Influence of biosurfactant and non-biosurfactant producing bacteria on phenanthrene removal from model soils by Julie Ann Eyre
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in
Chemical Engineering
Montana State University
© Copyright by Julie Ann Eyre (1999)
Abstract:
Polycyclic aromatic hydrocarbons (PAHs, e.g. phenanthrene) may occur in the environment as a result of fossil fuel combustion or as by-products from industrial processes, or from natural processes such as forest fires. PAHs include mutagenic and carcinogenic compounds, emphasizing the need for efficiently controlling and predicting their fate and transport in the subsurface.
Some PAH remediation technologies that are currently in use include pump and treat systems, soil vapor extraction, and excavation. Each of these technologies are expensive and time consuming treatments. Bioremediation offers a cost effective and efficient remediation option.
Because PAHs are hydrophobic, they tend to sorb strongly to soil particles. PAH biotransformation in soil is often limited by the rate at which PAHs can desorb from the soil. Previous research has shown that surfactant or biosurfactant addition increases the rate and extent of desorption and biotransformation of phenanthrene in soil. In this research, the expression “biosurfactant” is a surfactant produced by bacteria and “surfactant” is used for a synthetically produced surfactant.
Few studies have examined how in-situ biosurfactant production effects desorption and biotransformation. The general goal of this research project was to examine the effects of in-situ biosurfactant production on desorption and biotransformation of phenanthrene from two different types of model poly-tetra-fluoro-ethylene (PTFE) particles, porous and non-porous.
Results indicate that phenanthrene initially adsorbed to PTFE particles can be biotransformed by biosurfactant and non-biosurfactant producing bacteria. Instantaneous desorption was increased in the presence of both strains of bacteria. Biosurfactant producing bacteria were just as effective at phenanthrene biotransformation per cell mass than non-biosurfaptant producing bacteria. It was determined that the maximum concentration in the aqueous phase was a factor of the mass of phenanthrene initially adsorbed and the partition coefficient. Most likely once phenanthrene desorbs from the surface of the particles, it can re-adsorb either to the particle surface or on biomass present in the column, before it is carried put of the column.

INFLUENCE OF BIOSURFACTANT AND NON-BIOSURFACTANT PRODUCING
BACTERIA ON PHENANTHRENE REMOVAL FROM MODEL SOILS by
Julie Ann Eyre
A thesis submitted in partial fulfillment o f the requirements for the degree o f
Master o f Science in
Chemical Engineering
MONTANA STATE UNIVERSITY-BOZEMAN
Bozemani Montana
December, 1999

ii
APPROVAL of a thesis submitted by
Julie Eyre
This thesis has been read by each member of the thesis committee and has been-found to be satisfactory regarding content, English usage, format, citations, bibliographic style, and consistency, and is ready for submission to the College of Graduate Studies. ,
Dr. A. Cunningham
(co-chair)
Dr. J. Sears
(co-chair)
(Signatured
(Signature)
(Date)
2L&. /X / / f ?
(Date)
Dr. J. Sears
(Dept. Head)
Approved for the Department of Chemical Engineering
/
(Signature)
/% / f f ?
(bate)
Approved for the College of Graduate Studies
Dr. B. McLeod
(Graduate Dean) (Signature) y Z
Z (Date)

Ill
STATEMENT! OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of the requirements for a master’s degree at Montana State University-Bozeman, I agree that the Library shall make it available to borrowers under rules of the Library.
If I have indicated my intention to copyright this thesis by including a copyright notice page, copying is allowable only for scholarly purposes, consistent with “fair use” as prescribed in the U.S. Copyright Law. Requests for permission for extended quotation from or reproduction of this thesis in whole or in parts may be granted only by the copyright holder.
Signature
Date

ACKNOWLEDGEMENTS
I would like to thank my committee members: Al Cunningham, John
Sears, Anne Camper, and Phil Stewart for their support, guidance, and interest in this research project. Their ability and patience in answering my numerous questions was greatly appreciated.
Thanks also to Ryan Jordan, acting as my mentor throughout this research project. His insight and knowledge was a significant help to me. Finally thanks to the Bioremediation Lab group, for many hours of in-sight in the lab as well as on presentations and writing.
Completion of this, and almost every project at the Center for Biofilm
Engineering, would be nearly impossible without the assistance of the finest laboratory manager: John Neuman. John’s attention to detail and desire to help students, at all hours of the night, including weekends, was an invaluble asset.
This research was supported by the Center for Biofilm Engineering at
Montana State University, a National Science Foundation-sponsored engineering research center.

V
TABLE OF CONTENTS
Page
1. INTRODUCTION ................................................................... ........................
Purpose ........................................................................................
Background .....................................................................
Prior Research ..................................* ............................
Biotransformation of Initially Adsorbed PAHs
Apparent Cumulative Removal (Desorption Plus
Biotransformatipn)
Surfactant Enhanced Bioremediation
Bioavailability .......................
................. 10
12
1
1
2. MATERIALS AND METHODS ..............................................
Biotransformation and Desorption ..................................................
Materials .......
Media Recipes ...................................................................
Methods ..................................................................
Artificial Particle Contamination with [14C]
Phenanthrene and Phenanthrene
Growth of Bacteria using a Chemostat .............. 19
Biotransformation of Initially Adsorbed
Phenanthrene
20
Extraction of Adsorbed Phenanthrene .......... 22
Calculation of Phenanthrene Biotransformed ....
Calculation of Phenanthfene Desorbed by 19SJ and P15
.............
Effectiveness of Phenanthrene Biotransformation
22
23
24
15
15
15
17
18
18
Model of Experimental System .............................. 25
Biotransformation Constant .......
Phenanthrene Partition Coefficient .....
27
27
Diffusivity in Aqueous Phase ................................. 28
3. RESULTS ......................... 29
Biotransformation of Initially Adsorbed Phenanthrene ............ 29
Apparent Cumulative Removal (Desorption Plus 31
Biotransformation)
Effectiveness of 19SJ vs P l 5 ..... .......................... ................. 37
Data Analysis with Model of Flow Thru System ........................ • 38
4. DISCUSSION ................. 42
Biotransformation ........................................................................... 42

vi
Apparent Cumulative (Desorption Plus Biotransformation)
Removal
45
Effectiveness of Biosurfactant Producer Versus Non-
Biosurfactant Producer
48
Application of A Flow Thru Model to Experimental Data ......... 49
5. DISCUSSION AND CONCLUSIONS ............................ ........................
REFERENCES CITED ...................
51
55
APPENDICES ....................................
A - BIOTRANSFORMATION DATA AND GRAPHS ..............
B -CUM ULATIVE REMOVAL AND DESORPTION DATA
AND GRAPHS
C - EFFECTIVENESS DATA AND GRAPHS ........................ 93
D - MODEL DATA AND GRAPHS ............................................... 96
58
59
72

vii
LIST OF TABLES
Table Page
2.1. Lureani Bertani (LB) Media .......................................................
2.2. Bushnell Hass (BH) Medium ...................................................
17
17
2.3. SWF 1/1 OF Medium ..........................................................
3.1. Student t Test Resuts for Biotransformation Data ..................... 31
3.2. Student t Test Results for Cumulative Removal Data ..................... 33
3.3. Student t Test Results for Desorption Data ............................. 35
3.4. Mass Balance Results for Column Studies ....................................... 36
18
3.5. Direct Cell Count Results and Activity Per Cell Results .............. . 37
3.6. Cells Initially Attached on Porous and Non-Porous Columns .......... 37
3.7. Coefficients Used in Model & Results of Model ................................. 41

Vlll
Figure
1.1
LIST OF FIGURES
Chemical Structure of Phenanthrene ..............................
Page
3
1.2 Chemical Structure of Surfactant Molecule ..................................... 4
1.3. Interactions of biosurfactant with soil and PAH
2.1. Biotransformation Experimental System
.........................
....................................
4
20
3.1. Bjotransformation of Adsorbed Phenanthrene on Porous
Particles Monitored by COg Production
3.2. Biotransformation of Adsorbed Phenanthrene on Non-Porous
Particles, Monitored by CO2 Production
29
30
3.3. Apparent Cumulative Removal From Non-Porous Particles ....... 32
3.4. Apparent Cumulative Removal From Porous Particles .................. 33
3.5. Apparent Desorption From Non-Porous Particles ...................... 34
3.6. Apparent Desorption From Porous Particles ..............................
3.7. Equilibrium Isotherms Measured for Phenanthrene on Porous
(P) and Non-Porous (NP) Particles
3.8. Experimental Diffusivity Data .............
34
39
40

ix
ABSTRACT
Polycyclic aromatic hydrocarbons (PAHs, e.g. phenanthrene) may occur in the environment as a result o f fossil fuel combustion or as by-products from industrial processes, or from natural processes such as forest fires. PAHs include mutagenic and carcinogenic compounds, emphasizing the need for efficiently controlling and predicting their fate and transport in the subsurface.
Some PAH remediation technologies that are currently in use include pump and treat systems, soil vapor extraction, and excavation. Each o f these technologies are expensive and time consuming treatments. Bioremediation offers a cost effective and efficient remediation option.
Because PAHs are hydrophobic, they tend to sorb strongly to soil particles. PAH biotransformation in soil is often limited by the rate at which PAHs can desorb from the soil. Previous research has shown that surfactant or biosurfactant addition increases the rate and extent o f desorption and biotransformation o f phenanthrene in soil. In this research, the expression “biosurfactant” is a surfactant produced by bacteria and
“surfactant” is used for a synthetically produced surfactant.
Few studies have examined how in-situ biosurfactant production effects desorption and biotransformation. The general goal o f this research project was to examine the effects o f in-situ biosurfactant production on desorption and biotransformation o f phenanthrene from two different types o f model poly-tetra-fluoroethylene (PTFE) particles, porous and non-porous.
Results indicate that phenanthrene initially adsorbed to PTFE particles can be biotransformed by biosurfactant and non-biosurfactant producing bacteria. Instantaneous desorption was increased in the presence o f both strains o f bacteria. Biosurfactant producing bacteria were just as effective at phenanthrene biotransformation per cell mass than non-biosurfaptant producing bacteria. It was determined that the maximum concentration in the aqueous phase was a factor o f the mass o f phenanthrene initially adsorbed and the partition coefficient. Most likely once phenanthrene desorbs from the surface o f the particles, it can re-adsorb either to the particle surface or on biomass present in the column, before it is carried put o f the column.

I
CHAPTER I
INTRODUCTION
Polycyclic aromatic hydrocarbons (PAHs, e.g. phenanthrene) are hydrophobic and they tend to sorb strongly to soil particles. PAHs include mutagenic and carcinogenic compounds, emphasizing the need for efficiently controlling and predicting their fate and transport in the subsurface. Current remediation technologies include pump and treat systems, soil vapor extraction, and excavation. These technologies are expensive and time consuming. Bioremediation potentially offers a cost effective and efficient remediation option. Previous research has shown that surfactant or biosurfactant addition enhances bioremediation (Alexander et al., 1991; Falatko et a l, 1992; Alexander et a l, 1993; Alexander et a l, 1992; Michelic, 1993). In this research, the expression
“biosurfactant” is a surfactant produced by bacteria and “surfactant” is used for a synthetically produced surfactant.
Purpose
The purpose o f this research was to examine the effects o f in-situ biosurfactant production on desorption and biotransformation o f phenanthrene from two different types o f model poly-tetra-fluoro-ethylene (PTFE) particles, porous and non-porous.
Two different strains o f bacteria, Pseudomonas saccharophilia P15, a non­ biosurfactant producing bacteria, and Pseudomonas aeruginosa 19SJ, a biosurfactant

2 producing bacteria, were used in this research project to determine i f initially adsorbed phenanthrene could be biotransformed. The effects that these bacteria have on desorption o f initially adsorbed phenanthrene was also examined. The mass o f phenanthrene removed by desorption and biotransformation was combined together to determine the apparent cumulative removal. Apparent cumulative removal is equal to the mass o f phenanthrene removed by desorption plus the mass removed by biotransformation.'
Physical morphology o f subsurface soils can also affect the bioavailability o f sorbed hydrocarbons. Naturally occurring soil particles contain pores o f many different sizes, many o f which are smaller than the size o f most microorganisms. Bioavailability o f a PAH may be limited by desorption and diffusion out o f these soil pores. Two different types o f PTFE particles were used in this research to examine the effects o f porosity on bioavailability o f initially adsorbed phenanthrene, one without pores and one with pores unavailable to microorganisms.
Previous research has shown that the addition o f surfactants or biosurfactants can increase the rate o f desorption (DiVincenzo, 1.996), however if applied above the critical micelle concentration (CMC), biotransformation is inhibited (Michelic, 1993). If surfactants or biosurfactants are applied below the CMC, biotransformation can be enhanced (Alexander et or/., 1991; Falatko et a l, 1992; Alexander et a l, 1993; Alexander e ta l, 1992).
Background
Polycyclic aromatic hydrocarbons (PAHs, e.g. phenanthrene) may occur in the environment as a result o f fossil fuel combustion, as by-products from industrial

3 processes, or from natural processes such as forest fires. PAHs include mutagenic and carcinogenic compounds, emphasizing the need for efficiently controlling and predicting their fate and transport in the subsurface. Figure 1.1 shows the chemical structure o f
Figure 1.1 - Chemical structure o f a phenanthrene molecule.
phenanthrene.
Once phenanthrene is present in the environment, it can partition into four phases.
Phenanthrene can adsorb to soil surfaces, dissolve in the bulk aqueous fluid, vaporize into vapor, or remain among a non-aqueous phase liquid (NAPL). Since phenanthrene is hydrophobic, most phenanthrene will be present adsorbed to the soil surface. This research focuses on the saturated zone, where vapors are not present.
Previous research has shown that the addition o f surfactants or biosurfactants can increase the rate o f desorption (DiVincenzo, 1996), however if applied above the critical micelle concentration (CMC), biotransformation is inhibited (Michelic, 1993). If surfactants or biosurfactants are applied below the CMC, biotransformation can be

4 enhanced (Alexander et a/., 1991; Falatko et a i, 1992; Alexander et a i, 1993; Alexander e ta l, 1992).
A surfactant molecule, or monomer, has a hydrophobic tail and a hydrophilic head. An example of the chemical structure of a synthetic surfactant molecule, Triton X-
100 is shown in Figure 1.2. One surfactant molecule is known as a monomer. In
Figure 1.2 - Example o f chemical structure o f a synthetic surfactant molecule, Triton X-100. N is approxim ately equal to 10. The hydrophobic tail is the carbon chain on the right side o f the structure.
soil/aqueous environments, the hydrophobic tail will most likely sorb to the soil surface
(as long as the surface is hydrophobic), while the hydrophilic head will be associated with the aqueous phase (Figure 1.3). As surfactant concentration increases, binding sites are filled on the soil surface, causing sorbed monomers to form hemicelles. With
Figure 1.3 - Interactions o f biosurfactant with soil and PAH

increasing surfactant concentration,'the binding sites become occupied and micelles will form in the aqueous phase. The concentration at which this happens is referred to as the critical micelle concentration (CMC). When hemicelles and micelles are present, the tails form hydrophobic regions that will promote the redistribution o f PAHs from the aqueous phase (Figure 1.3).
Physical morphology o f subsurface soils can also affect the bioavailability o f sorbed PAHs. Naturally occurring soil particles contain pores o f many different sizes, many o f which are . smaller than the size o f most microorganisms but large enough to allow diffusion o f PAHs into the pores. Bioavailability o f a PAH may be limited by adsorption, desorption, and diffusion into and out o f these soil pores.
Prior Research
Biotransformation o f Initially Adsorbed PAHs. Experimental evidence has shown that microorganisms are most effective at biotransforming dissolved organic chemicals, and that the concentration in the bulk water determines the rate o f uptake (Bouwer et al,
1998). Even i f a bacterial cell is in direct contact with sorbed materials, there are theoretical arguments and experimental evidence suggesting that direct biotransformation o f sorbed organic molecules will be insignificant. Van Loosdrecht et al.
(1990) have described how geometric analysis suggests that only a small fraction o f the bacterial surface, can be in direct contact with the adsorbed materials on the soil surface. Most o f the bacterial surface cannot be in direct contact with the solid and diffusion directly from the solid particle into the microbial cell is slow. The direct transfer o f substrate between

6 bacterial cell and soil surface would be o f limited capacity compared to transfer through the aqueous phase (van Loosdrecht et a l, 1990).
Alexander et al: (1986) suggested that microorganisms may biotransform a substrate that has minimal aqueous solubility by one or a combination o f these mechanisms. It could be biotransformed the instant it dissolves in water, after its aqueous solubility has been biologically enhanced, or by mechanisms involving physical contact with the solid phase o f the substrate. Growth o f pure cultures o f bacteria on naphthalene, phenanthrene, and anthracene was faster on those solid substrates having higher water solubility (Alexander et al., 1986). This research only compared the rate o f dissolution and degradation, and did not address the mechanisms by which microorganisms biotransformed substrates with minimal aqueous solubility. Additional work must be performed which assess these mechanisms (Alexander et a l, 1986).
Studies have been performed which demonstrate that the biotransformation o f an organic substrate can occur when Tt is initially adsorbed. Alexander et al.
(1995) performed a column study measuring the biotransformation o f phenanthrene initially adsorbed in column studies under constant, intermittent, or no water flow conditions. In all three flow conditions phenanthrene was biotransformed; however, less was biotransformed under intermittent or no water flow conditions. The authors concluded that this result was due to lack o f oxygen present in the column under intermittent or no water flow conditions.
Comelissen et al.
(1998) examined the biotransformation and desorption rates o f
15 PAHs initially adsorbed to sediments. Diphasic biotransformation profiles were observed: an initial phase o f rapid biotransformation, followed by a later phase o f slow

7 biotransformation. Desorption profiles were also biphasic, indicating that a large fraction desorbs fast,, while a smaller fraction desorbs much slower. The authors concluded from the biphasic biotransformation profiles that the initially rapid biotransformation is limited by microbial factors, while the slower biotransformation is limited by mass transfer.
Carmichael et al.
(1997) examined the rates o f biotransformation o f soils impacted with aged PAHs. The rates o f desorption were much slower than biotransformation, suggesting that desorption may control biotransformation. By keeping the bulk liquid concentration essentially zero, this study also determined that bacterial cells were able to biotransform the PAHs the instant they desorbed.
The rate o f substrate dissolution or desorption may limit bacterial growth. Linear biotransformation trends can be explained through the use o f a first order mass transfer equation (Volkering et a l, 1992). Volkering et al.
(1992) compared theoretical biotrahsformation curves, derived using a first order mass transfer model, to experimental
. curves obtained from biotrahsformation o f solid naphthalene. The results indicated that the biotransformation o f solid naphthalene could be described using the first order mass transfer equation. These results indicate that biotransformation depends on desorption or mass transfer from the sorbed or solid phase to the aqueous phase.
Michelcic et al.
(1993) developed a model which assumed that sorbed substrate was not biotransformed. It assumed that both suspended and attached cells could biotransform soluble substrate and that desorption o f sorbed compounds was instantaneous. In other words the soil-water distribution was always at equilibrium. This model best fits experimental data o f biotransformation o f a sorbed substrate (Michelcic et a l, 1993).

8
Other studies suggest that sorbed compounds are available to microorganisms without prior desorption (Crocker, 1995). Remberger et al.
(1986) suggested that sorbed substrates are available for biotransformation, though it was not proven whether biotransfomation occurred in the sorbed state or after desorption.
I
Apparent Cumulative Removal (Desorption Plus Biotransformation'). For a remediation technology to be viewed as a viable option, it must be efficient at removing both dissolved PAHs in the groundwater and PAHs sorbed to soil, while minimizing the extent o f the PAH contamination o f groundwater and soil. When bioremediation is applied correctly it has the potential to biotransform the dissolved PAHs, increase the mass transfer from the soil to the groundwater, and minimize the extent o f the PAH contamination in the groundwater and soil.
One field application o f bioremediation is natural attenuation. A large amount o f
. research has been performed on the effects that natural attenuation has on the size o f dissolved plumes in the environment. Natural attenuation is the natural degradation o f compounds in the environment via indigenous microorganisms, volatilization, or abiotic reactions (i.e. hydrolysis) (Borden 1994). If indigenous organisms that have the capability o f biotransforming the constituent o f interest, and electron acceptors are , available (oxygen, nitrate, etc.) then natural attenuation o f the dissolved plume through biotransformation processes is possible. Biotransformation usually occurs at the edge o f the plume where dissolved oxygen and other electron acceptors are plentiful in the groundwater. Biotransformation on the edge o f the plume causes the plume to shrink, decreasing the mobility o f the contaminants.

9
Laboratory batch experiments have demonstrated that the presence o f bacteria can enhance the rate o f dissolution. Alexander et al.
(1986) determined in a batch experiment with octadecane that the biotransformation rate was 200 times faster than its spontaneous dissolution. This, result was consistent with the findings o f Goma et al.
(1974).
Alexander et tit/. (1991) performed batch experiments with phenanthrene-impacted high organic and low organic content soils. In these experiments it was found that even when non-detectable concentrations o f phenanthrene desorbed from the surface, phenanthrene was still biotransformed. Phenanthrene was biotransformed from the high organic soils at a much slower rate than from low organic soils, or soils that phenanthrene can readily desorbed from. The results in this study indicate that even if a PAH will n o t.
readily desorb from soil, bacteria still have the capability o f biotransforming it.
Rijnaarts et al.
(1990) performed an experiment to investigate the effects o f desorption, from soil aggregates, on the biotransformation kinetics o f
. oc-hexachlorocyclohexane (a-HOH). Desorption and biotransformation was shown to be controlled by intraparticle mass transfer processes. This was determined by applying two different "models to the experimental data. A general first order model and sorptionretarded radial diffusion model were created. When applied to experimental data, the sorption-retarded radial diffusion model fit the data best, indicating that intraparticle mass transfer controlled desorption and biotransformation rates (Rijnaarts et a l, 1990).
Harms (1996) found similar results as Rijnaarts et al.
(1990). Harms examined the effect o f substrate separation from bacteria on biotransformation rates, and concluded that biotransformation rates can be enhanced by promoting the effective diffusivity o f a substrate or by decreasing the average distance between the cells and the substrate.

10
Bosma et al.
(1997) created a model for biotransformation that takes into account the biochemical activity o f microorganisms and mass transfer o f a chemical to the microorganism. Through the application o f this model to experimental data, it was again found that mass transfer o f the substrate to the microorganism was the most critical component o f biotransformation and not the biochemical activity o f the microorganism.
Michelic et al.
(1993) reported that high substrate concentration and fast mass transfer rate to the cell surface o f the microorganism are two benefits to a microorganism growing on a solid surface with a desorbing substrate. The enhanced mass transfer from the solid surface to the microorganism is due to a shorter diffusion distance.
Furthermore, it is believed that facilitated nutrient uptake at a solid surface may enhance biotransformation, especially in systems where nutrient concentrations are low (Michelic e ta l, 1993).
Beilin et al.
(1993) investigated the effects o f bacterial biomass on the sorption and transport o f naphthalene in soils. It was determined that the presence o f bacteria actually decreased the sorption o f naphthalene in the soils. Bacteria were first grown on the soil and then naphthalene was flushed through the column. The study does not mention if the bacteria were hydrophobic or hydrophilic, but it does indicate that once the substrate enters the bulk aqueous phase, it is unlikely to sorb to the biomass (Beilin et al.,
1993).
Surfactant Enhanced Bioremediation. In recent years the field o f bioremediation has been investigating the application o f biosurfactants to hydrocarbon impacted sites.
Most o f this research has been investigating the effects o f surfactant addition as opposed

11 to stimulating bacteria in-situ to produce biosurfactant, or injecting bacteria capable o f producing biosurfactant. For this review “biosurfactant” is a surfactant produced by bacteria, while “surfactant” is a surfactant produced synthetically.
Research has been performed which shows that surfactants must be applied above the critical micellar concentration (CMC) to promote mobilization o f hydrocarbons from soil surfaces (DiVincenzo, 1996). However, surfactants applied above the CMC prohibit biotransformation o f adsorbed hydrocarbons (Michelic, 1993).
Alexander et al.
(1991) found that non-ionic surfactants applied below the CMC.
increased phenanthrene biotransformation even though it did not enhance the extent o f desorption. The authors believe that the rate o f instantaneous desorption was increased by the addition o f surfactant, perhaps by altering the strength o f sorption or complexation o f the substrate in some way that the compound becomes more available for microorganisms without appearing in the bulk solution (Alexander et al.
1991).
Alexander et al.
(1993) examined whether surfactants applied at a distance from phenanthrene could still stimulate phenanthrene biotransformation. Surfactants were applied at concentrations above the CMC and below the CMC. In both cases phenanthrene biotransformation was increased: however, the addition o f surfactant below the CMC had the greater effect on phenanthrene biotransformation. Again the authors concluded that the rate o f instantaneous desorption was increased by the addition o f surfactant, perhaps by altering the strength o f sorption or complexation o f the substrate in some way that the compound becomes more available for microorganisms without appearing in the bulk solution (Alexander et al.
1993).

12
The availability o f hydrocarbons associated with the micellar phase o f surfactants has been in question. Jaffe et al.
(1996) created a mathematical model which examined the effective bioavailability, o f phenanthrene in the micellar phase. The model simulated experimental data well, indicating that a fraction o f the micellar-phase hydrocarbons were directly bioavailable. For three different surfactants tested, the bioavailable micellar phase decreased with increasing surfactant concentration. The authors concluded that the optimum surfactant concentrations for biotransfomation are below the CMC (Alexander et a/. 1996).
Falatko et al: (1992) examined the effects o f biosurfactants on the solubility and biodegradation o f petroleum hydrocarbons. Biosurfactants used for this study were produced on two different types o f substrates, gasoline and a mixture o f glucose with vegetable oil. It was determined that both types o f biosurfactant increased the solubility, but the gasoline grown biosurfactants did not inhibit biotransformation, whereas biotransformation was inhibited by biosurfactants produced on the glucose and vegetable .
oil mixture (Falatko et al.
1992).
Bioavailabilitv. Bioavailability is generally defined as the availability o f a chemical to biotransformation, and it is determined by the extent to which a chemical is exposed to organisms (Hamelink et al.
1994). It is generally believed that bioavailability o f hydrocarbons is limited by their low aqueous solubility, assuming that hydrocarbons can only be biotransformed in the aqueous phase. This is best explained by noting that intrinsic microbial kinetics are often best described by models that incorporate the dependence o f substrate concentration on the rate o f biotransformation

13
(i.e. Michaelis-Menton or first-order kinetics). Biotransformation rates should be faster when substrate concentrations are higher (up to a point, where the substrate concentration may either inhibit microbial activity via toxicity or saturate the enzymes responsible for biotransformation). Since hydrocarbon concentrations in water are limited by their solubility, one would expect that their bioavailability would be limited as well.
Partitioning into non-aqueous phases (non aqueous phase liquids, soil particles, etc.) may decrease the aqueous phase concentration o f the hydrocarbon to values well below its solubility limit, further limiting its bioavailability and resulting in slow, biotransformation rates (Jordan et al.
1999).
Bioavailability is not necessarily limited by solubility alone. In multiphase systems, where hydrocarbons may partition among aqueous and non-aqueous phases, biotransformation may be limited by partition rates. If the characteristic time for mass transfer, or partitioning, to the aqueous phase is slower than the characteristic time for biotransformation, then mass transfer may limit bioavailability (Jordan et al.
1999).
Examining the physical morphology o f subsurface solids can provide insight into to how subsurface solids can impact the biotransformation o f sorbed hydrocarbons.
Naturally occurring particles contain pores o f many different sizes, many o f which are smaller than the sizes o f most microorganisms. For example, analysis o f one o f the coarser sands size from the Borden Aquifer in Ontario, Canada, indicated that roughly 50 percent o f the intraparticle pore volume resides in pores that are less than 0.1 pm in diameter. Pores with a diameter larger than I pm comprised about 12 percent o f the total pore space, while only about 5 percent o f the pore volume was attributed to pores larger than 2 pm. Most indigenous bacteria range from 0.5 to 1.0 pm, therefore most bacteria

14 will be physically excluded from most o f the intraparticle pores o f these grains. The mean diameter o f intraparticle pores occupied by bacteria has been estimated to be typically larger than 2 pm. This is likely to be larger than the intraparticle pore spaces in most natural soils (Bouwer et al.
1998).
Even if bacteria are in direct contact with sorbed hydrocarbons, there are theoretical arguments and experimental evidence to suggest that direct biotransformation o f sorbed hydrocarbons will be insignificant. Most experimental evidence collected to date shows that bacteria are indeed most effective in using dissolved hydrocarbons, and the concentration in the aqueous fluid determines the rate o f biotransformation and bioavailability (Bouwer et al.
1998).
The impact o f adsorption on biotransformation is still not fully understood.
Complex models have been developed, which attempt to fully describe the effect o f adsorption on biotransformation. These models are most often dependent on the conditions o f the model simulation (i.e. numerical calculations o f the models, are needed in order to estimate the overall effect o f adsorption/desorption on the rate and extent o f biotransformation).

15
CHAPTER 2
MATERIALS AND METHODS
Biotransformation and Desorption
Materials
Two types o f poly-tetra-fluoro-ethylene (PTFE) particles, porous and non-porous, were used for all experiments. Porous particles were approximately 0.5 mm in diameter, and contained pores 100 to 300 nm in diameter (Dolder AG, Switzerland). Non-porous particles were also approximately 0.5 mm in diameter, and were provided by Du Pont
(Wilmington, DE).
[14C] Phenanthrene was purchased from Sigma Chemical Company (St. Louis, MO;
>98% purity, uniformly labeled, specific activity = ImCi/ml). A solution o f this compound was prepared in 200ml o f HPLC grade methanol. Crystalline phenanthrene was purchased from Sigma Chemical Company (St. Louis, MO). A working solution o f .
[14C] phenanthrene (0.7 pg/ml) and phenanthrene (0.2 mg/ml) was prepared in 50% GC grade methanol and 50% water (v/v).
Two strains o f bacteria, Pseudomonas saccharophiiia P l 5 and Pseudomonas aeruginosa 19SJ, were used for all studies. 19SJ was provided by Eric Deziel from
Centre de Recherche en Microbiologie Appliquee at Universite du Quebec. 19SJ has a hydrophobic cell surface and can produce a biosurfactant. The hydrophobicity o f the cell

16
. surface increases during the first stage o f biosurfactant production (Deziel et al.
1996).
P 15 was provided by Michael Aitken from the Department o f Environmental Sciences and Engineering at the University o f North Carolina at Chapel Hill. P l 5 also has a hydrophobic cell surface, but does not produce biosurfactant (Stringfellow et al.
1994).
Stock cultures o f both organisms were prepared by streaking for isolated colonies on
R2A agar (Difco; Detroit, MI). Isolated colonies were transferred to fresh R2A agar and incubated until a confluent lawn was present. The culture was harvested and placed in a
2% peptone - 20% glycerol solution and stored • in 2.0 milliliter vials at -70°C.
Approximately 3 loops o f each frozen stock culture were used to inoculate chemostats, as described below.
Two types o f particle columns were used for the experiments. 12 inch long, 1Z2 inch diameter PTFE columns, with stainless steel fittings, were used to artifically contaminate particles. Glass Omnifit liquid chromatography columns, 50 millimeter long and 10 millimeter in diameter, with 100 pm PTFE frits, were used for biotransformation studies.
All pump tubing used for particle contamination was size 16 Tygon®. All other tubing was PTFE tubing with an inner diameter o f 1/16 inch and outer diameter o f 1/8 inch. All pump tubing used for biotransfofmation studies was size 14 Pharmed® (Cole
Parmer). Other tubing types and sizes used for biotransformation studies were PTFE, inner diameter o f 1/32 inch and outer diameter o f 1/16 inch; PTFE, 1/16 inch inner diameter and 1/8 inch outer diameter; Viton® size 16; and peroxide, size 16. All pump tubing used for chemostat operation was size 16 peroxide. Viton® size 16 was also used y for chemostat operation.

17
Media Recipes. The recipes used for experiments are listed in the tables below.
Table 2.1 - Lureani Bertani (LB) Media
Constituent Amount (grams per liter)
Tryptone
Yeast Extract
10
5
NaCl 10
Table 1.2 - Bushnell Hass (BH) Medium
Constituent Amount (grams per liter)
0.2
MgSO4t TH2O
CaCl2*2H20
NH
4
NO
3
0.02
1.0
KH2PO4
K2HPO4
FeCl
1 . 0
1.0
0.05
To make LB broth, the constituents in Table 2.1 were added to deionized water and autoclaved. To make Bushnell Hass (BH) media, the first three constituents from
Table 2.2 were combined with 500 milliliters o f deionized water in a glass bottle and sealed. The fourth and fifth constituents were combined with 500 milliliters o f deionized water in a second glass bottle and sealed. Both bottles were autoclaved and mixed, once the liquid temperatures cooled to room temperature, maintaining aseptic conditions. The last constituent was added from a filter sterilized solution.

18
Table 2.2 - SWF 1/1 OF Medium ■
Constituent Amount (grams per liter)
KH2PO4
Na2HPO4
NaNO2
M g S 0 4*7H20
CaCl2*2H20
F eS 0 4*7H20
0.7
0.9
2.0
. 0.4
0.1
0.001
To make a mineral salt medium designated SWF 1/1 OF, the first three constituents, from Table 2.3, were combined in a glass bottle with 500 milliliter o f.
deionized water and sealed. In a second glass bottle the last three constituents were combined and sealed. Both bottles were autoclaved, cooled to room temperature, and combined under aseptic conditions.
Methods
Artificial Particle Contamination with F14Cl Phenanthrene and Phenanthrene.
PTFE columns were packed with either porous or non-porous particles and washed in a constant up-flow configuration with 1:1 (v/v) GC grade methanol and deionized water
(D.I.) for 24 hours. After washing, a 100 ml aliquot o f the [14C] phenanthrene and phenanthrene working mixture was added every 10 minutes until the appropriate mass o f phenanthrene was added. Particles were washed with this mixture for 2 hours. At the end

19 o f the 24 hours the column was drained and the particles were air dried under a fume hood overnight. They were stored in glass vials at 4°C until used.
Growth o f bacteria using a chemostat. 10 ml o f Lureani Bertani (LB) media was inoculated with approximately 3 loops o f 19SJ and P l 5 from frozen stock cultures, under aseptic conditions. Two test tubes were inoculated for each strain, and incubated at 3 S0C on a rotary shaker for 24 hours. 100 ml o f Bushnell Haas (BH) media with 1% napthalene was inoculated with I ml culture grown in LB media. One flask was inoculated for each strain, and incubated at 3 S0C on a rotary shaker for 48 to 60 hours. Once the color o f the media turned from light yellow to dark orange or reddish brown, cells were spun down at
6000 rpm and washed in 5ml o f mineral salt medium designated SWF 1/1 OF. The washed cells were used to inoculate each 750 ml chemostat containing. SWF 1/1 OF and 1% phenanthrene. The chemostats consisted o f IL Pyrex® glass bottles. The chemostats were continuously stirred, and compressed air was diffused into the liquid to provide oxygen saturation. The reactors were operated in batch mode until the media turned from a clear liquid to a dark orange or reddish brown color (usually after 48 to 60 hours). At this point the reactors were switched to chemostat mode, with a residence time o f 5 hours. Each chemostat was run continuously for 48 hours. At this point the soil columns were inoculated with effluent from the chemostat for two hours, at a flow rate o f 0.3 ml/min.
Surface tension o f the chemostat cultures were measured using a tensiometer
(Fisher Scientific; Fair Lawn, NI) to determine if biosurfactants were being produced.
The lowest surface tension measurement recorded for Pseudomonas aeruginosa 19SJ was
45 dynes/cm, indicating biosurfactant production. The surface tension, was also measured

20 for Pseudomonas saccharophilia P 15, but no reduction in surface tension, 75 dynes/cm, was observed, indicating biosurfactant was not being produced.
Biotransformation of initially adsorbed T14CI phenanthrene and phenanthrene.
Experiments were performed in packed columns in up-flow configuration. Six
Packed Soil
Column,
Inoculated with Bacteria
Pump
50 mm
Column
Solid Phase _
Extraction Iu b e
NaOH
Desorption
Liquid
Reservoir
Waste
Figure 2.1 - Biotransformation Experim ental System.
experiments were performed in triplicate. Porous particles were first inoculated with either P l5 or 19SJ, and then washed with sterile media, or only washed with sterile media
(abiotic controls). Non-porous particles were first inoculated with P l 5 and 19SJ, and then washed with sterile media, or only washed with sterile media (abiotic controls).
Prior to starting any experiment, all columns, end pieces, tubing, and frits were sterilized by autoclaving for 15 minutes. The experimental setup is shown in Figure 2.1.
Omnifit columns were packed with porous (2.3 grams) or non-porous (1.12 grams) particles that were artificially contaminated with phenanthrene previously.
Biological experiments were performed by inoculating the columns with chemostat effluent for two hours, at a flowrate of 0.3 ml/min. At this flowrate the column had a

21 residence time o f 6 minutes. After inoculation, the columns were washed with sterile media. For abiotic experiments, columns were only washed with sterile media. The start o f the inoculation or washing o f the abiotic columns with sterile media was considered time zero for the studies.
Effluent from the columns was collected in a nitrogen sparging vessel, where CO
2 in the aqueous phase was transferred to the gas phase. Anaerobic conditions were maintained in the nitrogen sparging vessels, by sparging the column effluent with high purity nitrogen gas. The nitrogen sparging vessels were 40 ml borosilicate glass EPA vials. Small ports were added on the side, so that size 16 tubing would fit snugly over the port. A constant volume was maintained in the sparging vessels.
Produced CO
2
was collected in sodium hydroxide traps, which consisted o f 5 ml o f 5 M sodium hydroxide in glass test tubes. The sodium hydroxide traps were changed every 8 hours. Biotransformation was determined by analyzing I ml o f the sodium
• hydroxide traps mixed with 9 ml o f liquid scintillation cocktail (Ultima Gold) for 14C, using a liquid scintillation counter (Packard A l 900).
The aqueous effluent from the nitrogen sparging Vessel was pumped through Iml solid phase extraction (SPE) tubes. The SPE tubes were conditioned off-line with HPLC grade methanol and D.I. water. At pre-determined time intervals, the SPE tubes were taken off-line and eluted with Iml o f a 1:1 methanol :methlyene chloride (v/v) solution.
This solution was mixed with 9 ml o f liquid scintillation cocktail (Ultima Gold) and was analyzed for 14C using a liquid scintillation counter (A l 900 Packard). Experiments were ended by allowing the columns to be pumped dry. The columns were then disconnected from the tubing and nitrogen was blown through the column, to remove excess water

22 from the columns. The columns were then left open in a fume hood and allowed to dry overnight.
Particles were removed from the columns and extracted in accordance to the extraction protocol, below. The columns were reconnected to the tubing and washed with
1:1 methanol and methlyene chloride. One milliliter o f the wash fluid was mixed with 9 ml o f liquid scintillation cocktail (Ultima Gold) and was analyzed with a liquid scintillation counter for 14C ( A l900 Packard).
A mass balance was performed using the wash fluid, initial particle contamination, final particle contamination, and mass that was removed from the column.
Extraction o f adsorbed phenanthrene. A known mass o f PTFE particles was placed in a 7 ml glass liquid scintillation vial. Five milliliters o f a 1:1 methylene chloride and methanol (v/v) solution was added to liquid scintillation vial and placed on a test tube rotator (~ 20 rpm) for one hour. The particles in the vial were allowed to settle, and the supernatant was decanted into a 10 ml glass liquid scintillation vial. Five milliliters o f the
1:1 methylene chloride and methanol solution was added again, and mixed on the test tube rotator. The supernatant was decanted into the 10 ml scintillation vial, and Iml o f both supernatants was analyzed in a liquid scintillation counter.
Calculation o f Phenanthrene Biotransformed. Biotransformation was monitored by analyzing for [14C] CO
2
derived from the biotransformation o f uniformly labeled [14C] phenanthrene. Total CO2 produced and phenanthrene biotransformed were calculated from the measured values o f 14C using the equations below.

23
First, radio-labeled phenanthrene that was biotransformed was determined by:
K P h e M e k d t r a m fOrmed = where:
D P M I /MnolCO2
^
\A/MrnolCO2
. MWPh Eq [1]
DPM/ml = the measured disintegrations per minute (DPM) o f the sodium hydroxide trap
MWPh = molecular weight o f phenanthrene
DP¥ - . * mwp h * lVmolph
/MnolCO2 /igPhen /MnolCO2
Eq [2]
1 mol Phenanthrene produces 14 moles CO
2
Second, non-labeled biotransformed phenanthrene was determined by:
/igPhen.nonlabeledlransformed
= /igPhenlabeled.transformed
* WSPhetinonldbded Eq [3]
/igPhenlabeled
Third, CO
2
production was determined by:
UgCO2 = ugPhen.nonlabeled lransformed *----—— *
2 MWPhen l/molPh
* MWCO2 E<^ ^ where:
MWCO
2
= molecular weight o f CO
2
.
Calculation o f Phenanthrene Desorbed. First, labeled phenanthrene desorbed was calculated by:

24-
/UgPhenJabeled.desorbed =
' D PM '
, ml j
DPM
*\ml ugPhen
Eq [5] where:
DPM/nil = measured value o f DPM from SPE tube elutent
DPM/^gPhen = Value from original stock o f radio-labeled phenanthrene
Second, non-labeled phenanthrene was calculated by
/igPhen.nonlabeled.transformed
= /igPhenlabeledtransformed * fiugPhen.nonlabeled
/igPhenlabeled
Eq [6]
Effectiveness o f Phenanthrene Biotransformation by 19SJ and P l 5. A biosurfactant producing bacteria, Pseudomonas aeruginosa 19SJ, and a non-biosurfactant producing bacteria, Pseudomonas saccharophilia P l 5 were utilized for this study. The cells were grown in chemostats as described above, except a 100 microliter (pi) aliquot o f radio-labeled phenanthrene was added to each 750 ml chemostat containing SWF 1/1 OF media. The chemostats consisted o f IL Pyrex® glass bottles, and were continuously stirred and compressed air was diffused into the liquid, to saturate it with oxygen. The reactors were operated in batch mode until the media turned from a clear liquid to a dark orange or reddish brown color (usually after 48 to 60 hours). At this point the reactors were switched to chemostat mode with a residence time o f 5 hours. Each chemostat was run continuously for 48 hours.

25
To determine the number o f cells present in the chemostat before inoculating the particle columns, total cell counts were performed. Cells from 10 milliliters (ml) o f chemostat effluent were spun down at 6000 rpm and resuspended in 10 ml o f SWF 1/1 OF media. These cells were then diluted using serial dilution, and 200 pi o f the IO'3 dilution were filtered onto a 25 mm diameter, 0.2 micron pore size polycarbonate filter
(Osmonics). The cells were then stained with acridine orange, and counted under ultraviolet light microscopy. Ten fields o f visions were counted for each sample and the average cell concentrations and corresponding standard deviations were calculated.
The activity o f the bacteria was determined by analyzing I ml o f each chemostat effluent, mixed with 9 ml o f liquid scintillation cocktail (Ultima Gold) on a liquid scintillation counter (A l 900 Packard). Each sample was counted for 10 minutes. Once the number cells per ml present in the chemostat and activity (Disintegration Per Minute -
DPM) per ml o f cells in the chemostat were known, one could determine the amount o f
■ ml by the number o f cells per ml in the chemostat.
Columns packed with clean, porous, and non-porous particles, were inoculated with the same number o f cells, and then washed with SWF 1/1 OF media for one hour at
0.3 ml/min. After I hour the cells were extracted according to the extraction o f adsorbed phenanthrene protocol. The supernatant was counted using a liquid scintillation counter.
These data allowed for the determination o f the number o f cells attached per column.
Model o f Experimental System. A mass balance for a PAH in a flow through system containing liquids and solids can be written as:

26
-K c + V 1 iw j
Here, V is the volume o f bulk water, C is the concentration in the bulk fluid, M is the volume or mass o f solids, S is the sorbed concentration, ky is the first order biotransformation rate coefficient, v; is the average bulk fluid velocity, and A c0Iumn is area o f the column. Equation 7 can be rearranged to yield:
V /4 y i •column
Eq [8]
Here, R is ratio o f volume o f particles per volume o f liquid in the column, S is the mass o f PAH adsorbed per volume o f particles. If C is small in the bulk fluid, then dC/dt will remain constant or approximately equal zero. Equation 8 can be written as:
R - = dt kbC +
ViA C '
V
Eq [9]
Mass transfer from the solid surface to the bulk phase can also be described using a traditional mass transfer equation: j i§ = i„ [ c ,- c ] W O ] at
Here, km is the mass transfer coefficient estimated from a chemical engineering correlation for mass transfer in packed bed, and Cs is the maximum concentration o f phenanthrene that can be present in bulk aqueous phase. Cs is approximated with
Equation 11, shown below.
mass.sorbed
Q _ mass.soil
s kd
Eq [11]

27
Here, kd is the partition coefficient between the aqueous phase and the sorbed phase.
If C in Equation 9 and Equation 10 are o f similar orders o f magnitude, this indicates that the disappearance o f PAH from the surface is dependent on mass transfer properties, such as fluid flow rate in the column, concentration o f PAH sorbed, and biotransformation o f the desorbed PAH.
Biotransformation Constant. The biotransformation constant, kb, was determined by Equation 12 shown below.
j I Phenbiotransformed
T phen.desorbed
+ phen.on..tubing
Where:
Eq [12]
ViA
Eq [13]
Here, v, is the average pore velocity, A is the area o f the column, and V is the pore volume in the column. Cs was determined by Equation 14 shown below.
mass.initially.adsorbed
C = ------ • maSS-Spn
----------- Eq [14] kJ
Phenanthrene Partition Coefficient. Phenanthrene partition coefficient was determined using non-labeled and 14C labeled phenanthrene. Known masses o f particles,
0.5 grams, were placed in 20 ml glass liquid scintillation vials. Water with varying concentrations o f phenanthrene were added to the particles, and continuously stirred. 14C was measured in the aqueous phase after two hours, using a liquid scintillation counter to determine the mass remaining the aqueous phase. The mass o f non-radio labeled

28 phenanthrene was calculated from the mass o f 14C, measurements as described above.
The mass o f phenanthrene adsorbed to the particles was then calculated based on the phenanthrene that was removed from the aqueous phase and an equilibrium constant calculated.
The mass transfer coefficient was estimated using a chemical engineering correlation for mass transfer in packed beds shown in Equation 15 below.
* . E q [15]
Here, Dv is the difrusivity o f phenanthrene in the bulk aqueous fluid, Dp is the diameter o f the particles, N rc is the Reynolds Number, and N sc is the Schmidt Number.
Diffusivitv in Aqueous Phase. The diffusivity o f phenanthrene was determined experimentally. The diffusivity was measured using non-labeled and 14C labeled phenanthrene. A known ratio o f labeled and un-labeled phenanthrene was placed in an
18ml flask, flask I. Flask I was connected with flask 2 that contained water with nophenanthrene, by a glass capillary tube. The capillary was sized to reduce the possibility o f convection and diffusion, 0.5mm in diameter and 2cm long. The disappearance o f phenanthrene from flask I was measured over time. Both flasks were continuously stirred.

29
CHAPTER 3
RESULTS
Biotransformation o f Initially Adsorbed Phenanthrene
Six different column experiments were performed to determine if phenanthrene initially adsorbed to PTFE, porous or non-porous, particles could be biotransformed.
Both abiotic and biotic experiments were performed. In the biotic experiments the columns were inoculated with either Pseudomonas saccharophilia P l 5 or Pseudomonas aeruginosa 19SJ. Both bacteria have the capability o f degrading aqueous phase phenanthrene, but 19SJ produces a rhamnolipid biosurfactant and P 15 does not.
With both porous and non-porous particles, 14C was detected in the sodium
1.200
P15
19SJ
No Cells
O 0.400
0.000
Time (hrs)
Figure 3.1 - Biotransformation o f adsorbed phenanthrene on porous particles monitored by CO 2 production. Experiments were completed in triplicate, and the error bars, representing one standard deviation, are shown but some are smaller than the data points.

30 hydroxide traps for both strains but not for the abiotic control experiments. This indicates that biotransformation only occurred in the presence o f bacteria. Since adsorbed
8 . 0 0
>P15
1 9 S J
No Cells
” 2 . 0 0
T i m e (hr*)
Figure 3.2 - Biotransformation o f adsorbed phenanthrene on non-porous particles monitored by CO2 production.
Experiments were completed in triplicate, and the error bars, representing one standard deviation, are shown but some are smaller than the data points.
phenanthrene was the only source o f 14C and no 14C was detected when the bacteria were not present, the results indicate that initially adsorbed phenanthrene was biotransformed.
At this point it is important to note that 5 times more phenanthrene was adsorbed to the non-porous particles than the porous particles. Less CO
2
was produced on the porous particles than on the non-porous particles, which corresponds to the amount o f phenanthrene initially adsorbed. It is believed that creating the pores in porous particles alters the surface chemistry o f the particles, which explains the adsorption characteristics o f phenanthrene.
Looking at Figures 3.1 and 3.2 it appears that similar amounts o f CO
2
were produced for both types o f bacteria. It is important to note at this point that cell mass is not taken into account in these figures. It is possible that two different amounts o f 19SJ and P l5 are present. Specific CO
2
production rates (mass CO
2
produced/cell) will be discussed later in this chapter.

31
A statistical analysis (t Test) was performed on the biotransformation data to determine i f the biotransformation was statistically significant. Table 3.1 shows the results. When comparing 19SJ and P15, there was no statistical difference, or the P value was much greater than 0.05. When comparing either strain with the control, there was a statistical difference, or the P value was less than 0.05. The CO
2
production in the presence o f 19SJ and P l 5 was statistically significant when comparing it to the abiotic control.
Non-Porous
19SJ vs P15
19SJ vs Control
Pt 5 v s Control
Porous
19SJ vs P t5
198J vs Control
P15 vs Control
10
6.E-01
8.E-03
3.E-03
4.E-01
2.E-03
2.E-03
27
9.E-01
4.E-02
1.E-02
5.E-01
1.E-03
2.E-03
Time Points (Mrs)
38
9.E-01
50
2.E-01
8.E-03
2.E-03
2.E-01
1.E-03
5.E-05
75
2.E-02
3.E-06
9.E-06
1.E-01
2.E-03
5.E-03
99
2.E-01
3.E-03
5.E-06
Table 3.1 - Results o f Students t Test for biotransformatiori data at various time points. 19SJ vs
P15 was a two tailed, two sample equal variance t Test. 19SJ vs Control and P15 vs Control was a one tailed, two sample unequal variance t Test.
124
3.E-01
Apparent Cumulative Removal (Desorption Plus Biotransformation)
Column studies were performed to determine i f the separate addition o f Pl 5 and
19SJ, to porous and non-porous particle columns, would enhance the apparent cumulative removal, desorption plus biotransformation, o f phenanthrene initially adsorbed to. porous or non-porous PTFE particles. Biotransformation and desorption were monitored using uniformly radio-labeled phenanthrene. Biotransformation was measured by analyzing sodium hydroxide traps for 14C. Desorption was measured by analyzing the methylene chloride methanol elutent from the SPE tubes. Cumulative percent removal was calculated by adding phenanthrene desorbed with phenanthrene biotransformed and

32 dividing by the mass o f phenanthrene initially adsorbed. It is important to note that this cumulative percent removal does not account for phenanthrene that desorbs and re­ adsorbs to system tubing. Apparent cumulative removal only examines phenanthrene recovered in the liquid effluent from the column.
. . , PhenJesorbed + PhenMotransformed . . . . .
apparent .cumulative.removal
= ---------------------------------------- ---------- *100 initial.adsorbed
Eq [1]
Experiments were completed in triplicate, and the error bars, representing one standard deviation, are shown but some are smaller than data points. Figures 3.3 and 3.4
show the results o f this experiment.
19SJ
P15
No Cells
60 8
Time (hours)
Figure 3.3 - Apparent cum ulative removal from non-porous particles. Experiments were completed in triplicate, and the error bars, representing one standard deviation, are shown but some are smaller than data points.
The presence o f Pl 5 or 19SJ resulted in approximately 50 percent removal o f the initial mass o f phenanthrene adsorbed to non-porous particles, compared to approximately 35 percent for the abiotic control. Again this does not include the percent o f phenanthrene that re-adsorbed to the column tubing. From porous particles, P l5 or
19SJ removed approximately 50 percent o f the initial mass o f phenanthrene sorbed,

33
Figure 3.4 - Apparent cumulative removal from porous particles. Experim ents were com pleted in triplicate, and the error bars, representing one standard deviation, are shown but some are smaller than data points.
compared to approximately 10 percent for the abiotic control. The addition o f P l 5 or
19SJ enhanced the apparent cumulative removal o f phenanthrene from porous and nonporous particles, compared to the abiotic control experiments. General trends o f the removal curves are consistent with first order removal, similar to that seen in the biotransformation experiments.
A statistical analysis (t Test) was performed on the cumulative removal data to determine if the cumulative removal was statistically significant. Table 3.2 shows the results. When comparing 19SJ and P 15, there was no statistical difference, or the P value was much greater than 0.05. When comparing either strain with the control, there was a
3 6
11 me P< Dints (Mrs
9 12 15 24 36 48 59 72
Non-Rorous
19SJ vs P15
19SJ vs Control
P l5 vs Control
Porous
19SJ v s P15
19SJ vs Control
P I S v s Control
0.405
0.613
0.520
0.634
0.194
0.066
0.011
0.085
0.061
0.023
0.003
0.145
0.232
0.188
0.498
0.000
0.006
0.001
0.008
0.005
0.001
0.699
0.983
0.766
0.442
0.061
0.035
0.229
0.009
0.024
0.007
0.669
0.330
0.061
0.001
0.479
0.369
0.001
0.226
0.117
0.056
0.001
0.002
0.001
0.223
0.429
0.003
0.213
0.210
0.000
0.027
0.000
0.003
0.092
0.001
0.323
Table 3.2 - Results o f Students t Test for cumulative removal data at various time points. 19SJ vs
P15 was a two tailed, two sample equal variance t Test. 19SJ vs Control and P15 vs Control was a one tailed, two sample unequal variance t Test.

3 5 . 0 0
3 0 . 0 0
2 5 . 0 0
2 0 . 0 0
1 5 . 0 0
I 0 . 0 0
5 . 0 0
0 . 0 0
I 9 S J
P 1 5
N o C e l l s
T i m e ( h r s )
Figure 3.5 - Apparent desorption from non-porous particles. Experim ents were completed in triplicate, errors bars are shown, representing on standard deviation, but are smaller than some data points.
statistical difference, or the P value was less than 0.05, for most time points.
Figures 3.5 and 3.6 show the results o f apparent phenanthrene desorption from porous and non-porous particles, or the phenanthrene that was collected in the SPE tubes.
On non-porous particles, for 19SJ and P 15, approximately 15 to 20 percent was removed in the aqueous phase as compared to approximately 35 percent for the abiotic control. On porous particles, for 19SJ and Pl 5, approximately 17 to 18 percent was removed in the aqueous phase as compared to approximately 13 percent for the abiotic control. These
20 T-
15 - -
19SJ
P15
No Cells
Time (hrs)
Figure 3.6 - Apparent desorption from porous particles. Experiments were completed in triplicate, errors bars are shown, representing on standard deviation, but are smaller than some data points.

35 results indicate that less phenanthrene was present in the bulk fluid when both types o f bacteria were present, for non-porous particles. For porous particles, more phenanthrene was present in the bulk fluid when both types o f bacteria were present. Again examining
Figures 3.1, 3.2, 3.3, and 3.4 it appears that both strains removed similar amounts o f phenanthrene, but these results are not on a per cell mass basis.
A statistical analysis (t Test) was performed on the desorption data to determine if the cumulative removal was statistically significant. Table 3.3 shows the results. When comparing 19SJ and P 15, there was no statistical difference, on porous and non-porous particles, or the P value was much greater than 0.05. On non-porous particles, when comparing 19SJ or P5 with the control, initially the P values are greater than 0.05 but at later time points the P value is less than 0.05. These results indicate that the difference at later time points is statistically significant, as opposed to earlier time points. On porous particles, when comparing 19SJ or P l 5 with the control, initially P values are less than
Non-Porous
19SJ vs Pt 5
19SJ vs Control
Pt 5 v s Control
Porous
19SJ v s P15
19SJ vs Control
Pt 5 v s Control
3 6 lime
9
Points
12
[firs)
15 24 33 48 59 72
0.41
0.52
0.48
0.47
0.51
0.56
0.59
0.55
0.42
0.34
0.35
0.50
0.61
0.16
0.04
0.05
0.07
0.06
0.04
0.05
0.10
1.00
0.74
0.39
0.06
0.08
0.09
0.08
0.06
0.05
0.23
0.14
0.24
0.00
0.00
0.01
0.02
0.00
0.00
0.69
0.66
0.54
0.62
0.79
0.75
0.05
0.05
0.07
0.15
0.28
0.79
0.05
0.05
0.08
0.19
0.32
0:91
Table 3.3 - Results o f Students t Test for desorption data at various time points. 19SJ vs
P15 was a two tailed, two sample equal variance t Test. 19SJ vs Control and P15 vs
Control was a one tailed, two sample unequal variance t Test.
0.05 and at later time points P values are greater than 0.05. These results indicate that at earlier time points the difference is statistically significant, but not at later time points.
A mass balance was performed for each study discussed and Table 4 shows the results. The mass o f phenanthrene initially adsorbed to the particles was known for each

36 experiment, and is shown in Table 3.4. The mass o f phenanthrene remaining on the particles was determined by extracting the remaining mass on the particle using the extraction protocol. The mass adsorbed on the tubing was measured by washing the tubing with 5 milliliters o f 1:1 (v/v) MeCkrMeOH. The mass desorbed in the bulk aqueous fluid and the mass biotransformed were measured as described above. The percentage attributed to each phase; desorption, biotransformation, remaining on particles, and adsorbed to tubing; were determined by Equation 2 shown below.
Phase% mass.measured
* . . .
------- ------------------*100 initiaLmass.present
Eq [2]
Particle Type Bacteria
Non-Porous I as J
Non-Porous Pl 5
Non-Porous None
Porous i as j
Porous P15
Porous None
Mass or
Phenanthrene .
InitiaIIyAdsorbed
(ng) iti
16
20
3.0
3.0
9.9
% Desorbed
18
21
32
18
17
13
%
Biotransformed
27
29
0
30
31
0
% Remaining On
Particles ti
9
7
Iti
18
9
%
Adsorbed on Tubing
41
33
50
2b
28
57
%
Recovered
93
92
89
93
94
80
Table 3.4 - Mass balance results for column studies.
For non-porous particles approximately 90 percent o f the initially adsorbed phenanthrene was removed and recovered in the aqueous phase, through biotransformation, or adsorbed to the system tubing. For porous particles approximately
80 to 90 percent o f the initially adsorbed phenanthrene was removed and recovered in the aqueous phase, through biotransformation, or adsorbed to the system tubing. The mass balance results indicate that for porous and non-porous particles approximately 50
' -t .
percent (desorbed plus mass adsorbed on tubing) o f the initially adsorbed phenanthrene was desorbed, while 30 percent o f the initially adsorbed phenanthrene was biotransformed. It’s also important to note that approximately 5 times more phenanthrene was initially adsorbed to the non-porous particles than the porous particles.

37
Effectiveness o f 19SJ vs‘P15
A baterial attachment study was performed to attain an understanding o f the number o f bacteria from each strain that attach to particles. Cells were initially grown on non-labeled and radio-labeled phenanthrene, and the number o f cells per amount o f
|19SJ field # average std dev
9
10
6
7
8
—T
2
3
4
5
Dilution
IDRMTml
I-):
Field of View area (cm*)
1.00E-08 cells
72
20
7
22
28
Z5
12
47
47
23
34
~ TC
219|
P15 field #
9
10
6
7
8
4
5
2
3
Dilution cells
T TC-
'■
' V . '
TE
53
58
30
79
64
8
21
72
50
46
Z3
’i T 2 E tl Ol
1764|
Area of filter (cm*)
0.5
Table 3.5 - Results for calculations to determine cells/activitiy.
radioactivity, or disintegrations per minute (DPM), were determined as described above.
Table 3.5 shows the results.
Porous and non-porous columns were inoculated with the same number o f cells
Bacteria
19SJ
P15
19SJ
P15
Particle Type
Non-Porous
Non-Porous
Porous
Porous
DPM
50.56
63.55
51.59
62.72
Cells on Column CU2/Cells
1.516+09
4.17E+08
1.54E+09
4.12E+08
2fc-09
1E-09 l t - 1 0
6E-10
Table 3.6 - Results o f biotransformation per initial number o f cells attached experiment. The values in bold indicate the higher biotransformation per cell number for each particle type.

38 for both bacteria, and. washed with sterile media for I hour. At the end o f the experiment the columns were pumped dry and the bacteria extracted from the particles. ' The supernatant was counted on a liquid scintillation counter and the number o f cells initially attached to the particles was determined. Table 3.6 shows the results.
Applying this data to the biotransformation data in shown above, the carbon dioxide formed per initial cell number attached to the particles in the columns was calculated for the first time point. Table 3.6 shows the results.
For non-porous particles, it appears the 19SJ, the surfactant producer, produced more carbon dioxide per cell number initially attached than P l 5, the non-surfactant producer, but it’s still within an order o f magnitude. For porous particles, it appears that
P 15, the non-surfactant producer, produced slightly more carbon dioxide per cell number initially attached than 19SJ, the surfactant producer, but it’s still within an order o f magnitude. Due to the accuracy o f cell number calculations there is no significant difference in the biotransformation o f phenanthrene per initial number o f cells attached.
In other words, one strain does not appear to be more effective than the other at phenanthrene biotransformation.
Data Analysis with Model o f Flow Thru System
The development o f a simple model was described in Chapter I . Equations 9 and
10 from Chapter 2 will be used to analyze data collected from this work. All constants used for these calculations are shown in Table 3.8.
R - = dt
V : +
ViA C
V
Eq [3]

39
Eq [4]
Here, R is the volume o f particles divided by the bulk fluid volume in the column, S is the initially adsorbed phenanthrene divided by the volume o f particles, and C is the concentration o f phenanthrene in the bulk aqueous fluid. The biotransformation constant, kb, was determined by Equation 5 shown below.
I PhenJjiotransformed kb = r phen.desorbed
+ phen.on. Jubing
Where:
Eq [5]
Here, v, is the average pore velocity, A is the area o f the column, and V is the pore volume in the column. Cs was determined by Equation 7 shown below.
mass.initially.adsorbed
C
= ----------mass.soil
----------- Eq [7] kd
Here, kd is the solid aqueous partition coefficient. The partition coefficient was
0.50
y = 46.215x
m NP
♦ P
____ Linear (NP)
____ Linear (P)
0.E-HDO 2.E-03 4.E-03 6.E-03 8.E-03 1.E-02 1.E-02
M aq (ug/ml)
Figure 3.7 - Experimental equilibrium isotherms for partition coefficient calculation.

40 determined experimentally as described in Chapter 2. The isotherms developed for porous and non-porous particles are shown in Figure 3.7. A linear approximation is used for non-porous and porous particles, where the slope o f the line is equal to Icd.
The mass transfer coefficient was estimated using a chemical engineering correlation for mass transfer in packed beds shown in Equation 8 below.
Eq [8]
Here, Dv is the diffusivity o f phenanthrene in the bulk aqueous fluid, Dp is the diameter o f the particles,
N rc is the Reynolds Number, and N sc is the Sherwood Number.
The diffusivity o f phenanthrene was determined experimentally, as described in
Chapter 2. The experimental results are shown in Figure 3.8. The diffusivity was calculated to be 6.5* IO"5 cm2/s.
0.65
0.6
f
% 0.55
j "
0.45
0.4
0 100 200
Tim e (min)
300 400
Figure 3.8 - Phenanthrene disappearance from flask I for diffusivity experiments.
The calculated coefficients needed to determine the C in Equations 3 and 4, shown above, are displayed in Table 8. In Table 8, c l is C from Equations and c2 is C from Equation 4. The values that were calculated for C from Equations 3 and 4 are within an order o f magnitude, indicating that this model can be used to generally describe the mass transfer processes occurring in the experimental system examined for this research.

41
C o e f f c ie n t
K d S /d t (ug Phen/ml particles*hrs) t (min) kb (1/min)
C l lug/mlj k m (i/s )
CS (ug/ml)
C l tug/ml/
N o n - P o r o u s P o r o u s
1 9 S J P15
1 9 S J P15
0 .0 5
6.7
0.07
0.11
6.7
0.08
0 .0 2
6,7
0.10
0 2
6.7
0.10
3.b d fc-04 /.9Ufc-U3 1.Utifc-U3 1 .U « t- 0 3
2 .4
0.05
2 .4
0.05
2 .4
0.03
2 .4
0.03
2 .9U L -U 2 4 .1 tit-U 3 2 .1 U b -U 2 . 2 .U U t-U 2
Table 3 . 8 - Coefficients used for model, and calculated values o f C. C l is C in
Equation 3 and C2 is C in Equation 4.
This can be assumed because the coefficients used for this model were derived from experimental data and mass transfer correlations, which all contain an error o f measurement.

CHAPTER 4
DISCUSSION
Biotransformation
The first purpose o f this research was to determine i f initially adsorbed phenanthrene could be biotransformed by 19SJ and P 15. By comparing the results o f abiotic control experiments with biotic experiments, it was determined that initially adsorbed phenanthrene was biotransformed. By applying a statistical analysis to the data collected, it appears the phenanthrene biotransformed was statistically significant when compared to the control.
Previous research has been performed which indicates that initially adsorbed or solid phenanthrene could be biotransformed, once it desorbs to the bulk aqueous phase
(Alexander et al., 1986; Comelissen et a l, 1998; Carmichael et a l, 1997; Volkering et q l, 1992; Michelic et al., 1993). Numerous batch studies have been performed indicating that phenanthrene must first desorb before it can.be biotransformed (Comelissen et al.,
1998; Carmichael et al., 1997; Volkering et al., 1992; Alexander et al., 1986).
Unfortunately, batch experiments do not allow for examining mechanisms o f biotransformation, such as when phenanthrene is biotransformed, as it desorbs or later in
' a ; the bulk fluid. Corenelissen et al.
(1998) compared desorption rates and biotransformation rates in batch using Tenax TA beads to keep the bulk fluid free o f phenanthrene. Batch experiments also do not simulate environmental, or i in-situ, conditions because they are usually well mixed.

43
Column studies can be designed so that conditions within the column are closer to environmental or in-situ conditions. Column studies can provide information on when phenanthrene is biotransformed, as it desorbs or in the bulk fluid.
The system used for this research was designed with a column residence time o f 6 minutes. Once phenanthrene entered the bulk fluid it was quickly transported out o f the column. The column effluent was sparged- with nitrogen, to create an anaerobic environment, minimizing phenanthrene biotransformation outside the column. The characteristic time for phenanthrene biotransformation is on the order o f several hours; thus biotransformation occurred for this research only if bacteria could acquire the phenanthrene directly o ff the surface or the instant it desorbed. There is a possibility that phenanthrene re-adsorbs to the particle surface or to biomass. This possibility will be discussed in further depth during the discussion o f the model applied to data collected.
For both types o f bacteria 14C was detected in the sodium hydroxide traps, indicating that phenanthrene was biotransformed. The only phenanthrene present in the columns was initially adsorbed to the particles; thus indicating that initially adsorbed phenanthrene was biotransformed. The column design allows for the conclusion that phenanthrene was biotransformed on the surface o f the particles or as it desorbed from the particle, which agrees with work performed by Carmichael et al.
(1997).
Examining Figures 3.1 and 3.2, in Chapter 3, it appears that biotransformation profile was biphasic: an initial phase o f rapid biotransformation followed by a second phase o f slow biotransformation. Biphasic biotransformation profiles have been observed multiple times before in both batch as well as column studies (Comelissen et

44 a l, 1998). The mechanisms causing these profiles have been examined in numerous research projects.
When examining the initial shape o f the curves, or the initial phase o f rapid biotransformation, it appears that CO2 production was first order. This first order trend was most likely caused by microbial limiting factors (growth rates, toxicity o f substrate, electron acceptors), as also seen by Comelissen et al.
(1998). In many instances the microbial growth rate, at low concentrations, is described by a first order equation with respect to substrate concentration (Characklis, 1990). Microbial growth and biotransformation in these experiments can most likely be described by such a first order equation with respect to substrate concentration. Assuming biotransformation is a function o f phenanthrene concentration in the aqueous phase indicates that phenanthrene must first desorb from the surface before it is biotransformed. Volkering et al.
(1992) and
Manilal et al.
(1991) have also seen these trends. Both o f these experiments were performed in batch as opposed to columns, so phenanthrene concentration was only limited by the rate o f desorption.
The decreased biotransformation, or second, slow phase o f biotransformation, could be a factor o f two different mechanisms. It could first be explained by the fact that once the initial fraction is desorbed, a second fraction will desorb more slowly, or not all.
At this point, mass transfer o f phenanthrene from the surface to the bacteria limits biotransformation. Comelissen et al.
(1998) and Manilal et al.
(1991) saw similar effects in biotransformation experiments.
A mechanism for the slower biotransformation could be diffusion o f phenanthrene from the location at which it desorbs from the surface to the cell. The diffusion, or mass

45 transfer, could be limiting the biotransformatitin rate. If the rate o f phenanthrene diffusion is much slower than the rate o f biotransformation, then diffusion could be limiting biotransformation.
Apparent Cumulative (Desorption Plus Biotransformation') Removal
It was reported by Bosma et al.
(1997), Harms (1996), and Rijnaarts et al.
(1990) that biotransformation is limited by mass transfer o f the substrate to the cell. A ll three studies indicate that biotransformation could be enhanced by decreasing the distance between the cells and the substrate. One way to decrease distance is by inoculating the soil surface with bacteria. It was reported by Alexander et al.
(1991) and Alexander et al.
(1986) that biotransformation enhanced the overall removal o f substrates. In both o f these studies the bacteria were most likely adsorbed to the soil surface.
Assuming that phenanthrene must first desorb from a particle to be biotransformed, the cumulative removal o f phenanthrene is equivalent to the amount o f phenanthrene desorbed. Comparing the cumulative removal for biotic and abiotic experiments, the data indicates that the presence o f the bacteria enhanced the desorption.
It is also important to note that approximately 30 percent o f the desorbed phenanthrene re-adsorbed to tubing after leaving the column. This indicates higher removal rates than what is shown in the Figures 3.3 and 3.4, from Chapter 3.
Assuming that desorption must occur before biotransformation can occur, Figures
3.2 and 3.3 indicates that biotransformation increases desorption, or mass transfer from the adsorbed phase to aqueous phase.

46
For non-porous particles, less phenanthrene was present in the bulk aqueous phase collection flask when bacteria were present than in the abiotic controls. Possible explanations for this result could be that phenanthrene that reached the bulk fluid was removed by biotransformation, adsorbed later in the column to bacteria (since both strains were hydrophobic), or re-adsorbed on sites on the particles. From this data it can be concluded that the presence o f bacteria decreased the mobility o f phenanthrene in the presence o f non-porous particles, since less phenanthrene was present in the aqueous fluid leaving the column.
For porous particles, more phenanthrene was present in the bulk aqueous phase collection flask when bacteria were present than in abiotic controls. For porous particles, it can be concluded that the presence o f bacteria increased phenanthrene mobility. This might be explained by the fact that bacteria may be more widely dispersed, or the bacteria increased desorption in pores, but the- bacteria were not able to physically access the pores. This limited access to the pore could limit the ability o f the bacteria to biotransform the phenanthrene. However, the total amount desorbed in aqueous phase or re-adsofbed on the tubing was greater for the abiotic control than when bacteria were present. Phenanthrene mobility on porous particles was not increased on decreased in the presence o f bacteria when compared to the abiotic control.
When the pores were created in the. porous particles, it is possible that the surface chemistry o f the particles was altered. This alteration could affect binding abilities, and even inhibit phenanthrene adsorption to the pore walls. Altering o f surface chemistry could explain the small mass o f phenanthrene that was initially adsorbed to porous

47 particles. It is probable that the pores walls had little phenanthrene adsorbed to them, as radiation alters the pore walls.
Beilin et al.
(1993) reported that bacterial biomass decreased the adsorption o f naphthalene in soil columns. Beilin et al.
(1993) did not report i f the cells were hydrophobic or hydrophilic. The reason for the decreased mobility in the aqueous phase, in the presence o f bacteria for non-porous particles, could be that a significnat portion o f the phenanthrene that was desorbed was biotransformed. In the abiotic control experiments, once phenanthrene was desorbed it was transported out o f the columns.
A large amount o f research has been performed on the effects, that natural attenuation has on the size o f dissolved plumes in the environment. Natural attenuation is the natural degradation o f compounds in the environment via indigenous microorganisms, volatilization, or abiotic reactions (i.e. hydrolysis) (Borden 1994). If indigenous organisms that have the capability o f biotransforming the constituent o f interest, and electron acceptors are available (oxygen, nitrate, etc.) then natural attenuation o f the dissolved plume through biotransformation processes is possible.
Biotransformation usually occurs at the edge o f the plume where dissolved oxygen and other electron acceptors are plentiful in the groundwater. Biotransformation on the edge o f the plume causes the plume to shrink, by decreasing the mobility o f the contaminants. In the presence o f bacteria, phenanthrene mobility in the aqueous phase was decreased in this research. This decrease in contaminant mobility is similar to what was seen in the results from natural attenuation research.

48
Effectiveness o f Biosurfactant Producer Versus Non-Biosurfactant Producer
The effectiveness o f a biosurfactant and a non-biosurfactant producing bacteria at phenanthrene biotransformation was examined. By normalizing the production o f carbon dioxide to the initial number o f cells attached, it becomes clear that 19SJ, the biosurfactant producing bacteria, was as effective as P l 5, the non-biosurfactant producing bacteria, at phenanthrene biotransformation on non-porous particles and porous particles.
Previous research has shown that the addition o f surfactants above the CMC increases phenanthrene desorption but inhibits biotransformation (DiVincenzo, 1996;
Michelic, 1993). Previous research has also shown that the addition o f surfactants and biosurfactants below the CMC enhances phenanthrene biotransformation (Alexander et a l, 1991; Falatko et a l, 1992; Alexander et a l, 1993; Alexander et a l, 1992). Alexander et al.
(1993) found that the addition o f surfactants below the CMC enhanced phenanthrene biotransformation but did not increase mobilization. Alexander et al.
(.1991) suggested that the addition o f surfactants could alter the strength o f sorption or complex compounds to make them more available. Jaffe et al.
(1996) suggested that a portion o f the micellar phase is directly bioavailable. Falatko et al.
(1992) found that'the addition o f biosurfactants only increased biotransformation if the biosurfactant was produced on gasoline. It appears that the behavior o f surfactant produced depends on the carbon source.
In this research biosurfactant production was monitored by measuring the surface tension o f the column effluent in the nitrogen sparging vessel. The surface tension o f the

49 column effluent for 19SJ, biosurfactant producer, was, reduced from 75 dynes/cm to 45 dynes/cm, indicating that biosurfactant was being produced in the column. It is probable that the biosurfactant concentration was much below the CMC. The surface tension o f the column effluent for P 15, non-biosurfactant producer, did not decrease; it remained at 75 dynes/cm throughout the experiment. The minimal biosurfactant production by 19SJ did not enhance phenanthrene biotransformation or desorption.
The aqueous column effluent was monitored for phenanthrene in the biotransformation and the overall removal experiments. When comparing the abiotic controls with the biotic experiments it was demonstrated that the mobility o f phenanthrene in the aqueous phase was decreased when bacteria were present. When, comparing phenanthrene in the aqueous column effluent for 19SJ, biosurfactant producer, and P 15, non-biosurfactant producer, it appears that biosurfactant production did not increase phenanthrene mobility or biotransformation.
Application o f the Model to Experimental Data
The simple model that was derived and explained in the Chapter 2 was applied to experimental data collected in this research. This model incorporates the effects o f biotransformation and adsorptipn/desorption on phenanthrene removal from a packed particle column. Concentrations in the aqueous phase were calculated using Equations 9 and 10 from Chapter 2. These concentrations were within an order o f magnitude, indicating that this model can be used to describe the mechanisms affecting phenanthrene removal.

50
First this model assumes that the change o f concentration in the bulk aqueous phase remains approximately constant. This model also assumes that both biotransformation and desorption affect phenanthrene removal from the particle surface, and also assumes that the maximum concentration in the aqueous phase is a function o f the mass initially sorbed on the particle and the partition coefficient. Once phenanthrene desorbs from the particle surface the possibility exists for it to re-adsorb or be retarded in the column by a similar mechanism.
~ a

CHAPTER 5
DISCUSSION & CONCLUSION
This research project was an effort to investigate the effects that in-situ biosurfactaht production has on phenanthrene biotransformation and desorption from model poly-tetra-fluoro-ethylene (PTFE) “soils”. This was accomplished by comparing results from biotransformation and desorption experiments with two strains o f bacteria, a biosurfactant producing bacteria, Pseudomonas aeruginosa 19SJ, and a non-biosurfactant producing bacteria. Pseudomonas saccharophilia P15. The effect o f soil morphology on biotransformation was examined by using two different types o f PTFE “soil” particles, porous and non-porous. Porous particles were approximately 0.5 mm in diameter, and contained pores, 100 to 300 nm in diameter (Bolder AG, Switzerland). Non-Porous particles were approximately 0.5 mm in diameter, and were provided by Du Pont
(Wilmington, DB).
It was found that initially adsorbed phenanthrene, om porous and non-porous particles, could be biotransformed by 19SJ, the biosurfactant producer, and P 15, the non­ biosurfactant producer. The experiment was designed such that the residence time in the column was much faster than the characteristic time for biotransformation. Therefore, the bacteria needed to acquire the phenanthrene as it was desorbed from the particle surface or directly from the surface, otherwise it was transported out o f the column. The result

52 calculated from the application o f a model indicates that it is a possibility that phenanthrene re-adsorbs to the particles or biomass.
The biotransformation curves obtained by were biphasic; showing an initial rapid biotransformation rate, followed by a second, slower phase o f biotransformation. The initial rapid phase o f biotransformation, in these experiments can most likely be described by a first order kinetic equation with respect to substrate concentration in the aqueous phase. Assuming biotransformation is a function o f phenanthrene concentration in the aqueous phase this indicates that phenanthrene must first desorb from the surface before it can biotransformed.
The decreased biotransformation, or second, slow phase o f biotransformation,.
could be a factor o f two different mechanisms. It could first be explained by the fact .that
once the initial fraction is desorbed, a second fraction will desorb more slowly or not at all. The mechanism for the slower biotransformation could be diffusion o f phenanthrene from the surface location at which it desorbs to the cell. Comelissen et al.
(1998) and
Manilal et al.
(1991) saw similar effects in biotransformation experiments.
It was found that the apparent cumulative removal, biotransformation plus desorption, was enhanced in the presence o f 19SJ and P l 5 bn both porous and nonporous particles. Comparing these results with abiotic controls allows for the conclusion that mass transfer from the surface o f the particles to the aqueous phase was enhanced in the presence o f 19SJ or P15.
In the presence o f 19SJ and Pl 5, the mass o f desorbed phenanthrene that was transported out o f the columns was decreased when compared to abiotic controls on nonporous particles and remained the same on porous particles. The results o f the

53 experiments with the non-porous particles concurs with previous research, in which the presence o f phenanthrene biotransforming bacteria increased mass transfer, but did not increase mobility. The results o f the experiments with the porous particles might be explained by the fact that bacteria increased desorption in pores, but the bacteria were not able to physically access the pores, or the bacteria were widely dispersed on the particle surface. This lim ited. access to the pore could limit the ability o f the bacteria to biotransform the phenanthrene. The increased mass in the aqueous phase could also be explained by the fact that the surface chemistry o f the porous particles was altered when pores were created in the particles.
It was found that 19SJ did not biotransform more phenanthrene than P15 per number o f cells initially attached in the soil columns, on non-porous and porous particles.
This could be due to the fact that 19SJ only produced a minimal amount o f biosurfactant, much less than the CMC.
A simple model o f a flow through system was compared with experimental data collected for this research. It was determined that the maximum concentration in the aqueous phase was a factor o f the mass o f phenanthrene initially adsorbed and the partition coefficient. Most likely once phenanthrene desorbs from the surface o f the particles, it can re-adsorb either to the particle surface or on biomass present in the column, before it is carried out o f the column.
This research is only the first step in examining the stimulation o f indigenous bacteria to produce biosurfactants. From this research, it can be concluded that one biosurfactant producing bacterial species cannot biotransform more phenanthrene per number o f cells initially present than one non-biosurfactant producing bacterial species.

54
From these experiments we can conclude that adsorbed phenanthrene is not more available for a biosurfactant producer than for a non-biosurfactant producer.

55
REFERENCES CITED
Alexander, M., Thomas, TM., Yordy, TR., Amador, TA. (1986) Rates o f Dissolution and
Biodegradation o f Water-Insoluble Organic Compounds. Applied and
Environmental Microbiology 52,290-296.
Alexander, M., Aronstein, B.N., Calvillo, Y.M. (1991) Effect o f Surfactants at Low .
Concentrations on the Desorption and Biodegradation o f Sorbed Aromatic
Compounds in Soil. Environmental Science and Technology 25, 1991.
Alexander, M., Aronstein, B.N. (1993) Effect o f a Non-Ionic Surfactant Added to the
Soil Surface on the Biodegradation o f Aromatic Hydrocarbons Within the Soil.
Applied Microbiology Biotechnology 39, 386-390.
Alexander, M., Devare, M. (1995) Bacterial Transport and Phenanthrene Biodegradation in Soil and Aquifer Sand. American Journal o f Soil Science Society 59, 1316-
1,320.
Beilin, C.A., Rao, P.S.C. (1993) Impact o f Bacterial Biomass on Contaminant Sorption and Transport in a Subsurface Soil. Applied and Environmental Microbiology 59,
1813-1820.
Bordan R.C. (1994) HandbookofBioremediation.
Boca Raton, Florida: Lewis
Publishers.
Bosma, T.N.P., Middeldrop, P.J.M., Schraa, G., Zehnder, A.J.B. (1997) Mass Transfer
Limitation o f Biotransformation: Quantifying Bioavailability. Environmental
Science and Technology 31, 248-252.
Bouwer, E.J., Zhang, W., Ball, W.P. (1998) Bioavailability o f Hydrophobic Organic
Contaminants: Effects and Implications o f Sorption-Related Mass Transfer on
Bioremediation. Groundwater Monitoring and Remediation W inter 1998, 126-'
138.
Carmichael, L.M., Christman, R.F., Pfaender, F.K. (1997) Desorption and Mineralization
Kinetics o f Phenanthrene and.Chrysene Contaminated Soils. Environmental.
Science and Technology 31, 126-132.
Characklis, W. (1989) Biofilms.
N ew York: John And Wiley Sons, Inc.
Comelissen, G., Rigterink, H., Ferdinandy, M.A., VanNoort, P.C.M. (1998) Rapidly
Desorbing Fractions o f PAHs in Contaminated Sediments as a Predictor o f the
Extent o f Bioremediation. EnvironmentalScience and Technology 32, 966-970.

Crocker, F.H., Guerin, W.F., Boyd, A.S; (1996) Bioavailability o f Naphthalene Sorbed to
Cationic Surfactant-Modified Smectite Clay. Environmental Science and
Technology 29, 2953-2958.
Deziel, E., Paquette, G., Villemuf, R., Lepine, F., Bisaillon, J. (1996) Biosurfactant
Production by a Soil Pseudomonas Strain Growing on Polycyclic Aromatic
Hydrocarbons. Applied and Environmental Microbiology 62, 1908-1912.
DiVincenzo, J.P., Dentel, S.K. (1996) Sorption-Desorption o f 1,2,4-Trichlorobenzene on
Soil: Anionic Surfactant and Cationic Polyelectrolyte Effects. Journal o f
Environmental Quality 25,1193-1202.
Falatko, D.M., Novak, J.T. (1992) Effects o f Biologically Produced Surfactants on the
Mobility and Biodegradation o f Petroleum Hydrocarbons. Water Environment
Research 64, 163-169.
Hamelink, J.L., Landrum, P.F., Bergman, H.L., Benson, W.H. (1994) Bioavailability:
Physically, Chemical, and Biological Interactions.
Chelsea, Michigan: Lewis
Publishers. '
Harms, H. (1996) Bacterial Growth on Distant Naphthalene Diffusing through Water,
Air, and Water-Saturated and Nonsaturated Porous Media. Applied and
Environmental Microbiology 62, 2286-2293. .
Jaffe, P.R., Saumyen, G.'(1996) Biodegradation Kinetics o f Phenanthrene Partitioned into the Micellar Phase o f Nonionic Surfactants. Environmental Science and
Technology 30, 605-611.
Jones, W.L., Dockery, J.D., Vogel, C.R., Sturman, P.J. (1993) Diffusion and Reaction within Porous Packing Media: A Phenomenological Model. Biotechnology and
Bioengineering 41, 947-956.
Jordan, R.N., Cunningham, A.B. (1999) Surfactant-enhanced bioremediation: A review o f the effects o f surfactants on the bioavailability o f hydrophobic organic chemicals in soils. Bioavailability o f OrganicXenobiotics in the Environment.
Kluwer Academic Publishers, pp. 463-496.
Manilal, V.B., Alexander, M. (1991) Factors Affecting the Microbial Degradation o f
Phenanthrene in Soil. Applied Microbiology Biotechnology 35, 4 0 1 -405.
Mihelic, J.R., Lueking, D.R., Mitzell, J.R., Stapleton, J.M. (1993) Bioavailability o f
Sorbed-and Separate-Phase Chemicals. Biodegradation 4,141-153.
Remberger, F.F, Robinson, K.G., Shabtai, Y. (1986) Bioavailability o f Sorbed
Hydrocarbons. Environmental Science and Technology^, 141-150.

57
Rice, A.M. (1999) Direct Observation o f Individual Cell Behavior within a Biofilm. (Un-
Pub). Montana State University.
Rijnaarts, H.H.M, Bachmann, A., Jumelet, J.C., Zehnder, J.B., (1990) Effect o f
Desorption and Intraparticle Mass Transfer on the Aerobic Biomineralization o f a-Hexachlorocyclohexane in a Contaminated Calcareous Soil. Environmental
Science and Technology 24,1349-1354.
Stringfellow, W., Aitken, M.D. (1994) Comparative Physiology o f Phenanthrene
Degradation by Two Dissimilar Pseudomonads Isolated from a Creosote-
Contaminated Soil. Journal o f Canadian Microbiology 40, 432-43 8.
Van Loosdrecht, M.C.M., Lyklema, J., Norde, W., Zehnder, A.J.B. (1990) Influence o f
Interfaces on Microbial Activity. Microbiological Review 54, 75-87.
Volkering, F., Breure, A.M., Sterkenburg, A., van Andel, J.G. (1992) Microbial
Degradation o f Polycyclic Aromitc Hydrocarbons: Effect o f Substrate
Availability on Bacterial Growth Kinetics. Applied Microbiology Biotechnology
36, 548-552.

APPENDICES

59
APPENDIX A - BIOTRANSFORMATION DATA AND GRAPHS

16.00
14.00
12.00
10.00
P15
19SJ
No Cells
0.00 i
SCTime (hrsfS

Non - Porous Particles - loaded on 3/23
Innoculated with 19SJ
Sample
Beginning
3
4
1
2
5
6
7
Time Exposed (min) Total Time (hrs)
0
570
1050
660
725
1480
1425
O
10
27
.38
50
75
99
124
1 (DPM/ml) 2 (DPM/ml) 3 (DPM/ml) Average (DPM/ml)
0 0 0 0
699 680 980 786
325 854 497 559
. 368 370 252 330
182 248 259 229
244 231 235 237
199 219 256 224
259 244 230 244
Innoculated with P15
Sample Time Exposed (min) Total Time (hrs)
Beginning 0
1 570
2 1050
3 660
4 725
5 1480
6
7
1425
1555
0
10
27
38
50
75
99
124
1 (DPM/ml) 2 (DPIWmI) 3 (DPIWmI) Average (DPIWmI)
0 . 0 0 0
919 .
735 903 . 852
677 621 419 573
345 383 233 320
241 298 276
272
249 261 265 258
254 239 253 249
239 233 232 ’ 234

Non - Porous Particles - loaded on 3/23
Innoculated with 19SJ stdev ug phen labeled Stand. Dev. Ug phen not labeled stdev (0ZoremovaII)
0.000000
0.000
0.000
0.001085
0.001747
0.279
0.450
1.745
2.810
•
0.000437
0.000270
0.000043
0.113
0.069
0.011
0.703
0.434
0.070
0.000187
0.000094
0.048
0.024
0.301
0.150
Total DPM Collected ug Phen labeled
0 0.00000
586 0.00509
945 0.00362
1075 0.00214
1104 0.00148
1141 0.00153
1165 0.00145
1209 0.00158
Innoculated with P15 stdev
- 0.000000
0.000659
0.000878
0.000505
0.000184
0.000052
0.000053
0.OOOO26
Stand. Dev. Ug phen not labeled stdev (%removall)
0.000
0.000
0.170
1.060
0.226
1.413
0.130
0.812
0:047 0.297
0.013
0.014
0.007
0.084
0.084
0.042
Total DPM Collected ug Phen labeled
0 0.00000
652 0.00552
1025 , 0.00371
1145 0.00207
1216 0.00176
1275 0.00167
1324 0.00161
1358 0.00152

Non - Porous Particles - loaded on 3/23
Innoculated with 19SJ ug phen not labeled
0.000
1.310
0.932
0.550
0.382
0.394
0.374
0.407
ug Phen Degraded Labeled Cumulative ug phen degraded non-labeled ug C02 stdev C02
0.00000
0.00
0.00
0.00
0.00379
1.31
4.12
0.88
0.00612
2.24
7.05
1.41
0.00696
2.79
8.77
0.35
0.00715
3.17
9.97
0.22
0.00739
3.57
11.21
0.04
0.00754
3.94
12.39
0.15
0.00783
4-35 13.67
0.08
Innoculated with P15 ug phen not labeled ug Phen Degraded
0.000
1.420
0.954
0.533
0.453
0.431
0.415
0.391
0.0000
0.0042
0.0066
0.0074
0.0079
0.0083
. 0.0086
0.0088
ug phen degraded non-labeled ug C02 stdev C02
0.00
- 0.00
0.00
1.42
4.46
0.53
2.37
7.46
0.71
2.91
9.14
0.41
3.36
10.56
0.15
3.79
11.91
0.04
4.21
4.60
13.22
14.44
0.04
0.02
a

Non Porous Particles - loaded on 3/11
Sample Time Exposed (min) Total Time (hrs) " 1 (DPM/ml)
Beginning 0 0 0.0
i
2
480
870
8
23
32.2
42.2
3
4.
5 ,
1590
1800
1000
49
79
96
39.3
46.2
38.9
2 (DPM/ml) 3 (DPM/ml) Average (DPM/ml). Stand. Dev.
0.0
0.0
0.00
0.0E+00
44.0
44.0
0.08
4.4E-05
36.0
55.3
4.50
6.4E-05
47.3
41.7
2.77
2.7E-05
38.6
37.6
0.81
3.1E-05
50.5
47.0
5.44
3.8E-05

Non Porous Particles - loaded on 3/11
Total DPM Collected ug phen labeled ug phen not labeled Cumulative ug C02
. 0 0.0
0.0E+00 0.0E+00 0.0E+00
-200 0.0
1.4E-04 1.4E-04 4.4E-04
-377 0.0
7.5E-03 7.6E-03 2.4E-02
-563 0.0
4.6E-03 1.2E-02 3.8E-02
-759 0.0
1.4E-03 1.4E-02 4.3E-02
-932 0.0
9.1E-03 2.3E-02 7.1E-02

3.500
3.000
2.500
2.000
1.500
1.000
0.500
0.000
0
P15
1 9 S J
No Cells
20 40
Time (hrs)
60 80

Porous Particles - loaded on 3/23
Innoculated with 19SJ
Sample
Beginning
3
4
1
2
Time Exposed (min) Total Time (hrs)
' O
540
915
1455
1470
24
49
0
9
73
1 (DPIWmI)
0
281
281
273
280
2 (DPIWmI) 3 (DPIWmI) Average (DPIWmI) stdev ug phen labeled
0 0 0 0.000000
227 280 263 0.000200
293 329 301 0.000160
281 223 259 0.000202
229 276 261 0.000184
Porous Particles - loaded on 3/23
Innoculated with P15
Sample Time Exposed (min) Total Time (hrs)
Beginning 0
1
2
540
915
3 1455
. 4 1470 .
24
49
0
9
73
1 (DPIWmI)
0
260
281
274
2 (DPIWmI)
0
322
344
286
216 239
3 (DPIWmI)
0
278
326
312
173
Average (DPIWmI)
0
287
317
290
210 stdev ug phen labeled
0.000000
0.000207
0.000208
0.000125
0.000218

Porous Particles - loaded on 3/23
Innoculated with 19SJ
Stand. Dev. Ug phen not labeled stdev (%removall)
0.000
0.026
0.021
0.026
0.024
0.000
0.856
0.684
0.865
. 0.790
Total DPM Collected ug Phen labeled
0 0.000000
63
164
223
0.001700
0.001948
0.001677
284 0.001692
ug phen not labeled
0.0000
0.2219
. 0.2542
0.2189
0.2208
Porous Particles - loaded on 3/23
Innoculated with P15
Stand. Dev. Ug phen not labeled stdev (%removall)
0.000
0027
0.027
0.016
0.028
0.000
0.889
0.892
0.534
0.935
Total DPM Collected ug Phen labeled
0 0.000000
87
204
0.001855
0.002052
294 0.001880
304 0.001357
ug phen not labeled
0.0000
0.2422
0.2678
0.2454
0.1771

Porous Particles - loaded on 3/23
Innoculated with 19SJ
Ug Phen Degraded Labeled Cumulative ug phen degraded non-labeled ug C02
0.00000
0,000
0.00041
0.00106
0.00144
0.00184
0.222
0.476
0.695
0.916
0.000
0.697
1.496
2.184
2.878
stdev C02
0.00
0.08
0.07
0.08
0.08
Porous Particles - loaded on 3/23
Innoculated with P15 ug Phen Degraded
0.00000
0.00056
0.00132
0.00190
0.00197
ug phen degraded non-labeled ug C02
0.000
0.242
0.510
0.755
.0.000
0.761
1.602
2.373
0.932
2.930
stdev C02
0.00
0.09
0.09
0.05
0.09

Porous Particles - loaded on 3/5
Sample Time Exposed (min) Total Time (hrs)
Beginning 0
1 480
2 870
3 1590
4 1800
5 1000
49
79
96
0
8
23
1 (DPM/ml) 2 (DPIWmI) 3 (DPIWmI) Average (DPIWmI) Stand. Dev.
0.0
21.4
47.5
55.4
38.7
42.4
0.0
33.8
35.8
47.2
38.0
42.3
0.0
17.8
42.4
27.3
25.6
45.6
0.00
0.00
1.90
3.27
0.00
3.41
.
0.0E+00
5.4E-05
3.8E-05
9.4E-05
4.7E-05
1.2E-05

Porous Particles - loaded on 3/5
Total DPM Collected ug phen labeled ug phen not labeled Cumulative ug C02
0 0.0
0.0E+00 0.0E+00 0.0E+00
0 . 0.0
0.0E+00 0.0E+00 0.0E+00
-191 0.0
3.2E-03 3.2E-03 1.0E-02
-374
-574
. -757
0.0
0.0
0.0
5.4E-03
0.0E+00
5.7E-03
8.6E-03
8.6E-03
1.4E-02
2.7E-02
2.7E-02
4.5E-02

72
APPENDIX B - CUMULATIVE REMOVAL AND DESORPTION DATA AND
GRAPHS

N o n -P o r o u s P a r tic le s
50
40
30
_____________________________, i i
20
H
H J
♦
A
10 A
•
■ "
\
■
" e
-
♦
B
♦
19SJ
■ P15
*
No Ce Hs
O
O 50
100

Non-Porous
Total Phen Removed (C02 and Desorbed)
19SJ
Time
0
10
27
38
50.
75
99
124
Phen (C02) error bars C02 Time
0.00
0.00
1.31
1.97
0.93
0.55
0.38
3:18
0.79
0.49
0.39
0.08
0.37
0.34
0.41
0.17
83
97
106
123
36
49
59
73
12
15
26
6
9
0
3
Phen (Aqueous) error bars desorp Cumulative Tim
0.000
0.000
0
0.252
0.442
3
0.389
0.594
0.329
0.420
9
0.232
0.247
0.338
0.229
. 0.366
0.156
0.121
0.089
10
12
15
26
0.158
0.063
27
0.115
0.246
36
0.118
0.110
0.160
0.048
38
49
0.109
0.110
0.164
,
0.047
0.085
0.097
50
59
73
75
83
97
''X
99
123
124
Cumulative Phen
2.28
2.51
2.76
3.10
4.03
0.00
0.25
0.64
0.97
. 4.26
% Desorbed Error Bars
0.0
6.1
0.00
1.6
0.44
.4.0
. 0.59
0.42
14.2
1.97
15.7
0.37
17.2
0.16
19.4
25.2
26.6
0.12
3.18
0.09
4.81
30.0
0.79
4.97
31.0
0.06
5.35
5.46
33.4
34.1
0.49
0.25
5.58
5.98
6.09
6.19
6.57
6.73
7.14
34.9
37.3
38.0
38.7
41.0
42.1
44.6
0.16
0.08
0.048
0.047
0.34
0.097
0.17

C
Non-Porous
Total Phen Removed ( 0 0 2 and Desorbed)
P15
Time
27
Phen (C02) Time
0 0.000
10 • 1.420
0.954
38 0.533
50 0.453 ;
75
99
124
0.431
0.415
0.391
24
33
48
59
72
83
95
12
15
6
9
0
3
106
121 error bars C02 Phen (Aq) Time
0.000
0.000
1.198
1.596
0.327
0.460
0.918
0.335
0.406
0.094
0.095
0.048
0.258
0.224
0.324
0.202
0.190
0.179
0.168
0.159
0.153
0.143
0.192 '
33
38
48
50
59
12
15
24
27
72
75
83
95
99
121
124
9
10
0
3
6 error bars desorption
0.0
0.4
0.6
0.4
0.4
. 0.2
0.1
0.1
0.1
0.2
0.2
0.0
0.0
0.1
0.1
5.1
5.3
5.8
5.9
6.1
2.6
2.9
3.1
3.4
4.4
4.6
0.0
0.3
0.8
1.2
6.5
6.7
6.8
7.3
7.4
7.8
'
16
18
19
21
27
0
2
5
7
29
32
33
36
37
38
41
42
43
45
47
49
0.00
0.07
0.29
0.19
0.06
0.06
0.10
1.20
0.71
0.50
0.44
0.19
0.14
0.11
0.12
3

Cumulative
Non-Porous
5
6
0
4
8
10
14
22
32
49
62
80
96
Phen
0.0
0.1
0.2
0.4
1.0
1.4
2.1
3.1
4.0
5.0
5.4
5.9
6.3
% Desorbed
7
11
2
5
0
1
1
16
21
25
28
30
32
Error Bars
0.00
0.13
0.13
0.35
0.61
0.20
0.74
1.20
2.39
2.30
0.16
0.23
0.14
ON

50 T—
75 40
> o
S 30
■ P15
* No Cells
P o r o u s P a rticles
B
B
*
---------9 ----------
9
I
3
E
<D
I
20 o
1 0
B
A
A
A
* *
----------------
40 Time (hours) 60
A
80

Porous
Total Phen Removed (C 0 2 and Desorbed)
19SJ
Time
24
49
73
'
0
9
Phen (C02) std overall Time Phen (Aqueous) std overall
0.0000
0.0000
0 0.0000
0.00
0.2219
1.1151
3 0.0535
0.13
0.2542
0.8919
6 0.0665
0.20
0.2189
1.1275
9 0.0567
0.09
0.2208
1.0299
15 0.0641
0.14
24 0.0666
0.55
36 0.0641
0.17
48 0.0588
0.26
60
73
0.0542
0.0521
0.08
0.27
Cumulative Tim Cumulative Phen % Desorbed Error Bars
0 0.00
0 0.000
3 0.05
2 0.129
6 0.12
4 0.202
9 0.18
6 0.087
9 0.40
13 1.115
15 0.46
15 0.138
24 0.53
17 0.554
24 0.78
26 0.892
36 0.85
28 0.170
48 0.91
30 0.059
49 1.13
37 1.1275
60 1.18
39 0.08
73 1.23
40 0.27
73 1.45
48 1.0299

Porous
Total Phen Removed (C 0 2 and Desorbed)
P15
Time
.24
49
73
0
9
Phen (C02) std overall
0.0000
0.0000
0.2422
1.1591
0.2678
1.1626
0.2454
0.6965
0.177j
1.2187
Time Phen (Aq) std overall Time
0 0.0000
0.0000
3 0.0606
0.3529
6 .
0.0719
0.1526
9 0.0524
0.0889
15 0.0532
0.0801
24
36
0.0644
0.0590
0.3681
0.2888
48 0.0595
0.1775
60 0.0577
0.2842
73 0.0499
0.1298
9
15
24
24
36
48
49
60
73
73
0
3
6
9
Phen Cumulative
0.000
0.061
0.132
0.185
0.427
0.480
0.545
0.812
0.871
0.931
1.176
1.234
1.284
1.461
% Desorbed Error Bars
0 0
2 0.3529
4 0.1526
6 1.1591
14 0.0889
16 0.0801
18 0.3681
27 1.1626
29 0.2888
31 0.1775
39 0.6965
41 0.2842
42 0.1298
48 1.2187

Cumulative
Porous
Time
10
14
22
.32
6
8
49
62
80
96
4
5
O
Phen
0.5
0.7
0.8
1.0
0.0
0.1
0.1
0.2
0.3
0.4
1.1
1.2
1.3
% Desorbed error bars
0 0.00
.
1
1
0.07
0.03
2
3
0.19
0.35
*
4
• 5
0.20
0.20
0.21
. 7
8
10
0.16
0.15
11
12
13
0.04
0.12
0.22

35.00
30.00
25.00
$ 20.00
£15.00
10.00
1 9 S J
P15
No Cells
Time (hrs)

Teflon - Non Porous Particles
Phenanthrene Loaded on 3/23
Innoculated with 19SJ - Surfactant Producer
Sample Time Exposed (min)
Beginning
1
2
5
6
3
4
7
8
9
I
0
170
175
195
180
180
540
540
870
695
780 10
11
12
13
637
760
645
14 920
Total Time (hrs) 1 (DPM/ml)
0
3
12
15
6
9
95
106
121
24
33
48
59
72
83
72
65
71
59
67
69
0
123
187
147
115
109
103
134
130
2 (DPM/ml) 3 (DPM/ml)
0
109
155
178
145
208
333
189
134
80
72
67 .
66
63
89
69
72
63
66
77
0
222
359
267
157
128
172
90
80
61
Average (DPM/ml)
0
151
234
198
139
148
203
138
95
69
71
66
65
66
98
Mass Balance
Final Bead
Initial Bead
Phenanthrene Wash
Desorbed
Biotransformed
% Recovered
DPM/g
647.06
Total DPM
731.18
8500.00
3897.95
9605.00
3897.95
hot phen (ug) cold phen (ug) % Recovered
0.0047
1.22
0.0622
16.01
0.0252
6.50
2.90
4.35
8
100
41
18
27
93

Teflon - Non Porous Particles
Phenanthrene Loaded on 3/23
Innoculated with 19SJ - Surfactant Producer
Stand. Dev. (overall remo stdev ug phen not label Hot Desorbed (ug) Cold Desorbed (ug) Total Cold ug desorbed (ug)
0.00
0.00
0.00000
0.00
0.64
1.14
0.65
0.23 ..
0.55 >
1.23
0.52
0.10
0.18
0.10
0.04
0.09
0.20
0.08
0.00098
0.00151
0.00128
0.00090
0.00096
0.00131
0.25
0.39
0.33
0.23
0.25
0.34
0.23
0.35
0.11
0.06
0.02
0.00089
0.00062
0.00045
0.16
0.11
0.01
0.07
0.02
0.03
0.00 .
0.01
0.00
0.01
0.00046
0.00043
0.00042
0.00043
. 0.12
0.11
0.11
0.11
0.29
0.05
0.00064
0.16
0.00
0.25
0.64
0.97
1.20
1.45
1.79
2.02
2.18
2.29
2.41
2.52
2.63
2.74
2.90

Non-Porous particles - Loaded on 3/11
Beginning
1
2
3
4
5
6
7
8
9
10
11
12
Sample Time Exposed (min)
0
210
60
75
120
120
240
515
550
1020
. 780
1110
1000
Total Time (hrs) 1 (DPIWmI)
0.0
3.5
4.5
5.8
7.8
9.8
13.8
22.3
31.5
48.5
61.5
80.0
96.7
245
415
550
373
345
313
368
294
0
69
48
79
296
2 (DPIWmI)
0
80
75
127
419
283
602
823
523
657
305
319
264
3 (DPIWmI)
0
102
169
440
293
480
807
963
933
344
‘ 316
262
Average (DPIWmI) Stand. Dev. (overa
0 0.00
84 0.12
67 0.12
. 125 0.34
385 0.58
274 0.19
499 0.71
727 1.14
620 2.28
645 2.19
321 0.16
335 0.22
274 0.13
Mass Balance
Final Bead initial Bead
Phenanthrene Wash
Desorbed
Biotransformed
% Recovered
DPM/g
846.71
11890.72
Total DPM
956.78
13436.51
6673.37
hotphen (ug)
0.0062
0.0870
0.0432
cold phen (ug) % Recovered
1.39
19.49
9.68
6.32
0.01 .
7
100
50
32
0
89

Non-Porous particles - Loaded on 3/11 stdev ug phen not I hot desorbed (ug) cold desorbed (ug) Total Cold ug desorbed (ug)
0.00
0.00000
0.00
0.02
0.02
0.00054
0.00043
0.12
0.10
0.07
0.11
0.04
0.14
0.22
0.44
0.43
0.00081
0.00249
0.OO177
0.00323
0.00470
0.00401
0.00418
0.00207
0.18
0.56
0.40
0.72
1.05
0.90 -
0.94
. 1.4
2.1
3.1
4.0
0.03
0.04
0.03
0.00217
0.00177
0.46
0.49
0.40
5.0
5.4
5.9
6.3
0
0.1
0.2
0.4
1.0

1 9 S J
P15
No Cells
Time (hrs)

Teflon - Porous Particles
Phenanthrene Loaded on 3/23
Innoculated with 19SJ - Surfactant Producer
Sample Time Exposed (min)
Beginning
1
2
3
4
5
6
7
8
9
170
180
O
Total Time (hrs) 1 (DPM/ml)
0.0
2.8
5.8
180
360
545
695
8.8
14.8
23.9
35.5
755
685
48.1
59.5
790 72.7
2 (DPM/ml)
0.0
59.4
84.8
64.4
80.2
61.2
71.3 •
63.4
62.3
53.5
0.0
66.4
73.9
67.9
73.0
88.4
80.7
77.4
63.8
63.5
3 (DPM/ml)
0.0
644
77.5
68.9
74.5
87.0
75.5
68.0
66.6
68.2
Average (DPM/ml)
0.0
63.3
78.7
67.1
75.9
78.9
75.8
69.6
64.2
61.7
Mass Balance
Final Bead
Initial Bead
Phenanthrene Wash
Desorbed
Biotransformed
% Recovered
DPM/g
575.22
3189.09
1000.00
Total DPM
650.00
3603.67
1000.00
hot phen (ug)
0.0042
0.0233
0.0065
cold phen (ug) % Recovered
0.55
3.04
0.84
0.54
0.92
18
100
28
18
30
93

Teflon r Porous Particles
Phenanthrene Loaded on 3/23
Innoculated with 19SJ - Surfactant Producer
Stand. Dev. (overall remov stdev ug phen not label Hot Desorbed (ug)
0.00
0.00
0.000000
0.10
0.00
0.000410
0.15
. 0.00
0.000509
0.07
0.00
0.000434
0.11.
0.43*
0.00
0.01
0.000491
0.000511
0.13
0.20
0.06
0.21
0.00
0.01
0.00
0.01
0.000491
0.000450
0.000416
0.000399
Cold Desorbed (ug) Total Cold ug desorbed (ug)
0.000
0.000
0.053
0.066
0.053
0.120
0.057
0.064
0.177
0.241
0.067
0.307
0.064
0.059
0.371
0.430
0.054
0.052
0.484
0.537
OO

Porous particles - Loaded on 3/23
Innoculated with P15 - Non Surfactant Producer
Sample Time Exposed (min)
Beginning
1
2
3
4
5
6
7
8
9
;
O
Total Time (hrs) 1 (DPM/ml)
0
170 3
180 6
180
360
9
15.
545
695
24
36
755 48
685
790
60
73
64
63
72
61
68
61
2 (DPM/ml)
0 .
60 .
81
55
0
78
89
63
61
69
73
76
76
61
3 (DPM/ml)
0
88
76
68
68
61
77
85
59
65
Average (DPM/ml)
0
72
85
62
' 63
76
70
70
68
59
Mass Balance
Final Bead
Initial Bead
Phehanthrene Wash
Desorbed
Biotransformed
% Recovered
DPM/g
575.22
3189.00
1000.00
Total DPM
650.00 hot phen (ug)
0.0042
3603.57
1000.00
0.0233
0.0065
cold phen (ug) % Recovered
0.55
3.04
:
0.84
0.53
0.93
18
100
28
17
31
94

Porous particles - Loaded on 3/23
Innoculated with P15 - Non Surfactant Producer
Stand. Dev. (overall remov stdev ug phen not label Hot Desorbed
0.00
0.27
0.00
0.01
0.000000
0.000464
0.12
0.00
0.000551
0.07
0.00
0.000402
0.06
0.00
0.000407
0.28
0.01
0.000494
0.22
0.01
0.000452
0.14-
0.22
0.10
0.00
0.01
0.00
0.000456
0.000442
0.000382
Cold Desorbed
0.000
0.061
0.072
0.052
0.053
0.064
0.059
0.060
0.058
0.050
Total Cold ug desorbed
0.000
0.061
0.132
0.185
0.238
0.303
0.362
0.421
0.479
0.529

Teflon -Porous Particles
Phenanthrene Loaded on 3/5
Sample Time Exposed (min)
Beginning
1
2
3
4
5
6
7
8
9
10
11
12
515
550
1020
780
1110
1000
0
210
60
75
120
120
240
Total Time (hrs) 1 (DPIWmI)
0.0
3.5
4.5
5.8
7.8
9.8
13.8
22.3
31.5
48.5
61.5
80.0
96.7
2 (DPIWmI)
0
72
53
64
103
83
106
154
145
198
145 •
169
86
. o
79
59
108
186
112
152
203
217
142
167
128
3 (DPIWmI)
185
160
233
135
143
135
0
88
53
94
153
131
130
Average (DPIWmI) Stand. Dev. (overa
0
80
55
89
O C
129
180
163
216
141
160
116
0.00
0.07
0.03
0.19
0.35
0.20
0.20
0.21
0.16
0.15
0.04
0.12
0.22
Mass Balance
Final Bead
Initial Bead
Phenanthrene Wash
Desorbed
Biotransformed
% Recovered
DPM/g
580.59
6017.70
. 3845.63
Total DPM
656.07
6800.00
3845.63
hotphen (ug)
0.0042
0.0440
0.0249
cold phen (ug) % Recovered
0.95
9.87 '
5.58
1.32
0.01
10
100
57
13
0
80

Teflon -Porous Particles
Phenanthrene Loaded on 3/5 stdev ug phen not I hot desorbed (ug) cold desorbed (ug) Total Cold ug desorbed (ug)
0.00
0.00000
0.00
0.00
0.01
0.00
0.00051
0.00036
0.07
0.05
0.07
0.11
0.02
0.03
0.00057
0.00096
0.07
0.12
0.19
0.31
0.02
0.02
0.02
0.02
0.00070
0.00084
0.00117
0.00105
0.09
0.11 ■
0.15
0.14
. 0.40
0.51
0.66
0.79
0.97
0.01
0.00
0.01
0.02
0.00140
0.00091
0.00103
’ 0.00075
0.18
0.12
0.13
0.10"
1.09
1.22
1.32

93
APPENDIX C - EFFECTIVENESS DATA AND GRAPHS

19SJ field # average std dev
9
10
7
8
5
6
2
3
4
Dilution
1 -3 cells
Field of View area (cm2)
1.00E-08
94
22
12
47
47
23
34
20
7
22
28
26
13
P15 field #
9
10
7
8
5
6
3
4
Dilution cells
1 -3
2
Iilllli
28
53
58
30
79
64
8
21
72
50
46
23
1764
Area of filter (cm2)
0.5

Bacteria
19SJ
P15
Bacteria
19SJ
19SJ
P15
P15
Cells/DPM
2.99E+07
6.56E+06
Column Type DPM
Porous
Non-Porous
51.59
50.56
Porous
Non-Porous
62.72
63.55
Cells on Column C02/Cell
1.54E+09 1.08E-10
1.51E+09 2.03E-09
4.12E+08 5.59E-10
4.17E+08 8.19E-09
U l

96
APPENDIX D MODEL DATA AND GRAPHS

Isotherm
5ml of 1.11 ES DPM/ul into 1L
1L had 1.0 mg/I cold phenanthrene
P4
P5
NI
N2
NS
N4
NS
Sample
P1
P2
P3
Spike (ml) DPM (TO) "Total DPM Hot (ug)
1 90 9000 5.83E-02
0.75
0.5
"
50
40.54
5000 3.24E-02
• 4054 2.62E-02
0.25
0
11.18
0
90
1118
0
9000
7.24E-03
0.00E+00
5.83E-02 I
0.75
0.5
0.25
60
40
20
6000
4000
2000
3.88E-02
2.59E-02
1.29E-02
0 0 0 0.00E+00
Cold (mg) Initial Aq
0.41
0.23
0.19
0.05
0.00
0.41
0.27
0.18
0.09
0.00
DPM (Tl)
42.11
23.58
17.51
4.95
0
20
10
5
2
0
Total DPM
. .
4211
2358
Hot (ug)
2.73E-02
1.53E-02
1751
495
1.13E-02
3.20E-03
0
2000
1000
500
200
0
0.00E+00
1.29E-02
6.47E-03
3.24E-03
1.29E-03
0.00E+d0
3 l

Cold (mg) Final Aqueous Sorbed (mg)
0.19
concentration sorbed Aqueous Co
2.2E-01 .
. 4.4E-01 . 9.6E-03
0.11 -
0.08 ■*
0.02
1.2E-01
1.1E-01
2.8E-02.
2.4E-01
2.1E-01
5.7E-02
5.4E-03
4.0E-03
1.1E-03
0.00
0.0E+00
0.09
0.05
0.02
0.01
3.2E-01
2.3E-01
1.6E-01
8.2E-02
0.0E+00
6.4E-01
0.0E+00
4.6E-03
4.6E-01 . 2.3E-03
3.2E-01
1.6E-01
1.1E-03
4.6E-04
0.00
0.0E+00 O.OE+00 0.0E+00 c
VO
OO

0.65 i
c 0.5
Time (hrs)

Diffusion
Stock
5ml of 1.11E3 DPIWuI into 1L
1L had 1.0 mg/I cold phehanthrene
Spike Jar I with 1 ml and stir
T3
T4
T5
T6
T7
T8
Sample
TO
Tl
T2
Time (min) Total Time (min) I jar (DPM) / 200 ul Total DPM
0 0 146.79
2 2 134.07
4 ■* 6 138.18
36 42 140
20 62 139.3
100 162 119.81
55
125
840
217
342
1182
115.31
107.93
88.36
Hot ug Phen
13211.1
12066.3
12436.2
12600
12537
10782.9
10377.9 -
9713.7
7952.4
8.55E-02
7.81 E-02
8.05E-02
8.16E-02
. 8.11 E-02
6.98E-02
6.72E-02
6.29E-02
5.15E-02
Cold (ug)
6.03E-01
5.51 E-01
5.68E-01
5.75E-01
5.73E-01
4.92E-01
4.74E-01
4.44E-01
3.63E-01

Diffusion Coefficient Calculations
Diameter of glass tube (mm) Area of diffusion (mmA2) Mass flux (mg/min*mmA2) DC
1 0.785 1.72E-04 8.87E-03
D (cmA2/s)
6.47E-05

MONTANA STATE - BOZEMAN