of the (a) Tree Sapindus Mukurossi and (b) the Fruits ...... 18
Figure 4.1: Schematic Diagram of the CGA Generator ...... 38
Figure 4.2: Schematic Diagram of the One Dimensional Soil Column Flushing Experiment ...... 42
Figure 5.1 : Variation of (a) Surface Tension and (b) HCB Solubility with Concentration for Water, Methanol and Ethanol Extracts .... 58
Figure 5.2 : Variation of pH with Concentration for Natural Surfactant Solutions ...... 63
Figure 5.3 : Absorption Spectra for Natural Surfactant Solutions of Several Concentrations ...... 67
Figure 5.4 : Correlation Between Natural Surfactant Concentration and Absorbance at (a) 252 nm and (b) 292 nm ...... 69
Figure 5.5 : Absorption Spectra for Sterile and Non-Sterile Natural Surfactant Solutions at (a) 0.5% and (b) 1.5% Concentration ...... 71
Figure 5.6 : Determination of CMC of Natural Surfactant Solutions by (a) Surface Tension and (b) Viscosity Measurements ...... 74
Figure 5.7: Typical Size Distribution Curves for CGA Suspensions Generated with (a) 0.1%, (b) 0.5%, and (c) 1.0% Natural Surfactant Solutions...... 79
Figure 5.8: Typical Variation in Size Distribution with Time for CGA Suspensions Generated with (a) 0.5% and (b)1.0% Natural Surfactant Solutions...... 81
viii Figure 5.9: Bimodal Size Distribution Curves for CGAs Generated with (a) 0.5%, (b)1.0%, and (c) 1.5% Natural Surfactant Solutions ...... 83
Figure 5.10: Effect of Natural Surfactant Concentration on the (a) 10 Percentile and (b) 90 Percentile Size of CGA Suspensions . . . 84
Figure 5.11: Effect of Natural Surfactant Concentration on the Sample Loading Parameter (Dv) ...... 87
Figure 5.12: Effect of the Presence of Salt on the (a) 10 Percentile and (b) 90 Percentile Size of CGAs Generated with 1 % Natural Surfactant Solution ...... 89
Figure 5.13: Typical Curve for Drainage of Liquid from CGA Suspensions . 93
Figure 5.14: Comparison of (a) the Stability and (b) Quality of CGA Suspensions Generated with Several Concentrations of Natural Surfactant and SDS ...... 94
Figure 5.15: Solubility of HCB in Natural Surfactant Solution: (a) Variation up to 25% and (b) Correlation Between Solubility and Concentration ...... 98
Figure 5.16: Effect of Steam Sterilization and Filter Sterilization on the Solubility of HCB in Natural Surfactant Solutions ...... 104
Figure 5.17: Comparison of HCB Solubilizing Capacities of Natural Surfactant Solutions and Commercial Surfactants as Reported by Jafvert et al. (1994) ...... 106
Figure 5.18: Solubility of HCB in Sodium Dodecyl Sulfate (SDS) Solutions ...... 109
Figure 5.19: Desorption of HCB from Soil with Natural Surfactant Solutions: Variation of (a) Aqueous Phase HCB and (b) Percent Desorbed 112
Figure 5.20: Desorption Isotherms for HCB in Natural Surfactant Solutions in Non-Linear Form ...... 114
Figure 5.21: Desorption Isotherms for HCB (a) Langmuir and (b) Freundlich 116
ix Figure 5.22: Desorption of HCB from Soil with SDS Solutions: Variation of (a) Aqueous Phase HCB and (b) Percent Desorbed ...... 117
Figure 5.23: Removal of HCB from Soil Columns Using (a) 0.5 % Natural Surfactant Solutions and (b) 0.5% CGA Suspensions ...... 121
Figure 5.24: Removal of HCB from Soil Columns Using (a) 1.0% Natural Surfactant Solutions and (b) 1.0% CGA Suspensions ...... 122
Figure 5.25: Concentration of Natural Surfactant in the Column Effluent . 126
Figure 5.26: Effect of Natural Surfactant Concentration on the Removal of HCB from Soil Columns (a) 1.6 mg and (b) 100 mg HCB/Kg Soil ...... 128
Figure 5.27: Pressure Build-up Across the Soil Columns for the Flushing Experiment ...... 131
Figure 5.28: Effect of Alternating Flushing Media with Water on HCB Removal from Soil Columns (a) 1% Natural Surfactant and CGAs and (b) 1% and 2.5% Natural Surfactant...... 134
Figure 5.29: Effect of Alternating Flushing Media with Water on the Pressure Build-up Across the Soil Columns ...... 138
Figure 5.30: Aerobic Bioenhancement in Natural Surfactant Solutions (a) Growth of Microorganisms and (b) Reduction in T O C 141
Figure 5.31: Aerobic Bioenhancement Studies (a) Effect of HCB (b) Effect of Nutrients on the Microbial G row th ...... 144
Figure 5.32: TOC After 350 Hours for Different Treatments Under Aerobic Conditions ...... 147
Figure 5.33: Effect of Natural Surfactant Concentration on the Growth of Microorganisms ...... 149
Figure 5.34: Anaerobic Bioenhancement Studies: (a) Growth in Natural Surfactant Solutions (b) Effect of HCB ...... 151
Figure 5.35: Effect of Nutrients on the Anaerobic Growth of Microorganisms (a) Cultures without HCB and (b) Cultures with HCB ...... 156
x Figure 5.36: TOC After 25 Days for Different Anaerobic Treatments .... 158
Figure 5.37: Effect of Natural Surfactant Concentration on the Anaerobic Growth of Microorganisms ...... 159 ABSTRACT
A plant based surfactant extracted from fruit pericarps of Sapindus mukurossi
(Ritha) is employed for the first time to investigate remediation of contaminated soils.
A method for preparing natural surfactant solutions was developed and the solutions were characterized followed by generation and characterization of colloidal gas aphron
(CGA) suspensions produced with these solutions. The variation of hexachlorobenzene
(HCB) solubility with natural surfactant solutions was estimated and desorption of HCB from soils was investigated with batch and one-dimensional column experiments.
Bioenhancement of soil microorganisms in natural surfactant solutions was studied under both aerobic and anaerobic conditions.
The empirical formula for natural surfactant solutions was found to be
(C26H3iOio)n with a critical micelle concentration (CMC) of 0.1%. CGA suspensions generated from natural surfactant have (i) comparable size distribution characteristics,
(ii) more stability and (iii) lower gas fraction than those generated with commercial surfactants. Natural surfactant concentrations beyond 1.5% were not suitable for generating CGAs due to the low quality of the suspensions. Solubility of HCB in natural surfactant solutions increased linearly with surfactant concentration beyond the
CMC. The mass of dry Ritha powder required to solubilize 1 mg of HCB in one liter was found comparable to several commercial surfactants. Natural surfactant solutions were able to desorb HCB up to 90% of the HCB solubility in the respective solutions in batch studies. Natural surfactant solutions performed more efficiently than CGA suspensions of similar concentration in recovering HCB from one dimensional soil
columns, with removals as high as 80% of the solubility of the respective solutions.
HCB recovery also increased with increasing surfactant concentration. However,
natural surfactant concentrations beyond 2.5% developed a high pressure drop across
the soil columns and resulted in termination of the experiment. Natural surfactant
solutions supported microbial growth and degraded to a considerable degree under both
oxygen rich and anoxic conditions. The presence of a chlorinated organic compound,
HCB, did not affect the growth and did not undergo any transformations under both
aerobic and anaerobic conditions. Significantly higher growth in nutrient amended cultures indicate that the mictrobial populations may be nutrient limited. CHAPTER 1
INTRODUCTION
Contamination of subsurface soils and groundwater formations is a pervasive environmental problem that has proven to be extremely difficult to remediate. Early efforts at groundwater clean-up were characterized as being costly, time consuming and ineffective (Knox et al. 1984). There are about 30,000 potentially hazardous waste sites in United States and 1208 of these sites are included on the Superfund, National
Priority List (US EPA 1988, and Olsen and Kavanaugh 1993). CERCLA, enacted in 1980 by the US Congress, authorizes EPA to identify and clean-up the abandoned hazardous waste sites. Conventional "pump and treat" technologies are among the most widely used systems for the remediation of contaminated groundwater. However, these systems require a long time to make significant reductions in the quantity of organic contaminants (Mackay and Cherry 1989). The removal of hydrophobic organic compounds (HOCs) from contaminated soils is handicapped by their very low solubility in water and the high interfacial tension (Hunt and Sitar 1988). In recent years there has been dramatic increases in the number of technologies being promoted for subsurface remediation. Many of the technologies represent simple innovations of existing procedures (Kim and Gee 1993).
When surfactants are used with the existing "pump and treat" systems, the performance may be enhanced to a significant level. The surface activity of the surfactants comes from their amphiphilic structure, meaning that their molecules contain one soluble and one insoluble moiety. There are several classes of surfactants,
1 namely anionic, cationic, non-ionic, and zwitterionic, with both positive and negative charges. The surfactant molecules tend to concentrate at the interfaces and thus lower the interfacial tension considerably. Another very fundamental property of surfactants
is the formation of aggregates known as micelles at concentrations beyond the critical micelle concentration (CMC). The two very important applications of surfactants are
(1) solubilization of contaminants in surfactant micelles and (2) mobilization of residual liquids by reduction of the capillary forces trapping the liquid droplets in the aquifer porous medium. Several researchers proposed their use in the remediation of the abandoned hazardous waste sites (Ellis et al. 1985, Nash 1987, Gannon et al. 1989,
Abdul etal. 1990, Clarke et al. 1991, Edwards et al. 1991, Fountain et al. 1991, Liu etal. 1991, Abdul etal. 1992, Palmer and Fish 1992, Darji 1993, Kommalapati 1994 and others). Cationic surfactants tend to adsorb to soil significantly and thus are not very suitable for remediation (Peters et al. 1992). The cationic surfactants can however be used to modify the aquifer material and thus contain the contaminants
(Burris and Antworth 1992). Anionic surfactants and non-ionic surfactants generally do not undergo significant adsorption (Sabatini et al. 1992, and Liu et al. 1992).
Moreover the anionic surfactants can be recovered and reused (Gannon et al. 1989).
Several synthetic surfactants are being used for remediation of subsurface soils at this time. Large quantities of these surfactants are being used in these processes, and frequently the surfactant in question is toxic and recalcitrant to biodegradation.
Also the production of these chemicals produce by-products which has to be disposed off safely. The surfactants left in the subsurface may influence the biodegradation of the organic compounds significantly for better or worse (Laha and Luthy 1991, 1992,
Breuil et al, 1980, Oberbremer et al. 1990, Rouse et al. 1994, and others).
Surfactant interactions with bacterial cells also seem to be important (Swisher 1987, and Rouse et al. 1994). Surfactant recycle is now being investigated in laboratories and a technology is yet to be developed for complex wastes at the actual hazardous waste sites (Gannon et al. 1989, Underwood etal. 1993a, 1993b, 1995, and Abdul et al. 1992). Although researchers have conducted a number of encouraging laboratory studies, none of the field studies have been highly successful (Nash, 1987, Fountain et al. 1991, Abdul et al. 1992, West and Harwell 1992). One of the main reasons for the failure of soil flushing by surfactants is that the surfactants clog the porous media, making the process less efficient.
There is a need for better alternatives such as natural surfactants and biosurfactants which are produced by plants and microorganisms, respectively. One of the plant based natural surfactants derived from Sapindus mukurossi , commonly known as Soap Nut or Ritha in the Indian sub-continent, seems to have a great potential in remediation of contaminated soils (Mandava 1994). This plant is grown in tropical and subtropical regions of Asia. The dry powder from the fruit pericarp of this tree is extracted into water and is used as a detergent to wash hair and delicate fabric such as wool and also as folk medicine. Even though an appreciable amount of work has been done on isolating the saponin fraction and identification of the individual chemical compounds of this fruit extract, none has been done to explore its application to soil remediation, which is very much warranted. This research is focussed on the use of a plant based natural surfactant (Ritha) for remediation of
contaminated subsurface soils.
Another innovative technology that has been shown to have potential for in-situ
soil flushing is the use of microfoam or Colloidal Gas Aphron (CGA) suspensions,
which are generated from surfactant solutions (Fugate 1984, Longe 1989, Roy et al.
1992b, Daiji 1993, Kommalapati 1994). CGA suspensions consist of 65% gas by
volume and therefore form a low density fluid. These bubbles do not coalesce easily
and are remarkably different from conventional soap bubbles in their stability and flow
through properties (Sebba, 1982). Two important considerations in the application of
CGA suspensions are: i) their colloidal size, resulting in a large surface area to volume
ratio and ii) the existence of a double film of surfactant encapsulating the gas that
retards the coalescence of the bubbles. The structure, fundamental properties and
possible applications of CGAs have been extensively studied by Sebba (1971, 1982),
Sebba and Barnett (1981), Longe (1989) and Chaphalkar (1994). The main application
of CGA suspensions are in three areas: (1) flotation (2) soil flushing and (3) in situ- bioremediation. CGA suspensions can be employed for separating hydrophobic organic compounds from aqueous waste streams, and they are found to be more effective than conventional sparged air or solvent sublation (Chaphalkar et al. 1994).
Soil flushing of residual non aqueous phase liquids and other hydrophobic organic compounds with CGA suspensions has been very promising ( Longe 1989, Roy et al.
1992, Darji 1993, and Kommalapati, 1994). CGA suspensions are able to mobilize colloids from the soil matrix since air-water interface in motion is very effective in transporting colloids in a porous media (Wan and Wilson 1992). CGA suspensions can
also be used to transport engineered microorganisms into the subsurface to enhance the
in-situ biodegradation of in the contaminated subsurface soils (Michelsen et al. 1984,
Michelsen et al. 1988 Jackson and Roy 1995). In addition, CGA suspensions can be
employed to transport nutrients and oxygen along with the microorganisms (Michelsen
1988).
This dissertation is directed at studying the application of a plant based natural surfactant for remediation of contaminated soils. Solutions made from the fruit pericarps of Sapindus mukurossi, commonly known as Ritha are used in this research.
This is accomplished by a series of four tasks. A method is developed for preparing surfactant solutions from dry powder, and the properties such as critical micelle concentration (CMC), empirical formula are determined. Solubilization of hydrophobic organic compounds by natural surfactant solutions is studied and compared with commercial surfactants. CGA suspensions are generated using natural surfactant solution, and the size distribution and stability of the suspensions are studied. The application of surfactant solutions and CGA suspensions to soil flushing of contaminated soils is investigated. Bioenhancement of soil microorganisms in the presence of natural surfactant solutions under both aerobic and anoxic conditions is evaluated. The effect of a chlorinated organic compound on the bioenhancement of soil microorganisms is also appraised.
The organization of the dissertation is in traditional book style. Chapter 1 gives the introduction to the problem. Chapter 2 reviews the literature on the application of The organization of the dissertation is in traditional book style. Chapter 1 gives the introduction to the problem. Chapter 2 reviews the literature on the application of surfactants for remediation of contaminated soils followed by application of colloidal gas aphron suspensions. The role of surfactants in the biodegradation of hydrophobic organic compounds is also reviewed. The test compound, hexachlorobenzene, is reviewed subsequently. Finally the scope of the present work is outlined. Chapter 3 lists the objectives of the dissertation. Chapter 4 describes the materials and methods.
Chapter 5 deals with the results and discussion. The results are separated into 4 sub sections, (1) Preparation and characterization of natural surfactant solutions; (2) The study of colloidal gas aphron suspensions generated from natural surfactant solutions;
(3) Application of natural surfactant solutions to soil flushing; and (4) Bioenhancement in the presence of natural surfactant solutions. Chapter 6 lists the conclusions of the study followed by recommendations for further work in Chapter 7. CHAPTER 2
LITERATURE REVIEW
2.1 General
Since its inception in 1986, the Superfund Innovative Technology Evaluation
(SITE) program had demonstrated 44 technologies through 1992 (US EPA 1993). Kim
and Gee (1993) made an excellent review on the available technologies for hazardous
waste treatment. Conventional pump and treat technology has been the most widely
used treatment system for the remediation of aquifers (U.S. General Accounting
Office, 1986). The contaminated groundwater is extracted and treated at the surface.
The treated water may then be returned to the aquifer or discharged to surface water
bodies (Mercer et al. 1990). It is very well known at this time that the pump and treat
systems require long periods of time to make any significant reductions in the quantity
of contaminants associated with both the liquid and solid phases of the subsurface
matrix (Mackay and Cherry, 1989, Palmer and Fish 1992, Hoffman, 1993, and
others). Macdonald and Kavanaugh (1994) in their recent review reported that at 8 of
the 77 sites for which data was available, the clean-up goals were achieved using the
pump and treat systems. The report also reviewed the available innovative technologies, some of which are modifications of the existing pump and treat systems.
Chemical enhancements which can improve the rate of removal of contaminants from the subsurface can be a very economical alternative (Palmer and Fish 1992).
Surfactants have been widely recommended as a class of chemicals that can increase the rate of removal of contaminants from the aquifers by solubilizing significant amounts of the contaminants (Ellis et al. 1985, Nash 1987, Kile and Chiou 1989,
Gannon et al. 1989, Abdul et al 1991, Clarke et al. 1991, Edwards et al. 1991, Liu et al. 1991, and others). They can also be useful in removing non-aqueous phase liquids from aquifers by decreasing the interfacial tension between the contaminant and water (Fountain et al. 1991, Ang and Abdul 1991, Darji 1993, Kommalapati 1994 and others).
2.2 Surfactants in Remediation
The word surfactant is a contraction of the descriptive phrase, "surface active agent". Surfactants are in general composed of a hydrophobic moiety, often a long chain aliphatic (C10 to C20) group and a hydrophobic moiety that can be anionic, cationic, non-ionic, or zwitterionic (possesses both positive and negative charges). The surfactant molecules tend to concentrate at interfaces due to their amphiphilic structure and reduce interfacial tension (Rosen 1989). The reduced interfacial tension alters the wetting properties of the soil matrix and make surfactants useful in the enhanced oil recovery and the remediation of non aqueous phase liquids (NAPLs). Another important property of surfactants that is very useful in the remediation of contaminated soils is the formation of self aggregates called micelles beyond critical micelle concentration (CMC). Above the CMC, the surfactant molecules exist mainly as micelles and very few exist as individual amphiphiles with an equilibrium between the two (Void and Void, 1983). In aqueous solutions, the polar or ionic portions of the molecules are presented to the aqueous phase while the non-polar hydrocarbon tails of the molecules are clustered together away from contact with the water molecules.
These micelles may take a number of different shapes, although they are usually
roughly spherical in dilute solutions. The interior of a micelle consisting of the
hydrocarbon tails of the surfactant species is a non-polar phase and can dissolve
appreciable quantities of non-polar solutes which are virtually insoluble in normal
aqueous solutions (Rosen 1989). This process known as solubilization is very useful
in secondary oil recovery, cleaning and laundering, and micellar catalysis. The
amount of hydrophobic solute that can be incorporated into these surfactant micelles
can be many fold depending on the hydrophobicity of the compounds (Void and Void,
1983).
Considerable research has been done on the use of aqueous phase surfactants
for remediation of contaminated soils (Ellis et al. 1983, Nash 1987, Kile and Chiou
1989, Gannon et al. 1989, Abdul et al 1991, Jafvert and Heath 1991, Jafvert 1991,
Clarke etal. 1991, Edwards et al. 1991, Liu et al. 1991, Peters et al. 1992, Abdul et al. 1992, Edwards et al. 1994, Jafvert et al. 1994 Roy et al. 1994, 1995 and others).
The researchers reported that significant amounts of contaminants are solubilized by surfactants at concentrations above the critical micelle concentration. An extensive laboratory study was reported by Ellis et al. (1985) on the use of non-ionic surfactants for washing petroleum hydrocarbons, PCBs and chlorinated phenols from soils. They reported that the removal of PCB is not maximum at the maximum surfactant concentration and concentrations below the CMC are ineffective. There is some concentration at which the removal is maximum, which has to be determined from the 10 batch experiments. They noted that surfactant solutions are more effective than water
in removing hydrophobic organic compounds. It was suggested that leachate recycling
is necessary to conserve both water and surfactant and to reduce the cost of disposal
of the leachate. Kile and Chiou (1989) used 6 surfactants to study the water solubility enhancements of DDT and trichlorobenzene at concentrations above and below CMC.
Abdul et al. (1990) evaluated the relative suitability of 10 commercial surfactants for washing residual levels of automatic transmission fluid from a sandy material.
Edwards et al. (1991) determined the solubilities of polycyclic aromatic hydrocarbons
(PAHs) in micellar non-ionic surfactant solutions and found a linear correlation between the surfactant concentration beyond CMC and PAH solubility. They also reported that the partitioning of PAHs into aqueous surfactant solutions is linearly correlated with octanol/ water partition coefficient. Liu et al. (1991) studied the solubilization of PAHs in soil water systems and reported that the CMC of the surfactants in soil water systems will be more than the aqueous solutions due to the adsorption of surfactant by soil. Peters et al.( 1992) conducted surfactant screening experiments with 22 surfactants for mobilizing contaminants from the contaminated soil. Anionic surfactants resulted in the greatest degree of mobilization. Jafvert et al.
(1994) studied the solubility of hexachlorobenzene in several anionic and non-ionic surfactants. Valsaraj and Thibodeaux (1989), and Valsaraj et al. (1988) determined partition coefficients between micelles and water for several hydrophobic non-polar organics and correlated them with octanol/water partition constant. The experimental distributions of polycyclic aromatic hydrocarbons between sediment and soil and 11 aqueous phase containing an anionic surfactant sodium dodecyl sulfate were determined
(Jafvert and Heath 1991, Jafvert 1991 and Abriola et a l 1993). Concentrations of surfactants below the CMC do not effect the solubility of the organic compounds
(Valsaraj et a l 1989, Jafvert 1991, Liu et a l 1991). The ability of a surfactant to solubilize a hydrophobic compound from soil is dependent on (i) interaction of the compounds with the surfactant, (ii) sorption of the compounds on soil, (iii) sorption of surfactant on soil and its effect on increasing the wettability of soil and (iv) the partitioning of aqueous phase compound with the surfactant micelle (Liu et a l ,1991).
The presence of soil results in sorption of surfactant and hence the effective CMC will be greater than the aqueous phase CMC (Liu et a l 1991 and Vigoii and Rubin, 1989,
Pennell et a l 1993, Rouse et a l 1993, and Edwards et a l 1994). A significant portion of the contaminants in the soils are attached to the smaller sized particles or fines (i.e. silt, clay and humus). The addition of a surfactant or a chelant can enhance the effectiveness of washing from these fines, in particular surfactants increased the removal of organics from soil significantly (Esposito et a/. ,1988).
Wilson and his group worked extensively on soil clean-up by surfactant washing
(Clarke et a l 1991, 1993, Oma et a l 1991, Oma et a l 1993, Megehee et a l 1993,
Burchfield et a l 1994) and soil clean-up by in-situ surfactant flushing (Wilson 1989,
Wayt and Wilson 1989, Gannon et a l 1989, Wilson and Clarke 1991, Underwood et a l 1993a, and Underwood et a l 1993b, 1995). Their work includes all aspects of surfactant washing and in-situ flushing including solubility, washing, column flushing and mathematical modelling. They also reported encouraging results on the reuse of 12 anionic surfactant solutions after reclaiming the contaminant from the surfactant solutions. They also showed that the reclaimed surfactant was as effective as the virgin surfactant.
Nash (1987) tested the laboratory work of Ellis et al. (1985) at a field site contaminated with aromatic and chlorinated hydrocarbons and petroleum hydrocarbons.
The positive results obtained with soil columns at laboratory scale, however, were not substantiated by the field studies, which yielded ambiguous results. Abdul et al.
(1992) performed a field test of the surfactant washing at a site contaminated with
PCBs and oils and reported that the in-situ method is promising for the remediation of contaminated soil systems. Schmitt and Caplan (1987) reported a successful implementation of in-situ surfactant flooding along with in-situ bioremediation for clean-up of contaminated soils. The site was contaminated with petroleum hydrocarbons and the remediation scheme consisted of soil flushing with surfactant and aerobic biodegradation. They were able to remove 78.5% aliphatics and 90.1% aromatics. Clarke et al. 1991 demonstrated the washing of weathered-in PCBs from soil on a bench scale experiments. Surfactant recycle and reuse was studied at laboratory level using pure compounds by several researchers (Gannon et al. 1989 and
Underwood et al. 1993a, 1993b) and only anionic surfactants were found suitable for recycle. However, the actual wastes are complex mixtures of many compounds and the recovery of surfactant may not be feasible at the field scale. Moreover, the problem with the application of surfactant solutions is that the soil-surfactant interaction tend to clog the soil pores and result in the alteration of the hydraulic properties (Nash 13 1987, Liu 1993, Darji 1993, Kommalapati 1994). The preferential problem can however, be addressed by using surfactant solutions in the form of microbubble or colloidal gas aphron (CGA) suspensions (Darji 1993, and Kommalapati 1994).
2.3 Colloidal Gas Aphron (CGA) Suspensions
CGAs are micron size gas bubbles generated with a film of surfactant around them and are stable for hours. CGAs were first developed by Sebba (Sebba 1971) and were named as microfoams. A CGA contains about 65% of gas and is a class of
Kugelschaum foams. The CGA is typically 25-300 #im in size. These bubbles do not coalesce easily and are remarkably different from conventional soap bubbles in their stability and flow-through properties. Figure 2.1 shows the structures of (a) air bubble, (b) soap bubble and (c) CGA. As can be seen, the soapy film around the CGA has inner and outer surfaces with surfactants mono-layers adsorbed on them. The encapsulation retards the coalescence and improve the stability of bubbles significantly.
The CGA suspensions have viscosities similar to water, which make them suitable for pumping without the deterioration in quality. The present method of production of
CGAs has been developed by Sebba (1985a). A unit was developed in our laboratory by Chaphalkar et al. (1993) based on the design by Sebba (1985a).
There are number of potential applications for CGA suspensions in treating polluted soil and water. These can be grouped into three categories (a) flotation, (b) in-situ biodegradation and (c) soil flushing. The properties associated with CGAs widens the scope of technological application. CGA suspensions can be used to float 14
Water Air
Bulk Water Containing Surfactant (a) Air Bubble (b) Soap Bubble
Water Containing Surfactant
Bulk W ater Containing Viscous Water Surfactant "*“0 Surfactant Molecule
(c) Colloidal Gas Aphron
Figure 2.1 : Structure of (a) Air Bubble, (b) Soap Bubble and (c) Colloidal Gas Aphron 15 suspended particulates, bubble-entrained floes, and hydrophobic compounds from solution (Sebba and Barnett 1981, Sebba 1982, 1985b, Auten and Sebba 1984,
Honeycutt et al. 1983, Roy et al. 1992a, and Chaphalkar et al. 1994). CGA suspensions can be used to deliver oxygen and necessary nutrients for enhancement of biodegradation in saturated soil systems (Michelsen et al. 1984, 1985, 1988). Jackson and Roy (1995) used CGA suspensions to transport microorganisms through porous media and reported that CGA suspensions enhanced the transport of bacteria significantly over that of surfactant solutions and water. Application of CGA suspensions in soil flushing was reported by Fugate 1984, Longe (1989), Roy et al.
(1992b), Darji (1993), and Kommalapati (1994). Longe (1989) reported that the CGA is very effective in flushing a variety of hydrophobic organics from the soil and that
CGAs are more effective than surfactants at the same surfactant concentration.
Moreover, emulsification does not often occur with CGA flushing. Roy et al. (1992b) applied CGA suspensions and surfactant solutions for washing 2,4-D from soil and reported that the CGAs are more efficient than surfactants on a weight of contaminant per weight of surfactant basis. They suggested that CGAs will be more effective for hydrophobic compounds rather than hydrophilic compounds like 2,4-D. Darji (1993) used surfactant solutions and CGA suspensions to flush columns contaminated with hazardous oily waste from a Superfund site. The results indicate that the CGA suspensions are more effective than surfactant solutions in most of the cases and as effective in the remaining cases depending on the mode of operation. The main advantage of employing CGA suspensions is that the pressure build-up across the soil 16 columns was significantly lower compared to flushing with surfactant solutions. The surfactant-soil interactions and clogging of pores due to the colloids are thought to be the reasons. Kommalapati (1994) used automatic transmission fluid at residual levels for demonstrating the effectiveness of CGA suspensions over surfactant solutions.
Results are similar to those observed by Daiji (1993). These studies showed promising results and warrants further research. Chaphalkar (1994) studied the transport of CGA suspensions through soil columns using gamma ray densitometry and also modelled the transport process. He observed gas saturation levels as high as 90% in the columns when CGA suspensions are pumped into a initially water saturated column.
Although the commercial surfactants have shown good potential in terms of recovery of contaminants from the soils, their fate in the subsurface is still unknown.
Some of the synthetic surfactants are recalcitrant and are toxic themselves. When these surfactants are employed, they contaminate the aquifers increasing the load to the subsurface soils. Now the attention has been shifting towards biosurfactants which are produced by microorganisms (Wilson 1986, Falatko and Novak 1992, Desai and Desai
1993, Thangaman and Shreve 1994 and others). The studies indicate that these biosurfactants improve the hydrocarbon dispersion and bacterial attachment to the hydrophobic contaminants and thus enhance the solubility and increase the biodegradation rates of these hydrophobic compounds. Major classes of these surfactants include glycolipids, phospholipids and fatty acids, lipopeptide/ lipoproteins, polymeric surfactant and particulate surfactants. Another class of these surfactants, 17 produced from plants, known as natural surfactants also seem to have a great potential
for remediation (Mandava 1994).
2.4 Natural Surfactant (Ritha)
Another class of natural surfactants are derived from plants belonging to the genus Sapindaceae. These plants produce saponaceous substances called saponins, which form lather or foam in water. Sapindus mukurossi, Sapindus trifoliatus,
Sapindus laurifolius and Sapindus emarginatus are widely grown in India and Pakistan and other tropical and subtropical regions of Asia. Generally a mixture of these are sold in local markets (Gedeon 1954). About 56% of the fruit is pericarp and the remaining is seed. A picture of the tree and the fruits are shown in Figure 2.2. As can be seen from the photograph, the tree is of medium size with wide leaves. The fruits are golden brown in color and globular in shape with a diameter between 1 and
3 centimeters. These fruits are locally known as soap berry, soapnut or Ritha in the
Indian subcontinent.
The fruit pericarps of the plants belonging to the genus Sapindaceae have traditionally been used by man as soap substitute for fabric washing and bathing
(Oommachan, 1977, Uphof 1968, Bor 1953). This practice continues even today in
Asian countries despite the widespread use of commercial cosmetic shampoos. The recorded use of this product by man as washing soap does not site any toxic effects on human skin and eyes (Windholz 1983). The fruits of Sapindus mukurossi has also been used as a folk medicine (Kimata et al. 1983, Nakayama et al. 1986, and Kasai et al. Figure 2.2: Photograph of the (a) Tree Sapindus Mukurossi and (b) the Fruits 19 1988). It is used to treat epilepsy, chlorosis, and excessive salivation (CSIR, 1993).
Uppal and Mehta (1951) reported its utilization as a industrial textile auxiliary. Shetty
(1972) used Ritha as an air-entraining agent in the preparation of concrete.
Several researchers have isolated and identified the saponins from the fruit
pericarps of the plants of genus Sapindus (Sarin and Beri 1939, Uppal and Mehta,
1951, Gedeon 1954, Ranganna etal. 1963, Row and Rukmini, 1966a, 1966b, Kimata
et al. 1983, Nakayama et al. 1986, Kasai et al. 1988, Gupta and Ahmed 1990).
Saponins are complex substances and are essentially glycosides with their aglycones related to either sterols or triterpenes. The sterols consist of nitrogenin, getogenin, digitogenin and sarsasapogenin. The triterpenes consist of hedaragenin and oleanoloc acid (Karrer, 1950). Sarin and Beri (1939) reported that fruit pericarps contained
30.6% saponins, however, Gedeon (1954) quantified the saponin content using the method used by Sarin and Beri (1939) and determined the saponin fraction to be 10.1 % of the weight of the pericarp and 6.1% based on the weight of the nut. Sarin and Beri
(1939) used ethyl acetate to extract saponin from the pericarps. Uppal and Mehta
(1951) used water at the boiling temperature for the saponin extraction. Gedeon
(1954) also used boiling water in his work on large scale extraction. However, the extract does not contain saponins alone, but also all the water soluble substances such as gums, resins and proteins (Uppal and Mehta 1951, and Gedeon 1954). The water extract was cleaned further using several methods. Ranganna et al. (1963) also used water as the solvent in his large scale production of saponins. Row and Rukmini
(1966a) extracted pericarps of Sapindus mukurossi with water and later purified the extract for isolating the saponin, mukorosside (CS2Hm02i.2H20 ). They reported that saponin on hydrolysis with methanolic sulfuric acid gave hederagenin, D-glucose, D- xylose, L-arabinose and L-rhamnose. They also reported that D-glucose is the end sugar unit in glycoside. Acid hydrolysis furnishes O-dimethylhederagenin and a mixture of methylated sugars. Kimata et a l (1983) defatted the pericarps with hot benzene and the residue was extracted with hot methanol. The methanol extract after evaporating the solvent was chromatographed to isolate the components. They isolated several saponins, and named them mukurozi-saponins and an alphabet is used to differentiate each of the components. Shetty (1972) reported the saponification value of the fruit pericarp to be 138.2, saponification is the hydrolysis of an ester into the fatty acid and the alkali.
The saponins, although practically non-toxic to man upon oral injection, act as a powerful hemolytic when injected into the blood stream, dissolving the red corpuscles even at extreme dilutions (Windholz, 1983). However, the water extracts of the soap berry are mixtures of saponins and other gums and proteins and are believed to be safe to use. They are bitter in taste. The natural surfactant solutions can be handled with the usual care with which the other commercial surfactants are handled.
Extraction with water has been the most commonly used method both for scientific purposes and also for domestic uses such as for washing hair and fabric.
Fruit pericarps were soaked for several hours in water and the residue was separated.
This seem to be the simplest and most inexpensive method available. The water 21 solution thus obtained can be purified by several methods to isolate the saponin
fraction. However, the aim of the present investigation is to prepare natural surfactant
solutions with the least amount of energy expenditure. Preliminary experiments
conducted in our laboratory using natural surfactant solutions for remediation of soils
contaminated with naphthalene indicated that these solutions can desorb and solubilize
significant amounts of hydrophobic hydrocarbons (Mandava 1994). These solutions
can be utilized to generate colloidal gas aphron suspensions and for flushing
contaminated one dimensional soil columns (Mandava 1994). The optimum
temperature for preparing natural surfactant solutions is 28 °C and the extraction time
required is 3 hours. The preliminary work with natural surfactant solutions with
hexachlorobenzene as the solute also indicated that these solutions are comparable with
other commercial surfactants. The results are very encouraging and warrant further research.
2.5 Bioenhancement in the Presence of Surfactant
The reduced interfacial tension when surfactants are employed often results in the formation of emulsions thus increasing the surface area between immiscible liquids.
The micellar core of surfactants incorporates hydrocarbon compounds into solution.
These characteristics make surfactants a primary tool for the biodegradation of poorly soluble compounds (Breuil et al. 1980, Guerin and Jones 1988, Oberbremer et al.
1990, Aronstein et al. 1991, Laha and Luthy 1991, 1992, Edwards et al. 1992, Bury and Miller 1993, Rouse et al. 1995, and others). These studies have investigated the 22 role of surfactants in enhancing the biodegradation of the hydrophobic organic contaminants and reported enhancements, inhibitions and no apparent effect on the biodegradation of organic compounds in the presence of surfactants. Rouse et a l
(1994) made an excellent review on the influence of surfactants on microbial degradation of organic compounds. The role of surfactants has been primarily confined to their ability to increase the aqueous phase concentrations of the contaminants. The actual aqueous component of the compound apart from the micellar pseudo-phase may be greatly reduced under non-equilibrium conditions or when the hydrocarbon excess phase is depleted (Christian et a l 1985, and Rouse et al. 1994).
It has been shown that solubilization of hydrophobic organic compounds by surfactant micelles enhances biodegradation in pure bacterial cultures. But the degradative enhancement does not seem to hold for mixed populations in soil/sediment systems.
Biodegradation of hydrocarbons was inhibited by the use of non-ionic surfactants at levels above the CMC with mixed microbial cultures (Laha and Luthy, 1991, 1992).
They hypothesized that the inhibition might be due to interference with substrate transport into the cell or to reversible physical-chemical interferences with the activity of enzymes and other membrane proteins involved in the hydrocarbon degradation.
Mueller et al. (1990) however, reported an enhancement in the degradation of fluoranthene when the non-ionic surfactant Tween-80 was employed at concentrations above CMC. Aronstein et al. (1991) and Aronstein and Alexander (1992) reported an increase in the mineralization of phenanthrene and enhanced partitioning of the hydrocarbons to the aqueous phase was suggested as the reason for the increase. Van 23 Hoof and Rogers (1992) summarized their investigation with non-ionic surfactant on the biotransformation of HCB by reporting that micellar surfactant solutions in general suppressed the transformation, but at sub-CMC concentrations enhanced the dechlorination. Cationic and anionic commercial surfactants are noted for their damaging effects on cell membranes (Swisher, 1987). Tiehm (1994) reported that an increase in the concentration of an anionic surfactant sodium dodecyl sulfate (SDS) increased the inhibition of phenanthrene degradation.
The fate of the residual contaminants in the subsurface along with that of the surfactants is a crucial factor in deciding the remediation process and also the type of surfactant to be employed. There is a need to undertake an extensive study to ascertain the effect of the surfactants used in the remediation process on the biodegradation of the residual contaminants. In particular, the fate of surfactant in the subsurface needs to be investigated. The present work addresses some of these issues, but the main focus will be on the enhancement of soil microbial populations in the presence of natural surfactant solutions under both aerobic and anaerobic conditions. Preliminary toxicity and biodegradation experiments on natural surfactant with a strain of
Pseudomonas showed no significant inhibitory action towards biodegradation of 2,4- dichlorophenoxyacectic acid.
2.6 Test Organic Compound
Hexachlorobenzene (HCB) is an anthropogenic compound of much concern due to the large quantities being released into the environment, its extreme persistence, and 24 potential toxicity. HCB was used and still being used to a small extent as a fungicide.
However the present environmental concern is over the disposal of large quantities of
HCB produced annually as a by-product of several manufacturing processes. By far the largest quantities of HCB appear to be produced as the waste product of the chlorinated solvent industry (Quinlivan etal. 1976). In addition, significant quantities of HCB have been present as impurities or by-products in the production of certain pesticides. This is listed as a priority pollutant by EPA and is of particular interest in this work as it is a major contaminant at a local Superfund site.
Hexachlorobenzene is a stable persistent compound of low water solubility and moderate vapor pressure. HCB released to soil is likely to remain there for extended periods of time due to its strong adsorption to soil. A half life of 1530 days has been reported by Beck and Hansen (1974). It exists as a white powder at room temperature, with a solubility in water in the range of 5-110 pg/\ (Farmer et al. 1980, and
Montgomery 1990). HCB is soluble in several organic solvents such as benzene, hexane, chloroform and ether and also in fats and oils. Hence it tends to accumulate in the fatty tissues of animals. This compound may reasonably be anticipated to be a carcinogen. It irritates skin and may affect the liver, kidney and the reproductive system.
2.7 Scope of the Present Work
This study investigates the potential of using a plant-based natural surfactant, for remediation of contaminated soils. Fruit pericarps of Sapindus mukurossi obtained 25 from India are used for the research. A method for the preparation of natural surfactant solutions is developed. COD and TOC are measured along with pH, and an empirical molecular formula is derived. Surface tension and viscosity are measured and used for determining the CMC of the natural surfactant. The solubility of HCB in natural surfactant solutions are determined and compared with those of commercial surfactants available in the literature. Desorption studies on HCB in the presence of natural surfactant are performed and compared with SDS solutions. The performance of natural surfactant solutions in soil flushing process is investigated.
CGA suspensions are generated with natural surfactant solutions and their properties such as size distribution, and stability are determined. CGA suspensions are also used for flushing contaminated soils. The effect of alternating the CGA suspensions and surfactant solutions with water is studied. Bioenhancement of typical soil microorganisms in the presence of natural surfactant is evaluated under both aerobic and anaerobic conditions. The effect of a chlorinated aromatic hydrocarbon, HCB on the bioenhancement is also studied. The possible out come of this study to develop a solution for hazardous waste remediation using a natural product. CHAPTER 3
OBJECTIVES
The objective of this research is to study the application of a plant based natural
surfactant obtained from the fruit pericarps of Sapindus mukurossi in the remediation
of contaminated soils. Specific supporting objectives to achieve this goal are:
(a) Develop appropriate methods for preparing natural surfactant solutions and
characterize the solutions.
(b) Generate CGA suspensions using natural surfactant solutions and study the size
distribution, quality and stability of the CGA suspensions.
(c) Assess the ability of the natural surfactant to (i) solubilize the test compound,
HCB and (ii) compare the performance with commercial surfactants.
(d) Evaluate the effectiveness of natural surfactant solutions to desorb HCB from
soil and compare the performance to commercial surfactant, sodium dodecyl
sulfate (SDS).
(e) Determine and compare the efficiency of natural surfactant in the form of
conventional solution and CGA suspensions in eluting the test contaminants
from the laboratory soil columns.
(f) Study the effect of alternating the surfactant solutions and CGA suspensions
with water floods on the recovery of hexachlorobenzene.
(g) Study the bioenhancement of typical soil microorganisms in the presence of
natural surfactant under both aerobic and anaerobic conditions and also appraise
the effect of HCB on the bioenhancement.
26 CHAPTER 4
MATERIALS AND METHODS
4.1 Materials
4.1.1 Sapindus mukorossi (Ritha)
Dry fruits of Sapindus mukorossi were procured from the city of Calcutta,
India. Seed was removed from the fruit and only the outer pericarp was shipped from
India. About 10 kilograms of the fruit was obtained in a single batch to maintain homogeneity and consistency for the entire research work. Pericarps were dried in the oven at 50 °C for about 2 days. The pericarps were then ground in small batches in a coffee grinder and sieved through US Standard No. 20 sieve (840 pm). Batches of about 500 grams of the powder were prepared and stored in amber glass bottles with air-tight screw caps to prevent photo degradation and contact with atmospheric moisture.
4.1.2 Sodium Dodecyl Sulfate
An anionic surfactant, sodium dodecyl sulfate (SDS) (Life Technologies Inc.,
Gaithersburgh, MD) was used to prepare surfactant solutions. SDS is a 12 carbon straight chain surfactant. Structure and properties of SDS are described in Table 4.1.
SDS is a biodegradable and non-toxic surfactant. It is a widely used anionic surfactant and is employed in this study to compare the performance of natural surfactant solutions.
27 28 Table 4.1: Properties of the Surfactant, Sodium Dodecyl Sulfate
Structure CH3(CH2)10-CH2O-SO3-Na Molecular weight 288.38 CMC @25° C 8.08 mM Purity 5:99.5% Aqueous solution pH 7-7.5 Biodegradable yes 29 4.1.3 Hexachlorobenzene
Hexachlorobenzene (HCB), an aromatic chlorinated hydrocarbon having a very low solubility in water was used as a test compound. Structure, and physical and chemical properties of this compound are tabulated in Table 4.2. HCB was purchased from Aldrich Chemical Company (Milwaukee, WI). The compound was 99% pure and was used as supplied.
4.1.4 Soil
An uncontaminated soil from a local superfund site north of Baton Rouge, LA was selected for this study. The soil was air dried, homogenized, and kept in an oven overnight at 105 °C for drying. Soil was ground and the soil passing through a US
Standard No. 10 (2 mm) sieve was used for the experiments. Physical and chemical characterization of the soil was performed in accordance with methods of soil analysis
(ASA, 1986). Results are presented in Table 4.3. The total sand content of the soil is approximately 70% sand, out of which approximately 42% is classified as very fine sand. This soil is classified as a sandy loam and has a very low organic matter content
( 4.1.5 Basal Salt Media Nutrients for the bacterial cultures were added in the form of basal salt medium (BSM). Concentrated BSM (10X) was prepared by dissolving 58.0 g K2HP04 or 65.52 g K2HP04 3H20 , 45.0 g KH2P04, 20.0 g (NH4)2S04, 1.6 g MgCl2, 200 mg CaCl2, 20 Table4.2: Physical Properties of Hexachlorobenzene (Montgomery 1990) Molecular formula C6C16 Molecular weight 284.80 Boiling point 323-326°C Melting point @ 1 atm. 231 °C Water solubility @20°C 6.2 jig/1 Log octanol/water partition coefficient 5.31 Vapor pressure @25°C 1.9xl0'5 mm Hg Henry’s law constant @23°C 0.0013 atm-mVmole Specific density @ 20°C 2.049 ci \ cir j ci Molecular structure 31 Table 4.3: Physical and Chemical Characteristics of the Soil Physical Chemical Sand % 70.3 Calcium 1012 mg/kg soil Silt % 20.0 Magnesium 326 II Clay % 9.7 Potassium 62 i r Organic matter % 0.20 Sodium 28 n pH 7.3 Phosphorous 170 n C E C 8.0 meq/100 g 32 mg NaMo04, and 10 mg MnCl2 in one liter of water (Singhal and Roy, 1988). BSM (IX) was prepared by diluting the BSM (10X) solution. 4.1.6 Heterotrophic Media Heterotrophic media for anaerobic microorganisms was prepared by dissolving 300 mg KC1, 900 mg NH4C1, 90 mg CaCl2 2H20, 250 mg K2HP04, 250 mg KH2P04, 35 mg NaCl, 20 mg MgCl2 6H20, 159 mg Na2C03, 240 mg Na2S, 100 mg yeast extract, 500 mg peptone in one liter of DI water (Daniels et a l 1986, and Boopathy e ta l, 1993). 4.2 Experimental Methods 4.2.1 Glassware All the glassware used for the work was soaked in laboratory soap overnight. The glassware was then rinsed with tap water, acetone and finally deionized (DI) water rinse. The glassware needed for HCB work was cleaned with chromic acid and oven- dried overnight. The glassware and the solutions used in the biological experiments were sterilized for 15 minutes at a pressure of 20 psi and a temperature of 250°F in a Renaissance series sterilizer 3021 (Amsco Scientific Apex, NC). 4.2.2 Extraction of Fruit Pericarp into Different Solvents Water, methanol, ethanol and benzene:methanol (1:3) mixture were selected as the solvents for this work after reviewing the literature (Sarin and Beri 1939, Uppal 33 and Mehta 1951, Gedeon 1954, Row and Rukmini, 1966a, 1966b, Kimata et al, 1983, Nakayama et al. 1986, Kasai et al, 1988, and Guptha and Ahmed 1990). Approximately 10 grams of the dry fine powder obtained after passing ground pericarps through a US Standard No. 20 (840 ftm) sieve was added to the approximately 100 ml of the selected solvents. The mixture was stirred for 3 hours at room temperature and then filtered through a cloth. The filtrate was centrifuged at 10,000 rpm for 45 minutes. The supernatant was filtered through a 44 mm pre-filter (Coming Costar Corp. Oneonta, NY) and a metricel 0.45/nm membrane filter (Gelman Scientific, Ann Arbor, MI) in sequence. The filtrate was allowed to evaporate on a water bath at 70°C and 2 ml of dichloromethane was added to remove the remaining water. The dry paste obtained was re-dissolved in water and used as a stock solution for measuring surface tension and solubility of HCB. The concentration was defined as a percent that is grams of dry soapnut powder in 100 ml of the solvent. The stock solution prepared was generally 10%. 4.2.3 Standard Method for Preparation of Natural Surfactant Solutions The following procedure was adopted from the results of this study and those of Mandava (1994). The procedure does not require any expensive solvents or sophisticated equipment. The dry Ritha powder was weighted and added to DI water in a glass bottle of appropriate size. The weight of dry powder was 10 grams for every 100 ml of water. The mixture was stirred for 3 hours at room temperature. The un-extracted residue 34 was separated using a cloth, and the liquid was centrifuged at 10,000 rpm for 45 minutes at room temperature. The supernatant was filtered through a pre-filter, followed by a 0.45 /un filter. The solution thus obtained is used as a stock solution. The concentration of the stock was always 10% unless specified otherwise. ] 4.2.4 Surface Tension Measurements Surface tension measurements were made using a processor tensiometer K14 (KRT5SS GmbH Borsteler Chaussee 85-99a D-2000 Hamburgh 61 Germany). The machine enables measurements using the plate method in accordance with the Wilhelmy method (Void and Void 1983). The standard measuring device for the plate method is a rectangular platinum plate of exactly known geometry. An immersion depth of 2 mm was used. About 30 ml of sample was used for each measurement. The measurements were continued until an equilibrium value of surface tension was reached. The dilutions used were ranging from 0.0001% to 10%. All measurements were made in duplicates using natural surfactant solutions prepared independently at different times. 4.2.5 Viscosity Measurements Viscosity measurements were made using Bohlin VOR Rheometer System (Bohlin Instruments International AB, Lund, Sweden). A C-25 measuring system was used with a torque of 90 g-cm. Measurements were made with five different shear rates. Viscosity was measured at an interval of 240 seconds and an integration time 35 of 240 seconds. Duplicate measurements were made for each sample. Concentrations in the range of 0.01 to 1% were measured. 4.2.6 Solubility of Hexachlorobenzene in Natural Surfactant Solutions Experiments to determine the solubility of hexachlorobenzene in surfactant solutions was performed in Erlenmeyer flasks using several concentrations of surfactants in the presence of excess quantities of HCB ciystals. The flasks were closed with screw caps and sealed with Parafilm to make them air tight. The flasks were equilibrated by shaking for about 36 hours on a mechanical shaker before taking the samples. The samples were withdrawn using a pipette and centrifuged in teflon tubes for 15 minutes at 15,000 rpm. The supernatant was then analyzed for HCB. 4.2.7 Total Organic Carbon Total organic carbon (TOC) of the natural surfactant solution was measured using a Total Organic Carbon Analyzer, Model TOC-500 fixed with an ASI 502 auto sampler (Shimadzu Corporation, Kyoto, Japan). TOC was obtained directly by removing inorganic carbon by treating the sample with 2 drops of concentrated sulfuric acid, and passing nitrogen gas free of carbon dioxide through the sample. The combustion tube was filled with oxidation catalyst and heated to 680°C. While high purity air was allowed to flow into the combustion tube as a carrier gas, a prescribed volume of sample was injected into the tube. A one point calibration with a TOC standard in the range of expected concentration of the sample was used. Auto sampler 36 vials were cleaned with chromic acid and about 5 ml sample was filled into the vials. Samples were used in triplicate. Sample was diluted as needed for the appropriate measuring range of the machine. 4.2.8 Chemical Oxygen Demand The COD of the natural surfactant solutions was measured using the closed reflux colorimetric method described in Standard Methods (Greenberg et al. 1992). This method was developed by Jirka et al. (1975). The COD digestion reagent vials were purchased from HACH Company (Loveland, CO). The colorimetric method of measurement was performed to determine the COD using a pre-calibrated spectrophotometer, DR 2000 (Hach Chemical Company Loveland, CO) at a wavelength of 420 nm. All the samples were measured in triplicate. 4.2.9 Organic Nitrogen The organic nitrogen of natural surfactant samples was measured using the macro-Kjeldahl method (4500-Norg B) recommended in Standard Methods (Greenberg et al. 1992). Five hundred milliliters of the stock solution (10%) was used for the determination of organic nitrogen. 4.2.10 pH The pH of the natural surfactant solutions was measured using electronic method as described in Method 4500-H of the Standard Methods (Greenberg et al. 37 1992). Several concentrations of natural surfactant in duplicates were used for the measurements. 4.2.11 Absorbance Spectrum for Natural Surfactant Solution The UV/Visible absorbance spectrum of natural surfactant solutions was measured using a HP 8452A diode array spectrophotometer operated by HP 89531A MS-DOS TJV-VIS software (Hewlett-Packard Company, Wilmington, DE). The wavelength range used was from 190 to 820 nm. Several concentrations were used with duplicates. The absorbance of natural surfactant at several wavelengths was measured and correlated with the concentrations of natural surfactant. This method is very similar to the one reported very recently by Eaton (1995). 4.2.12 Generation of Colloidal Gas Aphrons Natural surfactant solutions prepared from fruit pericarps of Sapindus mukurossi and sodium dodecyl sulfate (SDS) were used to generate colloidal gas aphron (CGA) suspensions. Based on the method suggested by Sebba (1985), a unit was developed in our laboratory for generating CGA suspensions. A schematic diagram of the generator is shown in Figure 4.1. The unit consists of a horizontal disk which is connected to a high speed motor of 0.5 Hp with a stainless steel rod. The disk is mounted between two vertical plexiglass baffles and positioned 2 centimeters below the surface of the surfactant solution in a 3 liter plexiglass cylinder. The waves generated by the rotating disk strike against the baffles and upon re-entering the solution, entrain o{ fl* CGA Genetatot ipigure 41 39 air encapsulated by the soap film producing colloidal gas aphrons (CGAs). Surfactant solutions were fed into the cylinder at frequent intervals and the CGA suspensions were withdrawn at the required flow rate using a FMI lab pump model QG 20 (Fluid Metering Inc., Oysterbay, NY). Sebba (1985) estimated that 10,000 liters of CGA suspension can be generated with less than 1 kilowatt-hour of electricity. 4.2.13 Size Distribution of CGA Suspensions The size distribution of CGA suspensions generated from natural surfactant solutions and SDS solutions was determined using a Microtrac model 9210 standard range particle size analyzer (Leeds and Northrup, North Wales PA). This analyzer utilizes the phenomenon of forward scattered light from a laser beam projected through a stream of particles. The amount and direction of light scattered by the particles is measured by an optical detector array and then analyzed by a microcomputer which calculates the size distribution of the particles in the sample stream. The analyzer has a range from 0.69-704 /im. The sample was added manually to the reservoir and was mixed with the recirculating liquid so that a stream of well dispersed particles passes continuously through the transparent sample cell for analysis. About 10 ml sample was added to the reservoir, which holds about 300 ml of water. The analyzer was set to measure the particle size every 2 minutes. Several concentrations were used and the effect of the presence of salts was also investigated. 40 4.2.14 Stability and Quality of CGA Suspensions The stability was measured in terms of half life, the time required for half of the liquid content to drain (Longe, 1989), This was measured by transferring about 250 ml of CGA suspension into a 250 ml graduated cylinder and monitoring the drainage with time. The total liquid volume was measured by allowing the CGA suspensions to drain completely. The quality is measured knowing the total liquid volume after drainage and the initial volume of the CGA suspension taken into the graduated cylinder. 4.2.15 Soil Contamination Soil for the present work was obtained from a local Superfund site, north of Baton Rouge, LA. The prepared soil was spiked with hexachlorobenzene to study the performance of natural surfactant solutions in desorption and soil flushing. Appropriate quantity of HCB was dissolved in a beaker filled with petroleum ether and a known amount of soil was added slowly with continuous mixing. The mixture was then poured onto aluminum foil under a hood, and ether was allowed to evaporate. The dry soil was transferred into a bottle, and the bottle was tumbled on a tumbler for about a week. The soil was then extracted with hexane and acetone (1:1) mixture and analyzed using gas chromatograph for hexachlorobenzene to assess the initial concentration. 41 4.2.16 Desorption of Hexachlorobenzene from Soil Five grams soil contaminated as mentioned in the earlier section was weighed and added to several 125 ml erlenmeyer flasks. Surfactant solution (50 ml) of different concentrations above and below the CMC was filled into each of the flasks. The concentration of natural surfactant solution used were 0.05, 0.1, 0.5, 1.0, 1.5, and 2.5% in addition to water. The concentrations of SDS used were 5, 8, 15 and 35 mM. The flasks were shaken at room temperature for about 36 hours on a mechanical shaker. The samples were withdrawn and centrifuged to separate the soil particles before analyzing for hexachlorobenzene. 4.2.17 One Dimensional Soil Flushing Experiments Glass columns 10 cm long and 5.75 cm in diameter with a stainless steel top and bottom were used for all soil flushing experiments. The schematic diagram of the experimental set-up is shown in Figure 4.2. To prevent soil from being washed out of the column, the outlet end of the column was fitted with a fme wire mesh sandwiched between two coarse wire meshes. A soil packing procedure reported by Kommalapati (1994) was followed to achieve a bulk density similar to that observed in the field. It was packed by dropping the soil through a funnel in four equal portions of about 112.5 grams. Each layer was compacted by giving 25 blows with a compacting rod to obtain a bulk density of about 1.6 g/cm3 and a porosity of about 0.40, on the high side of the field bulk densities of regional soils (1.2-1.5 g/cm3). A 42 O Pressure Gauge wire m esh 10ct Soil flask -5.75cm- wire m esh Pump -CH CGA/ W ater/ Surfactant Figure 4.2: Schematic Diagram of the One Dimensional Soil Column Flushing Experiment 43 fine wire mesh sandwiched between two coarse meshes was placed on top of the soil to distribute the flow uniformly across the soil column. The packed column was kept in a vertical position and saturated with deionized water at a slow rate to remove the air bubbles. Experiments were conducted in downflow with water, natural surfactant solutions, and CGA suspensions. Concentrations used were 0.5%, 1.0%, and 2.5%. CGA suspensions were generated using 0.5% and 1.0% natural surfactant solutions. Deionized water was used for the experiments. The pressure at the influent end was monitored using an analog pressure gauge, and effluent samples were collected in sealed erlenmeyer flask. The effluent samples were analyzed for HCB using solid phase extraction and natural surfactant concentration using UV absorbance. The rate of pumping used for all the flushing fluids was about 2.5 ml/min. For the experiments where the flushing media was alternated with water, flushing media (CGA or natural surfactant ) were pumped for the first three pore volumes before switching to water. Three pore volumes were selected because the surfactant breakthrough seemed to occur after about three pore volumes. Water was pumped for two pore volumes and switched back to flushing media for two pore volumes. This procedure was repeated until a total of 16 pore volumes were collected. 4.2.18 Bioenhancement Studies Under Aerobic Conditions The typical soil microorganisms from the LSU campus, Baton Rouge, LA were collected and added to a sterilized 250 ml erlenmeyer flask along with 1% filter 44 sterilized natural surfactant solution to develop a culture for this study. The flasks were kept on a mechanical shaker at room temperature and allowed to grow. The log phase was reached in less than two days. This inoculum was used as a seed for aerobic studies. A 10% natural surfactant stock solution was prepared by extracting fruit pericarps with water as mentioned earlier. All the D1 water and the basal salt media (BSM) required for the experiment was sterilized along with the required glassware. Natural surfactant solution necessary for the entire experiment was filter sterilized using 0.45 /*m filter. For experiments where HCB was used HCB was solubilized in 10% natural surfactant solution and filter sterilized before adding to the flask. The acclimatized seed from the preliminary experiment was added to these flasks. Three concentrations of natural surfactant were used in the study. The details of the different treatments used are given in Table 4.4. The flasks were kept on a mechanical shaker at room temperature of 23 ± 2°C and stirred gently. Samples were taken at appropriate intervals, depending on the growth of the microorganisms. The samples were monitored for the growth of microorganisms by determining the absorbance of the samples at a wavelength of 540 nm (Koch 1981). A HP 8452A Diode Array Spectrophotometer (Hewlett-Packard Company, Wilmington, DE) was used for measuring the absorbance. About 2 to 3 ml sample was collected at different times to analyze for HCB and total organic carbon. The experiment was continued until the stationary phase was reached. 45 Table 4.4. Treatments Used in the Aerobic Bioenhancement Studies Symbol Description DI Nat. HCB in BSM Seed Wate Surf. Nat. r Surf. N Nat. Surf. 90 10 XX XX XX NS Nat. Surf. + Seed 88 10 2 NBS Nat. Surf. + BSM 78 10 XX 10 2 + Seed NH Nat. Surf. + HCB 90 XX 10 XX XX NHS Nat. Surf. + 88 XX 10 XX 2 HCB+ Seed NHB Nat. Surf. + 80 XX 10 10 XX HCB+ BSM NHBS Nat. Surf. + 78 XX 10 10 2 HCB+BSM+Seed LNS 0.1% Nat. Surf. + 97 1 XX XX 2 Seed LNBS 0.1% Nat. Surf. 4* 87 1 XX 10 2 BSM + Seed HNS 2% Nat. Surf. + 78 20 XX XX 2 Seed HNBS 2% Nat. Surf. + 68 20 XX 10 2 BSM + Seed 46 4.2.19 Bioenhancement Studies Under Anaerobic Conditions The anaerobic soil microorganisms from a local superfund site north of Baton Rouge, LA. were collected and added to a sterilized 125 ml serum bottles along with 100 ml of 1% filter sterilized natural surfactant solution to develop a culture for anaerobic study. The bottles were sealed and the culture was kept on a mechanical shaker at room temperature and the microorganisms are allowed to grow. The log growth phase was reached in about two weeks. This inoculum was used as a seed for the anaerobic bioenhancement studies. Anaerobic studies were conducted similarly to aerobic experiments. Serum bottles of 125 ml capacity were sterilized and filled with appropriate solutions as detailed in Table 4.5 except for natural surfactant solution and the seed. The anaerobic procedures followed by Boopathy et al. (1993) were used to keep the serum bottles anaerobic. The sterilized natural surfactant solution with or without HCB was first filled in to a sterilized serum bottle and sealed before degassing. An appropriate amount of natural surfactant solution was then transferred with anaerobic syringe to the culture bottles anaerobically with the help of a syringe and a blank degassed serum bottle filled with DI water. The acclimatized seed from the preliminary experiment was added to these flasks. Three concentrations of natural surfactant were used in the study. The details of the different treatments used are given in Table 4.5. The bottles were kept on a mechanical shaker at room temperature of 23 ± 2°C and stirred gently. Samples were taken anaerobically at appropriate intervals, depending on the growth of the microorganisms. The samples were monitored for the 47 Table 4.5. Treatments Used in the Anaerobic Bioenhancement Studies Symbol Description DI Nat. HCB in BSM Seed Wate Surf. Nat. r Surf. N Nat. Surf. 90 10 XXXXXX NS Nat. Surf. + Seed 85 10 5 NBS Nat. Surf. + BSM 75 10 XX 10 5 + Seed NH Nat. Surf. + HCB 90 XX 10 XX XX NHS Nat. Surf. + 85 XX 10 XX 5 HCB+ Seed NHB Nat. Surf. + 80 XX 10 10 XX HCB+ BSM NHBS Nat. Surf. + HCB + 75 XX 10 10 5 BSM + Seed LNS 0.1% Nat. Surf. + 95 1 XX XX 5 Seed LNBS 0.1% Nat. Surf. + 84 1 XX 10 5 BSM + Seed HNS 2%Nat. Surf. + 75 20 XX XX 5 Seed HNBS 2% Nat. Surf. + 65 20 XX 10 5 BSM + Seed AHN Nat. Surf. + HM 90* 10 XX XX , XX AHNS Nat. Surf. + HM + 85* 10 XX XX 5 Seed AHHN Nat. Surf.+HM + 90* XX 10 XX XX HCB AHHNS Nat. Surf. +HM + 85* XX 10 XX 5 HCB+ Seed * Heterotrophic media was added directly in stead of DI water and 10X media. 4 8 growth of microorganisms by determining the absorbance of the samples at a wavelength of 540 nm. Samples were analyzed for HCB and TOC at regular intervals. 4.2.20 Extraction of Hexachlorobenzene from Aqueous Surfactant Solutions The traditional liquid-liquid extraction did not work effectively in the presence of surfactant solutions. In this research a relatively new method called solid phase extraction is used. Commercially available Sep Pak CI8 cartridges were used in this research for extracting HCB from aqueous solutions. The cartridges were first activated by passing 5 ml of DI water followed by 5 ml of methanol and 5 ml of DI water. The aqueous sample was diluted if necessary and 5 ml of the sample was eluted through the cartridge at a rate of 5 ml/minute followed by a wash with 5 ml of DI water. The cartridge was then eluted with 5 ml of hexane at the same rate, which was collected and analyzed on a gas chromatograph. The samples containing high HCB concentrations were diluted to avoid saturation of the cartridge. For samples containing HCB at concentrations lower than 2 /ig/1, larger sample were passed through the cartridge so that the final concentration will be between 5 and 150 ^g/1. 4.2.21 Extraction of Hexachlorobenzene from Soil Extraction of HCB from soil was performed by adding 50 ml of hexane:acetone (1:1) mixture to about 5 ml of wet or dry soil in a 125 ml flask. When wet soil was added the water content was determined to correct for the amount of water. Acetone 49 was added to wet all the soil surface and extract HCB, particularly for sample with wet soil. All the flasks were shaken on a mechanical shaker for about 24 hrs. Samples were centrifuged and the organic layer was separated using a separatory funnel. Any residual water and the colloidal particles were removed by passing it through a sodium sulfate column. The sample was diluted if necessary and analyzed for HCB using GC. 4.2.22 Analysis of Hexachlorobenzene on Gas Chromatograph Hexachlorobenzene was analyzed on a high resolution gas chromatograph, HP 5890 series II fitted with a HP 7673 auto sampler and Ni63 electron capture detector (Hewlett-Packard Company, Wilmington, DE). The GC was fitted with a 30 meter PTE-5 capillary column, with a 0.32 mm internal diameter and 1.0 /im film thickness (Suppelco Inc., Bellefonte, PA ). The chromatographic conditions are: 1 fi\ splitless injection, 70 ml/min helium and 50 ml.min auxiliary gas, nitrogen, injection temperature 275 °C, Temperature program - 50°C (initial for 1 minute) to 270°C at 10°C per minute and hold for 3 minutes, total run time 26 minutes, ECD temperature 325 °C. Several HCB standards were analyzed and a calibration curve was prepared based on the height of the HCB chromatogram. The calibration was checked periodically for accuracy. Minimum detection limit for the method was 1 pico grams. 4.2.23 Statistical Analysis of the Data The linear regression analysis was performed on the solubility data to determine the relationship between surfactant concentration and the solubility of HCB. The 50 coefficient of determination (r2) was used to denote the strength of the correlation (Keller et al. 1988). The significance test was performed using Tukey’s multiple comparison method (Tukey 1953). All the comparisons in this study are multiple comparisons and the single pair comparisons are not very effective (Keller et al. 1988). For example, CGA suspensions were generated with 4 surfactant concentrations and the size distributions were monitored at 2 minute intervals for about 30 minutes. In this case we have 4 treatments and about 15 levels that need to be analyzed to determine whether there is a significance difference between the size distributions of the CGA suspensions generated with 4 natural surfactant concentrations. One has to perform either 6 hypothesis tests or test at a significance level of a/6. The first one increases the probability a of making a Type I error by the number of treatments and the second one lowers the overall power of the test and make the probability of a Type II error relatively high (Keller 1988). However, the Tukey test is a more powerful technique where a critical number is determined such that if any pair of means have a difference greater than this number, we conclude that their corresponding population means are different. This test does not place any limit on the number of comparisons. Tukey method of the General Linear Model (GLM) procedure available in the PC version of the SAS software was used. CHAPTERS RESULTS AND DISCUSSION 5.1 Development of Method for Hexachlorobenzene Analysis Hexachlorobenzene (HCB) is a highly hydrophobic organic compound. Due to the low water solubility of HCB, the traditional liquid-liquid extraction procedure has to be followed by a technique which concentrates HCB in the sample. Moreover, the presence of surfactant hinders the extraction of HCB due to many complexities such as formation of thick emulsion and the inconsistency with the recovery. Preliminary liquid-liquid extraction results yielded a very low and variable recovery of HCB from surfactant solutions. The present method makes use of commercially available Sep-Pak Clg cartridges which can retain HCB from the sample and can later be eluted with a non-polar solvent such as hexane. This is the first reported research to extract HCB from aqueous surfactant solutions and one of the three investigations reported in the literature to use cartridges for extracting HCB from any phase (Chiang et al. 1986, and Borra et al. 1989). The first investigation to use cartridges for extracting HCB was by Chiang et al. (1986), where, Florisil cartridges were employed for extracting semi-volatile organic compounds from adipose tissue. The recovery efficiency was 86.2% with a standard deviation of 4.6. The other group Borra et al. (1989) used C18 cartridges to extract semi-volatile compounds from wastewater samples. The recovery of HCB was 93% using this method. In the current study Clg cartridges were used to extract HCB from aqueous solutions of natural surfactant and sodium dodecyl sulfate (SDS). 52 The extraction of HCB from aqueous surfactant solutions was performed with three surfactant concentrations and three concentrations of HCB. Natural surfactant concentrations used were 1%, 5% and 10% and HCB concentrations employed were about 20, 50 and 100 fig/l. For each concentration of natural surfactant and HCB, a known amount of HCB was added to 12 flasks. Hexane (50 ml) was added to six flasks and 50 ml of natural surfactant solution was added to the remaining six flasks. The average concentration of HCB added to the flasks is calculated from the 6 hexane flasks. HCB concentration recovered from 6 natural surfactant flasks was used to calculate the average recovery for that concentration of natural surfactant. The results of some of the extractions are presented in Table 5.1 along with the corresponding HCB concentration in hexane. Columns 2 and 5 show the HCB concentration in hexane and columns 3 and 6 give the recovery of HCB from natural surfactant solutions. The columns 4 and 7 give the recovery percent for each surfactant and HCB concentration. The overall recovery of HCB is 93.7% with a standard deviation of 2.2%. The recovery obtained in this study is higher and the variation is lower than that obtained by the two earlier investigations. The method is much simpler than the traditional liquid-liquid extraction and the recoveries are higher and reproducible. The method does not involve specialized glassware and also saves several man hours. The method is adopted for the entire analysis including the work with sodium dodecyl sulfate (SDS). 53 Table 5.1: Recovery of Hexachlorobenzene from Natural Surfactant Solutions*-f Sample Hexane Surf. % Rec Hexane Surf. % Rec (Mg/1) WO overy (Mg/0 (Mg/D overy 1% Nat. Surf 10% Nat. Surf. 1 15.13 13.98 23.29 21.27 2 15.57 14.06 24.23 22.44 3 15.56 14.66 23.78 23.81 4 16.85 14.91 23.78 21.41 5 16.56 15.53 23.58 22.19 6 15.34 14.15 23.50 21.63 Ave. 15.84 14.55 91.86 23.29 22.13 93.39 1 51.52 18.03 51.52 48.23 2 50.27 49.58 50.27 47.89 3 46.99 45.84 46.99 44.27 4 57.89 51.2 57.89 53.78 5 43.27 42.68 43.27 40.08 6 48.13 47.88 48.13 46.91 Ave. 49.70 47.54 95.65 49.70 46.86 94.29 1 103.21 93.38 103.21 92.03 2 100.07 93.81 100.07 91.42 3 102.29 62.29 102.29 93.25 4 98.39 94.59 98.39 96.58 5 100.99 97.64 100.99 91.81 6 101.38 91.21 101.38 93.88 Ave. 101.05 93.82 92.84 101.05 93.16 92.19 * Partial data + Average Overall Recovery = 93.7%, Standard deviation = 2.2% 54 Extraction of HCB from soil was performed by adding hexane .acetone (1:1) mixture to soil and the extraction efficiency was about 86.3 % with a standard deviation of 7.4. This efficiency is much higher than that obtained by Pardue (1992) which was between 60 and 70%. Pardue used wetland sediments which contain large amounts of organic matter (28 g, and 230 g carbon per kg soil). Under these conditions HCB is likely to undergo irreversible adsorption with the humic material resulting in a low HCB recovery. The soil used in this study has only 0.23% organic matter and thus does not affect recovery significantly. The extraction efficiency remained unchanged for the soils treated with either of the surfactants, natural surfactant or SDS. 5.2 Preparation and Characterization of Natural Surfactant 5.2.1 Natural Surfactant Solutions Several solvents were used in the literature to extract and isolate the saponins from the fruits of a number of species belonging to the genus Sapindus (Sarin and Beri 1939, Uppal and Mehta 1951, Gedeon 1954, Row and Rukmini, 1966a, 1966b, Ranganna et al. 1963, Kimata et al, 1983, Nakayama et al. 1986, Kasai et al. 1988, and Gupta and Ahmed 1990). The Sapindus family trees include Sapindus emarginatus, Sapindus mukurossi, Sapindus laurifolius, Sapindus trifoliatus, and Sapindus delavayi. The saponins are believed to be responsible for the detergent properties of these fruits. The water soluble portion of the fruit is being used as a detergent for washing fabric and as shampoo. The most common solvent used in the 55 literature is water and the other solvents used are methanol, ethanol and in some cases benzene and methanol. The aim of all the above investigators was to isolate and purify the saponins, however, the objective of this research is to And a simple and less expensive extraction method without compromising on the performance of natural surfactant solutions. Fruit pericarp powder was extracted into different solvents i.e. water, methanol, ethanol and methanol: benzene (3:1) mixture and the surfactant properties were evaluated. The properties used to evaluate the performance of the different extracts are surface tension and solubility of HCB in these extracts. Dry Ritha powder was extracted with the above mentioned solvents and the un-extracted residue was separated. The solvents after extraction acquired different colors. The ethanol extract was a light solution with light yellowish golden color, methanol had a light golden brownish color and the methanol:benzene mixture had a darker golden color and the water extract had the darkest color of all the extracts. The colors of different extracts are in agreement with the weight of Ritha powder dissolved in each solvent. The weight of the un-extracted residue and the weight of dissolved residue obtained after evaporating the solvents are reported in Table 5.2. Three extractions were performed for each solvent, and the averages are included in the table. It should be noted that these values are reported to give an idea of how much of pericarp powder is extracted into different solvents. Mass balance performed on Ritha powder yielded a higher mass than the initial powder weight when the un-extracted residue and the dissolved components are combined. This error could be due to the moisture present in the paste Table 5.2: Extraction of Fruit Pericarps of Soap Berry into Different Solvents: Residue Remaining and the Residue Extracted Solvent Residue Disssol- Total Average (g) ved (g) (g) % Dissolved Water 1 4.39 10.71 15.00 2 4.46 10.37 15.00 3 4.21 10.98 15.00 71.2 Methanol 1 4.99 10.41 15.00 2 5.08 10.16 15.00 3 4.89 10.29 15.00 68.58 Ethanol 1 5.99 9.97 15.00 2 6.23 8.89 15.00 3 6.11 8.92 15.00 59.51 Methanol: Benzene (3:1) 1 4.07 6.41 10.00 2 4.19 6.22 10.00 3 Not available 10.00 63.15 57 after evaporating the solvents. As can be seen from Table 5.2, ethanol extracted the least amount of powder and water extracted the most. Methanol and methanol:benzene mixture are in the middle. The residue obtained after evaporating the solvents, water, methanol, ethanol, and methanol:benzene mixture was re-dissolved in DI water to make a 10% stock solution. The diluted solutions were used to measure surface tension and solubility of HCB. The solution made from methanol-.benzene mixture was used in the solubility study only. Surfactant molecules have a tendency to accumulate at the air/water interface due to their amphiphilic structure which has both hydrophobic and hydrophilic moieties. This affects the surface tension of water significantly (Rosen, 1989). Hence, surface tension was used in this study to determine the suitable solvent for preparing natural surfactant solutions. The results of the surface tension measurements are presented in Figure 5.1(a). The figure shows the variation of surface tension with concentration for natural surfactant solutions made from water, methanol, and ethanol extracts. The Y axis shows the surface tension in mN/m and the X axis shows the concentration of surfactant on logarithmic scale. The concentrations used ranged from 0.0001% to 10%. As can be seen from the figure there is no variation in surface tension between the natural surfactant solutions prepared with different solvents. The surface tension value drops significantly until a concentration of about 0.1 % in all the cases and remains almost constant beyond this concentration. The break in the surface tension curve indicates that the surfactant concentration is at the critical micelle concentration (CMC). Since there is no difference between the surface tension values, 80 58 70 60 e z 50 WE C .2 *55c 40 £ u u 30 Water Extract 42i- 3 Methanol Extract CO 20 -— Ethanol Extract 10 0 t— 0.0001 0.001 0.01 0.1 1 10 100 Log (Natural Surfactant Concentration ,%) 14 Water 12 ' • — Methanol u 10 ' ~ Ethanol £ ca MeOH:Benzene a 8 X o 6 o 4 CO 2 Lines: Regression Points: Measurements 0 2 4 8 10 Natural Surfactant Concentration (%) Figure 5.1 : Variation of (a) Surface Tension and (b) HCB Solubility with Concentration for Water, Methanol and Ethanol Extracts 59 it is not practical to evaluate the efficiency of the extracting solvent. Solubility measurements should be used to determine the performance of these extracts. Surfactants form micelles beyond the CMC and these micelles can incorporate significant amounts of hydrophobic organic compounds into their structure (Rosen, 1989). Solubilities of some of the hydrophobic compounds increase a thousand fold in surfactant solutions (Void and Void, 1983). Solubility of HCB was used in this study to evaluate the performance of different solvents employed for extracting Ritha. Excess amount of HCB crystals were equilibrated with several concentrations of natural surfactant and the samples were analyzed for HCB. The concentrations of natural surfactant tested were 0.1, 1.0, 2.5, 5 and 10%. Triplicate samples were independently prepared with each solvent and used for the study. Figure 5.1(b) shows the variation of HCB solubility with natural surfactant concentration for the four solvents i.e. water, methanol, ethanol and methanol .benzene (3:1) mixture. The Y axis indicates solubility in mg/1 and the X axis represents natural surfactant concentration in percentage. The regression data is shown by the lines and the points indicate the actual measurements. It is clear from the figure that the solubility of HCB increased with increase in surfactant concentration for all the extracts. HCB solubility for any concentration is the maximum for methanol extract, followed by methanol:benzene mixture, ethanol and water extract has the lowest solubility. The differences between the different extracts are higher for higher natural surfactant concentrations (5% and 10%) and the variation decreased with decrease in surfactant concentration. The variation between the HCB solubilities of different 60 extracts is not significant at 95 % confidence level for lower concentrations of natural surfactant (^2.5% ). Tukey multiple comparison method (Tukey, 1953) was performed to test for the significance. It should also be noted that there is a considerable variation in HCB solubility between the samples of the same extract. This variation is attributed to the preparation process. As mentioned in the earlier section, several steps were involved in preparing natural surfactant solutions such as extraction, centrifugation, filtration, evaporation of the solvent and subsequent re-dissolution in water. The variations at each stage have a cumulative effect on the HCB solubility. It will be clear from the later sections that the concentrations of natural surfactant used in practical applications are always less than about 2.5%. Hence, it does not seem logical to use an expensive and/or toxic organic solvents to extract Ritha and thus complicate the preparation technique and/or increase the cost of preparation. The very purpose of using a natural surfactant is to eliminate the use of expensive solvents and sophisticated chemical processes and thus save considerable time and money and also to reduce the production of toxic by-products. Hence it is concluded to extract Ritha into water, even though the methanol extract seem to be little more effective in solubilizing HCB. By using water as the solvent two additional steps, evaporation of solvent and redissolving in water are avoided. The only by product in the preparation of natural surfactant solutions is the un-extracted residue, which can be disposed as a biodegradable solid waste. This residue can even be used 61 as a fertilizer for agricultural lands. It should be noted that only about 30% of the mass remain as residue. For all the subsequent work in this research, a simple method of extraction with water is adopted. A known amount of pericaip powder was added to appropriate amount of DI water so that the concentration is 10% (e.g. 10 grams in 100 ml). After stirring the mixture was separated by centrifugation followed by filtration. The clear transparent dark golden color liquid obtained is used as a stock solution. About 5 to 10% of the sample is lost during the preparation process. This loss is due to the transfer of the sample during the different stages of separation of the residue, centrifugation and filtration. The concentration of the stock solution was calculated based on the initial water added rather than the final volume, because all the losses occur only after Ritha is extracted into the water. 5.2.2 Chemical Oxygen Demand(COD) The COD of the natural surfactant solutions was measured to determine the oxygen equivalent of the organic carbon. The COD gives the amount of oxygen that is required to oxidize the organic carbon to carbon dioxide and water. The COD can be correlated with organic carbon or biochemical oxygen demand (BOD) for the samples from the same source. COD was used to determine the empirical formula for the natural surfactant solutions. COD was measured after diluting the stock solution considerably. The COD of 10% natural surfactant solution is 124.3 grams/liter. 62 5.2.3 Total Organic Carbon (TOC) TOC as the name indicates is total organic carbon present in the sample. It can be measured either indirectly by subtracting inorganic carbon from total carbon or directly by measuring total carbon after eliminating inorganic carbon by treating with concentrated acid. TOC was used for monitoring biodegradation of natural surfactant under aerobic and anaerobic conditions and also to determine the empirical formula for natural surfactant solutions. The TOC of natural surfactant solutions was measured after diluting the sample significantly. The TOC of a 10% solution is 41.2 grams/liter. 5.2.4 Nitrogen and Phosphorous The nitrogen content of the natural surfactant solutions was measured using macro-Kjeldahl method for organic carbon. The nitrogen content of the natural surfactant solutions was undetectable by this method, when 2S0 ml of 10% solution was used. The phosphorous content of natural surfactant solutions was undetectable by the current method. Natural surfactant being a fruit pericaip, it generally contains carbon, hydrogen, and oxygen only. 5.2.5 pH The pH of the natural surfactant solutions was measured using a pH electrode. The measurements were made at room temperature (22±1°C). The variation of pH with natural surfactant concentration is plotted in Figure 5.2. The figure clearly shows that the natural surfactant solutions are acidic in nature. The pH of the solution varied 63 8 7 6 5 4 3 Sample 1 2 * Sample 2 1 0 0.001 0.01 0.1 1 10 Log (Natural Surfactant Concentration, %) Figure 5.2 : Variation of pH with Concentration for Natural Surfactant Solutions 64 from that of DI water to about 4.5 as the concentration increased from 0.005% to 1% and was invariable beyond that concentration. The acidic nature of the solution could be due to the hydrolysis of the glycosides present in the fruit pericarps (Row and Rukmini 1966a). 5.2.6 Empirical Formula The empirical formula of compound shows the relative numbers of the different atoms in a molecule. It can be determined either from the percentage composition of each element or from the oxidation of the compound. The latter is simple and convenient for organic compounds (Christensen and McCarty, 1975). Knowing the COD, TOC, nitrogen and the total weight of organic compound, empirical formula can be calculated from the generalized oxidation half reaction for an organic as given below: C,HbO0Nd + (2a-c)HaO = aC02 + dNH% + (4a+b-2c-4d) H+ + (4a+b-2c-3d)e' For determining the empirical formula, Ritha pericarp powder was dissolved in six pre-weighed flasks and the weight of the residue was determined carefully. The weight of organics dissolved in water was calculated from the initial weight and the weight of residue. The solutions were analyzed for COD and TOC. The following analysis was performed to determine the empirical formula (Appendix): weight of organic (W) used in the experiment = 67.99 grams total organic carbon of the organic = 41.225 grams 65 chemical oxygen demand of the organic = 124.325 grams organic nitrogen content of the organic= 0.0 grams Substituting these values in equations derived in the appendix and solving for the stoichiometric coefficients yield the following: a = 3.52, b = 4.17, c = 1.35 and d = 0 The ratio of carbon:hydrogen:oxygen:nitrogen is 2.6:3.1:1:0, From these values one can arrive at an empirical formula of C26H3lOi0 for natural surfactant. The molecular formula for natural surfactant is (C26H3iO|0)n, where ’n’ is a constant that needs to be determined either from vapor density method or cryoscopic method (Morrison and Boyd, 1970). The carbon content expressed as percent of total weight of compound is 60.6%. This is very close to that of saponins isolated from natural surfactant reported in the literature (Row and Rukmini 1966a and Row and Rukmini 1966b). They determined the carbon percent to be 57.8% and 61.79% respectively for Sapindus mukurossi and Sapindus emarginatus respectively. The molecular formula was reported to be 2H20 for mukorossi fruit (Row and Rukmini 1966a) and C47H760 17 for emarginatus fruit (Row and Rukmini 1966b). It should also be noted that the soap nuts available in markets are generally mixtures of these two. These molecular formulas are for the isolated saponin fraction only, however, the empirical formula obtained in this study incorporates all the water soluble constituents of soap nut. Another accurate method for obtaining molecular formula for pure compounds is mass spectrometry. However, natural surfactant solutions are mixtures of complex organic compounds and thus it is not practical to use the method without significant 66 purification of the sample. The empirical formula obtained in this study gives an idea on the ratio of carbon, hydrogen and oxygen in natural surfactant solutions. 5.2.7 Quantification of Natural Surfactant Using UV Absorbance Several researchers used UV absorption as a surrogate measure for selected organic constituents in fresh water, wastewater, and salt water (Dobbs et al. 1972, Bunch et al. 1961 and Foster and Morris, 1971). Eaton (1995) reported a standard method for UV absorbing organics, which is to be added as method 5910 to the nineteenth edition of 1995 Standard Methods. He used the method for measuring disinfection by-products. Natural surfactant is a mixture of organic compounds and should be able to absorb UV radiation and thus can be quantified using this method. Natural surfactant solutions of several concentrations were analyzed with UV/Visible spectrophotometer to study the optical properties of these solutions. The UV/Vis absorption spectra of natural surfactant solutions for concentrations 0.01, 0.1, 0.5, 1.0, and 2.0% are overlaid and presented in Figure 5.3. The figure shows the absorbance between the wavelengths of 190 nm and 820 nm. As can be seen from the figure, the absorbance at any wavelength increases with the increase in concentration and also new peaks start appearing as the concentration is increased. The peaks appear at several wavelengths as the surfactant concentration is increased from 0.1% to 2%. Natural surfactant solutions exhibited spectral properties which can be used in quantifying the solutions. Absorbance is used as a measure to quantify dissolved 2.8568-1 Absorbance Spectra Concentrations: 0.01j 0.1, 0.5, 1.0, and 1.5 V. 2.2855- 1.7141- CO Increasing Concentration CO 0.57137 0.0000 2 0 0 300 500 600 700 800 UflVELENGTH Figure 5.3 : Absorption Spectra for Natural Surfactant Solutions of Several Concentrations 6 8 organic carbon by several researchers (Eaton 1995, Moore, 1985, Dobbs et al. 1972 and others). It is suggested in the literature that by correlating UV absorbance with dissolved organic carbon, quantification of DOC is possible for natural streams, lakes, etc. Natural surfactant being a mixture of saponins and glycosides it is fairly complicated to quantify accurately without chromatographic separation of individual compounds. It is believed that UV absorbance can be a very useful and cheaper tool for quantifying natural surfactant solutions. The correlation between natural surfactant concentration and UV absorbance is shown in Figures 5.4 (a and b) for wavelengths 252, and 292 nm respectively. The spectra showed peaks at several wavelengths. However, the absorbance at lower wavelengths was not following Beer’s law and thus cannot be used. Absorbance is plotted on the Y axis and X axis shows surfactant concentration in percent. The line indicates the regression data and the points show the actual measurements. As can be seen from the figures, the correlation of natural surfactant concentration with UV absorbance is very strong. The lowest R2 value for the regressions is greater than 0.99. The slopes of the lines are steep. The correlations are valid only up to 1.2% and 1.5% of natural surfactant for wavelengths 252 and 292 nm respectively. The lowest concentration used in the study is 0.01%. The correlation however, is good for concentrations higher than 0.1%. The correlation equations for the two wavelengths are given below along with the corresponding R2 value. The absorbance is denoted by A and concentration by C Figure 5.4 : Correlation Between Natural Surfactant Concentration and Absorbance Absorbance and Concentration Surfactant Natural Between Correlation : 5.4 Figure Absorbance at 292 nm Absorbance at 252 nm 0.5 2.5 2.5 1.5 1.5 0 2 2 3 1 1 05 15 2 1.5 1 0.5 0 at (a) 252 nm and (b) 292 nm 292 (b) and nm 252 (a) at Regression Measured (For concentrations1.2%)0.50.992 upto - R‘2 (For concentrations1.5%) up to Absorbance = 0.05 +2.31* Cone. +2.31* Absorbance0.05 = bobne- .0 1.675* Cone. 0.002+ Absorbance- R‘2 - 0.994 - R‘2 Cone, of Natural Surfactant Natural of Cone, Measured Regression 70 ^252 - 0.05 + 2.31 *C (/f2= 0.992) (1) ^ = 0.002 + 1.675 *C (Rz= 0.994) (2) For any sample, knowing the absorbance at one or both of the above mentioned wavelengths one should be able to calculate natural surfactant concentration from these equations. These correlations are valid over different ranges of natural surfactant concentrations. The correlations are very useful in determining the concentration of natural surfactant in the one dimensional column effluent and thus establish the breakthrough curves for natural surfactant. 5.2.8 Effect of Sterilization The same batch of samples used for correlating UV absorbance with natural surfactant concentration were sterilized and used to study the effect of sterilization. Preliminary experiments with sterile natural surfactant solutions indicated significantly higher solubility of hexachlorobenzene. Mandava (1994) also observed increased solubility of naphthalene in sterile natural surfactant solutions. The increased solubility with sterilization prompted an investigation to study the absorbance spectra of these samples and compare with those of non-sterile samples. The absorbance spectra of sterile and non-sterile natural surfactant solutions are shown in Figures 5.5 (a and b) for concentrations 0.5, and 1.5% respectively. As can be seen from the figure the sterilized samples have higher absorbance than the non-sterile sample, even though the changes are not significant at certain wavelength ranges. It should be noted that l.tllS (a) 0.5% Sterile Non-Sterile’ 0.177##. 0.0000 300 300100 coo 700 #00 (b) 1.5% ■2 Sterile I.I31T- Non-Sterile 0.0000 300 300 coo 700 #00 Figure 5.5 : Absorption Spectra for Sterile and Non-Sterile Natural Surfactant Solutions at (a) 0.5% and (b) 1.5% Concentration 72 fraction of the sample used for non-sterile solution is sterilized and used so the differences in absorbance spectra are due to sterilization only. The samples were also analyzed in duplicate to confirm the observations. The differences even though are not apparent at lower wavelengths, they became significant at higher wavelengths particularly between 240 and 300 nm. When the individual spectrums are compared it appeared that the sterilized samples developed new peaks at lower concentrations than they would normally appear for non-sterile samples. This indicates that the constituents of natural surfactant probably undergo some chemical changes during the sterilization. However, it is not possible in this study to identify these individual compounds. Regression analysis performed between absorbance and concentration for sterilized samples yielded a similar correlation as non-sterile samples. The regression lines for sterile solutions are parallel to those of non-sterile samples at the corresponding wavelengths. The intercepts of the regression lines are however, higher than those for non-sterile natural surfactant solutions. 5.2.9 Critical Micelle Concentration A very fundamental and important property of surfactants is micelle formation. This phenomena affects not only detergency and solubilization, which depend on existence of micelles, but also surface tension and interfacial tension, that do not directly involve micelles (Rosen 1989). The surface active solutes form colloidal size clusters which are called micelles at concentrations beyond critical micelle concentration (CMC). This concentration is a narrow range rather than a single 73 concentration (Void and Void, 1983). The surfactant monomers aggregate into micelles of different shapes such as spherical, rod like micelles or prolate ellipsoids, flat lamellar and vesicles. The micellization affects almost every measurable physical property that depends on size or number of particles in solution such as conductivity, surface tension, detergency, osmotic pressure, interfacial tension, viscosity etc. All these properties will have a break in the curve plotted against the concentration of surfactant in the neighborhood of CMC (Preston 1948). Surface tension and viscosity measurements were used in this study to determine CMC of natural surfactant solutions. Figure 5.6(a) shows the variation of surface tension with natural surfactant concentration. Concentration of surfactant is represented by X axis on a logarithmic scale and Y axis shows the surface tension in mN/m. Natural surfactant concentrations ranging from 0.0005 to 10% in two independently prepared batches were used. The regression line for the data is plotted along with the actual measurements in the figure. As can be seen from the figure, there is a sharp break in the surface tension curve in the concentration range of 0.08 and 0.1%. This sharp break is due to the formation of micelles and the concentration is called CMC. The reason for having a range of concentrations for CMC rather than a single value is due to the fact that there can be variation in lengths of hydrophobic carbon tail groups of surfactant that is more hydrophobic surfactant isomers may form micelles at concentrations well below that of other compounds in the mixture. This concentration is taken as 0.1%. Surface tension of water is about 73 mN/m. As the concentration of natural surfactant iue56 Dtriaino M fNtrl ufcatSltos y a Surface (a) by Solutions Surfactant Natural of CMC of Determination : 5.6 Figure eso n ()Vsoiy Measurements Viscosity (b) and Tension Viscosity (Pa-s) Surface Tension (mN/m) 250 200 300 350 150 100 50 20 40 50 70 60 30 80 10 0 n 0.0001 0 - - - - - o NtrlSratn ocnrto, %) Concentration, (NaturalSurfactant Log * 1.85E-03 —*— Shear Rate (1/S) (1/S) Rate Shear a \ 0.001 Natural Surfactant Concentration (%) Concentration Surfactant Natural 0.2 ■ 7.30E-03 • ~ 2.92E-03 - 1.16E-02 • » — 4.60E-03 \ □ 0.01 0.4 □ • y y . ------0.1 0.6 j ------ Sml 1 Sample D • Sample 2 Sample • 0.8 Regression 1 1 .. 10 1 74 75 increased from 0.0005 to 0.1%, surface tension values dropped steadily and beyond 0.1 % the values became stable between 36 and 35 mN/m. These values are very similar to those reported in the literature for several commercial surfactants (Liu et at. 1992, Edwards et al. 1991, Abdul et al. 1991, Kile and Chiou 1989, and Vigon and Rubin, 1989). The reported minimum surface tension values for the commercial surfactants values are typically in the range of 30 - 40 mN/m. The minimum surface tension value for natural surfactant is found to be about 35.5 mN/m. This value is lower than that for sodium dodecyl sulfate (SDS), 42 mN/m (Kile and Chiou 1989) and is in the range reported for other commercial surfactants. Viscosity measurements for natural surfactant solutions at several shear rates are shown in Figure 5.6(b). Viscosity is shown on the Y axis in Pascal-second units and the X axis shows natural surfactant concentration in percent units. Hoffmann and Rehage (1987) studied the rheology of surfactants and suggested that viscosity measurements can be used to determine CMC. As can be seen from the figure, the viscosity increased from that of water at all the shear rates when a low concentration (0.01%) of natural surfactant was used. However, when the concentration was increased from 0.01% to 0.05% the values dropped and reached a minimum at 0.1% and beyond 0.1% the viscosity values increased. At low concentrations, where the surfactant is in the monomeric form, the hydrophilic tails of the surfactant are surrounded by ordered molecules and the viscous resistance increases. The formation of micelles however, releases the ordered water molecules and thus result in an abrupt change in viscosity. From the figure the CMC of natural surfactant is found to be 7 6 0.1 %. However, it should be noted that there should have been more data points in the neighborhood of CMC. When the natural surfactant concentration is further increased the viscosity will increase again due to the formation of more and more micelles and the resultant intermicellar interactions. 5.3 Generation and Characterization of Colloidal Gas Aphron (CGA) Suspensions Colloidal gas aphron suspensions are generated from commercial surfactant solutions and used in several applications (Sebba 1982, 1985b, Longe 1989, Roy et al. 1992a, 1992b, 1994, 1995, and Chaphalkar 1994). The most important parameter in characterizing CGA suspensions is the size distribution of bubbles as it influences not only the stability but also the rheological properties of the suspensions. This section is focussed on the size distribution and stability of CGA suspensions generated with plant-based surfactant solutions. The results are compared to those generated using commercial surfactants, sodium dodecylbenzene sulfonate (NaDBS), Tergitol, hexadecyltrimethylammonium bromide (HTAB) which are available in the literature. The effect of the presence of electrolyte on the CGA size distribution is also investigated by adding 200 mg/1 and 400 mg/1 sodium chloride to natural surfactant solutions. As discussed in Chapter 2, the distinguishing feature of the structure of colloidal gas aphron is that the encapsulating soap film has inner and outer surfaces with surfactant mono-layers adsorbed on them (Figure 2.1). The encapsulation retards the coalescence and improves the stability of the bubbles significantly. The water captured between the layers was found to have properties different from those of bulk water due to enhanced hydrogen bonding (Sebba 1982). The different mechanisms involved in the formation and subsequent stability of the bubbles are hydrodynamic forces, inter-bubble gas diffusion and inter-bubble collision. The CGA bubbles, during formation, are found to range from sub-micron to a hundred microns in size. However, with time the bigger bubbles disappear at the expense of smaller bubbles (< 25 microns)(Sebba, 1982). The turbulence during the generation causes inter bubble collisions which does not seem to affect the smaller bubbles (< 100 nm), but the larger bubbles (>350 ftm) coalescence, while the bubbles (150 - 300 /*m) seem to undergo bubble fission (Longe 1989). 5.3.1 Size Distribution Parameters The particle size analyzer used in the present study (Microtrac standard range particles size analyzer) uses a dimensionless parameter, Dv, to represent the volume of sample material in the circulating system which is also referred as sample loading. Chaphalkar (1994) used a similar system and suggested that about 50 ml of CGA suspension has to be added to the system which, contains about 250 ml of water to have an optimal range of loading for all suspensions used. However, for the present work, a more recent and sensitive instrument was used and about 10 ml of the sample added to a mixing chamber containing 300 ml water was found to be satisfactory from the preliminary experiments. Sample load as represented by Dv was used to determine the stability of CGA suspensions as was suggested by Chaphalkar (1994). The stability 7 8 was also measured by the gravity drainage method suggested by Longe (1989) and both the results are compared. The other important parameter in characterizing CGA suspensions is the representative size. Chaphalkar (1994) used mean volume diameter, mv to indicate the average size of the bubbles, as it was found to be more or less stable with time. It was calculated based on the sample volume in the system and represents the average size of the bubbles in the suspension. However, ’mv’ was not stable and reduced with time in our studies. The 10% and 50% sizes remained stable over the length of the run and 90% size was reducing with time. This indicates that the smaller bubbles in the CGA suspensions remain stable but the bigger bubbles undergo dynamic changes as indicated by the 90% size which directly influences the mean volume diameter. This kind of behavior was not reported by the earlier researchers and hence it is felt that all the size fractions should be studied. Moreover, it is necessaiy to see the dynamic changes that occur over the entire range of the bubbles rather than just the average size. A 10% (10 percentile) size represents a size in microns such that 10% of the bubbles are finer than this size. Similarly the bubbles finer than 50% size and 90% size are 50% and 90% of the total sample. The 10% and 90% sizes are also used to define the size range of the bubbles. 5.3.2 Size Distribution A typical size distribution for colloidal gas aphron suspensions is shown in Figure 5.7 for three concentrations of natural surfactant, 0.1, 0.5, and 1.0% 79 1M M U 71 U 0 1# L U n E S X MB 1M M C B u I I) F 15 V U 0 0 L IB L U U n II E E a 5 x x IN Z5 M B U 7B I r v a L U n E 1M Figure 5.7: Typical Size Distribution Curves for CGA Suspensions Generated with (a) 0.1%, (b) 0.5%, and (c) 1.0% Natural Surfactant Solutions 8 0 respectively. The X axis is on a logarithmic scale and represents size of the bubbles in microns and the Y axis represents cumulative volume in percent on one side and the differential volume in each size range on the other side. As can be seen from the figure, the size distributions follows a pattern similar to frequency distribution with most of the CGA having sizes in the middle (30-100 /zm). Close examination of the figure indicates that the size distribution of CGA suspensions generated with 0.1% (Figure a) has a wider range that is 10 - 500 /zm, where as the CGA suspensions generated with 0.5% (Figure b) and 1.0% natural surfactant (Figure c) have sizes ranging from 20 - 100 /zm. As will be discussed later, CGA suspensions generated with low concentrations of natural surfactant (0.1%) become increasingly unstable promoting the formation of bigger size bubbles. The typical variation of different size fractions, 10%, 50% 90% and mv, mean volume diameter, with time is shown in Figure 5.8 (a and b ) for natural surfactant concentrations 0.5% and 1.0% respectively. The Y axis represents the different sizes, 10%, 50%, 90% and mv in microns and the X axis shows the time in minutes. The range of particles as defined by the 10% and 90% sizes is between 30 and 300 /zm, except in two or three cases where it was up to 350 /zm. This range is the same as the one reported by Chaphalkar (1994) for CGA suspensions generated with several commercial surfactants. However, this range was much wider than that suggested by Sebba (1982) and Longe (1989). It should be kept in mind that, they determined size distribution under static conditions using photomicrographic methods. The present method uses a dynamic system where, the bubbles are pumped into a mixing chamber 350 M 300 (a) Natural Surfactant, 0.5% c o ok* is 250 10% > 50% « 200 90% 8 g 150 — mv * o 100 In M K 50 x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-r-x-x-x-*-*-*-*-*'*"* 0 10 20 30 40 50 60 70 350 (b) Natural Surfactant, 1% 300 10% I 250 > £ 50% gC 200 c— 90% s “ mv g 150 W-) fiR o 100 •X * X . 0 510 15 20 30 3525 40 Time (Minutes) Figure 5.8: Typical Variation in Size Distribution with Time for CGA Suspensions Generated with (a) 0.5% and (b)1.0% Natural Surfactant Solutions 82 and continuously re-circulated through a viewing cell and hence is considered to be appropriate for representing the CGAs from the applications stand point. The sample load in the system as indicated by the parameter, Dv starts with a high value about 1.0 - 3.0 depending on the concentration of natural surfactant and goes down exponentially with time. The size distribution was measured until Dv reaches a value of about 0.005. By this time, the bubbles generally become unstable and some bubbles become very big in size (400 - 700 fim) and eventually coalescence. The size distribution of CGA suspensions changes to a bimodal distribution wherein one mode is at the usual size 40-60 microns and the other one is at a larger size around 200 microns when the sample loading parameter, Dv becomes less than 0.01. The run was terminated at this time. A typical size distribution of the suspensions following a bimodal distribution is shown in Figure 5.9. This type of bimodal distribution observed for CGA suspensions generated with plant-based surfactant solutions was not reported in earlier studies (Sebba 1982, Longe 1989, and Chaphalkar 1994). As can be seen from the figure, the distributions are identical for all the concentrations, 0.5, 1.0 and 1.5%. About 60% of the bubbles have diameters less than 100 fim and the remaining bubbles have diameters between 100 and 500 /zm. This phenomena is suggested to be due to the Laplace pressure of the small bubbles. 5.3.3 Effect of Natural Surfactant Concentration on the CGA Size Distribution Figure 5.10 (a and b) shows the effect of natural surfactant concentration on the sizes 10% and 90% respectively. As can be seen from the figure, 10% size decreased 83 199- • ■ ' | TTT ------1 \i lur i, i ; i ! !: I >" ! 1 1 I 1 1 : 1(a).0.1% M ; I > 11 / i t i 1 : ; 1 i i 1 ..III' 1 I :iji/r i i : !' c r ! i •: i t ill/ i 1 u i i. ! t : i 1/ II H 1 ; ! ' j i .V I I I H E 1 i i i II " 1 • i i i i t > > i 1 i / r, ’ 111\ i l i i 1 ! • E i ' i i 11 i ■ i 1/ V i I u 11 \ | i i i » 1 i ' i i M u * / / I I V : till 1 1 i ' —Mil / A ! I ! ■ \ 1 ) ' I M ' 1 it 1 I 1 -r i | i ■ i i L ------i r 1 1 < 1 I m em o 49 C U n v U a 0 I L u II h R E E 199 59 9 1 T 69 V 59 V a L U n E Figure 5.9: Bimodal Size Distribution Curves for CGAs Generated with (a) 0.5%, (b)1.0%, and (c) 1.5% Natural Surfactant Solutions 50 8 4 (a) 40 eCO o Natural Surfactant Cone — °— 0.1% 20 u N -•*«-*- o.5% 1.0% 10 — *— 1.5% J------1------1______I______I______L 350 300 Cfle o o 250 ii " ? 200 * > N y O 6$ ’A oo\ 'Z 150 _N V3 100 50 0 10 20 3040 5060 70 Time (Minutes) Figure 5.10: Effect of Natural Surfactant Concentration on the (a) 10 Percentile and (b) 90 Percentile Size of CGA Suspensions 85 with increase in surfactant concentration but only marginally. As the concentration increased from 0.1 to 1.5%, the 10% size decreased from a mean of 35.5 to 31.5 pm. Tukey multiple comparison test with 95% confidence level was performed on the data. For 50% size also the increased concentration decreased the size (not shown). However, this significant decrease in size was noticed only for concentrations up to 1.0%. Beyond 1.0% the concentration did not significantly change the 50% size. The 90% size of CGA suspensions also showed a similar trend. An increase in natural surfactant concentration beyond 1.0% did not change the size significantly. The CGA suspensions with 0.1% natural surfactant have the lowest 90% size and 0.5% have the largest size which decreased when the concentration was increased. The CMC of plant-based natural surfactant is 0.1% and the bubbles are unstable at concentrations below CMC due to the non-availability of surfactant molecules to stabilize the enormous interfacial area of the microbubbles (Longe 1989). The significant differences in size distribution are due to the crowding of surfactant molecules on the bubble surface which in turn reduces the interfacial tension between the bubble and bulk water. This reduced interfacial tension reduce the bubble size. The same trend was seen for all the surfactant concentrations except 0.5 % in the case of 90% size. CGA suspensions generated with 0.5% natural surfactant did not follow the trend, instead the sizes were on the higher side at the beginning of the run and stabilized by the end of the run. This will be supported in the later sections by the fact that these suspensions are more stable than those generated with other natural surfactant concentrations. As the concentration of natural surfactant increased beyond 86 1.5%, the CGA suspensions became very thick, which prevented the air from entraining in the solution. This resulted in low gas fraction (poor quality) of the CGA suspensions. However, the suspensions are very stable as indicated by the longer liquid drainage time. Natural surfactant is a complex mixture of organic compounds which are believed to be responsible for the poor quality of the bubbles. The variation of sample loading parameter (Dv) with time for different concentrations of natural surfactant are plotted in Figure 5.11. As the concentration of surfactant increased from 0.1% to 1% the Dv value increased significantly. However, the changes were marginal when the concentration was increased from 1 to 1.5%. The bubbles generated with 0.5% surfactant are more stable and thus persisted for longer time than the bubbles generated with low concentration, 0.1 % or the higher concentrations, 1.0 and 1.5%. The Dv values decreased exponentially with time for all the suspensions. For CGA suspensions made with concentrations higher than 1.5%, the Dv value was very high (about 4) and took longer time to reach 0.005 where the run was stopped. This indicates that the bubbles are very stable, however, the quality (air content) of suspensions as determined from the drainage test is very low. As mentioned earlier the constituent compounds are thought to be responsible for this. 5.3.4 Effect of Electrolytes The effect of electrolytes on the size distribution of colloidal gas aphron suspensions was studied by adding 200 mg/1 and 400 mg/1 of sodium chloride to natural surfactant solutions before generating the CGA suspensions. The effect of salt Figure 5.11: Effect of Natural Surfactant Concentration on the Sample Loading Loading Sample the on Concentration Surfactant Natural of Effect 5.11: Figure Dv, Sample Loading 0.5 2.5 3.5 1.5 0 2 3 4 1 Parameter (Dv) Parameter 10 20 Time (Minutes) Time 30 Cone, of Natural Surfactant Natural of Cone, 40 — 1.5% *— — 50 0 1 . . 0 1 070060 % % 7 8 on the size distributions is summarized in Figure 5.12 (a and b) for 10% and 90% respectively. It is evident from the figures that the presence of electrolyte did not effect the size of the bubbles significantly. The 50% size also showed similar trend (not shown). Tukey multiple comparison test was performed with 95% confidence level and the results indicated no significant difference over the entire duration of the run. Lower amounts of salt (200 mg/1) used in earlier studies decreased the bubble size for ionic surfactants. This is due to the adsorption of one of the ions of the salt at the interface of the bubble along with the surfactant, resulting in the increase of the effective surfactant concentration (Longe, 1989). However, for non-ionic surfactants the electrolyte does not seem to affect the size distribution of the bubbles as the surface of the bubbles are neutral. The invariability in different sizes due to the addition of the salt indicate that the plant-based surfactant used to generate CGAs may be some what non-ionic in nature. The sample loading parameter, Dv was not affected when salt at concentration of 200 mg/1 and 400 mg/1 was used. However, there is a significant difference at 95% confidence level between the Dv value of the CGA suspensions generated with 200 mg/1 salt and 400 mg/1 salt. The Dv of CGAs generated with 400 mg/1 salt was lower at the beginning of the run but the value was after about 15 minutes and persisted for longer time indicating higher stability. 5.3.5 Comparison with CGAs Generated from Commercial Surfactants The size distributions of CGA generated with natural surfactant solutions are compared with those generated from commercial surfactants in our laboratory by 89 50 (a) 40 u c o 1 1% NS 20 1% NS+0.2g/l Salt 10 ■*«--- l% NS+0.4g/l Salt NS - Natural Surfactant i I _i______L 350 300 o 250 ~ 200 150 * 0. CO 100 c x 50 0 10 155 20 25 35 Time (Minutes) Figure 5.12: Effect of the Presence of Salt on the (a) 10 Percentile and (b) 90 Percentile Size of CGAs Generated with 1 % Natural Surfactant Solution Chaphalkar (1994). The results of this study are presented in Table 5.3 along with those for commercial surfactants, sodium dodecylbenzene sulfonate (NaDBS), Tergitol and hexadecyltrimethylammonium bromide (HTAB). The lower sizes (10% and 50% sizes) of the CGAs generated with natural surfactant correspond very well with the CGAs produced with the non-ionic surfactant, Tergitol. The ionic surfactants have sizes 15-60 microns larger for the 10% and 50% size. However, the 90% size of the CGAs in our study has significantly higher value than those of Tergitol, but these sizes are closer to those of ionic surfactants. This suggests that the fruit extract used for generating CGA suspensions could be a mixture of ionic and non-ionic compounds. 5.3.6 Stability of CGA Suspensions Stability of CGA suspensions is defined as its ability to resist change in bubble size, liquid content or degree of dispersion (Longe, 1989). Due to the constitution of the two phases of CGA suspensions, the bubble cannot retain its integrity without some external hydrodynamic force such as mixing or agitation. If agitation is discontinued, the bubbles tend to rise to the top and cream as a result of density difference between the gas and liquid phases. Longe (1989) suggested a simple method of gravity drainage under static conditions to determine the stability of CGA suspensions. CGA suspensions were filled into a graduated cylinder and the volume of the liquid drained was monitored with time. Stability was measured in terms of half life, the time need for 50% of the liquid content to drain. The total liquid volume collected by this method was also used for calculating the quality of the suspensions, i.e. the gas 91 Table 5.3. Comparison of the Size Distribution of CGAs Generated with Natural Surfactant and Commercial Surfactants (Chaphalkar, 1994) * Size Distribution of CGAs (microns) Surfactant 10% 50% 90% Natural Surfactant Concentration 0.1% 34 55 202 0.5% 33 65 305 1.0% 32 60 290 1.0% + 200 mg/1 salt 32 62 290 1.0% + 400 mg/1 salt 31 48 279 1.5% 31 52 295 DDBS Concentration Anionic 200 mg/1 50 123 244 500 mg/1 54 124 194 500 mg/1 + 200 mg/1 salt 44 98 187 750 mg/1 46 112 198 Tergitol Concentration Non-ionic 50 mg/1 31 58 106 100 mg/1 32 57 128 100 mg/1 + 200 mg/1 salt 32 58 123 1000 mg/1 34 63 112 HTAB Concentration Cationic 200 mg/1 52 152 258 328 mg/1 46 105 186 328 mg/1 + 200 mg/1 salt 42 83 171 500 mg/1 50 112 200 * The percentile distribution reported is at time = 2 minutes 92 fraction. Another method to determine the stability of CGA suspensions is to use the sample loading parameter, Dv, during the particle size analysis. The time required for the value of Dv to reach low values (0.005) can be used to represent the stability of the CGA suspensions. This method seems more appropriate as it is used under dynamic conditions and also with aqueous phase. In our study both the methods are used. A typical drainage curve for the CGA suspensions is shown in Figure 5.13. The figure shows the cumulative liquid volume with time for 0.5% natural surfactant concentration. Rate of drainage was large at the beginning but it slowed with time. The smaller bubbles appeared at the CGA/ liquid interface while the size increased towards the top. From this figure, half life is calculated and compared for CGA suspensions generated with different concentrations of natural surfactant and SDS. The advantages of using half life to represent CGA stability rather than the H-factor are discussed by Longe (1989). H-factor is defined as the height of the liquid column after one minute drainage in a 250 ml graduated cylinder (Suggs, 1987). This factor can be normalized by dividing with the final liquid height. The half lives of CGA suspensions generated with different concentrations of plant-based surfactant solutions are presented in Figure 5.14(a). The half lives for CGA suspensions generated with sodium dodecyl sulfate (SDS) are also presented. The figure shows that the stability of the suspensions increases with surfactant concentration. These results are also supported by the particles size analysis data presented earlier. The sample loading parameter, Dv, which represents the sample Figure 5.13: Typical Curve for Drainage of Liquid from CGA Suspensions CGA from Liquid of Drainage for Curve Typical 5.13: Figure Liquid Volume (ml) 20 40 50 30 60 90 10 70 80 0 0 ult = 67.1% = Quality 0.5% Natural Surfactant Surfactant Natural 0.5% 10 20 Time (Minutes) Time 30 40 —O • — Flask 1 Flask — ' Flask 2 Flask ' 50 Flask 3 Flask -o 60 93 iue51: oprsn f() h Saiiy n () ult o G Suspensions CGA of Quality (b) and Stability the (a) of Comparison 5.14: Figure Generated with Several Concentrations of Natural Surfactant and SDS and Surfactant Natural of Concentrations Several with Generated Quality (%) Half Life (Minutes) 20 70 80 12 16 10 0 (a) iP Z Vi © & Z CO Surfactant and Concentration and Surfactant IP irt Z tN w "Sb 73 + (N on tP "So Vi 73 + NS - Natural Surfactant Natural - NS 00 s eo Q QO CO CO e on m s Q CO E v/-\ cn to a w S to on E 4 9 95 material in the system was used by Chaphalkar (1994) to study the stability of CGA suspensions in a dynamic system. The CGA suspensions generated with 0.5% natural surfactant have the highest stability. The presence of salt seem to increase the half life. The stability of CGA suspensions generated with SDS solutions seem to be lower than that for CGAs generated with natural surfactant. Natural surfactant is a mixture of several glycosides and sugars and these constituents are believed to enhance the stability of CGA suspensions over those generated with commercial surfactants. The quality, defined as the gas fraction of CGA suspensions generated with different concentration of plant-based surfactant is shown along with that for SDS in Figure 5.14(b). The quality of CGA suspensions increased as the surfactant concentration increased from 0.1 to 0.5%, but as the concentration increased beyond 0.5%, the quality of the suspension decreased. CGA Suspensions with quality below 50% are not desirable as the CGAs loose air content and hence their distinguishing feature. The CGA suspensions generated with SDS solutions under the same conditions have a higher quality and the quality is more or less constant. The low quality of CGA suspensions generated with plant-based surfactant is attributed to the constituent complex organic compounds such as glycosides, sugars and lot of other compounds which may not have any significance to detergent properties of the extract. However, the main purpose of using a natural surfactant is to avoid the sophisticated, high energy intensive and waste generating manufacturing process of commercial surfactants. 96 5.4 Solubility of Hexachiorobenzene in Surfactant Solutions The main objective of this research is to study the suitability and limitations of natural surfactant solutions for flushing soils contaminated with hydrophobic organic compounds(HOCs) such as hexachiorobenzene. The first step is to determine the solubility of these HOCs in natural surfactant solutions and also to establish the variation of solubility of HOCs with natural surfactant concentration. Hexachiorobenzene, a chlorinated hydrophobic aromatic hydrocarbon is used as a test compound to study the solubilization phenomena of natural surfactant solutions. The solubility of HCB in natural surfactant solutions is compared to that in SDS solutions and other commercial surfactant solutions. The results of the solubility study will be utilized for selecting appropriate concentrations of the surfactant for the desorption and column flushing studies. Surface active agents form clusters known as micelles beyond critical micelle concentration (CMC). The hydrocarbon core of the micelles has the unique ability to solubilize many hydrophobic compounds which are otherwise only slightly or even insoluble in water (Tanford, 1980). The increase in solubility can be as high as thousand fold depending on the type of surfactant, hydrophobic compound and their interactions. 5.4.1 HCB Solubility in Natural Surfactant Solutions Natural surfactant stock solutions of 10%, 15% and 25% were prepared as discussed in the earlier section and 10% solution was used to make all the lower 97 concentrations ranging from 0.1 to 7.5%. Solubility of HCB was determined after equilibriating these surfactant solutions with excess quantity of HCB. All the concentrations were tested in duplicate flasks and each flask was analyzed in duplicate. The average values were used for regression and comparison to other surfactants reported in the literature. Figure 5.15(a) shows the variation of HCB solubility in natural surfactant solutions in the concentration range of 0.1% and 25%. Y axis shows the solubility of HCB in /ig/1 and X axis indicates concentration of natural surfactant in percent. The average of the four measurements is represented by a line. As can be seen from the figure, solubility of HCB increased by about thousand fold when 25% natural surfactant solution was used. Water solubility of HCB is about 50 /tg/1, which is in the range suggested in the literature (Montgomery, 1990). Reported HCB water solubility is ranging from as low as about 5 jxg/I (Miller et al ., 1985, and Keenaga et al. 1980) to as high as 110 /tg/1 (Metcalf et al. 1973). The solubility of HCB in 25% natural surfactant solutions was about 13000 /ig/1. There appears to be a linear relationship between solubility and natural surfactant concentration up to about 10% concentration. This linearity between the hydrophobic substance solubility and the surfactant concentration beyond CMC has been well established for commercial surfactants (Rosen, 1989, Edwards et a l , 1991, Liu etal. 1991, Roy etal. 1992b, Liu 1993, Jafvert et al. 1994, and others). The solubility of HCB beyond about 10% is not linearly increasing with concentration but follows saturation type curve. The Figure 5.15: Solubility of HCB in Natural Surfactant Solution: (a) Variation up to to up Variation (a) Solution: Surfactant Natural in HCB of Solubility 5.15: Figure 25% and (b) Correlation Between Solubility and Concentration and Solubility Between Correlation (b) and 25% HCB Solubility (ug/l) HCB Solubility (ug/1) 12000 10000 14000 4000 2000 5000 7000 4000 3000 2000 1000 6000 9000 6000 8000 8000 0 0 0 0 ■ Measured Points Measured ■ 2 5 Cone, of Natural Surfactant (%) Surfactant Natural of Cone, :egression Lines :egression 4 10 Average 6 15 8 pt 10% to up 20 10 25 98 99 possible reasons for this asymptotic value in solubility could be either that the micelles have a maximum capacity for HCB based on the molecular structure of HCB and surfactant or surfactant cannot be extracted efficiently for higher concentration solution. A 25% natural surfactant solution was prepared by adding 25 grams dry Ritha powder to 100 ml water. It was found that water cannot extract surfactant efficiently from the pericarp powder when such high concentrations are used. So it is not useful to prepare natural surfactant solutions at such high concentrations. However, natural surfactant concentrations up to 10% are successful in extracting Ritha efficiently. Only 10% stock solutions are prepared and used after appropriate dilution for the entire work. CMC of natural surfactant solutions was determined to be 0.1 % from surface tension and viscosity measurements. Regression analysis was performed on the solubility data to determine the solubility parameters. The analysis was done on three ranges of natural surfactant concentrations: 0.1 - 5%, 0.1 - 10%, and 0.1 -15% to determine the best range of concentrations where HCB solubility is linearly varying with surfactant concentration. Figure 5.15(b) shows the solubility data along with the regression lines for the concentration ranges 0.1 to 5%, and 0.1 to 10%. The regression in the lower range (0.1 - 5%) is fitting the data better for natural surfactant concentrations less than 5%. However this line gives much higher values than the measured values beyond 5%. The second regression was performed in the range of 0.1 and 10% and it gives higher values for the lower natural surfactant concentrations. However, this regression fits the measured values very well at higher concentrations. 100 The regression analysis on 0.1 to 15% range is off significantly at lower concentrations, even though it is linear at higher concentrations. Mandava (1994) used naphthalene as a test organic compound in her preliminary work on the application of natural surfactant solutions to soil flushing and reported that naphthalene solubility is increasing linearly with concentration in the range of 0.25 to 10%. From this study however, it is recommended that the range 0.1 to 5% should be used as the natural surfactant used for all practical applications is always less than 2.5%. The regression lines and the corresponding equations for HCB solubility in natural surfactant solutions along with the coefficient of determination (r2) are given below: Solubility = 119.1 +948.28 * (Nat.Surf.Cone.,%), r2 = 0.995, (for range 0.1 - 5%) Solubility=286.1 +808.3*(NatSurf.Cone.,%), r2 =0.992, (for range 0.1 - 10%) As can be seen from the above equations, the slope of the regression line decreases and the constant increases with increase in concentration range. The high constants for the second regression, cause the regression line to be higher than the measured HCB solubilities for lower natural surfactant concentrations. The slope of the regression line represent the maximum amount of HCB per mass of surfactant in equilibrium with solid phase HCB at standard temperature and pressure. These mass ratios can be converted to molar ratios if the molecular formulas are known. The molar solubilization ratio (MSR), defined as the number of moles of organic compound solubilized per mole of surfactant added to solution is commonly used to represent 101 solubility data (Edwards et al. 1991). However, the molecular formula for natural surfactant solutions is not determined and thus it is not possible to calculate MSR. Another approach in quantifying surfactant solubilization consists of characterizing the partitioning of the organic compound between micelles and monomeric solution with a mole fraction micelle - phase/aqueous-phase partition coefficient (K ^. The coefficient depends on the surfactant chemistry, solubilizate chemistry and temperature (Edwards et al, 1991). MSR and Km can be correlated with each other. This approach also can not be employed due to the lack of sufficient information on molecular formula. It was reported in the literature that low concentrations of dissolved and/or suspended particulate-bound natural organic matter can significantly enhance the solubility and stability of many hydrophobic organic compounds (Chiou et al. 1986, Wershaw et al. 1969, Hassett et al. 1982, Landrum et al. 1984 and others). It is believed that a partition-like interaction of the solute with the microscopic organic environment of dissolved organic molecules is responsible for the increased solubility of the solute. Natural surfactant being a mixture of saponins, glycosides, sugars and several other organic compounds may behave very similar to that of the natural dissolved organic matter and thus may improve the solubility of solutes. This can be in addition to that provided by the saponins, which are mainly responsible for the surfactant properties of the fruit pericarps of the trees belonging to this family (Gedeon 1954). 102 Natural surfactant being a fruit product is susceptible for biodegradation. One will suspect that solubilization of hydrophobic compounds by natural surfactant will be effected by the degradation of the constituent compounds of the surfactant. However, Mandava (1994) after studying the solubility of naphthalene in natural surfactant solutions over a period of 4 weeks concluded that there was only a marginal decrease in solubility over the period tested. She attributed this invariance in solubility to the inhibitive toxic action of the excess naphthalene on the microbial growth. Mandava (1994) also studied the effect of shelf time (storage time) and reported that naphthalene solubility does not decrease if natural surfactant solutions are properly refrigerated. She performed this study over a period of six weeks. In this research solubilization studies were always conducted with fresh natural surfactant solutions. 5.4.2 Effect of Natural Surfactant Sterilization on the Solubility of HCB The results of the preliminary experiment with HCB and those of Mandava (1994) indicated a significant increase in solubility of hydrophobic compounds in sterilized natural surfactant solutions. Natural surfactant being a fruit extract is prone for biodegradation and the degradation of the constituent compounds may decrease the solubility or the high temperature and pressure treatment that is given during the sterilization may breakdown natural surfactant into new compounds that may be more or less effective in solubilizing hydrophobic compounds than the original compounds. A study was conducted with 5 concentrations of natural surfactant and three treatments to study the effect of sterilization. In the first treatment natural surfactant solutions 103 were used without any treatment. For the second, the samples were steam sterilized. The third one consisted of autoclaved glassware and filter sterilized natural surfactant. The last two treatments will prevent any microbiological growth. The results of the study are summarized in Figure 5.16. This figure clearly shows that the results of the preliminary work are not correct and also the observations of Mandava (1994) are not true in our case. The absorption spectra for steam sterilized and non-sterile natural surfactant solutions indicated some differences. New peaks appeared at lower concentrations for sterile solutions than the non-sterile solutions indicating the possibility that the constituents are breaking into new compounds. However, the differences are not significant to cause any major changes in solubility. There is no significant difference in HCB solubility at 95% confidence level between steam sterilized and non-sterile natural surfactant solutions, even though the autoclave sterilized samples have solubilities lower than non-sterile samples. The filter sterilized samples have solubilities lower than both the autoclaved and non-sterile samples. The differences are significant only between the non-sterile and filter sterile samples. The reason for this could be the adsorption of surfactant by the filter which may reduce the effective concentration of the surfactant. However, about 100 ml of sample was filtered and the adsorption should not be a significant factor. 5.4.3 HCB Solubility in SDS Solutions SDS is a very commonly used surfactant in the laboratory for remediation of soils contaminated with hazardous wastes (Kommalapati 1994, Jafvert et al. 1994, Figure 5.16: Effect of Steam Sterilization and Filter Sterilization on the Solubility of of Solubility the on Sterilization Filter and Sterilization Steam of Effect 5.16: Figure HCB Solubility (ug/1) 2000 4000 3000 5000 7000 1000 6000 0 0 HCB in Natural Surfactant Solutions Surfactant Natural in HCB Filter * “ Autoclaved ' 1 Non-Sterile Cone, of Natural Surfactant (%) Surfactant Natural of Cone, 2 3 ons Measured Points: Lines: Regression Regression Lines: 4 5 6 104 Darji 1993, Jafvert and Heath 1991, Gannon et al. 1989, Kile and Chiou 1989, Valsaraj et at. 1988, etc.). This surfactant has been employed by Kommalapati (1994) in his application to flushing of non aqueous phase liquid (NAPL) from soil columns in the form of conventional solutions and colloidal gas aphron suspensions. In an unpublished work, Kommalapati (1993) applied SDS solutions for flushing oily hazardous wastes from undisturbed cores from a local superfund site. Currently CGA suspensions generated with SDS solutions are applied on a pilot scale study at a local superfund site (Roy et al. 1994). SDS is used in this study to compare the performance of natural surfactant solutions in solubilizing HCB and desorbing HCB from soil. Several other surfactants used in the literature for solubilizing HCB are also compared with natural surfactant solutions. Solubility of HCB in SDS solutions of several concentrations below and above CMC was measured in duplicate flasks. Figure 5.17 shows the variation of HCB solubility with SDS concentration. X axis shows the SDS concentration in g/1 and Y axis represents HCB solubility in //g/1. As can be seen from the figure there is a linear relationship between SDS concentration and HCB solubility beyond the CMC. A linear regression was performed between the solubility beyond the CMC and SDS concentration. The regression line has a high coefficient of determination (r2 = 0.990). The regression line is given by: Solubility = 460.65 * SDS Cone, (g/l) - 1523.26 iue51: ouiiyo C nSdu oey Slae SS Solutions (SDS) Sulfate Dodecyl Sodium in HCB of Solubility 5.17: Figure HCB Solubility (ug/l) 10000 12000 14000 000 4 2000 6000 8000 0 aurd Points P red easu M ■ 5 D Cnetain (g/I) Concentration SDS 10 15 geso line egression R 20 25 30 106 107 The equation can be changed to moles/liter by dividing the coefficient of SDS concentration in g/1 with the molecular weight. The constant term won't change. Maximum solubility obtained with 100 mM SDS solution is about 12 mg/1, which is about 250 times the water solubility. The solubilities obtained in our studies are in agreement with those reported by Jafvert et al. (1994) within the experimental error. However, it should be noted that the maximum concentration of surfactant used in their study was only 3.5 g/1. The correlation lines obtained in their study are based on just two data points. In this research, SDS concentrations as high as about 30 g/1 were used and also several concentrations were employed. Hence, for SDS solutions the values obtained in our study seem to be more appropriate and were used for comparison purposes. 5.4.4 Comparison Between Natural Surfactant Solutions and Commercial Surfactant Solutions Natural surfactant solutions were extracted from fruit pericarp and the fact that the extracts are not purified indicates that there are several other water soluble compounds which may not necessarily contribute towards the surfactant properties of these solutions. Dry weight of Ritha powder used for extraction is taken to represent the strength of natural surfactant solutions. For example a 10% solutions would mean that 10 grams of Ritha powder was extracted into 100 ml water. But, only about 70% of this powder (7 grams) was extracted into water and about 30% (3 grams) remains 108 as residue. The strength of natural surfactant solution should have been 7% rather than 10%. However, dry weight of Ritha powder is used to designate the strength of natural surfactant solution as it denotes the weight of raw Ritha that is to be weighed to prepare the necessary solutions. The comparison of natural surfactant solutions with commercial surfactant solutions is thus made based on the dry weight of Ritha powder rather than the dissolved net weight of extracted powder. The maximum solubility of HCB, 13 mg/1 for natural surfactant solutions was with 25% natural surfactant concentration. The maximum SDS concentration used, 100 mM (28.8 g/1) has solubilized about 12 mg/1 HCB. Natural surfactant solutions behaved exactly like commercial surfactant in terms of the linear relationship between surfactant concentration and solubility. Jafvert et al. (1994) used about 10 surfactants to solubilize HCB in aqueous solutions. They used very low concentration of surfactants for their experiments. However, the solubility parameter (millimoles of HCB per mole of surfactant) reported in their study is used here to compare the performance of natural surfactant solutions. The comparison between different surfactant will be done on the basis of surfactant required in one liter of water to solubilize a given amount (1 mg) of HCB. Grams of surfactant required in one liter water to solubilize 1 mg of HCB was calculated for the 10 surfactants from the data reported in the literature (Jafvert et al. 1994) and presented in Figure 5.18 along with the results from this study. As can be seen from the bar diagram, Brij 30, POE 10-LE and Tween 85 seem to be the most effective to solubilize HCB. Only a fraction of a Figure 5.18: Figure Surfactant (g) per Liter 12.00 16.00 0.00 4.00 8.00 Comparison of HCB Solubilizing Capacities of Natural Surfactant Surfactant Natural of Capacities Solubilizing HCB of Comparison Solutions and Commercial Surfactants as Reported by Jafvert Jafvert by Reported as (1994). Surfactants Commercial and Solutions n i - I i :E* CO o CO oo m £ c o Cu O W m i 13 cu in P m X H s I Surfactant o CN I c '•E in CO ca •c P X *7 in *—• in o e *5 .y M3 E in Prf 9 s oo co co Q CO This Study This CO a CO . N • M 1 cm a M u t al. et 109 110 gram of surfactant was needed to solubilize 1 mg of HCB in one liter water. About 2 to 3 grams of Brij 35, Tween 20, Tween 80 and Exxal F 5715 were sufficient to do the same job. Triton X-705 and Plutonic P-65 in large quantities (51 grams and 83 grams respectively) were required to solubilize 1 mg HCB in one liter solution. About 10.5 grams of Ritha and about 5.5 grams of SDS were required to solubilize 1 mg hexachiorobenzene. It should be noted that only 70% of Ritha is dissolved in water, which makes the net Ritha required to be about 7.5 grams. This clearly shows that natural surfactant solutions are comparable to other commercial surfactants in solubilizing a chlorinated hydrophobic organic compound, HCB. 5.5 Batch Desorption Studies Desorption studies are used to evaluate the efficiency of natural surfactant solutions in desorbing a chlorinated hydrophobic aromatic hydrocarbon, hexachiorobenzene from soil. The experiments were conducted with 6 different initial contamination levels and several concentrations of natural surfactant ranging from 0.1 to 2.5 %. The reason for limiting the higher surfactant concentration to 2.5 % was due to the fact that CGA suspensions can not be generated with such high concentration solutions and also that concentrations above 2.5% tend to clog the soil columns during flushing. SDS solutions were also employed in desorption studies to compare the performance of natural surfactant solutions. The amount of HCB desorbed from soil and solubilized by natural surfactant and SDS solutions was estimated as a percentage I l l of HCB initially present in the soil and is reported as percent recovery. Desorption isotherms are also presented. 5.5.1 Desorption Studies with Natural Surfactant Solutions Figure 5.19 (a) shows the plot between aqueous phase HCB concentration and natural surfactant concentration for different soil contamination levels. For higher contamination level, about 90 mg/kg, the aqueous phase HCB concentration was approaching the solubility of HCB in the respective surfactant solution. For soil contamination of 30 mg/kg the solubility of HCB approached for natural surfactant concentrations up to 1.5%. For lower contamination levels, practically all HCB was removed from the soil. It is clear from the figure that the solubility of HCB in natural surfactant solutions was the limiting factor and by repeating the washing process one would be able to clean the soil to a significant level. Figure 5.19(b) shows the recovery of HCB from contaminated soil by natural surfactant solutions for several initial HCB contamination levels. Y axis shows the recovery of HCB in percent of total initial HCB present on the soil. For the lowest contamination level employed, 0.6 mg/kg, about 90% of HCB was recovered from soil. Even though the contamination level was less than that can be solubilized by the natural surfactant concentrations used, only 90% of natural surfactant was desorbed. It is believed that irreversible adsorption, volatilization losses and other factors are responsible for not being able to desorb all the adsorbed HCB (Pardue 1992 and Mandava 1994). For other contamination levels, 1.6, 17 and 33 mg/kg about 70, 60 and 50% of HCB was iue51: eopin f C fo Si wt Ntrl ufcat Solutions: Surfactant Natural with Soil from HCB of Desorption 5.19: Figure HCB Desorbed from Soil (9b) Aqueous HCB Concentration (ug/1) 2000 2500 3000 1000 1500 500 70 0 " Contamination (mg/kg) Contamination Variation of (a) Aqueous Phase HCB and (b) Percent Desorbed Percent (b) and HCB Phase Aqueous (a) of Variation —06 ( 0.6 Q— — — 33 *— — Contamination (mg/kg) Contamination Ut — 1.6 o— — — — x— — / / ° 0.6 — 0.5 6 L 3 3 Cone, of Natural Surfactant Natural of Cone, 1 (a) . * • • A * . 1.5 2 2.5 113 recovered respectively from the soil. For the highest contamination level used, 90 mg/kg the percent recovery was varying linearly with natural surfactant concentration as was the case for solubility studies. These aqueous phase HCB concentrations were within 10 - 15% of HCB solubility in the respective natural surfactant solutions. Similar observations were made by Mandava (1994) for desorption of naphthalene from soil with natural surfactant solutions. Desorption isotherms were constructed from the experimental data with the Y axis representing HCB remaining on the soil and X axis showing the aqueous phase HCB concentration. The isotherms for natural surfactant solutions are shown in non linear form in Figure 5.20. As can be seen from the figure, the isotherms for all the concentrations are concave upwards throughout, which is an unfavorable condition for adsorption or in other words favorable situation for desorption (Wark and Warner, 1981). Presence of surfactant solutions reduces the adsorption of the hydrophobic organic compounds onto the soil (Edward et al. 1994). The soil used in our study has very low organic matter content (0.25%) and thus offer no resistance for the desorption of HCB. Natural surfactant solutions desorb HCB from soil and solubilize into their micelles. This is indicated by the fact that at low solid phase concentration, the aqueous phase concentrations are low and at higher solid phase HCB concentrations the aqueous phase concentrations are limited by the solubility of HCB in the respective solutions. Similar observations were made by Mandava (1994) for natural surfactant solutions with naphthalene as the hydrophobic organic compound. Regression analysis performed on these isotherms indicated that the isotherms can best be described by Figure 5.20: Desorption Isotherms for HCB in Natural Surfactant Solutions in Non- Non- in Solutions Surfactant Natural in HCB for Isotherms Desorption 5.20: Figure X/M, HCB Remaining on Soil, ug/g 100 20 40 30 50 90 70 60 80 10 0 50 00 50 00 50 00 50 4000 3500 3000 2500 2000 1500 1000 500 0 Linear Form Linear , qeu hs C Cnetain ug/1 Concentration, HCB Phase Aqueous C, Cone, of Natural Surf. Natural of Cone, 1.5% - 2.5% 0 . 1 % 114 115 either power law or exponential law. The power of the aqueous phase concentration in the power law is in the range of 1.3 to 1.6 and the coefficient of determination (r2 ) for the regression is in the range of 0.87 and 0.96. The exponential curve fit yielded a r2 value in the range of 0.81 to 0.96. Figure 5.21 (a and b) show the Freundlich and Langmuir isotherms respectively for three contamination levels. Freundlich isotherm is plotted on a log-log graph with HCB remaining on the soil on Y axis and the aqueous phase HCB on the X axis (Freundlich, 1926). It is clear from the plot that there is a linear relationship between the HCB remaining on the soil and the aqueous phase HCB concentration. This is particularly true for high contamination levels. When the contamination is reduced the linear correlation does not seem to be followed. It is believed that the mass transfer of HCB from soil to aqueous solutions is the limiting factor. At lower contamination levels when high surfactant concentrations are employed there is not enough HCB that can be desorbed. This is indicated by the drop in the isotherm (Figure 5.21a). Langmuir isotherm (Figure 5.21b) is plotted with reciprocal of HCB remaining on the soil on the X axis and the reciprocal of aqueous phase HCB concentration on the X axis (Langmuir, 1971). The same argument used for Freundlich isotherm seem to be valid for Langmuir isotherms also. 5.5.2 Desorption Studies with SDS Solutions Figure 5.22(a) shows the plot between aqueous phase HCB concentration that is desorbed from soil and concentration of SDS for different levels of soil 116 100 (a) Freundlich Isotherm Contamination (mg/kg) ’oco c o BJ} ‘c.E *e3 uE Pi - 4 ca o s 5 X 0.1 1 10 100 1000 10000 C, Aqueous Phase HCB Concentration (ug/1) 0.16 0.14 c o (b) Langmuir Isotherm 0.12 CO'o e o CO 0.1 c c e 0.08 ca 0.06 u as 1s 0.04 1 0.02 0.01 0.02 0.03 0.04 0.05 0.06 1/C, Aqueous phase Concentration (ug/1) Figure 5.21: Desorption Isotherms for HCB (a) Langmuir and (b) Freundlich 117 1800 Contamination (mg/lcg) 1600 1400 1200 1000 800 600 400 200 Contamination (mg/kg) 90 ° — 0.6 80 70 60 50 40 30 20 10 0 10 15 20 25 30 35 SDS Concentration (mM) e 5.2 Desorption of HCB from Soil with SDS Solutions: Variation of (a) Aqueous Phase HCB and (b) Percent Desorbed 118 contamination. For SDS solutions also the same trend as followed by natural surfactant solutions was observed. For low contamination levels, aqueous phase HCB was significantly less than the HCB solubility in the respective solutions. For higher contamination levels there was a sharp increase in the aqueous phase concentration of HCB from 8 mM to IS mM SDS concentration. This sharp increase was also noticed in solubility studies. It should be noted that CMC of aqueous SDS solutions is 8 mM, however, the presence of soil significantly alters the CMC of surfactants (Jafvert 1991 and Liu et a l 1992). The CMC of SDS in soil-water systems will be higher than 8 mM and thus there should be a significant increase in HCB recovery from soil when SDS concentration is increased from 8 to 15 mM. The aqueous phase HCB concentrations were within 10 - 15 % of the HCB solubility in respective solutions at higher contamination levels. Figure 5.22(b) is plotted with percent of HCB desorbed from the soil based on the initial contamination on the Y axis and natural surfactant concentration on X axis. For a lower contamination level, about 75% of HCB was recovered compared to 90% by natural surfactant solutions. Volatilization losses and irreversible adsoiption were thought to be responsible for the unavailability of the remaining adsorbed HCB. For other contamination levels, 1.6, 3.9, and 17 mg/ kg about 60 - 70% was recovered and for the higher contamination levels 33 and 93 mg/kg about 40 and 15% was recovered. There is a linear relationship between the recovery and surfactant concentration as noted in case of natural surfactant solutions. It is believed that by 119 repeating the washing process one would be able to recover significant amounts of HCB from soil and thus clean the soil. 5.5.3 Comparison Between Natural Surfactant and SDS Solutions As discussed in the last two sections of the desorption studies, natural surfactant solutions and SDS solutions have shown similar behavior in desorbing HCB from soil. The isotherms for natural surfactant solutions exhibit concavity upwards, which indicates that the systems are favorable for desorption (Figure 5.20). SDS solutions also exhibited similar upward concavity when non-linear isotherms were plotted (not shown). Both natural surfactant and SDS solutions were able to desorb as much as 90% of the total HCB on the soil for low contamination levels and desorbed about 90% of the solubility of HCB in the respective solutions with soils contaminated to higher level. This study clearly suggests that natural surfactant solutions are comparable in performance to commercial surfactants in solubilizing and desorbing hydrophobic compounds and should further be investigated. 5.6 Application of Natural Surfactant Solutions to Soil Flushing The results of solubility and desorption studies established that natural surfactant solutions are comparable to other commercial surfactants. In this section, flushing experiments with one dimensional columns were conducted to appraise the applicability of natural surfactant solutions and the CGA suspensions generated from these solutions to soil flushing. Three surfactant concentrations, 0.5, 1.0, and 2.5% and CGA 120 suspensions generated from these solutions were employed with soils contaminated with different amounts of HCB. Another study to evaluate the effect of alternating the natural surfactant solutions and CGA suspensions with water on the removal of HCB and the pressure build-up across the soil column was also conducted. 5.6.1 Flushing of Soils Contaminated to Different Levels o f HCB Results of the column flushing experiments are presented in Figures 5.23 and 5.24 respectively, for 0.5% and 1% natural surfactant in the form of conventional solutions and CGA suspensions. Y axis shows the cumulative removal of HCB from soil columns in ftg and the X axis represents the number of pore volumes collected. The columns were packed with soil contaminated to a level of 1.6, 70 - 80, and 100 mg HCB/kg soil. As can be seen from these graphs, depending on the contamination level, natural surfactant solutions were able to recover as much as 80% of HCB solubility in the respective solutions. If the contamination levels were low ( 2 mg/kg), the removals were limited by the mass transfer from the adsorbed phase to the aqueous phase. The removal of HCB during the first pore volume after saturation of the column was practically negligible. HCB recovery started increasing as more and more surfactant was pumped through the column in the form of either conventional solutions or CGA suspensions. This is because during the first pore volume the saturation water was replaced with surfactant solution and the effluent contained only water. Surfactants are known to undergo adsorption with soil (Liu et al. 1992, and Liu 1993) iue52: eoa o HB rm ol oun Uig a 0.5 (a) Using Columns Soil from HCB of Removal 5.23: Figure Cumulative Removal of HCB (ug) Cumulative Removal of HCB (ug) 100.00 120.00 160.00 180.00 140.00 100.00 200.00 120.00 140.00 160.00 180.00 20.00 40.00 60.00 80.00 20.00 40.00 60.00 80.00 0.00 2 6 1 1 14 12 10 8 6 4 2 0 Surfactant Solutions and (b) 0.5% CGA Suspensions CGA 0.5% (b) and Solutions Surfactant - - ■"**■* Contamination Contamination - U o- — 7mg/kg 67 113mg/kg 1.6 7 6 0.6 i i 6 3 mg/kg mg/kg mg/kg mg/kg N o. o f Pore Volumes Pore f o o. N mg/kg 0 * n OJ * 0 0*9 9 9 / P » 9 0 9 ---- b / (b) 09 0 * 90 (a) *9 i f % 0O j 0 tf/ • < • « * # A 09 9 0 9 0 P / 0f * ------0 9 J 0 0 g 9 0 m J W * a a / . * • . . 9 0 0 / a a 9 0 a a a 0 * a J 0 tr 0 i # 1 o <* / P i m a 0 0 9 / a 0 09 0 a 0 9 0 9 a * a 0 - - 0 a °* r o A ? 0 * 0 0 0 * Natural 121 Figure 5.24: Removal of HCB from Soil Columns Using (a) 1.0% Natural Surfactant Surfactant Natural 1.0% (a) Using Columns Soil from HCB of Removal 5.24: Figure Cumulative Removal of HCB (ug) Cumulative Removal of HCB (ug) 1 0.0 “+ 2*k /// / / 92m*/kg 200.00 +~ “ - 400.00 * 500.00 300.00 - - 300.00 0.0 — 16m/g / mg/kg 1.6 — - 700.00 0.0 Contamination * 900.00 0.0 - 600.00 800.00 * “ ‘ 0 — 0.6 mg/kg 0.6 — 0 ‘ “ * 800.00 0.0 - 100.00 , 200.00 300.00 400.00 500.00 100.00 600.00 700.00 900.00 800.00 000.00 0 0 . 0 / - Solutions and (b) 1.0% CGA Suspensions CGA 1.0% (b) and Solutions 2 6 1 1 14 12 10 8 6 4 2 0 ------1. I 1 ------♦■*-54 *o— o -* - * o — 82 mg/kg 82 * — o - 92 mg/kg 92 ~~a— Contamination °“ 92 mg/kg 92 °“ 1 Contamination ' ' S " . / 2m/g / < / / mg/kg 82 78 mg/kg 78 6 . 1 4 5 6 . 0 i.gnjg/jjg l rr OF O r r Cl ,T mg/kg / ' / mg/kg mg/kg # - * * # / / mg/kg N o. o f Pore Volumes Pore f o o. N / mg/kg ' A / / .«'** P.* / «•* .a ' p y / A / / / / (b) (b) T T ■p/ + / ,+ / / - # / ------*a +.+ a .-••a* ■ / ° / p q I ------123 and natural surfactant being a mixture of organics can adsorb to soil significantly. It took about three to four pore volumes for the surfactant breakthrough to occur. HCB concentrations increased steadily after the first pore volume and approach the maximum HCB concentration by the fifth pore volume. This maximum concentration was about 80% of the HCB solubility in the aqueous natural surfactant solution of the corresponding concentration. HCB concentration in the effluent remains more or less constant for the remainder of the experiment. The experiments were however, terminated at 12 pore volumes after establishing the HCB removal trend. All the experiments were continued for 12 pore volumes without stopping the runs. However, in the case of CGA suspensions the experiments were stopped run for 6 to 8 pore volumes before stopping and continued the following day. There was no remarkable difference in the recoveries due to the break in the experiment. It should be noted that CGA runs take about 30 hours as opposed to about 12 hours for natural surfactant solutions. The reason for the longer times in case of CGA suspensions was that CGA suspensions are only about 35 % liquid and it takes three times the time to collect one liquid pore volume. It took between 2 and 3 hours to collect one pore volume with CGA suspensions. Total HCB recovered after flushing with natural surfactant solutions or CGA suspensions for 12 pore volumes was about 200 ug for 0.5% and about 1000 fig for 1.0% natural surfactant. These amounts are about 20 and 100 times more than that recovered with water (10 ug) from soil columns contaminated to about 100 mg HCB / kg soil. This significantly higher removal of HCB is very encouraging for the fact 124 that natural surfactant solutions can be used as a supplement at the existing "pump and treat" facilities to enhance the performance. This will cut down the time required for clean-up of the hazardous waste sites from several decades to years and hence lower the treatment costs significantly. Soil was analyzed at the end of the column run and mass balance on HCB was performed to study the distribution of HCB in the column after flushing. The total HCB recovered in the effluent and HCB remaining on the soil after the column run were able to account for about 90% of the HCB in the column. The remaining 10% HCB is believed to be lost due to volatilization losses from aqueous phase effluent and from the soil and also any other error due to the non-homogeneous distribution of HCB on soil. HCB is a semi-volatile compound and volatilize significantly from aqueous solutions (Montgomery, 1990). The good mass balance obtained in this study provided confidence in our analytical techniques and experimental procedures. HCB remaining in the soil column was quantified by slicing the column into four approximately equal sections and each section was analyzed independently. For both CGA suspensions and natural surfactant solutions, the removal of HCB was mainly from the influent end and the HCB present at the effluent end did not change considerably. This indicates that the HCB removal starts from the soil at the influent end first and the removal continues towards the effluent end. In some cases, HCB concentration at the effluent end was higher than the initial contamination indicating that the desorbed HCB may be getting re-adsorbed along with the surfactant during its transport towards the effluent end. 125 The concentration of natural surfactant in the effluent was monitored using the IJV absorption method. The effluent samples were centrifuged and the concentration of natural surfactant was measured using the correlations discussed in section 5.1. Figure 5.25 shows the variation of natural surfactant concentration in the column effluent. As can be seen, natural surfactant concentration in the effluent takes about 3 to 4 pore volumes to reach a breakthrough and surfactant concentration remained constant beyond that. The delay in the breakthrough indicates that the natural surfactant solutions are interacting with the soil and in the process are getting adsorbed. Another important factor could be that the constituents of natural surfactant are undergoing preferential adsorption with soil. The effluent natural surfactant concentrations calculated using the correlations at different wavelengths yielded different results. This suggests that the certain components are getting adsorbed more than the other. Surfactants tend to adsorb to the soil surface due to their charge, dipole interaction and the presence of organic matter (Liu et al. 1992 and Liu 1993), Sorption of any surfactant on to the soil tend to retard the transport of surfactant and also decrease the amount of surfactant available for micellar solubilization (Liu et al, 1992). The breakthrough curves for natural surfactant solutions obtained in this study are similar to those obtained by Kommalapati (1992) with uncontaminated soil columns and SDS solutions. He studied the adsorption of SDS by soil during the column runs with both CGA suspensions and conventional surfactant solutions. Natural surfactant solutions in the form of conventional solutions and CGA suspensions undergo similar adsorption with soil. This is indicated by the similarity in the trend of the Figure 5.25: Concentration of Natural Surfactant in the Column Effluent Column the in Surfactant Natural of Concentration 5.25: Figure Cone, of Naural Surfactant in the Effluent 0.4 0.2 0.6 0.8 2 6 1 1 1 16 14 12 10 8 6 4 2 0 ♦ NS- Natural Surfactant Natural NS- ♦- . ■ « [ i i— ’ i * No. of Pore Volumes Pore of No. --o--- + 0.5 0.5 %NS + .% CGA 0.5% % NS 1 % CGA 1 126 127 breakthrough curves. Roy et al. (1992c) also reported that the CGA suspensions and surfactant solutions are transported through the column in much the same manner. 5.6.2 Comparison Between CGA Suspensions and Natural Surfactant Solutions Natural surfactant solutions with a concentration of 0.5, 1.0, and 2.5% and CGA suspensions generated with 0.5 and 1.0% were used to flush the soils contaminated with HCB. The effect of natural surfactant concentration on the removal of HCB from soil columns is shown in Figure 5.26 (a and b) for contamination levels 1.6 mg/kg and 100 mg/kg respectively. These figures also show the comparison between CGA suspensions and conventional natural surfactant solutions. As can be seen from the figures natural surfactant solutions have better performance than the CGA suspensions generated with the corresponding natural surfactant solutions, even though the differences are marginal. These results are in contrast to those reported in the earlier studies on CGA suspensions (Kommalapati 1994, Darji 1993, Roy et al. 1992b, and others). Kommalapati (1994) and Darji (1993) used SDS solutions and CGA suspensions generated from SDS solutions to recover residual levels of non aqueous phase liquids (NAPLs) from soil columns and found CGA suspensions to be better than the surfactant solutions. Roy et al. (1991) used 2,4-dichlorophenoxyacetic acid (2,4-D) and found CGA suspensions to be as efficient as surfactant solutions. Kongara (1994) however, used several surfactants and CGA suspensions generated from those to flush soils contaminated with naphthalene and reported that surfactants are better in recovering the contaminant than the CGA suspensions, even though the 128 120 Flushing Media 100 ° — 1% Nat. Surf. CQ U ' 0.5% Nat. Surf. K o • 1% CGA >O - 0.5% CGA Flushing Media =— 1 % Nat. Surf U 700 *” * 0.5% Nat. Surf <>•** 1% CGA °--- 0.5% CGA JS 300 4 6 8 10 12 14 No. of Pore Volumes Figure 5.26: Effect of Natural Surfactant Concentration on the Removal of HCB from Soil Columns (a) 1.6 mg and (b) 100 mg HCB/Kg Soil 129 differences are not very significant. This study also shows that the CGA suspensions are not as effective as natural surfactant solutions, however, the differences between the two are only marginal. It should be noted that the two studies mentioned above (Kommalapati 1994, and Darji 1993) where the CGA suspensions have better performance than surfactant solutions, the contaminant is a NAPL. In this study and that of Kongara (1994) the test organic is in an adsorbed phase. 5.6.3 Effect of Surfactant Concentration Figures 5.26 (a and b) depict the effect of surfactant concentration on the removal of HCB for both surfactant solutions and CGA suspensions for two levels of soil contamination. It is veiy clear from the figures that with the increase in the concentration of natural surfactant there is a significant increase in the removal of HCB from soil columns. The same is true for CGA suspensions. Surfactant solutions can solubilize more and more of the hydrophobic compounds as the surfactant concentration is increased. The higher the surfactant concentration, the higher is the amount of organic that can be solubilized. Similar trends were observed for all the contamination levels. When the natural surfactant concentration was increased from 1.0 to 2.5% the removal also increased significantly. However, the higher surfactant concentrations tend to clog the soil columns resulting in high pressure build-up and eventually in terminating the run. It should be noted that CGA suspensions cannot be generated with natural surfactant solutions beyond 1.5%. 130 5.6.4 Pressure Build-up Across the Soil Columns The pressure build-up phenomena often seem to be the controlling factor that determines the application of surfactants at the field scale (Nash, 1987). Figure 5.27 shows the variation of pressure build-up for different flushing media with the number of pore volumes. As can be seen from the figure the pressure variation is about the same for all the runs except for the one with 2.5% concentration. The general trend is that, pressure build-up increases with increase in surfactant concentration. However, the increase in pressure was not very significant when the concentration is increased from 0.5 to 1 %. The pressure increased gradually and approached about 60 psi by the end of 8 pore volumes in case of 2.5% natural surfactant. When the concentration was increased to 5% the pressure increased to 60 psi in 3 to 4 pore volumes (not shown in the plot) and the run had to be terminated. It is believed that the interaction of natural surfactant solutions with the soil are responsible for the increase in pressure. This behavior was reported in the literature (Ang and Abdul 1991, and Liu 1993, Kommalapati 1992 and 1994). However, it is not known at this time the specific interactions that are responsible for the increase in pressure build-up. There was no remarkable difference in pressure build-up between the CGA suspensions and natural surfactant solutions. This observation is in contrast to those made by Kommalapati (1994) with SDS solutions and automatic transmission fluid and Darji (1993) with SDS solutions and a heavy oily waste. They observed lower pressure build-up when CGA suspensions at the same concentration as surfactant solutions are used for flushing the Figure 5.27: Pressure Build-up Across the Soil Columns for the Flushing Experiment Flushing the for theSoil Build-up Columns Across Pressure 5.27: Figure Pressure Build-up (Psig) 20 30 40 50 10 60 0 4 1 16 12 8 4 0 NS - Natural Surfactant Natural - NS No. of Pore Volumes Pore of No. - X- Flushing Media Flushing .% NS 0.5% - .% NS 2.5% CGA 0.5% 1% NS 1% 131 132 residually saturated contaminated soil columns. However, they also noted that the pressure increased with increase in the surfactant concentration. The pressure build-up across uncontaminated soil columns has similar trend indicating that the pressure build up does not depend on the type of contaminant (Roy et al. 1992). The interactions between the soil and the surfactant, surfactant and the contaminant seem to be responsible for the pressure build-up. Ang and Abdul (1991) proposed that the formation of floe and micelle and also the presence of any fine suspended particles as the reason for pore clogging. The formation of "mat" due to the blocking of flow paths along a continuous front can halt fluid flow and results in terminating the run (Nash 1987). By increasing the surfactant concentration the solubility of the contaminants can be increased and thus increase the recovery, but the pressure build-up limits the concentration of the surfactant that can be employed. There should be a compromise between the two. Some of the columns were flushed with natural surfactant solutions and CGA suspensions for 16 pore volumes and the pressure build up seem to be stable after 4 to 5 pore volumes. Natural surfactant at a concentration of 1 % seem to be a very good compromise. CGA suspensions can be generated from natural surfactant solutions at this concentration, pressure build-up does not create any problem and the HCB removal was about 100 times more than that of water run. As will be seen from later sections natural surfactant can also be used as a carbon source at this concentration by soil microorganisms. 133 5.6.5 Effect of Alternating the Flushing Media with Water Surfactants when pumped at higher concentrations tend to clog the pores and some times the run has to be terminated due to the high pressure build-up. The CGA suspensions or surfactant solutions were used for flushing in a continuous mode. However, surfactant breakthrough occurs after about 3 to 4 pore volumes and the recovery of the contaminant becomes more or less stable. It is interesting to see whether one can avoid the pressure build-up problem by manipulating the flushing media. We may even be able to reduce the amount of surfactant needed by alternating the expensive surfactants and CGA suspensions with water. These manipulations should first be studied in the laboratory before one can use at the real sites. In this study two concentrations of natural surfactant, 1% and 2.5% and CGA suspensions generated with 1 % surfactant were alternated with water and the HCB removal and the pressure build-up phenomena are studied. The results of the study are summarized in Figure 5.28 (a and b) respectively for 1% and 2.5% natural surfactant. The soil was contaminated for all the runs in one batch to avoid any variation due to soil contamination. The contamination used was about 92 mg HCB /kg soil. Natural surfactant solutions recovered more HCB from soil columns in 16 pore volumes than the CGA suspensions of same concentration in a continuous mode of flushing (Figure 5.28a). Higher concentration of natural surfactant (2.5%) was able to recover more HCB than 1% natural surfactant solutions (Figure 5.28b). The alternate run with 2.5% natural surfactant and water followed the same trend as the 1 % run. The continuous mode of operation of the flushing media Figure 5.28: Effect of Alternating Flushing Media with Water on HCB Removal Removal HCB on Water with Media Flushing Alternating of Effect 5.28: Figure and 2.5% Natural Surfactant Natural 2.5% and rm ol oun () 1 (a) Columns Soil from Cumulative Removal of HCB (ug) Cumulative Removal of HCB (ug) 2000 2500 3500 3000 1000 1000 1500 1200 1400 200 400 600 800 500 > t S Water NS* 1 ft ■- —>* 3 — Flushing Media Media Flushing Flushing Media Flushing ----- ♦•-1ft • tt A G tftC *•• ■ ------IR NS* Water NS* IR - R NS 1R Water . t NS ft 2.5 . NS'Water R 2.5 NS * 1 Water CGA'Water 1* CGA'Water % aua Sratn ad Gs n () 1 (b) and CGAs and Surfactant Natural No. of Pore Volumes Pore of No. Solid Arrows: Flushing Media Flushing Arrows: Solid Dotted Arrows: Water Water Arrows: Dotted ■ ' NS - Natural Surfactant Natural - NS ■' 134 % 135 recovered twice as much HCB as the alternate runs recovered. In the alternating mode also the natural surfactant solutions were giving higher recoveries of HCB than the CGA suspensions under similar conditions. However, it should be noted that when the flushing media is alternated , only half of the surfactant is pumped. For example for a 16 pore volume experiment 9 pore volumes of surfactant was pumped and 7 pore volumes of water was used. HCB recovered at the end of 16 pore volumes was used to calculate the removal per gram of natural surfactant. The values are tabulated in Table 5.4. The table clearly indicates that more contaminant can be recovered per gram of natural surfactant with continuous flushing than alternating runs for all the flushing media. In the alternate mode of operation, natural surfactant solutions or CGA suspensions were pumped for about three pore volumes and then switched to water. The first pore volume of water after switching basically replaces the surfactant that was already pumped and thus the recovery does not reduce from that of earlier pore volume. However, for the second pore volume the amount of surfactant present in the effluent was low and thus could recover only small quantities of HCB. When the flushing media was changed back to natural surfactant solutions or CGA suspensions, the first pore volume basically replaces water and the recoveries become lower than the earlier pore volumes. By the second pore volume the recoveries will start increasing and the continue to increase for one pore volume even after switching to water. The process was repeated as the flushing media was switched after every two pore volumes. This can be seen from both the figures with a rise in the cumulative HCB removal curve for about three pore volumes and followed by a 136 Table 5.4: Comparison of Continuous and Alternate Flushing Description HCB recovered in 16 Recovery of HCB (pig) pore volumes (pig) per gram of natural surfactant 1% Nat. surf. 1321 68.8 1% Nat. surf, -w ater 512 48 1% CGA 1057 55 1% CGA —water 375 35 2.5% Nat. surf.1 3302 100 2.5% Nat. surf.—water* 1275 39 f Column runs were terminated after 11 pore volumes due to high pressure 137 stationary curve. Tamayo (1991) alternated CGA suspensions and surfactant solutions with water for flushing 2,4-D from soil columns and observed that there is no significant difference when the surfactant solutions or CGA suspensions are alternated with water. It should be kept in mind that 2,4-D is fairly water soluble and even water can flush all the 2,4-D. However, due to the hydrophobicity of HCB and its low water solubility, the alternating procedure did not help in recovering HCB. One of the main reasons for alternating natural surfactant solutions with water is to investigate the possibility of overcoming the pressure build-up across the soil columns. As mentioned in the earlier section, when natural surfactant solutions of 2.5% concentration are used there is very high pressure build-up across the soil columns and the column run had to be terminated. Tamayo (1991) reported that the pressure build-up across the column will reduce if the flushing media is switched to water. However, the results of this study are contrary to those of Tamayo (1991). Figure 5.29 shows the variation of pressure build-up across the soil columns for all the alternate runs and the corresponding continuous mode runs. The pressure behavior did not change significantly when the flushing media was alternated between natural surfactant solutions or CGA suspensions and water. The columns where 2.5% natural surfactant was used had to be terminated at about 11 pore volumes due to the high pressure build-up. However, at the lower concentration (1%) both the CGA suspensions and natural surfactant solutions have similar pressure trends and the alternating with water did not change the pressure build-up considerably. Figure 5.29: Effect of Alternating Flushing Media with Water on the Pressure Build Pressure the on Water with Media Flushing Alternating of Effect 5.29: Figure Pressure Build-up (Psig) 20 30 40 50 10 70 60 *«»<>>»X X0 X 5 0 5 20 15 10 5 0 NS - Natural Surfactant Natural - NS up Across the Soil Columns Soil the Across up ^ -• l / cl * .* .0* ' ' *“ v*-1— .0* .* * 9 c - . — • * / * —• p.' p.' —'AX— / / P —-X-—X X / / ~ * ~ / / No. of Pore Volumes Pore of No. ,p ' ' ,p Ai—■—r J • ' / / f ' O — x— Water x— — i / *-• f : tf f P k * X -- ♦ .o.. 1%CGA ...... o — ■— 1% NS NS 1%"Water ■— — —X * Flushing Media Flushing +- J* p 0 X X- X X ,.o-.o--o * 2.5 2.5 * '■ 2.5 1 % NS 1% % CGA'Water % % S Water NS NS 138 139 5.7 Bioenhancement in the Presence of Natural Surfactant Solutions Surfactants are known to increase the aqueous concentrations of poorly soluble organic compounds and interfacial areas between immiscible fluids beyond their critical micelle concentration (Ellis et al. 1985, Rosen, 1989, Kile and Chiou 1989, Edwards et al. 1992, Palmer and Fish 1992, Abdul et al. and others). The increased solubility and decreased interfacial tension can improve the accessibility of these substances to microorganisms. However, the literature suggests that the presence of surfactants may enhance, inhibit or may not have any effect on the biodegradation of organic compounds (Laha and Luthy 1991, 1992, Van Hoof and Rogers 1992, Aronstein and Alexander 1992, 1993, Bury and Miller 1993, Tiehm 1994, Rouse et al. 1995 and others). Rouse et al. (1994) made an excellent review on the factors involved and the current state of knowledge concerning the influence of commercial surfactants and bio-surfactants on the biodegradation of hydrocarbons. Another important factor in the bioenhancement studies is the compatibility of surfactants with the cell membrane of the bacteria (Swisher, 1987). Surfactants may form complexes with membrane proteins and exoenzyme and could potentially inhibit the microbial system. In this study, natural surfactant solutions are employed in the form of conventional solutions and CGA suspensions for the remediation of contaminated soils. The advantage of using a natural product in the remediation is that it does not add a toxic load to the subsurface and it may even enhance the degradation of other toxic contaminants present in the subsurface. It is very important to investigate the fate of natural surfactant that is used in the clean-up of subsurface soils. The effect of the 140 presence of a natural surfactant on the growth of soil microorganisms was investigated under both aerobic and anaerobic conditions. Natural surfactant concentration of 1% was used for all the experiments, as this dose seem to be the optimum concentration for flushing of contaminants from soil. The effect of natural surfactant concentration was studied with 0.1% and 2.0% solutions. Natural surfactant solutions were amended with nutrients and/or hexachlorobenzene to determine their effect on the microbial growth. Total organic carbon (TOC) and absorbance at 540 nmona spectrophotometer were monitored. 5.7.1 Aerobic Bioenhancement Studies The microbial populations isolated from surface soils around the university campus were acclimated in natural surfactant solutions for about a week and used as a seed inoculum for the bioenhancement experiments. The results of the duplicate bioenhancement studies are presented in Figure 5.30(a). The plot shows the growth of microorganisms as measured by the absorbance at 540 nm of the solutions versus the time of incubation in hours in a completely mixed batch system. Absorbance at 540 nm was used to monitor the bio-growth, as this was shown to correlate very well with the microbial populations (Koch 1981). The aerobic growth was monitored over a period of 360 hours. The microorganisms were in the lag phase for about 5 hours followed by a log growth or exponential growth phase for about 100 hours. The growth was very rapid during this phase. As can be seen from the figure the growth 141 3 2.5 2 1.5 Flask 1 Flask2 1 • Control 0.5 90 Flask 1 80 ~ Control 70 60 50 40 50 100 150 200 250 300 350 Time (Hours) e 5.: Aerobic Bioenhancement in Natural Surfactant Solutions (a) Growth of Microorganisms and (b) Reduction in TOC pattern followed a typical growth curve for pure cultures (Benefield and Randall, 1986). There was a significant increase in growth again after about 200 hours after which the stationary phase had begun. This could be due to the degradation of compound or compounds that were resistant during the earlier log growth. Due to the lack of precise quantification methods for natural surfactant and its constituent components, it was not possible to identify and quantify the individual compounds. However, TOC was used as a quantification tool for monitoring the changes in the total organic carbon content. Figure 5.30(b) shows the TOC at different stages of the growth. X axis shows the time and the Y axis indicates the TOC in grams per liter. As can be seen from the bars, the TOC was reduced by 27% during the first 48 hours and the degradation was slow beyond that point and the TOC after 360 hours was about 40%. The microorganisms were able to degrade natural surfactant to about 60% of the initial concentration by the end of 45 days. The experiment was repeated with a different seed inoculum and the growth curves followed identical trends. Swisher (1987) and Rouse et al. (1994) reported that surfactant interactions with the bacterial cell wall can be crucial in the biodegradation of surfactants and hydrocarbons in the presence of surfactants. However, the interactions of natural surfactant with the bacterial cells does not seem to inhibit the growth of microorganisms. 5.7.1.1 Effect of Hexachlorobenzene Hexachlorobenzene, a chlorinated hydrocarbon and a priority pollutant was used in this study to investigate the performance of natural surfactant solutions in soil 143 flushing of contaminated soils and hence was selected to study the effect of the hydrocarbons solubilized in natural surfactant solutions on the growth of microorganisms. HCB was solubilized in natural surfactant solutions and the microorganisms were added and monitored for bio-growth. Figure 5.31(a) shows the effect of HCB on the aerobic bioenhancement of soil microorganisms. The HCB concentration used was about 1 mg/1. As can be observed from the figure, there is no significant difference in growth curves for cultures with and without HCB. Tukey method for analysis of variance (Tukey 1953) with 95% confidence level was performed to test for the significance. There was no significant difference between the TOC of the two cultures. This clearly indicates that the presence of HCB at concentrations up to 1 mg/1 does not significantly effect the growth of the microorganisms in natural surfactant solutions. The concentration of HCB was monitored at regular intervals to determine whether there is any degradation or transformation due to the biological growth. Controls were also amended with HCB to account for the volatilization losses. There was no remarkable change in HCB concentration of the cultures after 360 hours when compared to the control flasks. It is known that HCB can not be degraded aerobically as it is in a fully oxidized form. However, it can be reduced to lower chlorobenzenes under anaerobic conditions (Fathepure et al. 1988). These lower chlorobenzenes can be oxidized under aerobic conditions (Fathepure and Vogel 1991). 144 2.5 cB V)s ■e NS 1 NS 2 ~ NS+HCB 1 0.5 NS+HCB2 3.5 cB NS 2 NS+BSM 1 0.5 NS+BSM 2 NS - Natural Surfactant 0 100 150 200 250 300 350 Time (Hours) Figure 5.31: Aerobic Bioenhancement Studies (a) Effect of HCB (b) Effect of Nutrients on the Microbial Growth 145 A test was performed to study the sorption of HCB onto the biomass over a period of 210 hours. HCB was added to the cultures without any additional carbon and the concentration of HCB and bio-growth was monitored over the test period. During this period there was no significant growth of microorganisms. HCB concentration in the aqueous phase did not change significantly indicating that HCB is not undergoing significant adsorption with biomass. As mentioned earlier, HCB is fully oxidized compound and it can not support aerobic biological growth. Rouse et al. (1994) in their work on sulfated and sulfonated surfactants reported that surfactant interactions such as sorption with biomass seem to be the main factor influencing the biodegradation of hydrocarbon. However, in our study, natural surfactant is degraded considerably and thus indicates that the bacterial cell and natural surfactant interactions are not inhibiting the growth of the cultures. 5.7.1.2 Effect of Nutrients Basal salt media (BSM) was added to the culture flasks to determine whether the system is nutrient limited. Figure 5.31(b) shows the difference between the cultures that were amended with nutrients and those without any nutrients. It is evident from the figure that the presence of nutrients doubled the growth in the first hundred hours. There was a considerable variation between the duplicate cultures amended with nutrients. The log phase was much steeper than the culture without nutrients. The cultures that were amended with the nutrients reached stationary phase by about 20 hours as against 100 hours by nutrient limited cultures. The nutrient limited cultures 146 however, had a significant increase in growth after around 200 hours and the differences in bio-growth between the two cultures narrowed down by the end of the run. This indicates that the system growth is limited by the nutrients. The continued growth in case of cultures without nutrients indicate that the nutrients are utilized as they are released from the endogenous phase. The endogenous phase around 150 hours was followed by a significant exponential growth. The nutrients released during the endogenous phase were utilized in the subsequent log phase. Addition of HCB to cultures that were amended with nutrients did not effect the growth pattern as was noted for cultures without nutrients (figure not shown). The growth curves were very similar to those for cultures without HCB. Figure 5.32 shows the variation of TOC for samples with natural surfactant, natural surfactant and HCB, natural surfactant and nutrients, natural surfactant, HCB and nutrients. It is very clear that the nutrient media is helping the microorganisms to degrade natural surfactant to significantly lower amounts than the cultures without nutrients. After about 15 days, the TOC was reduced by 40% in flasks without nutrients and by 55% in flasks where nutrients were added. The TOC values are very similar for systems amended with HCB. The TOC was reduced by about 42% and 61% for cultures without and with nutrients respectively. The degradation in nutrient amended systems continued after 15 days and reached about 72% by the end of 45 days as against 60% by nutrient limited cultures. This section clearly shows that the cultures are limited by nutrients and the presence of HCB does not have any effect on the bio-growth and the rate of growth. However, the nutrient limited cultures seem to Figure 5.32: TOC After 350 Hours for Different Treatments Under Aerobic Aerobic Under Treatments Different for Hours 350 After TOC 5.32: Figure % TOC 30 40 45 35 50 55 60 60 65 NaturalSurfactant Conditions 12 II i III lit III ■ aua uf + HCB Natural + Surf. 12 Treatments Nat. Surf. + BSM + Nat. Surf. ______12 ju ______ III, 12 I I . . - 147 148 catch up with the nutrient amended systems eventually. This probably is due to the utilization of the nutrients released by the bacteria in the endogenous phase. 5.7.1.3 Effect of Natural Surfactant Concentration Three concentrations of natural surfactant 0.1, 1 and 2% were used to study the effect of concentration on the growth of microorganisms. The later two concentrations are also amended with nutrients. Figure 5.33 shows the growth curves for the three concentrations. It is evident from the plots that the higher the surfactant concentration, the higher is the growth of the microorganisms. For cultures with very low concentration of natural surfactant (0.1 %) the growth as indicated by the absorbance increased only marginally (from 0.05 to 0.4) by the end of about 50 hours and remained stationary beyond that. Cultures with 2% natural surfactant has significantly higher growth than those with 1 % natural surfactant either with or without the nutrient media. The higher growth is probably due to the increased amounts of carbon available for synthesis and respiration. The cultures with 2% solutions with nutrient media showed a significantly higher growth than the corresponding cultures without nutrients. This again indicates that the cultures are nutrient limited. After about 15 days, however, the difference between the growth of cultures with and without nutrient media narrowed down. The TOC was reduced by about 50% for 0.1 %, by 40% for 1 % and by 44% for 2% natural surfactant cultures. For cultures with nutrients the TOC was reduced by 57% for 1% and by 56% for 2% natural surfactant solutions. The amount of carbon utilized during the growth period is significantly different. For higher Figure 5.33: Effect of Natural Surfactant Concentration on the Growth of of Growth the on Concentration Surfactant Natural of Effect 5.33: Figure Absorbance at 540 nm 0 2 3 4 5 7 6 1 8 Microorganisms 50 % S BSM NS+ 2% 1%NS 100 NS % NS 2% Time (Hours) Time 150 & 200 NS - Natural Surfactant Natural - NS 250 3000 * ------350 149 150 concentrations of natural surfactant higher amounts of carbon were utilized for the growth. This is indicated by the increased absorbance with increase in natural surfactant concentration in the concentration ranges tested. However, from the soil flushing experiments it is learned that natural surfactant concentrations higher than 2.5% are not suitable and concentrations lower than 0.1% are not effective due to the low solubility of HCB. The present study covers the range of natural surfactant concentrations that are generally useful in flushing experiments. 5.7.2 Anaerobic Bioenhancement Studies The subsurface soils are typically cutoff from the atmosphere and hence are under anoxic conditions. As mentioned in the earlier section, oxygen can be provided to the subsurface by injecting CGA suspensions. However, it is important to appraise the bioenhancement of soil microorganisms under anoxic conditions. Hexachlorobenzene is known to undergo reductive dechlorination under anaerobic conditions and in the presence of an additional carbon source (Fathepure et al. 1988, Pardue 1992). In this section the results of bioenhancement studies under anaerobic conditions in the presence of natural surfactant solutions are evaluated and the effect of the addition of nutrients and HCB are appraised. Figure 5.34(a) shows the growth curves for soil microorganisms under anaerobic conditions with natural surfactant as the only carbon source. The inoculum acclimatized with natural surfactant for about 2 weeks was used as seed for the studies. The concentration used was 1%, as this dose seems to be the optimum for flushing 151 1 0.9 (a) ’.N 0.8 f s A Vo. ------B e 0.7 / • -0 r-'S TCo 0.6 •V"' u 0.5 T Control oc C3 i i _ — JGiZ 0.4 i NS j inQ t •O i < 0.3 I NS 2 t 0.2 . P1 ( 0.1 re ------L. ... --1------1_ ------1------1------1------ 0 \ 0.9 0.8 eB 0.7 •o o U-lTf 0.6 ca o 0.5 co ea NS 1 •fi 0.4 coo X) NS 2 < 0.3 • NS+HCB1 0.2 NS+HCB 2 0.1 NS - Natural Surfactant 0 0 10 15 20 25 30 35 Time (Days) Figure 5.34: Anaerobic Bioenhancement Studies: (a) Growth in Natural Surfactant Solutions (b) Effect of HCB 152 contaminated soils. The cultures had a lag phase of about one day. The growth beyond the lag phase was rapid and reached a stationary phase by about a week. The anaerobic microorganisms showed an increase in growth again by the end of second week, which continued for about a week and by the fourth week the microorganisms are back in the stationary phase. Similar pattern was observed for aerobic cultures growing on natural surfactant. The exponential growth following the first stationary phase is believed to be due to the degradation of constituent components of natural surfactant that were resistant for degradation during the initial log phase. However, due to the lack of analytical techniques to identify the individual compounds, it was not practical to verify this hypothesis. However, TOC was used to monitor the total organic carbon disappearance. The TOC was reduced by about 40% by the end of 25 days for both the flasks. The TOC dropped by 32% by the end of 5 days and beyond 5 days the degradation of natural surfactant was slow. Anaerobic degradation does not yield as much energy as the aerobic process and is generally a very slow process. Under aerobic conditions oxidation of 1 mole of glucose will yield 686 Kcal as opposed to less than 100 Kcal by anaerobic microorganisms (Patrick, 1993). The cell yield for anaerobic process (0.06 mg per mg of organic matter) is much less than that of aerobic process (0.5 mg/mg)(Benefield and Randall 1986). In our study the maximum bio growth as recorded by the absorbance for anaerobic cultures was about one third of that for aerobic cultures. Similar to aerobic cultures, for the anaerobic microorganisms also the interactions of natural surfactant with cell walls are not interfering with the ability of bacteria to degrade the surfactants. Surfactant-cell interactions are reported to be 153 critical for some surfactants in determining the biodegradation capabilities of the microorganisms (Swisher 1987 and Rouse et al. 1994). It is also clear that natural surfactant solutions can support aerobic as well as anaerobic soil microorganisms and can readily serve as a carbon and energy source. 5.7.2.1 Effect of Hexachlorobenzene Natural surfactant solutions were amended with 1 mg/1 HCB and the cultures were monitored for biological growth and HCB. It should be kept in mind that HCB can be reduced under anoxic conditions to lower chlorobenzenes by suitable microorganisms. It is reported that HCB serves as an electron acceptor rather than a carbon source when there is availability of other carbon sources (Fathepure et al. 1988, Sims et al. 1990). This experiment was undertaken to determine whether the isolated soil microorganisms can use HCB as an electron acceptor while using natural surfactant as the carbon source. Figure 5.34(b) shows the comparison of the growth curves for cultures in the presence and absence of HCB under anaerobic conditions. As can be seen from the figure there is no significant difference at 95% confidence level between the cultures with HCB and without HCB except between 15 and 25 days. Natural surfactant solutions without HCB showed a significantly higher growth than the cultures with HCB between 15 and 25 days. The reason for this sharp increase in the microbial population is believed to be due to the availability of constituent components of natural surfactant that were resistant before. However, the growth curves beyond 25 days followed the same trend. The lack of precise analytical technique for monitoring the 154 individual components of natural surfactant limit the scope of this work. There were no significant differences in TOC values after 25 days between the cultures, indicating that HCB does not have either toxic effects or enhancement on the microbial populations at concentrations around 1 mg/1. Concentration of HCB was also monitored to determine the fate of HCB in the cultures, such as reduction or transformation. During the period of 35 days there seem to be no degradation or transformation of HCB into lower chlorobenzenes. When HCB was added as the only carbon source to anaerobic microorganisms, no growth was reported over a period of 45 days. Pardue (1992) suggested that HCB can be utilized as an electron acceptor by anaerobic microorganisms using soil organic matter as the sole carbon source. It is hypothesized that natural surfactant being readily available for anaerobic populations, should lower the redox potential low enough to be able to use HCB as an electron acceptor. Anaerobic cultures use HCB as an electron acceptor when the redox potential is in the range of -150 to -200 millivolts (Pardue 1992). Fathepure and Vogel (1991), Sims et al. (1990) and Fathepure et al. (1988) studied reductive dechlorination of HCB under anoxic conditions with fresh anaerobic sludge as the carbon source and reported similar observations. However, the inoculum used as a seed for this work was not acclimatized with HCB for longer times as reported in the literature (Fathepure et al. 1988, Sims et al. 1990, Pardue, 1992). It is however, believed that Ritha can be used as a carbon source while using HCB as an electron acceptor if the acclimated populations are employed in the work. The redox potentials were not monitored in this work as the experiments were conducted in sealed serum bottles. 155 5.7.2.2 Effect of Nutrients Nutrients in the form of basal salt media (BSM) and heterotrophic media are added to cultures to determine whether the systems are nutrient limited. BSM is generally used in aerobic studies and the heterotrophic media is a very general media which can support all the anaerobic heterotrophic microorganisms (Daniels et a l 1986, and Boopathy et al. 1993). Figures 5.35 (a and b) show the effect of different nutrient media on the cultures that were growing on natural surfactant and natural surfactant and HCB, respectively. As mentioned in the earlier section, the presence of HCB did not significantly effect the growth of anaerobic soil microorganisms, both the figures follow the same trend. The systems amended with heterotrophic media showed a faster and higher growth than the corresponding cultures without the nutrients in the first 10 days. However, there was no significant difference in the growth of microbial populations by the end of the experiment at 35 days. The cultures with BSM on the other hand had a significantly faster and higher growth than the systems with natural surfactant and/or without heterotrophic media. The growth in BSM is twice as high as the other two cultures. The cultures without the nutrients did not show any increase in growth beyond the stationary phase as was observed for aerobic microorganisms. The growth reached stationary phase by the end of the first week and remained in that phase for the remainder of the experiment, however, aerobic cultures had a second exponential phase after a prolonged stationary phase. 156 1.8 1.6 1.4 e s N ! e 1.2 o M- --*«•■• N2 NHm 1 o c !;A yo<5-'- S'*' * ‘ ^ v « ;P - - 0 - - NHm 2 ■s r~~? 8 , i.1 - • 5 : ; : — NB 1 < S 2i *— NB 2 1.2 ■ N K 1 E s NH 2 1 m9 N H H m l CO O 0.8 ccs NHB 1 «o - O ' 1 0.6 NHB 2 JDCO 0.4 0.2 N- Nat. Surf., H- HCB, B- BSM, Hm- Hetrotroph Media 0 30 40 50 Time (Days) Figure 5.35: Effect of Nutrients on the Anaerobic Growth of Microorganisms (a) Cultures without HCB and (b) Cultures with HCB The TOC at the end of 25 days for all the cultures is shown in Figure 5.36. As can be seen from the figure, the decrease in TOC for all the cultures is about the same. It should be noted that the initial concentration of natural surfactant used in cultures with heterotrophic media is higher than the other cultures. The percent of TOC reduced is in the range of 35 to 45% for all the cultures. The cultures with BSM degraded TOC by about 45 % and the cultures with heterotrophic media by about 40% as against 38% by cultures without any nutrients. These values are lower than those obtained for aerobic cultures. In aerobic studies the addition of nutrients improved the degradation by more than 10% as indicated by the percent TOC disappeared. 5.7.2.3 Effect of Natural Surfactant Concentration Natural surfactant concentrations 0.1, 1 and 2% were used to study the effect of concentration on the bioenhancement. The same concentrations were also amended with BSM. Figure 5.37 shows the growth curves for the three concentrations both without and with the nutrient media, BSM. This growth curve once again clearly shows that the nutrient media is significantly increasing the growth of microorganisms. The 1% cultures with nutrients has higher growth than the 2% cultures without nutrients. The cultures with 2% natural surfactant seem to have a lower growth than those with 1 % natural surfactant, however by about 25 days the cultures growing on 2% cultures were able to catchup with the 1% cultures. This probably is due to the limitation of nutrients and also possible toxic affects of higher natural surfactant Figure 5.36: TOC After 25 Days for Different Anaerobic Treatments Anaerobic Different for Days 25 After TOC 5.36: Figure % TOC 20 30 40 60 70 50 80 10 0 * CN *** 2 2 ' ■■ *' N -Nat. Surf., B -BSM, H - HCB, Hm -Heterotroph Media -Heterotroph Hm HCB, - H -BSM, B Surf., -Nat. N CN Treatment N(N CN CN E IN E 158 Figure 5.37: Effect of Natural Surfactant Concentration on the Anaerobic Growth of of Growth Anaerobic the on Concentration Surfactant Natural of Effect 5.37: Figure Absorbance at 540 nm 2.5 3.5 1.5 0 2 1 3 0 Microorganisms N -Natural Surfactant, B -BSM B Surfactant, -Natural N 10 m —-fj — 20 ie (Days) Time — o 04 50 40 30 - '- O - - “ 0.1% N 0.1% “ % NB 2% 2% N 2% 0.1 0.1 %NB 1 %NB 159 160 concentrations. But the cultures of same natural surfactant concentration when amended with BSM showed no inhibition and had a significantly higher growth. The cultures with 2% natural surfactant utilized about 67% carbon which is significantly higher than that observed (38%) for 1% and 0.1% cultures. The cultures amended with nutrients utilized 69%, 44% and 40% carbon for 2%, 1% and 0.1% natural surfactant respectively. The higher surfactant concentrations can provide higher amounts of degradable carbon for the microbial populations and thus the increased growth. It is evident again that the presence of nutrients did not significantly effect the amount of TOC degraded, even though the growth showed significant increase. This is in contrast to the observations made with aerobic studies where, the increase in the TOC utilized was about 10%. CHAPTER 6 CONCLUSIONS There has been a keen interest in the recent past on biosurfactants and natural surfactants which are produced by microorganisms and plants respectively. Fruits from the plants belonging to the genus Sapindaceae, commonly known as Soap Nut or Ritha in the Indian sub-continent are known for their detergent properties and has been traditionally used for washing fabric and hair. There were no reported investigations on the use of these natural detergents in the remediation of hazardous wastes. This research is aimed at studying the application of fruit pericarps of Sapindus mukurossi, for the remediation of soils contaminated with hazardous wastes. A comprehensive research was conducted to characterize and study the applications based on the results of preliminary investigations. The specific conclusions from this research are: • The natural surfactant solutions can be prepared using a very simple method, by extracting the dry powder obtained from fruit pericarps with water. • The water extract of Ritha is as effective as methanol, ethanol and methanol:benzene (3:1) extracts for concentrations below 2.5% in solubilizing a chlorinated aromatic hydrocarbon, hexachlorobenzene. • About 70% of the dry Ritha powder is extracted in to water and methanol and about 60% into ethanol and methanol: benzene (3:1) mixture. • Stock solutions are always prepared with 10% strength, which is 10 grams of dry Ritha powder in 100 ml of DI water. 161 The COD and TOC of 10% natural surfactant solutions are 124.3 grams/liter and 41.2 grams/liter and the organic nitrogen and phosphorous content is negligible. The. pH of natural surfactant solutions is about 4.5 for concentrations 1.0 % and above. The empirical molecular formula for natural surfactant solutions (water soluble fraction of the fruit extract) is determined to be (C26H31Oj0)n where n is a constant that needs to be determined. The UV absorption properties of natural surfactant solutions can be utilized to quantify natural surfactant. Very good correlations are found between the absorbance and natural surfactant concentration at several wavelengths. Two wavelengths 252 and 292 are selected however, as the solutions exhibited peaks in their spectrums at these wavelengths. The critical micelle concentration (CMC) is found to be in the neighborhood of 0.1% from the surface tension and viscosity measurements. The usefulness of a particle size analyzer is demonstrated for characterizing the CGA suspensions. The dynamic changes that occur when CGA suspensions are introduced into an aqueous system can be studied only using a particle size analyzer. The mean volume diameter used by earlier researchers can not be used to represent the dynamic changes in size distributions. The size fractions 10, 50 and 90 percentile, are used along with the mean volume diameter for studying the size distribution of CGA suspensions. The CGA suspensions generated from natural surfactant solutions have bubble diameters ranging from 10 to 300 fim . This range is similar to that reported for CGA suspensions generated with commercial surfactants. Increase in natural surfactant concentration decreased all the sizes, even though the decrease is marginal in some cases. The stability of CGA suspensions increased when the surfactant concentration was increased from 0.1 to 1.5%. The CGA suspensions generated with 0.5% natural surfactant seem to be more stable than those generated with other concentrations. The presence of salt did not effect the size distribution or stability significantly. The quality of CGA suspensions increased when natural surfactant concentration was increased from 0.1 to 0.5%, however, the quality decreased when natural surfactant concentrations were increased beyond 0.5%. CGA suspensions generated with natural surfactant concentrations beyond 1.5% have very low quality and are very thick in consistency. CGA suspensions generated with natural surfactant solutions have size distributions very similar to those generated with commercial non-ionic surfactant solutions. CGA suspensions generated with natural surfactant solutions are more stable than those generated with SDS solutions, even though the quality is lower. Solubility of hexachlorobenzene is increasing linearly with surfactant concentration beyond CMC. However, the solubility beyond 10% natural surfactant concentration is not linear and follows a saturation type curve. The inefficient extraction of Ritha into water and saturation of the surfactant micelles with HCB are thought to be responsible. Solubility of HCB in SDS solutions increase linearly with SDS concentration beyond CMC for concentrations up to 100 mM. Natural surfactant solutions are comparable with commercial surfactants in solubilizing HCB. About 10.5 grams of raw Ritha powder and about 5.5 grams of SDS are required in one liter water to solubilize 1 mg HCB. The commercial surfactants required ranged from 0.5 to 81 grams. Batch desorption studies show that the natural surfactant solutions are favorable for desorption like other commercial surfactant solutions such as SDS. Natural surfactant and SDS solutions are able to desorb HCB up to 90% of their solubilization capacity. The desorption isotherms follow both Freundlich and Langmuir isotherms with some limitations. Natural surfactant solutions either in the form of conventional solutions or CGA suspensions recovered HCB from soil columns very effectively when compared with water flood. Natural surfactant solutions are able to recover as much as 90% of their HCB solubilization capacity when the soils have high contamination levels. For low contamination soils, natural surfactant solutions recovered as much as 90% of the total HCB. Natural surfactant solutions seem to be more effective than the CGA suspensions in recovering HCB from soil columns. Increase in natural surfactant concentration increased the recovery of HCB significantly. The increased solubility of HCB enhanced the removal. The pressure build-up across the soil columns remained fairly low when natural surfactant was used at concentrations up to 1%. The pressure build-up increased up to 60 psi and resulted in the termination of the experiment after about 10 pore volumes when natural surfactant at 2.5% concentration was employed. Alternating the flushing media with water did not help to reduce the pressure build-up or to increase the removals per gram of Ritha. Natural surfactant solutions can support biological growth and serve readily as carbon source under oxygen rich conditions and both as carbon and energy source under anoxic conditions. HCB did not affect the bio-growth significantly under both aerobic and anaerobic conditions. Tests show that HCB does not adsorb to biomass. Addition of basal salt media (BSM) increased the bio-growth significantly under both aerobic and anoxic conditions suggesting that the cultures are nutrient limited. However, the percent reduction in TOC remained about the same under anaerobic conditions and was about 10% more for cultures with BSM for aerobic cultures. However, the addition of heterotrophic nutrient media to anaerobic cultures did not enhance the growth of microorganisms. The increase in natural surfactant concentration from 0.1 to 2.0% increased the growth significantly under both aerobic and anaerobic conditions. The percent TOC reduction was significantly higher for higher surfactant concentration under anaerobic cultures and did not change appreciably for aerobic cultures. HCB did not undergo any transformations over the period of the study under both aerobic and anaerobic conditions. CHAPTER 7 RECOMMENDATIONS Application of surfactants for both in-situ and ex-situ remediation of contaminated sites is shown to have a tremendous potential at laboratory level, and real sites. Commercial surfactants however, are manufactured with the aid of energy intensive and cost consuming process, not to mention the possible toxic by-products. This work is focussed on the applicability of plant based surfactants for the remediation of contaminated soils at laboratory level. It is evident that natural surfactant solutions are comparable to commercial surfactants in (a) solubilizing HOCs, (b) desorbing HOCs from soil, and (c) flushing organic contaminants from one dimensional soil columns. Natural surfactants also support microbial growth under both aerobic and anaerobic conditions. The following areas need to be researched further before one could establish the usefulness of natural surfactants and implement at the actual sites. • In the present study fruit pericarps of Sapindus mukurossi are employed. However, there are several other plants belonging to the genus Sapindaceae which produce fruits with similar characteristics. The fruit pericarps of these plants should also be studied for the possible application to remediation. • Characterization of natural surfactant solutions should be done with sophisticated methods. The present study estimated the empirical formula based on TOC, COD and organic nitrogen. However, these solutions should be analyzed with mass spectrometry and NMR to identify the constituent compounds and 167 determine the exact molecular formula. An analytical technique should be developed to quantify natural surfactant solutions. Feasibility studies should be conducted for the disposal of un-extracted residue as a fertilizer for agricultural lands. The interaction of natural surfactant solutions with soil need to be investigated to understand (a) adsorption characteristics, (b) to determine the effective surfactant available for solubilization and (c) pore clogging phenomena during soil flushing. The possible application of CGA suspensions generated with natural surfactant for recovering contaminants from aqueous streams should be studied. The biotic degradation or reduction of HOCs in the presence of natural surfactant solutions should be studied under both aerobic and anaerobic conditions Pilot scale testing is necessary before adopting the technology for a contaminated site. A mathematical model should be developed incorporating the fate of natural surfactant and HOCs in the subsurface. The model should include biological degradation, physical and chemical interactions between contaminants and natural surfactant. REFERENCES Abdul, S. A., Gibson, T. L., and Rai D. N., 1990. Selection of Surfactants for the Removal of Petroleum Products from Shallow Sandy Aquifers, Ground Water, 28(6): 920-926. Abdul S. A., Gibson T.L., Ang C. C., Smith J. C., and Sobczynki R. E., 1992. In- Situ Surfactant Washing of Polychlorinated Biphenyls and Oils from a Contaminated Site, Ground Water, Vol 30(2): 219-231. Abriola L. M., Dekker T. J., and Pennell K. D., 1993. Surfactant Enhanced Solubilization of Residual Dodecane in Soil Columns: 2. Mathematical Modeling, Environmental Science and Technology, Vol 27(12):2341-2351. Ang, C. C., and Abdul, S. A., 1991. Aqueous Surfactant Washing of Residual Oil Contamination from Sandy Soil, Ground Water Monitoring Review, Spring 1991: 121-127. Aronstein B. N., Calvillo Y. M., and Alexander M., 1991. Effect of Surfactants at Low Concentrations on the Desorption and Biodegradation of Sorbed Aromatic Compounds in Soil, Environmental Science and Technology, Vol 25 (10): 1728- 1731. Aronstein B. N ., and Alexander M ., 1992. Surfactants at low Concentrations Stimulate Biodegradation of Sorbed Hydrocarbons in Samples of Aquifer Sands and Soil Slurries, Environmental Toxicology and Chemistry, Vol 11(9): 1227-1233. Aronstein B. N., and Alexander M., 1993. Effect of a Non-Ionic Surfactant Added to the Soil Surface on the Biodegradation of Aromatic Hydrocarbons Within the Soil, Applied Microbiology and Biotechnology, Vol 39(3): 386-390. Auten W. L., and Sebba F., 1984. The Use of Colloidal Gas Aphrons (CGAs) for Removal of Slimes from Water by Floe Flotation, in Solid and Liquid Separations, Gregory J. (ed), Ellis Horwood, U.K. Beck J., and Hansen K. E., 1974. Degradation of Quintozene and Pentachloroaniline in Soil, Pesticide Science, 5(l):41-48. Benefield L. D., and Randall C. W., 1986. Biological Process Design for Wastewater Treatment, Prentice-Hall Inc., Englewood Cliffs, NJ. 169 170 Boopathy R., Kulpa C. F., and Wilson M., 1993. Metabolism of 2,4,6- Trinitrotoluene (TNT) by Desulfovibrio sp. (B strain), Applied Microbiology and Biotechnology, Vol 39: 270-275. Bor, N. 1., 1953. Manual o f Indian Forest Botany, Oxford University, Calcutta, India. Borra C., Andreolini F., and Novotny M., 1989. Use of Capillary Supercritical Fluid Chromatography and Microcolumn Liquid Chromatography for the Determination of Nonvolatile Organics in Aqueous Environmental Samples, Analytical Chemistry Vol 61(11): 1208-1210. Breuil C., and Kushner D. J., 1980. Effects of Lipids, Fatty Acids and Other Detergents on Bacterial Utilization of Hexadecane, Canadian Journal of Microbiology , Vol 26: 223-231. Bunch R. L., Barth E. F., and Ettinger M. B., 1961. Organic Materials in Secondary Effluents, J. Water Pollution Control Federation, Vol 33(2): 122-126. Burchfield S. B., Wilson D. J., and Clarke A. N., 1994. Soil Clean-up by Surfactant Washing: V. Supplementary Laboratory Testing, Separation Science and Technology, Vol 29(1): 47-70. Burris D. R., and Antworth C. P., 1992. In-Situ Modification of an Aquifer Material by a Cationic Surfactant to Enhance Retardation of Organic Contaminants, J. Contaminant Hydrology, Vol 10: 325-337. Bury S. J., and Miller C. A., 1993. Effect of Micellar Solubilization on Biodegradation Rates of Hydrocarbons, Environmental Science and Technology, Vol 27(1): 104-110. Caballero M., Cela R., and Perez-Bustamente J. A., 1989. Studies on the Use of Colloidal Gas Aphrons in Coflotation and Solvent Sublation Processes. A Comparison with the Conventional Technique. Separation Science and Technology, Vol 24(9 & 10): 629-640. Chaphalkar P. G., Valsaraj K. T., and Roy D., 1993. A Study of the Size Distribution and Stability of Colloidal Gas Aphrons Using Particle Size Analyzer, Separation Science and Technology, Vol 28(6): 1287-1302. Chaphalkar P. G., Valsaraj K. T., and Roy D., 1994, Flotation Using Microgas Dispersions for the Removal of Pentachlorophenol from Aqueous Solutions, Separation Science and Technology, Vol 29(7): 907-921. 171 Chaphalkar P. G., 1994. Characterization and Application o f Colloidal Gas Aphrons for Groundwater Remediation, Ph.D. Dissertation, Louisiana State University, Baton Rouge, LA. Chiang T,, Liao W., and Williams L. R., 1987. Use of Solid Phase Florisil Cartridges to Separate Fat from Semi volatile Organic Compounds in Adipose Tissue, J. Association Off, Analytical Chemists, Vol 70(1): 100-102. Chiou C. T., Malcolm R. L., Brinton T. I., and Kile D. E., 1986. Solubility Enhancements of Some Organic Pollutants and Pesticides by Dissolved Humic and Fulvic Acids, Environmental Science and Technology, Vol 29(5): 502-508. Christensen D. R., and McCarty P. L., 1975. Multi-Process Biological Treatment Model, J. Water Pollution Control Federation, Vol 47(11): 2652-2664. Christian S. D,, Smith G. A., Tucker E. E., and Scamehom J. F., 1985. Semiequilibrium Dialysis: A New Method for Measuring the Solubilization of Organic Solutes by Aqueous Surfactant Solutions, Langmuir, Vol 1(5): 564- 567. Clarke A. N., Plumb P. D., Subramanyan T. K., and Wilson D. J., 1991. Soil Clean up by Surfactant Washing: I Laboratory Results and Mathematical Modelling, Separation Science and Technology, Vol 26(3): 301-343. Clarke A. N., Oma K. H., Megehee M. M., and Wilson D. J., 1993. Soil Clean-up by Surfactant Washing: II. Design and Evaluation of the Components of the Pilot-Scale Surfactant Recycle System, Separation Science and Technology, Vol 28(13 & 14): 2103-2135. Daniels L ., Belay N ., and Rajagopal BS., 1986. Assimilatory Reduction of Sulfate and Sulfite by Methanogenic Bacteria, Applied Environmental Microbiology, Vol 56: 703-709. Darji M., 1993. Removal of Hazardous Oily Waste from Soil Matrix Under Different Flow Conditions, M.S. Thesis, Louisiana State University, Baton Rouge, LA. Desai J. D., and Desai A. J., 1993. Production of Biosurfactants, In Biosurfactants Kosaric N. (Ed), Marcel and Dekker, Inc., New York, NY. Dobbs R. A., Wise R. H., and Dean R. B., 1972. The Use of Ultraviolet Absorbance for Monitoring the Total Organic Carbon Content of Water and Wastewater, Water Research, Vol 6(10): 1173-1180. 172 Eaton A., 1995. Measuring UV-Absorbing Organics: A Standard Method, J. American Water Works Association, Vol 87(2):86-90. Edwards D. A., Luthy R. G., and Liu Z., 1991. Solubilization of Polycyclic Aromatic Hydrocarbons in Micellar Non-Ionic Surfactant Solutions, Environmental Science and Technology, Vol 25(1): 127-133. Edwards D. A., Laha S., Liu Z., and Luthy R. G., 1992. Solubilization and Biodegradation of Hydrophobic Organic Compounds in Soil-Aqueous Systems with Non-Ionic Surfactants, in Transport and Remediation of Subsurface Contaminants, D. A. Sabatini and R. C. Knox (eds.), ACS Symposium Series 491, American Chemical Society, Washington D. C. Edwards D. A., Adeel Z., and Luthy R. G., 1994. Distribution of Non-Ionic Surfactant and Phenanthrene in a Sediment/Aqueous System, Environmental Science and Technology, Vol 28(8): 1550-1560. Ellis W.D., Payne J. R., and McNaab G. D., 1985. Treatment of Contaminated Soils with Aqueous Surfactants, EPA/600/2-85/129, U. S. Environmental Protection Agency, Cincinnati, OH. Esposito M. P., Locke B. B., Greber J., and Traver R. P., 1988. Superfund Standard Analytical Reference Matrix Preparation and Results o f Physical Soil Washing Experiments, EPA/600/9-88/021, U. S. Environmental Protection Agency, Cincinnati, OH. Faltkao D. M., and Novak J. T., 1992. Effects of Biologically Produced Surfactants on the Mobility and Biodegradation of Petroleum Hydrocarbons, Water Environment Research, Vol 64(2): 163-169. Fanner W. J., Yang M., Letey J., and Spencer W. F., 1980. Land Disposal of Hexachlorobenzene Wastes: Controlling Vapor Movement in Soil, EPA-600/2- 80-119, U. S. Environmental Protection Agency, Cincinnati, OH. Fathepure B. Z., Tiedje J. M., and Boyd S. A., 1988. Reductive Dechlorination of Hexachlorobenzene to Tri- and Dichlorobenzenes in Anaerobic Sewage Sludge, Applied and Environmental Microbiology, Vol 54: 327-330. Fathepure B. Z., and Vogel T. M., 1991. Complete Degradation of Polychlorinated Hydrocarbons by a Two Stage Biofilm Reactor, Applied and Environmental Microbiology, Vol 57: 3418-3422. 173 Foster P., and Morris A. W., 1971. The Use of Ultraviolet Absorption Measurements for the Estimation of Organic Pollution in Inshore Sea Waters, Water Research, Vol 5(1): 19-27. Fountain J.C., Klimek A., Beikirch M. G., and Middleton T.M., 1991. The Use of Surfactants for In-Situ Extraction of Organic Pollutants from a Contaminated Aquifer, J . Hazardous Materials , Vol 28: 295-311. Fransisco V. R., 1992. In-Situ Surfactant Flushing of a Sandy Aquifer Contaminated with Gasoline Hydrocarbons, Ground Water, Vol 30 (5): 788-799. Freundlich H., 1926. Colloid and Capillary Chemistry, Metheun, London. Fugate E., 1984. The Use o f Colloidal Gas Aphrons to Facilitate Decontamination of Subsoil Hydrocarbon Spills, M. S. Thesis, Virginia Polytechnic Institute and State University, Blacksburg, VA. Gannon O. K., Bibring P., Raney K., Ward J. A., Wilson D. J., Underwood J. L., and Debelak K. A., 1989. Soil Clean-up by In-Situ Surfactant Flushing: III. Laboratory Results, Separation Science and Technology, Vol 24(14): 1073- 1094. Gedeon J., 1954. Saponins from Indian Soapnuts, J. Sci. IndustriRes., Vol 13B: 427- 428. Gee G. W., and Bauder J. W., 1986. Particle Size Analysis, in Methods of Soil Analysis Part I: Physical and Mineralogical Methods, Klute (ed), 2nd edition, American Society of Agronomy, Madison, WI. Greenberg A. E., Clesceri L. S., and Eaton A. D., 1992. Standard Methods for the Examination o f Water and Wastewater, 18th edition. APHA, AWWA and WEF, Washington D. C. Guerin W. F., and Jones G. E., 1988. Mineralization of Phenanthrene by a Mycobacterium sp., Applied Environmental Microbiology, Vol 54(4): 937-944. Gupta D. R., and Ahmed B., 1990. Emarginatosides B and C: Two New Saponins from Sapindus Emarginatus Fruits, Indian Journal of Chemistry Vol 29B: 268- 270. Harwell J. H., 1992. Factors Affecting Surfactant Performance in Groundwater Remediation Application, In Transport and Remediation of Subsurface 174 Contaminants, D. A. Sabatini and R. C. Knox (eds,), ACS Symposium Series 491, American Chemical Society, Washington D. C. Hassett J. P., and Anderson M. A., 1982. Effects of Dissolved Organic Matter on Adsorption of Hydrophobic Organic Compounds by River-Borne Particles, Water Research Vol 16(5): 681-686. Hoffman F., 1993. Ground-Water Remediation Using Smart Pump and Treat, Ground Water, Vol 31(1): 98-106. Hoffmann H., and Rehage H., 1987, Rheology of Surfactant Solutions, in Surfactant Solutions: New Methods of Investigation, Zana R. (ed). Marcel Dekker, Inc., New York, NY. Honeycutt S. S., Wallis D. A., and Sebba F., 1983. A Technique for Harvesting Unicellular Algae Using Colloidal Gas Aphrons, Biotechnology and Bioengineering Symposium No. 13: 567-575. Hunt J. R., and Sitar N., 1988. Non-Aqueous Phase Liquid Transport and Clean-up: 1. Analysis of Mechanisms, Water Resources Research, Vol 24(8): 1247-1258. Jackson W. A ., and Roy D ., 1995. Transport of Bacteria Through a Soil Matrix Using Colloidal Gas Aphrons, Applied Environmental Microbiology (In Review). Jafvert C. T., and Heath J. K., 1991. Sediment and Saturated Soil Associated Reactions Involving an Anionic Surfactant (Dodecylsulfate): 1. Precipitation and Micelle Formation, Environmental Science arid Technology, Vol 25(6): 1031-1038. Jafvert C. T., 1991. Sediment and Saturated Soil Associated Reactions Involving an Anionic Surfactant (Dodecylsulfate): 2. Partition of PAH Compounds Among Phases, Environmental Science and Technology, Vol 25(6): 1039-1045. Jafvert C. T., Van Hoof P. L., and Heath J. K., 1994. Solubilization of Non-Polar Compounds by Non-Ionic Surfactant Micelles, Water Research Vol 28(5): 1009-1017. Jirka A. M., and Carter M. J., 1975. Micro Semi-automated Analysis of Surface and Waste Waters for Chemical Oxygen Demand, Analytical Chemistry, Vol 47: 1397. Karrer P., 1950. Organic Chemistry, Elsevier Publishing Company, Inc., New York, NY. 175 Kasai R., Nishi M., Mizzutani K., Miyahara I., Moriya T., Miyahara K., and Tanaka O., 1988. Trifoliosode II, An Acyclic Sesquiterpine Oligoglycoside from Pericarp of Sapindus trifoliatus, Phytochemistry, Vol 27(7): 2209-2211. Keller G., Warrack B., and Bartel H., 1988. Statistics for Management and Economics: A Systematic Approach, Wadsworth Publishing Company, Belmont, CA. Keenaga E. E., and Goring C., 1980. Relationship Between Water Solubility, Soil Sorption, Octanol-Water Partitioning and Concentration of Chemicals in Biota, in Aquatic Toxicology, ASTM STP 707, Eaton J G., Parrish P. R., and Hendricks A. C., (eds)American Society for Testing and Materials, Philadelphia, PA. Kile D. E., and Chiou C. T., 1989. Water Solubility Enhancements of DDT and Trichlorobenzene by Some Surfactants Below and Above the Critical Micelle Concentration. Environmental Science and Technology, Vol 23(7): 832-838. Kim B. J., and Gee C. S., 1993. Hazardous Waste Treatment Technologies, Water Environment Research, Vol 64(4):430. Kimata H., Nakashima T., Kokubun S., Nakayama K., Mitoma Y., Kitahara T., Yata N., and Tanaka O., 1983. Saponins of Pericarps of Sapindus mukurossi Gaertn. and Solubilization of Monodesmosides by Bisdesmosides, Chemistry and Pharmaceutical Bulletin, Vol 31(6): 1998-2005. Knox R. C., Canter L.W., Kincannon D.F., Stover E.L., and Ward C.H., 1984. State-of-the Art of Aquifer Restoration, EPA/600/S2-84/182, U.S. Environmental Protection Agency, Ada, OK. Koch A. L., 1981. Growth Measurement in Manual of Methods for General Bacteriology, Gerhardt P., Murray R., Costilow R. N., Nester E. W., Wood W. A., Krieg N. R., and Phillips C. B. (eds), American Society for Microbiology, Washington D.C. Kommalapati R. R., 1992. In-Situ Flushing of Disturbed Uncontaminated Soil Cores from the PPI Site Using Surfactants and Colloidal Gas Aphron Suspensions, Unpublished Work, Louisiana State University, Baton Rouge, LA. Kommalapati R. R., 1993. In-Situ Flushing of Undisturbed Contaminated Cores from the PPI Site Using Surfactants and Colloidal Gas Aphron Suspensions, Unpublished Work, Louisiana State University, Baton Rouge, LA. 176 Kommalapati R. R., 1994. Soil Flushing of Non Aqueous Phase Liquids Using Conventional Surfactant Solutions and Colloidal Gas Aphron Suspensions, M.S. Thesis, Louisiana State University, Baton Rouge, LA. Kongara V., 1994. Application of Surfactant Solutions and Colloidal Gas Aphrons in Flushing Naphthalene from the Soil, M.S. Thesis, Louisiana State University, Baton Rouge, LA. Laha S., and Luthy R. G., 1991. Inhibition of Phenanthrene Mineralization by Non- Ionic Surfactants in Soil-Water Systems, Environmental Science and Technology, Vol 25(11): 1920-1930. Laha S., and Luthy R. G ., 1992. Effects of Non-Ionic Surfactants on the Solubilization and Mineralization of Phenanthrene in Soil-Water Systems, Biotechnology and Bioengineering, Vol 40(11): 1367-1380. Landrium P. F., Nihart S. R., Eadie B. J., and Graner W. S., 1984. Riverse-Phase Separation Method for Determining Pollutant Binding to Aldrich Humic Acid and Dissolved Organic Carbon of Natural Waters, Environmental Science and Technology, Vol 18(3): 187-192. Langmuir I. J., 1918. The Adsorption of Gases on Plane Surfaces of Glass, Mica and Platinum, J. American Chemical Society, Vol 40: 1361. Liu M., 1993. Mobilization of a Hydrophobic Organic Compound Using Surfactant for Soil Washing, Ph. D. Dissertation, Louisiana State University, Baton Rouge, LA. Liu Z., Laha S., and Luthy R. G., 1991. Surfactant Solubilization of Polycyclic Aromatic Hydrocarbons in Soil Water Suspensions, Water Science and Technology, Vol 23: 475-485. Liu Z., Edwards D. A., and Luthy R. G., 1992. Sorption of Non-Ionic Surfactants onto Soil, Water Research, Vol 26(10): 1337-1345. Longe T. A ., 1989. Colloidal Gas Aphrons: Generation, Flow Characterization and Application in Soil and Groundwater Decontamination, Ph. D. Dissertation, Virginia Polytechnic Institute and State University, Blacksburg. Macdonald J. A., and Kavanaugh M. C., 1994. Restoring Contaminated Groundwater: An Achievable Goal?, Environmental Science and Technology, Vol 28(8)362- 368. 177 Mackay D. M., and Cherry J. A., 1989. Groundwater Contamination, Pump and Treat, Environmental Science and Technology, Vol 23(6): 630-634. Mandava S., 1994. Application o f a Natural Surfactant from Sapindus Emarginatus to In-Situ Flushing o f Soils Contaminated with Hydrophobic Organic Compounds, M.S. Thesis, Louisiana State University, Baton Rouge, LA. Megehee M. M., Clarke A. N., Oma K. H., and Wilson D. J., 1993. Soil Clean-up by Surfactant Washing: IV. Modification and Testing of Mathematical Models Separation Science and Technology, Vol 28(17 & 18): 2507-22527. Mercer J. W., Skipp D. C., and Giffin D., 1990. Basics of Pump and Treat Ground water Remediation Technology, EPA-600/8-90/003, U. S. Environmental Protection Agency, Office of Research and Development, Ada, OK.. Metcalf R. L., Kapoor I. P., Lu P. Y., Schuth C. K., and Sherman P., 1973. Model Ecosystem Studies of the Environmental Fate of Six Organochlorine Pesticides, Environmental Health Perspect., June 1973: 35-44. Michelsen D. L., Wallis D. A., and Sebba F., 1984. In-Situ Biological Oxidation of Hazardous Organics. Environmental Progress, Vol 3: 103-110. Michelsen D. L., Wallis D. A., and Lavinder S. R., 1985. In-Situ Biodegradation of Dispersed Organics Using a Microdispersion of Air in Water. Proceedings of the 5lh National Conference on Management of Uncontrolled Hazardous Waste Sites, Washington D. C. Michelsen D. L., Smith J. W., and Suggs J. A., 1988. Use of Colloidal Gas Aphrons for In-Situ Biodegradation of Contaminated Groundwater, Proceedings of the 14h Annual Hazardous Waste Symposium, United States Environmental Protection Agency, Cincinnati, OH. Miller M, M., Wasik S. P., Huang G. L., Shiu W. Y., and Mackay D., 1985. Relationships Between Octanol-Water Partition Coefficient and Aqueous Solubility, Environmental Science and Technology, Vol 19(6): 522-529. Montgomery J. H., and Welkin L. M., 1986. Ground water Chemicals Desk Reference, Lewis Publishers, New York, NY. Moore T. R., 1985. The Spectrophotometric Determination of Dissolved Organic Carbon in Peat Waters, Soil Science Society of America Journal, Vol 49:1590- 1592. 178 Morrison R. T., and Boyd R. N., 1986. Organic Chemistry, 2nd edition, Allyn and Bacon, Inc., Boston, MA. Mueller J. G., Chapman P. J., BlattmannB. O., and Pritchard P. H., 1990. Isolation and Characterization of a Fluoranthene Utilizing Strain of Pseudomonas Paucomobilis , Applied Environmental Microbiology, Vol 56(4): 1079-1086. Nash J. H., 1987. Field Studies of In-Situ Soil Washing, EPA/600/2-87/110, United States Environmental Protection Agency, Cincinnati, OH. Nakayama K., Fujino H., Kasai R., Tanaka O., and Zhou J., 1986. Saponins of Pericarps of Chinese Sapindus delavayi (Pyi-shiau-tzu), a Source of Natural Surfactant, Chemistry and Pharmaceutical Bulletin , Vol 34(5): 2209-2213. Oberbremer A., Muller-Hurtig R., and Wagner F., 1990. Effect of Addition of Microbial Surfactants on Hydrocarbon Degradation in a Soil Population in a Stirred Reactor, Applied Microbiology, Biotechnology, Vol 32: 485-489. Olsen R. L., and Kavanaugh M. C., 1993. Can Groundwater Restoration be Achieved?, Water Environment Technology, Vol 5(3): 42-47. Oma, K. H., Clarke, A. N., Mutch, R. D., and Wilson, D. J., 1991. Surfactant Flushing/Washing: Design and Preliminary Testing of an Innovative Process for Recovery and Recycle of Surfactant, Presented at the I&EC Special Symposium, American Chemical Society, Atlanta, October 1991. Oma K. H., Clarke A. N., Megehee M. M., and Wilson D. J., 1993. Soil Clean-up by Surfactant Washing: III. Design and Evaluation of the Integrated Pilot-Scale Surfactant Recycle System, Separation Science and Technology, Vol 28(15 & 16): 2319-2349. Oommachan M., 1977. The Flora of Bhopal: Angiosperms, J. K. Jain Brothers, Bhopal, India. Palmer C. D,, and Fish W., 1992. Chemical Enhancements to Pump and Treat Remediation, EPA/540/S-92/001,United States Environmental Protection Agency, Ada, OK. Pardue J. H., 1992. Assimilation of Organic Contaminants in Wetlands: Processes and A Conceptual Model, Ph. D. Dissertation, Louisiana State University, Baton Rouge, LA. 179 Patrick W., 1993, Biogeochemistry of Wetland Soils and Sediments, Lecture Notes, Louisiana State University, Baton Rouge, LA. Pennell K. D., Abriola L. M., and Weber W. J., 1993. Surfactant Enhanced Solubilization of Residual Dodecane in Soil Columns: 1. Experimental Investigation, Environmental Science and Technology, Vol 27(12):2332-2340. Peters R. W., Montemagno C. D., Shem L., and Lewis B., 1992. Surfactant Screening of Diesel-Contaminated Soil, Hazardous Waste and Hazardous Materials, Vol 9(2): 113-136. Preston W. C., 1948. J. Phys. Colloid Chem. Vol 52: 84. Quinlivan S., Ghassemi M., and Santy M., 1976. Survey of Methods Used to Control Wastes Containing Hexachlorobenzene, U. S. Environmental Protection Agency, Office of Solid Waste Management Programs, Washington D. C., EPA-530/SW-120C. Ranganna S., Sastry M. V., and Siddappa G. S., 1963. Preparation of Saponins from Soapberry, Indian Journal of Technology Vol 1: 97-98. Rosen M. J., 1978. Surfactants and Interfacial Phenomena, John Wiley & Sons Ltd., New York. Rouse J. D., Sabatini D. A., and Harwell J. H., 1993. Minimizing Surfactant Losses Using Twin-Head Anionic Surfactants in Subsurface Remediation, Environmental Science and Technology, Vol 27(10): 2072-2078. Rouse J. D., Sabatini D. A., Suflita J. M., and Harwell J. H., 1994. Influence of Surfactants on Microbial Degradation of Organic Compounds, Critical Reviews in Environmental Science and Technology, Vol 24(4): 325-370. Rouse J. D., Sabatini D. A., and Harwell J. H., 1995. Influence of Anionic Surfactants on Bioremediation of Hydrocarbons, Manuscript in Preparation Row, L. R., and Rukmini C., 1966a. Chemistry of Saponins: Part 1 - Saponin of Sapindus Mukorossi Gaertn. Indian J. Chemistry, Vol 4: 36-38. Row, L. R., and Rukmini C., 1966b. Chemistry of saponins: Part II - Saponin of Sapindus Emarginatus Vahl. Indian J. Chemistry, Vol 4: 149-150. 180 Roy D., Valsaraj K. T., and Kottai S., 1992a. Separation of Organic Dyes from Wastewater by Using Colloidal Gas Aphrons, Separation Science and Technology, Vol 27(5): 573-588. Roy D., Valsaraj K. T., and Tamayo A., 1992b. Comparison of Soil Washing Using Conventional Surfactant Solutions and Colloidal Gas Aphron Suspensions, Separation Science and Technology, Vol 27(12): 1555-1568. Roy D., Valsaraj K. T., and Constant W. D., 1992c. In-Situ Soil Washing Using Surfactants and Colloidal Gas Aphrons to Enhance Pump and Treat Remediation at the PPI Site. Internal Report Submitted to College of Engineering, Louisiana State University, Baton Rouge, LA. Roy D., Valsaraj K. T., Constant W. D. and Darji M., 1994. Removal of Hazardous Oily Waste from a Soil Matrix Using Surfactants and Colloidal Gas Aphron Suspensions Under Different Flow Conditions, J. Hazardous Materials, Vol 38: 127-144. Roy D., Kommalapati R. R., Valsaraj K. T., and Constant W. D., 1995. Soil Flushing of Residual Transmission Fluid: Application of Colloidal Gas Aphron Suspensions and Conventional Surfactant Solutions, Water Research, Vol 29(2): 589-595. Sabatini D., Rouse J., Lipe M., and Harwell J., 1992. Surfactant Enhanced Subsurface Remediation: Minimization of Surfactant Losses and Recovery of Surfactant, Ground Water, Vol 30(5):795. Sarin J. L., and Beri M. L., 1939. Extraction of Saponin from Soap Nut, Industrial and Engineering Chemistry, Vol 31(6): 712-713. Schmitt E, K., and Caplan J. A., 1987. In-Situ Biological Clean-up of Petroleum Hydrocarbons in Soil and Groundwater, Proceedings of the 5lh Annual Hazardous Materials Management Conference, June 9-11, 1987, Atlantic City, NJ. Sebba F., 1971. Microfoams- An Unexplored Colloid System, J. Colloid and Interface Science, Vol 35(4): 643-646. Sebba F., and Barnett S. M., 1981. Separations Using Colloidal Gas Aphrons, Proceedings of 2 nd World Congress of Chemical Engineering, Montreal, Vol IV: 27-31. 181 Sebba F., 1982, Investigation of the Modes of Contaminant Capture in CGA (.MGD) Foams, Report OWRT/RU-82/10 to U. S. Department of the Interior, Office of Water Research and Technology. Sebba F., 1985a. An Improved Generator for Micron Sized Bubbles, Chemistry and Industry, Feb 1985, 91-92. Sebba F., 1985b. Separations Using Aphrons, Separation and Purification Methods, Vol 14(1): 127-148. Shea, P. T., and Barnett, S. M., 1979. Flotation Separation Using Micro Gas Dispersion, Separation Science and Technology, Vol 14(9):757-762. Shetty, M. S., 1972. The Use of Ritha Powder as an Air- Entraining Agent. Indian Concrete Journal, Vol. 46: 116-122. Sims J. L., Suflita J. M., and Russell H. H., 1990. Reductive Dehalogenation of Organic Contaminants in Soils and Ground Water, Ground Water Issue, EPA/540/4-90/054, United States Environmental Protection Agency, Washington D. C. Singhal N., and Roy D., 1988. Modeling Kinetics of 2,4-D Degradation by Two New Pseudomonas Isolates, The Chemical Engineering Journal, Vol 39: B37-B45. Spence J. T., Cotton J. W., Underwood B. J., and Duncan C. P., 1983. Elementary Statistics, 4th edition, Prentice-Hall, Inc., Englewood Cliffs, NJ. Suggs J. A., 1987. Generation of Microbubble Foam Using a Packed Bed, M. S. Thesis, Virginia Polytechnic Institute and State University, Blacksburg, VA. Swisher R. D., 1987. Surfactant Biodegradation, 2nd edition, Marcel Dekker Inc., New York, NY. Tamayo A., 1991. Application of Surfactant and Colloidal Gas Aphrons in Washing Soils Contaminated with 2,4-D, M.S. Thesis, Louisiana State University, Baton Rouge, LA. Tanford C., 1980. The Hydrophobic Effect: Formation of Micelles and Biological Membranes. Wiley, New York, NY. Thangamani S., and Shreve G. S., 1994. Effect of Anionic Biosurfactant on Hexadecane Partitioning in Multiphase Systems, Environmental Science and Technology, 28(12): 1993-2000. 182 Tiehm A., 1994. Degradation of Polycyclic Aromatic Hydrocarbons in the Presence of Synthetic Surfactants, Applied Environmental Microbiology, Vol 60(1):258- 263. Travis C. C., and Doty C. B., 1990, Can Contaminated Aquifers at Superfund Sites be Remediated?, Environmental Science and Technology, 24(10): 1464-1466. Tukey J. W., 1953. The Problem of Multiple Comparisons, (mimeographed) Princeton, NJ. Underwood J. L., Debelak K. A. Wilson D. J., and Means J. M., 1993a. Soil Clean up by In-Situ Surfactant Flushing: V. Micellar Solubilization of Some Aromatic Contaminants, Separation Science and Technology, Vol 28(8): 1527-1537. Underwood J. L., Debelak K. A., and Wilson D. J., 1993b. Soil Clean-up by In-Situ Surfactant Flushing: VI. Reclamation of Surfactant for Recycle, Separation Science and Technology, Vol 28(9): 1647-1669. Underwood J. L., Debelak K. A., and Wilson D. J., 1995. Soil Clean-up by In-Situ Surfactant Flushing: VII. Determination of Mass Transfer Coefficients for Reclamation of Surfactant for Recycle, Separation Science and Technology, Vol 30(l):73-87. United States Environmental Protection Agency, 1988. Technology Screening Guide for Treatment of CERCLA Soils and Sludges, Office of Solid Waste and Emergency Response, EPA/540/2-88/004, Washington D. C. United States Environmental Protection Agency, 1993. The Superfund Innovative Technology Evaluation Program, Office of Solid Waste and Emergency Response and Office of Research and Development, EPA/540/R-93/525, Washington D. C. United States Environmental Protection Agency, 1991, Guide for Conducting Treatability Studies Under CERCLA: Soil Washing, EPA/540/2-91/20B. United States General Accounting Office, 1986. Hazardous Waste :EPA’s Considerations of Permanent Remedies, GAO/RCED 86-178BR, July, 1986. Uphof, J. C. T, 1968. Dictionary of Economic Plants. Cramer, L. J, New York. NY. Uppal I. S. and Mehta R. L., 1951. Industrial Utilization of Saponin as Textile Auxiliary: Part I - Wetting and Lime soap Dispersing Properties, J. Sci. and Indust. Res. Vol 10B: 190-195. 183 Usher, G., 1974. A Dictionary of Plants Used by Man. Hafner Press, New York, NY. Valsaraj K. T., Gupta A., Thibodeaux L. J., and Harrison D. P., 1988. Partitioning of Low Molecular Weight Aliphatic Hydrocarbons Between Aqueous Phase and Surfactant Micelle Phases, Water Research, Vol 22(9): 1173-1183. Valsaraj K. T., and Thibodeaux L. J., 1989. Relationships BetweenMicelle-Water and Octanol-Water Partition Constants for Hydrophobic Organics of Environmental Interest, Water Research, Vol 23(1): 183-189. Van Hoof P. L., and Rogers J. F., 1992. Influence of Low Levels of Non-Ionic Surfactants on the Aerobic Dechlorination of Hexachlorobenzene, in Biosystems Technology Development Program. Bioremediation of Hazardous Wastes, EPA/600/R-92/126, United States Environmental Protection Agency, Washington D. C. Vigon B. W., and Rubin A. J., 1989. Practical Considerations in Surfactant Aided Mobilization of Contaminants in Aquifers, J. Water Pollution Control Federation, Vol 61(7): 1233-1240. Void R. D., and Void M. J., 1983. Colloid and Interface Chemistry, Addison-Wesley, Reading, MA. Wan J., and Wilson J. L., 1992, Colloid Transport and the Gas-Water Interface in Porous Media, in Transport and Remediation of Subsurface Contaminants, Sabatini D. A., and Knox R. C. (eds.), ACS Symposium Series 491, American Chemical Society,Washington D.C. Wark, K., and Warner, C. F., 1981. Air pollution: Its Origin and Control. Harper & Row, New York, NY. Wayt H. J., and Wilson D. J., 1989. Soil Cleanup by In-Situ Surfactant Flushing: II. Theory of Micellar Solubilization. Separation Science and Technology, Vol 24(12 & 13): 905-937. Wershaw R. L., Burcar P. J., and Goldberg M. C., 1969. Interaction of Pesticides with Natural Organicc Material, Environmental Science and Technology, Vol 3(3): 271-273. West, C. C., and Harwell, J. H., 1992. Surfactants and Subsurface Remediation, Environmental Science and Technology, Vol 26(12): 2324-2330. 184 Wilson D. J., 1989. Soil Clean-up by In-Situ Surfactant Flushing: I. Mathematical Modeling, Separation Science and Technology, Vol 24(11): 863-892 Wilson D. J., and Clarke A. N,, 1991. Soil Clean-up by In-Situ Surfactant Flushing: IV. A Two-Component Mathematical Model. Separation Science and Technology, Vol 26(9): 1177-1194. Wilson S. B., 1986. In-Situ Biosurfactant Production: An Aid to Biodegradation of Organic Groundwater Contaminants, Proceedings o f the NWWA/API Conference on Petroleum Hydrocarbons and Organic chemicals in Groundwater: Prevention, Detection and Restoration, November 13-15, 1985, Houston, TX. Windholz M.,(ed) 1983. The Merck Index: An Encyclopedia of Chemicals, Drugs, and Biologicals, Merck & Co., Inc., Rathway, NJ. APPENDIX A generalized custom half reaction for the oxidation of an unknown organic is (Christensen and McCarty (1975): C,HbOcN„ + (2a-c)HaO =aCOa + dNH% + (4a+b-2c-4d) H* + (4a+b-2c-3d)e' where a,b,c,d are stoichiometric coefficients for carbon, hydrogen, oxygen and nitrogen respectively. Let W = weight of organic, grams, D = chemical oxygen demand of organics, grams N = organic nitrogen content of organics, grams and C = organic carbon content of organics, grams. Then: W = 12a +b + 10c + 14d D = 8 (4a + b - 2c - 3d) N = 14d C = 12a These equations can be solved for a, b, c, and d and the resulting coefficients are: 185 If W, D, N, and C are known, the above equations are solved directly to determine a, b, c, and d, the stoichiometric coefficients for the empirical formula of the organic. VITA The author was born in Chodavaram, a small and beautiful village in the southern state of Andhra Pradesh in India on August 23, 1967. He is the last of the four children to Rangamma and Peraiah Choudary Kommalapati. The author spent all his childhood in the village. The author enjoyed learning farm work and working with the friendly villagers on the farm. After his high school and junior college, he joined V.R. Siddhartha Engineering College (Nagrajuna University, Guntur, India) and graduated with a Bachelor of Technology in Civil Engineering in 1988. He graduated from Regional Engineering College (Kakatiya University, Warangal, India) in 1990 with a Master of Technology in Structural Engineering. The author joined Louisiana State University in the Fall of 1990 in pursuit of Doctor of Philosophy in the Department of Civil and Environmental Engineering with an emphasis in Environmental Engineering. He obtained a Master of Science in Civil Engineering in 1994. He is a member of Honor Society, Phi Kappa Phi and The Scientific Research Society, Sigma Xi,. He is an Engineer in Training and member of National Society of Professional Engineers, ASCE, WEF, AWWA, A&WMA, AWRA and SSSA. His Ph.D. dissertation emphasizes the application of a plant based natural surfactant for the remediation of soils contaminated with hazardous wastes. 187 DOCTORAL EXAMINATION AND DISSERTATION REPORT Candidates Raghava Rao Kommalapati Major Field: Civil Engineering Title of Dissertation: Remediation of Contaminated Soils Using a Plant Based Surfactant Approved: d Chairman Dean of 'the Graduate School EXAMINING COMMITTEE: ^ * e^CU. t Date of Examination: May 15, 1995