Embed Size (px)
Citation preview
ENHANCING IN SITU PAH BIODEGRADATION
The Effects of Amendments on Bench-Scale Bioremediation
Systems
by
Avery Gottfried
A thesis submitted in partial fulfilment of the requirements for the degree of
Master of Engineering,
Department of Civil and Environmental Engineering.
The University of Auckland, 2009.
ii
ABSTRACT
Current research in the field of bioremediation is uncovering a growing number of
microorganisms with the metabolic potential to degrade PAHs in soil and water. In situ
bioremediation is based on encouraging the growth of microorganisms, either indigenous or
introduced, to improve the degradation of contaminants without excavating or transporting
the soil. The majority of PAHs sorb strongly to soil organic matter posing a complex barrier
to biodegradation. Biosurfactants can increase soil-sorbed PAHs desorption, solubilisation,
and dissolution into the aqueous phase, which increases the bioavailability of PAHs for
microbial metabolism. In this study, biosurfactants, carbon sources, metabolic pathway
inducers, and oxygen were tested as stimulators of microorganism degradation.
Phenanthrene served as a model PAH and Pseudomonas Putida ATCC 17484 was used as the
naphthalene and phenanthrene degrading microorganism for the liquid solutions, soil
slurries and column systems used in this investigation. Bench-scale trials demonstrated that
the addition of rhamnolipid biosurfactant increases the apparent aqueous solubility of
phenanthrene, and overall degradation by at least 20% when combined with salicylate and
glucose. In soil slurries containing salicylate, the effects of biosurfactant additions were
negligible as there was greater than 90% removal, regardless of the biosurfactant
concentration. An in situ enhancement strategy for phenanthrene degradation could focus
on providing additional carbon substrates to induce metabolic pathway catabolic enzyme
production, if degradation pathway intermediates are known. The results of experiments
performed in this study provide further evidence that future studies should focus on
enhancing the metabolic processes responsible for successful in situ bioremediation.
iii
ACKNOWLEDGEMENTS
I would like to thank the Commonwealth Scholarship and Fellowship Plan which awarded
me the funding to come to New Zealand and pursue my research interests in this unique
part of the World. While here I was lucky enough to be part of a diverse Environmental
Engineering research group headed by Dr. Naresh Singhal, who also provided supervision for
this work. Special thanks to Abel Francis, our laboratory technician, for providing training,
maintaining equipment, and helping with experimental setup; Dr. Simon Swift in the
Molecular Medicine and Pathology Department for guidance, laboratory training, and
laboratory facilities to develop the biological aspects of this research; and Roy Elliot for
sharing his microbiology expertise in designing experiments and assisting with many
laboratory techniques over the research period.
iv
TABLE OF CONTENTS
ABSTRACT ........................................................................................................................ II
ACKNOWLEDGEMENTS .................................................................................................... III
LIST OF FIGURES .............................................................................................................. VI
LIST OF TABLES .............................................................................................................. VIII
ABBREVIATIONS ........................................................................................................... IX
CHAPTER 1 INTRODUCTION AND THESIS OBJECTIVES ........................................................ 1
1.1 INTRODUCTION .................................................................................................. 1
1.2 THESIS OBJECTIVES ............................................................................................. 3
1.3 ORGANIZATION OF THE THESIS ........................................................................... 4
CHAPTER 2 LITERATURE REVIEW ....................................................................................... 5
2.1 CONTAMINATED SITES IN NEW ZEALAND ............................................................ 5
2.2 BIOREMEDIATION ............................................................................................... 6
2.3 POLYCYCLIC AROMATIC HYDROCARBONS ........................................................... 8
2.4 IN SITU BIOREMEDIATION ................................................................................. 11
2.5 BACTERIAL DEGRADATION ................................................................................ 13
2.5.1 PHENANTHRENE METABOLISM ......................................................................... 13
2.6 LIMITING FACTORS AND STRATEGIES FOR BIOREMEDIATION ............................ 18
2.6.1 BIOAVAILABILITY OF PAH CONTAMINANTS ....................................................... 20
2.6.2 BIOSTIMULATION AND BIOAUGMENTATION .................................................... 21
2.6.3 SURFACTANTS AND BIOSURFACTANTS .............................................................. 22
2.6.4 SORPTION AND DESORPTION ............................................................................ 27
2.6.5 IONIC STRENGTH AND PH EFFECTS ON BIOSURFACTANTS ............................... 33
2.6.6 BIOSURFACTANT MICROBUBBLE DISPERSIONS ................................................. 34
2.6.7 METABOLIC PATHWAY INDUCERS ..................................................................... 37
2.7 DETERMINING TRANSPORT PARAMATERS ........................................................ 41
CHAPTER 3 MATERIALS AND METHODS .......................................................................... 45
3.1 LABORATORY SUPPLIES AND MICROOGANIM STORAGE .................................... 45
3.1.1 CHEMICALS ......................................................................................................... 45
3.1.2 BIOSURFACTANT ................................................................................................ 45
3.1.3 MICROORGANISMS ............................................................................................ 45
3.1.4 MEDIA AND NUTRIENT SUPPLY ......................................................................... 46
3.2 CELL CULTURING ............................................................................................... 48
3.2.1 AGAR PLATES ..................................................................................................... 48
3.2.2 INOCULANT PREPARATION AND HARVESTING .................................................. 48
3.2.3 PLATE COUNTS ................................................................................................... 49
3.2.4 OPTICAL DENSITY ............................................................................................... 50
v
3.3 SOIL METHODS ................................................................................................. 51
3.3.1 SOIL PROPERTIES ................................................................................................ 51
3.3.2 SOIL CONTAMINATION ...................................................................................... 52
3.3.3 CONTAMINANT EXTRACTION ............................................................................ 53
3.4 EXPERIMENTAL PROCEDURES ........................................................................... 55
3.4.1 OBJECTIVE 1: LIQUID MEDIUM TESTS ................................................................ 55
3.4.1.1 Inoculant culture growth ............................................................................ 55
3.4.1.2 Phenanthrene dissolution ........................................................................... 56
3.4.1.3 Degradation trials ....................................................................................... 56
3.4.2 OBJECTIVE 2: SOIL SLURRY TESTS ...................................................................... 58
3.4.2.1 Phenanthrene Desorption in the Presence of Surfactants .......................... 58
3.4.2.2 Degradation Trials ...................................................................................... 59
3.4.3 OBJECTIVE 3: COLUMN TESTS ............................................................................ 61
3.4.3.1 Experimental Apparatus ............................................................................. 61
3.4.3.2 Pressure Measurement ............................................................................... 62
3.4.3.3 Micro-foam Generation and Stability ......................................................... 62
3.4.3.4 Column Packing and Unpacking ................................................................. 63
3.4.3.5 Experimental Operating Conditions............................................................ 64
3.5 ANALITICAL METHODS ...................................................................................... 66
3.5.1 PAH DETECTION ................................................................................................. 66
3.5.2 BIOSURFACTANT DETECTION ............................................................................ 67
3.5.3 CHLORIDE ANION ............................................................................................... 67
CHAPTER 4 RESULTS AND DISCUSSION ............................................................................ 68
4.1 OBJECTIVE 1: LIQUID CULTURES ........................................................................ 68
4.1.1 BACTERIA GROWTH ON VARIOUS SUBSTRATES ................................................ 68
4.1.2 EFFECT OF INOCULANT ACCLIMATIZATION AND PRE-TREATMENT .................. 70
4.1.3 BIOSURFACTANT SOLUBILITY ENHANCEMENT .................................................. 73
4.1.4 PHENANTHRENE DEGRADATION ....................................................................... 75
4.2 OBJECTIVE 2: SOIL SLURRIES .............................................................................. 78
4.2.1 PHENANTHRENE DISTRIBUTION IN THE PRESENCE OF BIOSURFACTANTS ....... 78
4.2.2 SOIL DEGRADATION ........................................................................................... 83
4.3 OBJECTIVE 3: COLUMN TESTS ............................................................................ 91
4.3.1 TRACER AND BIOSURFACTANT BREAKTHROUGH CURVE FITTING .................... 91
4.3.1.1 Tracer Breakthrough Curves ....................................................................... 91
4.3.1.2 Biosurfactant breakthrough ....................................................................... 92
4.3.2 PRESSURE DROP ASSOCIATED WITH MICROBUBBLE DISPERSION PUMPING ... 97
4.3.3 BIODEGRADING TRIALS IN CONTINUOUS FLOW SYSTEMS .............................. 100
CHAPTER 5 CONCLUSIONS AND RECOMENDATIONS ...................................................... 106
5.1 RECOMMENDATION FOR FUTURE WORK ........................................................ 109
CHAPTER 6 WORKS CITED ............................................................................................. 112
vi
LIST OF FIGURES
Figure 2.1 Factors that influence biodegradation systems in bioremediation. Adapted
from Singh and Ward (2004) ............................................................................................ 7
Figure 2.2 The diagrammatic structures of 16 U.S EPA priory pollutant PAH compounds.
Modified from Rogers et al. (2002) .................................................................................. 9
Figure 2.3 Bay, K and L regions of PAHs involved in the formation of metabolically active
epoxides. Adapted from Chauhan et al. (2008) ............................................................. 14
Figure 2.4 Illustration of common steps in the upper pathway for aerobic metabolism of
phenanthrene (Samanta, Chakraborti et al. 1999; Chauhan, Fazlurrahman et al.
2008; Jun 2008) .............................................................................................................. 16
Figure 2.5 Illustration of common steps in aerobic metabolism of naphthalene and one of
the lower pathways for aerobic metabolism of phenanthrene (Samanta,
Chakraborti et al. 1999; Chauhan, Fazlurrahman et al. 2008; Jun 2008) ....................... 17
Figure 2.6 Schematic diagram of physical changes that occur due to surfactant addition
above the CMC. Adapted from (Mulligan, Yong et al. 2001) ......................................... 24
Figure 2.7 Examples of typical glycolipid biosurfactants produced by Pseudomonas
aeruginosa ...................................................................................................................... 25
Figure 2.8 Schematic diagram of abiotic processes in a soil-aqueous-surfactant system
containing a non-ionic surfactant, phenanthrene and soil organic matter. Adapted
from Edwards et al. (1994). ............................................................................................ 30
Figure 2.9 Plasmid-encoded naphthalene (upper pathway) and salicylate (lower pathway)
degradation genes of NAH7 catabolic plasmid for Pseudomonas sp. Genes nahA-D
encode the upper pathway operon which encodes enzymes for the degradation of
naphthalene to salicylate and genes nahG-M encode the lower pathway operon,
where salicylate is further degraded to pyruvate and acetylaldehyde.The product
from nahR (a trans-acting positive control regulator) is the positive regulator for
both operons and is induced by salicylate. The location of each respective operon
promoter is shown and locations of genes encoding the naphthalene dioxygenase
complex are indicated. ................................................................................................... 40
Figure 3.1 Schematic diagram for serial dilutions to determine cfu/mL of solution .............. 50
Figure 3.2 Particle size distribution.......................................................................................... 52
Figure 3.3 Soil column setup for uplflow pumping experiments ............................................ 61
Figure 4.1 Partial growth curves for P.Putida until early stationary phase in four growth
medias of glucose (2g/L); naphthalene (0.5g/L); salicylate (0.5g/L); and naphthalene
(0.5g/L) + biosurfactant (1g/L) ....................................................................................... 69
Figure 4.2 Naphthalene degradation and cell growth in liquid cultures containing
different bacteria inoculant seeds which were pre-grown in seven different
solutions (s1 BHB+naphthalene+glucose grown for 1 week; s2 BHB+glucose; s3
BHB+naphthalene+glucose; s4 BHB+salicylic acid+glucose; s5 LB; s6
LB+naphthalene; s7 LB+salicylic acid; s2 - s7 grown overnight approximately 20
hours growth) ................................................................................................................. 71
Figure 4.3 Phenanthrene solubility enhancement as a function of biosurfactant
concentration. The equation refers to the fit of data above the CMC and � � �� ��� ................................................................................................................................... 73
vii
Figure 4.4 Phenanthrene degradation in liquid cultures containing BHB and/or
biosurfactant (1000mg/L), salicylate (100mg/L), and glucose (100mg/L). Data
presented is the average of triplicate measurements taken at 22 and 46 hours after
inoculation. ..................................................................................................................... 75
Figure 4.5 Phenanthrene desorption from soil in the presence of biosurfactant.
Desorption partitioning coefficient Kd calculated from the linear regression
trendline for each series of data. ................................................................................... 78
Figure 4.6 Phenanthrene desorption from contaminated soil (50, 100, 250, 500 mg/kg)
into aqueous solution in the presence of biosurfactant over a 48 hour period. ........... 80
Figure 4.7 Total phenanthrene concentration in solution and suspended/dissolved
organic matter in soil slurries containing BHB and/or biosurfactant (0.25, 1, 5 g/L),
salicylate (100mg/L), and glucose (100mg/L) over a 10 day period .............................. 84
Figure 4.8 Total remaining phenanthrene in soil slurries after 10 days of bioremediation,
results presented as phenanthrene remaining in mg/kg of dry soil. ............................. 86
Figure 4.9 Live cell counts (cfu/mL) taken from soil slurry solution over 10 days. Results
presented are averages from duplicate or triplicate plate counts. ............................... 88
Figure 4.10 Observed breakthrough curves and fitted breakthrough curve models using
CXTFIT inverse parameter estimation for (a) chloride with v = 1.54cm/h (b) chloride
with v = 77.87cm/h and (c) biosurfactant with v = 1.54 cm/h ....................................... 93
Figure 4.11 Microbubble dispersion breakthrough curve with a conservative tracer in the
liquid fraction. ................................................................................................................ 96
Figure 4.12 Pressure distribution across the length of the soil column during biosurfactant
(1 g/L) solution pumping. Data presented corresponds to depth in the column with
the highest pressure at the inlet, and the lowest pressure at -17cm assuming the
outlet is the datum. ........................................................................................................ 98
Figure 4.13 Pressure distribution across the length of the soil column during biosurfactant
microfoam (1 g/L) pumping. Data presented corresponds to depth in the column
with the highest pressure at the inlet, and the lowest pressure at -17cm assuming
the outlet is the datum. .................................................................................................. 99
Figure 4.14 Pressure distribution across the length of the soil column during biosurfactant
microfoam ( 5g/L) pumping. Data presented corresponds to depth in the column
with the highest pressure at the inlet, and the lowest pressure at -17cm assuming
the outlet is the datum. .................................................................................................. 99
Figure 4.15 Trial 1 phenanthrene distribution in soil column after 10 days continuous
upflow at 0.2 mL/min , phenanthrene in column effluent over 10 days. Column 1
influent solution biosurfactant 1g/L + salicylate 100 mg/L; Column 2 influent
solution biosurfactant 1 g/L. Soil distribution assuming effluent (0cm) is the top of
the column and influent (-37 cm) is the bottom of the column. ................................. 100
Figure 4.16 Trial 2 phenanthrene distribution in soil column after 10 days continuous
upflow with BHB broth at 0.5mL/min; phenanthrene in column effluent over 10
days. Column 1 influent pulse solution biosurfactant microfoam 1 g/L + salicylate
100mg/L; Column 2 influent pulse solution biosurfactant 1 g/L. Soil distribution
assuming effluent (0cm) is the top of the column and influent (-37 cm) is the
bottom of the column. ................................................................................................. 101
viii
LIST OF TABLES
Table 2.1 Chemical characteristics of EPA PAH priority pollutants. Adapted from data
sources (Aitken, Stringfellow et al. 1998; World Health Organization 1998; Rogers,
Ong et al. 2002) .............................................................................................................. 10
Table 2.2 Advantages and disadvantages for selecting in situ bioremediation. Adapted
from The Interstate Technology & Regulatory Council (2005) ...................................... 12
Table 3.1 BHB marine salts broth approximate formula per litre of prepared media ............ 47
Table 3.2 Soil properties .......................................................................................................... 51
Table 3.3 Soil slurry media constituents .................................................................................. 60
Table 3.4 Column trial experimental design ............................................................................ 65
Table 4.1 Rate of phenanthrene degradation in liquid cultures expressed as mg of
phenanthrene degraded / hour ..................................................................................... 77
Table 4.2 Calculated phenanthrene soil partitioning coefficient Kd, and phenanthrene
partitioning onto soil sorbed surfactant coefficient Ks .................................................. 81
Table 4.3 Total percentage removal of phenanthrene due to soil flushing and
biodegradation in soil column tests after 10 days continuous flow. ........................... 103
ix
ABBREVIATIONS
PAHs polycyclic aromatic hydrocarbon/s SOM soil organic matter IRZs in situ reactive zones CMC critical micelle concentration HOCs hydrophobic organic compounds BHB Bushnell-Hass marine salts Broth LB Lysogeny Broth PV pore volume HPLC high performance liquid chromatography TOC total organic carbon OD600 optical density at 600nm
1
CHAPTER 1 INTRODUCTION AND THESIS OBJECTIVES
1.1 INTRODUCTION
Interest in the bioremediation processes of soil contaminated with Polycyclic Aromatic
Hydrocarbons (PAH) has been growing over the past two decades. This interest stems from
the identification of microbes with the ability to degrade toxic xenobiotic compounds in soil
and water. Although bioremediation primarily relies on the catalytic roles of soil
microorganisms to break down contaminants into innocuous by-products, the
understanding of the microbial communities’ operation and behaviour in complex soil
systems remains limited. Bioremediation occurs in the natural environment where most
organisms are uncharacterized, and each site is unique in terms of its soil, microbes, and
contamination. These variable site characteristics create numerous challenges to
understanding the interactions taking place which actually contribute to the desired
decrease in harmful contamination. The biodegradation process is mostly treated as a
unknown ‘black box’ process where soil amendments are made and desired contaminant
removal is achieved without fully understanding the microbial processes that were
enhanced to bring about contaminant mineralization (Singh and Ward 2004). Recent
research has focused on the biochemical and physiological aspects of the bioremediation
process with an emphasis on determining key parameters that make the process more
efficient and reliable (Samanta, Singh et al. 2002). This includes improving the
bioavailability of the contaminants and understanding the metabolic pathways and the
enzymatic reactions that are used in contaminant breakdown, with the goal of identifying
the rate-limiting steps. Ultimately obtaining this knowledge will enable scientist to engineer
better bioremediation processes. Biotechnology and advanced molecular techniques are
2
now providing researchers with the tools to advance understanding in these areas. There is
tremendous potential for engineered bioremediation to make microorganisms more
effective and efficient in removing contaminants, accelerating the remediation process.
It is difficult to replicate the complexity of real PAH contaminated sites in constructed lab
scale systems. However, there is certainly a need to determine optimal treatment
conditions and degradative capabilities in a single bacteria strain in order to unravel the
underlying interactions. Simple systems, where most variables can be controlled and
monitored, show insight into the microbial response to specific variables that are altered,
and will offer advances for understanding the underlying complexities of in situ
bioremediation (Pieper and Reineke 2000). This research was carried out to determine why
specific soil amendments, including the addition of biosurfactants, have been shown to
increase (or decrease) overall contaminant degradation. Results from these trials provide
further evidence to the processes responsible for successful in situ bioremediation
treatment and contribute to the understanding and capabilities of these processes in PAH
field contaminated sites.
3
1.2 THESIS OBJECTIVES
The focus of this research was to enhance the degradation of phenanthrene in soil by the
microbe Pseudomonas putida ATCC 17484 (P.putida) with amendments that included co-
substrates, electron acceptors, and metabolic pathway inducers to the system.
Amendments were designed to increase contaminant bioavailability, enhance microbial
degrading activity, and increase the amount of contaminant degradation. To fully
understand the interactions between biodegradation, amendments, and soil, each process
was isolated and independently evaluated in order to focus on specific characteristics that
enhanced in situ biodegradation. The experiments were designed in stages to achieve each
specific objective:
OBJECTIVE 1: To study the effect of co-substrates, metabolic pathway inducers, and
inoculant pre-treatment on the degradation of phenanthrene and naphthalene by P.putida
in liquid cultures.
Task A: studied growth characteristics of P.putida in various substrates
Task B: determined changes in phenanthrene solubility in the presence of rhamnolipid
biosurfactant
Task C: determined contaminant degradation rates in liquid cultures with added
biosurfactant, glucose, and salicylate
OBJECTIVE 2: To evaluate the effects and monitor the changes in the degradation of
phenanthrene by P.putida due to various soil amendments to contaminated soil slurry.
Task A: determined soil characteristics and the contaminant desorption characteristics
in the presence of rhamnolipid biosurfactant
Task B: determined contaminant degradation rates in soil slurries with amendments
4
OBJECTIVE 3: To design a continuous-flow bench-scale micro-environment to model in situ
remediation and observe the degradation of phenanthrene by P.putida in a saturated
contaminated soil. This system was used to study the effects and flow of microfoam through
the system, and analyze the substrate transport parameters in a soil column.
Task A: determined transport parameters in soil columns using non reactive tracers
Task B: determined microfoam characteristics and evaluate pressure build up in the
soil during the injection of microfoam for various flow rates and microfoam
qualities
Task C: evaluated the efficiency of microfoam and various soil amendments on the
overall removal of phenanthrene from contaminated soil
1.3 ORGANIZATION OF THE THESIS
This thesis consists of five chapters:
Chapter One gives a brief introduction to bioremediation and gives an overview of research
objectives and thesis setup. Chapter Two defines the nature of the problem and discusses
the most important areas of investigation. It also provides an overview of the topic and
highlights key knowledge gained in similar areas, which are relevant to the work presented
in this thesis. Chapter Three presents all of the methods that were used in this study.
Experimental designs are presented in detail for each of the experiments that were
necessary to complete the three main objectives. Chapter Four summarizes the results
obtained and offers an interpretation and discussion of them. Chapter Five forms the
conclusion, and provides recommendations for future research.
5
CHAPTER 2 LITERATURE REVIEW
2.1 CONTAMINATED SITES IN NEW ZEALAND
As modern economies move to enhance environmental protection, more effective testing
methods, increased legislation, and stricter monitoring guidelines have been developed. The
result of this has been the location and acknowledgement of numerous sites of soil
contamination (Doyle, Muckian et al. 2008). Many strategies have been proposed, including
physical, chemical, and biological methods to restore contaminated soil sites. Polycyclic
Aromatic Hydrocarbons (PAHs) are present in many contaminated soil sites, stemming
primarily from the use of oil and petroleum-derivatives; including potentially hazardous,
carcinogenic, and toxic hydrocarbons. Sites with high PAH concentration can act as sources
because contaminants mobilize and leach offsite posing extra risks to groundwater, soil
fertility, and living systems (Singh and Ward 2004). PAHs significantly accumulate in surface
and subsurface soils and an increased concentration can result in a highly toxic
environmental site, necessitating cleanup. Depending on the site location and the level of
groundwater contamination such contamination can pose serious human health risks. In
New Zealand, surface and subsurface soil contamination has been linked to historical land
uses which include agricultural and horticultural activities, gas works, landfills, petrol
station, dry cleaners, sheep dips, and timber treatment sites. The New Zealand Ministry of
the Environment (2007) reports 1,238 contaminated sites resulting from industrial inputs
deemed as ‘Hazardous Activities and Industries List’ (HAIL). However, this number could be
over 50,000 when sites not currently on the HAIL list are considered. These include a large
number of urban sites contaminated unknowingly by fill materials—and such sites are
slowly being discovered (Auckland City Council 2007).
6
2.2 BIOREMEDIATION
The term bioremediation can be applied to any biological process that uses enzymes in
microorganisms, fungi, or green plants to break down undesired contaminants and
contribute to the restoration of the environment to its original condition. Biodegradation is
defined as the breakdown of organic compounds to less complex metabolites, or the
complete breakdown through mineralization into the inorganic minerals H20, C02, or CH4.
Understanding the bioremediation process requires the examination and interpretation of
both biochemical and physiological aspects. Knowledge of these processes will allow key
parameters to be manipulated and bioremediation optimized (Singh and Ward 2004).
Generally, the environmental conditions in which microbial processes are occurring must be
altered to encourage the desired outcomes. With bioremediation, a variety of factors
(Figure 2.1) can influence microbial growth and bioactivity which ultimately increase the
microbial, physiological and biochemical activity and enhance the biodegradation of
contaminants. Even if these factors are optimized, PAH degradation can remain slow unless
the mass transfer rates and the bioavailability of the PAHs for microbial metabolism are
increased (Cerniglia 1993). To effectively accelerate the removal of PAHs from contaminated
soils a greater understanding of not only the physical processes involved but the
physiology, biochemistry, molecular genetics and microbial ecology of the degrading strains
of microorganisms is required (Chauhan, Fazlurrahman et al. 2008). Consideration of the
biotic factors such as the production of toxic or dead-end metabolites, metabolic repression,
presence of preferred substrates and lack of co-metabolites or inducer substrates are also
important in optimizing the overall efficiency of the bioremediation process (Chauhan,
Fazlurrahman et al. 2008).
Figure 2.1 Factors that influence
7
Factors that influence biodegradation systems in bioremediation. Adapted from
Singh and Ward (2004)
in bioremediation. Adapted from
8
2.3 POLYCYCLIC AROMATIC HYDROCARBONS
PAHs are identified as a class of chemicals with two or more fused aromatic rings containing
solely carbon and hydrogen (Figure 2.2). They are formed as products in the incomplete
combustion or pyrolysis of fossil fuels and organic matter (Harvey 1991). PAHs are natural
components of fossil fuels such as petroleum or coal, but leakage and accidental spill of
these products can cause accumulation in the environment (Doyle, Muckian et al. 2008).
The World Health Organization (1998) reports that the PAHs contained in various
environmental wastes, including coal combustion residues, motor vehicle exhaust, used
motor lubricating oil, and tobacco smoke “are mainly responsible for their carcinogenic
potential.” The largest and most important PAH contaminated sites occur near large
industrial sources where individual PAH levels of up to 1g/kg of soil have been found. These
originate from concentrated emissions of combustion residues and storage and handling of
coal, coke, fly ash or liquid petroleum reserves. Soils present around crude-oil refineries,
fuel storage depots, petrol stations, gas works, landfills, incinerators and wood preservation
facilities are the most common areas where significant PAH accumulation has been
detected. PAH sources such as automobile exhaust have been shown to cause
contamination next to busy roadways in the range of 2-5mg/kg of soil, whereas background
levels of PAHs are 5-100µg/kg soil deriving from natural sources of atmospheric deposition
such as forest fires and volcanic eruptions (World Health Organization 1998).
The environmental fate of PAH contaminants is governed by the number of aromatic rings
present, and the nature of the linkage between the rings (Doyle, Muckian et al. 2008). PAHs
with low molecular weight are typically defined as those containing up to three aromatic
rings and tend to be more soluble and volatile. High molecular weight PAHs are generally
9
defined as those containing four or more aromatic rings and tend to be less soluble, less
volatile and have a tendency to accumulate in the environment as they sorb strongly to soil
organic matter (SOM) (World Health Organization 1998).
Figure 2.2 The diagrammatic structures of 16 U.S EPA priory pollutant PAH compounds.
Modified from Rogers et al. (2002)
The Environmental Protection Agency (EPA) Priority Pollutant List of 126 pollutants includes
16 PAH compounds (Figure 2.2), the important details of their chemical properties include
thier aqueous solubility (Table 2.1). The contaminants included on this list are regulated,
and the EPA has developed analytical testing methods for accurate detection in the
environment (U.S Environmental Protection Agency 2008). This frequently referenced list
originated from the 1972 Clean Water Act and the 1977 Clean Water Act Amendment, and
10
the only modifications to this list were the removal of 3 compounds from the list in 1981
showing the long recognition of PAHs as toxic compounds (Hendricks 2006; U.S
Environmental Protection Agency 2008).
Table 2.1 Chemical characteristics of EPA PAH priority pollutants. Adapted from data
sources (Aitken, Stringfellow et al. 1998; World Health Organization 1998; Rogers, Ong et
al. 2002)
U.S EPA PAH priority
compound
Chemical
Formula
Aqueous
solubility
Csat
(mg/L)
Molecular
weight
(g/mol)
n-Octanol
water
partition
coefficient
(log Kow)
Organic
carbon
partition
coefficient
(log Koc)
Naphthalene C10H8 31 128.17 3.37 3.11
Acenaphthylene C12H8 3.4 152.19 4.0 3.4
Acenaphthene C12H10 3.8 154.21 3.94 3.65
Fluorene C13H10 1.9 166.22 4.18 3.86
Phenanthrene C14H10 1.1 178.23 4.57 4.15
Anthracene C14H10 0.045 178.23 4.54 4.15
Fluoranthene C16H10 0.26 202.25 5.22 4.58
Pyrene C16H10 0.132 202.25 5.18 4.58
Chrysene C18H12 0.002 228.29 5.65 5.3
Benz[a]anthracene C18H12 0.011 228.29 5.91 6.14
Benzo[b]fluoranthene C20H12 0.0015 252.31 5.8 5.74
Benzo[k]fluoranthene C20H12 0.0008 252.31 6.0 5.74
Benzo[a]pyrene C20H12 0.0038 252.31 6.04 6.74
Dibenz[a,h]anthracene C22H14 0.0006 278.35 6.75 6.52
Indeno[1,2,3-c,d]pyrene C22H12 0.062 276.33 7.66 6.2
Benzo[g,h,i]perylene C22H12 0.00026 276.33 6.5 6.2
11
2.4 IN SITU BIOREMEDIATION
Conventional remediation techniques, including ex situ treatment, and removing the soil for
treatment and disposal at another, safer location are difficult to apply to many PAH
contaminated sites. The slow mobilization of contaminants, and leaching into ground water
tables have lead to situations which contaminated ground water supplies and contaminant
plumes migrate below city infrastructure and developed areas, make ex situ treatment and
soil removal impossible. Sites of PAH contaminations are varied, including: developed urban
and industrial areas, parks and natural environments, and spread over large geographical
areas. Many contaminated sites were discovered after industrial activities ceased, as PAHs
are essentially recalcitrant and persistent (Gómez, Alcántara et al. In Press). The increasing
number and widespread distribution of contaminated sites has encouraged the
development of in situ technologies such as heat based injection, air sparging, soil vapour
extraction, soil washing or flushing, bioremediation, enhanced bioremediation, microbial
filters, and others (Warith, Fernandes et al. 1999). In situ technologies offer an advantage
over ex situ techniques as they are designed to be implemented in place, without
transporting or disturbing the soil (Error! Reference source not found.). Furthermore, in situ
technologies have now been proven to be much cheaper alternatives to traditional
methods, saving time and resulting in a less invasive remediation design that can
complement the natural attenuation process (Suthersan and Payne 2005). In situ
bioremediation is a technology based on stimulating the growth of indigenous or introduced
microorganisms to improve the degradation of contaminants without excavating or
transporting the soil to other locations for treatment. In situ bioremediation can include
augmenting the natural microbial population with the addition of specific microbes that can
metabolize and grow on specific compounds.
12
Table 2.2 Advantages and disadvantages for selecting in situ bioremediation. Adapted
from The Interstate Technology & Regulatory Council (2005)
In situ reactive zones (IRZs) are designed to manipulate oxidation and reduction reactions,
and other biogeochemical processes to affect the mobility, transport, and fate of inorganic
and organic contaminants in the subsurface. Successful designs of IRZs need to account for
all the variables presented in Figure 2.1 to optimize the reactions required for the
biodegradation of target contaminants. Each contaminated site has different characteristics
and naturally variable conditions that need to be taken into account in an engineered
remediation solution to create an effective microbial reactive zone.
ADVANTAGES DISADVANTAGES It is usually less expensive than other
remediation options.
Complete contaminant destruction is not
achieved in some cases, leaving the risk of a
residual toxic intermediate. It is almost always faster than baseline
pump-and-treat.
Some contaminants are resistant to
biodegradation.
It may be possible to completely destroy
the contaminant, leaving only harmless
metabolic by-products (no ex situ waste
created).
Some contaminants (or their co-
contaminants) are toxic to the
microorganisms and prevent complete
metabolism and site restoration.
It can be designed with minimal
disturbance to the site and facility
operations, and also can be incorporated
into ongoing site development activities.
Biodegradation of organic species can
sometimes cause mobilization of naturally
occurring toxic inorganic species such as
manganese or arsenic.
It is not limited to a fixed area, typical of
chemical flushing or heating technologies,
because it can move with the contaminant
plume.
Alteration of groundwater redox conditions or
substrate supply can reduce the down
gradient effectiveness of natural
bioattenuation processes. It can treat both dissolved and sorbed
contaminants.
Uncontrolled proliferation of the
microorganism may clog the subsurface.
The processes usually use reagents that are
easily accepted by regulators and the
public.
The hydrogeology of the site may not be
conducive to enhancing the microbial
population.
13
2.5 BACTERIAL DEGRADATION
The biochemical pathways and enzymes responsible for the initial transformation stages are
usually specific to particular contaminants, but bacteria have the capacity to evolve new
catabolic pathways when exposed for long periods of time to specific contaminants. Due to
the complex mixture of low and high molecular weight PAHs present in some contaminated
sites, there tends to be incomplete bioremediation of higher weight PAHs even if aggressive
approaches are used to enhance the process (Singh and Ward 2004). Due to the
recalcitrance of high molecular weight it is difficult for any single microbial organism to use
them as sole energy and carbon growth, but they are more likely to be oxidized in a series of
steps by consortia of microbes (Perry 1979). Cerniglia (1993) stated that “a better
understanding of the metabolism, enzyme mechanisms, and genetics of polycyclic aromatic
hydrocarbon-degrading microorganisms is critical for the optimization of these
bioremediation processes” and this fact holds true 15 years later and remains an effective
motivation for research today.
2.5.1 PHENANTHRENE METABOLISM
Phenanthrene is a three-ring PAH with low aqueous solubility and is commonly used in
laboratory research as an ideal PAH contaminant for the study of various aspects of
microbial metabolism and physiology (Woo, Lee et al. 2004; Labana, Manisha et al. 2007).
Phenanthrene is the smallest PAH which has both a low aqueous solubility and contains an
“L-region” “bay-region,” and a “K-region” which is common in many higher ringed PAHs
(Figure 2.3). Bay-regions are locations where there is a terminal ring on one side of the bay
region (the terminal ring is also termed the A region), K-regions are areas of high electron
density in all resonance structures and L-regions are sites between two ring fusion points
14
(Yan 1985). The understanding of phenanthrene metabolism can be correlated to studies
on higher-ringed PAHs such as benzo[a]pyrene, benzo[a]anthrancene and chrysene. The
metabolism of bay-region and K-region is believed to be important in understanding the
degradation of both higher and lower ringed PAH compounds and phenanthrene serves as
the example (Xiang, Xian-min et al. 2006). The Bay-region dihydrodiol epoxides are believed
to be the main carcinogenic species and in benzo[a]pyrene these metabolites are cytotoxic,
cause DNA strand breaks and are also mutagenic (World Health Organization 1998). The
Bay, K, and L PAH regions (Figure 2.3) are involved in the formation of metabolically active
and highly reactive epoxides. PAH epoxides arise via metabolism of the parent PAH and
occur whenever oxygen atoms are added across double bonds, a process that can be
catalyzed by the action of enzymes or by an uncatalyzed oxidation process (Josephy and
Mannervik 2006).
There are several bacteria strains that are capable of degrading phenanthrene aerobically
and the more commonly identified strains are Pseudomonas sp, Rhodococcus sp.,
Mycobacterium flavescens, Mycobacterium sp., Flavobacterium sp., and Beijerinckia sp.,
which are capable of using phenanthrene as the sole carbon source and growth substrate
(Cerniglia 1993; Samanta, Chakraborti et al. 1999; Chauhan, Fazlurrahman et al. 2008 ).
Figure 2.3 Bay, K and L regions of PAHs involved in the formation of
metabolically active epoxides. Adapted from Chauhan et al. (2008)
15
Phenanthrene has two potential degradation pathways that are established based on the
bacteria present. These pathways take advantage of the biologically and chemically active
bay and K-region epoxides, which can be formed metabolically by enzymes present in
phenanthrene degrading bacteria (Samanta, Chakraborti et al. 1999). Both pathways share
the same common upper route (Figure 2.4) and are initiated by the double hydroxylation of
a phenanthrene ring by a dioxygenase enzyme to yield cis-3,4-dihydroxy-3,4
dihydrophenanthrene, which then undergoes enzymatic dehydrogenation to 3,4-
dihydroxyphenanthrene. From here the diol is cleaved and metabolized, and 1-hydroxy-2-
naphthoic acid remains and is degraded by one of the two routes termed the lower
pathways (Prabhu and Phale 2003).
The lower pathways consist of two separate routes for degradation depending on the
enzymes that are present in the organisms. In route one (Figure 2.5) 1-hydroxy-2-naphthoic
acid is degraded via the naphthalene pathway to salicylate and then further metabolized via
the formation of catechol or gentisic acid, while route two uses the phthalate pathway. Both
naphthalene and phenanthrene share a common upper metabolic pathway and organisms
that degrade phenanthrene via route one have the ability to degrade naphthalene,
salicylate and catechol (Kiyohara, Torigoe et al. 1994; Samanta, Chakraborti et al. 1999;
Prabhu and Phale 2003). Both oxygen and water are consumed during metabolism and H+
ions are produced, which can affect the pH of the environment if enough degradation
activity is occurring. Understanding the metabolic processes that are involved in the
degradation of phenanthrene are important when determining how additions such as
oxygen or salicylate will influence microbial activity, or determining why changes in pH or a
build up of intermediate metabolites is occurring.
16
Fig
ure
2.4
Ill
ust
rati
on
of
com
mo
n s
tep
s in
th
e u
pp
er
pa
thw
ay
fo
r a
ero
bic
me
tab
oli
sm o
f p
he
na
nth
ren
e
(Sa
ma
nta
, C
ha
kra
bo
rti
et
al.
19
99
; C
ha
uh
an
, F
azl
urr
ah
ma
n e
t a
l. 2
00
8;
Jun
20
08
)
17
Fig
ure
2.5
Ill
ust
rati
on
of
com
mo
n s
tep
s in
ae
rob
ic m
eta
bo
lism
of
na
ph
tha
len
e a
nd
on
e o
f th
e l
ow
er
pa
thw
ay
s fo
r a
ero
bic
me
tab
oli
sm o
f p
he
na
nth
ren
e (
Sa
ma
nta
, C
ha
kra
bo
rti
et
al.
19
99
; C
ha
uh
an
, F
azl
urr
ah
ma
n e
t a
l. 2
00
8;
Jun
20
08
)
18
2.6 LIMITING FACTORS AND STRATEGIES FOR BIOREMEDIATION
It is commonly assumed that bacteria can only access PAHs in the aqueous phase, and the
relatively low bioavailability of PAHs in this phase limits their consumption by the microbial
biomass. As this is a limiting factor, increasing the bioavailability of PAHs in the aqueous
phase by increasing mass transfer rates of PAHs from the soil into solution is essential in
furthering research in this area (Wick, Colangelo et al. 2001). Recent review papers
(Cerniglia 1993; Samanta, Chakraborti et al. 1999; Chauhan, Fazlurrahman et al. 2008 )
conclude that virtually all of the PAHs of concern are biodegradable, and that organisms
capable of degrading PAHs are ubiquitous in the natural environment. PAHs also have
strong hydrophobicity and associate with nonaqueous phases in soil and natural organic
matter where they are not bioavailable, meaning they are in a location where they are not
able to be adsorbed and metabolised by microorganisms. Because of these tendencies,
bioremediation of PAHs in the environment is usually incomplete, even when soil
amendments attempt to enhance the system (Singh and Ward 2004).
Looking specifically at the microbial kinetics, there are several methods possible to enhance
the rate of biodegradation of a PAH. The simplest way to determine what factors influence
this rate in a simple solution is to look at Monod growth kinetics (Equation 2.1). The Monod
growth kinetics take the same form as Michaelis-Menten kinetics with the assumption that
a certain number of new cells grow per unit mass of chemical transformed (Hemond and
Fechner-Levy 2000). This equation is the most commonly used in modelling growth kinetics
associated with PAH degradation.
19
Monod Growth Kinetics
� � �� · �� ��
Equation 2.1
Where:
• u : specific growth rate [T-1
]
• umax : maximum specific growth rate [T-1
]
• C : concentration of dissolved chemical [M/L3]
• Ks : half-saturation constant [M/L3]
The biodegradation rate depends on µmax, Ks and substrate concentration C. Therefore,
increasing µmax, decreasing Ks, increasing microbial cell density, or increasing contaminant
concentration will be sufficient strategies to enhance the biodegradation process. At low
contaminant concentrations, the rate at which bacteria can degrade the substrate can also
depend on the specific affinity for the substrate (Johnsen, Wick et al. 2005). Specific affinity
refers to the ratio of the maximal rate of substrate uptake and the half saturation constant,
and high affinities lead towards efficient contaminant removal at low concentrations due to
steeper concentration gradients and higher transfer rates between the substrate and the
cell (Johnsen, Wick et al. 2005). Enhancements in microbial growth kinetics can only occur if
no chemicals other than the contaminants are limiting the microbial community.
Specifically, oxygen and mineral nutrients must be in excess. Understanding the basic
parameters that influence the degradation rate of contaminants, highlights the importance
of increasing the bioavailability of contaminants for effective in situ bioremediation.
20
2.6.1 BIOAVAILABILITY OF PAH CONTAMINANTS
Significant PAH accumulation in the environment occurs in subsurface organic soil matter
due to the hydrophobicity and low aqueous solubility of PAHs. The majority of PAHs are
difficult to remove because they sorb strongly to soil organic matter. Long term
contamination of soil, which is commonly referred to as aging or weathering, is the result of
chemical oxidation reactions and slow chemical diffusion into small pores, both of which
decrease PAH bioavailability over time (Singh and Ward 2004). The degradation process
therefore involves the transfer of contaminants from the soil to the enzymes in the
microorganisms which begin the mineralization of the contaminant (Noordman 1999).
Contaminant characteristics such as molecular structure, solubility, and the octanal/water
partitioning coefficient (Kow) are relevant for substance sorption in or onto soil and can be
used to indicate the availability of the contaminant to bacteria in the soil.
It is the influence of the dissolution and desorption process of PAHs in soil that are often
cited as the rate limiting step in the degradation process. The slow transport of PAHs from
the soil matrix to bacteria is the slowest process and limits degradation (Mulder, Breure et
al. 2001; Prabhu and Phale 2003; Johnsen, Wick et al. 2005; Doyle, Muckian et al. 2008).
However, contaminant bioavailability can be species-specific, with different bacteria strains
able to access different contaminant pools in the soil-water system. Understanding the
interactions amongst bacteria will also provide further opportunities for enhanced
degradation of bioavailable contaminants (Dean, Jin et al. 2001). For instance, some
organisms have the ability to affect sorption kinetics on their own, through the production
of surface active agents termed biosurfactants. These can increase the apparent solubility of
PAHs in the aqueous phase and concomitantly increase the concentration gradient, allowing
21
improved mass transfer of contaminants from the soil to the aqueous phase (Pignatello and
Donald 1999). Dean et al. (2001) demonstrated that some of the sorbed phase
phenanthrene was bioavailable to certain Pseudomonas bacteria, and called into question
the frequently used assumption that only bulk aqueous phase contaminant is available for
degradation. Woo et al. (2001) included a term for sorbed phase biodegradation of
phenanthrene when modelling the process to account for the rapid degradation that
occurred in soil slurry tests. Kwok and Loh (2003) also proposed that bacteria which have
attached themselves directly to soil particles can utilize the nutrients sorbed at that
location. Several techniques have been developed to effectively enhance the bioavailability
of contaminants.
2.6.2 BIOSTIMULATION AND BIOAUGMENTATION
Enhanced biodegradation is usually accomplished through biostimulation and
bioaugmentation. Biostimulation refers to the modification of the environment via the
addition of oxygen, nutrients, other electron donors or acceptors, and surfactants. These
additions stimulate the existing bacteria and increase the number or rate at which the
organisms are degrading a contaminant. Biostimulation relies on making the natural
environment more favourable to the metabolic capacities of the indigenous microbial
populations, whereas bioaugmentation describes the addition of adapted microorganisms
to the environment that are capable of degrading contaminants that are present.
Depending on the characteristics of the contaminated site, either biostimulation or
bioaugmentation may be needed to achieve the desired outcomes. Ruberto (2006) found
that a combination of both techniques using fish meal for nutrient supply and surfactant
Brij700 with bioaugmentation using a psychrotolerant PAH degrading bacterial consortium
22
caused significant removal (46.6%) of phenanthrene whereas when each technique was
applied separately, insignificant reduction was observed (Ruberto, Vazquez et al. 2006). It is
commonly reported that either the availability of electron acceptors, or nutrient limitations,
are the cause of slow biodegradation processes at contaminated sites (Institute for Ecology
of Industrial Areas 1999). Laboratory studies often report high rates of biodegradation
compared to results actually achieved in the field with similar soil and bacteria types, which
can be due to the optimization of many variables such as temperature, mixing, nutrient
balances and nutrient delivery. These variable are sometimes impossible to replicate in the
field (Institute for Ecology of Industrial Areas 1999).
2.6.3 SURFACTANTS AND BIOSURFACTANTS
Surfactants are used to describe surface-active agents that lower the surface tension of a
liquid (Riser-Roberts 1998). Surfactants have both a hydrophilic group and a hydrophobic
group and can be described as either anionic or cationic depending on whether they release
an anion or a cation when dissociating in water. They are termed non-ionic if no net charge
is dissociated. Therefore an anionic surfactant has an anionic hydrophilic group at its head,
whereas a non-ionic surfactant has no net charge groups at its head. Anionic and non-ionic
surfactants tend to be the best solubilizers and are relatively non-toxic compared to cationic
surfactants (Oostrom, Dane et al. 2006). Jin (2007) ranked the toxicity of the studied
surfactants to bacterial activity in soil and determined the order of toxicity towards bacteria
as follows: non-ionic surfactants (Tween 80, Brij30, 10LE and Brij35) < anionic surfactants
(LAS) < cationic surfactants (TDTMA) (Jin, Jiang et al. 2007).
23
Suitable co-solvents or surfactants must be selected according to solution chemistry, proven
ability to solubilise PAH compounds, and compatibility with the remediation technique. In
addition, they must not be toxic or a threat to human health or the environment (Gómez,
Alcántara et al. In Press). The presence of surfactants in the bulk phase causes an increase in
the free energy of the system. In order to lower the free energy, the surfactant molecules or
monomers are concentrated at the surface and interface, and the surface tension is lowered
increasing the solubility of hydrophobic contaminants (Myers 1988). The surface tension will
decrease to a given value, known as the critical micelle concentration (CMC) beyond which
point it will remain constant (Figure 2.6). Once the concentration of surfactants is above the
CMC, the surfactants begin to aggregate to form micelles, vesicles, and lamellae. Surfactant
micelles increase the apparent aqueous solubility of hydrophobic particles by reducing the
interfacial tension between the oil phase and the aqueous phase. In contaminated systems
this results in PAHs partitioning within the hydrophobic micellar core of the micelles. This
creates higher apparent aqueous solubility as PAHs are dissolved both in aqueous solution
and inside surfactant micells which are present in the bulk aqueous phase (Error! Reference
source not found.) (Noordman, Ji et al. 1998; Cameotra and Bollag 2003; Makkar and
Rockne 2003).
Surfactants can also be produced by bacteria or yeasts from growth on various substrates
including sugars, oils, hydrocarbons and agricultural wastes. These are termed
biosurfactants (Lin 1996). In terms of surface activity, heat and pH stability, many
biosurfactants are comparable to synthetic surfactants (Lin 1996). Biosurfactants are
receiving increasing attention as they have lower toxicity and higher biodegradability
compared to their chemical counterparts (Rosenberg and Ron 1999). Specifically,
rhamnolipid biosurfactants produced by
extensively as they have excellent emulsifying power with a variety of hydrocarbon
vegetable oils (Wang, Fang et al. 2007)
glycolipidic surface-active molecules that are produced in mixtures of one or two rhamnoses
attached to β–hydroxyalkanoic acid
resulting in lengths of 8, 10, 12 and 14 carbons
Fang et al. 2007). The in situ
them potentially more cost effective while also using natural resources instead of chemical
inputs.
Figure 2.6 Schematic diagram of physical changes that occur due to surfactant addition
above the CMC. Adapted from
Surfactant
monomer
24
rhamnolipid biosurfactants produced by Pseudomonas aeruginosa have been studied
excellent emulsifying power with a variety of hydrocarbon
(Wang, Fang et al. 2007). Rhamnolipid (Figure 2.7) is the name given to the
active molecules that are produced in mixtures of one or two rhamnoses
hydroxyalkanoic acid. The length of the fatty acid chains can va
12 and 14 carbons (Soberón-Chávez, Lépine et al. 2005; Wang,
in situ production of biosurfactants at contaminated sites renders
ntially more cost effective while also using natural resources instead of chemical
Schematic diagram of physical changes that occur due to surfactant addition
above the CMC. Adapted from (Mulligan, Yong et al. 2001)
Surfactant
Micelle
have been studied
excellent emulsifying power with a variety of hydrocarbons and
the name given to the
active molecules that are produced in mixtures of one or two rhamnoses
an vary significantly,
Chávez, Lépine et al. 2005; Wang,
production of biosurfactants at contaminated sites renders
ntially more cost effective while also using natural resources instead of chemical
Schematic diagram of physical changes that occur due to surfactant addition
g et al. 2001)
25
Figure 2.7 Examples of typical glycolipid biosurfactants produced by Pseudomonas
aeruginosa
Research into the addition of surfactants and biosurfactants have produced mixed results,
from greatly enhanced rates of PAH degradation to the inhibition of PAH degradation
(Pieper and Reineke 2000; Makkar and Rockne 2003; Avramova, Sotirova et al. 2008). There
are several hypotheses explaining the mixed results. Beneficial results may be due to the
facilitation of bioremediation through increases in desorption, solubilisation, and dissolution
of PAHs from soil sorbed or solid phase contaminant into the aqueous phase, which results
in increased bioavailability of PAHs for microbial metabolism (Mulligan, Yong et al. 2001;
Makkar and Rockne 2003; Shin, Kim et al. 2004; Avramova, Sotirova et al. 2008). Negative
result led to an assortment of conclusions including:
• the preferential use of surfactants as a growth substrate by degrading
microorganisms;
• the toxicity of the applied surfactants preventing increased microbial growth;
26
• the toxicity of the PAHs resulting from the increased bioavailability that is caused by
the surfactant solubilization of PAHs;
• the reduction of PAH bioavailability due to the uptake into surfactant micelle which
then could not be available for bacteria;
• the sorption of surfactant into the soil blocking access to PAHs that could have been
further absorbed into the soil or causing PAH sorption into soil sorbed surfactants
(Garcia-Junco, Gomez-Lahoz et al. 2003; Shin, Kim et al. 2004; Avramova, Sotirova et
al. 2008).
An example of these mixed results was meaningfully demonstrated by Allen et al. (1999)
with the use of titron X-100 with Pseudomonas sp. strain 9816/11 and Sphingomonas
yanoikuyae B8/36. Triton X-100 increased the rate of oxidation of phenanthrene with strain
9816/11. Conversely, the surfactant inhibited the biotransformation of both naphthalene
and phenanthrene with strain B8/36 under the same conditions (Allen, Boyd et al. 1999).
These observations show an important knowledge gap in how surfactants truly alter the
biodegradation process and interact with bacteria. Considering that a non-ionic surfactant
could have contrasting effects on the ability to degrade PAHs by different bacteria, there is a
requirement for additional research relating to surfactants, including all stages of soil-water-
surfactant-bacteria interactions.
There is a recurring assumption that the remediation of PAHs in soil or soil-water systems
depends strongly on the desorption rates of the PAHs from the soil into the aqueous phase
(Jin, Jiang et al. 2007). It is assumed that once PAHs are in the bulk aqueous phase, it is
possible to use engineering treatment steps to enhance the remediation process and create
27
an effective bioremediation strategy. However, there are an increasing number of studies
that have demonstrated that bacteria can attach to soil particles and use the nutrients
sorbed to the soil surface (Dean, Jin et al. 2001; Wick, Colangelo et al. 2001). This could
explain why the addition of surfactants to some systems does not predictably enhance
contaminant biodegradation. As the natural role of biosurfactant is to increase the
bioavailability of contaminants by decreasing surface tension, there can be a reduction in
direct adhesion of bacteria to the desired contaminants of interest due to the decrease in
surface tension (Pieper and Reineke 2000).
The mixed effects of surfactant on biodegradation show the complex interactions between
the PAH, surfactant, microorganism, soil, and water in the environment. Due to variable that
are important in the bioremediation process all researchers have to provide caveats in the
conclusions section to isolate results to the unique system of bacteria, soil type,
contaminant, and test conditions that was studied.
2.6.4 SORPTION AND DESORPTION
The partitioning and transport processes (sorption, desorption, and dissolution) between
the soil and water phases of both contaminants and surfactants affect the overall
degradation of contaminants (Schlebaum, Schraa et al. 1999; Kraaij, Ciarelli et al. 2001;
Mulder, Breure et al. 2001; Zhou and Zhu 2005; Zhou and Zhu 2007; Wang and Keller 2008;
Zhu and Zhou 2008; Laha, Tansel et al. In Press). Soil organic matter and natural organic
matter is not homogeneous and PAHs strongly absorb to soot carbon, and more slowly
partition into humic matter (Jonsson, Persson et al. 2007). As PAHs adsorb onto the surface
28
of soil organic mater they slowly begin to penetrate further into cavities and diffuse into the
organic fraction over time. Landrum et al (1992) observed a continuous increase in the
partition coefficient of phenanthrene and pyrene into soil over a period of six months, after
in-lab contamination of the soil. The length of this process makes it impractical for the
determination of single sorption or desorption coefficients to model the process over the
long term. Schlebaum et al (1999) successfully modelled the sorption of hydrophobic
organic compounds (HOCs) from the soil matrix with a kinetic model using two separate
compartments. A Freundlich isotherm represented high affinity sites, and a linear sorption
isotherm and first order kinetics represented low affinity sites. Even if the amount of
organic matter is low, PAHs can still become trapped in pores and voids and these variables
will affect the efficiency and success of any remediation process.
It is not just the average aqueous concentration of the target contaminant that determines
its availability. The rate of mass transfer to microbial cells relative to the intrinsic substrate
utilization capacity of the microbial cells must also be considered because it determines the
bioavailability of the contaminant (Wick, Colangelo et al. 2001). As a result, limited
bioavailability occurs when the environment is unable to deliver the substrate at the rate
consumable by the microbial biomass. The biodegradation rate in the subsurface is often
reported as first-order even when total contaminant concentrations are high. Wick et al.
(2001) provided an explanation for these observations by considering that the mass
transport processes are slow for hydrophobic organic soil pollutants which cause the same
degradation rates to be obtained even when the substrate concentration has changed.
29
When enhancing bioremediation, it is important to consider the effect surfactants can have
on the desorption and dissolution of contaminants from the soil. When surfactants exceed
their CMC , it is well established that there is an increase in desorption of PAHs from the soil
(Noordman 1999). However, when surfactants adsorb to soil they increase the overall
organic content of the soil and provide additional sorption capacity. This can enhance
sorption of hydrophobic compounds onto soil sorbed surfactants. This influences the
amount of PAHs present in the aqueous phase, accessible for biodegradation (Edwards,
Adeel et al. 1994). Conversely, the micelles present in the bulk aqueous phase can greatly
enhance the solubilisation of the PAHs, causing increased desorption from soil. The
efficiency of surfactants at enhancing PAH desorption shows a strong dependence on the
soil composition, surfactant structure and concentration, and PAH properties as concluded
by Zhou and Zhu (2005).
The process of surfactant adsorption to the soil has been described as a three stage process
by Torrens et al.(1998). The first stage is controlled by electrostatic attraction between
surfactants and the soil surface. As the surfactant concentration increases, there is a
tendency for self-association of surfactant ions due to the electrostatic and hydrophobic
forces. This is analogous to the micelle formation but it leads to the formation of
hemimicelles which is the second stage. This stage is more rapid than the first stage, and
results in neutralization of the particle surface, causing the sorption process to slow. After
the second stage, micelle formation begins, which results in the reversal of the surface
charge. This greatly reduces surfactant sorption due to charge repulsion. The third stage is a
plateau region, and additional surfactant will be present in solution (Torrens, Herman et al.
1998). Figure 2.8 encapsulates the interactions that are believed to occur between the soil-
30
water-surfactant systems. The stages of sorption also result in the wetting of soil grains
which enables the washing out of the hydrophobic substances from the soil pores. For most
hydrophobic contaminants they can be assumed to be un-wetted with water, and
surfactants increase the wettablity of the hydrophobic surfaces through attachment and
sorption to the soil surface (Pastewski, Hallmann et al. 2006).
Figure 2.8 Schematic diagram of abiotic processes in a soil-aqueous-surfactant system
containing a non-ionic surfactant, phenanthrene and soil organic matter. Adapted from
Edwards et al. (1994).
31
The distribution of contaminants between the soil fraction and the aqueous phase is
generally described by the partitioning coefficient Kd. Kd refers to the ratio of the
concentration of contaminant in the soil fraction to the concentration in the aqueous phase.
In the simplest form, the equation Cs = KdCw where Cs is the concentration of the
contaminant sorbed by the soil fraction and Cw in the concentration of the contaminant in
the aqueous phase respectively. Kmc is another commonly used partitioning coefficient,
which defines the amount distributed between the aqueous phase and the surfactant
micelle phase. The total amount is commonly referred to as the apparent aqueous solubility
as there is more contaminant in the aqueous phase, although it is located inside the
surfactant micelle. There are many different theoretical models that are used to determine
the partitioning coefficient Kd, taking into account surfactant adsorption modelled by the
Langmuir isotherm, Kow, and the fraction of organic carbon, and PAH sorption (Huang and
Cha 2001). The following equation appears to be the most commonly used to describe PAH
partitioning within a soil-water-surfactant system (Zhu, Chen et al. 2003; Zhou and Zhu
2007; Wang and Keller 2008; Zhu and Zhou 2008).
��� � �� ����
� ��� ��� ��� ��� Equation 2.2
Where:
• ��� : ratio of sorbed PAH to mobile PAH in the aqueous solution (L/kg);
• Kd : PAH sorption coefficient with the soil in the absence of surfactant (L/kg);
• Qs : quantity of surfactant sorbed to the soil;
• Ks : solute distribution coefficient with the soil-sorbed surfactant (L/kg);
• Xmm and Xmc : surfactant monomer and micellar concentration in water (g/L);
• Kmm and Kmx : PAH partitioning coefficients with the surfactant monomer and
micellar phases (L/kg).
32
The overall factors that effect ��� are the partitioning of PAH to soil due to the presence of
sorbed surfactants (terms in the numerator in equation 2.2), and decreased PAH
partitioning to soil by the enhanced aqueous solubility of the PAH in the presence of
surfactant monomers and micelles (denominator in equation 2.2).
Depending on the quantity of surfactant added to the system, the majority may be in the
soil sorbed-phase (Laha, Tansel et al. In Press). The result of this is increased partitioning of
PAHs onto soil until the solubilisation by micellar phase surfactant is at a high enough
concentration to compete with the increased PAH sorption on the surfactant sorbed soil
(Laha, Tansel et al. In Press). However, the cation exchange capacity of the soil can
significantly affect the sorption of surfactants (Ks), and well as the ionic strength or pH of the
system.
33
2.6.5 IONIC STRENGTH AND PH EFFECTS ON BIOSURFACTANTS
Anionic surfactants are strongly affected by the presence of electrolytes in solution as they
can influence the solubilization capacity, cause precipitation of the surfactant from the
aqueous phase, and increase the adsorption to subsurface porous media (Stellner and
Scamehorn 1989; Jafvert and Heath 1991; Guiyun, Brusseau et al. 1998). Torrens et al (1998)
saw 67% rhamnolipid sorption to soil at low K+ concentrations (10mM) but this increased to
98% in the presence of 20mM K+ in solution. The ionic strength and presence of cations in
solution has been shown to further enhance the solubility of hydrophobic organic
contaminants in rhamnolipid solution. Guivun (1998) reported that both Na+ and Mg
2+
enhanced the solubility of PAHs as there was an increase in the interior volume of
rhamnolipid micelles in the presence of cations, and Mg2+
, being a divalent cation, had a
stronger affection on reducing the repulsion forces between anionic head groups (Guiyun,
Brusseau et al. 1998). However, Ca2+
had little affect on solubility, due to competing effects
between rhamnolipid precipitation and enhanced contaminant solubility. The presence of
cations also reduced the interfacial tension between rhamnolipid solutions and hexadecane
from 2.2 to 0.89 dyn cm-1
(Guiyun, Brusseau et al. 1998). A decrease in pH from 7 to 6 was
seen to have the same qualitative effect to the interfacial tension as the increase in Na+
concentration. The carboxyl group in the rhamnolipid head group has a pKa of 5.6, causing it
to become more protonated as the pH decreases, thereby reducing repulsion between the
head groups. A similar effect was seen by Shin (2004) as the apparent solubility of
phenanthrene was 3.8 times greater at a pH of 5.5 when compared with a pH of 7 in the
presence of 240 mg/L rhamnolipid. In another study, more rhamnolipid molecules were lost
by sorption to sand particles at a pH 4 than at both higher and lower pH values, explaining
why a dramatic decrease in apparent aqueous solubility of phenanthrene was seen at that
34
pH (Shin, Kim et al. 2008). These findings are particularly important in soil remediation as
subsurface matrix solutions contain electrolytes such as Ca2+
, Mg2+
, Na+, K
+, and Al
3+ which
can have affect the surfactant performance (Guiyun, Brusseau et al. 1998).
2.6.6 BIOSURFACTANT MICROBUBBLE DISPERSIONS
Microbubble dispersions, also known as colloidal gas aphrons (CGA) or microfoam, are a
series of micro-bubbles that were first investigated by Sebba (1971). Microfoam displays
colloidal properties because of its micron-sized bubbles (typically 0.7-100µm) and its unique
bubble structure which consists of multiple layers of surfactant monomers surrounding the
surface of the microbubble. In contrast, standard foam consists of just one layer of
surfactant monomers (Jauregi and Varley 1999; Wan, Veerapaneni et al. 2001; Larmignat,
Vanderpool et al. 2008). Microbubble dispersions can flow like water, and can be pumped
easily without collapse (Jauregi and Varley 1999). Surfactant microfoam technology is a
relatively new approach for enhancing in situ bioremediation, showing promising
advantages over air sparging or surfactant solution application. Foam can flow in a plug flow
manner, delivering oxygen or air uniformly (Wang and Mulligan 2004). Microbubble
dispersion flow is also capable of overcoming heterogeneity in porous media, enhancing
bacterial transport, and delivering oxygen and nutrients to the subsurface (Wan,
Veerapaneni et al. 2001; Choi, Park et al. 2008; Park, Choi et al. In Press). Foam and
microfoam technology is designed either to remove contaminants and/or act
simultaneously as an augmentation for existing technologies such as pump-and-treat and
bioremediation. It is designed to enhance the process and improve removal efficiencies and
cost effectiveness (Wang and Mulligan 2004). Foam stability reflects the ability of the
35
suspension to resist bubble collapse, and is typically measured as the time required for half
of the foam to collapse. The half-life for microfoam can range from minutes to days,
depending on the generation method, surfactant, and additions such as nutrients, bacteria
or soil particles.
Microbubble dispersions can facilitate mobilisation and transport of contaminants trapped
in porous media, and can take less pore volumes to achieve high contaminant removal when
compared to surfactant solutions (Wang and Mulligan 2004). Couto et al (In Press) saw 96%
removal in sandy soils using microfoam in soil flushing to remove diesel oil, versus 88%
removal with regular foam and 35% removal with surfactant solution. Park et al. (In Press)
saw a 2.2-fold increase in phenanthrene degradation when 3 pore volumes of microbubbles
were injected instead of 1 pore volume. There are several studies demonstrating the ability
of conventional foam to enhanced remediation of PAH contaminated soils, and the
beneficial transport mechanisms of foam (Chowdiah, Misra et al. 1998; Rothmel, Peters et
al. 1998). Microfoam appears to have an added advantage over conventional foam as
dispersion can be generated that contain less gas (60-70% versus up to 99% with
conventional foam) in smaller sized bubbles, making them easier to pump through the
subsurface (Roy, Kommalapati et al. 1995; Jauregi and Varley 1999).
Microbubble injection systems have been shown to be efficient oxygen delivery systems in
pilot scale tests that used microbubble generators that were encapsulated in pressurized
chambers that contained oxygen and biosurfactant solution. Leigh et al. (1997)
demonstrated that microbubbles generated using this method persistent in the subsurface
for longer periods of time and have different migration characteristics compared to air
36
bubbles injected in by typical air sparging. Using this generation method and a mixture of
anionic and non ionic surfactants, Wan et al. (2001) was able to generate microbubbles that
were still present in solution up to six weeks after generation.
Subsurface foam and microfoam flow is typically accompanied by a pressure drop due to the
flow characteristics. Higher-viscosity foams flow forward and fill up larger channels and pore
spaces. When the pressure drop builds up in the channel, the foam flows into less accessible
spill areas. This pressure dependent “clogging” process means that channelling, or poor
sweep, should not occur with the microbubble scouring as compared with surfactant
flushing (Riser-Roberts 1998). However, applications could be limited by the pressure drop
required to pump microbubbles into soil with low permeability (Riser-Roberts 1998; Choi,
Park et al. 2008; Park, Choi et al. In Press).
37
2.6.7 METABOLIC PATHWAY INDUCERS
Another biostimulation strategy that can enhance the intrinsic biodegradation rate of target
compounds is the addition of one or more known pathway intermediate catabolite. These
are usually produced by the bacteria when mineralizing a contaminant and they stimulate
growth, enzymatic expression, and ultimately increase the biodegradation of PAHs
(Ogunseitan and Olson 1993; Cho, Seung et al. 2006). This process is defined as co-
metabolism, where bacteria may co-utilize various substrates that compete with the
structurally similar primary substrate for the enzyme’s active sites (Mohan, Kisa et al. 2006).
The introduction of carbon sources that are metabolic pathway inducers into the soil can
enhance in situ bioremediation by stimulating the growth of specific indigenous micrograms
that are capable of degrading organic contaminants. Unfortunately, additional carbon
sources can also be used preferentially by soil bacteria causing diauxic growth which can
have a negative effect on the degradation process (Lee, Park et al. 2003).
A number of studies have used salicylate as a pathway inducer to enhance initial rates of
naphthalene and phenanthrene removal (Chen and Aitken 1999; Lee, Park et al. 2003; Woo,
Jeon et al. 2004; Lee, Lee et al. 2005; Powell, Singleton et al. 2008; Basu, Das et al. In Press).
Salicylate is the third intermediate formed in the degradation of naphthalene (Figure 2.5)
and it is also an intermediate formed in the degradation of phenanthrene for bacteria that
degrade phenanthrene via the naphthalene pathway. Most information about PAH
metabolism has been derived from the study of naphthalene catabolic plasmids in
Pseudomonas putida G7 (Yen and Serdar 1988). In the plasmid there are genes which
encode the pathway for naphthalene degradation (Figure 2.9) In the first operon, there are
genes which encode the pathway for conversion of naphthalene to salicylate, and in the
38
second operon are the genes which code for the conversion of salicylate via catechol meta-
cleavage to acetaldehyde and pyruvate (Eaton and Chapman 1992; Platt, Shingler et al.
1995). The regulatory mechanism for both operons is encoded in a third operon which acts
as the regulatory protein and positively regulates the two operons by the increased
presence of salicylate (Schell and Wender 1986; Atlas and Philip 2005). The principle
mechanism for the aerobic bacterial metabolism of naphthalene is via the oxidative action
of the naphthalene dioxygenase enzyme; that introduces molecular oxygen into the
aromatic ring. The naphthalene (upper pathway) and salicylate (lower pathway) degradation
genes located in the NAH7 catabolic plasmid from Pseudomonas sp. are regulated by
salicylate induction to both operons (Figure 2.9). Chen and Aitken (1999) showed that
salicylate greatly enhanced removal of fluoranthene, pyrene, benz[a]anthracene, chrysene,
and benzo[a]pyrene, all of which are high molecular weight PAHs which the strain
Pseudomonas saccharophila P15 could not use as a sole carbon for growth. This showed
that high-molecular weight PAH metabolism by this organism is induced by salicylate. Lee et
al. (2005) saw phenanthrene degradation rates 3.5-fold higher with Burkholderia cepacia
PM07 compared to the rates achieved without salicylate addition in aqueous solutions. They
also saw a decrease in phenanthrene removal with the addition of glucose (Lee, Lee et al.
2005). Basu et al. (In Press) determined Pseudomonas Putida CSV86 preferentially utilized
aromatics over glucose and co-metabolized them with organic acids, indicating that
intermediate metabolites enhance the mineralization rate of PAHs more effectively than
additional carbon sources (Basu, Das et al. In Press).
In other studies Cho et al. (2006) saw up to 12 times increase in the degradation of target
chemicals per equivalent cell mass with the addition of various intermediate metabolites
39
into solution. In this experiment all phenanthrene was soluble due to the addition of 1%wt
Triton X-100. Woo et al. (2004) saw up to a 3-fold increase in phenanthrene degradation
using salicylate in soil water systems, however addition of triton X-100 saw inhibitory effects
towards total phenanthrene mineralization. Other substances such as 1-hudroxy-2-
naphthoate, catechol, and pyruvate have also shown their potential as effective pathway
inducers to enhance in situ bioremediation (Cho, Seung et al. 2006; Basu, Das et al. In Press).
Chemotaxis is another strategy that can be used to enhance the degradation of
contaminants in the environment. Chemotaxis is “a complex process [in] which bacterial
cells detect temporal changes in the concentrations of specific chemicals, respond
behaviourally to theses changes and then adapt to the new concentration of the chemical
stimuli” (Samanta, Singh et al. 2002). It is not clear if it is the metabolism of the substrate or
if it is the binding of the substrates to the chemoreceptors that is the crucial inducer of
chemotaxic behaviour. The NAH7 plasmid in Pseudomonas putida, which encodes the
enzymes for the degradation of naphthalene and salicylate, also encodes the chemotaxis
towards these compounds (Samanta, Singh et al. 2002). This chemotaxis in Pseudomonas
putida was found to be homologous to chemotaxis, flagellar and mobility genes from other
known E.coli bacteria. The ability to foster chemotaxis phenomenon via metabolic
influences could be important to enhance in situ bioremediation.
40
Fig
ure
2.9
Pla
smid
-en
cod
ed
na
ph
tha
len
e (
up
pe
r p
ath
wa
y)
an
d s
ali
cyla
te (
low
er
pa
thw
ay
) d
eg
rad
ati
on
ge
ne
s o
f N
AH
7 c
ata
bo
lic
pla
smid
for
Pse
ud
om
on
as
sp.
Ge
ne
s n
ah
A-D
en
cod
e t
he
up
pe
r p
ath
wa
y o
pe
ron
wh
ich
en
cod
es
en
zym
es
for
the
de
gra
da
tio
n o
f n
ap
hth
ale
ne
to
sali
cyla
te a
nd
ge
ne
s n
ah
G-M
en
cod
e t
he
lo
we
r p
ath
wa
y o
pe
ron
, w
he
re s
ali
cyla
te i
s fu
rth
er
de
gra
de
d t
o p
yru
va
te a
nd
ace
tyla
lde
hy
de
.Th
e p
rod
uct
fro
m n
ah
R (
a t
ran
s-a
ctin
g p
osi
tiv
e c
on
tro
l re
gu
lato
r) i
s th
e p
osi
tiv
e r
eg
ula
tor
for
bo
th o
pe
ron
s a
nd
is
ind
uce
d b
y s
ali
cyla
te.
Th
e l
oca
tio
n o
f e
ach
re
spe
ctiv
e o
pe
ron
pro
mo
ter
is s
ho
wn
an
d l
oca
tio
ns
of
ge
ne
s e
nco
din
g t
he
na
ph
tha
len
e
dio
xy
ge
na
se c
om
ple
x a
re i
nd
ica
ted
.
41
2.7 DETERMINING TRANSPORT PARAMATERS
Laboratory tracer experiments are useful for flow characteristics in soil. The fundamental
mass balance of the system uses:
Inputs + Production – Outputs – Losses = Accumulation Equation 2.3
All contaminant transport and biodegradations models use this fundamental principal to
derive equations when approximating parameters such as convection and dispersion. These
assumptions allow the derivation of the partial differential equation referred to as the
convection-dispersion (Equation 2.4). The CDE is a mathematical model used in
quantitatively simulating the transport of solutes in porous media. The CDE is derived by
assuming the change in chemical flux into and out of a control volume is controlled by an
advection component (which is controlled by the velocity of the chemical), and dispersion
(which can be through of as mimicking diffusion in the sense that the dispersive flux appears
to be driven by concentration gradients) (Toride, Leij et al. 1995). The CDE for one-
dimensional transport of reactive solutes subject to adsorption, first-order degradation, and
zero-order production, in homogenous soil is given as: (Toride, Leij et al. 1993):
� ���� � � ����
��� � ����� � μ� " #$% Equation 2.4
The initial boundary conditions used to solve this equation assume that there is a fixed
known concentration of solute added to the system. This is expressed as the following:
• &'&( $∞, +% � 0 (exit condition where the concentration = 0)
• C(x,t) = 0 for x = 0
• C(x,0) = Ci (where the concentration of influent tracer is constant)
• C(0,t) = -./0 0 / 12 3 24
2 5 24 (tracer on/off after time t)
42
Where:
• C : dissolved aqueous chemical concentration;
• x and t : dimensionless space and time variable respectively;
• 6 � 1 " 89 :;< (Where => is the bulk density; �� is a partitining coefficient; and n is
the porosity);
• D : dispersion coefficient;
• ? : pore water velocity;
• @ : first order decay rate constant;
• A : zero order production rate constant.
When decay appears in such a system it can be due to strong sorption, chemical and
biological activity, or other physicochemical interactions of the solutes in the porous media.
The inclusion of decay represents the change in dispersive flux out and assumes that decay
affects the mass inside the controlled volume. Another common form of equation 2.4 uses
a term called the Peclet number (P) which is a dimensionless number relating the amount of
advection to dispersion (P=v/D).
The transport of solutes in soil and groundwater systems includes a large number of
complicated physical, chemical, and microbiological processes (Toride, Leij et al. 1993).
Variations to the standard CDE presented in equation 2.3 have been added to account for
the simultaneous effect of sorption (including zero and first order), convective transport,
molecular diffusion, hydrodynamic dispersion, zero-order production, and first order decay
(Chen, Wang et al. 2006). Equilibrium transport processes refer to exchange reactions that
are perceived as instantaneous and are commonly described by equilibrium isotherms
including linear, Freundlich or Langmuir type. However, these equilibrium models appear to
fail in situations where chemical transport processes are not at equilibrium (Nielsen, Van
Genuchten et al. 1986). This has led to the development of non-equilibrium transport
43
models that incorporate first order reactions and various chemical, kinetic and diffusion
limited rate laws to describe the non equilibrium transport. A familiar chemical non-
equilibrium model (Equations 2.5 & 2.6) includes one-site and two-site sorption. This means
sorption onto one site can be considered to be instantaneous (equilibrium) while sorption
onto the second site will be rate limited by first order kinetics (non-equilibrium) (Toride, Leij
et al. 1993). The two-site model presented is also used for physical non-equilibrium and is
termed a two-region (dual-porosity) type formulation which contains two distinct liquid
regions, one being mobile (flowing) and the other being immobile and the rate constants
refer to the mass transfer between the two regions which is modelled as a first-order
process.
B� ����� � �
C ������� � � ����
� � � D$E� � E�% � �� " #�$% Equation 2.5
$� � B%� ����� � D$E� � E�% � �� " #�$% Equation 2.6
Where:
• subscripts 1 and 2 refer to equilibrium and non equilibrium sites respectively;
• β : partitioning coefficient of adsorption sites that equilibrates with instantaneous
and kinetic adsorption sites or mobile and immobile liquid phase;
• ω : dimensionless mass transfer coefficient (Toride, Leij et al. 1995).
The two-site equilibrium and non-equilibrium equation is incorporated into a software
package called CXTFIT developed by Torride et al. (1995), which permits one to fit a variety
of analytical solutions to the concentration distributions observed in laboratory and field
tracer studies as a function of time and/or distance. The transport of PAHs in the presence
of surfactants (Linear Alkylbenzene Sulfonate) has been successfully modelled using the
two-site model presented in equations 2.5 and 2.6 using this CXTFIT software (Chen, Wang
44
et al. 2006). Noordman et al. (1998) were also successful in predicting the removal of
phenanthrene from soil using a rhamnolipid biosurfactant using a similar two site
formulation which accounted for both micellar solubilisation and admicellar sorption due to
the presence of the biosurfactant.
45
CHAPTER 3 MATERIALS AND METHODS
3.1 LABORATORY SUPPLIES AND MICROOGANIM STORAGE
The important characteristics and handling procedures for the chemicals, biosurfactants,
microorganisms, and nutrient broth media purchased for this study are detailed below.
3.1.1 CHEMICALS
HPLC grade acetonitrile, hexane, acetone, and dichloromethane were purchased through
Biolab New Zealand and supplied by Mallinckrodt Baker. Naphthalene technical crystals
were from B.D.H., London, England and from Ajax Finechem, Auckland, New Zealand.
Salicylic acid crystals were from B.D.H., London, England. Phenanthrene crystals were >96%
purity HPLC grade and were supplied by Sigma Aldrich.
3.1.2 BIOSURFACTANT
Biosurfactant was purchased from Jeneil Biosurfactant Co., LLC Saukville, Wisconsin, USA.
The biosurfactant is a glycolipid produced by Pseudomonas aeruginosa with the trademark
name JBR 425. Biosurfactant stock solution contained a 25% solution of Rhamnolipids Rha-
C10-C10 (termed R1 or RLL) with the molecular formula C26H48O9 and Rha-Rha-C10-C10 (termed
R2 or RRLL) with the molecular formula C32H58O13. The rhamnolipids are an anionic
surfactant with a pKa=5.6.
3.1.3 MICROORGANISMS
The microorganisms used in this study were Pseudomonas putida ATCC 17484 (P.putida),
obtained from the American Type Culture Collection and purchased through Cryosite
Distribution in Australia. This isolate is chemoheterotropic, and is from biotype B which is
cited to degrade naphthalene. P.putida are a gram-negative rod-shaped flagellated
46
bacterium which stains a pink colour when a gram stain test is performed for identification
under a microscope. P.putida are aerobic bacteria with an optimum growth temperature
between 25-30°C in a neutral pH environment. They are easily isolated from environmental
samples, and are found in most aerobic soil environments which makes it a good
representative isolate of the soil microbial consortium. For shorter storage periods, nutrient
agar plates inoculated with bacteria cultures were kept at 4°C and were re-streaked onto
fresh agar every two weeks from a single colony. For long term storage a stock of frozen
isolates, consisting of 0.5 mL of fresh overnight culture (approximately 1 x 108 – 1 x 10
9
cfu/mL) added to an equal volume of sterile 50% glycerol in a 1 mL sterile plastic tube, was
constructed and stored at -80˚C.
3.1.4 MEDIA AND NUTRIENT SUPPLY
All experiments, unless otherwise noted, were carried out using DifcoTM
Bushnell-Hass Broth
(BHB) as the nutrient supply. BHB is designed for study of microbial utilization of
hydrocarbons. It contains no carbon source, but provides all the trace elements necessary
for bacterial growth. It provides the monopotassium and diammonium hydrogen phosphate
to buffer the growth media, in an initial pH of 7.0 at 25°C. BHB was mixed at 3.27g/L and
autoclaved for 15 minutes at 121°C according to manufacturer’s instructions in 0.5 or 1L
increments (Table 3.1). Nutrient agar plates and Pseudomonas isolation agar plates, along
with a gram stain set with stabilized gram iodine, were purchased from Fort Richard
Laboratories, Auckland, New Zealand. Nutrient agar plates were also made in house by
adding 1.5% DifcoTM
agar to Lysogeny broth (LB). LB was purchased from USB corporation,
Cleveland, USA, and was made in a 20g/L solution which contains 10g/L casein peptone,
5g/L yeast extract, and 5g/L of sodium chloride. BHB and Agar were purchased from Becton
47
Dickinson and Company, Sparks, USA. Other solutes that were used include glucose (Glucosa
1-hidrato) and sodium chloride reagent grade, both supplied by Panreac and Scharlav,
Spain.
Table 3.1 BHB marine salts broth approximate formula per litre of prepared media
Approximate Formula Chemical Formula Concentration (g/L)
Magnesium Sulfate MgSO4 0.2
Calcium Chloride CaCl2 0.02
Monopotassium Phosphate KH2PO4 1.0
Diammonium Hydrogen Phosphate (NH4)2HPO4 1.0
Potassium Nitrate KNO3 1.0
Ferric Chloride F2Cl3 0.05
48
3.2 CELL CULTURING
This section details the bacteria storage and inoculant growth methods. Bacteria were
stored on agar plates, and single colonies were used to grow inoculants used in
experiments. Serial dilution plate counts and optical density at 600nm (OD600) methods are
presented to quantify the bacteria used in this study.
3.2.1 AGAR PLATES
Initially, bacteria were streaked out from a freeze-dried culture and allowed to grow at 28°C
until colonies were visible. For shorter storage periods, nutrient agar plates were inoculated
with bacteria cultures and re-streaked from a single colony onto fresh agar every two
weeks. They were grown for 24-48 hours at 28°C until colonies were large, and then stored
at 4°C to be used as inoculant for the liquid cultures. All agar plates were checked to ensure
colonies were of uniform shape, size, colour, and consistency. Periodic tests using
Pseudomonas selective agar and the gram-stain tests were employed to ensure everything
was aseptic and the only culture was indeed a gram-negative rod shaped Pseudomonas.
3.2.2 INOCULANT PREPARATION AND HARVESTING
The liquid cultures used to inoculate all aqueous phases, soil slurries, and soil column tests,
were prepared by transferring one loop from a single bacteria colony from a previously
cultivated plate, to 125 mL or 250 mL Erlenmeyer flasks with 50-100 mL of sterilized
medium. Flasks were then placed on a rotary shaker at 200 rpm and maintained at 25-30°C
overnight (12-20 hours) until the bacteria reached an OD600 of 1.0 to 2.0. This indicates that
they have reached their late exponential growth phase. Standard inoculant growth media
49
(used for all tests other than those outlined in section 3.4.1.1) were a BHB broth with
glucose as per minimal media of 2g/L. After overnight growth, bacteria were transferred to
sterile centrifuge tubes and spun at 4000g for 5 minutes, the supernatant was then poured
off, and cells were re-suspended in 0.85% saline (w/v) at room temperature. This process
was repeated before trials were performed to remove residual broth or carbon sources.
Finally, cells were concentrated to an OD600 of 1.5 to 2.0 depending on the aqueous phase
or soil slurry phase experiment and used as the inoculant for the degradation trials.
3.2.3 PLATE COUNTS
The plate count method was used routinely in all experiments as a means of enumeration of
the viable bacteria present. This method involved performing serial dilutions and plating the
dilution series to obtain a dilution of 10 – 100 colony forming units (cfu), which then can be
accurately enumerated visually. Dilutions were done in either 1 mL culture tubes or in 96-
well micro plates, depending on the number of samples necessary. Serial dilutions in 96-well
plates were performed by adding 10 µL of neat solution to 90 µL of saline solution, while
avoiding immersing the pipette tips in the saline solution. New sterile pipette tips were used
to mix the contents in the well and add 10 µL of the mixture to the next well, until a 1x10-7
dilution series was complete. In the 1 mL culture tubes, the same technique was used,
however 100 µL of neat solution was added to 900 µL of saline solution and 100 µL was
transferred for each dilution. Ten microlitres from each dilution were then transferred onto
nutrient agar plates, and placed in a 28°C incubator overnight, or until colonies were large
enough to enumerate (Figure 3.1). All serial dilutions were done in duplicate or triplicate.
The amount of 10 µL was chosen as the transfer volume to the plate because it evaporated
and soaked into the agar within a few minutes and did not allow bacteria to become
50
detached from the growing colonies. This method also allowed an entire dilution series to
be plated on a single agar plate. Final calculations for cfu/mL follow the formula:
cfu/mL = (colonies on the plate) * 10 (dilution# + 2)
Equation 3.1
3.2.4 OPTICAL DENSITY
Optical density (OD) at 600nm was also used as an indicator of cell density in solution. A
standard curve for OD600 and the density of cells per mL was constructed to calibrate the
absorbance value and obtain comparisons between readings. OD was measured by
transferring 1 mL of solution to a disposable UV-cuvette. The same spectrophotometer. The
spectrophotometer was used for all readings, and was zeroed by using the sample matrix
from a sterile stock solution.
CFU/ML ??
100/10 µL 100/10 µL 100/10 µL 100/10 µL 100/10 µL 100/10 µL 100/10
90/900 µL saline
1 2 3 4 5 6 7
x2 duplicate
10 µL of sample
Incubated the plate at 28ºC until colonies
visible and counted dilutions with
number of colonies between 10 and 100
cfu/mL = (colonies on the plate) * 10 (dilution# + 2)
Figure 3.1 Schematic diagram for serial dilutions to determine cfu/mL of solution
51
3.3 SOIL METHODS
Soil was obtained from a laboratory supply (original location unknown) of mixed sand and
silt, and the properties are outlined in the following section. The artificial soil contamination
and contaminant extraction techniques are also presented, along with the efficiency of the
contaminant extraction method.
3.3.1 SOIL PROPERTIES
The particle size distribution of the soil used in all experiments was determined using
American Society for Testing and Materials (ASTM) D6913 standard testing methods for
gradation of soils, using sieve analysis (Figure 3.2). Any material larger the 2mm opening
was discarded to obtain a homogeneous mixture that would be appropriate for bench-scale
testing. According to the ASTM standards, sand is 2mm to 0.05mm; silt is 0.05mm to
0.002mm; and clay is less than 0.002mm. The soil used can therefore be classified as loamy
sand using the soil texture triangle. The loss on ignition test method ASTM D2974-87 was
used to determine the organic content of the soil (Table 3.2). Soil pH was determined in a
1:1 soil slurry with distilled water. Porosity and density were determined gravimetrically
when soil column were packed (Table 3.2 Soil properties
Table 3.2 Soil properties
Soil Parameter Symbol/Units Value
Organic Content % 2.15
pH --- 5.4
Dry Density ρd /(g/cm3) 1.75
Unit Weight γd (kN/m3) 17.2
Specific Gravity Gs 2.8
Average porosity n 0.38
52
Figure 3.2 Particle size distribution
3.3.2 SOIL CONTAMINATION
Soil was sterilized by autoclaving it at 120°C in 100g increments three times, after which the
soil samples were plated on nutrient agar plates to ensure continued sterility. Sterile dry soil
was placed in 1 L shot bottles and spiked with phenanthrene dissolved in acetone. It was
then shaken vigorously for 5 minutes to promote homogeneous distribution of
phenanthrene in the soil. The amount of acetone added was sufficient to completely
saturate the soil, without producing excess liquid after shaking. Acetone was then
evaporated by allowing the sample to rest for 3 days at 30°C under a fume hood. The
contaminated soil was aged between 2 weeks and 1 month before each experiment. After
contamination, soil was re-autoclaved (as it was proven not to be sterile) and the
0.00
20.00
40.00
60.00
80.00
100.00
0.010.1110
Pe
rce
nt
Pa
ssin
g
Particle Size (µm)
53
phenanthrene concentration was analyzed before and after autoclaving. There was no
change in the concentration of phenanthrene in the soil, ensuring that no phenanthrene
volatilized during the autoclaving process.
3.3.3 CONTAMINANT EXTRACTION
The remaining phenanthrene in the soil samples was extracted ultrasonically using a
modified EPA method 3550B (U.S Environmental Protection Agency 1996). Several
researchers, including Lee et al. (1999), Song et al. (2002), and Son et al. (2003 ), have
applied a modified EPA 3550B method and have assessed the efficiency in comparison to:
Soxhlet extraction, Microwave-assisted extraction, Alkaline saponification, and Direct
solvent-extraction. Final recommendations indicate that at low contamination levels, all
methods produce similar results, although further study is needed since each soil sample is
unique. There are also abnormalities in the efficiency of the commonly preferred but less
rigorous ultrasonic and shaking methods due to lower solvent consumption, quick
experimental time, and less complex equipment (Song, Jing et al. 2002; Buco, Moragues et
al. 2004). These shortcomings have now been included in the recently updated EPA method
3550C and the results are comparable to those from the modified 3550B. U.S EPA (2007)
states that samples should be extracted using a solvent system that gives optimum,
reproducible recovery of the analytes from the sample matrix, at the concentrations of
interest. For the soil in this experiment, acetone was selected as the optimum extraction
solvent, as recoveries averaged 96% (range of 85-105%) with RSD of 6%. Dichloromethante,
acetonitrile, and hexane were tested in this study, but the recovery percentage and RSD
were less favourable than acetone.
54
Two grams of soil were placed with 10 mL of solvent [soil : solvent ratio of 1:5 (w/v)] in a 10
mL glass tube with Teflon-lined screw cap. All assays were conducted in triplicates. Acetone
was added to the sample which was disrupted by vortexing for 2 minutes, and
phenanthrene extracted for one hour in a Sonic Bath from Sonicator Instrument
Corporation N.Y Model SC-120. The contents in the tube were mixed for 24 hours on a
horizontal shaker table at room temperature between 22 - 27°C. Samples were centrifuged
for 10 minutes at 4000 rpm, and 1mL of supernatant collected with a glass syringe. This
liquid was filtered through a 0.2 μm poly(tetrafluoroethene) (PTFE) filter into 2mL amber
vials with PTFE-lined screw-caps. These were analyzed for phenanthrene concentration by
High Performance Liquid Chromatography (HPLC). Two grams of the same soil were weighed
and placed into a furnace at 104°C for one hour to determine the dry weight. Results
presented are on a dry weight basis. Control samples from the stock of contaminated soil
were run concurrently with the samples to ensure accuracy of the extraction procedure.
Moisture content in the soil did not appear to change the extraction results.
55
3.4 EXPERIMENTAL PROCEDURES
The following sections define the experimental procedures that were used for this research.
The first objective of the investigation was to determine how bacteria respond in aqueous
solutions to a combination of biosurfactant, glucose, and salicylate in order to enhance the
biodegradation of phenanthrene by measuring changes in cell growth and biodegradation
rates. The second objective aimed to use knowledge gained from these experiments to
determine how well each amendment would work in soil slurry environments. The third
objective was to create a more realistic in situ micro environment to conduct flow-through
experiments and to determine the effect of amendments on degradation rates and flow
characteristics.
3.4.1 OBJECTIVE 1: LIQUID MEDIUM TESTS
Liquid medium tests were performed to determine both cell numbers and phenanthrene
concentration in liquid media after various amendments. The following sections outline the
procedures followed to grow inoculant cultures in various types of media, to determine the
dissolution of phenanthrene in the presence of biosurfactant, and to determine the
degradation rate of phenanthrene due to amendments.
3.4.1.1 Inoculant culture growth
Amendments were made to seed growth media to evaluate their effect on seed culture
growth stages and the degradation of phenanthrene and naphthalene in the liquid culture
trials. Naphthalene, phenanthrene, and salicylic acid were first dissolved in acetone, and
then added to each sterile flask. The acetone was allowed to evaporate before liquid growth
media was added. These cultures were then used as different types of inoculants, and the
degradation trial method was followed, as outlined in section 3.4.1.3.
56
3.4.1.2 Phenanthrene dissolution
Batch dissolution experiments were conducted with phenanthrene at various biosurfactant
concentrations which ranged from 0 to 2,500 mg/L. All trials were conducted in triplicate.
Phenanthrene dissolved in acetone was added to a series of 50 mL glass tubes at a
concentration of 500 mg/L, well above the solubility limit of 1.3 mg/L. The acetone was
allowed to evaporate. Then 10 mL of biosurfactant solution at various concentrations was
added and the tubes placed in a shaker at 200rpm for 48 hours. This allowed them to reach
equilibrium conditions. The mixtures were then centrifuged at 2,500 rpm for 10 minutes to
separate the undissolved phenanthrene crystals from the supernatant. Samples were
filtered through a 0.2µm PTFE filters using a glass syringe, and transferred into 2mL amber
glass vials for analysis on HPLC. Samples were diluted 50% with acetonitrile prior to analysis
in order to obtain the same matrix solution as the prepared calibration standards for the
HPLC.
3.4.1.3 Degradation trials
Degradation trials in liquid media were conducted in 10 mL sterilized glass tubes with Teflon
lined screw caps. Naphthalene, phenanthrene and salicylic acid were first dissolved in
acetone, added to the sterile glass tubes, and the acetone was allowed to evaporate prior to
the addition of the liquid media. All tests were conducted in triplicate and 3 tubes served as
uninoculated controls during each sampling period. This was done to ensure that losses of
contaminate due to vapourization, total recovery of contaminant, and sterility was
accounted for and monitored throughout the experimental procedure. The following steps
summarize the procedure for determining cell concentration and phenanthrene removal in
batch degradation experiments.
1. Starting inoculant cultures were prepared and harvested as per section 3.2.2
57
2. Naphthalene, phenanthrene and salicylic acid were added, depending on the test
conditions and acetone was evaporated.
3. Two milliliters of sterile liquid media containing BHB as the nutrient source and
glucose or biosurfactant (depending on the test conditions) were added and the
tubes left to equilibrate overnight.
4. Tubes were then inoculated with a 1 to 100 dilution (20µL in 2mL) from the
harvested cell suspensions
5. Three tubes for each variable served as initial controls and were examined
immediately. The remaining samples were incubated at 28°C on a shaker table at
200rpm, and every 24 hours three tubes for each trial were removed and examined
as follows:
6. Live cell numbers were determination by drawing off 10µL from each sample and
using the serial dilution plate count method as outlined in section 3.2.3.
7. Phenanthrene or naphthalene concentration was determined by solvent extraction
through the addition of 2mL of hexane to each tube and vortexing for 2 minutes.
Samples were then allowed to stand for one hour until the water and hexane phases
separated and then samples were centrifuged at 2000 rpm for 10 minutes. One mL
of the supernatant was drawn off and placed in an amber 2 mL vial with PTFE lined
screw cap for determination of total phenanthrene or naphthalene concentration by
HPLC analysis.
58
3.4.2 OBJECTIVE 2: SOIL SLURRY TESTS
Soil slurry tests were performed to both determine cell number and the remaining
phenanthrene concentration after various amendments were made to the system. Batch
tests methods to determine the phenanthrene partitioning coefficient in the presence of
biosurfactant are outlined. The procedure followed to determine degradation rates of
phenanthrene in soil slurries due to amendments is explained and summarized.
3.4.2.1 Phenanthrene Desorption in the Presence of Surfactants
Batch desorption experiments were conducted with phenanthrene contaminated soil at
various biosurfactant concentrations, ranging from 0 to 1000mg/L. The soil samples used
were pre-contaminated with phenanthrene concentrations of 50, 100, 250 and 500 mg/kg.
Two grams of contaminated soil were added to a series of 50 mL glass tubes and 10 mL of
biosurfactant solution at concentrations of 0, 250, 500, and 1000 mg/L. The tubes were
placed on a shaker table at 200rpm. Tubes were removed at 0.5, 1, 4 ,8 ,24 , and 48 hours
for determination of phenanthrene in solution. The mixtures were centrifuge at 2,500 rpm
for 10 minutes to separate the soil particles from the supernatant. Samples were then
filtered through a 0.2µm PTFE filter using a glass syringe and transferred into 2 mL amber
glass vials for analysis on HPLC. Samples were diluted 50% with acetonitrile prior to analysis
in order to obtain the same matrix solution as the prepared calibration standards for the
HPLC.
59
3.4.2.2 Degradation Trials
Degradation trials in soil slurries were conducted in 125 mL Erlenmeyer flasks which were
capped with tinfoil during the experiments. Five grams of 500 mg/kg phenanthrene
contaminated soil were added to the flasks along with 50 mL of liquid media for a soil to
water ration of 1:10. All tests were conducted in triplicate and various amendments made
to the liquid media (Table 3.3). Four uninoculated controls were also run with the
experiment to determine the recovery efficiently of the soil extraction, to determine the
amount of phenanthrene in the solution phase during each sampling period, to ensure
sterility throughout the testing period, and to serve as a bases of comparison for the results.
The following steps summarize the procedure for determining cell concentration and
phenanthrene removal in batch degradation soil slurry experiments.
1. Starting innoculant cultures were prepared.
2. Five grams of soil were added to each tube, along with the liquid media
3. Tubes were inoculated with a 1 to 100 dilution (0.5mL in 50mL) from the harvested
cell suspensions to achieve a cell density of 1x107 cells/mL of liquid or 1x10
8 cells/g
of soil.
4. At set time periods on days 4, 7, and 10, samples were taken from the flasks to
determine the amount of phenanthrene in solution and number of active bacteria.
5. Determination of cell number by drawing off 10 µL and using the serial dilution plate
count method.
6. Determination of phenanthrene in suspension by drawing off 1 mL of liquid from
each flask into 2 mL amber HPLC vials. One mL of acetone was added to stop
degradation activity, and bring all phenanthrene into the soluble phase for detection
by HPLC. Everything was completed in triplicate.
60
7. On day 10, in addition to step four, the remaining liquid in the flasks was drawn off
and 10 mL of acetone was added to each flask, placed on the shaker table overnight
to enable the extraction of the remaining phenanthrene from the soil.
Table 3.3 Soil slurry media constituents
Biosurfactant Media and Amendments
0 g/L BHB (uninoculated control)
0 g/L BHB
0 g/L BHB + Salicylate 100mg/L
0 g/L BHB + Glucose 100 mg/L
0.25 g/L BHB (uninoculated control)
0.25 g/L BHB
0.25 g/L BHB + Salicylate 100mg/L
0.25 g/L BHB + Glucose 100 mg/L
1 g/L BHB (uninoculated control)
1 g/L BHB
1 g/L BHB + Salicylate 100mg/L
1 g/L BHB + Glucose 100 mg/L
5 g/L BHB (uninoculated control)
5 g/L BHB
5 g/L BHB + Salicylate 100mg/L
5 g/L BHB + Glucose 100 mg/L
3.4.3 OBJECTIVE 3: COLUMN TESTS
Continuous bench-scale column tests were
conducted using the same contaminated
soil as the slurry tests in order
enhanced bioremediation procedures in a
system replicating in situ treatment.
3.4.3.1 Experimental Apparatus
Column experiments were conducted using
two glass columns, set up identically
conduct parallel experiments comparing
variables in each column
Columns measured 37cm in length
in diameter with PTFE endplates and screw
caps purchased from Omnifit
Valve INC in Boonton, New Jersey
and 30 cm along the length
ceramic cups on the inside of the column to prevent soil clogging and washout during
sampling. The other ends of the ports were connected to PTFE 3
ports could be closed and pressure sensor equip
conducted in an up-flow mane
Parmer EW-07553-85 Masterflex
13 & 14 pharmed tubing. The tubing was
UNF fittings attached to column ports and end plates
61
3: COLUMN TESTS
scale column tests were
me contaminated
in order to optimize
enhanced bioremediation procedures in a
treatment.
Experimental Apparatus
xperiments were conducted using
up identically, to
conduct parallel experiments comparing
(Figure 3.3).
Columns measured 37cm in length and 5cm
with PTFE endplates and screw
Omnifit Bio-Chem
ersey, USA . Sampling ports were installed at 2.5, 5.0, 7.5, 10, 20
(Figure 3.3). PTFE sampling ports were covered with porous
ceramic cups on the inside of the column to prevent soil clogging and washout during
of the ports were connected to PTFE 3-way v
be closed and pressure sensor equipment attached. Experiments were
flow maner to maintain complete saturation in the column using
Masterflex L/S variable-speed modular drive at 1 to 100 rpm
. The tubing was connected to 1/16" O.D. PTFE tubing with 1/4"
attached to column ports and end plates.
Figure 3.3 Soil column setup
pumping experiments
. Sampling ports were installed at 2.5, 5.0, 7.5, 10, 20
were covered with porous
ceramic cups on the inside of the column to prevent soil clogging and washout during
way valves so that the
ment attached. Experiments were
column using a Cole-
1 to 100 rpm, with L/S
/16" O.D. PTFE tubing with 1/4"-28
Soil column setup for uplflow
pumping experiments
62
3.4.3.2 Pressure Measurement
Pressure changes that occurred along the length of the column were measured in five
separate locations using -100 to +200 kPa pressure transducers supplied by ICT International
in Armidale, NSW, Australia. A resolution of 0.1 kPa (1 cm H2O) could be attained for the
pressure transducers by using smart interfaces to maintain a stable reference voltage. Data
was recorded on a 16-bit Plug & Play Smart Logger supplied by ICT international which could
store up to 500,000 date-and-time stamped readings.
3.4.3.3 Micro-foam Generation and Stability
Micro-foam was generated based on a method proposed by Sebba (1985) using a high-
speed spinning disk method. One litre of surfactant solution was poured into a baffled
container and stirred at 8,000 rpm for three minutes. The disk was slowed down to 6,000
rpm for the duration of pumping experiments to maintain microfoam quality. The
temperature of the solution slowly increased over the course of the experiment starting at
25°C and rising to 30°C after approximately four hours of operation. Microfoam stability
measurements used the half-life method. After three minutes of intensive stirring, 100 mL
of microfoam was transferred to a 100 mL measuring cylinder. The height of the clear liquid
interface below the dispersion was recorded at various times, and after approximately 4
hours dispersion was complete and the final volume of the drained liquid recorded. The
stability of the micro-foam is reported as the time needed for draining half of the liquid
from the dispersion, based on a plot of time versus percentage drained.
63
3.4.3.4 Column Packing and Unpacking
All column tests were dry-packed using contaminated soil. Columns were first flushed with a
90% ethanol solution and left for 48 hours to sterilize all parts of the column system.
Afterwards, columns were flushed continuously with sterile miliQ water to remove any
remaining ethanol. Thirty grams of coarse sand was added to the column at the inlet to
prevent finer particles from settling into the inlet tubing. Contaminated soil was pre-
inoculated with P.putida prior to soil column packing. Harvested cells were concentrated to
an OD600 of 2.0 and 1mL per 100g of soil of the concentrated cell solution was added,
corresponding to a cell density of approximately 1x108 cells/g of soil. The soil was vigorously
shaken in sterile lab bottles prior to column packing to distribute the bacteria evenly
throughout the soil. Columns were then dry-packed with soil in 200g increments. After each
increment was added, the column was tapped and a sterile rod was used to pack the soil by
prodding the added increment 5 – 7 times. A 90µm PTFE frit was placed in the top cap of the
column to prevent any washout of soil particles. Columns were then purged with CO2 gas for
approximately 30 minutes to displace any oxygen. They were then saturated with sterile
BHB nutrient broth. Column unpacking was done in 7 equally spaced layers from top to
bottom. At each layer triplicate extractions were performed on the soil to determine
remaining phenanthrene concentration. Columns were repacked for each experiment that
was carried out using the same mass of soil and amount of packing.
64
3.4.3.5 Experimental Operating Conditions
In un-inoculated soil, column tracer breakthrough curve tests were conducted using a
potassium chloride solution in BHB broth at a flow rate of 0.2mL/min. A biosurfactant
breakthrough curve was attained using the same flow rate to determine the amount of
adsorption that occurs during biosurfactant addition. A flow rate of 0.2mL/min corresponds
to a flow velocity of 0.026 cm/min or 24 hours to pump one pore volume of solution
through the column.
Column tracer tests were then performed using a potassium chloride solution in BHB broth
with biosurfactant added at a concentration of 1000 g/L and 5000 g/L at a flow rate of 10
mL/min. The same solution was also used to conduct tracer tests using microfoam instead
of aqueous solution to determine the liquid fraction break through curve when using
microfoam, and determine pressure changes that occur during microfoam pumping across
the length of the column.
Three series of degradation trials were conducted with the solution amendments and flow
rates as outlined below in Table 3.4. Degradation trials were run for up to 10 days and on
each day samples were taken from the inlet, various sampling ports along the column
length, and the column outlet to determine phenanthrene and biosurfactant concentration
in solution, cfu/mL, and dissolved oxygen levels (using a MI-730 dip-type 02 microelectrode
connected to an OM-4 Oxygen meter from Microelectrodes INC, Bedford, New Hampshire,
USA). Influent solutions were all sterilized and were placed on magnetic stir plates to
maintain saturated levels of oxygen in the influent solution. Depending on the testing
conditions as indicated in Table 3.4, the influent was switched to a second solution for the
65
pulse duration simply by using a 3-way PTFE valve on the influent line. Special care was
taken to sterilize all tubing and equipment throughout the testing period. At the end of 10
days the column was unpacked and total remaining phenanthrene in soil was determined.
Table 3.4 Column trial experimental design
Column
number
Flow
conditions
Total flow Standard
influent
solution
Pulse
Influent
solution
Pulse
duration
Initial soil
contamination
TRIAL 1
1 0.2mL/min
10 days
2,880mL
(10.1 PV)
BHB
biosurfactant 1g/L
salicylate 100mg/L
n/a n/a 450 mg/kg
2 0.2mL/min
10 days
2,880mL
(10.1 PV)
BHB
biosurfactant 1g/l
n/a n/a 450 mg/kg
TRIAL 2
1 0.5mL/min
10 days
6,900mL
(24.3 PV)
BHB biosurfactant
1g/L salicylite
100mg/L as
microfoam
4 hours
on days
2,4,6 &
8
450 mg/kg
2 0.5mL/min
10 days
6,900mL
(24.3 PV)
BHB biosurfactant
1g/L salicylite
100mg/L as
liquid solution
4 hours
on days
2,4,6 &
8
450 mg/kg
66
3.5 ANALITICAL METHODS
3.5.1 PAH DETECTION
Independent stock solutions of naphthalene and phenanthrene were prepared by dissolving
0.01g of naphthalene or phenanthrene crystals in a small amount of hexane or acetonitrile,
then bulking up to 10 mL with the same solvent. This method was used to create stock
solutions of 1000 mg/L, stored in the dark at 4°C, and last up to three months. From the
stock solutions independent solutions ranging from 0.1 to 100mg/L were prepared to
construct 5 point calibration lines for each sample sequence that was run; determination
coefficients obtained were higher than 0.99 to insure good linearity and reproducibility over
the sampling range. To account for various matrix compositions multiple standards were
made in various solvent solutions. This insured the relative recovery was near 100% -- the
calibration standards along with given test samples were processed and run using the HPLC
in the same way (García-Falcón, Cancho-Grande et al. 2004). HPLC was equipped with a
Dionex P580 pump, Dionex ASI-100 Automated Sample Injector, Dionex UVD340S UV
detector and a Phenomenex C18 Column (150x4.6mm), used for the detection of
naphthalene and phenanthrene. HPLC analysis was performed isocratically with a mobile
phase of 30% water and 70% acetonitrile at a flow rate of 1.5 ml/min to elute the column,
with the sample injection volume being 10µl. The optimal UV detection wavelengths were
found to be 220nm and 250nm for Naphthalene and Phenanthrene respectively; these
wavelengths gave the greatest response height and area, and allowed for detection limits
below 0.01 mg/L for phenanthrene and 0.1 mg/L for naphthalene using this method.
Chromeleon Client Version 6.80 by Dionex Corporation was used for data collection and
analysis.
67
3.5.2 BIOSURFACTANT DETECTION
Biosurfactant was quantified using two different methods, the first using a total organic
carbon (TOC) analyzer to determine the relative amount of carbon present and relate back
to the concentration of rhamnolipid, and the second using the HPLC to quantify both the R1
and R2 rhamnolipids present in solution. TOC samples were collected in 40-mL pre-cleaned
glass vials were measured immediately after sapling using a TOC-V-CSH analyzer (Shimadzu
Corporation, Kyoto, Japan).
Biosurfactant was detected using a HPLC/ELSD following the method outlined by (Wang,
Fang et al. 2007). The HPLC gradient was obtained by the following method: starting at 8%
solvent B and holding for 1 minute, ramping to 75% solvent B in 20 minutes, holding 75%
solvent B for 10 minutes, backing to 8% solvent B in 1 minute and holding at 8% solvent B
for 5 minutes. Solvent A was 98:2 (v/v) water : acetonitrile with 0.1% acetic acid, and solvent
B is 10:90 water : acetonitrile with 0.1 % acetic acid. Rhamnolipid samples were filtered
through a 0.22um syringe filter prior to analysis using a 50µl sample injection volume.
3.5.3 CHLORIDE ANION
Chloride used as a tracer in the column tests was detected using a Dionex DX-120 Ion
Chromatograph (IC) with an AS9-HC column. The effluent was a Na2CO3 solution. A
potassium chloride stock solution was made to prepare calibration standards, and using this
method a detection limit of 0.1ppm was obtained.
68
CHAPTER 4 RESULTS AND DISCUSSION
4.1 OBJECTIVE 1: LIQUID CULTURES
4.1.1 BACTERIA GROWTH ON VARIOUS SUBSTRATES
Influences of glucose, naphthalene, salicylate, and biosurfactant on bacteria growth were
investigated to evaluate the rate and extent of growth that could be obtained from the
various carbon sources. Bacteria growing in suspended, shaken, cultures with crystalline
PAH above the aqueous solubility limit as the sole carbon source of energy, exhibit
characteristic growth curves involving four phases. The phases are: (i) lag phase, (ii)
exponential phase, (iii) a subsequent phase with pseudo-linear growth, and (iv) pseudo-
stationary phase (Volkering, Breure et al. 1992; Volkering, Breure et al. 1993; Wick,
Colangelo et al. 2001; Johnsen, Wick et al. 2005). The partial growth curves (Figure 4.1) for
P.putida in four different medias of glucose (2g/L), naphthalene (0.5g/L), salicylate (0.5g/L)
and naphthalene (0.5g/L) + biosurfactant (1g/L) showed P.putida was not just capable of
using naphthalene as a sole growth substrate, but was able to obtain the same amount of
growth as with highly water-soluble substrates such as glucose. Adding biosurfactant to
increase the solubility of naphthalene had no effect on increasing the growth rate beyond
that obtained without biosurfactant addition or with glucose, indicating that bioavailability
of naphthalene is not limiting the growth of P.putida in liquid cultures. Naphthalene has a
higher water solubility (31.4mg/L) compared to most other PAHs (<1mg/L), therefore the
dissolution of naphthalene crystals into solution occurred at a rate high to support the
organisms’ exponential growth. In a batch study by Wick et al. (2001), only a few anthracene
crystals (<2g/L) resulted in pseudo-linear growth due to low dissolution fluxes, whereas
exponential growth was only obtained when high amounts of solid anthracene (30g/L) were
69
provided. As the solubility of naphthalene is greater than 600-fold higher than anthracene it
could be assumed that the addition of 500mg/L of naphthalene crystals would have
provided dissolution fluxes that would produce enough substrate to support exponential
growth.
With this specific Pseudomonas putida ATCC 17484 strain, the biodegradation rate appears
to be controlled by the metabolic activity of the bacteria. Salicylate did not produce the
same rate of growth as glucose or naphthalene indicating a slightly lower amount of
metabolic activity occurring when salicylate is the sole carbon source. Cells grown in 500
mg/L and 1000 mg/L biosurfactant were also tested; after 24 hours the growth became
stationary with OD measuring 0.131 and OD measuring 0.235, respectively. The increased
biosurfactant concentration produced roughly twice the cell density, but the low amounts of
biomass produced overall indicates that biosurfactant is not a preferential carbon source.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
0 2 4 6 8 10 12 14
OD
60
0n
m
Time (hours)
Glucose Naphthalene Salicylate Naphthalene + Biosurfactant
Figure 4.1 Partial growth curves for P.Putida until early stationary phase in four growth medias
of glucose (2g/L); naphthalene (0.5g/L); salicylate (0.5g/L); and naphthalene (0.5g/L) +
biosurfactant (1g/L)
70
4.1.2 EFFECT OF INOCULANT ACCLIMATIZATION AND PRE-TREATMENT
Different acclimatization or pre-treatments of the seed inoculants were tested to determine
what affect the seed growth conditions have on the removal of naphthalene and
phenanthrene in subsequent batch studies. Bacteria for lab studies are often grown in the
presence of target contaminants to maintain the degradative phenotype, and to increase
the enzyme activity that is responsible for the degradation of the target compound. This is
done to improve removal efficiency and decrease lag time in cell growth when bacterial
seeds are added to the system where the contaminant is to be degraded. The affect of
seven inoculant pre-treatments were tested (Figure 4.2) on the subsequent degradation of
naphthalene over a four day period. All cultures obtained the same amount of cell growth
(between 1x108 and 1x10
9 cfu/mL) in the first 24 hours. This concentration was maintained
over the 4 day testing period (Figure 4.2). The results for the bacteria growth in each
separate system were proven to be statistically similar to each other and are presented as
an average from all the tested inoculant systems (single factor ANOVA F=0.88 p=0.51). The
total amount of degradation was similar in all systems after 96 hours. After 74 hours the
samples where BHB was the nutrient source during seed growth showed virtually total
mineralization of all the naphthalene, whereas the samples which had LB as the nutrient
source during seed growth required 96 hours before all the naphthalene was mineralized.
Growth of the seed for one week prior to inoculation showed a lag phase of approximately
24 hours before significant naphthalene removal was measured. After 24 hours
naphthalene removal was similar in all the systems regardless of the growth conditions of
the inoculant seeds. Guerin and Boyd (1995) obtained similar observations where cultures
of P.putida maintained a high degree of capacity to degrade naphthalene for several days
following entry into stationary phase. Naphthalene degradation activity was present
71
constitutively at low levels under all growth conditions and was rapidly induced to high
levels upon exposure to naphthalene. Naphthalene loss due to volatilization was substantial
over the 5 day experimental period as indicated by a 50% loss in the control samples.
Volatilization adds uncertainty to the results; therefore, it is difficult to draw further
conclusions from the study of aerobic naphthalene degradation as accounting for
naphthalene loss.
Figure 4.2 Naphthalene degradation and cell growth in liquid cultures containing different
bacteria inoculant seeds which were pre-grown in seven different solutions (s1
BHB+naphthalene+glucose grown for 1 week; s2 BHB+glucose; s3 BHB+naphthalene+glucose; s4
BHB+salicylic acid+glucose; s5 LB; s6 LB+naphthalene; s7 LB+salicylic acid; s2 - s7 grown overnight
approximately 20 hours growth)
1.E+02
1.E+03
1.E+04
1.E+05
1.E+06
1.E+07
1.E+08
1.E+09
0
1
10
100
1000
0 20 40 60 80 100
Ba
cte
ria
Co
nce
ntr
ait
on
(cf
u/m
L)
To
tal
Na
ph
tha
len
e R
em
ain
ing
(m
g/L
)
Time (hours)
s1 s2 s3 s4 s5 s6 s7 control Viable Cells
72
Different pre-treatments were also tested with regards to phenanthrene degradation and
showed little change in the degradation ability of the bacteria (data not shown). The only
added benefit was seen in pure phenanthrene solution with no amendments, where an
increase in total phenanthrene degradation occurred when the inoculant was cultured in a
naphthalene and BHB solution overnight and spiked with glucose 3 hours prior to
inoculation. However, there was no statistical difference between inoculants that were
added to phenanthrene solutions that contained amendments of salicylate, biosurfactant
and glucose. This indicates that pre-treatment and acclimatization of the bacteria had
minimal effect compared to amendments to the culture during degradation as discussed in
section 4.1.4.
Studies with a P.putida have shown the strongest expression of nah genes (genes
responsible for naphthalene degradation) in late-log phase growth, where the catabolic
operons were poorly induced in the early-exponential growth phase but strongly induced in
the late-exponential-growth phase (Hugouvieux-Cotte-Pattat, Kohler et al. 1990; Guerin and
Boyd 1995). The results obtained in this study support these observations and indicate that
harvesting the seed cultures in the late-exponential or early stationary phase will have the
best ability to enhance the degradation process, and it is less important to acclimatize
P.putida bacteria seeds prior to inoculation.
73
4.1.3 BIOSURFACTANT SOLUBILITY ENHANCEMENT
The apparent aqueous solubility enhancement of phenanthrene in a biosurfactant solution
can be expressed as a function of both the interaction with the surfactant monomers and
with the surfactant micelles once the concentrations exceeds the respective CMC (Zhu,
Chen et al. 2003; Wang and Keller 2008).
���
��� � " ������ " ������ (Equation 4.1)
• FG� is the apparent phenanthrene solubility at a total stoichiometric surfactant
concentration of Xmn and Xmc (g/L)
• FG is the intrinsic phenanthrene solubility in water in the absence of the surfactant
(mg/L)
• HI<and HIJ are the surfactant monomer and micelle concentration in water (g/L)
• �I<and �IJ are the phenanthrene portioning coefficient with the surfactant
monomer and micellar phases (L/g)
0
5
10
15
20
25
30
35
40
0 500 1000 1500 2000 2500
Ph
en
an
thre
ne
Co
nce
ntr
ati
on
(S
*w
/Sw
)
Biosurfactant Concentration (mg/L)
S*w/Sw = 1
Figure 4.3 Phenanthrene solubility enhancement as a function of biosurfactant
concentration. The equation refers to the fit of data above the CMC and
� � ��� ��⁄
74
FG� and FGwere measured directly on the HPLC and the results of the Phenanthrene
solubility enhancement are presented ( ). Linear regression was conducted on the solubility
data for values of FG� FG L 1⁄ , using the average of triplicate measurements of
phenanthrene solubility at various biosurfactant concentrations above the CMC. Similar
linear relationship for surfactant enhanced solubility curves above the CHC have been
determined by a number of researchers (Kim, Park et al. 2001; Shin, Kim et al. 2004; Yu, Zhu
et al. 2007). It should be noted that the value for Sw (phenanthrene solubility in water)
obtained in this experiment was 1.12 mg/L and this value was used in subsequent
calculations. There appears to be no consensus for the phenanthrene solubility in water as
results are reported in the range of 0.4-1.6 mg/L for temperatures ranging from 8.5-30°C;
there is no consensus for a single solubility value in water at 25°C (Verschueren 1983). It is
assumed there is a lack of interaction between surfactant monomers and phenanthrene, as
phenanthrene Kow = 104.57
and according to literature only extremely hydrophobic organic
compounds such as DDT with at Kow of 106.36
are known to associate with surfactant
monomers in the aqueous phase, allowing Kmn to equal 0 in Equation 4.1. This allows
FG� FG � 1⁄ before the CMC is reached, and after the CMC has been exceeded there is a
linear increase in solubility with increasing surfactant concentration, with R2 = 0.9979 for the
data presented (Figure 4.3). Using equation 4.1 the CMC can be solved for by equating
FG� FG L 1⁄ with the linear regression line and this gives a CMC of 63 mg/L. This value
correlates well with Shin et al. (2008), who reports a CMC with a similar rhamnolipid of 0.1
mmol/L at a pH of 7. This is 56 mg/L assuming that the biosurfactant product supplied was
an equal mix of mono to di-rhamnolipied by weight. It was also shown by (Shin, Kim et al.
2008) that pH can have a significant effect on the CMC of the rhamnolipid and can cause
considerable changes in the solubility of phenanthrene in the presence of rhamnolipid.
75
4.1.4 PHENANTHRENE DEGRADATION
Phenanthrene degradation took place in all systems over a 46 hour period, with the addition
of BHB broth and/or biosurfactant (1.0 g/L), salicylate (100 mg/L), and glucose (100 mg/L)
(Figure 4.4). Increased removal of phenanthrene occurred in systems which contained
salicylate, showing a 3.5 fold increase in removal compared to no amendment addition. The
addition of biosurfactant enhanced the amount of degradation in all systems with 1.5 to 2.6-
fold increase in phenanthrene removal versus equivalent systems with no biosurfactant
addition.
Figure 4.4 Phenanthrene degradation in liquid cultures containing BHB and/or biosurfactant
(1000mg/L), salicylate (100mg/L), and glucose (100mg/L). Data presented is the average of
triplicate measurements taken at 22 and 46 hours after inoculation.
0
10
20
30
40
50
60
70
80
90
100
BHB BHB + Salicylic
Acid
BHB+ Glucose BHB +
Biosurfactant
BHB +
Biosurfactant
+ Salicylic Acid
BHB +
Biosurfactant
+ Glucose
% P
he
na
nth
ren
e R
em
ov
al
After 22 hours After 46 hours
76
Bacterial uptake of solubilized compounds is hypothesized to be influenced by
biosurfactants, as the uptake of biosurfactant-solubilized molecules has been found to be
faster than the uptake of dissolved (i.e monodispered) molecules (Johnsen, Wick et al.
2005). Results here indicate P.putida degradation occurs faster in the presence of
biosurfactant. However, salicylate addition, with no biosurfactant addition, increased the
total degradation of phenanthrene 30% more than system with only biosurfactant addition.
Glucose was shown to improve phenanthrene degradation which is attributed to the
presence of an additional carbon source that increased bacteria growth in the system.
Phenanthrene removal was nearly 2-fold more due to salicylate addition compared to the
glucose system; therefore it can be assumed that it is not just the increased bacterial growth
due to the presence of an additional carbon source that is responsible for the increased
removal in the salicylate systems. This indicates a greater amount of phenanthrene
degradation could be achieved through amendments that specifically cause metabolic
pathway induction, when compared to increased solubility brought about by the addition of
biosurfactant, or increased growth brought about by additional carbon substrates.
Salicylate enhanced the rate of degradation in the first 22 hours, although this rate was not
sustained, and decreased considerably from 2.2 mg/hr to 0.88 mg/hr in the following 24
hours (Table 4.1). The addition of biosurfactant increased initial degradation rates, and
sustained relatively higher degradation rates over the 46 hour period compared to systems
without biosurfactant. Determining the length of time apparent enhancement strategies
influence the system for has important implications for the design of in situ treatment
strategies. Powell et al. (2008) determined that the rate at which a carbon source is made
available or the instantaneous concentration in the medium (spike addition versus
77
continuous addition) considerably changes the outcomes achieved. Therefore it is important
to determine what method works best for a particular system when evaluating remediation
strategies. Salicylate offers the best initial and total phenanthrene removal, and
biosurfactant augmented these results and could offer benefit over a longer period of time.
Table 4.1 Rate of phenanthrene degradation in liquid cultures expressed as mg of
phenanthrene degraded / hour
Media Constituents Degradation rate
after 22 hours
Degradation rate
22 hours - 46 hours
Average over
46 hours
BHB 0.47 0.43 0.45
BHB + salicylate 2.20 0.88 1.51
BHB + glucose 1.11 0.52 0.80
BHB + biosurfactant 1.45 0.89 1.16
BHB + biosurfactant + salicylate 2.60 1.40 1.98
BHB + biosurfactant + glucose 1.89 1.33 1.59
78
4.2 OBJECTIVE 2: SOIL SLURRIES
4.2.1 PHENANTHRENE DISTRIBUTION IN THE PRESENCE OF BIOSURFACTANTS
Desorption experiments were conducted to determine how increasing rhamnolipid
concentrations affected the amount of phenanthrene in the aqueous phase. Samples were
removed at succeeding time intervals of (1, 4, 8, 24, and 48 hours) to determine the change
in aqueous phenanthrene concentration over time, as well as the final equilibrium
phenanthrene concentration after 48 hours. Equilibrium phenanthrene concentration for
increasing surfactant concentrations of 0, 250, 500 and 1000 mg/L (Figure 4.5) indicate that
equilibrium desorption isotherms can be fitted with a linear equation and described with a
single soil-water partitioning coefficient Kd. Similar linear desorption isotherms were found
by Haung and Cha (2001) and were used to calculate the desorption of PAHs in the
0
100
200
300
400
500
600
0 1 2 3 4 5
Ph
en
an
thre
ne
in
so
il u
g/g
Phenanthrene in solution mg/L
0 mg/l 250 mg/l 500 mg/l 1000 mg/lKd = 829 Kd = 520 Kd = 217 Kd = 96
Biosurfactant Concentration
Figure 4.5 Phenanthrene desorption from soil in the presence of biosurfactant.
Desorption partitioning coefficient Kd calculated from the linear regression trendline for
each series of data.
79
presences of a rhamnolipid biosurfactant. Results indicate that phenanthrene interaction
with the biosurfactant solution is independent of the phenanthrene concentration in the
soil in the range of soil contamination studied (50 to 500 mg/kg), as aqueous solubility limits
were not reached. The partitioning coefficient Kd decreased with increasing biosurfactant
concentration, with Kd decreasing by a multiple of 8.5, from 829 to 96, due to the addition
of 1000 mg/L of biosurfactant.
The sorption of surfactant onto soils is known to have a significant effect on the
performance of surfactant enhanced desorption of contaminants (Zhou and Zhu 2007; Zhu
and Zhou 2008; Laha, Tansel et al. In Press). Although biosurfactants form a mobile micellar
pseudophase, they are also adsorbed by the soil matrix and can lead to phenanthrene
partitioning into the adsorbed biosurfactants (Zhou and Zhu 2007). This in turn enhances
the sorption of phenanthrene on the soil. The process of increased phenanthrene
adsorption to soil appears to be responsible for the trend observed over a 48 hour period
(Figure 4.6). The initial amount of phenanthrene in solution increased for approximately 4 to
8 hours, and then began to decrease over the remainder of the experiment. This is believed
to be caused by the sorption of biosurfactant to the soil surface, which in turn enhanced the
sorption of phenanthrene back onto the soil. The sorption sites on the soil would become
‘available’ after phenanthrene initially desorbs into solution, leaving sites available for
biosurfactants to sorb to the soil surface. This would increase the overall organic content of
the soil, causing sorption of phenanthrene back onto the soil before an equilibrium
concentration is reached. Garcia-Junco et al. (2003), saw a similar occurrence where
biosurfactant may have promoted sorption of PAH to the soil by modifying the surface
hydrophobicity due to the orientation of the hydrophobic moieties of the first
80
Figure 4.6 Phenanthrene desorption from contaminated soil (50, 100, 250, 500 mg/kg) into
aqueous solution in the presence of biosurfactant over a 48 hour period.
0
1
2
3
4
5
6
Ph
en
an
thre
ne
in
solu
tio
nm
g/L
50 mg/kg 100 mg/kg 250 mg/kg 500 mg/kg
Biosurfactant 1000 mg/L
0
1
2
3
4
5
Ph
en
an
thre
ne
in
solu
tio
n m
g/L
Biosurfactant 500 mg/L
0
0.5
1
1.5
2
2.5
3
Ph
en
an
thre
ne
in
solu
tio
n m
g/L
Biosurfactant 250 mg/L
0
0.2
0.4
0.6
0.8
0 10 20 30 40 50
Ph
en
an
thre
ne
in
solu
tio
n m
g/L
Time Hours
Biosurfactant 0 mg/L
81
biosurfactant layer that adsorbs to the surface in the aqueous phase. PAHs may also
partition into the hemimicelles that are formed as a layer on the soil surface by sorbed
biosurfactants. The time dependent phenanthrene sorption does not appear to occur in the
1000 mg/L biosurfactant solution (Figure 4.6). At lower biosurfactant concentrations the
amount of surfactant sorption to the soil accounts for a majority of the added surfactant;
whereas at high biosurfactant concentrations (1000 mg/L) there is still a significant amount
of biosurfactant in the aqueous phase which can act to increase the overall aqueous
solubility of phenanthrene.
Rhamnolipid adsorption to soil types with similar amounts of organic matter have been
reported as high as 2,500 mg/kg, with a rhamnolipid partitioning coefficient of 34 L/kg
(Huang and Cha 2001). However, if there is enough biosurfactant remaining in solution (as
with the 1000 mg/L solution), it appears the amount of rhamnolipid sorption to the soil has
no significant effect on the overall PAH desorption. Low levels of biosurfactant increase the
affinity of the solids for aqueous phenanthrene due to biosurfactant sorption, whereas at
levels well above the CMC, the Kd values were lower due to competition between micellar
and sorbed biosurfactants for phenanthrene partitioning.
Table 4.2 Calculated phenanthrene soil partitioning coefficient Kd, and phenanthrene
partitioning onto soil sorbed surfactant coefficient Ks
Biosurfactant Concentration (mg/L) Kmc (L/kg) Kd* (L/kg) Log Ks
250 8,894 520 4.40
500 11,535 217 4.51
1,000 12,585 96 4.52
82
Using equations 2.2 and 4.1, the distribution coefficient Ks (the solute distribution
coefficient with the soil-sorbed surfactant) which represents the quantity of phenanthrene
sorbed to the soil-sorbed biosurfactant can be solved using the Kd values determined, and
assuming Qs equals 34 (L/kg) from (Huang and Cha 2001). The calculated Ks values for
phenanthrene partitioning onto soil sorbed surfactant (Table 4.2) follow a similar trend as
reported by Zhu and Zhou (2008). Ks values ranging from 3.75 to 4.75 on the log scale were
reported, depending on the concentration and type of surfactant added. The trend Zhu and
Zhou (2008) determined showed when 0 to 200 mg/L of titron X-100 were added, values of
Ks increased from 4.0 to 4.25. Surfactant addition above 200 mg/L did not increased Ks any
further and values were constant for the addition of 200 mg/L to 2000 mg/L of surfactant.
This is similar to the trend (Table 4.2) were Ks is 4.4 at 250 mg/L and increases to a constant
value near 4.5 at 500 mg/L and 1000 mg/L. Ks has a strong relation to the amount of
surfactant sorbed to the soil, and accounts for a large amount of increased PAH sorption to
soil that occurs due to the addition of surfactant. This can be attributed to the three stage
surfactant adsorption process that was described by Torrens et al. (1998). At low surfactant
concentrations, the sorbed surfactant molecules are spread out over the soil surface. As the
surfactant concentration increases the monomers form a monolayer or hemimicelles, which
is considered as a weak partition phase for PAHs (Zhu and Zhou 2008). As the surfactant
concentration increases the soil surface reaches a point where the entire solid surface is
covered by a hemimicelles and begins to form surface micelles (admicelle) which are
bilayers on the soil surface that have greater sorption capacity for PAHs. Finally surfactant
sorption plateaus as micelles begin to return to solution and the soil has research surfactant
sorption capacity and equilibrium conditions exist.
83
4.2.2 SOIL DEGRADATION
Phenanthrene degradation occurred in the combined aqueous phase and suspended
organic matter in soil slurries containing biosurfactant (0, 0.25, 0.5, 1.0 g/L), salicylate (100
mg/L), and glucose (100 mg/L) inoculated with P.putida (Figure 4.7). Both natural organic
mater and biosurfactants appear to increase the apparent aqueous solubility of
phenanthrene as the concentration in the control flasks averaged 32 mg/L with no
biosurfactant present. This increased to 38 mg/L in the presence of 1 g/L biosurfactant. The
results presented (Figure 4.7) represent both the aqueous solubility and the phenanthrene
concentration in the suspended organic matter; as samples were taken from the liquid in
the soil slurries but were not filtered to remove suspended soil particles prior to the
addition of acetone to extract the phenanthrene. Filtered samples were taken during the
first sampling period from the control flasks and there was 1.24, 3.01, 6.27, 6.88 mg/L of
phenanthrene for 0, 0.5, 1.0, 5.0 g/L biosurfactant concentrations respectively. This
correlated with the expected values from the phenanthrene desorption results present in
section 4.2.1. The total phenanthrene concentrations (Figure 4.7) are a result of the
enhanced solubility due to the addition of biosurfactant, in addition to approximately 32
mg/L of phenanthrene that was present in the suspended organic matter and dissolved
organic matter. Similar results were obtained by Cho et al. (2002) where apparent solubility
of phenanthrene and other PAHs was accounted for by the sum of the phenanthrene
solubility in natural organic matter solution plus the solubility in a non-ionic surfactant
solution. Roskam and Comans (In Press) also observed PAH levels in solution that were up to
an order of magnitude higher in batch tests versus column tests due to the large dissolved
organic carbon molecules that were generated by more vigorous mixing.
84
The amount of phenanthrene present in solution was constant over sampling periods of 4,
7, and 10 days. Equilibrium existed between concentration of phenanthrene being degraded
by the bacteria, and the rate of phenanthrene desorption from the soil to the aqueous
phase. There was 1 to 4-fold less phenanthrene present in solution in flasks which contained
salicylate, regardless of the amount of biosurfactant present. There were 5.6, 2.3, 1.5, and
1.7 mg/L of phenanthrene present in the systems which contained biosurfactant at 0, 0.25,
0.1
1
10
Co
ntr
ol
BH
B
BH
B +
Sa
licy
lite
BH
B +
Glu
cose
Co
ntr
ol
BH
B
BH
B +
Sa
licy
lite
BH
B +
Glu
cose
Co
ntr
ol
BH
B
BH
B +
Sa
licy
lite
BH
B +
Glu
cose
Co
ntr
ol
BH
B
BH
B +
Sa
licy
lite
BH
B +
Glu
cose
No Biosurfactant Biosurfactant 0.25 g/L Biosurfactant 1 g/L Biosurfactant 5 g/L
Ph
en
an
thre
ne
(m
g/L
)
Day 4 Day 7 Day 10
Figure 4.7 Total phenanthrene concentration in solution and suspended/dissolved organic
matter in soil slurries containing BHB and/or biosurfactant (0.25, 1, 5 g/L), salicylate
(100mg/L), and glucose (100mg/L) over a 10 day period
85
1, and 5 g/L respectively. As the amount of biosurfactant increased, the amount of
phenanthrene in solution decreased, until a point between 1 g/L and 5 g/L biosurfactant
concentration. This signifies increased microbial degradation due to the addition of
biosurfactant until a point between 1 g/L and 5 g/L where the additional biosurfactant did
not improved degradation. Biosurfactant also appeared to enhance degradation in systems
with glucose, as the average concentration of phenanthrene in solution decreased from 2.5
mg/L with no biosurfactant present, to 0.5 mg/L in 5 g/L biosurfactant solution.
Biosurfactant had little effect on systems with salicylate as average amounts of
phenanthrene were 0.4 mg/L in all salicylate systems. There was a 4-fold increase in the
concentration of phenanthrene in the 5 g/L biosurfactant and salicylate system, compared
to the system with no biosurfactant.
Mass balance calculations were performed to determine total phenanthrene remaining
including both phenanthrene in the soil and aqueous phase after 10 days of incubation.
Results are presented as total phenanthrene in mg/kg of dry soil (Figure 4.8). The greatest
phenanthrene removal was in systems that contained salicylate with greater than 90%
phenanthrene removal achieved in all instances. There was also a significant decrease in the
concentration of phenanthrene remaining due to the addition of biosurfactant with 86, 90
and 91 percent removal due to the addition of 0.25, 1.0 and 5.0 g/L of biosurfactant
respectively compared to 68% removal when no biosurfactant was added. Biosurfactant
also improved the amount of removal when glucose was present, but the effects were less
significant when compared to the system with no biosurfactant and only glucose, with
removals increasing by 1.6, 1.9 and 1.6-fold with the addition of 0.25, 1.0 and 5.0 g/L of
biosurfactant respectively. With salicylate the effects of biosurfactant were almost non
86
existent as the total removal was 1.2-fold more due to the presence of 0.25 g/L
biosurfactant, and there was no improvement and a decrease in total removal when 1.0 and
5.0 g/L biosurfactant was added. From the results obtained it appears that almost complete
removal was obtained in most systems, and the concentration of phenanthrene in solution
changed very little over 10 days.
Total live cell counts over the 10 day experimental period (Figure 4.9) represent both the
cell concentration in the aqueous phase and cells that could have been attached to the
suspended organic matter, as solution was drawn off from each flask and not filtered to
remove suspended soil matter prior to serial dilutions. The same level of active growth near
0
20
40
60
80
100
120
140
160
180
BH
B
BH
B +
Sa
licy
lite
BH
B +
Glu
cose
BH
B
BH
B +
Sa
licy
lite
BH
B +
Glu
cose
BH
B
BH
B +
Sa
licy
lite
BH
B +
Glu
cose
BH
B
BH
B +
Sa
licy
lite
BH
B +
Glu
cose
No Biosurfactant Biosurfactant 0.25 g/L Biosurfactant 1 g/L Biosurfactant 5 g/L
To
tal
ph
en
an
thre
ne
re
ma
inin
g (
mg
/kg
)
Figure 4.8 Total remaining phenanthrene in soil slurries after 10 days of bioremediation, results
presented as phenanthrene remaining in mg/kg of dry soil.
87
4x107 cfu/mL occurred in all samples with no biosurfactant. This suggests salicylate
improved degradation through means other than increased quantities of live cells. All live
cell counts were in the range of 1x107 to 1x10
9 cfu/mL with biosurfactant generating
positive effects on the total cfu/mL when glucose was present. Biosurfactant caused no
significant change on the amount of cfu/mL in systems with salicylate. Theses results
support a hypothesis by Johnsen et al. (2005) that PAH degrading population in soil are
mostly not growing, and cells are in a pseudo-stationary phase where transient growth only
replaces decaying and washed out cells until the habitat’s mass transfer-controlled carrying
capacity is reached again .
The growth characteristics correspond to the trends observed in the concentration of
phenanthrene remaining in the aqueous phase (Figure 4.7). There appeared to be more
growth present in glucose systems as biosurfactant concentrations increased from 0 to 5
g/L, and this corresponds to the lower amount of phenanthrene present in aqueous solution
for the same increase in biosurfactant concentration. There was roughly the equivalent
phenanthrene concentration in all systems with salicylate regardless of the amount of
biosurfactant present, and this corresponds to the equivalent live cell concentration
between 5x107 and 1x10
8. Powell et al. (2008) observed a similar trend where salicylate
degrading bacteria increased in abundance substantially after enrichment by continuous
addition of salicylate in batch cultures but did not increase in abundance in response to the
spike addition. However Powell et al. (2008) suggested that enrichment with salicylate can
select for naphthalene-degrading bacteria, but does not select for organisms responsible for
degrading PAHs of higher molecular weight, as phenanthrene and benzo[a]pyrene
degradation where not enhanced. This result actually depends on the strain of bacteria
88
present and the metabolic pathway that exists in the bacteria, as it has been observed in
this research that salicylate can enhance phenanthrene degradation in cultures that are
capable of both phenanthrene and naphthalene degradation. The ability to induce the
metabolism or cometabolism of a target compound can be a sufficient bioremediation
strategy. However, information about the metabolic pathways intermediates that induce
the catabolic enzymes being utilized by the bacteria are required.
Figure 4.9 Live cell counts (cfu/mL) taken from soil slurry solution over 10 days. Results
presented are averages from duplicate or triplicate plate counts.
1.00E+06
1.00E+07
1.00E+08
1.00E+09
0 2 4 6 8 10
Pse
ud
om
on
as
Pu
tid
a 1
74
84
cfu
/mL
Time Days
BHB BHB SALICYLITE BHB GLUCOSE
No Biosurfactant
0 2 4 6 8 10Time Days
5 g/L Biosurfactant
1.00E+06
1.00E+07
1.00E+08
1.00E+09
0 2 4 6 8 10
Pse
ud
om
on
as
Pu
tid
a 1
74
84
cfu
/mL
Time Days
BHB BHB SALICYLITE BHB GLUCOSE
1g/L Biosurfactant
0 2 4 6 8 10Time Days
0.25 g/L Biosurfactant
89
Research by Grimm and Harwood (1997) with Pseudomonas Putida G7 and strain NCIB
9816-4 indicated that salicylate, the compound that directly induces naphthalene
degradation, was also an inducer of naphthalene chemotaxis. This gene along with the
naphthalene degrading genes is encoded in the NAH7 plasmid. Chemotaxis enhances the
ability of motile bacteria, such as Pseudomonas Putida ATCC 17484, to locate and degrade
organic compounds. It is probable that this process can be used by bacteria to move
towards phenanthrene, as a common metabolic pathway exists for the degradation of both
naphthalene and phenanthrene in P.Putida. The improved phenanthrene removal observed
in systems with salicylate did not show a relative increase in live cell concentration (cfu/mL),
however increases in phenanthrene degradation were observed. This can be explained by
the induction of chemotactic behaviour due to salicylate addition. Chemotactic responses
require a concentration gradient of the attractant for a response to occur, and the greater
the concentration gradient the more effective a chemotactic response could be in
enhancing the degradation of the contaminant (Samanta, Singh et al. 2002). Since
biosurfactant did not enhance the degradation of phenanthrene in the presence of
salicylate, this could be due to the increased mass transfer of phenanthrene into the
aqueous phase which decreased the concentration gradient. This would negatively impact
the advantage of chemotactic responses as concentration gradients in the systems were
decreased due to biosurfactant addition. The result of chemotactic attraction can lead to an
increase in phenanthrene bioavailability due to bacteria migrating towards high
concentrations and result in the observed biodegradation rate increasing. The strain of
P.Putida used in this research is capable of attaching to solid surfaces such as soil or solid
phase PAHs and using the nutrients and contaminants directly, indicating that solubility
enhancement of phenanthrene is only one of the mechanisms available to enhance the
90
degradation rate (Dean, Jin et al. 2001). Similar results demonstrating the chemotactic
response have been obtained in diffusion limited systems with naphthalene (Marx and
Aitken 2000; Ortega-Calvo, Marchenko et al. 2003). The positive chemotaxis of P.putida
towards naphthalene due to the metabolic induction by salicylate is most likely extended to
the degradation of phenanthrene, as this may account for the increased removal that is
achieved with salicylate addition.
Results from soil slurry tests are similar to results obtained in aqueous tests (section 4.1.4).
A greater amount of phenanthrene degradation could be achieved through amendments
that specifically induce metabolic pathway induction, when compared to increased solubility
brought about by the addition of biosurfactant, or increased growth brought about by
additional carbon substrates. Biosurfactant addition showed improved phenanthrene
removal when compared to the system with no amendments. Increasing biosurfactant
addition from 0.25 g/L to 5 g/L did not produce a large benefit for the 20-fold increase in
quantity applied. The overall results after 10 days showed nearly complete phenanthrene
removal in most systems; therefore it is not possible to determine the rate at which specific
amendments enhanced the biodegradation in soil slurries. This information would be
important to asses the overall efficiency of each amendment, as some systems could have
achieved the quantity of phenanthrene removal observed in a shorter time frame than the
10 day results studied.
91
4.3 OBJECTIVE 3: COLUMN TESTS
4.3.1 TRACER AND BIOSURFACTANT BREAKTHROUGH CURVE FITTING
The steady-state breakthrough curves for conservative tracer Cl- were analyzed with local
equilibrium convective-dispersive transport model to determine the dispersion (D) and
retardation (R) transport parameters using the computer program STANMOD (Simunek, M.
Th. van Genuchten et al. 2003). STANMOND is a windows based user platform that
incorporates the CXTFIT 2.0 mathematical models to evaluate solute transport in porous
media using analytical solutions to the convection-dispersion solute transport equation
(Toride, Leij et al. 1995). The retardation factor for the rhamnolipid biosurfactant was
determined from the frontal limb of the rhamnolipid breakthrough curve following a
method presented by Noordman et al.(1998).
4.3.1.1 Tracer Breakthrough Curves
Chloride breakthrough curves at a steady-state flow of 0.2 mL/min (pore water velocity =
1.542 cm/h) were fitted using the deterministic equilibrium CDE to determine the simple
hydrodynamic characteristics of the soil columns (Figure 4.10a). The symmetrical
breakthrough
curves obtained with the conservative tracer showed no
physical non-
equilibrium and were used to estimate dispersivity (D), with an average fitted value of 34
cm2/h, with R equal to 1. The chloride breakthrough curves fitted well with the CDE model
parameters, indicating the column system was stable and the hydrodynamic conditions
were similar in separately packed columns.
Chloride breakthrough curves at a high flow of 10 mL/min (pore water velocity = 77.87
cm/h) were fitted using the same deterministic equilibrium CDE to determine the
92
hydrodynamic characteristics of the soil columns (Figure 4.10b). Breakthrough curves
obtained with the conservative tracer showed slight tailing, and the average fitted
dispersion (D) was 749 cm2/h and R equal to 0.9. The curve could not be fitted if R was equal
to 1, and R equal to 0.9 was the best parameter estimate to fit the observed data. This
indicates at high flow rates, some short circuiting of the tracer flow might have occurred at
the column inlet reducing the effective flow length in the soil column. The dispersion to
velocity difference increased the Peclet number to 3.84, compared to 1.68 for the lower
flow rate indicating more advective transport, and less dispersive transport is responsible
for the tracer movement at a high flow rate.
4.3.1.2 Biosurfactant breakthrough
Compared to chloride breakthrough curves, biosurfactant breakthrough produced an
asymmetrical curve. After 10 PV of pumping the biosurfactant concentration in the effluent
was 70% of that in the influent (Figure 4.10c). It is evident from the gradual rising limb of the
biosurfactant breakthrough curve that significant adsorption had occurred, causing
retardation of the biosurfactant flow. Equilibrium CDE models were not able to simulate the
data. A non-equilibrium CDE model assuming one-site chemical adsorption obtained the
best fit to the rising limb of the biosurfactant breakthrough curve. The rapid drop in the
tailing limb could not be fitted to the non-equilibrium model, and data was not collected for
the extent of tailing that occurred for the breakthrough curve, therefore the tailing limb was
not modelled and information about the entire transport process could not be obtained.
Using this method D equalled 34 cm2/h, R equalled 6.59, ω equalled 0.067, and β equalled
0.15.
93
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6
C/C
o
Pore Volume
Cobs(x,t) High Flow
Cfitted(x,t) High Flow
v = 77.87 cm/h
D = 749 cm2/h
R = 0.9
(b)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 5 10 15 20
C/C
o
Pore Volume
Cobs(x,t) Biosurfactant
Cfitted(x,t) Biosurfactant
v = 1.54 cm/h
D = 34 cm2/h
R = 6.59
(c)
Figure 4.10 Observed breakthrough curves and fitted breakthrough curve models
using CXTFIT inverse parameter estimation for (a) chloride with v = 1.54cm/h (b)
chloride with v = 77.87cm/h and (c) biosurfactant with v = 1.54 cm/h
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6
C/C
o
Pore Volume
Cobs(x,t) Low Flow
Cfitted(x,t) Low Flow
v = 1.54 cm/h
D = 34 cm2/h
R = 1.0
(a)
94
4.3.2 MICROBUBBLE DISPERSION FLOW CHARACTERISTICS
Stability of the microbubble dispersions improved with increasing biosurfactant
concentrations from 1 g/L to 5 g/L. The increased stability can be identified by the longer
half drainage times (data not shown) and the increased gas hold-up, which describes the
fraction of the microbubble dispersion that is in gas phase versus liquid phase. Based on the
collection of effluent from the column during the microbubble pumping experiments there
was an average of 30% liquid, and 70% gas breakthrough from the 1 g/L biosurfactant
microbubble dispersion, and 24% liquid, and 76% gas breakthrough from the 5g/L
biosurfactant microbubble dispersion.
Visual observations of the microbubble dispersion saw separation of the dispersion into
liquid and gas phase directly after injection into the column. There were only a few short
instances over the four hours of pumping that visible microbubbles could be observed a few
centimetres beyond the inlet of the column. This could be due to a variety of factors
including: microbubble stability; low flow rate of the microbubble dispersion versus half-life
of the suspension; interaction of the microbubble dispersion with dissolved organic matter;
and the sorption of surfactant onto the soil. In a separate test microbubble dispersions were
pumped at a high flow rate (1PV in 3 minutes) through a small (volume = 10 cm3) column
containing clean washed medium grain sand. The microbubble dispersion transported
through the column without separating into liquid and gas phases and microbubbles were
collected in the effluent. Oliveira et al. (2004) determined the presence of fine particles and
hydrophobic fine particles interfere with the foam breaking process, and with the solution
foaming ability. Sorption of biosurfactant also leads to a reduction of the surfactant
concentration which in turn would decrease the foam stability (Oliveira, Oliveira et al. 2004).
95
Increasing surfactant concentration during continuous biosurfactant pumping initially
increases the adsorption of biosurfactant on the particle surfaces rendering them less
hydrophobic and would decrease the foam breaking ability over time
The effluent liquid flow rate occurred at a constant rate throughout the experiment. The
liquid front was observed to be moving slower than the gas phase, as gas breakthrough in
the effluent would occur within minutes of microbubble injection. To determine the liquid
flow characteristics a conservative tracer (chloride) was added to the microbubble
dispersion during generation and effluent was collected to determine the breakthrough
curve for the liquid fraction of the microbubble dispersion. The pumping flow rate was set
to 10cm3/min per minute, which describes the flow rate of the microbubble dispersion
including both the gas and liquid phases. The equivalent liquid flow rate based on the
quantity of liquid present in the microbubble dispersion was 2.37 mL/min for the 5 g/L
microbubble dispersion. Using the liquid fraction flow rate, data was modelled for the 5g/L
microbubble dispersion with the equilibrium CDE and R equall to 1.37, and D equall to 32.1
cm2/h producing the best model (Figure 4.11).
Visualisation of microbubble dispersion flow patterns by Choi et al. (2008) in sand media
showed chemical surfactant sodium dodecyl sulphate microbubble suspensions separated
into liquid and gas phase directly after injection. The liquid phase showed faster movement,
whereas the gas from of the microbubble suspension flowed in a plug-flow manner (Park,
Choi et al. In Press). The liquid fraction did not advance ahead of the gas phase in these
experiments, but the flow was in a low dispersion (D = 32.1 cm2/h) plug-flow manner. The
gas fraction was considered an immobile fraction (decreasing the relative permeability) and
96
characteristic break through curves were obtained (Figure 4.11). This behaviour is consistent
with foam drainage in porous media where flow of the liquid fraction is through the
interstices of the immobile foam (Zitha, Nguyen et al. 2006). Zitha et al. (2006) concluded
the role of trapped gas during foam flow was not as important as it might have been
believed previously by most foam flow researchers. The microbubble dispersions in this
research acted more like standard foam where the process of breakage of the foam and
regeneration caused a relatively large decrease in liquid mobility due to foam development
(Chowdiah, Misra et al. 1998).
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
C/C
o
Pore Volume
Cobserved(x,t) microfoam 1 g/L
Cobserved(x,t) microfoam 5 g/L
Cfitted(x,t) microbublle dispersion 5g/L
v = 18.4 cm/h
D = 32 cm2/h
R = 1.37
Figure 4.11 Microbubble dispersion breakthrough curve with a conservative tracer in the liquid
fraction.
97
4.3.2 PRESSURE DROP ASSOCIATED WITH MICROBUBBLE DISPERSION PUMPING
The flow propagation of the microbubble dispersion was accompanied by a large pressure
drop greater than 50 kPa after 10 pore volumes pumping with 5 g/L biosurfactant
microfoam (Figure 4.12, Figure 4.13, & Figure 4.14). The results from the pumping show the
increase in biosurfactant concentration from 1g/L to 5g/L cause pressure build up in the soil
after 10 PV to be nearly 40% higher at the inlet. This is possibly due to the increased stability
and stronger films that make up foam as the biosurfactant concentration increased. This
creates more effective displacing or blocking agents in the porous media which eventually
creates pressure build up in the soil column. A variety of measures including foam quality,
flow rate, and surfactant type have shown to greatly influence the magnitude of the
pressure drop that is associated with foam flow. With standard foam it is assumed the
presence of the gas phase makes foams more compressible. Higher quality foam will have
larger bubbles and thinner liquid films which causes the pressure gradients in soil to
decrease as the foam quality is increased (Chowdiah, Misra et al. 1998).
Pressure profiles (Figure 4.13; Figure 4.14) demonstrate the pressure gradient due to the
presence of foam at the inlet was significantly higher than the pressure in the remainder of
the column. The pressure drop was significantly lower from 17cm depth to the surface, and
this small pressure drop corresponds to single phase water flow, versus larger pressure
drops creted by foam flow (Apaydin and Kovscek 2001). The retarded flow of the
microbubble dispersion resulted in larger pressure drops which increased over the duration
of the experiment. When the pressure drop builds up in a channel, the foam flows into less
accessible spill areas. This pressure dependent “clogging” process indicates that channelling
or poor sweep efficiency should not occur, allowing microbubbles to overcome subsurface
98
heterogeneity. This process also causes increases in pressure that can be undesirable if to
large (Riser-Roberts 1998). It is considered necessary to maintain the pressure drop below
22.6 kPa/m to prevent mechanisms such as soil heaving occurring due to the injection of
foam at high pressure (Chowdiah, Misra et al. 1998).
0
10
20
30
40
50
60
0 2 4 6 8 10 12
Pre
ssu
re (
kP
a)
Pore Volumes
Biosurfactant 1g/L solution: Flow = 10 mL/min
Inlet -37cm
-34.5cm
-32cm
-29.5cm
-27cm
-17cm
Figure 4.12 Pressure distribution across the length of the soil column during biosurfactant (1
g/L) solution pumping. Data presented corresponds to depth in the column with the highest
pressure at the inlet, and the lowest pressure at -17cm assuming the outlet is the datum.
99
0
10
20
30
40
50
60
0 1 2 3 4 5 6 7 8 9 10 11 12
Pre
ssu
re k
Pa
Pore Volumes
Biosurfactant 5 g/L microfoam: Flow =
10cm3/minInlet -37cm
-34.5cm
-32cm
-29.5cm
-27cm
-17cm
0
10
20
30
40
50
60
0 1 2 3 4 5 6 7 8 9 10 11 12
Pre
ssu
re k
Pa
Pore Volumes
Biosurfactant 1 g/L microfoam: Flow =
10cm3/min
Inlet -37cm
-34.5cm
-32cm
-29.5cm
-27cm
-17cm
Figure 4.14 Pressure distribution across the length of the soil column during biosurfactant
microfoam ( 5g/L) pumping. Data presented corresponds to depth in the column with the highest
pressure at the inlet, and the lowest pressure at -17cm assuming the outlet is the datum.
Figure 4.13 Pressure distribution across the length of the soil column during biosurfactant
microfoam (1 g/L) pumping. Data presented corresponds to depth in the column with the highest
pressure at the inlet, and the lowest pressure at -17cm assuming the outlet is the datum.
100
4.3.3 BIODEGRADING TRIALS IN CONTINUOUS FLOW SYSTEMS
Biodegradation trials in pre-inoculated soil columns were operated as a continuous flow
system over a 10 day period. Soil profiles of the remaining phenanthrene in soil over the
column depth indicate phenanthrene mobilization and transport due to the upward flow of
biosurfactant solution (Figure 4.15; Figure 4.16). Inlet concentrations of phenanthrene were
up to 2-fold lower than the concentration remaining in the upper half of the soil column.
Significant dissolved organic matter was present in the effluent as TOC readings averaged
near 50 mg/L, and there was a visible brown colour to the effluent samples.
-40
-35
-30
-25
-20
-15
-10
-5
0
0 50 100 150 200 250 300 350 400 450 500
De
pth
in
co
lum
n (
cm)
Phenanthrene remaining in soil (mg/kg)
Column 1
Column 2
0
5
10
15
20
25
30
35
40
45
0 50 100 150 200 250
Ph
en
an
the
re m
g/L
Time Hours
Phenanthrene in Efluent
Figure 4.15 Trial 1 phenanthrene distribution in soil column after 10 days continuous upflow at
0.2 mL/min , phenanthrene in column effluent over 10 days. Column 1 influent solution
biosurfactant 1g/L + salicylate 100 mg/L; Column 2 influent solution biosurfactant 1 g/L. Soil
distribution assuming effluent (0cm) is the top of the column and influent (-37 cm) is the bottom
of the column.
101
Four pore volumes of continuous biosurfactant pumping occurred before notable amounts
of phenanthrene were detected in the effluent solution (Figure 4.15). After 4 days the
amount of phenanthrene detected in the effluent significantly increased to over 30 mg/L.
This indicates that biosurfactant sorption to the soil was occurring for the first 4 PVs and not
contributing to an increase in the apparent aqueous phase phenanthrene concentration.
Depending on the quantity of biosurfactant added to the system, the sorbed-phase
biosurfactant can account for the majority of the added biosurfactant (Laha, Tansel et al. In
Press). The result of this process indicates there is increased partitioning of phenanthrene
-40
-35
-30
-25
-20
-15
-10
-5
0
0 50 100 150 200 250 300 350 400 450
De
pth
in
co
lum
n (
cm)
Phenanthrene remaining in soil (mg/kg)
Column 1
Column 2
Figure 4.16 Trial 2 phenanthrene distribution in soil column after 10 days continuous upflow
with BHB broth at 0.5mL/min; phenanthrene in column effluent over 10 days. Column 1
influent pulse solution biosurfactant microfoam 1 g/L + salicylate 100mg/L; Column 2 influent
pulse solution biosurfactant 1 g/L. Soil distribution assuming effluent (0cm) is the top of the
column and influent (-37 cm) is the bottom of the column.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 48 96 144 192 240
Ph
en
an
the
re m
g/L
Time Hours
Phenanthrene in Efluent
Pulse injections
102
onto soil, until the biosurfactant is at a high enough concentration above the CMC in the
aqueous phase to promote phenanthrene desorption from the surfactant sorbed soil (Laha,
Tansel et al. In Press). The amount of biosurfactant sorption occurring for the first 4 days
most likely decreased the overall biosurfactant concentration below the CMC. After this
period the rate of sorption to the soil decreased and the concentration became high enough
in solution to exceed the CMC. The amount of phenanthrene in the effluent samples from
continuous pumping without biosurfactant (Figure 4.16) maintained a relatively constant
aqueous phenanthrene concentration near 1mg/L with some noted increases to 1.2mg/L
after pulse injection of both the microbubble dispersion and biosurfactant solution.
It has been demonstrated that the application of surfactant enhanced soil washing or
flushing of PAH contaminants generally does occur until after the sorption of surfactant on
the soil reaches saturation (Zhou and Zhu 2007; Zhu and Zhou 2008). No advantageous
results are demonstrated until after the sorption of surfactant on the soil reaches
saturation. This effect was evident in the biosurfactant pumping trials, indicating that higher
flow rate would be beneficial in the initial stages of a biosurfactant enhanced remediation
process to speed up the sorption process. The adsorption and desorption process appears
to be a non-equilibrium process where two-site sorption models can account for the
sorption behaviour of both PAHs and surfactants (Noordman 1999; Chen, Wang et al. 2006).
In surfactant flushing experiments, after biosurfactant sorption reaches saturation, the
removal of fast desorbing phenanthrene fractions can occur quickly, and the slow desorbing
fractions could be efficiently removed at very low purging rates (Schlebaum, Schraa et al.
1999). Overall, results of this study demonstrate that surfactant sorption to the solid phase
can lead to increases in phenanthrene retardation when biosurfactant concentrations are
103
low. This effect would be desirable if the treatment objective was to immobilize
phenanthrene; however, the effect is undesirable in surfactant enhanced phenanthrene
removal applications (Ko, Schlautman et al. 1998).
Table 4.3 Total percentage removal of phenanthrene due to soil flushing and
biodegradation in soil column tests after 10 days continuous flow.
Trial 1: Continuous Biosurfactant Flushing (0.2mL/min)
Column 1: biosurfactant and salicylate Column 2: biosurfactant only
Soil Flushing 10.8% ± 0.1% 7.9% ± 0.1%
Biodegradation -0.7% ± 7.4 % 10.5% ± 6.9%
Biodegradation/hr --- 0.20 mg phenanthrene/hr
Total Removal 10.1% ± 5.4% 18.4% ± 4.8%
Trial 2: Biosurfactant and Salicylate Pulse Injection (0.5mL/min)
Column 1: microfoam pulse Column 2: liquid pulse
Soil Flushing 1.4% ± 0.1% 1.4% ± 0.1%
Biodegradation 27.8% ± 7.7% 21.6% ± 7.2%
Biodegradation/hr 0.51 mg phenanthrene/hr 0.40 mg phenanthrene/hr
Total Removal 29.3% ± 5.8% 23.0% ± 5.2%
Mass balance calculations were performed to determine the amount of biodegradation that
occurred (Table 4.3). The total amount of biodegradation was calculated by taking the total
initial phenanthrene amount, subtracted by the total remaining phenanthrene after 10 days
and the total amount of phenanthrene that was present in the effluent due to soil flushing.
With an average of 5% error on soil extractions (both to obtain the initial soil contamination,
and to obtain the total remaining concentration) the error becomes over 7% when the mass
balance calculations are performed. This shows accuracy of the soil extraction method
strongly influences the overall accuracy of the results obtained. However flushing data
obtained on the HPLC had very little error as phenanthrene is already in the aqueous phase
and easily quantified. This highlights the importance in determining accurate soil extraction
methods to obtain statistically significant results, and the importance conducting tests in
triplicate or greater in order to report experimental results with certainty.
104
The results from the second column trial showed cell elution from column 1 averaging an
order of magnitude larger than column 2 with an average of 5x107, versus 5x10
6 cfu/mL
respectively, for samples taken from the effluent at the same time as phenanthrene
measurements. The final amount of cells in the soil when unpacking was completed
followed the same trend, where there was approximately one order of magnitude more
cells present in column 1. Significant cell elution was observed by Yolcubal et al.(2002) in
sand column tests which increased in the presence of substrate in the influent solution.
These results suggest that experiments which exhibited the greatest degradation were
associated with the production of new cells both in the soil and in the aqueous phase.
Degradation rates in the soil columns were much lower than phenanthrene degradation in
soil slurry batch tests indicating that contaminant mass transfer and quantity of available
oxygen would be responsible for the differences observed. Park et al. (2001) determined
that batch systems degradation followed zero order kinetics and was independent of the
concentration range, while column tests exhibited decreased rates at concentrations less
than 100 mg/L of naphthalene. Half-saturation constant Ks, as described by Michaelis-
Mention and Monod growth kinetics (Equation 2.1) were elevated in column tests due to
length dependent transfer of substrate to cell surfaces in column tests. Whereas kinetic
parameters in well mixed batched systems create shorter and more uniform mass transfer
distances (Park, Zhao et al. 2001). This can also be explained using non-equilibrium
conditions present in diffusion control sorption process. Vigorous shaking allows diffusion to
generally be eliminated as a rate-limiting step in batch studies (Nielsen, Van Genuchten et
al. 1986). In soils with flowing water, the sorption rate can be limited by the rate at which
ions are transported to the exchange sites, and can be determined by physical non-
105
equilibrium models (Nielsen, Van Genuchten et al. 1986). Physical non-equilibrium models
use the same formulation as the chemical non-equilibrium models presented, (Equation 2.4
& 2.5) but two-site partitioning in soil water phases are defined as mobile (flowing) and
stagnant (immobile) phases (Nielsen, Van Genuchten et al. 1986). This assumes the
convective dispersive transport is confined to the faction of the liquid filed mobile pores,
and the stagnant non moving liquid decreases the overall mass transfer rate (Nielsen, Van
Genuchten et al. 1986).
Oxygen appears to have been a rate limiting amendment in soil columns tests. The
increased flow rate and microfoam pulse increased the total amount of degradation (Table
4.3). The system with microfoam pulse injection saw 27.8% degradation versus 21.6% in the
same setup with a liquid pulse instead of a microfoam pulse injection every 48 hours. The
low flow rate saw only 10.5 % degradation with continuous biosurfactant supply and no
degradation with biosurfactant and salicylate addition. These results are contrary to what
were expected based on preliminary batch soil slurry tests results (Section 4.2.2). The
explanation for the observed biodegradation rates can be explained assuming oxygen was
the limiting amendment. Added salicylate at a low flow rate in the aqueous phase would
have been preferentially used by the bacteria as an available carbon source. This amount of
added bacterial activity due to the salicylate degradation would have decreased the oxygen
concentration prior to any phenanthrene degradation, resulting in no phenanthrene
degradation due to oxygen limitations. The increased flow rate in the second trial, delivered
increased oxygen to the system which resulted in higher total amounts of degradation, and
microfoam addition further increased the added oxygen to the system resulting in 27.8%
degradation.
106
CHAPTER 5 CONCLUSIONS AND RECOMENDATIONS
The focus of this research was to enhance in situ biodegradation of phenanthrene in soil. In
order to determine why specific physical variations have been shown to increase (or
decrease) overall contaminant biodegradation, bench-scale systems were created to
quantify the effects of rhamnolipid biosurfactant, salicylate, glucose, and biosurfactant
microbubble dispersions on the overall biodegradation process. The conclusions of these
experiments are briefly described below.
Liquid Cultures
• Higher levels of phenanthrene degradation can be achieved through amendments
that target metabolic pathway induction, as phenanthrene removal was doubled
upon addition of salicylate, as compared to the addition of glucose.
• The addition of rhamnolipid biosurfactant increases the apparent aqueous solubility
of phenanthrene, and augmented the amount of degradation in all systems in
combination with added salicylate and glucose.
Soil Slurries
• Biosurfactant addition caused competition between micellar and sorbed
biosurfactants for phenanthrene partitioning. The addition of rhamnolipid
biosurfactant above the CMC enhanced the desorption of phenanthrene from soil,
decreasing Kd over 8-fold from the addition of 1 g/L biosurfactant.
• Biosurfactant additions improved biodegradation by reducing both the aqueous
phase phenanthrene, and total remaining phenanthrene in soil slurries with at least
86% total phenanthrene removal in systems containing 0.25g/L biosurfactant versus
67% removal without biosurfactant .
107
• With salicylate present, the effects of biosurfactant were almost nonexistent as
greater than 90% removal occurred in all systems containing salicylate regardless of
the biosurfactant concentration. Although bioavailability is commonly perceived as
the rate limiting step, an in situ enhancement strategy does not need to increase
bioavailability via increased aqueous solubility by adding a biosurfactant (Shin, Kim
et al. 2004).
• The positive chemotaxis of P.putida towards naphthalene, due to the metabolic
induction by salicylate, is most likely extended to the degradation of phenanthrene,
as this accounts for the increased removal achieved with salicylate present.
• Chemotactic attractive behaviour of P.putida towards phenanthrene leads to an
increase in phenanthrene bioavailability. As a result, the observed biodegradation
rates increased.
Column Systems
• Non-equilibrium CDE models assuming one-site chemical adsorption obtained the
best fit to the rising limb of the biosurfactant breakthrough curve. Adsorption and
desorption appears to be a non-equilibrium process where two site sorption models
in CXTFIT can account for the sorption behaviour of both PAHs and surfactants
(Noordman 1999; Chen, Wang et al. 2006).
• Degradation rates in the soil columns were much lower than phenanthrene
degradation in soil slurry batch tests. Enhancement strategies that increased delivery
of oxygen to the system result in higher total amounts of degradation. Other
amendments have little effect when oxygen is limiting the system.
108
Understanding the metabolic pathways and the enzymatic reactions utilized by
Pseudomonas putida ATCC 17484 during contaminant breakdown yields information about
the induction ability of salicylate. Liquid culture experiments were relatively simple and less
time consuming, and provided good preliminary data to asses the viability of amendments.
Soil organic matter affected the efficiency of biosurfactant addition due to partitioning
processes, demonstrating the importance of maintaining a biosurfactant concentration in
solution above the CMC to enhance the overall phenanthrene desorption from the soil.
Results from these trials provide further evidence to the importance of determining the
metabolic processes that are responsible for successful in situ bioremediation, as salicylate
proved to be the best amendment in non-oxygen limiting environments.
109
5.1 RECOMMENDATION FOR FUTURE WORK
Due to the observations of microbubble dispersion separation into liquid and gas phases
upon injection, it is apparent that microbubble dispersion flow is dependent on the
conditions of the soil and the stability of the microfoam generated. Microbubble stability,
low flow rates of dispersion, half life of the microbubble suspension, interactions between
the microbubble dispersion and the dissolved organic matter, as well as sorption of
surfactant to soil could negatively impact the effectiveness of microbubble dispersions as
amendments to in situ bioremediation. Further studies detailing the microbubble dispersion
flow in simple sand, versus systems with higher amounts of organic materials and fine
particulates, would help determine the efficiency of microbubble dispersions.
A limitation to the studies performed in this research was the continuous operation of the
microbubble generating system. The spinning disk motor could not be run for longer than
four hours without overheating, which limited the scope of the research trials significantly.
Improved methods for generating the microbubble dispersion, such as methods that
encapsulate the spinning disk apparatus inside a pressure vessel have been shown to create
microfoam that is more stable and can be used in continuous operation (Wan, Veerapaneni
et al. 2001). Microbubble generation and injection under pressure have been shown to
minimize microbubble loss due to gas dissolution (Wan, Veerapaneni et al. 2001). A
comparison of Generation methods of sonication versus mechanical agitation showed that
the generation method greatly affected the size of microbubbles. Therefore the use of a
suitable generation method should be an important consideration for future work (Xu,
Nakajima et al. 2008).
110
The length of time required for effective biodegradation experiments in soil (and application
in the field) is difficult to determine. Although batch tests showed steady phenanthrene
removal, scaling up to column tests requires longer project timelines depending on the
amendments made. This constrains the number of trials that can be run, the number of
variables that can be monitored, and therefore the amount of data that can be collected
obtained. Developing or using existing modelling software is an important tool to simulate
transport parameters, estimate biodegradation rates, and determine effective treatment
strategies. It would also be beneficial to characterize microbubble flow based on the
predicted pressure build up in the soil. This would determine oxygen delivery, and could be
used to determine the amount of oxygen available for uptake by bacteria.
The overall results in soil slurries after 10 days of biodegradation showed nearly complete
phenanthrene removal in most systems. Therefore it is not possible to determine the rate
at which each amendment enhanced the biodegradation in soil slurries. This info is essential
to asses the overall efficiency of each amendment, as some systems may have achieved the
same quantity of phenanthrene removal in shorter time frames. The recent development of
an HPLC method to accurately determine rhamnolipid biosurfactant concentration will also
be an important tool in future research, allowing the quantification of biosurfactant
sorption and its utilization by microbial communities.
Contaminant bioavailability can be species-specific, with different microbial strains capable
of accessing different contaminant pools in the soil-water system. Thus, understanding the
interactions between microbes will provide further opportunities for enhanced
biodegradation of bioavailable contaminants (Dean, Jin et al. 2001). The recent acquisition
111
of a variety of PAH degrading bacterial strains (graciously donated by Landcare Research)
which were characterized from contaminated sites in New Zealand, will allow for the
development of more complex systems. This will enable the study of interactions between a
variety of bacteria and their model environments. Research with different strains to
determining metabolic influences will be an important tool to use, and gain an
understanding of how to control subsurface environments with amendments that target
known metabolic pathway induction.
Work is underway to develop fluorescent protein reporter organisms capable of degrading
target contaminants. This will allow population growth dynamics to be monitored in situ
with a fiberoptic spectroscopic probe system with fluorescence linked to nah operon
promoter activity. This technique will offer unique opportunities to monitor bacterial
growth characteristics in a non-destructive manner which is linked to salicylate
mineralization.
The results of this research indicate that a sound understanding of the microbial processes
that occur, aid in determining the most efficient strategy to enhance the in situ
biodegradation process. A better understanding of metabolic pathways and inducers for
degradation of higher-ringed PAHs for more bacterial strains in the environment will result
in the identification of rate-limiting steps in the process. This will help future researchers
target specific enhancement strategies to make in situ biodegradation more reliable and
efficient. Microbubble dispersions show promise for effective oxygen delivery to the
subsurface and will help to extend the feasibility of in situ bioremediation to more complex
systems.
112
CHAPTER 6 WORKS CITED
Aitken, M. D., W. T. Stringfellow, et al. (1998). "Characteristics of phenanthrene-degrading bacteria
isolated from soils contaminated with polycyclic aromatic hydrocarbons." Canadian Journal
of Microbiology 44(8): 743-752.
Allen, C. C. R., D. R. Boyd, et al. (1999). "Contrasting Effects of a Nonionic Surfactant on the
Biotransformation of Polycyclic Aromatic Hydrocarbons to cis-Dihydrodiols by Soil Bacteria."
Appl. Environ. Microbiol. 65(3): 1335-1339.
Apaydin, O. G. and A. R. Kovscek (2001). "Surfactant Concentration and End Effects on Foam Flow in
Porous Media." Transport in Porous Media 43(3): 511-536.
Atlas, R. and J. Philip (2005). Bioremediation - applied microbial solutions for real-world
environmental cleanup, ASM press.
Auckland City Council. (2007). "Urban Soils Review." Retrieved December 21, 2008, from
http://www.aucklandcity.govt.nz/council/documents/soilquality/default.asp.
Avramova, T., A. Sotirova, et al. (2008). "Effect of Triton X-100 and rhamnolipid PS-17 on the
mineralization of phenanthrene by Pseudomonas sp. cells." International Biodeterioration &
Biodegradation 62(4): 415-420.
Basu, A., D. Das, et al. (In Press). "Sequential utilization of substrates by Pseudomonas putida CSV86:
Signatures of intermediate metabolites and online measurements." Microbiological
Research In Press, Corrected Proof.
Buco, S., M. Moragues, et al. (2004). "Analysis of polycyclic aromatic hydrocarbons in contaminated
soil by Curie point pyrolysis coupled to gas chromatography-mass spectrometry, an
alternative to conventional methods." Journal of Chromatography A 1026(1-2): 223-229.
Cameotra, S. S. and J.-M. Bollag (2003). "Biosurfactant-Enhanced Bioremediation of Polycyclic
Aromatic Hydrocarbons." Critical Reviews in Environmental Science and Technology 30(2):
111-126.
Cerniglia, C. E. (1993). "Biodegradation of polycyclic aromatic hydrocarbons." Current Opinion in
Biotechnology 4(3): 331-338.
Chauhan, A., Fazlurrahman, et al. (2008). "Bacterial metabolism of polycyclic aromatic hydrocarbons:
strategies for bioremediation." Indian Journal of Microbiology 48(1): 95-113.
Chen, J., X. J. Wang, et al. (2006). "Modeling surfactant las influenced PAHs migration in soil
column." Water, Air, and Soil Pollution 176(1-4): 217-232.
Chen, S.-H. and M. D. Aitken (1999). "Salicylate Stimulates the Degradation of High-Molecular
Weight Polycyclic Aromatic Hydrocarbons by Pseudomonas saccharophila P15."
Environmental Science & Technology 33(3): 435-439.
Cho, H.-H., J. Choi, et al. (2002). "Combined Effect of Natural Organic Matter and Surfactants on the
Apparent Solubility of Polycyclic Aromatic Hydrocarbons." J Environ Qual 31(1): 275-280.
Cho, H. Y., H. W. Seung, et al. (2006). "Effects of Intermediate Metabolites on Phenanthrene
Biodegradation." Journal of Microbiology and Biotechnology 16(6): 969-973.
Choi, Y. J., J. Y. Park, et al. (2008). "Flow Characteristics of Microbubble Suspensions in Porous Media
as an Oxygen Carrier." CLEAN - Soil, Air, Water 36(1): 59-65.
Chowdiah, P., B. R. Misra, et al. (1998). "Foam propagation through soils for enhanced in-situ
remediation." Journal of Hazardous Materials 62(3): 265-280.
Couto, H. J. B., G. Massarani, et al. (In Press). "Remediation of sandy soils using surfactant solutions
and foams." Journal of Hazardous Materials In Press, Corrected Proof.
Dean, S. M., Y. Jin, et al. (2001). "Phenanthrene Degradation in Soils Co-Inoculated with
Phenanthrene-Degrading and Biosurfactant-Producing Bacteria." J Environ Qual 30(4): 1126-
1133.
Doyle, E., L. Muckian, et al. (2008). Chapter 2 Microbial PAH Degradation. Advances in Applied
Microbiology, Academic Press. Volume 65: 27-66.
113
Eaton, R. W. and P. J. Chapman (1992). "Bacterial metabolism of naphthalene: Construction and use
of recombinant bacteria to study ring cleavage of 1,2-dihydroxynaphthalene and subsequent
reactions." Journal of Bacteriology 174(23): 7542-7554.
Edwards, D. A., Z. Adeel, et al. (1994). "Distribution of Nonionic Surfactant and Phenanthrene in a
Sediment/Aqueous System." Environmental Science & Technology 28(8): 1550-1560.
García-Falcón, M. S., B. Cancho-Grande, et al. (2004). "Stirring bar sorptive extraction in the
determination of PAHs in drinking waters." Water Research 38(7): 1679-1684.
Garcia-Junco, M., C. Gomez-Lahoz, et al. (2003). "Biosurfactant- and biodegradation-enhanced
partitioning of polycyclic aromatic hydrocarbons from nonaqueous-phase liquids."
Environmental Science and Technology 37(13): 2988-2996.
Gómez, J., M. T. Alcántara, et al. (In Press). "A two-stage process using electrokinetic remediation
and electrochemical degradation for treating benzo[a]pyrene spiked kaolin." Chemosphere
In Press, Corrected Proof.
Grimm, A. C. and C. S. Harwood (1997). "Chemotaxis of Pseudomonas spp. to the polyaromatic
hydrocarbon naphthalene." Appl. Environ. Microbiol. 63(10): 4111-4115.
Guerin, W. F. and S. A. Boyd (1995). "Maintenance and induction of naphthalene degradation
activity in Pseudomonas putida and an Alcaligenes sp. under different culture conditions."
Appl. Environ. Microbiol. 61(11): 4061-4068.
Guiyun, B., M. L. Brusseau, et al. (1998). "Influence of cation type, ionic strength, and pH on
solubilization and mobilization of residual hydrocarbon by a biosurfactant." Journal of
Contaminant Hydrology 30(3-4): 265-279.
Harvey, R. G. (1991). Polycyclic Aromatic Hydrocarbons: Chemistry and Carcinogenicity. Cambridge,
Cambridge University Press.
Hemond, H., F and E. Fechner-Levy, J (2000). Chemical Fate and Transport in the Environment.
London, Academic Press.
Hendricks, D. W. (2006). Water Treatment Unit Processes: Physical and Chemical, CRC Press.
Huang, C. P. and D. Cha (2001). Desorption of PAHs in the Presence of Surfactant. Electrochemical
Processes for in-situ Treatment of Contaminated Soils. Newark, US Department of Energy.
Hugouvieux-Cotte-Pattat, N., T. Kohler, et al. (1990). "Growth-phase-dependent expression of the
Pseudomonas putida TOL plasmid pWW0 catabolic genes." J. Bacteriol. 172(12): 6651-6660.
Institute for Ecology of Industrial Areas (1999). Bioremediation of Petroleum Hydrocarbon-
contaminated Soils. Department of Energy USA. Katowice, Poland.
Jafvert, C. T. and J. K. Heath (1991). "Sediment- and saturated-soil-associated reactions involving an
anionic surfactant (dodecylsulfate). 1. Precipitation and micelle formation." Environmental
Science & Technology 25(6): 1031-1038.
Jauregi, P. and J. Varley (1999). "Colloidal gas aphrons: potential applications in biotechnology."
17(10): 389-395.
Jin, D., X. Jiang, et al. (2007). "Effects of concentration, head group, and structure of surfactants on
the degradation of phenanthrene." Journal of Hazardous Materials 144(1-2): 215-221.
Johnsen, A. R., L. Y. Wick, et al. (2005). "Principles of microbial PAH-degradation in soil."
Environmental Pollution 133(1): 71-84.
Jonsson, S., Y. Persson, et al. (2007). "Degradation of polycyclic aromatic hydrocarbons (PAHs) in
contaminated soils by Fenton's reagent: A multivariate evaluation of the importance of soil
characteristics and PAH properties." Journal of Hazardous Materials 149(1): 86-96.
Josephy, D. and B. Mannervik (2006). Chapter18 Polycyclic Aromatic Hydrocarbon Carcinogens.
Molecular Toxicology, Oxford University Press: 589.
Jun, O. (2008). Phenanthrene Degradation. Biocatalysis/Biodegradation Database, University of
Minnestoa.
Kim, I. S., J.-S. Park, et al. (2001). "Enhanced biodegradation of polycyclic aromatic hydrocarbons
using nonionic surfactants in soil slurry." Applied Geochemistry 16(11-12): 1419-1428.
114
Kiyohara, H., S. Torigoe, et al. (1994). "Cloning and characterization of a chromosomal gene cluster,
pah, that encodes the upper pathway for phenanthrene and naphthalene utilization by
Pseudomonas putida OUS82." J. Bacteriol. 176(8): 2439-2443.
Ko, S.-O., M. A. Schlautman, et al. (1998). "Partitioning of Hydrophobic Organic Compounds to
Sorbed Surfactants. 1. Experimental Studies." Environmental Science & Technology 32(18):
2769-2775.
Kraaij, R. H., S. Ciarelli, et al. (2001). "Bioavailability of lab-contaminated and native polycyclic
aromatic hydrocarbons to the amphipod Corophium volutator relates to chemical
desorption." Environmental Toxicology and Chemistry 20(8): 1716-1724.
Kwok, C. K. and K. C. Loh (2003). "Effects of Singapore soil type on bioavailability of nutrients in soil
bioremediation." Advances in Environmental Research 7: 889-900.
Labana, S., K. Manisha, et al. (2007). Diversity, Biodegradation and Bioremediation of Polycyclic
Aromatic Hydrocarbons. Environmental Bioremediation Technologies. S. N. Singh and R. D.
Tripathi. Berlin, Springer Press: 409-443.
Laha, S., B. Tansel, et al. (In Press). "Surfactant-soil interactions during surfactant-amended
remediation of contaminated soils by hydrophobic organic compounds: A review." Journal of
Environmental Management In Press, Corrected Proof.
Landrum P, Eadie B, et al. (1992). "Variation in the bioavailability of polycyclic aromatic
hydrocarbons to the amphipod Diporeia spp. with sediment aging." Environmental
Toxicology and Chemistry 11(8): 1197-1208.
Larmignat, S., D. Vanderpool, et al. (2008). "Rheology of colloidal gas aphrons (microfoams)."
Colloids and Surfaces A: Physicochemical and Engineering Aspects 322(1-3): 199-210.
Lee, D. S., M. W. Lee, et al. (2005). "Effects of salicylate and glucose on biodegradation of
phenanthrene by Burkholderia cepacia PM07." Journal of Microbiology and Biotechnology
15(4): 859-865.
Lee, K., J.-W. Park, et al. (2003). "Effect of additional carbon source on naphthalene biodegradation
by Pseudomonas putida G7." Journal of Hazardous Materials 105(1-3): 157-167.
Lee, P.-H., S. K. Ong, et al. (1999). "Extraction Method for Analysis of PAHs in Coal-Tar-Contaminated
Soils." Practice Periodical of Hazardous, Toxic, and Radioactive Waste Management 3(4):
155-162.
Leigh, D., D. Jerger, et al. (1997). Injetion of Surfactant/Ozygen Microbubbles as a Delivery System
for In Situ Bioremediation. 4th International in Situ and on-Site Bioremediation Symposium.
B. Alleman, C and A. Leeson. New Orleans, LA, Battle Press. 2: 603 - 608.
Lin, S.-C. (1996). "Biosurfactants: Recent advances." Journal of Chemical Technology & Biotechnology
66(2): 109-120.
Makkar, R. S. and K. J. Rockne (2003). "Comparison of synthetic surfactants and biosurfactants in
enhancing biodegradation of polycyclic aromatic hydrocarbons." Environmental Toxicology
and Chemistry 22(10): 2280-2292.
Marx, R. B. and M. D. Aitken (2000). "Bacterial Chemotaxis Enhances Naphthalene Degradation in a
Heterogeneous Aqueous System." Environmental Science & Technology 34(16): 3379-3383.
Mohan, S., T. Kisa, et al. (2006). "Bioremediation technologies for treatment of PAH-contaminated
soil and strategies to enhance process efficiency." Reviews in Environmental Science and
Biotechnology 5(4): 347-374.
Mulder, H., A. M. Breure, et al. (2001). "Application of a mechanistic desorption-biodegradation
model to describe the behavior of polycyclic aromatic hydrocarbons in peat soil aggregates."
Chemosphere 42(3): 285-299.
Mulligan, C. N., R. N. Yong, et al. (2001). "Surfactant-enhanced remediation of contaminated soil: a
review." Engineering Geology 60(1-4): 371-380.
Myers, D. (1988). Surfactant Science and Technology. New York, VCH Publishers Inc.
New Zealand Ministry for the Environment (2007). Environment New Zealand 2007. Wellington.
115
Nielsen, D. R., M. T. Van Genuchten, et al. (1986). "Water Flow and Solute Transport Processes in the
Unsaturated Zone." Water Resources Research 22(9S): 89S - 108 S.
Noordman, W., W. Ji, et al. (1998). "Effects of Rhamnolipid Biosurfactants on Removal of
Phenanthrene from Soil " Environ. Sci. Technol. 32(12): 1806 -1812.
Noordman, W. H. (1999). Biosurfactant-enhanced soil remediation : effect of rhamnolipid on the
bioavailability and microbial uptake of hydrophobic compounds. science, University of
Groningen. Phd: 116.
Ogunseitan, O. A. and B. H. Olson (1993). "Effect of 2-hydroxybenzoate on the rate of naphthalene
mineralization in soil." Applied Microbiology and Biotechnology 38(6): 799-807.
Oliveira, R. C. G., J. F. Oliveira, et al. (2004). The effect of hydrophobic fine particles on the foam
flushing remediation process. Surface and Colloid Science: 293-297.
Oostrom, M., J. H. Dane, et al. (2006). "A Review of Multidimensional, Multifluid Intermediate-Scale
Experiments: Nonaqueous Phase Liquid Dissolution and Enhanced Remediation." Vadose
Zone J 5(2): 570-598.
Ortega-Calvo, J. J., A. I. Marchenko, et al. (2003). "Chemotaxis in polycyclic aromatic hydrocarbon-
degrading bacteria isolated from coal-tar- and oil-polluted rhizospheres." FEMS Microbiology
Ecology 44(3): 373-381.
Park, J. H., X. Zhao, et al. (2001). "Comparison of biodegradation kinetic parameters for naphthalene
in batch and sand column systems by Pseudomonas putida." Environmental Progress 20(2):
93-102.
Park, J. Y., Y. J. Choi, et al. (In Press). "Microbubble suspension as a carrier of oxygen and acclimated
bacteria for phenanthrene biodegradation." Journal of Hazardous Materials In Press,
Corrected Proof.
Pastewski, S., E. Hallmann, et al. (2006). "Physicochemical aspects of the application of surfactants
and biosurfactants in soil remediation." Environmental Engineering Science 23(4): 579-588.
Perry, J. J. (1979). "Microbial cooxidations involving hydrocarbons." Microbiol. Mol. Biol. Rev. 43(1):
59-72.
Pieper, D. H. and W. Reineke (2000). "Engineering bacteria for bioremediation." Current Opinion in
Biotechnology 11(3): 262-270.
Pignatello, J. J. and L. S. Donald (1999). The Measurement and Interpretation of Sorption and
Desorption Rates for Organic Compounds in Soil Media. Advances in Agronomy, Academic
Press. Volume 69: 1-73.
Platt, A., V. Shingler, et al. (1995). "The 4-hydroxy-2-oxovalerate aldolase and acetaldehyde
dehydrogenase (acylating) encoded by the nahM and nahO genes of the naphthalene
catabolic plasmid pWW60-22 provide further evidence of conservation of meta-cleavage
pathway gene sequences." Microbiology 141(9): 2223-2233.
Powell, S. N., D. R. Singleton, et al. (2008). "Effects of Enrichment with Salicylate on Bacterial
Selection and PAH Mineralization in a Microbial Community from a Bioreactor Treating
Contaminated Soil." Environ. Sci. Technol. 42(11): 4099-4105.
Prabhu, Y. and P. S. Phale (2003). "Biodegradation of phenanthrene by Pseudomonas sp. strain PP2:
novel metabolic pathway, role of biosurfactant and cell surface hydrophobicity in
hydrocarbon assimilation." Applied Microbiology and Biotechnology 61(4): 342-351.
Riser-Roberts, E. (1998). Remediation of Petroleum Contaminated Soils: Biological, Physical, and
Chemical Processes. Boca Raton, Florida, Lewis Publishers.
Rogers, S. W., S. K. Ong, et al. (2002). "Natural Attenuation of Polycyclic Aromatic Hydrocarbon-
Contaminated Sites: Review." Practice Periodical of Hazardous, Toxic, and Radioactive Waste
Management 6(3): 141-155.
Rosenberg, E. and E. Z. Ron (1999). "High- and low-molecular-mass microbial surfactants." Applied
Microbiology and Biotechnology 52(2): 154-162.
116
Roskam, G. D. and R. N. J. Comans (In Press). "Availability and leaching of polycyclic aromatic
hydrocarbons: Controlling processes and comparison of testing methods." Waste
Management In Press, Corrected Proof.
Rothmel, R. K., R. W. Peters, et al. (1998). "Surfactant foam/bioaugmentation technology for in situ
treatment of TCE-DNAPLs." Environmental Science and Technology 32(11): 1667-1675.
Roy, D., R. R. Kommalapati, et al. (1995). "Soil flushing of residual transmission fluid: Application of
Colloidal Gas Aphron suspensions and conventional surfactant solutions." Water Research
29(2): 589-595.
Ruberto, L. A. M., S. C. Vazquez, et al. (2006). "Phenanthrene Biodegradation in Soils Using an
Antarctic Bacterial Consortium." Bioremediation Journal 10(4): 191 - 201.
Samanta, S. K., A. K. Chakraborti, et al. (1999). "Degradation of phenanthrene by different bacteria:
evidence for novel transformation sequences involving the formation of 1-naphthol."
Applied Microbiology and Biotechnology 53(1): 98-107.
Samanta, S. K., O. V. Singh, et al. (2002). "Polycyclic aromatic hydrocarbons: environmental pollution
and bioremediation." Trends in Biotechnology 20(6): 243-248.
Schell, M. A. and P. E. Wender (1986). "Identification of the nahR gene product and nucleotide
sequences required for its activation of the sal operon." Journal of Bacteriology 166(1): 9-14.
Schlebaum, W., G. Schraa, et al. (1999). "Influence of Nonlinear Sorption Kinetics on the Slow-
Desorbing Organic Contaminant Fraction in Soil." Environ. Sci. Technol. 33(9): 1413-1417.
Sebba, F. (1971). "Microfoams--an unexploited colloid system." Journal of Colloid and Interface
Science 35(4): 643-646.
Sebba, F. (1985). "Separations Using Aphrons." Separation & Purification Reviews 14(1): 127-148.
Shin, K.-H., K.-W. Kim, et al. (2008). "Rhamnolipid Morphology and Phenanthrene Solubility at
Different pH Values." J Environ Qual 37(2): 509-514.
Shin, K. H., K. W. Kim, et al. (2004). "Combined effects of pH and biosurfactant addition on
solubilization and biodegradation of phenanthrene." Applied Microbiology and
Biotechnology 65(3): 336-343.
Simunek, J., M. Th. van Genuchten, et al. (2003). Studio of Analytical Models for Solving the
Convection-Dispersion Equation (STANMOD) I. G. W. M. Center. Golden, Colorado School of
Mines: CXTFIT 2; CHIAN; CFITM; 3DADE; N3DADE; SCREEN.
Singh, A. and O. P. Ward, Eds. (2004). Biodegradation and Bioremediation Soil Biology. Berlin
Heidelberg, Springer-Verlag.
Soberón-Chávez, G., F. Lépine, et al. (2005). "Production of rhamnolipids by Pseudomonas
aeruginosa." Applied Microbiology and Biotechnology 68(6): 718-725.
Son, A.-J., J.-U. Lee, et al. (2003 ). "Chemical and Ecotoxicity Assessment of PAH-Contaminated Soils
Remediated by Enhanced Soil Flushing." Environmental Engineering Science 20(3): 197-206.
Song, Y. F., X. Jing, et al. (2002). "Comparative study of extraction methods for the determination of
PAHs from contaminated soils and sediments." Chemosphere 48(9): 993-1001.
Stellner, K. L. and J. F. Scamehorn (1989). "Hardness tolerance of anionic surfactant solutions. 1.
Anionic surfactant with added monovalent electrolyte." Langmuir 5(1): 70-77.
Suthersan, S. S. and C. F. Payne (2005). In Situ Remediation Engineering. London, CRC Press.
The Interstate Technology & Regulatory Council (2005). Overview of In Situ Bioremediation of
Chlorinated Ethene DNAPL Source Zones. E. R. I. o. t. S. (ERIS). Washington DC, Interstate
Technology & Regulatory Council, Bioremediation of Dense Nonaqueous Phase Liquids (Bio
DNAPL) Team.
Toride, N., F. J. Leij, et al. (1993). "A Comprehensive Set of Analytical Solutions for Nonequilibrium
Solute Transport With First-Order Decay and Zero-Order Production." Water Resources
Research 27(7): 2167–2182.
Toride, N., F. J. Leij, et al. (1995). CXTFIT Model: Estimates parameters in several equilibrium and
non-equilibrium transport models from laboratory or field tracer experiments USDA
Agricultural Research Service
117
Torrens, J. L., D. C. Herman, et al. (1998). "Biosurfactant (Rhamnolipid) Sorption and the Impact on
Rhamnolipid-Facilitated Removal of Cadmium from Various Soils under Saturated Flow
Conditions." Environ. Sci. Technol. 32(6): 776-781.
U.S Environmental Protection Agency. (1996). "Test methods for evaluating solid wastes
physical/chemical methods: Method 3550B Ultrasonic Extraction." Retrieved July 9, 2008,
2008, from http://www.epa.gov/sw-846/pdfs/3550b.pdf.
U.S Environmental Protection Agency. (2007). "Test methods for evaluating solid wastes
physical/chemical methods: Method 3550C Ultrasonic Extraction." Retrieved July 9, 2008,
2008, from http://www.epa.gov/SW-846/pdfs/3550c.pdf.
U.S Environmental Protection Agency. (2008, October 18, 2008). "Toxic and Priority Pollutants: Clean
Water Act Analytical Test Methods." Retrieved December 21, 2008, from
http://www.epa.gov/waterscience/methods/pollutants-background.htm.
Verschueren, K. (1983). Handbook of environmental data on organic chemicals. New York, Van
Nostrand Reinhold.
Volkering, F., A. M. Breure, et al. (1993). "Effect of micro-organisms on the bioavailability and
biodegradation of crystalline naphthalene." Applied Microbiology and Biotechnology 40(4):
535-540.
Volkering, F., A. M. Breure, et al. (1992). "Microbial degradation of polycyclic aromatic
hydrocarbons: effect of substrate availability on bacterial growth kinetics." Applied
Microbiology and Biotechnology 36(4): 548-552.
Wan, J., S. Veerapaneni, et al. (2001). "Generation of stable microbubbles and their transport
through porous media." Water Resources Research 37(5).
Wang, P. and A. A. Keller (2008). "Partitioning of hydrophobic organic compounds within soil-water-
surfactant systems." Water Research 42(8-9): 2093-2101.
Wang, Q., X. Fang, et al. (2007). "Engineering bacteria for production of rhamnolipid as an agent for
enhanced oil recovery." Biotechnology and Bioengineering 98(4): 842-853.
Wang, S. and C. N. Mulligan (2004). "An evaluation of surfactant foam technology in remediation of
contaminated soil." Chemosphere 57(9): 1079-1089.
Warith, M., L. Fernandes, et al. (1999). "Design of in-situ microbial filter for the remediation of
naphthalene." Waste Management 19(1): 9-25.
Wick, L. Y., T. Colangelo, et al. (2001). "Kinetics of Mass Transfer-Limited Bacterial Growth on Solid
PAHs." Environmental Science & Technology 35(2): 354-361.
Woo, S., C. Jeon, et al. (2004). "phenanthrene biodegradation in soil slurry systems: Influence of
salicylate and triton X-100." Korean Journal of Chemical Engineering 21(2): 412-418.
Woo, S. H., M. W. Lee, et al. (2004). "Biodegradation of phenanthrene in soil-slurry systems with
different mass transfer regimes and soil contents." Journal of Biotechnology 110(3): 235-
250.
Woo, S. H., J. M. Park, et al. (2001). "Evaluation of the interaction between biodegradation and
sorption of phenanthrene in soil-slurry systems." Biotechnology and Bioengineering 73(1):
12-24.
World Health Organization (1998) "Environmental Criteria 202. Selected Non-Heterocyclic Polycyclic
Aromatic
Hydrocarbons."
Xiang, L., M. Xian-min, et al. (2006). "Experiments and modelling of phenanthrene biodegradation in
the aqueous phas by a mixed culture." Journal of Environmental Sciences 18(1): 147-153.
Xu, Q., M. Nakajima, et al. (2008). "A comparative study of microbubble generation by mechanical
agitation and sonication." Innovative Food Science & Emerging Technologies 9(4): 489-494.
Yan, L. (1985). "Study of the carcinogenic mechanism for polycyclic aromatic hydrocarbons--
extended bay region theory and its quantitative model." Carcinogenesis 6(1): 1-6.
Yen, K. M. and C. M. Serdar (1988). "Genetics of naphthalene catabolism in pseudomonads." Critical
reviews in microbiology 15(3): 247-268.
118
Yolcubal, I., S. A. Pierce, et al. (2002). "Biodegradation during Contaminant Transport in Porous
Media: V. The Influence of Growth and Cell Elution on Microbial Distribution." J Environ Qual
31(6): 1824-1830.
Yu, H., L. Zhu, et al. (2007). "Enhanced desorption and biodegradation of phenanthrene in soil-water
systems with the presence of anionic-nonionic mixed surfactants." Journal of Hazardous
Materials 142(1-2): 354-361.
Zhou, W. and L. Zhu (2005). "Distribution of polycyclic aromatic hydrocarbons in soil-water system
containing a nonionic surfactant." Chemosphere 60(9): 1237-1245.
Zhou, W. and L. Zhu (2007). "Efficiency of surfactant-enhanced desorption for contaminated soils
depending on the component characteristics of soil-surfactant-PAHs system." Environmental
Pollution 147(1): 66-73.
Zhu, L., B. Chen, et al. (2003). "Interactions of Organic Contaminants with Mineral-Adsorbed
Surfactants." Environmental Science & Technology 37(17): 4001-4006.
Zhu, L. and W. Zhou (2008). "Partitioning of polycyclic aromatic hydrocarbons to solid-sorbed
nonionic surfactants." Environmental Pollution 152(1): 130-137.
Zitha, P., Q. Nguyen, et al. (2006). "Coupling of Foam Drainage and Viscous Fingering in Porous
Media Revealed by X-ray Computed Tomography." Transport in Porous Media 64(3): 301-
313.
