Bioremediation approaches for organic pollutants: A critical perspective
In-Vessel Composting Bioremediation of Electrical Insulating ...
-
Upload
khangminh22 -
Category
Documents
-
view
0 -
download
0
Transcript of In-Vessel Composting Bioremediation of Electrical Insulating ...
In-Vessel Composting Bioremediation of Electrical
Insulating Oil-Contaminated Soil
2009
Lucinda Alice Rose Warner
Imperial College London
A thesis submitted to the University of London in partial fulfilment of the
requirements for the degree of Doctor of Philosophy in the Division of
Biology and Diploma of Imperial College London
1
Declaration
The work presented in this thesis was carried out at Wye College, Division of
Biology, Faculty of Natural Sciences, Imperial College London from April 2005.
This thesis is the result of my own work and any quotation from, or description
of the work of others is acknowledged herein by reference to the sources,
whether published or unpublished. This thesis is not the same as any that I have
submitted for any degree, diploma or other qualification at any other University.
No part of this thesis has been or is being concurrently submitted for any such
degree, diploma or other qualification. It is less than 50,000 words long,
excluding the bibliography and the appendices.
Lucinda Warner (10/12/08)
2
Abstract
Electrical insulating oils are used to insulate electrical transmission devices. The
oils consist of saturated and aromatic hydrocarbons and are released into the
environment through leaks and accidental releases. Electrical insulating oils in
soil can be measured as total petroleum hydrocarbons (TPH). Contaminated soils
are disposed of to landfill, but this is unsustainable. Therefore, in-vessel
composting bioremediation is an attractive alternative treatment.
The aims of this study were to optimize in-vessel composting bioremediation of
electrical insulating oil-contaminated soils, to achieve efficient removal of TPH
contamination and to examine the effects of soil properties, contaminant ageing
and compound physico-chemical properties on TPH removal during composting.
For identification and quantification of TPH, samples were extracted using ASE®
and analyzed by GC-MS. A preliminary pilot-scale composting study was
evaluated to determine the effectiveness of composting on TPH losses that were
14.3 to 61.5%. Composting process optimization at bench-scale found optimal
operating conditions of 0.8:1 soil to green waste ratio, 38°C operating
temperature and 60% moisture content, which resulted in enhanced TPH removal
of 75.6 to 86.1%. By composting two contrasting oil-contaminated soils, it was
shown that differing soil characteristics had no significant effect on TPH
biodegradation during composting. Contaminated soils were aged for 3 months
to examine contaminant ageing effects on composting, but no significant
difference in TPH losses from aged soils and freshly contaminated soils was
found. Loss of some contaminants was influenced by their physico-chemical
properties, more hydrophobic compounds being less susceptible to degradation.
3
The microbial community involved in composting was dominated by bacteria;
Gram-positive bacteria and actinomycetes were more abundant than Gram-
negative bacteria, fungi were less numerous. This study demonstrates that
composting bioremediation has potential for the clean-up of electrical insulating
oil-contaminated soils.
4
Acknowledgements
I would like to thank Dr Angus Beck and Prof Nick Russell for their continuing
input, support and help during this project, your help has been very much
appreciated. I would also like to thank Dr Joe Lopez-Real and Dr Blanca
Antizar-Ladislao for introducing me to the subject of bioremediation as an
undergraduate for my final year project, and to Dr Lopez-Real for letting me
know about this PhD opportunity. Also thanks to Prof Mike Jeger for letting me
change my project.
Thanks to Crown BioSystems, for the provision of compost samples and input
from their staff members; Ian Grant, Joe Rother and Muffadel Ezzi. Also to
EPSRC for funding this project.
Thank you to Mark Bennett for his help with the GC-MS, his guidance in using it
as well as the regular maintenance. Thanks to Colin Ladley for his technical
support and problem solving, as well as checking I was OK in the lab. Thanks
also to Ian Browning and John Haines for their help. I would like to thank Tim
Lawrence from EDF Energy for his technical advice as well as provision of
samples. Also to Brockhurst Forestry for the provision of green waste samples.
I would like to thank my parents for their moral support and encouragement, also
thanks to my brothers, Oliver and Hugo. I would also like to thank Liam Harris
for his support during the last three years. Thank you to my friends from Wye
and SAF for helping to keep me partially sane during this experience. Thanks
5
also to Julie Hoare for the provision of much needed coffee and to Lee Marsh for
the provision of other beverages at the end of a long day.
6
Table of Contents
Title Page 1
Declaration 2
Abstract 3
Acknowledgements 5
Table of Contents 7
Table of Figures 13
List of Tables 19
Chapter 1. Introduction 21
1.1. Electrical Insulating Oils in the Environment and Their Remediation 21
1.2. Project Hypotheses 23
1.3. Aims and Objectives 24
1.4. Deliverables 26
1.5. Organization of Thesis 27
Chapter 2. Electrical Insulating Oils: Contamination of Soils and Their
Remediation 28
2.1. Electrical Insulating Oils 28
2.1.1. Composition 30
2.1.2. UK Electricity Network 31
2.1.3. Amount of Oil 31
2.1.4. Electrical Insulating Oil Pollution 32
2.2. Oil Transport and Fate in Soil 32
2.3. Hydrocarbon Degrading Organisms 37
2.3.1. Metabolism of Hydrocarbons 40
2.3.2. Environmental Conditions Affecting Degradation 45
7
2.4. Toxicity and Effects of Oils 48
2.5. Remediation 51
2.5.1. Remediation and Clean-Up Standards 52
2.5.2. Bioremediation 52
2.5.3. Composting Bioremediation 54
2.6. Conclusions 60
Chapter 3. Development and Validation of Methods for the Analysis of Total
Petroleum Hydrocarbons in Soil and Compost Samples 61
3.1. Total Petroleum Hydrocarbons 61
3.1.1. Overview of Method Development 62
3.2. General Preparation Methods 63
3.2.1. Cleaning of Glassware 63
3.2.2. Preparation of Samples 64
3.2.3. Preparation of Standards 64
3.2.4. Determination of Moisture and Ash Contents of Soil and Compost. 66
3.3. Extraction of Soil and Compost Samples 67
3.3.1. Extraction Methods 68
3.3.2. Results of Soxhlet and ASE® Extraction Methods 72
3.4. Clean-up of Soil and Compost Extracts 75
3.4.1. Clean-Up Column Methods 75
3.4.2. Results of Clean-Up of Extracts 78
3.5. GC-MS Analysis of Samples 80
3.5.1. Method Development 81
3.6. Quality Assurance and Quality Control 86
8
3.6.1. Continuous Validation of Extraction Methods Using Surrogate
Standard Recoveries 88
3.6.2. Surrogate Standard Recovery Method 89
3.6.3. Surrogate Recovery Results 90
3.6.4. Continuous Analysis 90
3.6.5. Statistical Analysis 92
3.7. Conclusions 92
Chapter 4. Pilot-Scale Composting Study of In-Vessel Composting
Bioremediation of Electrical Insulating Oil-Contaminated Soil 94
4.1. Introduction 94
4.2. Materials and Methods 96
4.2.1. Preparation of Time Zero Samples 99
4.2.2. Preparation and Analysis of Samples 100
4.3. Results of Pilot-Scale Study 101
4.3.1. Contribution of Hydrocarbon Groups to TPH 101
4.3.2. Relative Rates of TPH Loss 103
4.3.3. Kinetics of TPH Loss 104
4.3.4. Remediation Targets 105
4.3.5. Importance of Treatment Parameters 108
4.4. Discussion 110
4.5. Conclusions 113
Chapter 5. Laboratory Optimization of Bench-Scale In-Vessel Composting
of Electrical Insulating Oil-Contaminated Soil 114
5.1. Introduction 114
5.1.1. Hypotheses 118
9
5.1.2. Aims and Objectives 118
5.2. Materials and Methods 119
5.2.1. Contaminated Soil 119
5.2.2. Green Waste 120
5.2.3. Bench-Scale Composting System 121
5.2.4. Soil to Green Waste Ratio and Temperature Optimization 123
5.2.5. Moisture Content Optimization 124
5.3. Optimization Results 125
5.3.1. Soil to Green Waste and Temperature Optimization Results 125
5.3.2. Moisture Content Optimization Results 132
5.4. Discussion 134
5.5. Conclusions 141
Chapter 6. Effect of Soil Properties on the Removal of TPH Contamination
During In-Vessel Composting 143
6.1. Introduction 143
6.1.1. Hypothesis 145
6.1.2. Aims and Objectives 145
6.2. Materials and Methods 146
6.3. Results 148
6.4. Discussion 152
6.5. Conclusions 155
Chapter 7. Effect of Contaminant Ageing on the Composting
Bioremediation of Electrical Insulating Oil-Contaminated Soils 157
7.1. Introduction 157
7.1.1. Hypothesis 159
10
7.1.2. Aims and Objectives 159
7.2. Materials and Methods 160
7.3. Results 165
7.4. Discussion 170
7.5. Conclusions 173
Chapter 8. Effect of Contaminant Physico-Chemical Properties on Removal
of Contamination During In-Vessel Composting 174
8.1. Introduction 174
8.1.1. Aims and Objectives 178
8.2. Materials and Methods 179
8.3.1. Relationships Between Compound Physico-Chemical Properties and
Losses During Composting Optimization 180
8.3.2. Relationships Between Physico-Chemical Properties and Compound
Losses in Freshly Contaminated and Aged Contaminated Soils 186
8.3.3. Differences in soil types 194
8.3.4. Patterns of Compounds 195
8.4. Discussion 196
8.5. Conclusions 202
Chapter 9. PLFA Analysis of Microbial Communities During Composting
Bioremediation 203
9.1. Introduction 203
9.1.1. Hypothesis 205
9.1.2. Aims and Objectives 206
9.2. Materials and Methods 207
9.2.1. GC-MS Analysis 208
11
9.3. Results of PLFA Analysis 212
9.3.1. Optimized Composting Results 212
9.3.2. Aged Treatments Composting Results 215
9.5. Discussion 222
9.6. Conclusions 228
Chapter 10. Conclusions 229
References 235
Appendices 278
Appendix 4.1: List of DEFRA 17 PAF1s 278
12
Table of Figures
Figure 2.1. General design of substation transformer 29
Figure 2.2. Design of underground cables (adapted from National Grid 2005) 30
Figure 2.3. Representation of bacterial oxidation of alkanes (adapted from
Fritsche and Hofrichter 2001) 42
Figure 2.4. Bacterial degradation of aromatic hydrocarbons (adapted from
Cerniglia 1992) 44
involved in a chemical
63
Figure 3.1. Schematic representation of the steps
analytical method
Figure 3.2. Schematic of an ASE® machine
Figure 3.3. Packing of an ASE® extraction cell
68
70
Figure 3.4. Schematic of Soxhlet apparatus (adapted from PLT Scientific 2005)
71
Figure 3.5. Recovery of TPH analytes using Soxhlet and ASE® extraction
methods from; a) compost samples and b) soil samples .73
Figure 3.6. Average recovery of all surrogate standards from blank, compost and
soil samples using ASE® and Soxhlet extraction 74
Figure 3.7. Diagram of clean-up column packing from Mills et al. (1999) 76
Figure 3.8. Diagram of clean-up column packing from Antizar-Ladislao et al.
(2005) 77
Figure 3.9. Recovery of TPH analytes using clean-up columns from; a) compost
extracts and b) soil extracts 79
Figure 3.10. Average recovery of all surrogate standards using both clean-up
columns 80
13
Figure 3.11. Chromatograms of analyzed samples of; a) standard mix of all
target analytes, b) composted contaminated soil extract and c) contaminated
soil extract 84
Figure 3.12. Percentage recovery of total surrogates from project surrogate
recovery experiments 90
Figure 3.13. Percentage variation of sample analyte total response during
repeated analysis of a sample . 91
Figure 4.1. Steel barrels illustrating internal vanes and filled height 97
Figure 4.2. Insulation of steel barrels 97
Figure 4.3. Contribution of saturates and aromatics to TPH at the 'start' (Time 1)
of composting 102
Figure 4.4. Contribution of saturates and aromatics to TPH at the end of
composting 102
Figure 4.5. Rates of normalized relative loss of TPH per day from treatments
103
Figure 4.6. Slowest, fastest and intermediate kinetic losses of TPH from
treatments 105
Figure 4.7. Initial and final concentrations of TPH from treatments 106
Figure 4.8. Estimated initial and final concentrations of TPH in compost samples
106
Figure 4.9. Initial and final concentrations of aromatics after composting
bioremediation 107
Figure 4.10. Estimated initial and final concentrations of aromatics in the
compost during treatments 108
14
Figure 5.1. Diagram of bench-scale composting bioremediation optimization
122
Figure 5.2. Photo of compost reactors showing the attachment of the reactors to
the inlet and outlet tubing and air reservoir 123
Figure 5.3. Initial and final TPH concentrations at each experimental
temperature for soil to green waste ratios of; a) 0.7:1, b) 0.8:1, and c) 0.9:1
127
Figure 5.4. Initial and final TPH concentrations in soil control treatments at each
experimental temperature 128
Figure 5.5. Change in average TPH concentration during composting for soil to
green waste ratios of; a) 0.7:1, b) 0.8:1, c) 0.9:1, and d) soil control
129
Figure 5.6. Kinetic losses of TPH at three test temperatures during composting
with soil to green waste ratios of; a) 0.7: 1, b) 0.8:1, c) 0.9:1, and d) soil
controls 131
Figure 5.7. Average initial and final TPH concentrations in moisture content
treatments (0.8:1 soil to green waste ratio, 38°C) and soil control 132
Figure 5.8. Change in average TPH concentration during composting 133
Figure 5.9. Kinetics of loss from moisture content treatments and soil control
134
Figure 6.1. Initial and final TPH concentrations in gault clay and brick earth
compost samples 148
Figure 6.2. Initial and final TPH concentrations in gault clay and brick earth soil
controls 149
15
Figure 6.3. Change in TPH concentration in gault clay and brick earth during
composting .. 150
Figure 6.4. Change in TPH concentration in gault clay and brick earth soil
controls 150
Figure 6.5. Kinetic losses of TPH gault clay and brick earth during composting
152
Figure 6.6. Kinetic losses of TPH from gault clay and brick earth soil controls
152
Figure 7.1. Average inside and outside temperatures during ageing of
contaminated soils over 3 months 164
Figure 7.2. Percentage loss of TPH from soils with different contaminant ageing
treatments during composting 166
Figure 7.3. Percentage loss of TPH from soil controls with different contaminant
ageing treatments 166
Figure 7.4. Changes in TPH concentration during composting of a) gault clay
and b) brick earth with different ageing treatments 167
Figure 7.5. Changes in TPH concentration of a) gault clay and b) brick earth
controls with different ageing treatments 168
Figure 8.1. Relationships between losses of saturate and aromatic compounds
and physico-chemical properties of; a) molecular weight, b) boiling point, c)
log Kow, d) log Koe, e) aqueous solubility and f) vapour pressure 182
Figure 8.2. Relationships between increasing compound molecular weight and
loss of compounds in; freshly contaminated a) gault clay, b) brick earth and
aged contaminated c) gault clay, d) brick earth 188
16
Figure 8.3. Relationship between increasing compound boiling point and loss of
compounds in; freshly contaminated a) gault clay, b) brick earth and aged
contaminated c) gault clay, d) brick earth 189
Figure 8.4. Relationship between decreasing aqueous solubility and loss of
compounds in; a freshly contaminated a) gault clay, b) brick earth and aged
contaminated c) gault clay, d) brick earth 190
Figure 8.5. Relationship between increasing Kow point and loss of compounds in;
freshly contaminated a) gault clay, b) brick earth and aged contaminated c)
gault clay, d) brick earth 191
Figure 8.6. Relationship between increasing compound Koc and loss of
compounds in; freshly contaminated a) gault clay, b) brick earth and aged
contaminated c) gault clay, d) brick earth 192
Figure 8.7. Relationship between increasing compound vapour pressure and loss
of compounds in; freshly contaminated a) gault clay, b) brick earth and aged
contaminated c) gault clay, d) brick earth 193
Figure 9.1. Chromatogram of FAME standards showing peaks of compounds
used for identification 210
Figure 9.2. Chromatogram of extracted compost sample 211
Figure 9.3. Ratio of Gram-negative to Gram-positive bacteria in optimized
composting condition samples 213
Figure 9.4. Ratio of Gram-negative bacteria to actinomycetes in optimized
composting condition samples 213
Figure 9.5. Ratio of bacteria to fungi in optimized composting condition samples
214
17
Figure 9.6. Ratio of actinomycetes to fungi in optimized composting condition
samples 215
Figure 9.7. Ratio of Gram-negative to Gram-positive bacteria in gault clay with
different contaminant ageing treatments 217
Figure 9.8. Ratio of Gram-negative to Gram-positive bacteria in brick earth with
different contaminant ageing treatments 217
Figure 9.9. Ratio of Gram-negative bacteria to actinomycetes in gault clay
composted treatments 218
Figure 9.10. Ratio of Gram-negative bacteria to actinomycetes in brick earth
composted treatments 218
Figure 9.11. Ratio of bacteria to fungi in gault clay with different ageing
treatments 220
Figure 9.12. Ratio of bacteria to fungi in brick earth with different ageing
treatments 221
Figure 9.13. Ratio of actinomycetes to fungi in fresh- and aged-contamination
gault clay compost samples 221
Figure 9.14. Ratio of actinomycetes to fungi in fresh- and aged-contamination
brick earth compost samples 222
Figure 9.15. Changes in microbial group ratios and loss of TPH during
composting under optimized conditions 226
Figure 9.16. Changes in microbial group ratios and loss of TPH during
composting in composted oil-contaminated gault clay 227
Figure 9.17. Changes in microbial group ratios and loss of TPH during
composting in composted oil-contaminated brick earth 227
18
List of Tables
Table 2.1. Distribution of n-tetradecane and n-eicosane in the environment 33
Table 2.2. Toxicity of hydrotreated light naphthenic distillates 50
Table 2.3. Suitability of treatments for oil and hydrocarbon wastes (adapted from
Harris et al. 1998) 51
Table 2.4. Composting bioremediation studies of hydrocarbons 59
Table 3.1. Compounds included in TPH analysis 62
Table 3.2. Relative performances of Soxhlet and ASE® 67
Table 3.3. Quantification and confirmation ions of target compounds 82
Table 3.4. Quality assurance and quality control measures 87
Table 4.1. Treatments used in the Crown BioSystems pilot-scale study 98
Table 4.2. Sampling intervals of treatments (days after start of composting) ... 98
Table 4.3. Losses of compound groups from treatments 109
Table 4.4. Losses of petroleum hydrocarbon contamination from bioremediation
studies . 110
Table 5.1. Average percentage loss and kinetic rate loss constant of TPH from
soil to green waste and temperature optimization 131
Table 5.2. Rate constants for removal of TPH from other bioremediation studies
140
Table 6.1. Soil properties of gault clay and brick earth 147
Table 7.1. Percentage variation of contaminant spiking in each soil from
10,000ppm 162
Table 7.2. Rate constant (k) and R2 values for TPH losses from freshly
contaminated and aged contamination gault clay and brick earth
169
19
Table 8.1. Ranges of physico-chemical property values of compound groups
found in TPH 177
Table 8.2. R2 values for relationships between compound physico-chemical
properties and compound loss for increases at operating temperatures 184
Table 8.3. R2 values for relationships between compound physico-chemical
properties and compound loss for changes in soil to green waste ratio 185
Table 8.4. R2 values for relationships between compound physico-chemical
properties and compound loss using different moisture contents 186
Table 8.5. R2 values for relationships between compound physico-chemical
properties and percentage compound loss in freshly contaminated and aged
contaminated gault clay and brick earth 195
Table 9.1. Composted samples used for PLFA analysis 207
Table 9.2. FAME compounds of interest for microbial group identification 210
20
Chapter 1. Introduction
1.1. Electrical Insulating Oils in the Environment and Their Remediation
Electrical insulating oils are used principally to insulate high voltage electrical
equipment such as transformers and cables. Naphthenic distillates are used in
transformers, whereas linear alkyl benzene (LAB) oil is mainly used in cables.
Due to high performance requirements, standards ensure that electrical insulating
oils from different manufacturers are interchangeable (Shell Lubricants 2005).
Electricity transmission devices contain vast quantities of oils. For example, a
high voltage transformer can hold 20,000-32,000L, low voltage pole
transformers each contain 100L and underground cables contain about 4.5x106L
in total (National Grid Transco 2004). Transformers and other equipment used at
National Grid substations contain about 100x106L of oil (National Grid 1998).
Soils can become contaminated with electrical insulating oils due to leaks from
underground cables, pole transformers and substation transformers, caused by
third party damage, ageing and fracturing of components. In 2003, 4,000 tonnes
of soil contaminated with naphthenic oil was landfilled (CBS 2005). In 2006/07,
the National Grid estimated that 16,419L of cable oil was lost to the environment
from their underground cables (National Grid 2008a). This large number of
potential contaminant sources located around the UK, plus the large amounts of
oil contained within them, pose a threat to soil and the wider environment if they
are damaged or incorrectly managed.
Electrical insulating oils contain long-chain hydrocarbons, polychlorinated
biphenyls (in older transformers and capacitors) and polycyclic aromatic
21
hydrocarbons (National Grid 1998; Nynas Naphthenics 2001; Pillai et al. 2005)
which have the potential to persist, bioaccumulate and become toxic to
organisms, thereby raising concern for their impacts on ecosystem processes,
wildlife and human health (Antizar-Ladislao et al. 2004). The hydrophobicity of
the compounds results in their strong sorption to soils, which may be a key
limiting-factor for biodegradation in situ (Hwang and Cutright 2002), and their
presence as complex mixtures serves to exacerbate this problem. Therefore, soils
contaminated with electrical insulating oils need to be treated in situ or ex situ
using engineering approaches to bring the contamination under control, ensuring
that the cleaned soil is suitable for multiple uses.
The cost of disposing of oil-contaminated soil to a licensed special landfill is
£500/tonne (including labour/transport). Both the high running expense and the
fact that in July 2004 there was been a reduction in the number of UK special
waste landfill sites, from 230 to 14 (Defra 2004), means that landfill is an
unsustainable disposal method. Therefore, the electrical industry is exploring
methods of treating contaminated soil, such as bioremediation. An attractive
option is the composting of contaminated soil with green waste, by controlled ex-
situ in-vessel composting to degrade the oil contamination to acceptable levels,
so that the soil can be reused at transformer sites or sold for use in other
industries such as construction. For this purpose, industry focuses on an overall
reduction in the Total Petroleum Hydrocarbon (TPH) measurement, but from an
environmental and toxicological perspective, it would be more useful to
investigate changes in individual compounds and classes of compounds, because
22
of variations in the environmental behaviour and toxicity of individual oil
constituents.
Composting bioremediation has been demonstrated to be effective for a number
of contaminants, including polycyclic aromatic hydrocarbons, chlorophenols,
polychlorinated biphenyls, explosives, pesticides and petroleum hydrocarbons
(Lain et al. 1997; Semple et al. 2001; Namkoong et al. 2002; Antizar-Ladislao
et al. 2005a). Therefore, it is possibly a suitable method for use in the
remediation of electrical insulating oil-contaminated soil. Furthermore, a green
waste stream is readily available within the electrical industry from management
of vegetation around electrical installations. Thus, two waste issues could
potentially be resolved by in-vessel composting bioremediation.
1.2. Project Hypotheses
Composting of electrical insulating oil-contaminated soils with green waste will
result in the degradation of oil contamination, mediated by microbial
communities. The degradation of contaminants will be affected by different
levels of composting parameters, resulting in lower or higher contaminant losses.
Electrical insulating oils interact with soils, and these interactions and processes
vary depending on soil characteristics, which will affect the degradation of
contamination during composting, because the differing characteristics of
different soil types will result in different losses.
23
Ageing of contamination in soils will result in more persistent contamination,
which will be more resistant to removal during remediation, as compared to
freshly contaminated soils.
1.3. Aims and Objectives
Little information is available on the behaviour of electrical insulating oils in soil
or how composting bioremediation can be used to remove them. In light of this,
the aims of this project were to:
1) Optimize laboratory-scale composting conditions, resulting in removal of
contamination (allowing re-use of end product at transformer sites or sold for
use in other industries such as construction).
2) Understand the behaviour of electrical insulating oils in different soils during
composting bioremediation to demonstrate how soil characteristics affect
remediation.
3) Determine the effect of ageing on electrical insulating oil behaviour and
remediation.
4) Identify the composition of microbial communities involved in the
degradation of contamination.
24
To meet these aims, the following objectives were established:
1) Develop and validate a chemical analytical method for the analysis of
electrical insulating oils in soils and compost, including an extraction method
and GC-MS analysis. Determine contamination from TPH (total petroleum
hydrocarbons), both saturates and aromatics, including polycyclic aromatic
hydrocarbons.
2) Use a bench-scale simulation model of commercial-scale in-vessel
composting systems with control of temperature, moisture content and
aeration, to determine the optimal physical and chemical parameters
(temperature, moisture content, and waste mix-ratios) required for
degradation of electrical insulating oils.
3) Contaminate two contrasting types of soil with oil and compost them at the
pre-determined optimum conditions to determine the effects of different soil
properties on removal of contamination during composting bioremediation.
4) Contaminate soils with oil and age them using natural and artificial
techniques, then compost aged soils and freshly contaminated soils to
determine the effects of contaminant ageing on removal of TPH during
composting.
5) Use data from optimization, soil properties and ageing experiments to
analyze the behavioural patterns of individual compounds and classes of
compounds during composting, and determine the effects of physico-
25
chemical properties of the contaminants on their susceptibility to degradation
and removal during in-vessel composting.
6) Use phospholipid fatty acid analysis (PLFA) of composted contaminated soil
samples in order to identify microbial groups present during composting
bioremediation and how these change during composting.
1.4. Deliverables
The project will provide a robust chemical analytical method developed
specifically to identify and quantify electrical insulating oil compounds in soil
and compost samples.
An optimized composting process will be developed, resulting in the removal of
contamination to an acceptable level, allowing for the re-use of the end product.
It will give an indication of optimum composting parameters, although these
might depend on the composting system. In addition, optimization at the
laboratory scale will allow scope for conversion to larger commercial-scale
composting operations.
It will provide an understanding of the effects of properties of soils and
chemicals themselves, of ageing, and of the microbial communities during in-
vessel composting bioremediation of electrical insulating oil-contaminated soils.
Additionally, analysis of the composting process will provide better
26
understanding of the rates of contamination loss and residual contamination at
the end of the composting process.
1.5. Organization of Thesis
Following this Introduction, Chapter 2 provides an overview of the relevant
literature concerning electrical insulating oils, their characteristics, the
distribution and fate of organic contaminants in the environment, losses,
microbial degradation of hydrocarbons, toxicity of electrical insulating oils and
composting remediation. Chapter 3 discusses the development and validation of
chemical analytical methods used to quantify TPH. In Chapter 4 a pilot-scale
study carried out by Crown BioSystems is introduced and the results presented.
In Chapter 5 the laboratory optimized losses of contamination from composting
are presented in order to determine optimal operating parameters. The effects of
different soil characteristics on the behaviour of contamination during in-vessel
composting of contaminated soil and its removal are explored in Chapter 6. The
problem of contaminant ageing is introduced in Chapter 7 and the effect it has on
the removal of contamination during composting is discussed. In Chapter 8 the
behaviour of individual compounds and classes of compounds during the
composting process is examined and compared to that of TPH. Chapter 9
examines the dynamics of the microbial community involved in the composting
process and the removal of contamination. Finally, in Chapter 10 conclusions are
drawn from the project and are presented within the context of the potential for
clean-up of electrical insulating oil-contaminated soils.
27
Chapter 2. Electrical Insulating Oils: Contamination of Soils and
Their Remediation
2.1. Electrical Insulating Oils
Electrical insulating oils are important in the transmission of electrical power:
they insulate, cool, and lubricate high voltage electrical equipment such as
transformers, circuit breakers, capacitors and cables (Shell Lubricants 2005).
Their principal function is electrical insulation, so they need high dielectric
strength (Lipscomb 1988; El-Gayara et al. 2008) to minimize current loss.
Additionally, as electricity transfer generates vast amounts of heat, the oils
transfer heat away from the core of electrical equipment. They also protect the
internal components of electrical equipment (e.g. copper, aluminium, and
cellulose insulation) from corrosion and oxidation. Under normal conditions,
electrical insulating oil is expected to match the transformer life of 25 to 50 years
(Shell Lubricants 2005). Due to performance requirements, standards ensure that
electrical insulating oils from different manufacturers are interchangeable. Four
basic standards have evolved: in the United States, specifications are defined by
the American Society for Testing and Materials (ASTM); most European
specifications are defined by the International Electrotechnical Commission
(IEC); and the UK uses the British Standard (BS) [BS 148] (Shell Lubricants
2005).
Electrical insulating oils (treated light naphthenic oils) are refined mineral oil
obtained from the fractional distillation of crude petroleum. Once distilled, it is
treated to remove impurities (polar compounds containing nitrogen or oxygen) to
28
Low voltage bushing High voltage bushing
Core
Radiator grills
Windings
Oil-filled tank
Figure 2.1. General design of substation transformer based on Harlow 2003
make it suitable as an insulating and cooling liquid (Lipscomb 1988; Rouse
1998).
Naphthenic oils are used in transformers and at times in underground cables,
although linear alkylbenzene (LAB) oils are more often used (Energy Networks
Association 2007). Transformers reduce high voltages for industrial and
domestic use. The oil is primarily used to cool and insulate the transformer. Oil-
filled transformers consist of a steel core around which conductors (windings),
usually copper rectangular strip conductors, are coiled. The core and conductors
are insulated from each other using paper and presswood board. The copper and
coil assembly is placed in a tank filled with oil. Oil circulates around the
transformer, either by convection in smaller transformers, or by pumps in larger
transformers (Harlow 2003). The general structure of an oil-filled transformer is
shown in Figure 2.1.
29
Tough plastic cover protects against corrision
Aluminium or lead sheath to retain insulating fluid
The conductor is insulated by a thick wraping of fluid impregnated paper
Oil-filled cables consist of a copper conductor insulated from the ground by
layers of paper tapes. These are enclosed within an aluminium sheath and
protective plastic layer. The paper tapes are impregnated with oil, which is kept
under pressure (National Grid 2005). The design of underground cabling is
shown in Figure 2.2.
Figure 2.2. Design of underground cables (adapted from National Grid 2005)
2.1.1. Composition
Naphthenic insulating oils are defined as 'distillates (petroleum), hydrotreated
light naphthenic', which is a complex combination of hydrocarbons obtained by
treating a petroleum fraction with hydrogen in the presence of a catalyst. It
consists of hydrocarbons having carbon numbers predominantly in the range of
C15 to C30. The CAS number is 64742-53-6 and the EINECS number is 265-156-
6. Examples of insulating oils are Nynas Nytro 10GBN produced by Nynas
Naphthenics and Shell Transformer Oil 148 (Halligan 2005; Svensson 2005).
The compounds present in the distillate are normal-, iso- and cyclo-paraffins;
30
aromatics are also a major constituent, e.g. naphthalenes and polycyclic aromatic
hydrocarbons (PAHs) (Lipscomb 1988). In Nynas Nytro 10GBN, the percentage
weight of each molecular type is 72.5% saturates, 27.4% aromatics and 0.1%
polar molecules (test method D2007) (Svensson 2005).
2.1.2. UK Electricity Network
The UK electricity distribution network is 773,376km in length (466,699km of
underground cable and 306,677km of overhead lines) (Energy Networks
Association 2005). The National Grid (UK) high-voltage transmission system in
England and Wales consists of approximately 7,242km of high-voltage overhead
line, 660km of high-voltage underground cable and 337 substations (National
Grid 2008a). EDF Energy generates 7.6% of the UK's electricity. In the
southeast region, EDF Energy manage and maintain 180,000km of underground
and overhead cables (EDF Energy 2008a) and 66,300 distribution substations
(EDF Energy 2008b).
2.1.3. Amount of Oil
Electricity transmission devices can contain vast quantities of oils. For example,
high voltage transformers hold 20,000-32,000L each, low voltage pole
transformers contain 100L each and underground cables contain about 4.5x106L
in total (National Grid Transco 2004). A typical 400/132 kV Supergrid
transformer holds approximately 100x106L of insulating oil, the equivalent of
four oil tankers. The National Grid transformers contain 67x106L of oil, with
31
another 13x106L in small oil-volume equipment and a further 20x106L in other
equipment (National Grid 1998). These large quantities of oil pose a risk to the
environment if they are accidentally released.
2.1.4. Electrical Insulating Oil Pollution
Soils become contaminated with electrical insulating oils due to leaks from
underground cables, pole transformers and substation transformers. This can be
caused by third party damage, e.g. contractors hitting cables during construction;
ageing, resulting in cracks and multiple leaks from deteriorating components, as
well as cracking of welding due to stress. There is also a risk of a spillage when
oil is removed from equipment during maintenance (Harlow 2003; National Grid
Transco 2004). Losses of oil from underground cables in 2006/2007 from the
National Grid (2008b) were estimated at 16,419L. In 2003, 4,000 tonnes of soil
contaminated with naphthenic oil was landfilled (CBS 2005).
2.2. Oil Transport and Fate in Soil
The 'form' in which contaminants are present in soil, and their potential to be
lost or persist, depend on the soil physico-chemical characteristics,
environmental conditions and the properties of the chemicals themselves (Doick
et al. 2005a).
Once released, hydrocarbons become distributed into different environmental
compartments: air, biota, sediment, soil and water (Environment Agency 2003).
32
The physical and chemical properties of hydrocarbons affect their distribution.
The redistribution and fate of hydrocarbons depend on biotic and abiotic
processes (Fine et al. 1997). The fate of petroleum products released into the soil
is varied, as these products generally consist of complex mixtures of
hydrocarbons with differing physical and chemical properties (Fine et al. 1997),
such as aqueous solubility, hydrophobicity, lipophilicity and molecular structure
(Semple et al. 2003). An example of the distribution of two hydrocarbons is
shown in Table 2.1.
Table 2.1. Distribution of n-tetradecane and n-eicosane in the environment (ECB
2000)
Air
(%)
Water
(%)
Soil
(%)
Sediment
(%)
Suspended matter
(%)
Biota
(%)
n-Tetradecane
n-Eicosane
76.6
0.0
0.0
0.0
22.8
97.7
0.5
2.2
0.0
0.1
0.0
0.0
Electrical insulating oils are light non-aqueous phase liquids (LNAPLs) (Doick et
al. 2005a). They form discrete globules with high surface tensions, which move
through the soil vadose zone and accumulate in pools on reaching the water table
(Beck and Jones 1995a; Soga et al. 2004; Mohamed et al. 2007). Terrestrial oil
spills are characterized primarily by vertical movement of oil into soil (Leahy
and Colwell 1990; Soga et al. 2004). As LNAPLs move downwards through the
soil profile, they leave behind a trail of residual contamination within the soil
pore space comprizing isolated droplets, globules and ganglia that have become
33
disconnected from the main body of contaminants. The volume fraction of the
pore space occupied by the residual contamination is referred to as the residual
saturation (Beck and Jones 1995a; Saenton et al. 2002; Soga et al. 2004; Paria
2008). NAPLs present a long-term source of water pollution because the
contaminants continue to enter the aqueous phase to replace contamination which
is transported away, biodegraded or removed through remediation (Alexander
1994; Saenton et al. 2002).
Persistence of contaminants is related to compound hydrophobicity and is a
consequence of 'ageing', microbial recalcitrance, and/or formation of bound
residues (Semple et al. 2001; Doick et al. 2005a). There are two major
mechanisms responsible for the persistence of contaminants, namely transport-
related and sorption-related mechanisms (Beck et al. 1996; Doick et al. 2005a).
As the length of time an organic chemical is in contact with soil increases, its
availability decreases (Wang et al. 2001), which is termed 'ageing'. Ageing
results in decreased chemical and biological availability to degradation, transport
and transformation processes. The main mechanisms involved in the ageing are
sorption and diffusion (intraorganic matter diffusion and sorption-retarded pore
diffusion). The nature and extent of ageing depends on pollutant properties, soil
properties including soil organic matter (quality and quantity), and soil inorganic
constituents (pore size and structure), microbial activity and contaminant
concentration (Smith et al. 1997; Semple et al. 2001; Semple et al. 2003;
Latawiec et al. 2008; Paria 2008).
34
Transport-related mechanisms are those in which sorption/desorption is
controlled by water and air flow. Soils may develop distinct structural
arrangements due to particle size distribution and management, causing
variations in fluid flow (Beck et al. 1996; Yaron et al. 1996; Kamra and Lennartz
2005). Sorption processes are adsorption (constituents concentrate at the
interface of two phases) and absorption (components transferred from the bulk
state of one phase to another) (Haws et al. 2006). Sorption-related mechanisms
are divided into two classes: diffusive mass transfer and chemisorption. Diffusive
mass transfer mechanisms include film diffusion, pore diffusion, retarded
intraparticle diffusion and intraorganic matter diffusion (Beck et al. 1996;
Pignatello and Xing 1996; Fine et al. 1997; Gao et al. 2006). Chemisorption is
the covalent bonding of a functional group in an organic chemical to specific
sites on/within the organic matter and/or mineral content of soil solids (Beck and
Jones 1995b; Beck et al. 1996; Fine et al. 1997; Doick et al. 2005b).
Loss mechanisms of contaminants in soil environments include volatilization,
biodegradation, transformation, uptake by biota, leaching and physical mass
transport, e.g. run-off and soil erosion (Beck and Jones 1995b; Beck et al. 1996;
Semple et al. 2001; Madsen 2003; Doick et al. 2005a). Contaminants can be
removed from the soil at varying rates and to varying extents (Semple et al.
2003). Generally, the loss of organic chemicals from soils is biphasic: initially,
there is a short period of rapid loss followed by a longer period of slow
dissipation. Pollutant volatility, hydrophobicity and affmity for organic matter
govern the relative importance of each phase (Semple et al. 2001; Weber and
Kim 2005).
35
Photodegradation by sunlight, is limited, because most of the hydrocarbon
components have little tendency to partition to air (ECB 2000). Photochemical
transformation occurs in the top 0.1 to 0.5cm of soil, where light breaks down
chemicals into smaller molecules (photolysis) (Miller et al. 1989; Maliszewska-
Kordybach 2005).
Volatilization does not contribute substantially to PAH removal from soils,
although losses by volatilization occur, but mainly for low molecular weight
compounds (Maliszewska-Kordybach 2005). Alkanes in the range of C10 to C15
volatilize readily, whereas alkanes ranging from C16 to C20 tend not to be lost by
volatilization (Namkoong et al. 2002).
Transfer to organisms can also occur. Plants can take up contaminants via root
uptake, vegetative uptake of vapour from air, or uptake by external shoots of soil
dust (Beck et al. 1996; Esteve-Turrillas et al. 2005). Uptake of organic chemicals
by earthworms occurs through passive diffusion from soil solution through the
outer membrane of the worm or via sorption of the compounds from soil material
in the gut (Krauss et al. 2000; Jager et al. 2003). Exposure to terrestrial
vertebrates occurs through ingestion of contaminated material, absorption
through skin, or via inhalation. Ingestion is the primary route, with intentional
and accidental ingestion of soil and sediment being significant sources of
contaminant exposure (Smith et al. 2007). Soil ingestion by grazing farm
animals has been estimated to be in the range of 1-10% of their daily intake
(Rychen et al. 2008). Ingested soil may represent a potentially major route of
uptake for soil-borne organic contaminants by ruminants: 'bound residues' may
36
become 'available' within the gastrointestinal tract, followed by either
bioaccumulation or further metabolic losses or removal in milk or livestock
products, allowing transfer to humans (Beck et al. 1996; Rychen et al. 2008).
Food consumption represents a major source of xenobiotic contamination for
humans, with plant-based foods being a major proportion of dietary exposure
(Wild et al. 2005).
Microbial biodegradation is a major process affecting the persistence of organic
compounds in soils and is one of the primary mechanisms by which
hydrocarbons are eliminated from the environment (Leahy and Colwell 1990;
Semple et al. 2003). The rate of microbial decomposition is a function of several
factors: availability of contaminants to microorganisms that have the catabolic
ability to degrade them; the numbers of degrading microorganisms present in the
soil; the activity of degrading microorganisms, molecular structure and
concentration of the contaminant (Alexander and Scow 1989; Allard and Neilson
1997; Semple et al. 2003). Biodegradation is also dependent on factors including
temperature, degree of acclimation and chemical partitioning tendencies
(Raymond et al. 2001).
2.3. Hydrocarbon Degrading Organisms
Hydrocarbon-degrading microorganisms (bacteria, actinomycetes and fungi) are
ubiquitously distributed in soil (Atlas 1995; Chaillan et al. 2004; Hamamura et
al. 2006). The levels of hydrocarbon-utilizing microorganisms generally reflect
the degree of contamination (Leahy and Colwell 1990; Capelli et al. 2001).
37
Increased carbon (from oil) increases microbial activity and the proportion of oil-
decomposing bacteria (Reid et al. 2000; Semple et al. 2003). When oil pollutants
are present, hydrocarbon-degrading populations increase, typically to 10% of the
microbial community (Atlas 1995).
Individual microorganisms can only metabolize a limited range of hydrocarbon
substrates, so mixed populations with a broad range of enzymatic abilities are
required to degrade the range of hydrocarbons (Riffaldi et al. 2006). Therefore,
synergistic relationships between hydrocarbon-degrading microorganisms can
develop: one species may remove toxic metabolites that affect the activities of
another; the second species may degrade compounds that the other can only
partially degrade (Mohamad-Ghazali et al. 2004; Riffaldi et al. 2006).
Additionally, it has been reported that hydrocarbons interfere with chemotaxis
for the natural substrates of non-degrading microorganisms, favouring
hydrocarbon-degrading microbes (Capelli et al. 2001). The catabolic activity can
develop by adaptation, through the induction or repression of specific enzymes,
the development of new metabolic capabilities through genetic changes, and
selective enrichment of organisms able to transform the target contaminants
(Semple et al. 2003). There are two defining characteristics of hydrocarbon-
oxidizing microorganisms: membrane-bound, group-specific oxygenases and
mechanisms for optimizing contact between the microorganisms and water
insoluble hydrocarbons (Ron and Rosenberg 2002).
Fungi and bacteria are the main hydrocarbon-degrading organisms in soil (Balba
et al. 1998) and the ability to degrade hydrocarbons is exhibited by a wide
38
variety of genera (Jorgensen et al. 2000). More than 200 species of bacteria,
fungi and algae can degrade hydrocarbons (Onifade and Abubakar 2007). The
most important hydrocarbon-degrading bacteria genera in soil environments are
Achromobacter, Acinetobacter, Alcaligenes, Arthrobacter, Bacillus,
Flavobacterium, Rhodococcus, Nocardia, Micrococcus, Corynebacterium,
Mycobacterium, Pseudomonas and Sphingomonas. The fungal genera include
Trichoderma, Mortierella, Aspergillus, Penicillium, Candida, Fusarium, Mucor,
Rhodotorula and Sporobolomyces (Leahy and Colwell 1990; Balba et al. 1998;
Salanitro 2001; Van Gestel et al. 2003; Antizar-Ladislao et al. 2004).
White-rot fungi are able to degrade a wide range of contaminants including
PAlis due to their extracellular non-specific enzymatic complex that is used to
degrade lignin (Andersson et al. 2000; Canet et al. 2001). Examples of white rot
fungi that have been used in hydrocarbon bioremediation are Phanerochaete
chrysosporium, Pleurotus ostreatus, Bjekandera adusta, Trametes versicolor and
Coriolus versicolor (Canet et al. 1999; Mancera-Lopez et al. 2008).
The low aqueous solubility of hydrocarbons and the use of a membrane-bound
oxygenise in degradation, make it essential for bacteria to come into direct
contact with hydrocarbons. There are two strategies for enhancing contact:
adhesion/desorption and emulsification. Adhesion to the hydrocarbon/water
interface occurs through hydrophobic interactions, mediated by hydrophobins on
the cell surface and emulsification of oil contamination is mediated by
surfactants (Ron and Rosenberg 2002; Ward et al. 2003). Many hydrocarbon-
degrading organisms produce extracellular emulsifiers (biosurfactants) (Tellez et
39
al. 2002), including glycerides, glycolipids, phospholipids and lipopeptides
(Kanga et al. 1997). These reduce surface tension, producing micelles of oil and
therefore increase surface area (Kanga et al. 1997; Ron and Rosenberg 2002),
allow higher rates of degradation than water solubility of the compounds permits
(de Jonge et al. 1997) and also increase the transport of oil through soil (Molnar
et al. 2002).
2.3.1. Metabolism of Hydrocarbons
Hydrocarbon degradation occurs through the partial or complete metabolism to
carbon dioxide, metabolites and cell constituents (Salanitro 2001), although not
all hydrocarbons are mineralized to carbon dioxide, giving rise to secondary
metabolites (Margesin et al. 2000; Van Gestel et al. 2003).
Hydrocarbons differ in their susceptibility to microbial attack, and their
biodegradation potential is correlated to their chemical structures (Wang et al.
2001). They are generally ranked in the following order of decreasing
susceptibility and biodegradability: alkanes > branched alkanes > low-molecular-
weight aromatics > cycloalkanes > resins > asphaltenes (Capelli et al. 2001;
Labud et al. 2007). Short chain alkanes are preferentially degraded (Namkoong
et al. 2002). The recalcitrance of PAHs to biodegradation is directly proportional
to molecular weight (Wammer and Peters 2005). The degradation of each
fraction also depends on the particular biodegradation capabilities of indigenous
microorganisms (Cole 1998; Jorgensen et al. 2000; Capelli et al. 2001).
40
Organic materials are degraded by microorganisms for the generation of energy
and synthesis of new cell tissue. In most biological treatment systems, these
activities occur spontaneously. These processes are: oxidation (dissimilatory
process), which produces carbon dioxide, water, energy and by-products; and
synthesis (assimilatory process), which produces new cells (Wang et al. 2001).
The initial steps in the metabolism of aliphatic, cyclic, and aromatic
hydrocarbons by bacteria and fungi involve the oxidation of the substrate by
oxygenases, for which molecular oxygen is required. The availability of oxygen
in soils depends on rates of oxygen consumption, soil type and waterlogging
(Leahy and Colwell 1990). Oxidation products of most organic compounds are
usually hydroxylated products, aldehydes, ketones, carboxylic acids, carbon
dioxide and water. These intermediates are usually more water-soluble and more
biodegradable than the parent compounds (Wang et al. 2001).
Oxidative metabolism of alkanes is primarily through terminal methyl group
oxidation, although sub-terminal or diterminal oxidation may also occur (Fritsche
and Hofrichter 2001). Oxidative metabolism requires molecular oxygen and
NADH or NADPH co-factors in the presence of an inducible mono or mixed-
function oxygenase enzyme system. Fatty acids derived from even or odd carbon
number alkanes are known to be readily incorporated into the membrane of
consuming microorganisms (Rosenberg and Ron 1996; Salanitro 2001). The
general oxidation sequence of an alkane is represented in Figure 2.3.
41
Terrhinal oxidation Sub-terminal oxidation
CH3-(CH2)n-CH2-CH3 alkane
CH3-(C1-12)n-CH2-CH2OH primary alcohol
NAD++ NADH
CH3-(CH2)„-CH2-CHO aldehyde
NAD+1- NADH
CH3-(CH2)„-CH2-COOH fatty acid
CH3-(CH2)n-CH(OH)-CH3 secondary alcohol
NAD++ NADH
CH3-(CH2)n-CHO-CH3 methyl ketone
IL NAD+ NADH 02 H2O
CH3-(CH2)n-O-CHO-CH3 acetyl ester
X N
CH3-(CH2)n-CH2-CH2OH primary alcohol
41--- fatty acid 411(
acetate
[3-oxidation
acetyl coA
`ir intermediary metabolism
Figure 2.3. Representation of bacterial oxidation of alkanes (adapted from
Fritsche and Hofrichter 2001)
Oxidation of alkanes by bacteria is initiated by an enzyme system formed by a
membrane-bound alkane hydroxylase and two soluble proteins. Metabolism of
alkanes normally proceeds via sequential oxidation of a terminal methyl group to
render alcohols, aldehydes and finally fatty acids. In the case of Pseudomonas
putida GPo 1 the initial oxidation step is performed by an alkane hydroxylase
system composed of a membrane-bound non-heme iron monooxygenase (alkane
42
hydroxylase) and two soluble proteins, rubredoxin and rubredoxin reductase,
• which act as electron carriers between NADH and the hydroxylase for
conversion of an alkane into an alcohol. The alcohol can be further oxidized to
an aldehyde and acid prior to proceeding into the 13-oxidation and tricarboxylic
acid cycles (Marin et al. 2003; Van Hamme et al. 2003).
Aromatic hydrocarbons are degraded by a number of mechanisms. The
transformation mechanisms are primarily cometabolic (Wang et al. 2001).
Bacteria initially oxidize aromatic compounds by incorporating two atoms of
molecular oxygen into the benzene ring of a PAH to form a cis-dihydrodiol,
catalyzed by a multi-component dioxygenase. The cis-dihydrodiol is further
oxidized to aromatic dihydroxy components (catechols), which are then
channelled through ortho- or meta-cleavage pathways (Rosenberg and Ron 1996;
Samanta et al. 2002; Johnsen et al. 2005), leading to central intermediates such as
protocatechuates and catechols, which are further converted to tricarboxylic acid
cycle intermediates (Kanaly and Harayama 2000). Ring cleavage can occur
either between the two hydroxyl groups by ortho cleavage, or adjacent to the two
hydroxyl groups by meta cleavage (Cerniglia 1992). Both ring cleavage
reactions are catalyzed by specific dioxygenases (catechol 1,2 dioxygenase for
ortho cleavage and catechol 2,3 dioxygenase for meta cleavage). The product of
the ortho-cleavage (cis, cis muconate) is transferred to the instable enollactone,
which is in turn hydrolyzed a dicarboxylic acid, which is activated by transfer to
CoA, followed by thiolytic cleavage to acetyl-CoA and succinate. Oxygenolytic
meta-cleavage yields 2-hydroxymuconic semialdehyde, which is metabolized by
the hydrolytic enzymes to formate, acetaldehyde and pyruvate. These are then
43
North() pathway
cis,cis muconic acid
utilized in central metabolism (Fritsche and Hofrichter 2001). A representation of
bacterial degradation of aromatic hydrocarbons is shown in Figure 2.4.
co aromatic
02 dioxygenase
H OH
H OH
cis, cis, dihydrol dehydrogenase
NAD± NADH
CCLOH
meta pathway- catechol
CHO COON
OH
OH
2-hydroxymuconic semialdehyde
Figure 2.4. Bacterial degradation of aromatic hydrocarbons (adapted from
Cerniglia 1992)
Although hydrocarbon degradation is primarily aerobic, alkanes and PAHs can
also be degraded under anaerobic conditions by bacteria. These reactions may
take place under Fe(III)-reducing, denitrifying and sulfate-reducing conditions,
44
by anoxygenic photosynthetic bacteria or in syntrophic consortia of proton-
reducing and methanogenic bacteria (Van Hamme et al. 2003).
2.3.2. Environmental Conditions Affecting Degradation
Several environmental conditions are important for the microbial degradation of
contaminants. Therefore, understanding the environmental factors influencing
biodegradation is critical to the success of site remediation (Wang et al. 2001).
Limited degradation efficiency results from low temperatures, anaerobic
conditions, low nutrient levels and low bioavailability (Margesin et al. 2000;
Romantschuk et al. 2000; Coulon et al. 2005).
Temperature influences hydrocarbon biodegradation by affecting the physical
nature and composition of oil and its bioavailability, the nature of the microbial
community and their rates of hydrocarbon metabolism (Van Gestel et al. 2003;
Coulon et al. 2005). At low temperatures oil viscosity increases, slowing the
onset of biodegradation; additionally, rates of enzymatic activity decrease.
Higher temperatures increase the rates of hydrocarbon metabolism to a
maximum, typically in the range of 30 to 40°C, above which hydrocarbon
membrane toxicity is increased (Leahy and Colwell 1990; Margesin and
Schinner 2001; Wang et al. 2001). Elevated temperatures also enhance
contaminant availability by increasing solubility and mass transfer/diffusion
(Feitkenhauer et al. 2003; Van Gestel et al. 2003).
45
The availability of nitrogen and phosphorus affects hydrocarbon degradation
(Rosenberg and Ron 1996). Phosphorus is generally plentiful, but unavailable
because of its poorly soluble natural forms. Nitrogen is likely to be present as
dinitrogen gas within the soil pores, but is seldom in bioavailable forms such as
amino acids, ammonium or nitrate (Wang et al. 2001). Nitrogen is required for
amino acids and phosphorus is involved in energy generation and transfer (e.g. as
ATP). In theory, the optimal C: N: P ratio for degradation is 100:10:1, which can
be used to estimate the required nutrient amendment (Ferguson et al. 2003).
Organic and inorganic fertilizers can be used to stimulate biodegradation in soils
low in nutrients (Ferguson et al. 2003; Xu and Obbard 2003). Other
micronutrients are also needed, but they are usually present naturally (Wang et
al. 2001).
Biodegradation requires water for microbial growth, diffusion of nutrients, and
transport of waste products (Wang et al. 2001; Zhou and Hua 2004).
Hydrocarbon biodegradation in terrestrial ecosystems may be limited by the
available water. Optimal rates of biodegradation at 30 to 90% water saturation
have been demonstrated (Leahy and Colwell 1990). The pH of the environmental
medium is also a key factor in microbial growth and hydrocarbon mineralization
is favoured by near neutral pH conditions (Margesin and Schinner 2001). Most
microorganisms cannot tolerate pH levels above 9.5 or below 4.0 (Wang et al.
2001), and the optimal pH range for contaminant mineralization is generally 5.5
to 8.5 (Zhou and Hua 2004).
46
Petroleum compounds are best biodegraded under aerobic conditions. In soils the
oxygen content depends on microbial activity, soil texture, water content and
depth (Rhykerd et al. 1999). Mineralization in soils is severely limited when the
oxygen content is below 10% (Vasudevan and Rajaram 2001). Organisms that
depend on aerobic respiration require an adequate supply of molecular oxygen. If
molecular oxygen is not available, NO3-, S042+, CO2 and some organic
compounds can be used. However, oxygen carries the highest electron energy
potential, allowing the organisms to derive more energy through organic
oxidation. Studies evaluating anaerobic degradation indicate appreciably lower
degradation rates than those with aerobic conditions (Wang et al. 2001).
Contaminants have minimum threshold levels for their biodegradation. At very
low substrate concentrations, microorganisms may lack an energy source to
sustain cellular functions, or biodegradative enzymes may no longer be induced
(Gray et al. 2000). The bioavailability of contaminants to soil microorganisms is
also a potential limiting factor, being influenced mainly by the hydrophobicity of
the compounds, their sorption onto soil matrix or their volatilization (Yerushalmi
et al. 2003).
47
2.4. Toxicity and Effects of Oils
The persistence of organoxenobiotics in the environment is a matter of
significant public, scientific and regulatory concern because of the potential
toxicity, mutagenicity, carcinogenicity and ability to bioconcentrate up the
trophic ladder (Semple et al. 2001). Petroleum hydrocarbons are toxic to
microorganisms, plants, animals and humans (Rosenberg and Ron 1996; Labud
et al. 2007). The toxicity of aromatic hydrocarbons is relatively high, whereas
that of straight-chain paraffins is relatively low. Hydrocarbons stunt plant growth
if the soil concentration is higher than about 1% by weight (Wang et al. 2001).
Several individual PAHs such as benzo[a]pyrene, chrysene, indeno[1,2,3-
c,d]pyrene, and benzo[b]fluoranthene have produced carcinogenic, mutagenic,
and genotoxic effects in animal experiments (Chen and Liao 2006).
A general measure of the presence of crude oil or petroleum product in soils is
Total Petroleum Hydrocarbon (TPH), which is defined as the measurable amount
of petroleum-based hydrocarbon in environmental media (e.g. soil, water,
sediments). However, it is not a direct indicator of the risk, i.e. mobility, toxicity,
and exposure to human and environmental receptors, posed by contamination.
There are three approaches used to estimate the potential human health risks
posed by TPH contamination:
1) Indicator: estimated risk characterized by indicator compounds, appropriate
for evaluation of the carcinogenic risk.
2) Surrogate: risk characterized by a single surrogate compound.
48
3) Whole Product: toxicity and mobility of TPH are based on that of a whole
product of similar character (Gustafson 2002).
Transformer oil (CAS number 64742-53-6) is not in the priority list No 793/93
(as foreseen under Council Regulation, EEC) on the evaluation and control of the
risks of existing substances. However, it is classified as a category 2 carcinogen
(may cause cancer); additionally, it is labelled as toxic (ECB, 2005). The
occupational exposure limit value in the UK are limited to 5mg M-3 , with a short-
term exposure limit value set at 10mg m3 for 10min (IUCLID 2000). Toxicity
tests have been carried out for hydrotreated light naphthenic distillates (Table
2.2).
In the case of Nytro 10GBN, related health issues include irritation of the
respiratory tract due to inhalation of vapours and/or mists, prolonged skin contact
causing defatting and possible irritation, and eye contact that may also cause
irritation (Nynas Naphthenics 2004). In the environment Nytro 10GBN is a risk
for contamination of earth, soil and water (Nynas Naphthenics 2004).
49
Table 2.2. Toxicity of hydrotreated light naphthenic distillates (IUCLID, 2000)
Test Subject Measure Amount
Toxicity and
carcinogenicity
Acute oral toxicity
Acute inhalation toxicity
Acute dermal toxicity
Genetic toxicity in vivo
Carcinogenicity
Rat
Rat
Rabbit
Rat cells
Mouse (dermal)
LD50
LC50
LD50
-
-
>5000mg/kg bodyweight
2.18mg/1 after 4h exposure
>5000mg/kg bodyweight
Increase in aberrant cells
Formation of cancerous
papillomas; some components
arc carcinogenic
Aquatic Toxicity
Acute/prolonged toxicity
to fish
Chronic toxicity to fish
Acute toxicity to aquatic
invertebrates
Chronic toxicity to
aquatic invertebrates
Toxicity to aquatic plants
Salmon
Pimphales promelas
Daphnia magna
Ceriodaphnia spp.:
Daphnia magna
Algae: Scenedesmus
subspicatus
LC50 (96h
exposure)
7d exposure
EC50, 48h
exposure
7d exposure
21d
exposure
EC50 (96h
exposure)
>1000mg/1
>5000mg/1, larval survival and
growth are affected
>1000mg/1, survival and
reproduction affected
552mg/1, survival and
reproduction affected
>1000mg/1, reproduction rate
affected
>1000mg/I, growth and
biomass affected
Toxicity to soil
organisms
Toxicity to soil
organisms
Pseudomonas
fluorescens
EC20 (6h
exposure)
>1000mg/1
50
2.5. Remediation
The most common techniques used in the UK for dealing with contaminated soils
are covering or excavation and removal to landfill; however, neither method
degrades or destroys the contaminants and they are not environmentally fail-safe
(Scullion 2006). Petroleum-contaminated soil is currently treated using three
processes: physical, chemical and biological. The most common physical
methods of treatment are disposal to landfill and incineration. Chemical
treatment includes direct injection of chemical oxidants into contaminated soil
and groundwater. Biological treatment involves the breakdown of contamination
to non-toxic forms using microbiological processes (Sarkar et al. 2005; Anastasi
et al. 2008). The suitability of these techniques for the remediation of soils
contaminated with oil and related hydrocarbon wastes is listed in Table 2.3.
Table 2.3. Suitability of treatments for oil and hydrocarbon wastes (adapted from
Harris et al. 1998)
Treatment technology
Vap
our e
xtra
ctio
n
Cont
ainm
ent
Ther
mal
trea
tmen
t
Vitr
ifica
tion
Solv
ent e
xtra
ctio
n
Pum
p an
d tre
at
cn
cC o u.1
Semi-volatile organics o ,/ ,/ / x x
Oil o ,./ ,/ ./ x x
Coal tars x o o ./ V ./
O ,/
,/ ✓
o x
x, inappropriate; appropriate; o, potential use under certain circumstances.
51
2.5.1. Remediation and Clean-Up Standards
Electricity distributors have implemented policies to reduce the risk posed by
electrical insulating oils. The electricity industry is also progressively bringing in
equipment based upon using insulating media other than oil, thus reducing the
problem at source (Energy Networks Association 2005). An alternative to oil-
filled cables is cross-linked polyethylene (XLPE) cables (National Grid 2005).
Additionally, silicone-filled sub-station transformers can be used as an
alternative to oil-filled units (Patel 2004). The National Grid has implemented
strategies to improve leak detection and repair. They have identified cable joints
in environmentally sensitive locations (near surface waters, groundwaters and
borehole abstractions) and aim to replace all these joint bays to reduce the risk of
leakage. Prompt action is taken to repair any leak in accordance with the
operating code agreed between the Electricity Association and the Environment
Agency (Energy Networks Association 2007). The National Grid also monitors
the amount of oil pumped into cables to replace that lost. The amount of oil lost
in 2003/2004 was 31,596L, which was reduced by 48.03% to 16,419L in
2006/2007 (National Grid 2008a).
2.5.2. Bioremediation
Bioremediation is the use of living organisms (primarily microorganisms) to
remove environmental pollutants from soil, water and gases (Atlas 1995; Madsen
2003; Sarkar et al. 2005). Bioremediation may be approached using in situ
(applied directly to a contaminate site) or ex situ technology (after contaminated
material is removed from contaminated sites) (Madsen 2003). Organic
52
compounds are metabolized under aerobic or anaerobic conditions by the
biochemical processes of microorganisms (Sarkar et al 2005). Mineralization of
organic pollutants is their complete biodegradation to water and carbon dioxide,
but it requires using the pollutant as a substrate for growth. However, many
xenobiotics are not used productively as a source of carbon and energy for
growth but are transformed by enzymes with an affinity for their structures. This
process is termed co-metabolism. It is defined as a non-specific enzymatic
reaction, in which a substrate competes with the structurally similar primary
substrate for the enzyme active site. An initial co-metabolic transformation may
pave the way for subsequent attack by another organism (Zhou and Hua 2004;
Johnsen et al. 2005).
Bioremediation relies on two sources of competent microorganisms. The first
depends on the metabolic capacities of the indigenous microbial populations to
degrade contaminants and is accomplished through environmental modification
(e.g. aeration or fertilizer application) to overcome limiting factors, which is
termed biostimulation (Atlas 1995; Sakar et al. 2005). The second uses
exogenous microbial populations added to the medium; these cultures are
selected for their degradation activities and the process is termed
bioaugmentation (Atlas 1995; Scullion 2006).
Bioremediation has several advantages over landfill disposal and incineration,
such as the conversion of toxic wastes to non-toxic end products, a lower cost of
disposal, reduced health and ecological effects and long-term liabilities
associated with non-destructive treatment methods (Sarkar et al. 2005)
53
2.5.3. Composting Bioremediation
Composting bioremediation is an ex-situ technology that relies on mixing organic
material with contaminated soil or sludge (Kriipsalu et al. 2007). As the compost
matures the pollutants are degraded by the microflora, resulting in the loss and
stabilization of pollutants (Semple et al. 2001; Scullion 2006). Composting has
enormous potential for bioremediation, since it can sustain diverse populations of
microorganisms, including bacteria, actinomycetes and lignin-degrading fungi.
These microorganisms have the potential to biotransform a variety of pollutants
into carbon dioxide, water, less toxic substances and/or lock up pollutants within
the organic matrix (Semple et al. 2001; Namkoong et al. 2002). Bulking agents
are added to the compost mixture to increase porosity and serve as sources of
easily assimilated carbon for biomass growth. Aerobic metabolism generates
heat, resulting in significant temperature increases that bring about changes in the
microbial population and physiology in the compost mixture (Antizar-Ladislao et
al. 2005a). Composting bioremediation has two goals: first to maximize removal
of the contaminants and, second, to produce a re-usable end product.
Under normal conditions, composting proceeds through three phases: (1)
mesophilic phase, (2) thermophilic phase and (3) maturation phase. At first,
temperatures are mesophilic peaking at approximately 35°C, at which stage
mesophilic bacteria and fungi are dominant. As the temperature rises, mesophilic
activity declines and the mesophilic community self-destructs at approximately
45-50°C. Activity by indigenous thermophilic microbes, especially
actinomycetes, is then initiated, with temperatures peaking at approximately
55°C. The temperature rises further beyond the point of maximal microbial
54
activity and self-limitation starts to occur at 60°C. The system brings itself to
near microbial extinction at 65-70°C (Finstein and Hogan 1993; Cole 1998).
In the first stages the soluble and easily degradable carbon sources, such as
monosaccharides, starch and lipids are utilized. In the next stage microorganisms
start to degrade proteins. After the easily degradable carbon sources have been
consumed, more resistant compounds such as cellulose, hemicellulose and lignin
are degraded and partly transformed into humus (Tuomela et al. 2000). In the
final compost stage (maturation), most digestible organic matter has been
consumed and the composted material is considered stable (Antizar-Ladislao et
al. 2005a; Mohan et al. 2006).
The advantages of composting bioremediation are simplicity of operation and
design, and relatively high treatment efficiency (Namkoong et al. 2002). By
adding organic material to contaminated soil, microbial activity is enhanced,
including the activity of specific degraders found in the contaminated soil or
introduced with the organic material (Jorgensen et al. 2000). It also stimulates the
production of biosurfactants, thereby increasing the bioavailability of petroleum
components (Olds College Composting Technology Centre 2005). It also offers
the advantages of enhanced opportunities for co-metabolism and greater
contaminant availability due to higher solubility and mass transfer at higher
temperatures (Margesin and Schinner 2001).
Modifications in the physical and chemical microenvironments within the
compost can serve to increase the diversity of the microflora to which the
55
contaminant is exposed and enhances co-metabolism (Haderlein et al. 2001;
Antizar-Ladislao et al. 2004). Composting improves the physical structure of soil
by reducing its bulk density, and increasing soil aggregation and porosity,
thereby improving the contaminated soil environment for indigenous or
introduced microbial degradative activity (Scelza et al. 2007).
Temperature is an important environmental variable in composting efficiency,
because it affects not only the physiological reaction rates and population
dynamics of microbes (Laine and Jorgensen 1997), but also most of the
physicochemical characteristics of the environment (Antizar-Ladislao et al.
2005a). The raised temperatures achieved during composting also accelerate the
relatively slow chemical reactions in soil by directly increasing rates of
compound mass transfer to the aqueous phase (Cole 1998; Semple et al. 2001).
Elevated temperatures stimulate degradation and enhance the contaminant
bioavailability by increased solubility and mass transfer (Van Gestel et al. 2003;
Feitkenhaur et al. 2003; Antizar-Ladislao et al. 2004).
Aerobic composting gives a higher degree of decomposition for most
compounds. Additionally, aeration helps to sustain microbial activity by
maintaining an open structure in the compost (Edwards 1998) and is also
important in removing carbon dioxide and excess moisture (Stelmachowski et al.
2003; Mohee and Mudhoo 2005). The supply of air increases microbial
metabolic activity and cools the compost, allowing higher rates of biodegradation
than would normally occur in conventional composting (Edwards 1998; Antizar-
56
Ladislao et al. 2004). Addition of bulking agents, such as bark chips or straw,
increases porosity and aerobicity of the matrix (Semple et al. 2001).
Composting can be carried out using several methods: the composting mix can
then be spread on land (land farming), or be placed in piles (windrows), or
introduced into a large vessel equipped with some form of aeration. Land-
farming is the older practice of treating wastes by adding, for example, oil sludge
and nutrients to agricultural land and mixing using agricultural practices
(Jorgensen et al. 2000; Short 2003). Windrows consist of stacked contaminated
material to a height of 2-4m into which nutrients and air (forced aeration/turning)
are introduced to speed up biodegradation. In-vessel composting takes place in a
partially or completely enclosed container, in which environmental conditions
can be controlled. Enclosed vessels closely approximate a laboratory incubator,
with the compost and its associated microflora being exposed to a more even
temperature profile (Antizar-Ladislao et al. 2004). The method provides
operators with more control, enabling them to select suitable operating
parameters (e.g., temperature, moisture content and mix ratios) to promote
microbial activity and contaminant degradation (Edwards 1998; Antizar-Ladislao
et al. 2007). A contained vessel is desirable when the compost contains
hazardous chemicals (Alexander 1994).
In-vessel composting systems allow a higher degree of process control than is
possible with systems exposed to the external environment, thus providing an
overall, higher process efficiency (Kim et al. 2008). Decomposition is
accelerated by creating ideal conditions for microorganisms. The compost
57
material is brought up to optimal temperature relatively quickly and air is
supplied to increase metabolic activity. It can deal with a wide range of waste
types and can be carried out near the waste source, with the control of leachate,
odours and pathogenic organisms (Edwards 1998).
Composting bioremediation has been used to treat contaminants, including
petroleum hydrocarbons and polycyclic aromatic hydrocarbons, chlorophenols,
polychlorinated biphenyls, pesticides and explosives at laboratory and/or field-
scales (Semple et al. 2001; Godoy-Faimdez et al. 2008). A large number of
hydrocarbons found in soil, such as saturates, polycyclic aromatic hydrocarbons
(PAHs) and BTEX (benzene, toluene, ethyl benzene and xylene) compounds all
degrade during the composting process (Cole 1998; Semple et al. 2001; Mohan
et al. 2006). Examples of hydrocarbon composting bioremediation studies and
their effectiveness are shown in Table 2.4.
58
Table 2.4. Composting bioremediation studies of hydrocarbons
Contaminant Organic
matter
Period of
composting
Contaminant
removal (%)
Reference
Diesel TPH Compost 30d 69.3-99.1 (Namkoong et
al. 2002)
Diesel Biowaste 12 weeks 85 (Van Gestel et
al. 2003)
Fuel oil Sawdust 56d >55 (Godoy-
Fahndez et al.
2008)
Diesel oil Food waste 15d 79 (Joo et al.
2007)
PAHs Greenwaste 105-111d 64.1-80.9 (Antizar-
Ladislao et al.
2005a)
Diesel oil Bark chips 5 months 71 (Jorgensen et
al. 2000)
PAHs Sewage
sludge and
yardwaste
130d 68 (Moretto et al.
2005)
PAHs Mushroom
compost
100d 37-80 (S agek et al.
2003)
Phenanthrene Compost
feedstock
126d 25-42 (Hesnawi and
McCartney
2006)
59
2.6. Conclusions
Electrical insulating oils used to insulate electrical transmission devices can be
accidentally released into soils and, therefore, the contaminated soils need to be
cleaned up. Once they are in soils, the behaviour of oils is affected by soil
characteristics and interactions within the soil matrix, compound ageing as well
as compound physico-chemical properties. As conventional clean-up methods
such as covering or removal to landfill do not degrade or destroy the
contaminants and are unsustainable, alternative methods such as in-vessel
composting bioremediation may provide the solution to cleaning up these
contaminated soils. The scientific feasibility of using composting bioremediation
for the clean-up of soils will be explored in this study. The potential effects of
soil properties on the fate of oil contamination and therefore the removal of
contamination during composting will also be examined. The persistence of TPH
contamination due to ageing will be explored with respect to the removal of aged
contamination compared to that of fresh contamination during composting.
Additionally, the effect of compound physico-chemical properties, which affect
compound susceptibility to degradation during composting will be examined.
Finally, the changes in relative proportions of major groups of bacteria and total
fungi during the composting process will be examined, to give an indication of
the microbial groups involved in the composting process.
60
Chapter 3. Development and Validation of Methods for the
Analysis of Total Petroleum Hydrocarbons in Soil and Compost
Samples
In order to analyze and quantify Total Petroleum Hydrocarbon (TPH)
contamination in soil and compost matrices, a robust chemical analytical method
was developed. The main challenge in doing so was to overcome the
heterogeneity that is inherent in compost and in highly TPH-contaminated soils,
in such a way as to allow any changes in TPH concentrations occurring during
experiments conducted to be measured in confidence. This required extraction
and analytical protocols that were reproducible, and to achieve this rigorous
quality assurance and control was practiced.
3.1. Total Petroleum Hydrocarbons
A general measure of the presence of crude oil or petroleum product in soils is
Total Petroleum Hydrocarbon (TPH), defined as the measurable amount of
petroleum-based hydrocarbon in environmental matrices (Gustafson 2002). TPH
in this study was taken to be the target saturates (C10-C35) and aromatics (single-
ringed and polycyclic aromatic hydrocarbons) (Table 3.1).
61
Saturate fraction Aromatic fraction
Decane Undecane Dodecane Tridecane Tetradecane Pentadecane Hexadecane Heptadecane Pristane Octadecane Phytane Nonadecane Eicosane Heneicosane Docosane Tricosane Tetracosane Pentacosane Hexacosane Heptaco sane Octacosane Nonacosane Triacontane Hentriacontane Dotriacontane Tritriacontane Tetratriacontane Pentatriacontane
Biphenyl Dibenzothiophene Naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz[a]anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3,c,d)pyrene Dibenzo(a,h)anthracene Benzo(g,h,i)perylene Benzo(e)pyrene Perylene
Table 3.1. Compounds included in TPH analysis
3.1.1. Overview of Method Development
The general steps of a chemical analytical method are shown in Figure 3.1,
which also indicates the methods tested for this project. In order to develop the
methods for this project each analytical stage was examined to determine the best
method to analyze the target analytes in soil and compost matrices. An analytical
paper by Mills et al. (1999) was used as a starting point for the development of
the method.
62
Sample
Surrogate standard addition
Soxhlet Extraction
Accelerated Solvent Extraction
Column Clean-up
Internal standard addition
GC/MS
Figure 3.1. Schematic representation of the steps involved in a chemical
analytical method
3.2. General Preparation Methods
Throughout experimental procedures, general methods for the preparation of
equipment and standards were used, which are described in the following sub-
sections.
3.2.1. Cleaning of Glassware
Prior to use, glassware was washed with detergent, rinsed with distilled water,
acetone and finally with distilled water. It was then dried in a drying oven, before
being wrapped in foil and baked at 500°C overnight. After baking, it was left to
cool and stored in a clean-glassware cupboard ready for use.
63
3.2.2. Preparation of Samples
Soil samples were air dried in fume cupboards and then homogenized using a
food processor, and fmally passed through a 2mm sieve. The <2mm soil samples
were stored in sealed glass containers at room temperature.
Green waste was also homogenized in food processors, to ensure that there was
an even mix of components and that no large pieces were left. After
homogenization, the green waste fragments were at maximum —10mm long.
Samples were kept at 4°C for short-term storage and at -18°C for longer periods
(to prevent decomposition).
3.2.3. Preparation of Standards
The chemical standards used in method development and sample analysis were
used as surrogate and internal standards as well as calibration standards.
Tetracosane-d50 and dodecane-d26 (CDN Isotopes, 98% purity) were used as
surrogate standards and were supplied as solids. To make stock solutions,
31.2mg of tetracosane-d50 and 25.2mg of dodecane-d26 were weighed out into a
clean 10m1 volumetric flask. The compounds were then dissolved in
acetone/dichloromethane (1:1, v/v) and made up to 10m1, giving stock solutions
with final concentrations of 312mg/L and 252mg/L.
The surrogate standard solution comprized naphthalene-d8, acenaphthene-dio,
phenanthrene-dio, chrysene-d12, perylene-d12 (IS mix 26, QMX Labs 99.5%
purity), dodecane-d26 and tetracosane-d50. A stock solution was prepared by
64
transferring 0.25m1 of IS mix 26, 3.968m1 of dodecane-d26 and 3.205m1 of
tetracosane-d50 using micro-syringes into a clean 10m1 volumetric flask and
made up to 10m1 with acetone/dichloromethane (1:1, v/v). The final
concentration of the solution was 100mg/L of each component.
The internal standard solution comprized nitrobenzene-d5 (Ultra Scientific, 99%
purity), 2-fluorobiphenyl (Ultra Scientific, 99% purity) and 5-a-androstane (Ultra
Scientific, 99% purity). A stock solution was prepared by transferring 5Opt of
nitrobenzene-d5, 25µL of 2-fluorobiphenyl and 254, of 5-a-androstane into a
clean 10m1 volumetric flask using a micro-syringe. The volume was made up to
10m1 with acetone/dichloromethane (1:1, v/v) resulting in a final concentration
of 5mg/L of each component.
Calibration standards for each compound or mix of compounds were made up for
calibrating the GC-MS. The calibration concentrations for each target analyte
were 5mg/L, lmg/L, 0.5mg/L, 0.1mg/L, 0.05mg/L and 0.01mg/L. Firstly, the
5mg/L and lmg/L solutions were prepared, the appropriate volume of the neat
standard was transferred to a clean 10m1 volumetric flask and made up to volume
with acetone/dichloromethane (1:1, v/v) to make 5mg/L and lmg/L solutions. To
make the other concentrations lml of 5mg/L was transferred to a clean 10m1
volumetric flask and made up to volume resulting in a 0.5mg/L solution, lml of
this was taken and made up to 10m1 to give 0.05mg/L. The same serial dilution
method was used to make 0.1mg/L and 0.01mg/L.
65
All standards were stored in sealed glass vials at 4°C until use. Standards were
kept for six months, replaced when needed and new ones made after six months.
3.2.4. Determination of Moisture and Ash Contents of Soil and Compost
Sample moisture content was determined by analyzing 5g sub-samples in
triplicate. Samples were weighed into pre-weighed labelled 25ml beakers, then
covered with foil and re-weighed. The samples were dried at 110°C for 24h and
cooled in a desiccator prior to weighing. Moisture content was calculated using
equation (3.1).
[ wet weight — dried 1 x 100 = % moisture wet weight
Ash content was determined by heating the oven-dried samples (from the
moisture content determination) in a muffle furnace at 550°C for 12h. After
ashing, the samples were cooled in the oven, then transferred to a desiccator to
cool to room temperature before re-weighing. The ash content of the samples
was calculated using equation 3.2.
[ dry weight — ashed weight x 100 = % ash content dry weight
(3.2)
(3.1)
66
ASE(g) Soxhlet
Extraction time
Solvent use
Number of samples
Cost per sample
Relative safety
12-18mina
15-40mr
24a
$14'
Limited solvent exposureb
4-481?
200-500mr
Labour/ resource dependent
$27a
Exposure to boiling solventsc
3.3. Extraction of Soil and Compost Samples
The aim of sample extraction is to effectively remove the analytes of interest
from the matrix rapidly and quantitatively for analysis (Saim et al. 1997). Several
approaches can be used to extract organic analytes from environmental matrices.
These include Soxhlet, supercritical fluid extraction (SFE), microwave assisted
extraction (MAE), subcritical water extraction and accelerated solvent extraction
(ASE®), as well as others (Saim et al. 1997; Hawthorne et al. 2000).
For this project, two extraction methods (Soxhlet and accelerated solvent
extraction [ASE®]) were compared for recovery and efficiency of the processes
and to demonstrate that ASE® is a more efficient method with greater analyte
recoveries for the extraction of soil and compost samples. Soxhlet was chosen for
comparison against ASE®, as it is a widely used technique for the extraction of
hydrocarbons from soil. ASE® extraction is a newer, innovative and more
efficient technique that allows more complicated extraction procedures to be
carried out and has several advantages over Soxhlet extraction (Table 3.2.)
(Dionex 2003).
Table 3.2. Relative performances of Soxhlet and ASE®
a - Dionex 2003; b - Dionex Corporation 1999; C - Cyberlipid 2003
67
pump
solvent bottle
collection vial air cylinder — ___ nitrogen cylinder
oven
extraction cell
3.3.1. Extraction Methods
Accelerated solvent extraction (ASE®) is a relatively new automated extraction
method. It also allows complicated extractions to be carried out through the use
of multiple extraction programmes for each sample.
Samples to be extracted were loaded into stainless steel extraction cells, which
were placed in the carousel of the machine. Solvent was pumped into the cells
and within the oven of the machine the cell is heated and pressurized to set
parameters. The temperature and pressure were held constant for a set time, and
clean solvent is pumped into the cell. The solvent is then purged from the cell
using nitrogen gas and is recovered in collection vials, ready for clean-up or
analysis. A schematic of an ASE® machine is shown in Figure 3.2.
Figure 3.2. Schematic of an ASE machine
The extraction method for the ASE® 200 extraction was adopted from Dionex
Application Notes (Note 338: Extraction of total petroleum hydrocarbon
contaminants [diesel and waste oil] in soils by accelerated solvent extraction
(Dionex 2004). The program conditions were: oven 175°C, pressure 2000psi,
68
heat-up for 8min, static for 5min, solvent dichloromethane/acetone (1:1, v/v),
flush volume 75% of extraction cell volume and nitrogen purge for 60s.
For extractions, the samples were loaded into 22ml ASE® stainless steel
extraction cells, packed with a glass fibre disk, copper turnings (to remove
sulphur), 3g anhydrous sodium sulphate and 7g of sample mixed together; the
remaining space was filled with HydromatrixTM, following the approach of
Antizar-Ladislao et al. (2005b) (Figure 3.3). Before extraction, the cells were
spiked with a surrogate standard mix (naphthalene-cis, acenaphthene-dio,
phenanthrene-dio, chrysene-d12, perylene-d12, dodecane-d26 and tetracosane-d5o).
The extraction cells were placed in the ASE® machine carousel and extracted
using the program conditions described above. During continuous extractions the
machine was programmed to rinse the system between each sample extraction to
avoid the transfer of contamination between samples. Additionally, blanks were
used during extractions to monitor contamination from reagents used and/or the
machinery. A blank was used at the start, end and between sample repeats in an
extraction run. The extracts were measured in cleaned measuring cylinders and
stored in clean, sealed glass vials at -18°C until analysis.
Cells were cleaned after each batch of extractions. The cells were taken apart and
the components washed with detergent and rinsed with acetone. The parts were
dried in an oven, the cells then re-assembled and stored in a designated cupboard.
69
Bottom cap of extraction cell
Top cap of extraction cell
Hydromatrix packing material
Mix: 7g of sample and 3g of anhydrous sodium sulphate
Copper turnings
Figure 3.3. Packing of an ASE® extraction cell
Soxhlet extraction is a continuous solvent extraction method, using solvent
reflux. The sample to be extracted is placed in a porous extraction thimble, which
retains the insoluble material, while allowing the soluble analytes to be extracted.
The extraction thimble is placed in the extraction chamber. Below the extraction
chamber is the solvent reservoir. The sample is extracted by heating the solvent
in the reservoir using a heating mantle. As the solvent heats up it vaporizes and
rises to the condenser, where it condenses to a liquid and drips into the sample in
the extraction chamber. The extraction chamber fills with solvent and, when
almost full, the chamber automatically empties through the siphon side-arm to
the solvent reservoir. This cycle is repeated throughout the extraction period. A
schematic of the Soxhlet apparatus is shown in Figure 3.4. Prior to use the
extraction chambers and solvent reservoirs were cleaned and the condensers
rinsed with dichloromethane.
70
► Water out
Condenser
Water in
Extraction chamber
Extraction thimble
Solvent reservoir
Siphon
Bypass sidearm
Heating mantle
Figure 3.4. Schematic of Soxhlet apparatus (adapted from PLT Scientific 2005)
The extraction method used was adapted from Mills et al. (1999). Samples were
extracted for 24h using acetone: dichloromethane (1:1, v/v), heated at 70°C. For
extraction, 7g of sample was weighed into an extraction thimble (Whatman
cellulose acetate extraction thimbles, single thickness, 30mm x 100mm). To this
3g of anhydrous sodium sulphate was added. On top of this mixture a plug of
cleaned glass wool was placed to prevent loss of sample. Prior to extraction the
thimbles were spiked with a surrogate standard mix (naphthalene-d8,
acenaphene-d10, phenanthrene-dio, chrysene-d12, perylene-d12, dodecane-d26 and
tetracosane-d50). The extraction thimbles were placed in the extraction thimbles
and wetted with 50m1 of solvent. The solvent reservoir was filled with 150m1 of
solvent, a few anti-bumping crystals and copper turnings. The condenser water
was turned on and the solvent heated at 70°C for 24h. After the extraction period
71
the heat was turned off and the extract was allowed to cool to room temperature.
The extracts were collected and their volume measured. As large volumes of
extract were obtained, they were reduced using rotary film evaporation (Buchi
Rotavapor R-124 with a Biichi B-721 vacuum controller), using 250m1 round-
bottomed flasks with the water bath at 45-50°C. The extracts were reduced to
—20m1 using a negative pressure of 300mBar.
3.3.2. Results of Soxhlet and ASE Extraction Methods
The efficiency of the recovery of TPH analytes for both extraction methods was
compared for compost and soil extracts. The mean recovered amounts (mg kg-1)
of each sample analyte from each extraction method were plotted against each
other (Figures 3.5a and b).
The slope of the trend line for the recovery of TPH analytes for both compost
and soil was greater than 1 (1.09 [R2 = 0.85] and 1.51 [R2 = 0.90] respectively),
which shows that recovery of TPH analytes was greater using ASE® than Soxhlet
and is therefore a more efficient extraction process for the recovery of TPH, with
greater recoveries of analytes using ASE® ranging from 4.36 to 92.3% in
compost extracts and 1.97 to 97.1% in soil extracts greater than using Soxhlet.
This indicates that ASE® is a more efficient recovery method, however, as the
slopes of the lines were near to 1, this indicates that the methods were
comparable.
72
0 O
0
ASE
extra
ctio
n
16
0
0 0 A
SE
extra
ctio
n
50
45
40
35
30
25
20
15
10
5
0 0 5 10 15 20
TPH analyte recovery(mg kg 1) Soxhlet extraction
25 30
(a)
0 2 4 6 8 10
TPH analyte recovery(mg kg-1) Soxhlet extraction
0
12
14
(b)
Figure 3.5. Recovery of TPH analytes using Soxhlet and ASE W extraction
methods from; a) compost samples and b) soil samples
The recovery of surrogate standards from spiked blanks, compost and soil
samples was also examined, by comparing the percentage recovery of all
surrogates (Figure 3.6). The recovery of surrogates was high for all matrices and
both extraction methods (81.3 to 95.7%), therefore both are efficient extraction
73
methods. The high surrogate recovery shows that sample matrix effects are
negligible and therefore extraction results in high analyte recovery for analysis.
Although the recovery of surrogates was greater using ASE® (5.51 to 17.3%
greater), examination of the error bars indicate that the methods are comparable.
Soxhlet extraction
0 ASE extraction
0 Blank
Compost
Soil
Matrix type
Figure 3.6. Average recovery of all surrogate standards from blank, compost and
soil samples using ASE and Soxhlet extraction
Using ASE® to extract soil and compost samples results in higher recovery of
TPH analytes and surrogate standards than conventional Soxhlet extraction,
although the two methods were comparable, due to similar recoveries. Therefore,
for this study ASE® was used as the extraction method due to its rapid sample
turnover, ease of use and lower resource requirements.
0 0
a)
a) 2
co
110 100 --90 80 70 60 -50 -40 -30 -20 -10-
74
3.4. Clean-up of Soil and Compost Extracts
Clean-up is used to separate compounds of interest from interferences
(impurities) in extracted samples, because extraction results in the extraction of
both types of compound. Clean-up columns work by using interactions between
the column sorbent and compounds of interest or the interferences. If the analytes
of interest are bound to the sorbent they need to be eluted using solvents, whilst
if the interferences are sorbed then the sorbent is discarded. Clean-up columns
are available as commercial cartridges or can be prepared in the laboratory. For
this project two clean-up columns were compared for recovery of surrogates and
analytes of interest from extracted spiked blanks, compost and soils.
3.4.1. Clean-Up Column Methods
Two clean-up columns were used to test and compare efficiency. The first was
adapted from Mills et al. (1999) and the second was from Antizar-Ladislao et al.
(2005b), both were selective elution clean-ups.
The column adapted from Mills et al. (1999) used x0.4 of the recommended
amounts to fit in the column, apart from the sodium sulphate. Therefore, the
volume of eluents used was also proportional (Figure 3.7). The silica and
alumina were activated by heating in a shallow tray at 130°C for >16h and 400°C
for 24h, respectively (Mills et al. 1999). The solvent used for elutions was
acetone: dichloromethane (1:1, v/v).
75
Chromatography column
Anhydrous sodium sulphate (2g)
Activated silica gel (8g)
Activated alumina (4g)
Glass wool
Sand (2g)
J Figure 3.7. Diagram of clean-up column packing from Mills et al. (1999)
The column was first eluted with 16m1 of solvent, which was discarded. The
sample extract (5m1) was added to the column and the sample tube rinsed with
solvent (2m1) and added to the column. The column was then eluted with 16m1 of
solvent and the elutions collected and measured.
The column used in Antizar-Ladislao et al. (2005b) was packed as shown in
Figure 3.8. The florisil was activated by heating it at 130°C for 12h. The column
was initially washed with 10m1 of dichloromethane: acetone (1:1, v/v), which
was discarded, before eluting 5m1 extract and then leaving to dry for lmin. The
column was then eluted with 10m1 of dichloromethane: acetone (1:1, v/v), the
elutions were collected and measured (Antizar-Ladislao et al. 2005a; Antizar-
Ladislao et al. 2005b).
76
Chromatography column
Anhydrous sodium sulphate (2g)
Activated florisil (1g)
Glass wool
Figure 3.8. Diagram of clean-up column packing from Antizar-Ladislao et al.
(2005b)
The efficiency of the columns was tested using surrogate standards, by
measuring their recovery. The clean-up column was also validated by using
extracts with surrogate standards. Surrogate standards were added in two ways.
The first method involved adding the surrogate standards directly to the column.
This was done by adding 100µ1 of 100mg/L of naphthalene-d8, acenaphthene-dio,
phenanthrene-dio, chrysene-d12, perylene-d12, dodecane-d26 and tetracosane-d5o in
dichloromethane to 5m1 of dichloromethane: acetone (1:1, v/v), as the sample to
be cleaned-up. The second method used blank, soil and compost samples that had
100µ1 of 100mg/L of surrogate standard added prior to extraction. Clean-up for
each method and extract type was carried out in triplicate using 5m1 of extract.
The elutions were analyzed using GC-MS to determine the recovery of the
surrogate standards, and the TPH analytes was measured and compared for each
method.
77
3.4.2. Results of Clean-Up of Extracts
The efficiency of the recovery of TPH analytes for the clean-up methods was
compared for compost and soil extracts (Figures 3.9a and b). The mean
recovered amounts (mg kg-I) of each analyte in the sample from each method
were plotted against each other.
The slope of the trend line for the recovery of TPH analytes for compost was
1.09 (R2 = 0.93) and for the recovery from soil extracts was 1.15 (R2 = 0.92),
which indicates that the clean-up column adapted from Mills et al. (1999) was
marginally more efficient for the recovery of TPH analytes, however, as the
value of the slopes was close to 1 this indicates that the clean-up methods were
comparable.
78
60
rn 50 E
‘-• c 40 E 8 3 2 0 6 a) 30 5,
20
10
Column 1, Mills et al. 1999
Column 2, Antizar-Ladislao et al. 2005b
(a)
• •
0 10 20 30 40
50
60
TPH analyte recovery (mg kg-1) Column 2
(b)
Column 1, Mills et al. 1999
Column 2, Antizar-Ladislao et al. 2005b
0 10 20 30 40 50
60 70
TPH analyte recovery (mg kg-1) Column 2
Figure 3.9. Recovery of TPH analytes using clean-up columns from; a) compost
extracts and b) soil extracts
The recovery of surrogate standards was also examined by comparing the mean
percentage recovery of all surrogates from different matrices (Figure 3.10). The
recovery of surrogates was similar for each clean-up column, with no significant
difference. The surrogate recovery from the soil and compost samples were low
(71.9 to 72.5% and 70.5 to 72.9% respectively). Although these recoveries were
within an acceptable range they were lower than those recovered from the un-
79
cleaned extracts, showing that analytes are lost during clean-up, although
recoveries were lower than preferred, the methods were reproducible.
Sur
roga
te re
cove
ry (%
)
100 -90 -80 -70 -60 -50 -40 -30 -20 -10- 0
o Column 1
o Column 2
Surrogate Blank Compost
Soil
Matrix
Figure 3.10. Average recovery of all surrogate standards using both clean-up
columns (column 1, Mills et al. 1999; column 2, Antizar-Ladislao et al. 2005b)
Although the columns were comparable, it was decided that a clean-up step
would not be used due to the loss of analytes and the time involved in setting up
the clean-up columns.
3.5. GC-MS Analysis of Samples
In order to analyze the sample extracts a GC-MS method had to be developed to
achieve the resolution and response suitable for quantification. GC-MS analysis
was used as it provides a high degree of chemical separation and resolution,
allowing for the quantification of a range of oil chemicals (Rogues et al. 1994).
80
Analysis was carried out using a Hewlett-Packard GC-MS HP6890-5973 system
fitted with a split/splitless injector, Hewlett Packard GC autosampler and a HP5-
MS (cross-linked 5% phenyl methyl siloxane) fused-silica capillary column (30m
x 0.25mm x 0.25µm film thickness). The carrier gas was helium at a flow rate of
1.0m1/min, maintained by electronic pneumatic control.
3.5.1. Method Development
The GC-MS analysis GC program was adapted from Mills et al. (1999), by
running analyte standards under different conditions, adjusting inlet temperature,
oven temperature and ramping rate to improve separation, resolution and
response of analytes. The development of the GC programme allowed the
separation of 58 analytes (target analytes [aromatics and saturates], internal
standards and surrogate standards). The final GC program was: injection port
temperature 290°C, 1.00pL sample injected, initial oven temperature was 40°C
held for 4min, rising at a ramp rate of 8°C min-1 to 300°C final temperature held
for 15min.
In order to quantify the target analytes, a selective ion monitoring (SIM) method
was developed, which allows for monitoring of individual ions and gives greater
sensitivity (Wang and Fingas 1995). It works by detecting only those compounds
possessing the selected mass fragments in certain time-windows. Quantification
and qualification ions for each analyte were identified, and expected proportions
were programmed in to create the SIM method. The ions used for each
compound in the SIM method are shown in Table 3.3.
81
Table 3.3. Quantification and confirmation ions of target compounds Compound Quantification ion Confirmation ions Decane 57 71, 85, 55 Undecane 57 71, 85, 55 Dodecane 57 71, 85, 55 Tridecane 57 71, 85, 55 Tetradecane 57 71, 85, 55 Pentadecane 57 71, 85, 55 Hexadecane 57 71, 85, 55 Heptadecane 57 71, 85, 55 Octadecane 57 71, 85, 55 Nonadecane 57 71, 85, 55 Eicosane 57 71, 85, 55 Heneicosane 57 71, 85, 55 Docosane 57 71, 85, 55 Tricosane 57 71, 85, 55 Tetracosane 57 71, 85, 55 Pentacosane 57 71, 85, 55 Hexacosane 57 71, 85, 55 Heptacosane 57 71, 85, 55 Octacosane 57 71, 85, 252 Nonacosane 57 71, 85, 55 Triacontane 57 71, 85, 55 Hentriacontane 57 71, 85, 55 Dotriacontane 57 71, 85, 55 Tritriacontane 57 71, 85, 99 Tetratriacontane 57 71, 85, 55 Pentatriacontane 57 71, 85, 99 Pristane 57 71, 85, 55 Phytane 57 71, 85, 55 Biphenyl 154 153, 152, 155 Dibenzothiophene 184 139, 185 Naphthalene 128 127, 129 Acenaphthylene 152 151, 150, 153 Acenaphthene 153 154, 152, 151 Fluorene 166 165, 163, 167 Phenanthrene 178 176, 179, 152 Anthracene 178 176, 179, 177 Fluoranthene 202 200, 203, 201 Pyrene 202 200, 203, 201 Benz[a]anthracene 228 226, 229 Chrysene 228 226, 229, 227 Benzo(b)fluoranthene 252 126, 113, 250 Benzo(k)fluoranthene 252 250, 253, 126 Benzo(a)pyrene 252 126, 253, 207 Indeno(1,2,3,c,d)pyrene 276 138, 124, 111 Dibenzo(a,h)anthracene 278 139, 279 Benzo(g,h,i)perylene 276 138, 137, 125 Benzo(e)pyrene 252 253, 250 Perylene 252 250, 253 Dodecane-d26 66 50, 82, 98 Tetracosane-d50 66 50, 82, 98 Naphthalene-d5 136 108, 137, 134 Acenaphthene-d10 162 164, 160, 163 Phenanthrene-d10 188 184, 189, 160 Chrysene-d12 240 236, 241, 239 Perylene-d12 264 265, 260, 263 Nitrobenzene-d5 82 54, 128, 70 2-Fluoro-biphenyl 172 171, 170, 173 5-a-Androstane 245 260, 203, 81
82
Example chromatograms showing the separation, resolution and response of the
target analytes using the finalized GC-MS SIM method are shown in Figures
3.11a to c. The analyte compounds listed in Table 3.4 are represented by the high
response peaks in Figure 3.11a. The chromatograms of compost and soil samples
(Figures 3.11b and c), show the peaks of analytes of interest, which have been
identified using SIM. Due to the complex chemical nature of the samples some
of the compound peaks are not as well resolved as the standard peaks (as shown
by the shape of the chromatograms), due to the presence of a number of
compounds in the sample.
83
45-00 10.00 15.00 20.00 25.00 30.00 35.00 40_00
1800000
1600000
1400000
1200000
1000000
800000
600000
400000
200000
TIC lvv151105-12.D (c)
ado°. anon.
nob ea..°
TicKr...219 1 099-00. 19
(a)
150009 140009
120000
120000
110000
100000
90000
90000
70000
a0000.
50000
40000-
30000
20000
10000
1 1 4_ 100 15_00 2 5. 00 30.0035.00 4 0! 00 .45' 00
Abundance
TIC: Ny151105-10.0
(b)
1 0.00 1 5.00 20 00 25.00 30.00 35.00 40.00 45.00
Abundance
Figure 3.11. Chromatograms of analyzed samples of; a) standard mix of all
target analytes, b) composted contaminated soil extract and c) contaminated soil
extract
84
65000 60000 55000, 50000, 45000 40000 35000 30000 25000 20000 15000 10000 5000
Prior to the analysis of experimental sample extracts, calibration of the GC-MS
was performed using internal standards. Calibration ensures GC-MS efficiency
and accuracy throughout the project, to give true to values for analyzed extracts.
Each compound standard or compound standard mixture was calibrated over the
range 5mg/L to 0.01mg/L. The calibration solution (0.9m1) was transferred to a
GC-MS vial using a micro-syringe, then 0.1m1 of the internal standard mix was
added. Analysis of each calibration concentration for each compound or mix was
performed in triplicate. The results of calibration were then used to calculate the
relative response factor (RRF) for each compound to be used in the calculation of
contaminant concentrations (Equation 3.3). Additionally, the precision of the
instrument at each calibration level was examined by examining the linearity of
the compound responses (R2) and variations (RSD values). The regression
coefficient R2 was used to show if the response of the calibration range 0.01-
5mg/L was linear. The RSD values show the precision of the calibration levels.
To quantify the data obtained from GC-MS analysis the following equations
were used to calculate the concentrations of contaminants in mg kg t . The
concentration of contaminants in the sample extracts was calculated as:
(3.3)
[ internal standard response
RRF value
Concentration =
analyte response
x
internal standard concentration
Note: RRF (relative response factor) is a measure of the relative response of an analyte compared
to an internal standard
85
[ concentration x volume of extract (3.4) Amount (mg/kg) —
weight of ashed sample
The weight of the contaminants in the ashed samples was calculated as;
3.6. Quality Assurance and Quality Control
Quality assurance is the overall approach taken to minimize the potential for
errors and provide adequate confidence that results are reliable and accurate
(Crosby et al. 1995; ATSDR 2001; Reeve 2002). Quality assurance measures
include sampling, sample storage, replicates, reagent purity and apparatus
cleanliness, instrument checks and calibration (Reeve 2002). Quality control is
the measures used to ensure that the results are valid and quantifiable, which
include blanks, surrogate spikes, replicates, internal standards and calibration
standards (Reeve 2002). The measures taken in this study are summarized in
Table 3.4 and details of measures taken are discussed throughout this chapter.
86
Table 3.4. Quality assurance and quality control measures
Measures Monitoring
Sample acquisition Samples were uniquely labelled with
type, source and date of acquisition
Chemical and reagent acquisition Analytical-grade chemicals were used
where possible. The source and expiry
date were checked and noted
Preparation of samples and reagents Soil samples were air dried and stored
in sealed containers at room
temperature. Compost samples were
stored in sealed bags at -18°C.
Chemicals and reagents were stored
under conditions specified by the
manufacturers
Cleanliness of glassware All glassware was washed and stored
in a designated glassware cupboard
Maintenance of equipment All electrical appliances were checked
on a regular basis and the calibration of
machinery required for analysis was
performed prior to use
Documentation and record-keeping Information on samples and
experiments was recorded in a
notebook
Calibration of equipment The GC-MS was auto-tuned before use
and calibrated using calibration
standards
87
Replication Where possible samples were analyzed
in triplicate
Recovery Soil, compost and blanks were spiked
with a known concentration of
surrogate standard solution before
extraction to measure recoveries.
Internal standards were added for GC-
MS analysis to quantify target analytes
Blank A reagent blank was included in each
experiment
Quality control measures were used in the development of the method and for
validation of methods during the project, such as continuous validation of the
extraction method using surrogates and testing recovery of analytes during GC-
MS analysis.
3.6.1. Continuous Validation of Extraction Methods Using Surrogate
Standard Recoveries
Surrogate standards are used to test the recovery of analytes, and monitor for
matrix effects. The surrogate standards are compounds that are similar to the
target analytes being tested (Crosby et al. 1995). The surrogate standards used
were; dodecane-d26 and tetracosane-dso for the saturates; naphthalene-d8 for the
2-ringed aromatics; acenaphthene-dio and phenanthrene-dio for the 3-ringed
aromatics; chrysene-d12 for the 4-ringed aromatics; and perylene-d12 for the 5-
88
and 6-ringed aromatics. An acceptable range of surrogate recovery is 80-120%
for extraction. However, for calculation of analytes when the recovery is below
90%, the concentrations of target analytes are adjusted depending on the
recovery achieved.
The efficiency of the extraction method was monitored throughout the project,
rather than carrying out just an initial validation of the method, which assumes
performance will be the same throughout. Testing the efficiency at regular
intervals shows the continuing validity and efficiency of analyte recovery for
analysis.
3.6.2. Surrogate Standard Recovery Method
The recovery of surrogate standards was tested in three different matrix types:
blanks (HydromatrixTM), soil and composted soil. Seven-gram samples of each
matrix were extracted in triplicate using ASE® method (Section 3.4.1). Prior to
extraction, the samples were spiked with 100111 of 100mg/L surrogate standard
mix (dodecane-d26, tetracosane-d50, naphthalene-d8, acenapthene-d8, chrysene-du
and perylene-d12). Additionally, un-spiked blanks were used to check for
contamination in samples during packing or from the machine. Surrogate
recoveries of the extraction method were carried out every 6 months during the
project (five recovery experiments in total). The percentage recoveries of
surrogates were calculated (Equation 3.5) to examine the efficiency of extraction.
surrogate concentration found Recovery (%) =
x 100 surrogate concentration added
(3.5)
89
3.6.3. Surrogate Recovery Results
The recoveries of total surrogate standards in each matrix type for each recovery
experiment were compared (Figure 3.12). There was high surrogate recovery for
each of the recovery experiments (89 to 102%) throughout the project,
demonstrating that during the project the extraction method results in high
analyte recovery, giving accurate analyte results. Recoveries for all the matrix
types were high, indicating that matrix effects did not interfere with analyte
recovery. Additionally, differences in recovery between the different matrix
types over the project duration was <10% (4.51 to 8.09%), showing the
consistency of the extraction method for all matrix types.
120 -
100
80- >as 0
60 -
cn 40 -
20 -
El Blank
Compost
❑ Soil
0
1
2 3
4
5
Recovery analysis
Figure 3.12. Percentage recovery of total surrogates from project surrogate
recovery experiments
3.6.4. Continuous Analysis
Performance of the GC-MS analytical method was also verified with continuous
analysis of one sample, to show the response consistency when a large number of
90
samples were analyzed consecutively. For this, a compost extract sample was
analyzed 15 times and the responses analyzed. Variation in the total response of
sample analytes was compared by using the response from the first repeat and
calculating the percentage variation of the other repeats compared to it (Figure
3.13).
1.5 -
• • • • •
0.5 - • c • • o 0• • ...o co L. j3 -0.5 -
• •
• • •
-1 - • -1.5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Repeat
Figure 3.13. Percentage variation of sample analyte total response during
repeated analysis of a sample
There was little variation in the sum compound responses between repeat
analyses (-1.11 to 1.30%). Thus, analysis of multiple samples gives reliable and
consistent results. The variations in responses of compounds over time were
generally small, mostly below 5% (the range was 0.22 to 9.92%). Therefore,
repeated analysis does not result in large variations of analyte response away
from the true value. Two compounds had high RSD values (7.31% and 9.92%),
which could have been due to variations within the extract or analytical
variability. However, since they were <10%, they were regarded as being within
91
an acceptable range. Therefore, analyzing large numbers of extracts would result
in accurate responses of analytes of interest during the project.
3.6.5. Statistical Analysis
Experimental results were statistically analyzed using Genstat 8.1. Comparison
of TPH removal under different composting conditions was investigated using
paired t-tests. Analysis of variance (ANOVA) was used to test the effects of
treatment parameters on the removal of TPH during composting. Statistical
significance was set at p<0.05 unless otherwise stated.
3.7. Conclusions
The analytical method used for this project involved extraction of samples using
ASE® and analysis of the extracts using GC-MS in SIM mode. The ASE® system
was used as the extraction method due to its high analyte recovery, rapid
turnover of samples and straightforward use. No clean-up step was used, as
analytes were lost during this stage and the process was time consuming.
Therefore, it was decided to analyze the sample extracts from the ASE® and
replace the GC capillary column frequently to avoid compound responses
becoming less well resolved. The analysis of extracts using GC-MS has been
shown to be consistent and accurate for the analysis of TPH from soil and
compost extracts. The development of robust analytical protocols and adherence
to good QA/QC practices achieved low variation in analytical responses during
GC-MS analysis coupled with good reproducibility during extraction. Thus, it
92
was possible to compare, with confidence, not only measurements made at
different times during an individual composting experiment, but also to compare
the results obtained between different experiments.
93
Chapter 4. Pilot-Scale Composting Study of In-Vessel
Composting Bioremediation of Electrical Insulating Oil-
Contaminated Soil
4.1. Introduction
Although mathematical composting models can be used to provide an idea of
how control measures affect different conditions during the process, they rely
heavily on estimated parameter values, therefore, it is necessary to verify the
assumptions used by experimental testing (Korner et al. 2003). Pilot-scale studies
assist in understanding the effect of the composting parameters on compost and
the interactions that occur within composting systems (VanderGheynst et al.
1997). They give an indication on the usefulness and efficiency of the processes
being tested and potential for scaling-up to industrial-scale (Korner et al. 2003;
Kim et al. 2008). Different scales can be used from laboratory-scale to pilot
scale.
Crown BioSystems carried out a pilot-scale study on the potential of using green
waste collected from vegetation management around electrical transmission
devices, in the bioremediation of electrical insulating oil-contaminated soil. Their
approach was to use an accelerated bioremediation process, combining the two
waste streams in a system where composting parameters could be controlled.
The aim of the study was to develop a cost-effective process for the clean-up of
electrical insulating oil-contaminated soil. The clean-up had to reach a standard
that was acceptable for re-use of materials on site, such as at a sub-station, or re-
application of soil onto a site.
94
Their objectives were:
1) To develop a suitable mid-scale composting system and operating procedures
(measurements, monitoring and aeration).
2) Determine optimal composting conditions: green waste to soil ratio; green
waste quality (C:N, amount of leaf material compared to wood); addition of
nutrients (nitrogen); moisture content; aeration and temperature ranges.
Samples were obtained from this study once the experimental phase had been
completed, to enable the analysis of TPH contamination and better understand
the usefulness of their composting method. ..However, there were a number of
limitations to this study, which need to be established at the outset, because they
have a marked affect on the results obtained and the conclusions reached. Firstly,
no Time 0 sample was available, so comparisons within and between treatments
were difficult as Time 1 had to be used as the 'initial' concentration. In an effort
to overcome this a Time 0 sample was prepared from available soil and green
waste samples, to provide reasonable hypothetical starting concentrations against
which the other time interval concentrations could be normalized (Section 4.2.1).
However, the concentrations of most of the target analytes were lower than their
respective concentrations in the Time 1 samples collected from the Crown
BioSystems study. This may have been due to the available soil sample used for
the Time 0 sample not being representative of the soil used in the trial.
Additionally, the green waste samples available had been oven dried and
therefore volatiles, including short chain saturate volatiles and volatile PAHs,
would have been lost. Therefore, the results presented in this chapter are with
95
reference to the concentrations found in the Time 1 samples. Secondly, only a
single composite sample was provided for each sampling interval for each
treatment. Therefore, the extent of variation within treatments was impossible to
determine. Since only a single sample was taken from each treatment these were
thoroughly homogenized in a food processor to ensure that they were as
representative as possible of the sample for each barrel. Additionally, there were
no treatment replicates, because the number and times of sampling intervals, as
well as the length of composting was different for the treatments. Combined with
the lack of Time 0 samples, this made comparison within and between treatments
difficult.
4.2. Materials and Methods
The contaminated soil was from a transformer leak; it was homogenized and
large lumps of clay were broken up. Two green wastes in the form of wood and
vegetation chippings were used: poplar and alder. Green waste in this context
was the fresh material collected from the management of vegetation around
electrical transmission devices.
Modified steel barrels were used as mesocosms. Holes were drilled for sampling,
ventilation and monitoring and vanes were fitted internally to aid mixing of
contents (Figure 4.1). The barrels were mounted on rollers attached to wooden
pallet platforms to allow turning. All the mesocosms were insulated with rock-
wool and carpet underlay (Figure 4.2).
96
Figure 4.1. Steel barrels illustrating internal vanes and filled height
Figure 4.2. Insulation of steel barrels
The treatments used combined different green wastes, with different soil to green
waste ratios, external heating, fertilizer addition and varying duration of
composting. The treatments are shown in Table 4.1 followed by their respective
sampling intervals in Table 4.2.
97
Table 4.1. Treatments used in the Crown BioSystems pilot-scale study
Treatment Green S:GW Fertilizer Heating Duration
waste (v/v) (days)
1 poplar 1:1 x x 134
2 poplar 1:4 x x 134
3 poplar 1:4 x x 98
4 poplar 1:4 ✓ x 98
5 alder 1:2 x x 110
6 alder 1:4 ✓ ✓ 110
7 alder 1:4 ✓ ✓ 134
8 alder 1:4 x x 110
9 alder 1:4 x ✓ 110
10 alder 1:4 ✓ ✓ 98
11 alder 1:4 ✓ 3( 134
12 alder 1:4 ✓ x 98
13 alder 1:4 x x 98
14 alder 1:4 ✓ x 110
Treatment 4: fertilizer was urea, added in three doses
Other treatments used ammonium nitrate as fertilizer, added in three doses to adjust total nitrogen
to 3% (w/w)
Heating provided by electric blankets
Table 4.2. Sampling intervals of treatments (days after start of composting)
Sampling interval
Treatment number 1 2 3 4
1, 2 71 83 134
3, 4 35 47 98
5, 6, 8, 9, 14 9 36 68 110
7, 11 9 36 68 134
10, 12, 13 9 36 58 98
98
Moisture content was adjusted to 50-60% (w/w) by adding water to set
proportions of soil and green waste and was maintained by adding more water
when required. Aeration of the composting mixture was carried out twice a week
by rotating the barrels, to mix the composting mixture; also an extractor fan was
used for forced aeration.
Sub-samples of the starting materials (soil and green waste) were collected prior
to composting, for determination of initial moisture content, composition (ratio
of leaf to woody matter) and chemical analysis (TPH - total petroleum
hydrocarbons).
Throughout the trial, temperature and carbon dioxide emissions were monitored
using data loggers. Prior to sampling, the barrels were turned to mix the contents,
then sub-samples were collected from different positions within the barrel and
combined to form a composite sample for that treatment and time-point.
4.2.1. Preparation of Time Zero Samples
The Time zero samples were prepared in triplicate from samples of soil and dried
green waste provided by Crown BioSystems. The soil was homogenized in a
pestle and mortar and passed through a 2mm sieve to remove stones and debris.
The green waste was also homogenized in a food processor to reduce the size and
make it more uniform. The Time 0 samples were made up by volume, using a
25ml beaker, to the proportions specified (soil: green waste): poplar (1: 1 and
1:4) and alder (1:2 and 1:4). The components were mixed and homogenized in a
99
500m1 beaker, and the moisture content was adjusted to 60% by the addition of
the appropriate amount of distilled water. The components were re-mixed to
homogenize the wetted mixture. The samples were then extracted using ASE®
and analyzed using GC-MS (as described in Chapter 3).
4.2.2. Preparation and Analysis of Samples
The samples were stored at -18°C and prior to analysis were defrosted at room
temperature. To ensure the samples were homogeneous, they were blended in a
food processor to ensure even distribution of green waste and contaminated soil,
as well as to reduce the size of the green waste to make it a uniform size and ease
mixing with the contaminated soil. The samples were extracted using ASE®, and
the extracts were analyzed using GC-MS (as described in Chapter 3).
Additionally, samples were dried and ashed to calculate the moisture content and
total organic matter (Chapter 3).
For analysis, the term Total Petroleum Hydrocarbons (TPH) in this case was
taken to be the sum of all target saturates and aromatics, which are listed in
Section 3.1.
Due to the limitations outlined in the introduction to this chapter the data was
normalized to the first sample interval, Time 1, which was taken as the 'start'
concentration. Therefore, Time 1 was taken as having a value of 1, Time 2 was
T2/T1, Time 3 was T3/T1 and Time 4 was T4/T1, thus reporting each sample
interval for the treatment as a fraction of their 'initial' Time 1 concentrations.
100
Normalizing the data in this way made the results more comparable, allowing
some conclusions to be drawn. However, it must be acknowledged that the key
limitation in doing so is that the kinetics of TPH losses during the earliest part of
the experiment when lag-phase or rapid loss might have occurred remains
unknown. Furthermore, there were no replicates of samples from treatments, so
that no statistical analysis of the results was possible to verify the significance of
differences resulting from treatments and treatment parameters. In these respects
the experimental design was weak, but nevertheless the availability of these
pilot-scale samples was exploited because it is in an effort to glean some useful
information in an area that has seen little research to date.
4.3. Results of Pilot-Scale Study
These results are presented with respect to the limitations of the experimental
design, which were noted in the Section 4.1.
4.3.1. Contribution of Hydrocarbon Groups to TPH
The importance of saturates in terms of their contribution to TPH at the 'start'
(Time 1) of the process was small (Figure 4.3), since they contributed 8.9 to
43.1%. However, aromatics contributed the most to TPH (56.9 to 91.1%). The
same pattern was exhibited at the end of the process (Figure 4.4), with saturates
contributing 4.8% to 32.5% and aromatics contributing 67.5 to 95.2% of TPH in
treatments. This suggests, from an engineering point of view, that saturates are
less important in the composition of TPH than aromatics, and that the focus of
101
remediation should be on removing the aromatic fraction. The losses of each of
these hydrocarbon fractions varied between the treatments. The loss of saturates
ranged from 28.2 to 66.6%, while the aromatic fractions losses ranged from 11.3
to 59.5%. Therefore, although both fractions were removed to some extent, the
aromatic fraction had lower removal than the saturates.
2 120000 co
100000 - a)
80000 3) 60000 - a
• 40000
20000 -0 c 0 (.) 0
• TPH
❑ Aromatics
0 Saturates
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Treatment
Figure 4.3. Contribution of saturates and aromatics to TPH at the 'start' (Time 1)
of composting (for treatments, see Table 4.1)
2 70000 U) co 60000 0 3) 50000
E 40000
30000
bird 20000
8 10000 a 0 • 0
■ TPH ❑ Aromatics 0 Saturates
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Treatment
Figure 4.4. Contribution of saturates and aromatics to TPH at the end of
composting (for treatments, see Table 4.1)
102
4.3.2. Relative Rates of TPH Loss
The rate of TPH loss was determined by using the normalized data in order to
enable direct comparisons to be made between treatments. The rate of loss using
the normalized data was calculated as the normalized amount lost (i.e. 1 —
fraction left at the end). This loss was calculated as a percentage loss, which was
then divided by the number of days composting to give the normalized
percentage loss per day.
Overall, the normalized relative rates of loss of TPH were highly variable
between composting treatments (Figure 4.5), ranging from 0.14 to 0.56% day-1,
Treatment 14 had the highest rates of removal, with other treatments having rates
approaching this, while Treatment 4 had the lowest removal. This variation in
removal rates could be due to the variation in the removal of aromatics, as the
patterns of rate loss are similar for treatments, which also suggests that aromatics
contribute an important fraction to TPH, thereby affecting its removal.
Nor
mal
ized
loss
of T
PH ( %
day
1 0.6 -
0.5 -
0.4 -
0.3 -
0.2 -
0.1 -
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 Treatment
Figure 4.5. Rates of normalized relative loss of TPH per day from treatments
(for treatments, see Table 4.1)
103
4.3.3. Kinetics of TPH Loss
First order kinetic analyses were performed for the compost treatments and
controls. The pseudo-first-order kinetic approximation was applied using the
linear integrated form of equation (4.1):
In (C/C0) = -kt (4.1)
where C is the concentration at time t, Co is the concentration at t = to, k is
the first-order constant of removal (obtained by linear regression) and t is time
(Admon et al. 2001; Antizar-Ladislao et al. 2005a).
For these results the sampling intervals relate to the four intervals at which the
treatments were sampled. The number of days at which they were sampled is
shown in Table 4.2.
The loss kinetics of TPH varied between all the treatments, with fast, slow as
well as loss rates between these loss patterns. Examples of these differences in
loss kinetics of TPH are illustrated in Figure 4.6.
Treatment 9 exhibited the slowest rate of TPH loss (k = 0.04 day 1), Treatment 12
exhibited a kinetic loss between these two other rates (k = 0.10 day-1), while
Treatment 14 exhibited fastest kinetic loss of TPH (k = 0.20 day-1), which was
double the loss of Treatment 12. The R2 values for these treatments were low (R2
values: Treatment 9 = 0.72, Treatment 12 = 0.88 and Treatment 14 = 0.81),
which could indicate that these were not first-order, but since there was no Time
0 sample this cannot be drawn as a definite conclusion because it is not known
104
♦ Treatrrent 9
0 Treatrrent 12
Treatrrent 14
RN,
what occurred in the initial stage of composting. It should also be noted that the
number of sample points was very limited, so the data should be interpreted with
caution.
0 fa
-0.1
-0.2 -
-0.3 -
-0.4 -
-0.5 -
-0.6 -
-0.7 -
-0.8
• • •
0
4
Treatment 9 k= 0.40 day' Treatment 12 k= 0.10 day' Treatment 14 k= 0.20 day'
es •
1
2 3 4
Sampling interval
Figure 4.6. Slowest, fastest and intermediate kinetic losses of TPH from treatments
4.3.4. Remediation Targets
There is no set acceptable level for TPH, so it cannot be stated whether the
concentrations are acceptable for re-use of the end product (Figure 4.7).
Although there is no set level, it is encouraging that in all treatments there has
been a reduction in TPH ranging from at least 14.3 to 61.5% of their
concentrations prior to composting bioremediation. This suggests that there is
scope to optimize the composting process and reduce final contamination even
further. The changes in concentration for the original compost sample mass were
also estimated, using ash weight and weight of sample analyzed (Figure 4.8). The
values for the compost sample (soil and green waste mix) were lower than those
105
for the ash, because the sample mass was greater. This shows that, for the end
product, final concentrations approached an acceptable level, which could be
further improved through optimization.
0) 140000 .
E E 120000 -
x 100000 -
c
a.
1 0000 - 8
Ittill11111,1111
o c ..8 60000 - tic vs 40000
i 20000 - c U 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Treatment
Figure 4.7. Initial and final concentrations of TPH from treatments
■ Initial 0 Final
80000 -70000 - 60000 - 50000 -40000 -30000 -20000 -10000 -
0
MI Initial 0 Final
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Treatment
Figure 4.8. Estimated initial and final concentrations of TPH in compost samples
Although there is no limit for TPH, the aromatics (i.e. PAHs) have a limit of
100mg/kg that has been set for a sum of 17 PAHs (Defra 2004) (Appendix 4.1).
106
Y_ 5)
Z o
o g- O 0 0
0 0 U
(mg/
kg o
f ash
) C
once
ntra
tion
of a
rom
atic
s 100000 90000 80000 70000 60000 50000 -40000 30000 20000 10000
0
■ Initial 0 Final
In the present study, it was apparent that this statutory level was not been
achieved and levels greatly exceeded the recommended (Figure 4.9).
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Treatment
Figure 4.9. Initial and final concentrations of aromatics after composting
bioremediation
In terms of aromatics in the compost mixture the results are shown in Figure
4.10. As with total TPH, the concentrations of the aromatics are lower per
kilogram, but the limit is still greatly exceeded. Optimization of the composting
process with a focus on the aromatic fraction of the TPH should result in greater
reductions and the possibility of reaching the accepted statutory level.
107
Con
cent
ratio
n o
f aro
mat
ics
(mg /
kg o
f com
p ost
)
60000 -
50000 -
40000 -
30000 -
20000 -
10000 -
R hhh
• Initial
0 Final
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Treatment
Figure 4.10. Estimated initial and final concentrations of aromatics in the
compost during treatments
Overall, treatments 3, 9 and 14 resulted in the lowest concentrations of TPH and
aromatics in the compost mass, showing that these treatments have greater scope
for realizing the accepted levels for contamination. The parameters of the
treatments could be further improved, possibly resulting in the accepted criteria
being reached. Examination of the parameters that result in the best removal are
analyzed in the next section.
4.3.5. Importance of Treatment Parameters
Within the context of this study the conditions of the treatments appear to affect
the extent of losses of contamination. The losses of TPH from treatments were
categorized into small, medium and large values from the normalized data for
each treatment. The categories were assigned by examining the distribution of
the normalized ETPH data for the treatments, then categorizing the data into
groups, depending on the loss. The categories were set arbitrarily and were as
108
follows: small, a loss of 0.2 or less; medium, 0.21 to 0.4; and large, 0.41 or more.
These categories were used to identify parameters important in the composting
process (Table 4.3).
Table 4.3. Losses of compound groups from treatments
Small losses Medium losses Large losses
Treatment
1
2
3
4 ✓
✓
✓
✓
5 ✓
6 ✓
7 ✓
8 ✓
9 ✓
10 ✓
11 ✓
12 ✓
13 ✓
14 ✓
There were large losses of TPH from Treatments 6, 8, 11, 12 and 14. Treatments
6, 11, 12 and 14 had added N (ammonium nitrate), whereas Treatment 8 did not,
but the indication is that the addition of a nitrogen source results in larger losses
of compounds. There is no evidence that external heating makes a difference to
the amount of TPH lost, as most of these treatments except Treatment 6 had no
external heating.
109
Overall, for this study it appears that the addition of ammonium nitrate as a
nitrogen sources increased the losses from treatments, but external heating does
not have an effect. Additionally, the first four treatments utilized poplar as its
green waste, but the rest used alder, and those treatments had generally higher
losses for the compound groups.
4.4. Discussion
The results of the Crown BioSystems pilot-scale study indicate that there is some
potential for the use of composting bioremediation as a clean-up method for
electrical insulating oil-contaminated soil. However, the removal of TPH was
low (14.5 to 61.5%) compared to other bioremediation studies (Table 4.4);
therefore further, optimization of the composting process is needed.
Table 4.4. Losses of petroleum hydrocarbon contamination from bioremediation
studies
Loss (%) Treatment
Length
Contamination Reference
98.1- 98.4 30d Diesel (Namkoong et al. 2002)
93.8 — 96.2 8wks TPH (Sarkar et al. 2005)
84 12wks diesel (Van Gestel et al. 2003)
79 360d TPH (Sabato et al. 2004)
64.5 — 82.5 12 months TPH (Balba et al. 1998)
110
In terms of the oil fractions being removed, their contribution to TPH is also
important. The saturates contributed only a small fraction to TPH at the 'start'
and end, and therefore they pose little risk. However, the aromatics contributed
the greatest proportion to TPH, thereby posing a greater risk to the environment
and health, particularly in view of their generally slow rates of removal, which
means that the focus of remediation needs to be on them. It should be noted,
however, that because there was no Time 0 sample taken for proper comparisons
to be made, the 'start' proportions of the fractions may have been different, with
saturates providing a far greater proportion of the initial TPH, although the
general pattern indicates that this was not so.
The removal rates of TPH in some treatments shows the potential of this process
for removing oil contamination. The relatively fast removal rates reduce the
potential risk of the contamination and also reduce the time needed for
remediation, which has been a criticism of bioremediation in the past (Cole
1998). The composition of the contaminating oil appears to affect the rate of loss
of TPH, as the losses of TPH follow a similar pattern to that of the aromatic
fraction (Figures 4.3 and 4.4).
The kinetic losses of TPH from treatments were variable, which may have been
due to variations within the barrels making conditions more or less favourable for
degradation. Additionally, the different treatment conditions used would have
modified the composting environment resulting in different rates of kinetic loss.
111
Although there is no statutory limit for TPH in soil, the reductions in
concentration in all the treatments is encouraging, indicating that there is scope
for using composting to remove contamination, although optimization is needed.
The larger reduction of TPH in some treatments shows that the modification of
conditions within the treatments allows greater removal. Therefore, further
optimization of the process is likely to result in far greater aromatic and TPH
losses.
The variation in rates of loss, loss kinetics and overall TPH removal appears to
some extent to depend on the treatment conditions, with several treatments
having high removal. Large losses of TPH were found in several treatments
(Table 4.3). Their common parameters were the addition of a nitrogen source as
ammonium nitrate. The addition of nitrogen provides an utilizable source of
nitrogen and overcomes the nutrient limitations (Coulon et al. 2005). Nitrogen
becomes limiting as the carbon to nitrogen ratio increases, resulting in slower
rates of decomposition (Nordstedt et al. 1993).
Most of the treatments did not have external heating, so an external heat source
may not be necessary as heat is generated within the matrix during composting,
with mesophilic temperatures ranging from 30-45°C (Golueke 1991), which
should be sufficient for degradation to occur. The high temperatures also help to
desorb contaminants and make them more available for degradation (Cole 1998)
and also increase enzyme activity (Leahy and Colwell 1990). The additional
external heating was provided by electric blankets in a relatively uncontrolled
manner with heat transfer efficiency to the contents depending on how close the
112
barrel was to the mantle. The length of the treatments was also generally 110
days, so the process takes a few months: optimization may reduce this length of
time and also remove the contamination. Alder appears to be a better green waste
than poplar, as it had more leafy material, providing a more readily utilizable
nutrient source for the microorganisms (Tuomela et al. 2000).
4.5. Conclusions
In conclusion, there is scope for using composting bioremediation for the
remediation of electrical insulating oil-contaminated soil, although the rates and
total removal of TPH varied, and the greatest removal, still left an unacceptable
quantity of TPH contamination. Further optimization of the composting process
could reduce contamination further to acceptable set levels, so that the compost
can be re-used.
113
Chapter 5. Laboratory Optimization of Bench-Scale In-Vessel
Composting of Electrical Insulating Oil-Contaminated Soil
5.1. Introduction
The Crown BioSystems pilot-scale study analysis (Chapter 4) reported TPH
losses ranging from 14.3% to 65.5%. Although this shows scope for the use of
composting bioremediation to remove electrical insulating oil associated TPH
contamination, the removal was not great enough for the composting process to
be used commercially. Therefore, this suggests that composting processes need
to be optimized further in order to achieve greater and faster removal of TPH,
making them more efficient and achieving contaminant target values in a short
time period. Optimization can be achieved through the control of composting
parameters, such as soil to green waste ratio, operating temperature and moisture
content of the compost mix. By combining different levels of each of these
parameters optimal bench-scale contaminant removal operating parameters can
be found, which can be used to inform the development of pilot-scale and full
commercial-scale operations. To determine the optimal parameters for
composting of electrical insulating oil-contaminated soil bench-scale
experiments were designed, based on the methodology previously developed by
Antizar-Ladislao et al. (2005b) for their investigation of composting-
bioremediation of coal-tar contaminated soils.
Composts have excellent potential for bioremediation, as they are capable of
sustaining diverse populations of microorganisms, with the ability to degrade a
variety of organic contaminants (Scelza et al. 2007). Composting bioremediation
114
of hazardous wastes has been shown to be effective in biodegrading PAHs,
chlorophenols, polychlorinated biphenyls, explosives and petroleum
hydrocarbons at laboratory and field-scales (Semple et al. 2001; Antizar-Ladislao
et al. 2004).
In-vessel composting systems control and accelerate organic waste
decomposition by creating ideal conditions for the microorganisms to degrade
contaminants. This is achieved by bringing the compost mix up to the optimum
temperature as quickly as possible. The supply of air increases the microbial
metabolic activity and cools the compost, allowing higher rates of microbial
activity than would normally occur in conventional composting (Edwards 1998;
Antizar-Ladislao et al. 2004). In-vessel composting takes place in a partially or
completely enclosed container, in which environmental conditions can be
controlled (e.g., temperature, moisture content and mix ratios). Enclosed vessels
closely approximate a laboratory incubator, with the compost and its associated
microflora being exposed to a more even temperature profile (Antizar-Ladislao et
al. 2004).
The advantages of in-vessel composting over windrow systems is that it requires
less space, provides better parameter control and provides a higher process
efficiency (Kim et al. 2008). It also accelerates stabilization of waste material
(Clean Merseyside Centre 2005). Aeration helps to sustain microbial activity, by
maintaining an open structure in the compost (Edwards 1998) and is also
important in removing carbon dioxide and excess moisture (Stelmachowski et al.
2003; Mohee and Mudhoo 2005).
115
The factors that affect biodegradation rates of contaminated wastes include the
numbers of microbes, sufficient oxygen or nutrient availability, temperature and
water availability (Godoy-Fanndez et al. 2008), which can be controlled during
composting bioremediation.
Adding an organic matrix such as green waste to contaminated soil increases the
general microbial activity and the activity of specific degraders, which might be
found in the contaminated soil or introduced with the organic material (Jorgensen
et al. 2000). Modifications in the physical and chemical microenvironments
within the compost can increase the diversity of the microflora to which the
contamination is exposed. Additionally, substances within the organic matrix
may enhance co-metabolism (Antizar-Ladislao et al. 2004; Haderlein et al.
2006). Using the correct ratio of contaminated soil to organic amendment is
important, because an inappropriate ratio may retard or inhibit microbial activity
(Namkoong et al. 2002). In the experiments described in this chapter, different
soil to green waste ratios were examined to find the best ratio for the effective
treatment of electrical insulating oil-contaminated soils.
Water is essential to microbial activity by easing microbial migration and
colonization, and the diffusion of substrates and metabolic wastes. However,
water also tends to plug pores and impede gas exchange. Therefore, an
appropriate balance should be maintained (Finstein and Hogan 1993). Organic
materials have a wide range of moisture contents, but a value of 50-60% (w/w)
provides the best conditions for composting. Contents below 45% will inhibit
microbial growth, while excess water, above 70 % will inhibit the flow of air
116
(Edwards 1998). Therefore, the optimum moisture content for composting
bioremediation needs to be established, which was done by testing three different
moisture contents (40, 60 and 80%).
Temperature affects microbial metabolism and population dynamics (Laine and
Jorgensen 1997). High temperatures achieved during composting (50°C or
higher) accelerate the relatively slow chemical reactions in soil (Cole 1998).
Elevated temperatures stimulate hydrocarbon degradation and enhance the
contaminant bioavailability to metabolism by increased solubility and mass
transfer (Feitkenhauer et al. 2003; Van Gestel et al. 2003; Antizar-Ladislao et al.
2004). Elevated temperatures can also increase the rate of enzyme reactions
involved in the biodegradation process (Antizar-Ladislao et al. 2004). Research
suggests that a temperature range between 40 and 45°C allows maximum
diversity to be expressed for bacteria, actinomycetes and fungi (Antizar-Ladislao
et al. 2004). Additionally, a constant temperature is better for the removal of
contamination than varying the temperature, as variations may negatively affect
native microflora (Antizar-Ladislao et al. 2006). To find the optimum operating
temperature for the degradation of contamination, three temperatures were used:
38, 55 and 70°C.
This chapter considers the engineering dimension of the composting process, the
set-up of the compost system and the operating parameters. It investigates the
effects of controlling specific parameters (soil to green waste ratio, operating
temperature and moisture content) in a laboratory-scale composting system, in
117
order to determine the optimum combination of parameters for composting to
remove TPH contamination.
5.1.1. Hypotheses
Optimization of composting parameters (moisture content, soil to green waste
ratio and temperature) will result in higher rates of contamination loss and
greater overall contaminant removal. Optimized levels of these parameters are
envisaged to be a mesophilic temperature (between 35 and 50°C), an
intermediate soil to green waste ratio e.g. 0.8:1, and a moisture content ranging
from 50 to 70%.
5.1.2. Aims and Objectives
The aims of the optimization experiments were to:
1) Determine the optimal operating parameters (soil to green waste ratio,
temperature and moisture content) for composting bioremediation of
electrical insulating oil-contaminated soil.
2) Assess the impact of these different operating parameters on removal of
contamination.
118
In order to achieve these aims, the following objectives were developed:
1) Set up a bench-scale in-vessel composting system for electrical insulating oil
contaminated-soil.
2) Using this system, compost contaminated soil under different levels of the
test parameters.
3) Quantify loss of contamination from treatments to determine the optimal
operating conditions.
5.2. Materials and Methods
Two experiments were carried out to optimize the composting process. Firstly,
the soil to green waste ratio and temperature were optimized and then the second
experiment optimized the moisture content using the pre-determined optimum
temperature and soil to green waste ratio conditions. This methodology
introduces the composting materials, a description of the bench-scale composting
system and sets out how the experiments were performed.
5.2.1. Contaminated Soil
The electrical insulating oil-contaminated soil was obtained from EDF Energy.
The soil was London clay heavily contaminated with electrical insulating oil
from a transformer leak, collected from a site in north Kent, United Kingdom.
The initial concentration of TPH was —21, 000 mg kg' of ash.
119
The soil was air dried in a fume cupboard, before being homogenized.
Homogenization was achieved by breaking up larger pieces in a food processor,
then grinding the soil using a pestle and mortar. The ground soil was then passed
through a 2mm sieve to remove stones and other debris, and collected. The dried
ground soil was stored in an air-tight, labelled container at room temperature
until use.
The moisture and ash content of the soil was determined, by firstly drying 5g
sub-samples of the wet soil at 110°C for 24h and calculating the moisture
content. The dry samples were then ashed in a muffle furnace at 550°C for 12h,
and the ash content calculated (as described in Chapter 3). The average moisture
content was 2.30% + 0.2, the average ash content was 92.0% ± 0.01, and the pH
was 6.03 ± 0.01.
5.2.2. Green Waste
The green waste was in the form of vegetation chippings from vegetation
management around EDF Energy power lines in Kent. It was provided by
Brockwells Forestry Ltd (contractor used by EDF Energy). The chippings
consisted of beech, oak, holly, sycamore and conifer. The green waste was
homogenized in a food processor to reduce the size of the chippings and provide
a more even distribution of the green waste components; some of the larger
pieces of wood were removed. If the green waste was not used immediately, it
was stored at -18°C, to prevent decomposition and was defrosted at room
temperature prior to use.
120
Prior to packing the composting reactors, the moisture content of the green waste
was determined by drying 5g samples in triplicate (as described for the soil) and
the dry weight was determined. The amounts of green waste and soil needed for
the different ratios were determined on a dry weight basis. The average moisture
content of the green waste was 50.2% ± 0.3. The pH of the green waste was 5.68
± 0.1.
5.2.3. Bench-Scale Composting System
The bench-scale composting system was the same as that used by Antizar-
Ladislao et al. (2005b). It consisted of air reservoirs (in and out) attached to
compost reactors, an air pump and incubators (Figures 5.1 and 5.2). The air
reservoirs were made from a section of plastic pipe, sealed at the top and bottom.
Holes were drilled into the top, into which small PVC tubes were inserted. The
inlet reservoirs had a steel tube inserted, reaching to the bottom of the reservoir.
The reservoirs were filled through this tube with distilled water. The compost
reactors stood vertically in incubators, with air flowing continuously up through
the compost mixture. Airflow to the compost reactors was provided by 100% oil-
free diaphragm pumps (Model PXW- 600-DIOV, VP1, 5 1 min"'; Fisher
Scientific). Inlet air passed through a reservoir filled with distilled water to
moisten it as it entered the compost, which prevented excess drying of the
composting mix. Exhaust gas from the compost reactors passed through an
empty air reservoir and the gases were vented outdoors. Tubes not attached to
any reactors were blocked off with ties, to prevent air being lost through least
121
2 1101A1ir outlet (vented outside)
Air inlet
resistance. The reactors were placed in incubators to maintain a constant
temperature profile for the treatments.
The compost reactors themselves were 200m1 clear glass jars (Fisher Scientific),
which provided easily monitored and controlled conditions. The lids of the jars
were drilled with two holes, one in the centre and one nearer the edge. Through
the centre hole a stainless steel tube, that reached the bottom of the jar, was
inserted. In the second hole a small PVC tube was inserted. When the reactors
were attached to the aeration system, the inlet air tube was attached to the steel
tube and the exhaust gas outlet tube was attached to the PVC tube. Each reactor
had a unique label, showing the time interval to be sampled, the replicate
number, and the treatment.
1. Diaphragm pump 4. Compost reactor 2. Incubator 5. Outlet air reservoir 3. Inlet water-filled air reservoir
Figure 5.1. Diagram of bench-scale composting bioremediation optimization
122
Figure 5.2. Photo of compost reactors showing the attachment of the reactors to
the inlet and outlet tubing and air reservoir
5.2.4. Soil to Green Waste Ratio and Temperature Optimization
To determine the optimal conditions soil to green waste ratio and temperature,
three soil to green waste ratios (0.7: 1, 0.8:1 and 0.9:1, based on dry weight) and
three temperatures (38, 55 and 70°C) were used in combination. For the control
conditions, pure soil was used. Treatments were carried out in triplicate and were
composted for 98 days. Destructive sampling in triplicate occurred at 0, 7, 21, 45,
56, 80 and 98 days. Two hundred and fifty two reactors were set-up, each
containing about 60g of the composting mixture (electrical insulating oil
contaminated soil and green waste mixed to give the investigated soil: green
waste ratios).
The weights of soil and green waste used for each of the ratios were determined
on a dry weight basis. The soil and green waste mix were prepared by weighing
the green waste in a 500m1 beaker and adding the soil. The two components were
123
then thoroughly mixed using a spatula, so that the soil coated the green waste.
The moisture content was adjusted accordingly using distilled water. The mixture
was packed into the reactors and the total weight of the reactors was noted. The
packed reactors were then placed in the incubators and attached to air reservoirs.
The moisture content of the compost mix was maintained at 60% and was
measured periodically and maintained gravimetrically by comparing the weight
with the start weight of the reactors, and adding distilled water.
At each sampling interval, destructive sampling was carried out on the
appropriate compost reactors. The compost samples were extracted using ASETM
and the extracts were analyzed using GC-MS analysis (as described in Chapter
3). Additionally the moisture and ash contents of the samples were determined
(as described in Section 5.2.1). At each sampling date, the contents of the
selected reactors were emptied into 500m1 beakers and homogenized. Sub-
samples were collected in triplicate for moisture and ash content analyses and
TPH determination.
5.2.5. Moisture Content Optimization
After the optimal temperature and soil: green waste ratio was determined, the
optimal moisture content was established in combination with the pre-determined
optimum temperature and soil: green waste ratio. Three moisture contents were
used (40%, 60% and 80%). Pure soil was used as controls and the treatments
were carried out in triplicate. Treatments were composted for 98 days with
124
destructive sampling taking place in triplicate at 0, 7, 21, 45, 54, 80 and 98 days.
The reactors were set-up using the same method as the soil to green waste and
temperature optimization experiment. The moisture content of the mix was
adjusted to the appropriate level using distilled water. Throughout composting
the moisture content was adjusted regularly to maintain the experimental
moisture contents. Destructive sampling was carried out at the same time
intervals and using the same methods as described in the soil to green waste and
temperature optimization methodology (Section 5.2.4).
5.3. Optimization Results
The results of each of the optimization experiments are displayed in separate
sections, showing losses of TPH in the various treatment combinations. The TPH
contamination concentrations are expressed as mg TPH per kg of ash, in order to
provide a constant basis from which losses can be compared, because during the
composting process there is a reduction in the mass of the compost.
5.3.1. Soil to Green Waste and Temperature Optimization Results
The changes in TPH concentration over the composting period for the soil to
green waste ratios tested are illustrated in Figures 5.3a-d. The 0.7:1 soil to green
waste ratio (Figure 5.3a) had the lowest loss of TPH during the composting
treatment, with an average loss of 52.8% for all temperatures used. The highest
percentage of TPH loss was from the treatments using 0.8:1 soil to green waste
ratio (average loss 75.6%), with the 0.9:1 ratio having an average loss of 60.3%.
125
There was a significant difference in TPH loss between the 0.7:1 and 0.8:1 ratios
(p<0.05), and the 0.8:1 and 0.9:1 ratios (p<0.05). Therefore, the greatest TPH
loss occurs in treatments with higher soil to green waste ratios.
In terms of the effect of temperature (Figures 5.3a-d and Table 5.1), in all the
different soil to green waste ratios those using 38°C had the greatest losses (63.5
to 86.1%), followed by the 55°C treatments (50.5 to 76.4%), with the 70°C
treatments having the lowest removal (44.6 to 64.4%). There was a significant
difference between 38 and 70°C temperature based treatments (p<0.05).
Therefore, the mesophilic temperatures result in higher TPH removal than the
(thermophilic) 70°C treatment. This indicates that lower temperatures stimulate
more microbiological activity and therefore degradation than do the higher
temperatures, which might impede microbiological activity.
126
III 1: Ili Hi III Ii III 111 111'11 11111111 I 1 111 11
.1 111.11
1H1111117-----: -. •-=_E=
250000
,u) 200000
• 150000
100000
a. 50000 I-
1111 Initial Ei Final
250000 -
• 200000
• 150000 CD
CI 100000
a_ 50000 I-
0
Initial E3 Final
(a)
250000 - :2
200000 - 6
150000 - a)
E co 100000 -
a. 50000 -
I • Initial O Final
38 55
70
Temperature (°C)
(b)
38
55
70
Temperature (°C)
(c)
38
55
70
Temperature (°C)
Figure 5.3. Initial and final TPH concentrations at each experimental
temperature for soil to green waste ratios of; a) 0.7:1, b) 0.8:1, and c) 0.9:1
127
Losses of TPH (23.6 to 30.0%) also occurred in the soil controls (Figure 5.4. and
Table 5.1) in all of the temperatures used. This indicates that a proportion of the
losses from the composting treatments are attributable to abiotic and biotic losses
that would occur in soil left untreated. Additionally, losses were greater for the
lower temperature treatments that the 70°C treatments.
250000
2 200000 -d) co
"6 150000 - sc
E 1:s) 100000 -
x i— a' 50000 -
0 Initial
0 Final
o
38
55
70
Temperature (°C)
Figure 5.4. Initial and final TPH concentrations in soil control treatments at each
experimental temperature
The temporal changes in concentration over the composting period for the
treatments varied (Figures 5.5a-d). The change in concentration for the 0.7:1 soil
to green waste ratio was a generally slow or steady decrease (Figure 5.5a). A
faster decrease in concentration was found in the 0.8:1 ratio treatments (Figure
5.5b), with an increase in the losses towards the end of the composting period.
The treatments with a ratio of 0.9:1 (Figure 5.5c) also had a fast decrease in
concentration, but not as fast as with the 0.8:1 ratio. The changes in concentration
for the soil controls were very slow with little loss of TPH, indicating that losses
128
(a) (b)
0 0
250000 -
200000 - a co
5 150000 - a) .te E 100000 -
x 1— 50000 -
—.-38°C - 55°C ---70°C
250000
200000 -
150000 -
100000
50000
(d)
--*-38°C 55°C
—A-70°C
(c) 250000
200000
150000
100000
50000
0
250000 -
2- 200000 - a a
:13 150000 -1:a .Ne E 100000 -
50000 -
0
—*-38°C - o 55°C
due to microbial action were likely to be more significant than abiotic losses. This
shows that the soil to green waste ratio has an effect on how the TPH
contamination is removed during composting. The lowest ratio appears to have a
slowing effect on the removal of contamination especially at the start of
composting. The 0.8:1 ratio increases the rate at which TPH is lost compared to
the lower ratio. The highest ratio (0.9:1) also seems to retard the removal of TPH,
more than the lower ratios at the start of composting, although the removal
increases towards the end.
0 7 21 45 54 80 98 0 7 21 45 54 80 98
Time (days) Time (days)
0 7 21 45 54 80 98
0 7 21 45 54 80 98
Time (days)
Time (days)
Figure 5.5. Change in average TPH concentration during composting for soil to
green waste ratios of; a) 0.7:1, b) 0.8:1, c) 0.9:1, and d) soil controls
129
First order kinetic analyses were performed on the data from the compost
treatments and controls. The pseudo-first-order kinetic approximation was
applied using the linear integrated form of equation (5.1):
In (C/Co) = -kt (5.1)
where C is the concentration at time t, Co is the concentration at t = to, k is
the first-order constant of removal (obtained by linear regression) and t is time
(Admon et al. 2001; Antizar-Ladislao et al. 2005).
The kinetics of TPH loss for the soil to green waste and temperature treatments
are shown in Figures 5.6a-d and Table 5.1. For the 0.7:1 soil to green waste
ratios a good fit was found for 55°C and 70°C (R2 = 0.95 and 0.89 respectively).
However, the fit for 38°C was poor (R2 = 0.86), indicating TPH removal was not
a first order reaction, although it had the largest rate constant (k = 0.009 day').
The 0.8:1 and 0.9:1 soil to green waste ratios at each temperature were all first
order loss reactions (0.8:1 R2 values; 38°C = 0.94, 55°C = 0.96 and 70°C = 0.97,
and 0.9:1 R2 values; 38°C = 0.98, 55°C = 0.99 and 70°C = 0.89). For the 0.8:1
soil to green waste ratio the fastest removal was 38°C (k = 0.020 day'). In
contrast for the 0.9:1 ratio the fastest removal was at 55°C (k = 0.012 day'),
although there was little difference to that at 38°C (k = 0.011 day'). The soil
controls were also all first order degradation reactions, although the rate
constants were lower (R2 values; 38°C = 0.91, 55°C = 0.98 and 70°C = 0.92).
There was little variability in the values of the kinetic losses for the various
compost treatments, which is generally difficult to achieve for materials such as
compost and highly contaminated soils.
130
(c) -0.2
(d)
• 38°C a 55°C A 70°C
- - - • 70
-1.2
s -0.4
-0.8
0.1
-0.1
-0.3
i."3 -0.5
-0.7
-0.9
0.25
-0.25
-0.75 6-̀) .5- -1.25
-1.75
-2.25 0 10 20 30 40 50 60 70 80 90 100
0 10 20 30 40 50 60 70 80 90 100
Time (days)
Time (days)
0 10 20 30 40 50 60 70 80 90 100
Time (days) 0 10 20 30 40 50 60 70 80 90 100
Time (days)
Figure 5.6. Kinetic losses of TPH at three test temperatures during composting
with soil to green waste ratios of; a) 0.7:1, b) 0.8:1, c) 0.9:1, and d) soil controls
Table 5.1. Average percentage loss and kinetic rate loss constant of TPH from
soil to green waste and temperature optimization
Soil to green waste ratio
Control 0.7:1 0.8:1 0.9:1
Temperature (°C) 38 30.0 63.5 86.1 65.3
Rate constant (day-I) 0.003 0.009 0.020 0.011
55 24.8 50.5 76.4 66.5
Rate constant (day-1) 0.003 0.007 0.015 0.012
70 23.6 44.6 64.4 49.2
Rate constant (day-I) 0.003 0.006 0.010 0.008
131
250000
gi) 200000
150000 '7a)
100000
a. 50000
Ea Initial
0 Final
5.3.2. Moisture Content Optimization Results
The dependence of TPH loss on moisture content was measured during
composting using the optimal soil to green waste and temperature conditions
found in the first experiment (0.8:1 soil to green waste ratio, 38°C). The results
are given in Figure 5.7. The greatest overall loss of TPH (75.1%) was in the 60%
moisture content treatments. The 40 and 80% moisture content treatments had
similar losses of 65.6 and 61.4% respectively. There were significant differences
in TPH loss between all the different moisture content treatments (p<0.01). There
were significant differences between 40 and 60% moisture content treatments
and 60 and 80% moisture content treatments (p<0.05). Loss average loss of TPH
in the soil controls was 26.2%.
Control
40 60
80
Moisture Content (%)
Figure 5.7. Average initial and final TPH concentrations in moisture content
treatments (0.8:1 soil to green waste ratio, 38°C) and soil control
Temporal changes in the average TPH concentration over the composting period
are shown in Figure 5.8. The pattern of change in concentrations is similar in the
132
40 and 60% moisture content treatments. There is a generally slow reduction in
TPH concentration until day 45, after which the rate of loss increases. The 80%
moisture content treatment had a more steady change in concentration over time.
The soil controls had a very small change in concentration, throughout
composting, with a larger change in concentration towards the end of the
composting period. This suggests that moisture content has an effect on loss. A
moisture content of 40% gave the slowest start in TPH reduction, suggesting it
was too low and affected degradation. The 60 and 80% moisture contents gave
similar patterns, suggesting that these are better for degradation of TPH, although
removal from the 80% treatments was slower, suggesting some degree of
inhibition.
250000 -
2 200000 -y
_o 150000 - CD
ac
E a) 100000 -
ct• 50000 -
o -40
80 --x-- control
0
0 7 21 45 54 80 98
Time (days)
Figure 5.8. Change in average TPH concentration during composting
The kinetics of TPH loss are shown in Figure 5.9. The first-order kinetics were
carried out using the same method as described in Section 5.3.1. All the moisture
content compost treatments had good fits, exhibiting first order degradation
133
0.1
-0.1
-0.3
-0.5
-0.7
40% k = 0.011 dayl 60% k= 0.015 day' 80% k= 0.010 day' Control k= 0.002dayl
kinetics (R2 values: 40% = 0.89, 60% = 0.94 and 80% = 0.97). The fastest
removal was for 60% moisture content treatments (k = 0.015 day-1) with 40 and
80% having similar kinetic constants (k = 0.011 day4 and k = 0.010 day-1
respectively). However, the soil controls had a poor fit (R2 = 0.78), indicating
that the losses were not first order.
c 40% n 60%
80% x control
— — - 60% 40%
- - . 80% — - - — - control
0 10 20 30 40 50 60 70 80 90 100 Time (days)
Figure 5.9. Kinetics of loss from moisture content treatments and soil control
From these experiments, the optimal composting conditions for the removal of
TPH are concluded to be a combination of a soil to green waste ratio of 0.8:1, a
constant temperature of 38°C and a moisture content maintained at 60%.
5.4. Discussion
In all of the compost treatments, the addition of amendments resulted in an
increased reduction of TPH compared to soil controls. The addition of
134
amendments such as green waste helps to sustain populations of degrading
microorganisms, and can also improve the contaminated soil environment for
indigenous microorganisms (McFarland and Qui 1995; Antizar-Ladislao et al.
2006; Ostberg et al. 2007). The observed stimulatory effect may be due to
increased microbial biomass. Additionally, there may be increased bioavailability
of the substrate or changes in the microbial composition towards organisms able
to degrade hydrocarbons (Ostberg et al. 2007).
The greatest removal of TPH during composting was found in treatments using a
soil to green waste ratio of 0.8:1 (at all temperatures). Losses were generally
greater in treatments with the higher soil to green waste ratios (0.8:1 and 0.9:1),
than the lowest ratio used (0.7:1). These findings were in agreement with
Antizar-Ladislao et al. (2005b) for the removal of PAHs during composting.
Namkoong et al. (2002) found that amendment increased the biodegradation
rates, but higher mix ratios did not increase the degradation rate of TPH. The
potential advantages of amendments may be offset by the possible inhibition of
the degradation process (Lee et al. 2008). The lower TPH loss in the 0.7:1 ratio
treatments may have been due to preferential degradation of the green waste
rather than the target contaminants, as Van Gestel et al. (2003) found with the
composting of biowaste and the mineralization of diesel hydrocarbons. This
would be due to inhibition of contaminant mineralization (Namkoong et al. 2002;
Ostberg et al. 2007; Lee et al. 2008). The greater amount of green waste in this
ratio treatment may have provided an alternative carbon source that is more
easily degraded (Lame and Jorgensen 1997; Lee et al. 2008). Kriipsalu et al.
(2007) also found that organic material interfered with degradation.
135
The two lower temperatures used (38 and 55°C) had the greatest removal of TPH
from all the soil to green waste ratio treatments used. Antizar-Ladislao et al.
(2005b) also found this pattern in the composting of PAHs. The results of these
optimization experiments demonstrate that there is likely viability for the use of
composting at a commercial-scale to treat soils contaminated with TPH.
The low losses at the treatments using 70°C could have been due to the fact that
temperatures in excess of 60°C have been shown to reduce the activity of the
microbial community (inhibiting microbial diversity and the enzymatic potential
of the system) (Liang et al. 2003; Antizar-Ladislao et al. 2004; Cai et al. 2007).
The reduced metabolic activity at this higher temperature would lead to less TPH
biodegradation. Therefore, the losses observed at 70°C may have been due
mainly to volatilization with some microbial breakdown (Lazzari et al. 1999;
Antizar-Ladislao et al. 2005b). This would indicate that the leading mechanism
of removal at 38°C was biological, whereas at 70°C it was volatilization, and
most likely a combination of these two mechanisms at 55°C (Antizar-Ladislao et
al. 2005b).
The higher losses at the lower temperatures indicate that these temperatures
enhance biological degradation of TPH though stimulation of microorganisms
present and their metabolic activity (Antizar-Ladislao et al. 2006). In this study,
it appears that a temperature of 38°C enhances the biological removal of PAHs,
which might occur due to a promotion of the native microbial population and
activity (Antizar-Ladislao et al. 2006). A constant temperature of 38°C enhances
mesophilic microbial populations (Antizar-Ladislao et al. 2005b; 2006). Other
136
studies indicate that lower temperatures might allow more microbial activity
(Liang et al. 2003), although temperatures below 20°C slow or stop the
composting process (Laine and Jorgensen 1997). The optimum temperature for
the degradation of PAHs by composting has been reported to be between 40 and
45°C, which represents the overlap point of the optimal conditions for
mesophilic and thermophilic microorganisms (Moretto et al. 2005). This allows
for maximum diversity to be expressed for bacteria, actinomycetes, and fungi
(Antizar-Ladislao et al. 2006). Mesophilic populations could play an important
role in PAH degradation (Amir et al. 2005).
The greatest removal of TPH occurred in the treatments using 60% moisture
content, with lower losses from 40 and 80% moisture content treatments.
Moisture content is an important environmental variable as it provides a medium
for the transport of dissolved nutrients required for the metabolic and
physiological activities of microorganisms (Liang et al. 2003) and affecting
microbial activity and PAHs bioavailability (Cai et al. 2007). Low moisture
content values could cause early dehydration during composting, slowing or
stopping biological process by decreasing microbial populations (Atagana 2004).
The lower losses in the 40% moisture content treatments may have been due to
low moisture content causing dehydration within the compost matrix, which
slows or stop biological processes by decreasing microbial populations (Atagana
2004). In previous studies, the minimum required moisture content for increasing
microbial activity was 50%, whereas maximum activity was found in the range
of 60-70% (Tiquia et al. 1996; Liang et al. 2003). This agrees with the maximum
TPH losses at 60% moisture content found in the experiment described here,
137
where degradation by microorganisms was at its greatest. Low losses at 80%
moisture content were likely to be due to some degree of water logging due to
excess water, which resulted in anaerobic conditions and reduced microbial
activity (Laing et al. 2003).
Although the reduction in TPH was generally greater than that of the Crown
BioSystems pilot-scale study (50.5 to 86.1% compared to 14.3 to 61.5%), there
were still large amounts of residual contamination. The amount of residual
contamination in the compost was estimated from the ashed weight of samples,
the amounts remaining after the composting period ranged from (16.2 to 59.0g
kg' of compost for soil to green waste and temperature optimization and 26.0 to
37.8g kg"' of compost for moisture content optimization). The high residual
contamination may have been due to the highly contaminated nature of the soil to
start with, as well as sequestration and persistent residues (discussed later in this
Section). Such high starting concentrations mean that, inevitably, not all the TPH
contamination could be removed during the process. Therefore, this aspect of the
composting bioremediation process needs to be examined in order to overcome
this problem and to achieve lower and acceptable end-points.
Although there is no set limit for TPH as a measure, from the results of this
study, an acceptable level of residual TPH has not been achieved, as there were
still high residual levels of contamination in the samples. Therefore, although
this process has been optimized, and has improved on the results gained from the
pilot-scale study, further work is needed on the process to reach these end levels.
For example, a longer composting period may be needed.
138
The changes in TPH concentration during the various composting conditions
tested were mainly biphasic. However, the pattern differed to that typically
reported, in which there is generally an initial rapid decrease, followed by a
slower phase of loss (Potter et al. 1999; Namkoong et al. 2002; Sarkar et al.
2005; Antizar-Ladislao et al. 2005b). The loss of TPH during most of the
treatments showed a slow loss at first, which increased subsequently. The slow
onset of degradation may have been due to the microorganisms not using the
TPH compounds as a carbon source at the start of composting (Sarkar et al.
2005), but using components in the green waste instead (Lee et al. 2008).
Moretto et al. (2005) found that degradation of PAHs only became significant
after 30 days, which could have been due to re-starting of biological activity after
initial inhibition by the addition of the contaminated material. There may also
have been a delay due to the microorganisms adapting to the hydrocarbon
contamination (Atagana 2004; Scherr et al. 2007). Over time, indigenous
microbial species become as effective in degrading contaminants in soil compost
systems as inoculated systems (McFarland and Qui 1995). Additionally, there
may have been some inhibition of microorganisms due to the contamination
(Hughes et al. 2007). Another reason for the lag in TPH degradation may have
been that at the start of composting TPH compounds were initially adsorbed to
sites where they were temporarily less bioavailable (Amir et al. 2005). In
addition, the difference may have been due to a change in nutrient availability
and a shift in the microbial community degraders over time affecting removal
(Hartlieb et al. 2003). Additionally, the observed changes in concentration may
have been part of the initial fast phase of TPH loss and a slow TPH loss phase
had not been reached during the composting period. The observed pattern of loss
139
in this study may have been due to an insufficiently short timescale, so only the
start of the loss pattern may have been observed. Therefore, with a longer
composting period a typical biphasic pattern may be observed.
The rate constants for TPH loss (Table 5.1 and Figure 5.9) were lower than those
found in other TPH bioremediation studies (Table 5.2), although the R2 values of
the present study were generally higher. This indicates that the rate of TPH loss
during composting bioremediation could be greater. However, as the remediation
technique, experimental conditions and contamination are different to this study,
this affects the rate loss constant.
Table 5.2. Rate constants for removal of TPH from other bioremediation studies
Rate constant (day 1)R2 Reference
1.25 x 10-5 0.89 (Sarkar et al. 2005)
0.021 0.91 (Admon et al. 2001)
0.026 0.83 (Paudyn et al. 2007)
During the composting process the percentage ash content of the compost
samples decreased by 3-6%. This demonstrates that the total organic content of
the samples was decreasing indicating mineralization occurred (Antizar-Ladislao
et al. 2005b).
Losses of TPH from the composting treatments will have resulted from abiotic
and biotic processes (Civilini 1994; McFarland and Qui 1995). Abiotic losses
140
such as volatilization are important for lower molecular weight compounds,
whereas higher molecular weight compounds are biologically degraded (Park et
al. 1990; Lee et al. 2001; Sarkar et al. 2005). For example, biotic mechanisms are
responsible for removal of PAHs containing more than 3-rings (Hansen et al.
2004; Moretto et al. 2005). However, as electrical insulating oils are non-volatile
(Nynas Naphthenics 2004), volatilization losses would not have been a major
contributor to loss. Losses of TPH from the treatments may have been due to
degradation of the compounds by microorganisms present in the soil and green
waste, such as strains of white rot fungi (Baheri and Meysami 2002) and
indigenous microflora (Cai et al. 2007). Losses of TPH were observed in all the
control soils, and such losses from untreated soils would also contribute to losses
of TPH from compost treatments (23.6 to 30.0% in soil to green waste
optimization and temperature optimization and 26.2% in moisture content
optimization). Losses from the soil controls could have been abiotic and biotic.
Sequestration of contaminants may also have made TPH compounds less
available. Organic matter content, micropore volume and surface area affect
sequestration (Bogan and Sullivan 2003). High molecular weight compounds
adsorb to humic matter (Chung and Alexander 2002; Amir et al. 2005), resulting
in the large amounts of residual TPH contamination.
5.5. Conclusions
A greater removal of TPH than the pilot-scale study has been achieved through
optimization of the composting process. Additionally, a set of operating
141
conditions for in-vessel composting with high removal of TPH has been
developed that can be applied to treatments in the future. Different composting
parameters result in varying patterns of TPH loss and overall loss. This indicates
that these different parameters have an effect on the compost matrix micro-
environments and also the behaviour of the TPH contamination.
The losses from the composting treatments were as high as 86.1% and indicate
the viability of using in-vessel composting bioremediation of electrical insulating
oil-contaminated soil. However, high residual contamination remains in the
composted samples, meaning that the end product cannot be re-used and needs to
undergo further treatment. Therefore, more experiments need to be conducted in
order to improve the process before it could be scaled-up. Further optimizing
treatments may include the addition of fertilizers, using longer composting
periods, or further treatments such as land farming (Antizar-Ladislao et al.
2005b; Hansen et al. 2004) to achieve lower contamination end-points.
Another aspect that needs to be examined in more detail is how the starting
contamination concentrations affect the removal of contamination during
composting. From this experiment, it can be seen that highly contaminated soils
will leave residual contamination. Therefore, it is possible that diluting the
contaminated soil before composting may help in achieving low and acceptable
end concentrations. However, this does increase the amount of contaminated
material involved in the process, which is already a concern about composting
bioremediation and the addition of uncontaminated green waste to contaminated
soils (Cole 1998).
142
Chapter 6. Effect of Soil Properties on the Removal of TPH
Contamination During In-Vessel Composting
6.1. Introduction
Soils play an important role in the environmental fate of organic contaminants.
An organic contaminant in soil may undergo several processes: it may be lost (by
biodegradation, leaching, volatilization) or it may chemically bind with mineral
surfaces or organic matter (Northcott and Jones 2000a; Maliszewska-Kordybach
2005; Doick et al. 2005b).
The fate and behaviour of organic contaminants in soil is dependent on the
physico-chemical characteristics of the compound as well as the physical,
chemical, and biological properties of the soil (Fine et al. 1997; Semple et al.
2003; Maliszewska-Kordybach 2005; Aichberger et al. 2006; Jonsson et al.
2007). The physico-chemical properties of the contaminants that are important
include hydrophobicity and molecular structure, and these are considered in
Chapter 8. Here, important soil properties including soil type, organic matter
content, moisture content and structure are considered.
Soil organic matter determines the sorption and bioavailability of organic
contaminants (Aichberger et al. 2006; Hofman et al. 2008). The degradability of
organic contaminants decreases as soil organic matter content increases (Jonsson
et al. 2007). It has been identified as a preferential sorption site for contaminants,
when organic carbon content is present above 0.01-0.2% (Northcott and Jones
143
2000a; Aichberger et al. 2006; Ehlers and Loibner 2006). The properties, as well
as the amount of organic matter, affect bioavailability (Hofrnan et al. 2008).
Even in soils with low organic content, entrapment of organic contaminants in
pores and voids also reduces bioavailability (Maliszewska-Kordybach 2005;
Jonsson et al. 2007). Diffusion into micropores with diameters less than 0.1µm
results in low availability (Aichberger et al. 2006). Organic contaminants can be
more concentrated into fine soil fractions (clay and fine silt) than in the larger
particle size fractions (sand) (Amellal et al. 2001). Therefore, biodegradation
rates are usually higher in sandy soils than in clay rich soils (Rizzo et al. 2008).
Soil moisture content also affects the sorption of organic compounds in soil. In
wet soils, initial sorption can be dramatically reduced compared to initially dry
soils, due to a hindrance of organic compounds partitioning into organic matter
or diffusing into soil nanopores and also possibly by competition between
introduced contaminants and water molecules for specific sorption sites (Kottler
et al. 2001).
Contrasting biodegradation trends in dissimilar soils are due to dissimilar
microenvironments affecting the abundance and diversity of the microbial
community contaminant as well as contaminant bioavailability (Davis and
Madsen 1996; Bundy et al. 2002; Scherr et al. 2007). Microbial activity may be
affected by different soil characteristics such as soil texture and structure that
affect accessibility and bioavailability of the contaminants (Amellal et al. 2001).
Soil pH can also affect degradation-competent microorganisms, as well as the
144
physical properties of soil organic matter influencing its sorption capability
(Maliszewska-Kordybach 2005).
Different soil characteristics have varying affects on the behaviour and
availability of contamination. An understanding is needed of the potential affects
that differing soil characteristics might have on the remediation of TPH during
in-vessel composting. This chapter compares the removal of electrical insulating
oil-TPH contamination during composting from two contrasting soils, in order to
determine the affects that the different soil properties may have on the removal
of TPH.
6.1.1. Hypothesis
Different soils have distinct characteristics that affect the rates and overall
removal of TPH contamination, due to differing interactions with TPH
contamination. Variations in characteristics such as texture, organic matter and
moisture content will result in higher or lower TPH removal, depending on the
characteristics of the soil.
6.1.2. Aims and Objectives
The aim of this experiment was to examine the effect of the differing
characteristics in two soils on the removal of TPH contamination during in-
vessel composting bioremediation.
145
To achieve this, several objectives were set;
1) Contaminate two contrasting soil types with transformer oil to 10,000ppm to
simulate a naturally-occurring heavy leak.
2) Compost the contaminated soils under optimized composting conditions.
3) Compare the removal of TPH contamination in the different soils during
composting.
6.2. Materials and Methods
To examine the effects of different soil types, soils were artificially
contaminated; the methods used for collection, spiking and ageing of soils drew
on experience and recommendations from several methods reported by Northcott
and Jones (2000a), Lee et al. (2003) and Doick et al. (2003).
Two uncontaminated, contrasting soils were collected from the Imperial College
at Wye Estate; they were a heavy-texture gault clay and a lighter-texture brick
earth. The gault clay was from the Evesham soil series (calcareous pelosol) and
the brick earth was from the Hamble soil series (argillic brown earth) (Burnham
1994). The characteristics of the soils were previously determined (Gove 2001)
and are shown in Table 6.1. Approximately 3kg of each soil was collected from a
depth of 5-15cm.
146
Table 6.1. Soil properties of gault clay and brick earth
Gault clay Brick earth
Soil series Evesham Hamble
Texture Clay Sandy loam
Fine sand (%) 19 61
Coarse sand (%) 4 9
Silt (%) 26 23
Clay (%) 46 13
Organic matter (%)
Moisture content (%)
pH
3.8 ± 0.1
34.6 ± 0.15
7.7
4.8 ± 0.1
19.2 ± 0.11
7.6
Prior to spiking, the soils were forced-rubbed through a 9.5mm sieve to remove
stones and vegetation material, a large gauge sieve was used to minimize
disruption to the soil structure. The soils were spiked with transformer oil,
provided by EDF Energy, to 10,000ppm on a wet weight basis, to simulate a
heavy electrical insulating oil leak. For spiking, a 2kg sample of wet soil was
weighed into an acetone-cleaned glass bowl, transformer oil (20g) was added to
the soil and the mixture blended for 5 minutes by hand using a metal spoon.
Between the spiking of each soil sample, the mixing bowl and spoon were
washed and rinsed with acetone to remove any oil contamination. To check the
homogeneity of the spikes, 5g sub-samples of the spiked soils were taken at
random in triplicate, extracted and analyzed (as described in Chapter 3). The
percentage variation in contamination in the samples was ± 0.77 and ± 1.19% for
gault clay and brick earth respectively, showing that the spiking and
homogenization resulted in an even distribution of the contamination.
147
The contaminated soils were composted under optimized conditions (Chapter 5:
soil to green waste ratio of 0.8:1, 38°C and 60% moisture content). Controls
consisted of the spiked soils. The samples were composted with green waste for
98 days and destructive sampling was carried out at 0, 7, 21, 45, 54, 80 and 98
days in triplicate (as described in Chapter 5).
6.3. Results
The residual TPH concentrations at the end of composting in both types of soil
were comparable (Figure 6.1) with losses from gault clay and brick earth of 80.5
and 77.6% respectively, which were not significantly different (p>0.05). A
similar pattern of residual contamination was also observed in the soil controls
(Figure 6.2).
5000 IS 7
-se0) 4000 a) E • 3000 p 2
2 2000 o 0 a o • 1000 x 0-I—
Initial
0 Gault clay 0 Brick earth
Final
Time (days)
Figure 6.1. Initial and final TPH concentrations in gault clay and brick earth
compost samples
148
12000 2 cn 03 "a 10000 a• ) ..c • 8000 E
o = 6000 ..r.
62 *e. 4000 o o c o • 2000 x o. I-
Initial
o Gault clay control 0 Brick earth control
Final
Figure 6.2. Initial and final TPH concentrations in gault clay and brick earth soil
controls
Although the overall changes in TPH concentration during the composting
process were comparable for both types of soil (Figure 6.3), the gault clay had
greater losses of TPH until 45 days compared to the brick earth, which had
slower and steadier losses. After 45 days, the pattern of change in TPH
concentration for both types of soil was comparable. This suggests that the
different soil characteristics of the different soils, or differences in their
biological activity, might have had a small effect on the loss of TPH at the start
of composting, but these effects become less significant towards the end of the
composting period and appeared to be of little importance throughout the
composting period as a whole. The changes in TPH concentration in the soil
controls (Figure 6.4) were comparable until 21 days, after which the gault clay
has greater TPH losses. Despite this, no significant differences were observed
over the whole composting period.
149
w. 12000 - 0
10000 a) E 8000 C o 473 coo 6000
4000 C uo
2000 - o.
0
—*--Gault clay control
Brick earth control
4.1.• 5000 - 0
co 4000 a)
E 3000 - o co nscn
2000 0 c.) C c.) 1000 -
0
Gault clay ---1:1— Brick earth
0 7 21 45 54 80 98
Time (days)
Figure 6.3. Change in TPH concentration in gault clay and brick earth during
composting
0 7 21 45 54 80 98
Time (days)
Figure 6.4. Change in TPH concentration in gault clay and brick earth soil
controls
First order kinetic analyses were performed for the composted gault clay and
brick earth samples as well as for the controls. The pseudo-first-order kinetic
approximation was applied using the linear integrated form of equation (6.1).
150
In (C/Co) = -kt (6.1)
where C is the concentration at time t, Co is the concentration at t = to, k is
the first-order constant of removal (obtained by linear regression) and t is time
(Admon et al. 2001; Antizar-Ladislao et al. 2005a).
The losses of TPH from both types of soil during composting and from the
controls were first order reactions. The R2 values for the losses were all high and
linear (composted gault clay R2 = 0.96, composted brick earth R2 = 0.97, gault
clay control R2 = 0.97 and brick earth control R2 = 0.94), which shows that there
was little variation between the samples. The kinetic loss constants of TPH
during composting (Figure 6.5) were the same for both soils (k = 0.016 day'),
which suggests that although differences in the loss of TPH were observed
between the two soils during composting they were not significant enough to
affect the kinetic loss constant of TPH. The kinetic loss constants in the gault
clay controls (Figure 6.6) were greater than in the brick earth (k = 0.0063 and
0.0059, respectively), although this was not significant (p>0.05), again
suggesting that different soil characteristics do not have a significant effect on
the loss of TPH for the two soils considered here.
151
0 -0.2 -0.4
-0.6
O -0.8 -1
-1.2 -1.4
-1.6 -1.8
0.1
0
-0.1
-0.2
o -0.3
E -0.4
-0.5
-0.6
-0.7
-0.8
• Gault clay control o Brick earth control
Gault clay control — — — - Brick earth control
20 40 60 80 100
• Gault clay o Brick earth
Gault clay — — — - Brick earth
-2 20 40 60 80 100
Time (days)
Figure 6.5. Kinetic losses of TPH gault clay and brick earth during composting
Time (days)
Figure 6.6. Kinetic losses of TPH from gault clay and brick earth soil controls
6.4. Discussion
The losses in TPH from the different soils were not significantly different during
composting, despite marked differences in the physical chemical characteristics
and types of the two soils investigated. Thus, it appears that the composting
152
process was sufficiently rigorous to overcome any effects resulting from
differences in soils and/or the nature of the electrical insulating oil itself, was the
limiting factor on TPH losses rather than soil factors.
Comparable losses were observed despite the differences in organic matter, 4.8%
in brick earth and 3.8% in gault clay, and this difference was not significant
enough to result in greater contaminant sorption in the brick earth due to the
larger amount of organic matter, which might have resulted in less bioavailable
contamination and lower biodegradation (Trindade et al. 2005; Scherr et al.
2007; Hofman et al. 2008). The differences in sand content (fine and coarse
sand) also did not have an affect on the loss of TPH; brick earth had a higher
content than the gault clay (70% and 23%, respectively), although higher sand
contents can result in poor microbial proliferation and diversity (Scherr et al.
2007), which might have resulted in lower removal of TPH in the brick earth.
Additionally, the higher moisture content in gault clay (34.6%) compared to the
brick earth (19.2%), did not result in water molecules out-competing the organics
for adsorption sites, so less contamination would have been sorbed in the gault
clay (Kottler et al. 2001) and therefore more TPH would have been available for
degradation. Finally, the differences in soil characteristics did not result in
significant differences between the soil conditions of the two soils. Therefore, the
microbial and structure and density in the soils would have been comparable,
resulting in similar metabolic capacities and different degradation patterns
(Bundy et al. 2002; Scherr et al. 2007). Despite these differences in soil
characteristics, there was no significant difference in TPH loss between the two
soils.
153
The nature of the oil may have been a limiting factor on the removal of TPH
during composting, rather than the soil characteristics. Because electrical
insulating oil is a non-aqueous phase liquid (NAPL), the contaminants may have
preferentially remained in this oil-phase rather than interacting with the soil
(Huesemann et al. 2002). Therefore, the contamination would have been
available for degradation irrespective of soil characteristics. Additionally, the
large amount of contamination added to the soils may have negated any soil
properties that would have had an effect on the loss of TPH, so losses in both
soils would be comparable. It has been noted that at high concentrations of
organic contaminants, saturation of sorption sites occurs (Chung and Alexander
1999). Therefore, as the adsorption sites become saturated due to the large
quantity of oil in the soils, more of the oil will be available for degradation in
both soils. Additionally, the physicochemical properties of the contaminants may
have had more effect than the soil properties, controlling the behaviour of
contaminants in the soils (Stokes et al. 2006).
The comparable losses may have also been due to the optimized composting
conditions being efficient enough to overcome any limiting effects of the soil
properties or the oil during composting. Modifications of the composting
environment by the operating temperature, moisture content and addition of
green waste, would have stimulated microbial diversity and activity, compound
bioavailability and reaction rates (Finstein and Hogan 1993; Jorgensen et al.
2000; Antizar-Ladislao et al. 2004). Therefore the TPH contamination would
have been potentially equally removed from each soil.
154
The losses of TPH during composting were all first order reactions and even
under the optimized composting conditions, not all of the TPH contamination
was removed. Therefore, composting for a longer period may result in greater
overall TPH losses and also in second order loss reactions, as loss of
contaminants from soils decreases over times, resulting in a second, slower stage
of loss (Admon et al. 2001; Semple et al. 2003). A slower stage after the initial
fast loss of the more available contamination might result in more apparent
differences in TPH losses from the two contrasting soils.
6.5. Conclusions
No significant difference in the loss of TPH between the two soils was observed,
indicating that soil properties do not significantly affect the loss of TPH during
composting bioremediation. This may have been due to the nature of the oil itself
as well as the amount of contamination added to the soils. Alternatively, the
effectiveness of the composting treatment may have meant that the differences in
the soil had no effect. Additionally, the length of composting may have been a
factor, and longer composting periods may show a more pronounced difference
in the loss as removal of TPH slows down. However, longer composting periods
from a commercial process point of view would be uneconomic given the slow
loss phase, as the optimum composting period would equate to the point between
the fast and slow removal stages, so there would be no benefit to further
composting. Additionally, although no significant difference was found in the
removal of TPH as a measurement, there may be differences in the loss patterns
of individual compounds or groups of compounds between the two soils, which
155
are explored in Chapter 8. The implications of this experiment are that soil
properties do not appear to have an affect on TPH loss from different soils.
Therefore, the composting bioremediation of different contaminated soils will
not be affected by the properties of the soils themselves.
156
Chapter 7. Effect of Contaminant Ageing on the Composting
Bioremediation of Electrical Insulating Oil-Contaminated Soils
7.1. Introduction
Organic compounds within soils undergo a time-dependent reduction in
desorption, biodegradation, toxicity and uptake by organisms (Kottler et al. 2001;
Nam and Alexander 2001; Bogan and Sullivan 2003; Doick et al. 2003). Over
time, compounds become sequestered, which is known as ageing (Sun et al.
2003; Ncibi et al. 2007). Ageing is associated with a continuous diffusion and
retention of compounds into remote and inaccessible regions within the soil
matrix, thereby limiting the availability of compounds to abiotic and biotic loss
processes (Northcott and Jones 2001; Palomo and Bhandari 2006). The length of
time needed for a compound to become sequestered depends on the chemical or
the mixture of chemicals, soil characteristics, environmental conditions and any
operational land management practices (Alexander 2000; Stokes et al. 2006).
Longer contact times lead to higher retention, reduced desorption and greater
contaminant sequestration (Palomo and Bhandari 2006).
Ageing is the result of chemical and physical processes. Interactions between
hydrophobic contaminants and soil particles include partitioning into or sorption
onto humic matter, surface adsorption/binding onto mineral surfaces or diffusion
into micropores of soil particles (Kottler et al. 2001; Doick et al. 2003;
Huesemann et al. 2003; Sun et al. 2003; Latawiec et al. 2008). Depending on
compound properties, ageing begins with partitioning of contaminant molecules
into (and sorption onto) organic matter at the surface of soil particles and
157
diffusion into macropores (<100nm in diameter) followed by diffusion into
micropores (diameters of 0.3-1.0nm) (Alexander 2000; Bogan and Sullivan
2003; Sun et al. 2003).
Several soil characteristics are important in the ageing of contaminants.
Increased soil organic matter contents are correlated with decreased
bioavailability (Kottler et al. 2001), because it increases sorption of organic
contaminants to the soil matrix, decreasing the rate and extent of biodegradation
(Scherr et al. 2007). Additionally, the quality and quantity of organic matter
affect ageing, due to heterogeneous structures and the presence of reactive
functional groups such as carboxylic acids, phenolic acids and amines (Hutchison
et al. 2001; Palomo and Bhandari 2006). The inorganic make-up of the soil (e.g.
clay content, mineral/trace element supply), its pore size and structure, as well as
soil microbial activity, wetting and drying cycles and contaminant concentration
all affect sequestration (Reid et al. 2000; Doick et al. 2003). Research also
suggests that microorganisms facilitate the sequestration of contaminants in order
to reduce the toxicity of a contaminant (Pravecek et al. 2006).
The ultimate result of ageing is the movement of compounds from accessible soil
compartments into less accessible or even inaccessible compartments (Reid et al.
2000; Alexander 2000), resulting in lower physical and biological availability
(Reid et al. 2000; Kottler et al. 2001), and making them more resistant to
biodegradation (Trindade et al. 2005; Ncibi et al. 2007). Decreased contaminant
mobility and toxicity reduce the risk to human health, or to other environmental
receptors (Bogan and Sullivan 2003; Sun et al. 2003). Limited contaminant
158
bioavailability is the principal reason for the slow or incomplete biodegradation
of hydrocarbons in aged soils (Huesemann et al. 2003). This makes it more
difficult to remove contamination using biologically-based processes such as
composting bioremediation. This chapter explores the effect that contaminant
ageing has on the removal of TPH contamination during composting
bioremediation by comparing the removal of TPH from freshly contaminated and
aged contaminated soils.
7.1.1. Hypothesis
Ageing of electrical insulating oil in soils results in more persistent
contamination than in freshly contaminated soils, resulting in slower contaminant
removal rates and lower overall removal during composting.
7.1.2. Aims and Objectives
The aim of this experiment was to determine the effect of contaminant ageing on
the removal of TPH contamination during composting bioremediation.
To achieve this, several objectives were set:
1) Contaminate soil samples with transformer oil to simulate a transformer leak.
2) Age the contaminated soils either 'artificially' in the laboratory or 'naturally'
in the field.
159
3) Compost the aged soils and freshly contaminated soils under predetermined
optimized composting conditions.
4) Analyze the composted samples and compare loss of TPH contamination
from freshly contaminated and aged contaminated soils.
7.2. Materials and Methods
The methods used for collection, spiking and ageing of soils drew on experience
and recommendations from several methods reported by Northcott and Jones
(2000a), Lee et al. (2003), Doick et al. (2003) and Daoust et al. (2006).
Uncontaminated soils were collected from two sites on the Imperial College,
Wye campus estate. The soils were a brick earth soil (Hamble soil series: an
argillic brown earth) and a gault clay (Evesham soil series: a calcareous pelosol)
(Burnham 1994). Approximately 4kg (wet weight) of each soil was collected
from a depth of 5 - 15cm.
Prior to spiking with oil, the soils were forced-rubbed through a 9.5mm sieve to
remove stones and vegetation material. A portion of each soil was stored in
sealed plastic bags at field wetness at 4°C. These samples were then freshly
contaminated immediately prior to composting.
The remaining soil samples were separated into 2kg portions for the four ageing
treatments: 1) brick earth - artificial ageing; 2) brick earth - natural ageing; 3)
160
gault clay - artificial ageing; and 4) gault clay - natural ageing. The soils were
artificially spiked to 10,000ppm (wet weight) with transformer oil (provided by
EDF Energy) to simulate a transformer leak. The soils were weighed in an
acetone-cleaned glass bowl and the transformer oil (20g) was added; the oil and
soil were mixed using a clean steel spoon and the mixture was homogenized for
5min. Between spiking of different samples the mixing bowl and spoon were
cleaned with acetone to remove any oil contamination.
Homogeneity of the spikes was tested to ensure that the oil had been evenly
distributed in the soil samples, because heterogeneity may result in localized sub-
systems of high and low contaminant concentration, giving differences in soil-
contamination interactions and contaminant bioavailability. The homogeneity of
spiking was examined by taking sub-samples of 5g randomly from each soil in
triplicate. The soil samples were extracted and analyzed as described in Chapter
3. The contamination in each sample was determined and the percentage
variation from 10,000 ppm concentration was calculated (Table 7.1). The
percentage variation of the spiked samples of soil from 10,000ppm was 3.19% at
worst, showing that the spiking homogeneity was high and acceptable for
investigation.
161
Table 7.1. Percentage variation of contaminant spiking in each soil from
10,000ppm
Percentage variation (1 %)
Gault clay Brick earth
Ageing treatment
Freshly contaminated 1.77 2.19
Naturally aged 1.84 1.56
Artificially aged 2.23 3.19
The spiked soils were then either 'artificially' aged indoors using wetting-drying
cycles or 'naturally' aged outdoors by exposure to prevailing environmental
conditions. Ageing of the soil samples was carried out for 3 months, which
simulated a field situation when a major spill would be cleaned up within a
relatively short time of the soil being contaminated. For artificial ageing, soils
were placed into pre-cleaned amber glass jars, with wide necks to aid soil drying,
which were kept at room temperature in the dark. These soils underwent wetting
and drying cycles consisting of 2 days at field moisture content followed by 12
day drying periods. During the wetting stage lids were kept on and during drying
the lids were removed to allow samples to dry. During the 3 month ageing period
the soils underwent six wetting-drying cycles. The moisture content was
measured on the 10th day of the drying period to determine the amount of water
to be added, by taking a 5g sample randomly from the jar, determining the
moisture content (as described in Chapter 3) and adjusting the moisture content
with distilled water. For artificial ageing, indoor temperatures close to those
outdoors were sought. This was determined by testing the temperatures at
162
different locations in the laboratory over 3 days to find those that were
comparable to external ambient temperatures.
The soils to be naturally aged were placed in cleaned plastic containers, which
had small holes drilled in the bottom to aid soil drainage, in order to prevent
waterlogging and anaerobic conditions. This container was placed inside a larger
container that served to collect excess water draining from the soils. The inner
container was supported on glass stoppers to prevent the soil sitting in drained
water. The outer container was emptied periodically to prevent water build-up.
The containers were buried, leaving 1 inch of the outer container above ground.
The containers were covered with black mesh and wire cages to allow moisture
to pass through, but prevent animal access and disturbance. The weather
conditions were recorded daily over the three months of ageing, to show that
temperatures for each experiment were comparable. Temperatures ranged from 8
to 15°C indoors and 0 to 12°C outdoors; indoor temperatures were on average
4.3±1.4°C higher than those outdoors (Figure 7.1).
163
18 16
IS 14 0 — 12 a) I5 10 6- 8
6 a) - 4
2 0
--•— Inside temperature - Outside temperature
CD 0 N
C.
CD 0 N CO 0
CD 0 N
LO
CO 0 N
N N
CO 0 N
CY>
0
1.0 O
N- 0
CN1
N-- 0
N. 0
0 (7.0 N
N. 0 N
0,1 0
r- r, 0 0 N C\1 C) 0) CD 0
N- 0
0,
0 C..)
N- 0 CO
Date
Figure 7.1. Average inside and outside temperatures during ageing of
contaminated soils over 3 months
The stored samples of uncontaminated soils were spiked with transformer oil (as
described above) as the freshly contaminated samples. Prior to composting the
aged soils were mixed using a metal spatula. The aged and freshly contaminated
soil samples were composted with green waste using the optimal conditions
described in Chapter 5 (0.8:1 soil to green waste ratio, 38°C and 60% moisture
content). Samples of freshly contaminated and aged-contaminated (`naturally'
and 'artificially') soils were used as experimental controls during the composting
period. Composting was carried out for 98 days, with destructive sampling
performed at 0, 7, 21, 45, 54, 80 and 98 days (as described in Chapter 5).
164
7.3. Results
During composting, the overall loss of TPH (Figure 7.2) was greater in the
freshly contaminated soils than in the aged samples (5.21 to 9.02% for gault clay
and 4.87 to 17.9% for brick earth). In the aged soils, the artificially aged samples
had greater losses than the naturally aged soils (3.81 to 6.32% and 13.1 to
21.01% in gault clay and brick earth, respectively). However, none of the
differences in total TPH loss were statistically significant (p>0.05) between aged
and freshly contaminated soils. This suggests that contaminant ageing had no
affect on the removal of TPH during composting, or that 3 months was not long
enough to allow the contamination to become sequestered and the ageing
phenomenon to develop. Therefore, there was little difference between the
freshly contaminated and aged soils. A similar pattern of TPH loss was observed
in the soil controls (Figure 7.3), also indicating that in untreated soils ageing did
not significantly affect the loss of TPH. Losses from the gault clay were greater
than from the brick earth soil, which was more pronounced in the aged soils than
in the freshly contaminated soils, especially in the naturally aged soils, although
these were not significant.
165
90 80 70
• 60 T) 50
40 °- 30
20 10 0
Gault clay Brick earth Gault clay Brick earth Gault clay Brick earth fresh fresh natural natural artificial artificial
Ageing treatment
Figure 7.2. Percentage loss of TPH from soils with different contaminant ageing
treatments during composting
50 - 45 - 40 35 -
y 30 -0 25- _
20 - o. 1— 15 -
10 - 5 - 0
Gault clay Brick earth Gault c ay Brick earth Gault clay Brick earth fresh fresh natural natural artificial artificial
Ageing treatment
Figure 7.3. Percentage loss of TPH from soil controls with different contaminant
ageing treatments
During the composting of gault clay samples (Figure 7.4a) the changes in TPH
concentration were comparable for all treatments. Although differences in the
loss patterns were observed, the error bars indicate there was a large amount of
166
"6 5000
▪ 4000 -
E • 3000 -0
4E' • 2000 -a)
1000 - x 1-- 0
• 5000
▪ CO • 4000 -
F • 3000 -o 2 r2 -5 2000 - V
1000 -
a 0
variation within the treatments, thus rendering the differences non-significant
statistically. A similar result was obtained for the brick earth soil with the
different ageing treatments (Figure 7.4b). The overall change in TPH
concentration during composting was similar for all ageing treatments in both
soils, indicating that ageing did not affect the removal of TPH during composting
in this study. This might be due to the ageing period not being sufficient, or that
the prevailing environmental conditions had little or no effect.
0 7 21 45 54 80 98
Time (days)
(a)
- Freshly contaminated
- Naturally aged r, Artificially aged
(b)
Freshly contaminated —o— Naturally aged
-A.- , Artificially aged
0 7 21 45 54 80 98
Time (days)
Figure 7.4. Changes in TPH concentration during composting of a) gault clay
and b) brick earth with different ageing treatments
167
In the soil controls (Figures 7.5a and b), the freshly contaminated samples of
both soils had larger changes in TPH concentration than did the aged
contaminated soils. Losses from the freshly contaminated soils were slower at the
start of the experiment compared with the aged soils, with losses increasing over
the aged soils at 21 days and 45 days for brick earth and gault clay, respectively.
The pattern of TPH loss for the ageing treatments was comparable for each
treatment in both soils, again indicating that ageing did not have an effect or that
the ageing period was not long enough to show an effect.
0)
p
10000
8000
6000
a) 4000
3 2000
0
a)
10000
a) 8000
E c 6000
4E. 4000 41) 0
2000
0
(a)
Freshly contaminated --a--- Naturally aged
A Artificially aged
0 7 21 45 54 80 98
Time (days)
(b)
Freshly contaminated
—a—Naturally aged Artificially aged
0 7 21 45 54 80 98
Time (days)
Figure 7.5. Changes in TPH concentration of a) gault clay and b) brick earth
controls with different ageing treatments
168
First order kinetic analyses were performed for the soils and the soil controls.
The pseudo-first-order kinetic approximation was applied using the linear
integrated form of equation (7.1):
In (C/Co) = -kt (7.1)
where C is the concentration at time t, Co is the concentration at t = to, k is
the first-order constant of removal (obtained by linear regression) and t is time
(Admon et al. 2001; Antizar-Ladislao et al. 2005a).
The rate constant and R2 values for each of the soils with the different ageing
treatments are shown in Table 7.2. Again, it can be seen that the rates of loss of
TPH from each treatment in both soils were comparable, with no significant
differences seen in the rates of loss between the freshly contaminated soils and
the aged soils. This shows that, in this study, ageing of contaminated soils had no
affect on contaminant loss.
Table 7.2. Rate constant (k) and R2 values for TPH losses from freshly
contaminated and aged contamination gault clay and brick earth
Gault clay Brick earth
k (day') R2 k (day') R2
Composted Freshly contaminated 0.016 0.96 0.016 0.97
Naturally aged 0.013 0.91 0.010 0.98
Artificially aged 0.013 0.94 0.013 0.97
Control Freshly contaminated 0.0063 0.94 0.0055 0.97
Naturally aged 0.0092 0.96 0.0057 0.99
Artificially aged 0.0095 0.97 0.0073 0.97
169
7.4. Discussion
Total losses of TPH during composting from freshly contaminated soils were
higher than those from aged soils. The greater losses from fresh soils may have
been due to the compounds being more bioavailable to microbial degradation
than the aged contaminants that would have become sequestered and therefore
less available for biodegradation (Northcott and Jones 2001).
Although losses of TPH from freshly contaminated soils were greater than from
aged soils, differences were not large and not significantly different. Huesemann
et al. (2002) measured the extent of biodegradation of fresh and aged PAHs and
found similar effects. However, generally in other ageing studies, aged
contamination resulted in lower losses than fresh contamination. For example,
White et al. (1998) observed only 23.8% removal of phenanthrene after 151
days, compared to 40.8% in freshly contaminated soils, and Allan et al. (2006)
found that only 9.9% of added phenanthrene was removed after 100 days
compared to 68.8% in freshly spiked soil.
Comparable losses in the present study may have been due to the ageing period
not being long enough for the majority of compounds to be sequestered and,
therefore, become less bioavailable. Sorption is a slow process and may take
longer time periods of hundreds of days (Sharer et al. 2003). Therefore, in 3
months of ageing little differences in sorption may have occurred so leaving
contaminants comparably available for degradation as freshly added
contaminants despite differences in the artificial and natural ageing conditions
the contaminated soils were exposed to here. In other studies, longer periods of
170
ageing have been needed for contaminants to become less bioavailable, for
example 365 days for naphthalene, 200 days for phenanthrene and 203 days for
anthracene (Alexander 2000).
Another reason for the comparable TPH losses in the present study may have
been due to the concentration of oil used to spike the soils. In other ageing
studies the soils were generally spiked with single compounds at low
concentrations (i.e. 1 to 10mg/kg of soil), which may mean that most, if not all of
the contamination will sorb to soil mineral and organic matter more readily and
become increasingly sequestered with time (Huesemann et al. 2002). Here, the
electrical insulating oil investigated was a non-aqueous phase liquid (NAPL), so
it is likely that within the time period of the investigation, intramolecular
attraction of contaminants within the NAPL itself were greater than their
attraction to soil components so the waste itself controlled the contaminant
bioavailability rather than soil processes. This is also supported by the results
presented in Chapter 6 that show that differences in soil characteristics had little
effect on potential for loss of TPH by composting bioremediation. These finding
are supported by Huesemann et al. (2002) who reported that if the NAPL
characteristics did not change during ageing, then ageing would not have an
effect on contamination biodegradation kinetics.
Similar losses for all treatments during composting may also have been observed
due to the extraction method used. ASE® is an efficient extraction method
(Chapter 3) and uses high temperatures and pressures to extract analytes of
interest. This vigorous extraction method may have resulted in extraction of
171
contaminants that had become sorbed and less available for degradation to be
extracted (Kelsey et al. 1997; Northcott and Jones 2000b). Therefore, the
recovery of analytes might not have been representative of the fraction that was
bioavailable, although all experiments and their results are to some extent
ultimately operationally-defined by their design and methods used. Therefore, it
appears that similar amounts were lost in fresh and aged contaminated soils.
At the start of composting, the concentration of TPH was lower in the aged soils
compared to the fresh soils. This may have been due to losses of contamination
during the ageing process; during the natural ageing process losses were 11.09
and 9.36% for gault clay and brick earth respectively, while losses during
artificial ageing were 14.8 and 11.02% for gault clay and brick earth. Losses
could have been by volatilization of lower molecular weight contaminants,
leaching (especially in the naturally aged soils) and degradation (Hatzinger and
Alexander 1995; Beck et al. 1996; Allan et al. 2006).
The slower losses of TPH from freshly contaminated soils at the start of
treatment, especially for the brick earth, may have been due to adaptation by the
soil microbial community to the added contamination, slowing the degradation of
TPH, compared to that in the aged soils, whose communities may have already
become adapted due to prior exposure during ageing (Lee et al. 2003). However,
in the freshly contaminated soils, once the indigenous community of
microorganisms had become acclimatized to the contamination, activity would
have increased over that in the aged soils (Zucchi et al. 2003; Trindade et al.
172
2005). This variation in loss of TPH from the freshly contaminated soils would
have resulted in overall similar losses for them as for the aged soils.
Composting stimulated the removal of TPH contamination compared to the soil
controls, which could be due to stimulation of the soil microbial populations by
organic matter and the addition of microorganisms present on the green waste
used for composting (Trindade et al. 2005).
7.5. Conclusions
The loss of TPH from freshly spiked soils and aged contaminated soils subjected
to composting bioremediation was comparable, which suggests that the ageing of
a soil contained with a NAPL at high concentration for 3 months does not affect
the removal of TPH. Additionally, it suggests that in terms of contamination aged
soils do not have to be immediately treated, although it is more environmentally
responsible to do so. However, it should be noted that longer periods of
contaminant ageing may result in more recalcitrant contamination than
demonstrated in this study. Therefore, experiments should be performed using
contaminants that have been aged for longer periods to determine if longer
contaminant ageing reduces the removal of TPH by composting bioremediation.
Additionally, the affect of ageing may have been more marked for individual
compounds rather than TPH as a whole: this aspect is explored in Chapter 8.
Overall, the results are encouraging for the clean-up of contaminated soils after a
recent spill, with the best clean-up results achieved before ageing effects have
set-in
173
Chapter 8. Effect of Contaminant Physico-Chemical Properties
on Removal of Contamination During In-Vessel Composting
8.1. Introduction
Different classes of compounds and individual compounds have different
physico-chemical properties that affect their behaviour and fate in soil (de Maagd
et al. 1998; Ribeiro and Ferreira 2003; Semple et al. 2003). These properties
include molecular weight, water solubility, vapour pressure and indices of
sorption such as Koe and Kow (Table 8.1). Properties affect compound
susceptibility to microbial activity as well as the rate and extent of
biodegradation (Haus et al. 2001; Huesemann et al. 2003; Godoy-Faimdez et al.
2008). These properties also govern the distribution of a compound among
environmental phases and are used to predict its environmental fate (Staikova et
al. 2004). The environmental partitioning of organic compounds occurs between
the liquid phase, the gas phase, the dissolved phases in water and the solid phase.
Partitioning is expressed by partitioning coefficients, such as octanol—water
partition coefficient (log K0 ) and organic carbon-water partition coefficient (log
Kos) (Ballschmiter 1996; Beyer et al. 2002).
The susceptibility of hydrocarbons to degradation varies with compound group
and size (Englert et al. 1993). There is a general inverse relationship between
compound molecular weight and its potential for, as well as rate of, degradation
(Wang et al. 2001; Lima et al. 2005; Wammer and Peters 2005), with decreasing
susceptibility from alkanes to aromatic compounds (Capelli et al. 2001; Labud et
al. 2007). The structure of compounds also contributes to degradation patterns
174
(Leblond et al. 2001), with the more soluble isomer of an aromatic compound
being preferentially degraded (Lima et al. 2005). Hydrocarbon susceptibility to
degradation is correlated to their structures and molecular weight (Wang et al.
2001). Therefore, the removal of higher molecular weight hydrocarbons will be
slower and less than that of lower molecular weight compounds.
Organic pollutant persistence in soil is related to hydrophobicity (de Maagd et al.
1998). Indications of compound hydrophobicity are given by compound aqueous
solubility and octanol-water partition coefficients (log Kow) (Reid et al. 2000;
Verbruggen et al. 2000; Tolls et al. 2002). Water solubility indicates the potential
for chemical movement and distribution; highly soluble compounds are more
distributed in the environment (Verschueren 1996). Low hydrocarbon solubility
results in reduced microbial degradation due to slow movement from a non-
aqueous to the aqueous phase liquid in which they are metabolized.
Hydrophobicity also affects compound sorption and desorption (Bardi et al.
2000; Huesemann et al. 2003). Hydrophobicity decreases compound availability
and degradation during composting.
The octanol-water partition coefficient (log Kow) of a compound is the ratio of its
concentration in octanol and in water at equilibrium (USGS 2008). It is an
estimate of compound hydrophobicity and partitioning tendency from water to
lipids or organic matter (Mackay et al. 1992). It is used to predict the soil
partitioning, biological uptake and biomagnification of a compound
(Verschueren 1996; Staples et al. 1997; Sharer et al. 2003). The boiling point of a
compound is the temperature at which the vapour pressure is equal to the
175
atmospheric pressure. The boiling points of organic compounds give indications
about other physical properties and structural characteristics. As the chain length
or number of rings increases, the boiling points increase (Ophardt 2003). Vapour
pressure is the pressure of a compounds vapour phase at equilibrium with
compounds pure phase. It is important in modelling fate and transport of
atmospheric pollutants, as it determines whether a compound will volatilize and
how fast; pollutants with high vapour pressures tend to volatilize faster
(Dunnivant and Anders 2006; Hayes and Soni 2006).
Soil organic carbon content is often the most important determining factor in soil
adsorption of compounds, particularly non-ionic compounds, and their mobility
(Sabi* et al. 1995). Compound soil-water partition coefficient (log Ic,e) is used
to indicate compound mobility in soil-water systems (Baker et al. 2000; Ferreira
2001; Krauss and Wilcke 2001). Compounds with higher log Koe values are less
mobile than those with lower values (Ferriera 2001). Therefore, compounds with
higher log Ice values will be less available for degradation.
The physico-chemical properties of contaminants affect the availability of a
contaminant to biodegradation and other loss mechanisms. Therefore, some
compounds are less susceptible to degradation, making it harder to remove them
during composting bioremediation. This chapter examines the relationships
between the loss of . compounds and their physico-chemical properties during
composting bioremediation and how these are affected by different operating
parameters, contaminant ageing and soil type.
176
Table 8.1. Ranges of physico-chemical property values of compound groups found in TPH
Compound group Examples Structure Molecular weight
Boiling point (°C)
Solubility (mg/L at
25°C)
Log K.
Log Le
Vapour pressure
(mm Hg at 25°C)
Saturates (C10-C20)
decane pentadecane eicosane
pentadecane NVVV\AA
142.3 - 282.6
bcde
174.1 - 343 bcde
0.0019 - 0.052 bed
5.01 - , n n 1V.Z,
bcd 3.24 - 5.89 h
1.43- 4.62x 10-6 d
Saturates (C21-C35)
heneicosane hexacosane triacontane pentatriacontane
triacontane -13 296.3 - 492.9 cd
d
356.5 - 490 cde
2.47x 10 2.93, 10_8
-
10.65 -
17.53 6.16 - 9.88 h
8.7x 10-5 - 5.4x 10-12 d
2/3-Ring aromatics biphenyl naphthalene phenanthrene fluorene
naphthalene a so -- Oil
phenanthrene a -..
128.2 - 184.3 def
217.9 - 340 def 1.69 - 31 cd 3.30 -
4.6 cd 3.26 - 4.32 h
0.085 - 6.53x10-6 d
4-Ring aromatics fluoranthene chrysene SO chrysene a 00 202.3 -
228.9 df 383 -
435 dfg 0.002 - 0.26 d
5.16 - 5.81df
4.84 - 5.37 h
9.2x10-1° - 6.2X10-9 d
5/6-Ring aromatics benzo(b)fluoranthene benzo(a)pyrene benzo(ghi)perylene indeno(123cd)pyrene
benz(a)pyrene a WOO.
IS
benzo(ghi)perylene a 41a. Wirli
252.3 - 278.4 cdf
480 - 550 cdf
0.025 - 0.004 ed
5.78 - 6.75 d
5.37 - 6.43 h
5x10-7- 1,(10-10 d
a Lima et al (2005); b Verbruggen et al (2000); Toxnet (2008); d Syracuse Research Corporation (1999); e Chemistry, Oxford University (2007); Ribeiro and Ferriera (2003); g Ferreira (2001) and; ChemSpider (2008)
177
8.1.1. Aims and Objectives
Increasing values of properties such as molecular weight, boiling point, log Kow,
log Ko, and decreasing values of properties such as aqueous solubility and
vapour pressure make compounds less available and lead to lower removal of
contaminants. The aims of this chapter were to determine the relationships
between compound physico-chemical properties and loss of the compound under
optimized composting conditions, then examine the effect of changing the
composting conditions on these relationships. In addition, the relationship
between physico-chemical properties and loss of aged contamination compared
to fresh contamination were examined.
To achieve these aims, the following objectives were set:
1) Correlate losses of individual compounds with their physico-chemical
properties to determine the strength of relationships.
2) Compare relationship patterns found in optimized composting conditions
with different composting operating parameters to determine if composting
conditions affect relationships patterns.
3) Compare loss and compound property relationships in freshly contaminated
soils to those in aged contaminated soils.
178
8.2. Materials and Methods
The relationships between physico-chemical properties and the percentage loss
of a compound were examined using correlations. Physico-chemical data on
molecular weight, boiling point, octanol-water partitioning, organic carbon-water
partitioning aqueous solubility and vapour pressure was compiled for each
compound, and compounds were put in order of increasing or decreasing values
for each physico-chemical property, which are summarized in Table 8.1.
Percentage losses of each compound during composting were calculated from
optimization, ageing and soil property experiments. For each physico-chemical
property, the individual compound losses were plotted against the physico-
chemical values of compounds. Aqueous solubility and vapour pressure
compound properties had a wide range of values, therefore, these were plotted on
log scale in order show the relationship clearly.
Models for the relationships were plotted using Microsoft Excel. The linearity of
the relationship was calculated to give a R2 value, which showed the strength of
the relationship between losses of compounds and their physico-chemical
properties values. The slope of the trend-line also indicated if the losses and the
properties were negatively or positively correlated. Correlations were carried out
for losses of compounds during the optimization of composting condition
experiments, the contaminant ageing experiments and soil property experiments.
Values for relationships were then compared to determine the effect of differing
operating conditions (changes in soil to green waste ratio [0.7:1 or 0.9:1 at 38°C],
temperature [55 or 70°C at 0.8:1], moisture content [40 or 80% at 38°C and
179
0.8:1), contaminant ageing (freshly contaminated compared to naturally aged
soils) as well as the effect of different soil types (gault clay and brick earth).
8.3.1. Relationships Between Compound Physico-Chemical Properties and
Losses During Composting Optimization
Figure 8.1 shows the relationship between increases in compound physico-
chemical property values and losses of individual compounds under optimal
composting conditions (0.8:1 soil to green waste ratio, 38°C and 60% moisture
content). No relationship was observed between losses of saturate compounds
with molecular weight (R2 = 0.079), boiling point (R2 = 0.046), aqueous
solubility (R2 = 0.030), octanol-water partitioning [log K°„] (R2 = 0.079), organic
carbon-water partitioning [log Koe] (R2 = 0.079) and vapour pressure (R2 =
0.0019). As values of these properties increased or decreased, losses of saturate
compounds did not decrease significantly, with the slope values of the models
being close to 0. This suggests that the physico-chemical properties of saturate
compounds do not significantly affect their removal during composting
bioremediation under the optimized conditions. Although it is possible that
compound physico-chemical properties interact with each other in more complex
way to control losses, the very low R2 values observed indicate that this was
unlikely and, therefore, under optimized conditions it was the degrading
organisms that dominated the compound loss processes making the chemical
characteristics of little importance in the loss process. Additionally, it can be seen
from Figure 8.1 that the losses of some of the lower molecular weight saturates
were less than the mid-molecular weight compounds. This could be due to
180
preferential degradation of the green waste than these compounds (Ostberg et al.
2006). Alternatively, the microorganisms present during composting may have
had specific ranges of compounds that could be degraded (Rapp and Gabriel-
hirgens 2003). Therefore, the lower molecular weight compounds may not have
been degraded as much due to lack of degradation capacity possessed by the
microorganisms. Additionally, the presence of the longer chains may have
inhibited the degradation of the shorter chains (Plohl et al. 2002).
By comparison, stronger relationships between physico-chemical properties and
losses of aromatic compounds were observed under the optimum composting
conditions (Figure 8.1). Increasing/decreasing property values and losses of
aromatic compounds were negatively correlated, with compounds such as
benzo(k)fluoranthene and benzo(ghi)perylene having lower losses than
compounds such as naphthalene and acenaphthene. The values of the model
slopes were further from 0 than the saturates, showing differences between
compound physico-chemical value property extremes of 48.3% and above. There
was a relationship between molecular weight (R2 = 0.75), boiling point (R2 =
0.75), log (R2 = 0.75), log lc, (R2 = 0.77) and loss of compounds. A positive
relationship was also observed for aqueous solubility (R2 = 0.65), although this
was not as strong as the other properties. A relatively weak relationship was
observed between decreasing vapour pressure and losses of aromatics (R2 =
0.37). This suggests that compound molecular weight, boiling point, log Kow, log
1(0, and aqueous solubility influence the loss of aromatic compounds, as the
values of these properties increase.
181
Loss
of c
ompo
und
(%;
100 -90 -80 -70 -60 -50 -40 -30 -20 -10-
• Saturates a Aromatics
Saturates Aromatics
L
(e)
tri** • +*
*Et*
(0 a a «•*'
t.! sg`E
--- g
1 0.01 0.0001 1E-06 1E-08 1E-10 1E-12
Vapour pressure (log mm Hg)
0
Loss
of c
ompo
und
(%)
100 -90 -80 -70 -60 -50 -40 -30 -20 -10-
El a 6,4 • e+ **I
E iii
a
100
(a) 90 - 80 70 60 50 40 30 20 10 0
100 - 90 -
.+4 *Pi 80
=▪ • 70 - o 60 - EE 50 -
o 40 - u, • 30 - In 3 20 -
10 -
0
Ck a i« • • (c)
fC 7
a
0
(b)
100 150 200 250 300 350 400 450 500 200 250 300 350 400 450 500 550 3 6 9 12 15 18
Molecular weight Boiling point (°C)
log Kow
Loss
of c
ompo
und
(*A]
100 (d) 100
90 - -°. — 90-
80 - 62 a 44 • e. 1.1 ' 80 -
70 - 4;."-- 4 -IM
-o a 70 - a
60 - 4 •,. a 0 • 60 - . 50 - q .2. E 50 -
8 40 - 40 - g
Pi 30 - . 6 30 to
8 20- 20 - a
—I 10 - 10- 0
3 4 5 6 7 8 9 10
log Koc Aqueous solubility (log mg 1:1)
• Saturates a Aromatics
Satruates Aromatics
0 100 0.1 0.0001 1E-07 1E-10 1E-13
Figure 8.1. Relationships between losses of saturate and aromatic compounds and physico-chemical properties of; a) molecular weight, b)
boiling point, c) log Kow, d) log Koe, e) aqueous solubility and f) vapour pressure
182
Under non-optimum composting conditions, increasing the composting system
operating temperature from 38°C to 55 or 70°C, resulted in stronger relationships
between loss of saturate compounds and molecular weight, boiling point, log Kow
and log Ice (Table 8.2), than those observed at the optimal 38°C, with compound
loss being negatively correlated with these physico-chemical properties.
However, increasing the temperature to 55°C resulted in a weaker relationship
between aqueous solubility and losses of saturate and aromatic compounds than
at 38°C. A further increase to 70°C led to a stronger relationship between
aqueous solubility and loss of saturate compounds, but a weaker relationship for
aromatic compounds. Increasing the temperature resulted in a stronger
relationship between the loss of saturate compounds and compound vapour
pressure, but for aromatic compounds a weaker relationship was observed at both
higher temperatures. This suggests that, in general, increasing the operating
temperature results in stronger relationships for saturate compounds, resulting
from lower removal of larger saturates. However, temperature changes did not
affect the relationship for aromatic compounds, suggesting that their physico-
chemical properties influence loss more than does environmental temperature.
These observations support earlier assertions that it is likely the nature of the
microbial community rather than the compound physico-chemical properties that
are a limiting factor on contaminant losses. The microbial community may be the
limiting factor in the loss of contamination as they are able to overcome
compound characteristics such as hydrophobicity through production of
degradation enzymes, adhesion/desorption to hydrocarbons and production of
biosurfactants (Ron and Rosenberg 2002; Tellez et al. 2002; Ward et al. 2003;
183
Molnar et al. 2002), which make hydrocarbons more readily degradable and
increase rates of degradation.
Table 8.2. R2 values for relationships between compound physico-chemical
properties and compound loss for increases at operating temperatures
Temperature (°C)
Group Molecular weight
Boiling point
Aqueous solubility
Log K.,
Log K0
Vapour pressure
38 saturates 0.079 0.046 0.029 0.079 0.75
0.079 0.77
0.0019 0.37 aromatics 0.75 0.75 0.65
55 saturates 0.44 0.50 0.0029 0.44 0.44 0.41 aromatics 0.78 0.80 0.19 0.77 0.75 0.15
70 saturates 0.63 0.67 0.075 0.63 0.63 0.29 aromatics 0.72 0.71 0.46 0.73 0.77 0.22
Changing the soil to green waste ratio also resulted in differing relationships than
those observed under optimized conditions between compound physico-chemical
properties. Increasing or decreasing the soil to green waste ratio (to 0.7:1 or
0.9:1) resulted in stronger relationships between molecular weight, boiling point,
log Kow, log Ko, and vapour pressure and loss of saturate compounds (Table 8.3),
with increased property values being negatively correlated to compound losses.
Conversely, aromatic compounds had weaker relationships between loss and
these properties compared to values in optimized conditions. Decreasing the soil
to green waste ratio resulted in weaker relationships between losses of saturate
and aromatic compounds and aqueous solubility, whereas an increased ratio
resulted in a stronger relationship for loss of saturate compounds and a
comparable relationship for aromatic compounds. This suggests that altering the
184
soil to green waste ratios affects the relationships between compound loss and
their physico-chemical properties. Changes in the relationship between losses of
compounds and physico-chemical properties could have been due to differences
in the microbial community at the different soil to green waste ratios compared
to the community present under optimized conditions (Chapter 9). The different
soil to green waste ratios may have resulted in lower numbers of bacteria
especially Gram-positive bacteria and actinomycetes during composting,
therefore resulting in less degradation.
Table 8.3. R2 values for relationships between compound physico-chemical
properties and compound loss for changes in soil to green waste ratio
Ratio Group Molecular weight
Boiling point
Aqueous solubility
Log Kow
Log 1(0e
Vapour pressure
0.7:1 saturates 0.49 0.49 0.013 0.47 0.49 0.31 aromatics 0.31 0.34 0.27 0.29 0.31 0.22
0.8:1 saturates 0.079 0.046 0.029 0.079 0.079 0.0019 aromatics 0.75 0.75 0.65 0.75 0.77 0.37
0.9:1 saturates 0.52 0.53 0.075 0.52 0.52 0.27 aromatics 0.61 0.64 0.69 0.59 0.61 0.37
Variable relationships between losses and different properties were observed
with different sample moisture contents (Table 8.4). Decreasing the moisture
content to 40% resulted in stronger relationships for the loss of saturate
compounds with all the physico-chemical properties, whereas there were weaker
relationships for aromatics for most of the properties apart from vapour pressure,
from which the relationship was stronger than at 60%. Increasing the moisture
content to 8 0 % resulted in comparable relationships to 60% moisture content
185
between molecular weight, log Kota and log Ice and loss of saturates and
aromatics and for loss of aromatics with boiling point. Stronger relationships
were observed for the loss of saturates and boiling point and vapour pressure.
There was a weaker relationship between aqueous solubility and losses of
saturate and aromatic compounds. These differing relationships between
physico-chemical properties and compound losses at different moisture, suggest
that different moisture contents have an effect on the relationship between
compound properties and loss of the compounds.
Table 8.4. R2 values for relationships between compound physico-chemical
properties and compound loss using different moisture contents
Moisture content
Group Molecular weight
Boiling point
Aqueous solubility
Log K.
Log K.
Vapour pressure
(%) 40 saturates 0.15 0.21 0.049 0.15 0.16 0.45
aromatics 0.59 0.59 0.50 0.57 0.54 0.50 60 saturates 0.079 0.046 0.029 0.079 0.079 0.0019
aromatics 0.75 0.75 0.65 0.75 0.77 0.37 80 saturates 0.078 0.11 0.0093 0.077 0.078 0.37
aromatics 0.79 0.78 0.27 0.75 0.69 0.15
8.3.2. Relationships Between Physico-Chemical Properties and Compound
Losses in Freshly Contaminated and Aged Contaminated Soils
In freshly contaminated soils (Figures 8.2a and b) no relationship was observed
between increasing molecular weight and losses of saturates (gault clay R2 =
0.097 and brick earth R2 = 0.035), whereas after ageing (Figures 8.2c and d) a
186
relationship, albeit a very weak one, was observed with molecular weight with
losses of saturate compounds being negatively correlated (gault clay R2 = 0.48
and brick earth R2 = 0.31). The slope value decreased after ageing, from 0.032
(freshly contaminated) to -0.11 for gault clay and from 0.022 to -0.093 for brick
earth. A relationship was observed for losses of aromatic compounds and
molecular weight in freshly contaminated soil (gault clay R2 = 0.89 and brick
earth R2 = 0.77). However, after the contamination was aged a weaker
relationship for the loss of aromatic compounds (gault clay R2 = 0.77 and brick
earth R2 = 0.51) was observed (Figures 8.2c and d). The slope values increased
after ageing; for gault clay from -0.46 to -0.39 and for brick earth from -0.39 to -
0.26. Although, after ageing it would be expected that the compound physico-
chemical properties would have more of an effect on the loss of compounds, it
appears that it is the microbial community present during composting has a
greater influence on the loss of compounds. This could be due to adaptation to
contamination by the microbial community during ageing (Lee et al. 2003), as
well as adaption of microorganisms in the freshly contaminated soils (Zucchi et
al. 2003; Trindade et al. 2005), resulting in biodegradation of the contamination.
187
Loss
of
com
pou
nd (
%)
100 90 80 70 60 50 40 30 20 10 0 100
(d) (c) 100 90 80 -70 -60 -50 -40 -30 20 -10-
0
*i
Loss
of
com
pou
nd
(%)
• Saturates o Aromatics
- Saturates • - • • Aromatics
500 400
100 - 90 - 80 70 60 50 -40 -30 -20 10 0 100 200 500 100 400 200 300
Molecular weight 300
Molecular weight
(a) • ••6=
E; • U •
f
• Saturates o Aromatics
- Saturates - - - • Aromatics
200
300
400
500 Molecular weight
100 (b) 90 - 80- i 1 ' iii
a
a fL 0
±_t_i____
0
50 'HI f i 40
30
20 i 4 10
0 100 200 300 400 500
Molecular weight
Figure 8.2. Relationships between increasing compound molecular weight and
loss of compounds in; freshly contaminated a) gault clay, b) brick earth and aged
contaminated c) gault clay, d) brick earth
In freshly contaminated soil no relationship (Figures 8.3a and b) was observed
between compound boiling point and losses of saturates (gault clay R2 = 0.062
and brick earth R2 = 0.021), both soils had slope values close to 0. However, a
stronger relationship was observed in aged contaminated soil (Figures 8.3c and
d), with a negative correlation between increasing boiling point and loss of
saturates and decreased steeper gradients (gault clay R2 = 0.46 and brick earth R2
= 0.29). Strong relationships were observed for loss of aromatic compounds and
increasing boiling point in freshly contaminated soil (gault clay R2 = 0.85 and
188
• Saturates El Aromatics
- Saturates - - - •Aromatics
0 0
Loss
of c
omp
ound
(%)
100 -90 - 80 -70 -60 -50 -40 -30 -20 -10 -
(c) Lo
ss o
f com
poun
d ( %
)
100 90 80 70 60 50 40 30 20 -10 -
(d)
brick earth R2 = 0.79), whereas in aged soil (Figures 8.3c and d) it was weaker
for brick earth (R2 = 0.53) but comparable for gault clay (R2= 0.81). Lo
ss o
f com
pou
nd (%
)
Loss
of c
ompo
und
( %)
100 - 90 80 70 60 - 50 - 40 f. 2 30
0
10 0 200 250 300 350 400 450 500 550
Boiling point (°C)
100 -90 -80 -70 -60 -50 -40 -30 -20 -10 - 0 200 250 300 350 400 450 500 550
Boiling point (°C)
(a)
• Saturates Aromatics Saturates
Aromatics
200 250 300 350 400 450 500 550 200 250 300 350 400 450 500 550
Boiling point (°C)
Boiling point (°C)
Figure 8.3. Relationship between increasing compound boiling point and loss of
compounds in; freshly contaminated a) gault clay, b) brick earth and aged
contaminated c) gault clay, d) brick earth
No relationship between increasing compound aqueous solubility and loss of
saturates (Figures 8.4a and b) was observed in freshly contaminated soil (gault
clay R2 = 0.0019 and brick earth R2 = 0.084), whereas after ageing a stronger
relationship (gault clay R2 = 0.36 and brick earth R2 = 0.23) was observed
(Figures 8.4c and d), with positive correlations between compound loss and
physico-chemical property value and increased line gradients. In freshly
189
Lo
ss o
f com
pou
nd (
%)
100 90 80 70 -60 -50 -40 -30 20 10 -
(c)
Lo
ss o
f co
mpo
und
(VD)
100 -90 -80 -70 -60 -50 -40 -30 -20 -10 -
* it - if if
• Saturates o Aromatics
— Saturates - - - • Aromatics
0
contaminated soil (Figures 8.4a and b), aromatic compounds had a relatively
weak relationship between loss and aqueous solubility (gault clay R2 = 0.31 brick
earth R2 = 0.19). In aged soil the relationship was slightly weaker than freshly
contaminated soils for gault clay (R2 = 0.23), but a similar relationship to fresh
contamination was found in brick earth (R2 = 0.17).
(a) (b) 100 -
80 -i ii m 1 i If i;
•• . ti 90 -
70 60 - t et .. r • 'f i ' if 50 - 40- -- - - i 30
"Ili,I 1
20 f 10- 0 i
1E-13 100 0.1 0.0001 0.0000001 1E-10 1E-13
Aqueous solubility (log mg L-1)
• Saturates a Aromatics
— Saturates - - - • Aromatics
0 100 0.1 0.0001 0.0000001 1E-10
Aqueous solubility (log mg L1)
100 0.1 0.0001 0.0000001 1E-10 1E-13
Aqueous solubility (log mg I:1)
Figure 8.4. Relationship between decreasing
100 - 90 - 80 - g 70 -
TT
5 60 -0 -
40 - 30 -20 -10 - 0
100 0.1 0.0001 0.0000001 1E-10 1E-13
Aqueous solubility (log mg L-1) aqueous solubility and loss of
(d)
compounds in; a freshly contaminated a) gault clay, b) brick earth and aged
contaminated c) gault clay, d) brick earth
There was no relationship between compound log K0W and loss of saturates
(Figures 8.5a and b) in freshly contaminated soil (gault clay R2 = 0.0096 and
190
(a) (b)
80 70 60 50 40 30 20 10
♦ Saturates ® Aromatics
- Saturates - • - •Aromatics
a.
0 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 3
log Kow 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
log Kow
♦ Saturates o Aromatics
-Saturates • - • .Aroamtics
Loss
of c
ompo
und
(%)
100 90 80 70 60 50 40 30 20
0 10
0
100 -90 -80 -70 -60 -50 -40 -30 -20 -10 -
0
100 90 80 70 60 50 40 30 20 10
Loss
of c
ompo
ud (%
)
brick earth R2 = 0.035), but a stronger relationship was observed for aromatic
compounds (gault clay R2 = 0.85 and brick earth R2 = 0.31). After ageing a
stronger relationship (R2 = 0.48) was observed for losses of saturates, with
gradients decreasing in value (Figures 8.5c and d) and log Kow, whereas there
was a slightly weaker relationship for aromatics than freshly contaminated soil
for gault clay (R2 = 0.76) and an even weaker one for brick earth (R2 = 0.49).
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 log Kow
Figure 8.5. Relationship between increasing Kow point and loss of compounds in;
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 log kow
18
freshly contaminated a) gault clay, b) brick earth and aged contaminated c) gault
clay, d) brick earth
No relationship in freshly contaminated soils (Figures 8.6a and b) was observed
for log Koc and losses of saturate compounds (gault clay R2 = 0.097 and brick
191
(a)
0
100 90 80 70 60 50 40 30 20 10
Loss
of c
ompo
und
(%)
a
• Saturates a Aromatics
- Saturates • • - .Aromatics
100
90 80 70 60 50 40 30 20 10
0
(b)
z
3 4 5 6 7 8 9 10 log Koc
(d)
a
f •
earth R2 = 0.35). After ageing (Figure 8.6c and d), the relationship was stronger
between Ko, and loss of saturate compounds (gault clay R2 = 0.46 and brick earth
R2 = 0.31) and negative gradients were observed. The relationship between log
Ka, and loss of aromatics was strong in freshly contaminated soils (gault clay R2
= 0.81 and brick earth R2 = 0.77), whilst after ageing (Figures 8.6c and d) the
relationship was the same for gault clay (R2 = 0.81) but weaker for brick earth
(R2 = 0.55).
Loss
of c
ompo
und
(%)
3 4 5 6 7 8 9 10 log Koc
100 i4, iczat + (c) 100 - 90 - 90 80
f 80
70 i ..4
70 -
1 50 60 60 - 50 es 40 40 - 30 30 - 20 20 10- 10 0 ' 0
• Saturates Es Aromatics
- Saturates - - • Aromatics
4 5 6 7 8 9 10 3 4 5 6 7 8 9 10 log Koc log Koc
Figure 8.6. Relationship between increasing compound Koc and loss of
compounds in; freshly contaminated a) gault clay, b) brick earth and aged
contaminated c) gault clay, d) brick earth
192
(a) • 100
90 Is
Is
• Saturates O Aromatics
- Saturates - - • Aromatics
0.01 0.0001 1E-06 1E-08 1E-10 Vapour pressure (log mm Hg)
80 70 60 50 40 30 20 10 0
1E-12
100 90
3 80 a 70 E 60 ._-) 50
N• 40 3 30
20 10 0
(c) 100 90 80 70 60 50 40 30 20 10
• Saturates O Aromatics
- Saturates - - - • Aromatics
0
1 0.01 0.0001 1E-06 1E-08 1E-10 1E-12 Vapour pressure (log mm Hg)
100 90 80
c▪ 70 O • 60 o. E 50
40 • 30 o 20 J
10 0
(b)
-1 41 + f
(d)
0.01 0.0001 0.000001 1E-08 1E-10 1E-12 Vapour pressure (log mm Hg)
In freshly contaminated soils (Figures 8.7a and b) no relationship was observed
between vapour pressure and losses of saturate compounds gault clay soil (R2 =
0.046), whereas there was a weaker relationship for brick earth (R2 = 0.13). After
ageing (Figures 8.7c and d) the relationship remained weak between vapour
pressure and loss of saturate compounds in gault clay (R2 = 0.096) and brick
earth (R2 = 0.096). The relationship between vapour pressure and loss of
aromatics was weak in freshly contaminated soils (gault clay R2 = 0.15 and brick
earth R2 = 0.16), while after ageing (Figures 8.7c and d) the relationship was
comparable for both gault clay (R2= 0.15) and brick earth (R2 = 0.12).
1 0.01 0.0001 1E-06 1E-08 1E-10 1E-12
Vapour pressure (log mm Hg)
Figure 8.7. Relationship between increasing compound vapour pressure and loss
of compounds in; freshly contaminated a) gault clay, b) brick earth and aged
contaminated c) gault clay, d) brick earth
193
These results indicate that the ageing of contamination results in stronger
relationships between saturate compound loss and physico-chemical properties,
reducing removal of these compounds during composting of aged soils.
However, ageing resulted in weaker relationships between compound properties
and aromatic losses, which may be due to smaller losses of these compounds in
the aged soils. The losses from aged soils may have been smaller due to the
effects of ageing on the compounds; even the effects of ageing were not
significant (Chapter 7), some ageing effects may have led to compound
sequestration and, therefore, some persistent contamination, resulting in lower
losses.
8.3.3. Differences in soil types
In freshly contaminated soils (Table 8.5), there were weaker relationships in
brick earth soil between loss of saturate and aromatic compounds with molecular
weight, boiling point, log K, and Log Koc. However, saturates had a stronger
relationship for aqueous solubility and vapour pressure, whereas aromatics had a
weaker relationships with aqueous solubility, but the relationship for vapour
pressure was comparable to that of the gault clay. Weaker relationships were
observed in the aged brick earth compared to aged gault clay for aromatic and
saturate compounds for all the physico-chemical properties apart from vapour
pressure, which had a comparable relationship to gault clay. This suggests that
different soil types affect the relationship between losses of compounds and their
physico-chemical properties. This could be due to gault clay having a finer
texture than brick earth due its higher clay and silt content of 72% (compared to
194
36% in brick earth). This finer texture means that the gault clay has a higher
microporosity and, therefore, the contamination would have become more
concentrated within these finer fractions than in the brick earth (Amellal et al.
2001). Therefore, in the gault clay the contamination may have been less
available (Aichberger et al. 2006) and biodegradation rates may have been higher
in the sandier brick earth (Rizzo et al. 2008), resulting in weaker relationships in
brick earth, as the compounds were more readily removed.
Table 8.5. R2 values for relationships between compound physico-chemical
properties and percentage compound loss in freshly contaminated and aged
contaminated gault clay and brick earth
Gault clay Brick earth Fresh Aged Fresh Aged
Molecular weight saturates 0.097 0.85
0.035 0.48 0.31 aromatics 0.77 0.77 0.51
Boiling point saturates 0.062 0.46 0.021 0.29 aromatics 0.85 0.81 0.79 0.53
Aqueous solubility saturates 0.0019 0.36 0.084 0.19 aromatics 0.31 0.23 0.23 0.17
Log K., saturates 0.097 0.48 0.035 0.31 aromatics 0.85 0.76 0.78 0.49
Log Koc saturates 0. 097 0.48 0.79
0.035 0.31 aromatics 0.81 0.77 0.55
Vapour pressure saturates 0.046 0.096 0.13 0.096 aromatics 0.15 0.15 0.16 0.13
8.3.4. Patterns of Compounds
In terms of individual compounds, low molecular weight saturate compounds
had high losses in all the treatments examined. For example, pentadecane (C15)
had an average loss of 84.6% and eicosane (C20) had a relatively high removal of
195
67.5%. Larger saturate compounds had lower losses; for example, triacontane
(Cm) had losses of 57.8% and pentatriacontane (C35) had losses of 54.4%.
There was a more marked difference between 2/3-ringed aromatics and 5/6-
ringed ones. Losses of naphthalene (2-ringed) were generally high, on average
78.9%, while losses of phenanthrene (3-ringed) were also relatively high with an
average of 60%. In contrast, the benzo(a)pyrene (5-ringed) had an average loss of
35.6% and the dibenzo(ah)anthracene (6-ringed) had an average loss of 31.2%,
other 5 and 6 ring compounds had similar losses to these.
8.4. Discussion
Under optimized composting conditions no relationships were found between the
losses of saturate compounds and their physico-chemical properties. This may
have been due to saturate compounds being a readily and a preferentially
degradable fraction of oil (ATDSR 1999; Namkoong et al. 2002) with Cio to C24
alkanes being rapidly degraded (Englert et al. 1993). Therefore, these compounds
would easily have been removed despite physico-chemical properties, so there
would have been little or no discernible effect of compound properties on their
losses. Additionally under optimized conditions, there would have been large
numbers of microorganisms such as Gram-positive bacteria and actinomycetes
present (Chapter 9), involved in degradation, and the results suggest that the
chemical properties were not limiting on the effectiveness of these
microorganisms in removing contamination.
196
Additionally, under optimized composting, conditions would have been more
favourable in terms of moisture content, temperature and available carbon in
terms of soil to green waste ratio for microbial breakdown of saturates, thereby
overcoming any problems that physico-chemical properties might pose
(Oleszczuk and Baran 2003). The addition of green waste would have stimulated
microbial activity, both metabolism and co-metabolism (Jorgensen et al. 2000;
Antizar-Ladislao et al. 2004; Haderlein et al. 2006; Oleszczuk 2007),
temperature would have affected microbial activity and chemical reaction rates
(Liang et al. 2003; Antizar-Ladislao et al. 2004) and moisture content would
have supported microbial activity and movement (Finstein and Hogan 1993).
Therefore, saturates, including the larger compounds, would have been easily
removed. The presence of Gram-positive bacteria and actinomycetes (Chapter 9),
of which a number are hydrocarbon degraders (Hassen et al. 2001), shows that
there were microorganisms present during composting under optimized
conditions that were probably responsible for degradation during composting.
Stronger relationships were observed for the loss of aromatic compounds and
their physico-chemical properties. This may have been due to decreasing
degradability with increasing ring number and molecular weight (Lima et al.
2005; Lors and Mossmann 2005; Wammer and Peters 2005; Jonsson et al. 2007).
Low molecular weight aromatic compounds (e.g. naphthalene) are easily
biodegraded, but higher molecular aromatic compounds (e.g. perylene) are
relatively recalcitrant to biodegradation (ATSDR 1999; Lima et al. 2005; Lors
and Mossmann 2005). Therefore, they are not preferentially degraded
(Namkoong et al. 2002), resulting in lower losses of larger compounds. This is
197
because low molecular weight aromatics (2/3-ringed) have relatively high water
solubility, whereas high molecular weight aromatics (5/6-ringed) are relatively
hydrophobic with low aqueous solubility and high octanol—water partitioning
coefficients (Jonsson et al. 2007). Higher molecular weight compounds have
larger values for physico-chemical properties such as log K„,, log Koc, boiling
point and smaller values for aqueous solubility and vapour pressure, which
makes them less available for degradation and results in greater correlations of
loss with these properties.
Changing the composting operating parameters (increasing temperature, altering
soil to green waste ratio and decreasing moisture content) resulted in stronger
relationships between losses of saturates and their physico-chemical properties.
This may have been because composting conditions were less favourable for the
microbial community, resulting in less degradation of larger saturate compounds.
Therefore, the physico-chemical properties of these compounds would have been
more important factors affecting compound removal under these different
conditions. Changes in operating parameters may have made conditions
unfavourable by destroying some degraders (Potter et al. 1999), reducing
microbial activity (Liang et al. 2003; Cai et al. 2007) or denaturing enzymes
(ATSDR 1999) with increasing temperature. Decreasing the soil to green waste
ratio may have resulted in preferential degradation of the green waste rather than
most of the saturate compounds (Van Gestel et al. 2003; Ostberg et al. 2007).
Increasing the soil to green waste ratio may have inhibited the degradation
(Namkoong et al. 2002; Ostberg et al. 2007; Lee et al. 2008) due to less available
carbon in the form of green waste and a greater amount of contamination present
198
in the mix. Decreasing the moisture content to 40% may have resulted in
dehydration of the composting mix (Atagana 2004), resulting in less microbial
activity and therefore less degradation, while increasing it may have led to water-
logging and anaerobic conditions (Edwards 1998), which also reduces
degradation. The decrease in degradation of higher saturates would have resulted
in stronger relationships, as physico-chemical properties have more of an
influence on loss of saturates. Weaker relationships between loss of aromatic
compounds and their physico-chemical properties may have been due to lower
losses for aromatic compounds, including low molecular weight compounds such
as naphthalene, which are regarded as being more readily degradable than higher
molecular weight aromatics (Wammer and Peters 2005). As with the change in
relationship for saturate compounds, different operating parameters may have
made conditions less favourable for the degradation of aromatics by the
microbial community. Additionally, the use of a lower soil to green waste ratio
(0.7:1) would have provided a higher organic carbon content resulting in more
sorption of aromatic compounds, making them less available for degradation
(Jonsson et al. 2007), which would have become greater with increasing ring
number (Oleszczuk 2007). Lower removal of aromatics may also have been due
to the hydrophobicity of the aromatic compounds affecting their degradation, as
they are primarily oxidized in solution and not in the sorbed state (Jonsson et al.
2007).
Ageing of electrical oil-contaminated soil resulted in stronger relationships
between losses of saturated compounds and their physico-chemical relationships.
This may have been due to the effect of contaminant ageing decreasing the
199
chemical and biological availability of contaminants (Semple et al. 2003; Palomo
and Bhandari 2006; Jonsson et al. 2007), thereby resulting in lower removal of
contamination, compared to freshly added compounds, which would have been
more degradable (Northcott and Jones 2001; Semple et al. 2003). Therefore, the
increasing values of physico-chemical properties would result in larger saturated
compounds being harder to remove.
Weaker relationships were observed for aromatic compounds in aged soils,
which might have been due to lower overall removal of all the aromatic
compounds (5.7 to 10.8% lower in aged gault clay and 7.4 to 19.9% lower in
brick earth). Additionally, there may have been a more dominant role for
microbes relative to physico-chemical properties in aged soils despite lower
losses, which could have been due to adaptation of the microbial community
within the soil during the ageing period (Lee et al. 2003). Lower removal of
compounds could have been due to sequestration of the compounds during
ageing resulting in limited availability for degradation (Semple et al. 2003;
Pollard et al. 2008). Additionally, slow release of contaminant molecules into the
aqueous phase of soil limits the rate of contaminant biotransformation in aged
soils (Pollard et al. 2008). In freshly contaminated soils, aromatics would have
been more bioavailable than those in aged soils (Jonsson et al. 2007), so greater
removal of lower aromatics would have led to stronger relationships.
The relationships between the losses of saturate and aromatic compounds were
weaker in the brick earth compared to the gault clay. This may have been due to
less removal of the lower molecular weight compounds that are generally more
200
degradable; in freshly contaminated soils, losses from brick earth were 2.4 to
21.1% (saturate) and 10.2 to 17.8% (aromatic) lower than in gault clay. In aged
soils, they were 4.2 to 14.8% (saturate) and 6.7 to 25.8% (aromatic) lower than in
gault clay. Additionally, different soils have varying microenvironments that
support distinct microbial communities (Bundy et al. 2002; Scherr et al. 2007).
Therefore, in gault clay the microbial community may have had more degraders,
resulting in more degradation than in the brick earth, especially of the lower
molecular weight saturates and aromatics, resulting in stronger relationships.
In general, differences in average loss between shorter-chained and longer-
chained saturate, as well as 2/3-ringed aromatics and 5/6-ringed aromatics, are
due to the latter having higher physico-chemical property values for molecular
weight, boiling point, log Kow, log Ko, and lower values for aqueous solubility
and vapour reducing their bioavailability (Table 8.1). These values make these
compounds less available and more resistant to degradation (Huesemann et al.
2003), resulting in lower removal of these compounds and preferential
degradation of more available compounds. For example, molecular weight and
structure affect microbial attack, with decreasing degradability as molecular
weight increases (Leblond et al. 2001; Wammer and Peters 2005). Lower
aqueous solubilities, higher log Kow and log Km values of these compounds mean
that they are less soluble and therefore are less available to degradation (Bardi et
al. 2000; Huesemann et al. 2003) and more persistent due to sorption and
decreased mobility (Reid et al. 2000; Ferreira 2001; Verbruggen et al. 2000).
Low vapour pressures also mean that these compounds do not volatilize and be
lost abiotically (Dunnivant and Anders 2006; Hayes and Soni 2006).
201
8.5. Conclusions
The results suggest that it is the microorganisms present during composting that
control the loss of compounds to a greater extent than the physico-chemical
properties of the compounds. Additionally, the results show that under non-
optimal composting conditions and in aged-contaminated soils, the effect of
saturate compound physico-chemical properties is more pronounced, probably
due to conditions being unfavourable for degradation resulting in lower removal
of longer chain contaminants. The unfavourable conditions may also have
resulted in the lower loss of aromatic compounds, resulting in weaker
relationships between changes in property values and losses of compounds.
Patterns of greater loss from short chain and 2/3-ringed aromatics indicate that
physico-chemical properties affect the loss of longer chains and 5/6-ringed
aromatics more than the smaller compounds. Under optimized composting
conditions, saturate compounds were relatively easily removed and 2/3-ringed
and 4-ringed aromatics had relatively high removal, suggesting that optimal
composting conditions result in reduced effects of physico-chemical properties
and greater overall compound degradation. Changes in operating parameters,
contaminant ageing and differences in soil all affect the relationships between
compound physico-chemical properties and their losses, due to effects on
compound availability and the microbial community during composting.
202
Chapter 9. PLFA Analysis of Microbial Communities During
Composting Bioremediation
9.1. Introduction
This chapter examines the structure of the microbial community involved in the
composting bioremediation of electrical insulating oil contaminated soil. This
will assist in identifying and understanding which microbial groups were
important in the degradation of contaminants during the composting process. It
also provides insight as to how factors such as soil characteristics and
contaminant-ageing affect the composting microbial community.
Aerobic composting passes through four major microbiological phases:
mesophilic (30-45°C), thermophilic (45-75°C), cooling and maturation (Finstein
and Hogan 1993). Substantial changes occur in microbial populations and
species abundance during these stages. The greatest microbial diversity has been
observed in the mesophilic stage when mesophilic bacteria and fungi are
dominant. Spore-forming bacteria and thermophilic fungi characterize the
thermophilic stage (Tuomela et al. 2000). Recolonization by mesophilic fungi
and bacteria occurs during the cooling phase (Cole 1998).
Examples of hydrocarbon and PAH degraders that have been found in
composting bioremediation include Gram-negative bacteria such as Alcaligenes
spp., Pseudomonas spp. (Peressuttia et al. 2003; Antizar-Ladislao et al. 2004),
Flavobacteriurn spp. (Hassen of al. 2001), Mycobacterium spp., Rhodococcus
spp. and Gordonia spp.(Antizar-Ladislao et al. 2008). Gram-positive bacteria
203
(including actinomycetes) include Micrococcus spp., Planococcus spp.,
Staphylococcus spp. and Bacillus spp. (Hassen et al. 2001). Fungal degraders
include Bjekandera spp., Phanerochaete chrysosporium, Penicillium spp.,
Cunninghamella elegans (Antizar-Ladislao et al. 2004), Pleurotus spp. and
Trametes spp. (Anastasi et al. 2008).
Temperature is an important environmental variable in composting because it
affects physiological reaction rates and population dynamics (Antizar-Ladislao et
al. 2005b). This can be due to increases in the rates of reaction rates of enzymes
involved in biodegradation (Antizar-Ladislao et al. 2004), stimulating
hydrocarbon degraders and also enhancing contaminant availability through
increased solubility and mass transfer (Van Gestel et al. 2003). Additionally, co-
oxidation may be enhanced due to the range of alternative substrates present.
Addition of an organic component such as green waste to the contaminated soil
enhances the general microbial activity and the activity of specific degraders
(Jorgensen et al. 2000), as it supplement nutrients and carbon sources in
contaminated soil (Namkoong et al. 2002). Modifications in the physical and
chemical microenvironments can also serve to increase the diversity of
microflora (Antizar-Ladislao et al. 2004).
Lipids are a common component of microbial membranes; they include
phospholipids, glycolipids and other complex lipids. Most fatty acids are derived
from membrane phospholipids (Green and Scow 2000). Different
microorganisms have differing membrane lipid fatty acid composition, allowing
them to be distinguished by analysis as fatty acid methyl ester derivatives
204
(PLFA, phospholipids fatty acid) using GC-MS (Cavigelli et al. 1995; Keith-
Roach et al. 2002; Schelble et al. 2005). Gram-positive bacteria mainly have
fatty acid compositions dominated by iso- and anteiso-methyl branched-chain
fatty acids with an odd number of carbon atoms (commonly C15 and C17). Gram-
negative bacteria generally have monounsaturated fatty acids that tend to have an
even number of carbon atoms (C14-C20) (Schelble et al. 2005). Eukaryotic
membrane lipids contain polyunsaturated fatty acids (Green and Scow 2000),
which are rare in prokaryotes. Determining PLFA gives an indication of the
major microbial groups present (fungi, actinomycetes, and Gram-positive and
Gram-negative bacteria) and their relative proportions, i.e. population sizes of the
microbes (Klamer and Baath 1998). Because PLFA are rapidly degraded in the
environment once microbes die, PLFA analysis measures the current living
community (Carpenter-Boggs et al. 1998). Thus, analysis of PLFA can be used
to provide information to help elucidate what microbiological groups might be
important during composting bioremediation and provide insight as to the nature
of the microbiological degradation processes involved.
9.1.1. Hypothesis
During composting bioremediation different microbial groups are present in the
compost matrix, breaking down substrates and contaminants. Over the
composting period the population sizes of these groups will change, with
different groups becoming dominant over time, as they adapt to the changing
conditions and contamination within the composting matrix.
205
9.1.2. Aims and Objectives
The aims of the experiments were:
1) Identify major microbial groups present in composted contaminated soil
samples to understand which groups are degrading the contamination.
2) Determine the changes in the relative proportions of the microbial groups
over the composting period.
3) Investigate how the communities may be affected by different parameters,
including soil characteristics and contaminant ageing.
In order to achieve these aims, several objectives were set:
1) Extract PLFA from composted samples using a modified Bligh and Dyer
method and convert these into methyl esters (FAME) for GC-MS analysis.
2) Use an established GC-MS programme and develop a GC-MS SIM method
to identify and quantify fatty acid methyl esters of interest.
3) Analyze sample extracts containing FAME to identify microbial groups
present, and quantify markers using an internal standard to calculate ratios of
the groups present.
206
Experiment Samples
Optimization Optimized conditions
Freshly contaminated gault clay
Naturally aged gault clay
Artificially aged gault clay
Freshly contaminated brick earth
Naturally aged brick earth
Artificially aged brick earth
Ageing
9.2. Materials and Methods
The samples used for this experiment (Table 9.1) were obtained from the
optimized conditions composting experiment (composting conditions: 38°C,
0.8:1 soil to green waste ratio and 60% moisture content), which used a
transformer leak-contaminated soil and the contaminant ageing compost
experiment (artificially contaminated fresh and aged oil-contaminated soil, two
types of soil). Samples from these experiments were taken at each sampling
interval (0, 7, 21, 45, 54, 80 and 98 days). The samples were stored at -20°C and
prior to extraction the samples were defrosted overnight at room temperature.
Table 9.1. Composted samples used for PLFA analysis
Lipid was extracted from 2g sub-samples of the compost samples using a
modified Bligh and Dyer method (Antizar-Ladislao et al. 2008). The samples
were weighed in 30m1 acid-washed glass centrifuge tubes. To the weighed
samples a mixture of chloroform, methanol and water was added in a ratio of
1:2:0.8 (by volume). The samples were left to extract for 1 h, then centrifuged at
207
2500 rpm for 30 min in a bench centrifuge. The supernatant was removed to a
clean centrifuge tube and equal volumes of water and chloroform were added to
bring the ratio of chloroform, methanol and water to 1:1:0.9 (by volume),
resulting in a two-phase system. The lower phase was removed to a methylation
tube and brought to dryness under a stream of nitrogen in a sample concentrator.
The extracted lipid containing PLFA was then converted to fatty acid methyl
ester (FAME) derivatives by transmethylation, by adding 2 ml of 2.5% (v/v)
H2SO4 in dry methanol. The dry methanol was made by adding 100m1 of
methanol to lOg of anhydrous sodium sulphate in a stoppered flask. The samples
were heated at 80°C for 1 h in a heating block, checking for evaporation and
topping up with dry methanol if evaporation occurred. After heating, the samples
were allowed to cool, then 2m1 of 0.9% (w/v) sodium chloride was added to stop
the reaction. The FAME were extracted 3 times with 2m1 aliquots of light
petroleum spirit (b.p. 60-80°C). For each addition of petroleum spirit, 80% of the
top layer was removed to a clean tube for analysis. The extracts were combined
and dried under a nitrogen gas stream and stored at -20°C and re-dissolved in
fresh light petroleum spirit (b.p. 60-80°C) (2m1) prior to analysis.
9.2.1. GC-MS Analysis
The extracts were analyzed using a Hewlett Packard 6890 series GC system with
a 5973 series mass selective detector. The sample (1111) was injected using a
pulsed splitless injection mode with the inlet at 260°C, using helium as the
carrier gas, and the flow rate was maintained at lml/min by electronic
pneumatics control. The oven programme started at 40°C, held for 3min, ramped
208
to 150°C at 10°C/min, ramped to 230°C at 3°C/min and finally ramped to 300°C
at 30°C/min and held at 300°C for 5min. The mass spectrometer was operated in
selective ion monitoring mode (SIM).
The SIM method was developed by analyzing a FAME standard (Sigma Aldrich,
purity 99.‘:%) consisting of 26 fatty acid methyl esters and a methyl 12-
hydroxyoctadecanoate standard using the GC-MS programme described above in
scan mode. The compounds were identified, and both quantitative and qualitative
ions for each fatty acid were identified. An auto-SIM program was performed on
the output and 22 SIM windows were created for the identification of the
compounds.
To identify microbial groups of interest in the samples, several compounds were
used as markers. For Gram-negative bacteria 18:1 (11c) was used. Gram-positive
bacteria were identified using i15:0, a15:0, i16:0 and i17:0. Fungi and yeasts
were identified by 18:2 (9c, 12c). Actinomycetes were identified using 12-OH
18:0 (Schelble et al. 2005; Antizar-Ladislao et al. 2008). The names of the
compound of interest with their quantification and qualification ions are shown
in Table 9.2.
209
Table 9.2. FAME compounds of interest for microbial group identification
Compound Acronym Quantitative Ion Qualitative Ion
Me 13-methyltetradecanoate i15:0 74 213
Me 12-methyltetradecanoate al5:0 74 199
Me 14-methylpentadecanoate i16:0 74 270
Me 15-methylhexadecanoate i17:0 74 284
Me cis, cis-9,12- octadienoate 18:2 (9c, 12c) 81 294
Me cis-11-octadecanoatc 18:1 (11c) 55 264
Me 12-hydroxyoctadecanoate 12-OH 18:0 187 301
A chromatogram of the FAME standard in SIM mode analysis is shown in
Figure 9.1. It shows the peaks of all the FAME present, with compounds of
interest numbered. By way of example a chromatogram of a compost sample is
shown in Figure 9.2, showing the peaks of compounds present in samples.
Abundance
TIC: LW STANDARD SIM.D
260000 240000 220000 200000 180000 160000 140000 120000 100000 80000 60000 40000 20000
5
12.0014.0016.0018.0020.0022.0024.0026.0028.0030.0032 0034.0036.00 Time-->
Peaks: 1- i15:0; 2- a15:0; 3- i16:0; 4- i17:0; 5- 18:2(9c, 12c), 6- 18:1 (11c); and 7- 12-OH
Figure 9.1. Chromatogram of FAME standards showing peaks of compounds
used for identification
210
.41 unclanca
-r I-N/V2 3 007-1 2.0 1 4. 0 0 0 0
1 30000
1 2 0000
1 1 0000
1 00000
90000
50000
70000
50000
50000
.4-00 0 43
30000
20000
1 0000
s,DO 30 00 N
20.00 25.00 " " •
35.00 40.00 ekg
45.00 Ti rn
Figure 9.2. Chromatogram of extracted compost sample
To quantify the fatty acid methyl esters in the samples an internal standard of
nonadecanoic acid [19:0] (Sigma Aldrich, 99.5% purity) was used, because it
was not present in any of the samples. A range of internal standard
concentrations (2, 10, 20 and 50m/m1) were analyzed using GC-MS to identify
the optimal concentration to use in relation to analyte responses. The different
concentrations were made from a 1mg/m1 stock solution prepared using lmg of
solid standard dissolved in 1 ml of petroleum spirit (b.p. 60-80°C). The
appropriate volumes of this were diluted in more petroleum spirit to provide the
desired range of concentrations. These standards were analyzed using GC-MS
and their responses compared to those of the sample analytes. The most suitable
concentration of standard to use was found to be 50ug/ml. To quantify the
analytes of interest their relative response factors were calculated using the
internal standard response. The relative response factors were then used to
calculate the amounts of FAME in the samples.
211
The relative importance of the microbial groups and their changes over time were
expressed as ratios calculated using the amounts of each marker from the
compost samples. These ratios included: Gram-negative to Gram-positive
bacteria; total bacteria to fungi; actinomycetes to fungi; and Gram-negative
bacteria to actinomycetes.
9.3. Results of PLFA Analysis
The relative proportions and dynamics of the microbial groups present
throughout composting, in relation to the effects of soil type and contaminant
ageing on communities during composting are presented.
9.3.1. Optimized Composting Results
The ratio of Gram-negative to Gram-positive bacteria in optimal composting
conditions is shown in Figure 9.3. There was a general decrease in ratio
throughout the composting period, suggesting that Gram-negative bacteria may
be decreasing in number or that Gram-positive bacteria are increasing in number
during composting.
212
Gra
m -y
e/ G
ram
+ve
B
acte
ria
Rat
io
8-, 7 6 - 5 - 4 - 3 - 2 - 1
0
0 7 21 45 54 80 98
Time (days)
Figure 9.3. Ratio of Gram-negative to Gram-positive bacteria in optimized
composting condition samples
The Gram-positive actinomycetes are less abundant compared to Gram-negative
bacteria at the start of composting (Figure 9.4.), suggesting that they may be less
significant in the compost starting materials. The ratio drops 4-fold from the start
to 21 days, remains low during the middle of composting process, then increases
until 80 days, where it then levels out. This suggests that actinomycetes become
more prominent during composting, then decline towards the end of composting.
a) 1 0 - a) • 9 E ▪ 8 - o c 7- ;..7. "t1 o 6 - -a -.r., 5 - ..c cc a) W C.) 4- co co 3 -
(4') 2 -
g 1 -
6 0
7 21 45 54 80 98
Time (days)
Figure 9.4. Ratio of Gram-negative bacteria to actinomycetes in optimized
composting condition samples
213
2.5 -
2 - I
1.5 -
0.5 -
Bact
eria
l/ Fun
gal R
atio
I
0
During the optimized composting conditions, the total bacteria increase relative
to the fungi in composting (Figure 9.5.). There is a steady increase in ratio (day 0
to 45 days) that levels out towards the end of composting, with a small
fluctuation around 80 days. This indicates that either the total bacteria increase or
conversely, the proportion of fungi in the sample decreases.
0 7 21 45 54 80 98
Time (days)
Figure 9.5. Ratio of bacteria to fungi in optimized composting condition samples
The relative proportions of actinomycetes to fungi is shown in Figure 9.6. The
ratio increases from the start of composting by more than 4-fold from day 0 to 45
days, reaching a peak in the middle of the composting period, then halving by the
end of composting. This suggests that fungi are more prevalent at the beginning,
then actinomycetes increase in relative numbers before declining and fungi
becoming more dominant.
214
9-
8-
7- I
I
>, 3 - E g 2-
0
0 7 21 45 54 80 98
Time (days)
Figure 9.6. Ratio of actinomycetes to fungi in optimized composting condition
samples
9.3.2. Aged Treatments Composting Results
The general trend in the ageing experiment treatments was a decrease in the ratio
of Gram-negative relative to Gram-positive bacteria at the start of composting
that levelled off at the end of composting for all three ageing treatments (Figures
9.7 and 9.8). This suggests that Gram-positive bacteria are increasing in number
during the first half of composting and remaining important until the end of
composting, or that the Gram-negatives decrease in number during composting.
Contaminant ageing has an effect on the ratio of Gram-negative to Gram-positive
bacteria in gault clay during composting (Figure 9.7). At the start of composting
the aged-contaminated soils have relatively less Gram-positive bacteria than the
freshly contaminated soil compost. The naturally aged soil has a greater ratio,
than the artificially aged soil, which is higher than in the freshly contaminated
gault clay. This pattern generally continues through to 54 days, with values of the
215
ratio decreasing for all the treatments. After 54 days, the freshly contaminated
gault clay ratio increases to above that of the aged ratios. This suggests that the
aged soils have relatively fewer Gram-positive bacteria than the freshly
contaminated soils. In all the treatments the number of Gram-positive bacteria
increases relative to the Gram-negatives during composting period until 54 days.
Towards the end of the composting the Gram-positive bacteria decrease or the
Gram-negative bacteria start to increase again in freshly contaminated gault clay.
A similar pattern in the ratio of Gram-negative to Gram-positive bacteria occurs
in brick earth (Figure 9.8). At the start of composting the aged contaminated soils
have relatively fewer Gram-positive bacteria than the freshly contaminated soil.
This pattern generally continues through to 54 days, with the ratio values
decreasing for all the treatments. After 54 days the freshly contaminated brick
earth and naturally aged ratio increases to above that of the artificially aged
ratios. This suggests that the aged soils have relatively fewer Gram-positive
bacteria or more Gram-negative bacteria than the freshly contaminated soils. In
all the treatments, the relative number of Gram-positive bacteria increases during
composting period until 54 days. Towards the end of composting, the Gram-
positive bacteria decrease or the Gram-negative bacteria start to increase again in
the freshly contaminated and artificially aged brick earth.
Brick earth tends to have a slightly higher ratio of Gram-negative to Gram-
positive bacteria than gault clay in all the ageing treatments (Figures 9.7 and 9.8),
although the difference is generally not large. This suggests that brick earth
216
generally has relatively less Gram-negative bacteria or more Gram-positive
bacteria than the gault clay. G
ram
-vet
Gra
m +
ve B
acte
ria
Rat
io 12
10
8
6
4
2
0
• Gault Fresh 0 Gault Naturally Aged
El Gault Artificially Aged
0 7 21 45 54 80 98
Time (days)
(Fresh: Freshly contaminated soils)
Figure 9.7. Ratio of Gram-negative to Gram-positive bacteria in gault clay with
different contaminant ageing treatments
12 -
10 -
8 - 0 Brick Fresh 0
6 - 0 Brick Naturally Aged Brick Artificially Aged
4 - 11
2 - I is
•=Ci
II '44 0
0 7 21 45 54 80 98
Time (days)
Figure 9.8. Ratio of Gram-negative to Gram-positive bacteria in brick earth with
different contaminant ageing treatments
217
Gra
m -y
e/ G
ram
+ve
Bac
teri
a
as
• 7
E 47.
5 o
2 w ▪ fa 4
3 ca • 2
ns E
0 7 21 45 54 80 98
a) a)
E
c r.
o 473.
"a3 (a
E
8 -
7 -
6 -
5 -
4-
3-
2 -
1-
0
o Brick Fresh o Brick Naturally Aged
Brick Artificially Aged
r
The numbers of actinomycetes relative to Gram-negative bacteria is shown in
Figures 9.9 and 9.10. For both the gault clay and brick earth the ratios are quite
high at the beginning of composting, then decrease in the middle of composting,
finally increasing until the end of composting. This suggests that that the same
interaction between these groups is happening as in the optimized conditions.
• Gault Fresh E1 Gault Naturally Aged o Gault Artificially Aged
Time (days)
Figure 9.9. Ratio of Gram-negative bacteria to actinomycetes in gault clay
composted treatments
7 21 45 54 80 98
Time (days)
Figure 9.10. Ratio of Gram-negative bacteria to actinomycetes in brick earth
composted treatments
218
The ratio of bacteria to fungi (Figure 9.11 and 9.12) generally increases in all the
treatments. After a period of little change at the start of composting (day 0 to 21
days), the ratio increases quite rapidly (day 45 to day 80, from 2.83 to 4.81), then
remains stable to the end of composting. This suggests that during the
composting process the relative number of bacteria increases substantially or
those of the fungi decline and remain at that level until the end of composting.
As with the Gram-negative to Gram-positive bacteria ratio, the ratio of bacteria
to fungi is generally higher in brick earth than gault clay (Figures 9.11 to 9.12.),
suggesting that brick earth has relatively more bacteria than gault clay. At the
beginning of composting freshly contaminated gault clay has the highest bacteria
to fungi ratio, then naturally aged and finally artificially aged (Figure 9.11). By
21 days the naturally aged ratio increases above that of the fresh, until 80 days
when the ratio of the fresh increases above it again. The artificially aged ratio
increases above fresh and naturally aged at 21 days, and this generally remains
above the values of the other treatments. This suggests that, at the start of
composting, the freshly contaminated gault clay has less fungi or more bacteria,
than naturally aged or artificially aged. The naturally aged samples relative
decrease in bacteria at a greater rate than the freshly contaminated samples, until
80 days when the relative number of fungi in the fresh decreases above that of
the natural. In the artificially aged the relative number of fungi increases at a
greater rate than the rest of the treatments, indicating that ageing has an effect on
the make-up of the microbial community.
219
• Gault Fresh Gault Naturally Aged Gault Artificially Aged
0 7 21 45 54 80 98
Time (days)
Figure 9.11. Ratio of bacteria to fungi in gault clay with different ageing
treatments
At the start of composting in brick earth there is little difference in the ratio in the
ageing treatments, suggesting that the numbers of fungi and bacteria in each are
similar (Figure 9.12.). At 45 days, the ratios increase in the treatments, with
naturally aged brick earth increasing slightly more than freshly contaminated and
much more than artificially aged. However, fresh continues to increase more than
the naturally aged as does the artificially aged. At the end of composting the
fresh and artificially aged contaminated brick earth ratios level off, whereas the
naturally aged increases steadily until the end of composting. This suggests that
from 45 days, the relative amount of fungi decreases in all treatments with the
freshly contaminated soil having the biggest change, followed by the naturally
aged and artificially aged, with the fresh and artificial reaching their maximum
around 80 days, whereas the ratio in naturally aged climbs steadily from 45 to 98
days. This indicates that ageing has an effect on microbial community structure.
220
7
6 0
Ce 5
0) 4
t 3 co 2
1
0 Brick Fresh
❑ Brick Naturally Aged
Brick Artificially Aged
0 7 21 45 54 80 98
Time (days)
Figure 9.12. Ratio of bacteria to fungi in brick earth with different ageing
treatments
There was a general increase in the ratio of actinomycetes to fungi in all of the
ageing treatments (Figures 9.13 and 9.14), suggesting that over the composting
period the relative numbers of actinomycetes increased compared to that of
fungi.
• Gault Fresh
o Gault Naturally Aged
in Gault Artificially Aged
0 7 21 45 54 80 98
Time (days)
Figure 9.13. Ratio of actinomycetes to fungi in fresh- and aged-contamination
gault clay compost samples
221
❑ Brick Fresh o Brick Naturally Aged
Brick Artificially Aged
0 7 21 45 54 80 98
Time (days)
Figure 9.14. Ratio of actinomycetes to fungi in fresh- and aged-contamination
brick earth compost samples
9.5. Discussion
The increase in relative numbers of Gram-positive bacteria during the
composting period in both the optimized conditions and with the different ageing
treatments could be due to the ability of Gram-positive bacteria, with their broad
range of metabolic capabilities, to use a wider range of, and structurally more
complex hydrocarbons than Gram-negative bacteria are capable of using
(Peressuttia et al. 2003). This would give them a competitive advantage over
Gram-negative bacteria in the composting of electrical insulating oil-
contaminated soil. Additionally, during the general organic material composting
processes the numbers of Gram-positive bacteria increases over the Gram-
negatives (Hellmann and Shann 1997).
The number of actinomycetes increased during composting compared to Gram-
negative bacteria, but then declined towards the end of composting in both the
0
cc
E 0
4 -
3.5 -
3-
2.5 -
2-
1.5 -
1-
0.5 -
0
222
optimized and ageing treatments. This could have been due to the slower growth
rate of actinomycetes compared to other bacteria (Ma et al. 2003), so they would
have taken longer to proliferate and establish themselves in the composting mix.
Additionally, they are poor competitors at high nutrient levels, so at the start of
composting when available nutrients were high and more available (Tuomela et
al. 2000) they would not have been able to compete as well with other bacteria
and fungi present. However, as nutrients were depleted they would have
increased in number, as they can utilize the more complex substrates that remain
(Steger et al. 2007). This would also explain the increase in numbers of
actinomycetes relative to fungi in the optimized composting and ageing treatment
compost, which has also been observed by Klamer and Baath (1998).
The relative decrease in actinomycetes numbers at the end of composting in the
optimized composting compared to Gram-negative bacteria and fungi suggests
that these groups have some advantage over the actinomycetes by the end of
composting, or that the numbers of actinomycetes naturally decreased.
In the optimized compost and the ageing treatment composts, the number of
bacteria increases over that of the fungi. At the start of composting the ratio of
fungi to bacteria was high (0.71), which can be attributable to the introduction of
fungi into the composting mixture from the green waste fraction. The decline in
the amount of fungi could be due to the greater ability of bacteria to grow on a
wider range of hydrocarbons compared to fungi, and that fungi can be inhibited
by some hydrocarbons (Hughes et al. 2007), leading to a decline in fungal
numbers in the electrical insulating oil-contaminated compost mix. Additionally,
223
there may be competition between the fungi introduced from the green waste and
microorganisms indigenous to the contaminated soil (Andersson et al. 2000).
Contaminant ageing seems to lead to higher ratios of Gram-positive to Gram-
negative bacteria, as well as bacteria to fungi, in both soil types of the aged soils
than freshly contaminated soils at the start of composting. However, towards the
end of composting the ratios in the freshly contaminated soils exceeded those of
the aged contaminated soils. The larger ratios in the aged soils may be due to the
hydrocarbon pollution influence on microbial populations, leading to selection of
microorganisms and acclimatization to contamination (Scherr et al. 2007). Thus,
in the soils that had been aged the microorganisms will have become
acclimatized to contamination during the ageing process (Trindade et al. 2005).
Therefore, they would have an advantage over microorganisms in freshly
contaminated soils, which would not be acclimatized, therefore, the populations
would be larger in aged soils.
The greater ratios of Gram-negative to Gram-positive bacteria, and bacteria to
fungi values found in the brick earth soils compared to the gault clay may have
been due to the brick earth having a higher organic matter content than the gault
clay (4.8% and 3.8% respectively). This difference in organic matter content may
have resulted in differing community structure and density in the soils, since high
organic matter may stimulate microbial population size and activity (Labud et al.
2007). Additionally, high organic matter allows communities to recover more
quickly from environmental and contamination stresses (Scherr et al. 2007).
224
The relationship between changes in microbial group ratio and loss of TPH
contamination during composting are represented in Figures 9.15 to 9.17, which
show the percentage TPH lost during composting with the changes in microbial
group ratios. During composting under optimized conditions (Figure 9.15) there
is a sharp decrease in the ratio of Gram-negative to Gram-positive bacteria at the
start of composting (day 0 to 21), as the removal of TPH contamination starts;
the decrease in ratio then levels out, decreasing only slightly until the end of
composting. This suggests that at the start of composting the Gram-positive
bacteria are more adapted to removing TPH and therefore increase in number or
that the Gram-negative bacteria decrease in number as they are less adapted to
contamination. The ratio of bacteria to fungi increases at the beginning of
composting until 54 days, in parallel with the loss of TPH; the ratio then levels
off as the loss of TPH slows. This pattern suggests that bacteria play more of a
role in TPH degradation than fungi, as the total bacteria increases over the fungi,
or that the fungi are less adapted to TPH contamination and, therefore, decrease
in number.
225
- 100
- 90
- 80
- 70
- 60
- 50
- 40
- 30
- 20
- 10
7 —
6 —
5 —
0 4 —
cL▪ • 3 —
2 —
1 —
TPH
rem
oved
(%)
= TPH removed
—421-- Gram -ve/Gram +ve bacteria ratio
—4— BacteriaVfungal ratio
0
0
I I
7 21 45 54 80 98
Time (days)
Figure 9.15. Changes in microbial group ratios and loss of TPH during
composting under optimized conditions
The relationships between loss of TPH for the aged contamination composting
experiment are shown in Figures 9.16 and 9.17, which illustrate the average TPH
loss and microbial group ratios for all the ageing treatments (freshly
contaminated, naturally and artificially aged). The pattern of the Gram-negative
to Gram-positive bacteria ratio is similar in gault clay and brick earth (Figures
9.16 and 9.17). There is an initial sharp decrease (0-45 days), as TPH loss
increases, and there is a small increase in ratio at the end of the composting
period. This suggests that Gram-positive bacteria play more of a role than Gram-
negative bacteria in the degradation of TPH, as in the optimized composting. At
the start of composting the ratio of bacteria to fungi in the gault clay and brick
earth samples was low, as was the TPH loss; the ratio increased after day 45,
when overall TPH loss was reaching its maximum. This again suggests that
bacteria play a greater role in the removal of TPH contamination than fungi.
226
o 8 COCO • 6 —
14 —
12 —
10 —
cM,
- 100 - 90 - 80 - 70 - 60
- 50 - 40 - 30 - 20
- 10 0
ist.2
0
a. I-
O TPH removed
Gram -ve/Gram +ve bacteria ratio
—•— Bacterial/fungal ratio
4 —
2 — •
0 I I 1 I 1 I
14 —
12 —
10-
O 8— tr. co CL 6 —
4 — —
2—
0 1 0 7
- 100 90
- 80 70 60 50 40 30 20 10 0
E===1TPH removed
TP
H lo
st (%
)
- Gram -ve/Gram +ve bacteria ratio
—4,— Bacterial/fungal ratio
1
I I I I 21 45 54 80 98
Time (days)
Figure 9.16. Changes in microbial group ratios and loss of TPH during
composting in composted oil-contaminated gault clay
0 7 21 45 54 80 98
Time (days)
Figure 9.17. Changes in microbial group ratios and loss of TPH during
composting in composted oil-contaminated brick earth
227
9.6. Conclusions
During the composting of electrical insulating oil-contaminated soil, overall,
bacteria appear to be the dominant group involved in degradation during the
process, with Gram-positive bacteria, including actinomycetes, being more
important. During composting Gram-positive bacteria increase in number over
the Gram-negative bacteria. Although fungi are abundant at the beginning of
composting, their numbers decrease as the bacteria proliferate in the compost
matrix. The type of soil also has an effect on the microbial community, quantity
and its activity, depending on its organic matter content. Additionally,
contaminant ageing has an effect on community structure and quantity of
organisms, probably due to the toxic effect of oil and the ability of some
organisms to adapt to contamination.
However, because PLFA analysis does not show that the lipids found are actually
from the degraders, other techniques could be used to build on the results
presented here. For example organisms could be isolated from the compost
samples by culturing on media containing hydrocarbons such as hexadecane to
identify the organisms degrading the contamination. Or DNA/RNA methods
could be used to identify the microorganisms present, sequence analysis of rDNA
fragments can be used to phylogenetically identify corresponding organisms and
DNA sequence information used to identify an organism can be used to design
phylotype-specific polymerase chain reaction (PCR) primers for determining the
abundance of organisms. Also, new methods of screening for mRNA of the key
enzymes involved in e.g. hydrocarbon degradation can be used (Ritchie et al.
2000; Thomassin-Lacroix et al. 2001).
228
Chapter 10. Conclusions
This study examined a range of aspects relating to the clean-up of electrical
insulating oil-contaminated soils using in-vessel composting bioremediation,
including optimization of the composting operating parameters, the effects of
different soil properties, contaminant ageing and compound properties, as well as
examining the microbial communities involved in the composting process.
A robust chemical analytical method was developed for the identification and
quantification of electrical insulating oil-contaminants in soil and compost
samples, in order to measure TPH contamination. This method consisted of an
efficient extraction method using accelerated solvent extraction (ASE®)
combined with consistent and precise compound analysis using GC-MS in SIM
mode. The method gave high analyte recoveries, was consistent and repeatable,
producing reliable and precise results for the analysis of soil and compost
samples, which are inherently heterogeneous. This method can be reliably used
in future studies for the identification and quantification of TPH contaminants.
Optimizing the in-vessel composting conditions used for treatment of oil-
contaminated soil was achieved at bench-scale. As was hypothesized,
composting of electrical insulating oil-contaminated soil with green waste
resulted in removal of oil contamination. The optimal conditions for the removal
of contamination in this study were a soil to green waste ratio of 0.8:1, an
operating temperature of 38°C and a moisture content of 60%, which resulted in
75.6 to 86.1% removal of TPH. During the composting optimization
experiments, the use of different levels of the operating parameters, lower or
229
higher soil to green waste ratios (0.7:1 and 0.9:1), different moisture contents (40
and 80%) and higher operating temperatures (55 and 70°C), resulted in lower
losses of TPH than under the optimized conditions. Using different operating
parameters resulted in differing conditions within the composting matrix that
resulted in lower removal of oil-TPH contamination. These lower losses could
have been due to inhibition of microbial degradation and reduced microbial
diversity leading to unfavourable conditions such as high temperatures, amount
of contamination and availability/over-availability of water. Although a high
amount of contaminant removal was achieved under optimized conditions, the
high levels of residual contamination remaining meant that the end product of the
process was unsuitable for re-use. Therefore, future work should focus on further
optimization of the composting conditions in order to achieve higher removal of
contamination to allow for product use. However, if further optimization did not
result in significantly improved TPH removal compared to this study, then
mixing composted contaminated soil with 'clean' soil (i.e. uncontaminated soil)
or with green waste compost until an acceptable level of contamination was
achieved allowing for the re-use of the end product. Future studies should
examine the addition of nutrients such as nitrogen fertilizers, use of different
compost materials, e.g. foodwaste mixtures or sewage sludge, longer composting
treatment periods and the possibility of using a final polishing step to remove
residual contamination. After further optimization has achieved sufficient TPH
removal, scaling-up experiments should be used to test the efficiency of the
optimized conditions at a larger commercial scale.
230
Despite investigating two very different soils, it was found that soil type does not
affect the loss of contamination during in-vessel composting. However, this
could have been due to the nature and amount of the contamination itself, or the
optimized composting conditions negating any effects of the soil properties.
However, longer composting periods may result in more pronounced soil
property effects, which could be investigated in future work. Overall, this study
showed that the type of soil that is contaminated will not affect the removal of
contamination during composting, so the same process can be used for all types
of soil without any operational adjustments, making it a general clean-up
method.
Ageing of oil contamination for 3 months did not result in a significant
difference (p>0.05) between the loss of TPH contamination from freshly
contaminated soils compared to aged contaminated soils during composting.
Therefore, up to 3 months ageing does not result in more persistent
contamination and does not affect the removal of contamination during
composting. However, as this may be a relatively short ageing time for
contaminated soils, longer ageing periods should be examined, e.g. 12 months, to
determine if this results in more persistent contamination resulting in lower
losses during in-vessel composting bioremediation. Additionally, ageing soils
outdoors at different times of the year may result in more pronounced effects of
sequestered contaminants, due to varying environmental conditions.
There were relationships between loss of individual saturate and aromatic
compounds and their physico-chemical properties. Larger more hydrophobic
231
compounds, especially among the aromatic compounds, had lower removal than
smaller more soluble compounds, with losses of shorter chain saturates being up
to 26.8% higher than longer chains, and 2/3-ringed aromatics having losses up to
46.7% higher than 5/6-ringed aromatics. The strength of the relationship between
changing physico-chemical properties and losses of compounds changed under
non-optimal conditions and with contaminant ageing. Relationships between
compound loss and physico-chemical properties for saturate compounds were
stronger but weaker for aromatic, which could be due to unfavourable conditions
making it harder for microbial degradation of compounds.
The microbial community present during composting bioremediation, which was
potentially responsible for contaminant degradation, was dominated by bacteria,
with Gram-positive bacteria and actinomycetes being important. Gram-positive
bacteria tended to increase in number over Gram-negative bacteria during
composting, and although fungi were abundant at the start of composting their
numbers declined as bacterial numbers increased. In order to get a more accurate
idea of hydrocarbon degraders present during composting, future work could
isolate hydrocarbon degraders on a hydrocarbon-containing medium, e.g.
hexadecane, to identify hydrocarbon-degrading groups present in composted
samples. To identify individual microorganisms, DNA/RNA methods could be
used to could identify hydrocarbon degrading organisms, as well as identifying
the mRNA of the key enzymes involved in hydrocarbon degradation (Ritchie et
al, 2000; Thomassin-Lacroix et al, 2001).
232
This study has demonstrated the scientific basis for the use of in-vessel
composting bioremediation, which could be used for the industrial problem of
electrical insulating oil contaminated soils. If the composting process established
in this study were used at a commercial scale, then potentially a site that could
treat 200 tonnes of compost at a time (i.e. at a soil to green waste ratio of 0.8:1,
which is 89 tonnes of contaminated soil mixed with 111 tonnes of green waste),
with a treatment length of 3months, could treat 356 tonnes of contaminated soil a
year (800 tonnes of compost). If around 4000 tonnes of soil were contaminated
with naphthenic oil in a year 11 sites within the UK would be needed to treat the
soil. As there are around eight large electrical transmission companies in the UK,
including EDF Energy, e.on (Central Networks), Scottish Power and Western
Power Distribution, from a practical viewpoint each company could manage one
large or two smaller composting bioremediation facilities depending on the size
of their distribution area and the region they operate in.
In-vessel composting bioremediation is a viable treatment method for the clean-
up of electrical insulating oil-contaminated soil that has the potential to be used
on a commercial scale by electrical transmission companies as an alternative to
landfilling contaminated soils. It is a cheaper alternative that is more
environmentally friendly and treats the waste rather than burying it. It also
potentially solves the problem of two waste streams produced by electrical
transmission companies (contaminated soils and green waste from vegetation
management), reducing the overall volume of waste and providing a useful end
product. Additionally, as soil properties and contaminant ageing do not affect
TPH removal, the optimal composting process and operating parameters
233
identified in this study can be used for a range of soils that may or may not have
been freshly contaminated. Further optimization of the process to further clean-
up the soils and successful scaling-up of the process will result in a commercial
application for the clean-up of soils.
234
References
Admon, S., Green, M. and Avnimelech, Y. (2001) Biodegradation kinetics of
hydrocarbons in soil during land treatment of oily sludge. Bioremediation
Journal 5, 193-209.
Aichberger, H., Loibner, A.P., Celis, R., Braun, R., Ottner, F. and Rost, H.
(2006) Assessment of factors governing biodegradability of PAHs in three soils
aged under field conditions. Soil and Sediment Contamination 15, 73 - 85.
Alexander, M. (1994) Biodegradation and Bioremediation. London: Academic
Press.
Alexander, M. (2000) Aging, bioavailability, and overestimation of risk from
environmental pollutants. Environmental Science and Technology 34, 4259-
4265.
Alexander, M. and Scow, K.M. (1989) Kinetics of biodegradation in soil In
Reaction and Movement of Organic Chemicals in Soil eds. Sawhney, B.L. and
Brown, K. pp.243-269. Madison, Wisconsin: Soil Science Society of America.
Allan, I.J., Semple, K.T., Hare, R. and Reid, B.J. (2006) Prediction of mono- and
polycyclic aromatic hydrocarbon degradation in spiked soils using cyclodextrin
extraction. Environmental Pollution 144, 562-571.
235
Allard, S.-A. and Neilson, A.H. (1997) Bioremediation of organic waste sites: A
critical review of microbiological aspects. International Biodeterioration and
Biodegradation 39, 253-285.
Amellal, N., Portal, J.-M. and Berthelin, J. (2001) Effect of soil structure on the
bioavailability of polycyclic aromatic hydrocarbons within aggregates of a
contaminated soil. Applied Geochemistry 16 1611-1619.
Amir, S., Hafidi, M., Merlina, G., Hamdi, H. and Revel, J.C. (2005) Fate of
polycyclic aromatic hydrocarbons during composting of lagooning sewage
sludge. Chemosphere 58, 449-458.
Anastasi, A., Varese, G.C., Bosco, F., Chimirri, F. and Archisio, V. (2008)
Bioremediation potential of basidiomycetes isolated from compost. Bioresource
Technology 99, 6628-6630.
Andersson, B.E., Welinder, L., Olsson, P.A., S., 0. and Henrysson, T. (2000)
Growth of inoculated white-rot fungi and their interactions with the bacterial
community in soil contaminated with polycyclic aromatic hydrocarbons, as
measured by phospholipid fatty acids. Bioresource Technology 73, 29-36.
Antizar-Ladislao, B., Beck, A.J., Spanova, K., Lopez-Real, J. and Russell, N.J.
(2007) The influence of different temperature programmes on the bioremediation
of polycyclic aromatic hydrocarbons (PAHs) in a coal-tar contaminated soil by
in-vessel composting. Journal of Hazardous Materials 144, 340-347.
236
Antizar-Ladislao, B., Lopez-Real, J. and Beck, A.J. (2004) Bioremediation of
polycyclic aromatic hydrocarbon (PAH)-contaminated waste using composting
approaches. Critical Reviews in Environmental Science and Technology 34, 349-
289.
Antizar-Ladislao, B., Lopez-Real, J. and Beck, A.J. (2005a) Laboratory studies
of the remediation of polycyclic aromatic hydrocarbon contaminated soil by in-
vessel composting. Waste Management 25, 281-289.
Antizar-Ladislao, B., Lopez-Real, J. and Beck, A.J. (2005b) In-vessel
composting-bioremediation of aged coal tar soil: effect of temperature and
soil/green waste amendment ratio. Environment International 31, 173-178.
Antizar-Ladislao, B., Lopez-Real, J. and Beck, A.J. (2006) Bioremediation of
polycyclic aromatic hydrocarbons (PAH) in an aged coal-tar-contaminated soil
using different in-vessel composting approaches. Journal of Hazardous
Materials B137, 1583-1588.
Antizar-Ladislao, B., Spanova, K., Beck, A.J. and Russell, N.J. (2008) Microbial
community structure changes during bioremediation of PAHs in an aged coal-tar
contaminated soil by in-vessel composting. International Biodeterioration and
Biodegradation 61, 357-364.
Atagana, H.I. (2004) Co-composting of PAH-contaminated soil with poultry
manure. Letters in Applied Microbiology 39, 163-168.
237
Atlas, R.M. (1995) Bioremediation of petroleum pollutants. International
Biodeterioration and Biodegradation 35, 317-327.
ATSDR (1999) Toxicological Profile for Total Petroleum Hydrocarbons (TPH).
Atlanta: ATSDR. Available from: www.atsdr.cdc.gov/HAC/HAGM/appc.html.
Accessed: 7/6/05.
ATSDR. (2001) Appendix C. Data Evaluation Criteria. Available from:
www.atsdr.cdc.gov/HAC/HAGM/appc.html. Accessed: 7/06/05.
Baheri, H. and Meysami, P. (2002) Feasibility of fungi bioaugmentation in
composting a flare pit soil. Journal of Hazardous Materials B89, 279-286.
Baker, J.R., Mihelcic, J.R. and Shea, E. (2000) Estimating K. for persistent
organic pollutants: limitations of correlations with K.W. Chemosphere 41, 813-
817.
Balba, M.T., Al-Awadhi, N. and Al-Daher, R. (1998) Bioremediation of oil-
contaminated soil: microbiological methods for feasibility assessment and field
evaluation. Journal of Microbiological Methods 32, 155-164.
Ballschmiter, K. (1996) Persistent, ecotoxic, and bioaccumulative compounds
and their possible environmental effects. Pure and Applied Chemistry 68, 1771-
1780.
238
Bardi, L., Mattei, A., Steffan, S. and Marzona, M. (2000) Hydrocarbon
degradation by a soil microbial population with 13-cyclodextrin as surfactant to
enhance bioavailability. Enzyme and Microbial Technology 27, 709-713.
Beck, A.J., Johnson, D.L. and Jones, K.C. (1996) The form and bioavailability of
non-ionic organic chemicals in sewage sludge-amended agricultural soils. The
Science of the Total Environment 185, 125-149.
Beck, A.J. and Jones, K.C. (1995a) Kinetic constraints on the in-situ remediation
of soils contaminated with organic chemicals. Environmental Science and
Pollution Research 2, 244-252.
Beck, A.J. and Jones, K.C. (1995b) Limitations to the in-situ remediation of soils
contaminated with organic chemicals in relation to the potential to achieve clean-
up criteria. In Contaminated Soil eds. van den Brink, W.J., Bosman, R. and
Arendt, F. pp.327-336. Netherlands: Kluwer Academic Publishers.
Beyer, A., Wania, F., Gouin, T., Mackay, D. and Matthies, M. (2002) Selecting
internally consistent physicochemical properties of organic compounds.
Environmental Toxicology and Chemistry 21, 941-953.
Bogan, B.W. and Sullivan, W.R. (2003) Physicochemical soil parameters
affecting sequestration and mycobacterial biodegradation of polycyclic aromatic
hydrocarbons in soil. Chemosphere 52, 1717-1726.
239
Bundy, J.G., Paton, G.I. and Campbell, C.D. (2002) Microbial communities in
different soil types do not converge after diesel contamination. Journal of
Applied Microbiology 92, 276-288.
Burnham, C.P. (1994) Soils. In The Natural History of a Country Estate: Wye
College, Kent eds. Watt, T.A. and Chapman, G.P. pp.53-70. Wye: Wye College
Press.
Cai, Q.-Y., Mo, C.H., Wu, Q.-T., Zeng, Q.-Y., Katsoyiannis, A. and Ferard, J.-F.
(2007) Bioremediation of polycyclic aromatic hydrocarbons (PAHs)-
contaminated sewage sludge by different composting processes. Journal of
Hazardous Materials 142, 535-542.
Canet, R., Birnstingl, J.G., Malcolm, D.G., Lopez-Real, J.M. and Beck, A.J.
(2001) Biodegradation of polycyclic aromatic hydrocarbons (PAHs) by native
microflora and combinations of white-rot fungi in a coal tar contaminated soil.
Bioresource Technology 76, 113-117.
Canet, R., Lopez-Real, J.M. and Beck, A.J. (1999) Overview of polycyclic
aromatic hydrocarbon biodegradation by white-rot fungi. Land Contamination
and Reclamation 7, 191-197.
Capelli, S., Busalmen, J.P. and de Sanchez, S.R. (2001) Hydrocarbon
bioremediation of a mineral-base contaminated waste from crude oil extraction
240
by indigenous bacteria. International Biodeterioration and Biodegradation 47,
233-238.
Carpenter-Boggs, L., Kennedy, A.C. and Reganold, J.P. (1998) Use of
phospholipid fatty acids and carbon source utilization patterns to track microbial
community succession in developing compost. Applied and Environmental
Microbiology 64, 4062-4064.
Cavigelli, M.A., Robertson, G.P. and Klug, M.J. (1995) Fatty acid methyl ester
(FAME) profiles as measures of soil microbial community structure. Plant and
Soil 170, 99-113.
CBS (2005) Bioremediation of Electrical Insulating Oils: Pilot Composting
Studies July 2004 - April 2005: Crown BioSystems
Cerniglia, C. (1992) Biodegradation of polycyclic aromatic hydrocarbons.
Biodegradation 3, 351-368.
Chaillan, F., Le Fleche, A., Bury, E., Phantavong, Y., Grimont, P., Saliot, A. and
Oudot, J. (2004) Identification and biodegradation potential of tropical aerobic
hydrocarbon-degrading microorganisms. Research in Microbiology 155, 587-
595.
241
Chemistry, Oxford University. (2007) The Physical and Theoretical Chemistry
Laboratory, Oxford University, Chemical and Other Safety Information.
Available from: http://physchem.ox.ac.uk/msds/. Accessed: 12/6/08.
ChemSpider. (2008) Database of Chemical Structures and Property Predictions.
Available from: http://www.chemspider.com/. Accessed: 9/7/08.
Chen, S.-C. and Liao, C.-M. (2006) Health risk assessment on human exposed to
environmental polycyclic aromatic hydrocarbons pollution sources. Science of
the Total Environment 366, 112-123.
Chung, N. and Alexander, M. (1999) Effect of concentration on sequestration
and bioavailability of two polycyclic aromatic hydrocarbons. Environmental
Science and Technology 33, 3605-3608.
Chung, N. and Alexander, M. (2002) Effect of soil properties on bioavailability
and extractability of phenanthrene and atrazine sequestered in soil. Chemosphere
48, 109-115.
Civilini, M. (1994) Fate of creosote compounds during composting.
Microbiology Europe 2, 16-24.
Clean Merseyside Centre. (2005) Guide to Selecting an In-Vessel Composting
System. Available from: wwww.Envirolinknorthwest.co.uk/Envirolink/Library/.
Accessed: 6/2/07.
242
Cole, M.A. (1998) An Analysis of Composting as an Environmental Remediation
Technology: United States Environmental Protection Agency.
Coulon, F., Pelletier, E., Gourhant, L. and Delille, D. (2005) Effects of nutrient
and temperature on degradation of petroleum hydrocarbons in contaminated sub-
Antarctic soil. Chemosphere 58, 1439- 1448.
Crosby, N.T., Day, J.A., Hardcastle, W.A., Holcombe, D.G. and Treble, R.D.
(1995) Quality in the analytical chemistry laboratory. Chichester: John Wiley
and Sons.
Cyberlipid. (2003) Automatic Soxhlet Extraction. Available from: www.
Cyberlipid.org/extract/extr0010.htm: Accessed: 5/6/06.
Daoust, C.M., Bastien, C. and Deschenes, L. (2006) Influence of soil properties
and ageing on the toxicity of copper on compost worm and barley. Journal of
Environmental Quality 35, 558-567.
Davis, J.W. and Madsen, S. (1996) Factors affecting the biodegradation of
toluene in soil. Chemosphere 33, 107-1130.
de Jonge, H., Freijer, J.I., Verstraten, J.M. and Westerveld, J. (1997) Relation
between bioavailability and fuel oil hydrocarbon composition in contaminated
soils. Environmental Science and Technology 31, 771-775.
243
de Maagd, P.G.-J., Hulscher, D.T., van den Heuvel, H., Opperhuizen, A. and
Sijm, D.T.H.M. (1998) Physicochemical properties of polycyclic aromatic
hydrocarbons. Aqueous solubilities, n-octanol/water partition coefficients and
Henry's law constants. Environmental Toxicology and Chemistry 17, 251-257.
Defra (2004) The Landfill (England and Wales) Regulations 2005: A
Consultation Paper: Department for Environment, Food and Rural Affairs.
Dionex. (2003) Accelerated Solvent Extraction. Available from:
www.dionex.com/app/tree.taf?asset_id=56146. Accessed: 5/6/06.
Dionex. (2004) Application note 338 Extraction of total petroleum hydrocarbon
contaminants (diesel and waste oil) in soils by accelerated solvent extraction.
Available from: http://wwwl.dionex.com/en-us/by_technique/extraction.
Accessed: 6/4/05.
Dionex Corporation. (1999) ASE® 200 Accelerated Solvent Extractor: It's Time
to Accelerate. www.dionex.com/en-us/webdocs/4036 ASE200 Bro. Accessed:
05/06/06.
Doick, K.J., Burauel, P., Jones, K.C. and Semple, K.T. (2005a) Effect of
cyclodextrin and transformer oil amendments on the chemical extractability of
aged [14C] polychlorinated biphenyl and [14C] polycyclic aromatic hydrocarbon
residues in soil. Environmental Toxicology and Chemistry 24, 2138-2144.
244
Doick, K.J., Klingelmann, E., Burauel, P., Jones, K.C. and Semple, K.T. (2005b)
Long-term fate of polychlorinated biphenyls and polycyclic aromatic
hydrocarbons in an agricultural soil. Environmental Science and Technology 39,
3663-3670.
Doick, K.J., Lee, P.H. and Semple, K.T. (2003) Assessment of spiking
procedures for the introduction of a phenanthrene-LNAPL mixture into field-wet
soil. Environmental Pollution 126, 399-405.
Dunnivant, F.M. and Anders, E. (2006) A Basic Introduction to Pollution Fate
and Transport: An integrated approach with chemical modelling, risk assessment
and environmental legislation. New Jersey: Wiley Interscience.
ECB. (2000) IUCLID Data Sheet Product ID 64742-53-6. Available from:
http://ecbjrc.it/esis/esis.php?PGM=ein&DEPUIS----autre. Accessed: 10/06/05.
ECB. (2005) EISIS: European Chemical Substances Information System.
Available from: http://ecb.jrc. it/esis/esis.php?PGM=ei,-.5tDEPUIS=autre.
Accessed: 10/6/05.
EDF Energy. (2008a) EDF Energy Press Kit. Available from:
http://www.edfenergy.com/media-centre/press-kit/edf-energy-press-briefing.
Accessed: 6/8/08.
245
EDF Energy. (2008b) About EDF Energy Networks. Available from:
http://www.edfenergy.com/products-services/networks. Accessed: 6/8/08.
Edwards, M.A. (1998) A Guide to In-Vessel Composting plus A Directory of
Systems. Coventry: The Composting Association.
Ehlers, G.A.C. and Loibner, A.P. (2006) Linking organic pollutant
(bio)availability with geosorbent properties and biomimetic methodology: A
review of geosorbent characterisation and (bio)availability prediction.
Environmental Pollution 141, 494-512.
El-Gayara, M.S., Gohara, G.A., Ibrahima, A.M., Ibrahimb, H.M. and Aly, A.M.
(2008) Transformer oils prepared from the vacuum distillates of Egyptian crude
paraffinic petroleum. Fuel Processing Technology 89, 254-261.
Energy Networks Association. (2005) Key Facts 2002/2003. Available from:
http://www.energynetworks.org/netguide_04.asp#facts. Accessed: 5/4/05.
Energy Networks Association. (2007) Guidance on the National Operating Code
for the Management of Fluid-Filled Cable Systems. Available from:
www.energynetworks.org/spring/she/pdfs/environment/ea_ena joint_guidance_O
1.pdf. Accessed : 30/1/08.
246
Englert, C.J., Kenzie, E.J. and Dragun, J. (1993) Bioremediation of petroleum
products in soil. In Principles and Practices for Petroleum Contaminated Soils
eds. Englert, C.J. and Kostecki, P.T. pp.111-129. Boca Raton: Lewis.
Environment Agency. (2003) Principles for Evaluating the Human Health Risks
from Petroleum Hydrocarbons in Soils: A Consultation Paper. Bristol:
Environment Agency.
Esteve-Turrillas, F.A., Scott, W.C., Pastora, A. and Dean, J.R. (2005) Uptake and
bioavailability of persistent organic pollutants by plants grown in contaminated
soil. Journal of Environmental Monitoring 7, 1093-1098.
Feitkenhauer, H., Muller, R. and Markl, H. (2003) Degradation of polycyclic
aromatic hydrocarbons and long chain alkanes at 6070 degrees C by Thermus
and Bacillus spp. Biodegradation 14, 367-372.
Ferguson, S.H., Franzmann, P.D., Revill, A.T., Snape, I. and Rayner, J.L. (2003)
The effects of nitrogen and water on mineralisation of hydrocarbons in diesel-
contaminated terrestrial Antarctic soils. Cold Regions Science and Technology
37, 197-212.
Ferreira, M.M.C. (2001) Polycyclic aromatic hydrocarbons: a QSPR study.
Chemosphere 44, 125-146.
247
Fine, P., Graber, E.R. and Yaron, B. (1997) Soil interactions with petroleum
hydrocarbons: Abiotic processes. Soil Technology 10, 133-153.
Finstein, M.S. and Hogan, J.A. (1993) Integration of composting process
microbiology, facility structure and decision making. In Science and Engineering
of Composting: Design, Environmental, Microbiological and Utilization Aspects
eds. Hoitink, H. and Keener, T. pp.1-21. Ohio: Renaissance.
Fritsche, W. and Hofrichter, M. (2001) Aerobic degradation by microorganisms.
In Environmental Processes II- Soil Decontamination, Biotechnology Voll lb ed.
Klein, J. pp.145-155. Germany: VCH.
Gao, Y., Xiong, W., Ling, W. and Xu, J. (2006) Sorption of phenanthrene by
soils contaminated with heavy metals. Chemosphere 65, 1355-1361.
Godoy-Faimdez, A., Antizar-Ladislao, B., Reyes-Bozo, L., Camario, A. and
Saez-Navarrete, C. (2008) Bioremediation of contaminated mixtures of desert
mining soil and sawdust with fuel oil by aerated in-vessel composting in the
Atacama Region (Chile). Journal of Hazardous Materials 151, 649-657.
Golueke, C.G. (1991) Principles in composting. In The BioCycle Journal of
Waste Recycling ed. BioCycle. pp.14-27. Pennsylvania: JG Press Inc.
Gove, L. (2001) Effect of biosolid application to agricultural land on the
potential for transport of nutrients (N and P) and heavy metal contaminants (Cd,
248
Cu, Ni, Pb and Zn) in soils In Department of Agricultural Sciences. Wye
Imperial College.
Gray, M.R., Banerjee, D.K., Dudas, M.J. and Pickard, M.A. (2000) Protocols to
enhance biodegradation of hydrocarbon contaminants in soil. Bioremediation
Journal 4, 249-257.
Green, C.T. and Scow, K.M. (2000) Analysis of phospholipid fatty acids (PLFA)
to characterize microbial communities in aquifers. Hydrogeology Journal 8, 126-
141.
Gustafson, J.B. (2002) Using TPH in Risk-Based Corrective Action, Shell
Development Company. Available from: http://www.epa.gov/swerustl/rbdm.
Accessed: 5/5/05.
Haderlein, A., Legros, R. and Ramsay, B. (2001) Enhancing pyrene
mineralization in contaminated soil by the addition of humic acids or composted
contaminated soil. Applied Microbiology and Biotechnology 56, 555-559.
Haderlein, A., Legros, R. and Ramsay, B.A. (2006) Pyrene mineralization
capacity increases with compost maturity. Biodegradation 17, 293-302.
Halligan, J. (2005) Personal communication, Lubricants Technical Advisor, Shell
U.K. Oil Products Limited.
249
Hamamura, N., Olson, S.H., Ward, D.M. and Inskeep, W.P. (2006) Microbial
population dynamics associated with crude-oil biodegradation in diverse soils.
Applied and Environmental Microbiology 72, 6316-6324.
Hansen, L.D., Nestler, C., Ringelberg, D. and Bajpai, R. (2004) Extended
bioremediation of PAH/PCP contaminated soils from the POPILE wood
treatment facility. Chemosphere 54, 1481-1493.
Harlow, J.H. (2003) Electric Power Transformer Engineering. Boca Raton,
Florida: CRC Press.
Harris, J.A., Birch, P. and Palmer, J. (1998) land Restoration and Reclamation:
Principles and Practice. Harlow: Longman.
Hartlieb, N., Ertunc, T., Schaeffer, A. and Klein, W. (2003) Mineralization,
metabolism and formation of non-extractable residues of 14C-labelled organic
contaminants during pilot-scale composting of municipal biowaste.
Environmental Pollution 126, 83-91.
Hassen, A., Belguith, K., Jedidi, N., Cherif, A., Cherif, M. and Boudabous, A.
(2001) Microbial characterization during composting of municipal solid waste.
Bioresource Technology 80, 217-225.
250
Hatzinger, P.B. and Alexander, M. (1995) Effect of aging of chemicals in soil on
their biodegradability and extractability. Environmental Science and Technology
29, 537-545.
Haus, F., German, J. and Junter, G.-A. (2001) Primary degradability of mineral
base oils in relation to their chemical and physical characteristics. Chemosphere
45, 983-990.
Haws, N.W., Ball, W.P. and Bouwer, E.J. (2006) Modeling and interpreting
bioavailability of organic contaminant mixtures in subsurface environments.
Journal of Contaminant Hydrology 82, 255- 292.
Hawthorne, S.B., Grabanski, C.B., Martin, E. and Miler, D.J. (2000)
Comparisons of Soxhlet extraction, pressurized liquid extraction , supercritical
fluid extraction and subcritical water extraction for environmental solids:
recovery, selectivity and effects on sample matrix. Journal of Chromatography A
892, 421-433.
Hayes, T.D. and Soni, B.K. (2006) Vapor pressure characterization to predict
contaminant releases from MGP site source area soils Soil and Sediment
Contamination 15, 511 - 527.
Hellmann, R.F. and Shann, J.F. (1997) Microbial community changes during the
composting of municipal solid waste. Microbial Ecology 33, 78-85.
251
Hesnawi, R.M. and McCartney, D.M. (2006) Impact of compost amendments
and operating temperature on diesel fuel bioremediation. Journal of
Environmental Engineering and Science 5, 37-45.
Hofman, J., Rhodes, A. and Semple, K.T. (2008) Fate and behaviour of
phenanthrene in the natural and artificial soils. Environmental Pollution 152,
468-475.
Huesemann, M.H., Hausmann, T.S. and Fortman, T.J. (2002) Microbial factors
rather than bioavailability limit the rate and extent of PAH biodegradation in
aged crude oil contaminated model soils. Bioremediation Journal 6, 321-336.
Huesemann, M.H., Hausmann, T.S. and Fortman, T.J. (2003) Assessment of
bioavailability limitations during slurry biodegradation of petroleum
hydrocarbons in aged soils. Environmental Toxicology and Chemistry 22, 2853-
2860.
Hughes, K.A., Bridge, P. and Clark, M.S. (2007) Tolerance of Antarctic soil
fungi to hydrocarbons. Science of the Total Environment 372, 539-548.
Hutchison, K.J., Hesterberg, D. and Chou, J.W. (2001) Stability of reduced
organic sulfur in humic acid as affected by aeration and pH. Soil Science Society
of America Journal 65, 704-709.
252
Hwang, S. and Cutright, T.J. (2002) Impact of clay minerals and DOM on the
competitive sorption/desorption of PAHs. Soil and Sediment Contamination 11,
269-291.
IUCLID (2000) Data Set for Distillates (petroleum), hydrotreated light
naphthenic, CAS number 64742-53-6: European Commission- European
Chemicals Bureau.
Jager, T., Baerselman, R., Dijkman, E., De Groot, A.C., Hogendoorn, E.A., De
Jong, A., Kruitbosch, J.A.W. and Peijnenburg, W.J.G.M. (2003) Availability of
polycyclic aromatic hydrocarbons to earthworms (Eisenia andrei, oligochaeta) in
field-polluted soils and soil-sediment mixtures. Environmental Toxicology and
Chemistry 22 767-775.
Johnsen, A.R., Wick, L.Y. and Harms, H. (2005) Principles of microbial PAH
degradation in soil. Environmental Pollution 133, 71-84.
Jonsson, S., Persson, Y., Frankki, S., van Bavel, B., Lundstedt, S., Haglund, P.
and Tysklind, M. (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, 86-96.
Joo, H.S., Shoda, M. and Phae, C.-G. (2007) Degradation of diesel oil in soil
using a food waste composting process. Biodegradation 18, 597-605.
253
Jorgensen, K.S., Puustinen, J. and Suortti, A.-M. (2000) Bioremediation of
petroleum hydrocarbon-contaminated soil by composting in bioplies.
Environmental. Pollution 107, 245-254.
Kamra, S.K. and Lennartz, B. (2005) Quantitative indices to characterize the
extent of preferential flow in soils. Environmental Modelling and Software 20,
903-915.
Kanaly, R.A. and Harayama, S. (2000) Biodegradation of high-molecular-weight
polycyclic aromatic hydrocarbons by bacteria. Journal of Bacteriology 182,
2059-2067.
Kanga, S.A., Bonner, J.S., Page, C.A., Mills, M.A. and Autenrieth, R.L. (1997)
Solubilization of naphthalene and methyl-substituted naphthalenes from crude oil
using biosurfactants. Environmental Science and Technology 31, 556-561.
Keith-Roach, M.J., Bryan, N.D., Bardgett, R.D. and Livens, F.R. (2002)
Seasonal changes in the microbial community of a salt marsh, measured by
phospholipid fatty acid analysis. Biogeochemistiy 60, 77-96.
Kelsey, J.W., Kottler, B.D. and Alexander, M. (1997) Selective chemical
extractants to predict bioavailability of soil-aged organic chemicals.
Environmental Science and Technology 31, 214-217.
254
Kim, J.-D., Park, J.-S., In, B.-H., Kim, D. and Namkoong, W. (2008) Evaluation
of pilot-scale in-vessel composting for food waste treatment. Journal of
Hazardous Materials 154, 272-277
Khmer, M. and Baath, E. (1998) Microbial community dynamics during
composting of straw material studied using phospholipid fatty acid analysis.
FEMS Microbiology Ecology 27, 9-20.
Korner, I., Braukmeier, J., Herrenklage, J., Leikam, K., Ritzkowski, M.,
Schlegelmilch, M. and Stegmann, R. (2003) Investigation and optimization of
composting processes- test systems and practical examples. Waste Management
23, 17-26.
Kottler, B.D., White, J.C. and Kelsey, J.W. (2001) Influence of soil moisture on
the sequestration of organic compounds in soil. Chemosphere 42, 893-898.
Krauss, M. and Wilcke, W. (2001) Predicting soil-water partitioning of
polycyclic aromatic hydrocarbons and polychlorinated biphenyls by desorption
with methanol-water mixtures at different temperatures. Environmental Science
and Technology 35, 2319-2325.
Krauss, M., Wilcke, W. and Zech, W. (2000) Availability of polycyclic aromatic
hydrocarbons (PAHs) and polychlorinated biphenyls (PCBs) to earthworms in
urban soils. Environmental Science and Technology 34, 4335-4340.
255
Kriipsalu, M., Marques, M., Nammari, D.R. and Hogland, W. (2007) Bio-
treatment of oily sludge: The contribution of amendment material to the content
of target contaminants, and the biodegradation dynamics. Journal of Hazardous
Materials 148, 616-622.
Labud, V., Garcia, C. and Hernandez, T. (2007) Effect of hydrocarbon pollution
on the microbial properties of sandy and a clay soil. Chemosphere 66, 1863-
1871.
Laine, M.M., Haario, H. and Jorgensen, K.S. (1997) Microbial functional activity
during composting of chlorophenol-contaminated sawmill soil. Journal of
Microbiological Methods 30, 21-32.
Laine, M.M. and Jorgensen, K.S. (1997) Effective and safe composting of
chlorophenol-contaminated soil in pilot scale. Environmental Science and
Technology 31, 371-378.
Latawiec, A.E., Swindell, A.L. and Reid, B.J. (2008) Environmentally friendly
assessment of organic compound bioaccessibility using sub-critical water.
Environmental Pollution In Press, 1-7.
Lazzari, L., Sperni, L., Salizzato, M. and Pavoni, B. (1999) Gas chromatographic
determination of organic micropollutants in samples of sewage sludge and
compost: behaviour of PCB and PAH during composting. Chemosphere 38,
1925-1935.
256
Leahy, J.G. and Colwell, R.R. (1990) Microbial degradation of hydrocarbons in
the environment. Microbiological Reviews 54, 305-315.
Leblond, J.D., Schultz, T.W. and Sayler, G.S. (2001) Observations on the
preferential biodegradation of selected components of polyaromatic hydrocarbon
mixtures. Chemosphere 42, 333-343.
Lee, P.-H., Ong, S.K., Golchin, J. and Nelson, G.L.S. (2001) Use of solvents to
enhance PAH biodegradation of coal tar-contaminated soils. Water Research 35,
3941-3949.
Lee, P.H., Doick, K.J. and Semple, K.T. (2003) The development of
phenanthrene catabolism in soil amended with transformer oil. FEMS
Microbiology Letters 228, 217-223.
Lee, S.-H., Oh, BA. and Kim, J.-G. (2008) Effect of various amendments on
heavy mineral oil bioremediation and soil microbial activity. Bioresource
Technology 99, 2578-2587.
Liang, C., Das, K.C. and McClendon, R.W. (2003) The influence of temperature
and moisture contents regimes on the aerobic microbial activity of a biosolids
composting blend. Bioresource Technology 86, 131-137.
257
Lima, A.L.C., Farrington, J.W. and Reddy, C.M. (2005) Combustion-derived
polycyclic aromatic hydrocarbons in the environment-A review. Environmental
Forensics 6, 109-131.
Lipscomb, T.G. (1988) Mineral Insulating Oil Manufacture and Safekeeping. In
Electrical Insulating Oils ed. Erdman, H.G. pp.5-24. Philadelphia: American
Society for Testing and Materials.
Lors, C. and Mossmann, J.-R. (2005) Characteristics of PAHs intrinsic
degradation in two coke factory soils. Polycyclic Aromatic Compounds 25, 67-
85.
Ma, Y., Zhang, J.Y. and Wong, M.H. (2003) Microbial activity during
composting of anthracene-contaminated soil. Chemosphere 52, 1505-1513.
Mackay, D., Shiu, W.Y. and Ma, K.C. (1992) Illustrated Handbook of Physico-
Chemical Properties and Environmental Fate for Organic Chemicals. Boca
Raton: Lewis.
Madsen, E. (2003) Report on Bioavailability of Chemical Wastes with Respect to
the Potential for Soil Bioremediation. Washington.
Maliszewska-Kordybach, B. (2005) Dissipation of polycyclic aromatic
hydrocarbons in freshly contaminated soils- the effect of soil physicochemical
properties and aging. Water, Air and Soil Pollution 168, 113-128.
258
Mancera-Lopez, M.E., Esparza-Garcia, F., Chavez-Gomez, B., Rodriguez-
Vazquez, R., Saucedo-Castafieda, G. and Barrera-Cortes, J. (2008)
Bioremediation of an aged hydrocarbon-contaminated soil by a combined system
of biostimulation—bioaugmentation with filamentous fungi. International
Biodeterioration and Biodegradation 61, 151-160.
Margesin, R. and Schinner, F. (2001) Biodegradation and bioremediation of
hydrocarbons in extreme environments. Applied Microbiology and
Biotechnology 56, 650-663.
Margesin, R., Zimmerbauer, A. and Schinner, F. (2000) Monitoring of
bioremediation by soil biological activities. Chemosphere 40, 339-346.
Marin, M.M., Yuste, L. and Rojo, F. (2003) Differential expression of the
components of the two alkane hydroxylases from Pseudomonas aeruginosa.
Journal of Bacteriology 185, 3232-3237.
McFarland, M.J. and Qui, X.J. (1995) Removal of benzo(a)pyrene in soil
composting systems amended with the white-rot fungus Phanerochaete
chrysosporium. Journal of Hazardous Materials 42, 61-70.
Miller, G.C., Hebert, V.R. and Miller, W.W. (1989) Effect of sunlight on organic
contaminants at the atmosphere-soil interface. In Reaction and Movement of
Organic Chemicals in Soil eds. Sawhney, B.L. and K., B. pp. 99-110. Madison,
Wisconsin: Soil Science Society of America.
259
Mills, M.A., McDonald, T.J., Bonner, J.S., Simon, M.A. and Autenrieth, R.L.
(1999) Method for quantifying the fate of petroleum in the environment
Chemosphere 39, 2563-2582.
Mohamad-Ghazali, F., Rahman, R.N., Bakar Salleh, A. and Basri, M. (2004)
Biodegradation of hydrocarbons in soil by microbial consortium. International
Biodeterioration and Biodegradation 54, 61-67.
Mohamed, A.M.I., El-Menshawy, N. and Saif, A.M. (2007) Remediation of
saturated soil contaminated with petroleum products using air sparging with
thermal enhancement. Journal of Environmental Management 83, 339-350.
Mohan, S.V., Kisa, T., Ohkuma, T., Kanaly, R.A. and Shimizu, Y. (2006)
Bioremediation technologies for treatment of PAH-contaminated soil and
strategies to enhance process efficiency. Reviews in Environmental Science and
Bio-Technology 5, 347-374.
Mohee, R. and Mudhoo, A. (2005) Analysis of the physical properties of an in-
vessel composting matrix. Powder Technology 155, 92-99.
Molnar, M., Fenyvesi, E., Gruiz, K., Leitgib, L., Balogh, G., Muranyi, A. and
Szejtli, J. (2002) Effects of RAMEB on bioremediation of different soils
contaminated with hydrocarbons Journal of Inclusion Phenomena and
Macrocyclic Chemistry 44, 447-452.
260
Moretto, L.M., Silvestri, S., Ugoa, P., Zorzi, G., Abbondanzi, F., Baiocchi, C.
and Iacondini, A. (2005) Polycyclic aromatic hydrocarbons degradation by
composting in a soot-contaminated alkaline soil. Journal of Hazardous Materials
B126, 141-148.
Nam, K. and Alexander, M. (2001) Relationship between biodegradation rate and
percentage of a compound that becomes sequestered in soil. Soil Biology and
Biochemistry 33, 787-792.
Namkoong, W., Hwang, E.-Y., Park, J.-S. and Choi, J.-Y. (2002) Bioremediation
of diesel-contaminated soil with composting. Environmental Pollution 119, 23-
31.
National Grid. (1998) Environmental Issues Impacts to Land and Water.
Available from: www.nationalgrid.com/uk/library/archive/performance_reports.
Accessed: 5/4/05.
National Grid. (2005) Overhead or Underground? A Question of Insulation.
Available from: www.nationalgrid.com/uk/library/brochures/. Accessed: 29/7/05.
National Grid. (2008a) Cable oil releases. Available from:
http://www.nationalgrid.com/corporate/Our+Responsibility/Reporting+our+Perf
ormance/Environment/Cable+oil/. Accessed: 12/5/08.
261
National Grid. (2008b) Electricity Transmission. Available from:
http://www.nationalgrid.com/corporate/Our+Businesses/transmission/. Accessed:
07/8/08.
National Grid Transco. (2004) Responsibility: Environmental Performance-
Cable Oils. Available from : http://www.nationalgrid.com/uk/. Accessed: 5/4/05.
Ncibi, M.C., Mahjouba, B. and Gourdon, R. (2007) Effects of aging on the
extractability of naphthalene and phenanthrene from Mediterranean soils.
Journal of Hazardous Materials 146, 378-384.
Nordstedt, R.A., Barkdoll, A.W. and Schroeder, R.M. (1993) Composting Yard
Wastes. In Science and Engineering of Composting: Design, Environmental,
Microbiological and Utilisation Aspects ed. Hoitink H. and Keener, H. Ohio:
Renaissance.
Northcott, G.L. and Jones, K.C. (2000a) Spiking hydrophobic organic
compounds into soil and sediment: A review and critique of adopted procedures.
Environmental Toxicology and Chemistry 19, 2418-2430.
Northcott, G.L. and Jones, K.C. (2000b) Experimental approaches and analytical
techniques for determining organic compound bound residues in soil and
sediment. Environmental Pollution 108, 19-43.
262
Northcott, G.L. and Jones, K.C. (2001) Partitioning, extractability, and formation
of nonextractable PAH residues in soil. 1. Compound differences in aging and
sequestration. Environmental Science and Technology 35, 1103-1110.
Nynas Naphthenics. (2001) Base Oil Handbook. Available from:
www.nynas.com/sa/node.asp?node=44. Accessed: 26/4/05.
Nynas Naphthenics. (2004) Nytro 10GBN Safety Data Sheet. Available from:
http://www.nynas.com/templates/ProductList_2659.aspx?epslanguage=EN.
Accessed: 16/5/05.
Olds College Composting Technology Centre. (2005) Composting Based
Bioremediation. Available from: www. o cc i. ab.ca/PDF%20Fi les/Compo st ing.
Accessed: 13/5/05.
Oleszczuk, P. (2007) Changes of polycyclic aromatic hydrocarbons during
composting of sewage sludges with chosen physico-chemical properties and
PAHs content. Chemosphere 67, 582-591.
Oleszczuk, P. and Baran, S. (2003) Degradation of individual polycyclic
aromatic hydrocarbons (PAHs) in soil polluted with aircraft fuel. Polish Journal
of Environmental Studies 12, 431-437.
Onifade, A.K. and Abubakar, F.A. (2007) Characterization of hydrocarbon-
degrading microorganisms isolated from crude oil contaminated soil and
263
remediation of the soil by enhanced natural attenuation. Research Journal of
Microbiology 2, 149-155.
Ophardt, C.E. (2003) Boiling Points and Structures of Hydrocarbons. Elmhurst
College. Available from: http://www.elmhurstedu/—chm/vchebook. Accessed:
22/7/08.
Ostberg, T.L., Jonsson, A.P., Bylund, D. and Lundstrom, U.S. (2007) The effects
of carbon sources and micronutrients in fermented whey on the biodegradation of
n-hexadecane in diesel fuel contaminated soil. International Biodeterioration
and Biodegradation 60, 334-341.
Ostberg, T.L., Jonsson, A.P. and Lundstrom, U.S. (2006) Accelerated
biodegradation of n-alkanes in aqueous solution by the addition of fermented
whey. International Biodeterioration and Biodegradation 57, 190-194.
Palomo, M. and Bhandari, A. (2006) Impact of aging on the formation of bound
residues after peroxidase-mediated treatment of 2,4-DCP contaminated soils.
Environmental Science and Technology 40, 3402-3408.
Paria, S. (2008) Surfactant-enhanced remediation of organic contaminated soil
and water. Advances in Colloid and Interface Science 138, 24-58.
264
Park, K.S., Sims, R.C., Dupont, R. and Doucetta, W.J. (1990) Fate of PAH
compounds in two soil types: Influence of volatilization, abiotic loss and
biological activity. Environmental Toxicology and Chemistry 9, 187-195.
Patel, D. (2004) Dry type transformers: the better alternative to liquid-filled.
Electricity Today 2, 8-10.
Paudyn, K., Rutter, A., Rowe, K.R. and S., P.J. (2007) Remediation of
hydrocarbon contaminated soils in the Canadian Arctic by landfarming. Cold
Regions Science and Technology 53, 102-114.
Peressuttia, S.R., Alvarez, H.M. and Pucci, O.H. (2003) Dynamics of
hydrocarbon-degrading bacteriocenosis of an experimental oil pollution in
Patagonian soil. International Biodeterioration and Biodegradation 52, 21-30.
Pignatello, J.J. and Xing, B. (1996) Mechanisms of slow sorption of organic
chemicals to natural particles. Environmental Science and Technology 30, 1-11.
Plohl, K., Leskovgek, H. and Bricelj, M. (2002) Biological degradation of motor
oil in water. Acta Chimica Slovenia 49, 279-289.
Pillai, I., Ritchie, L., Heywood, R., Wilson, G., Pahlavanpour, B., Setford, S. and
Saini, S. (2005) Development of an improved analytical method for the
determination of carcinogenic polycyclic aromatic hydrocarbons in transformer
oil. Journal of Chromatography 1064, 205-212.
265
PLT Scientific (2005). Soxhlet Apparatus. Available from
http://favorit.smeboss.com/albums. Accessed: 21/01/08.
Pollard, S.J.Y., Hough, R.L., Kim, K.-H., Bellarby, J., Paton, G., Semple, K.T.
and Coulon, F. (2008) Fugacity modelling to predict the distribution of organic
contaminants in the soil:oil matrix of constructed biopiles. Chemosphere 71,
1432-1439.
Potter, C.L., Glaser, J.A., Chang, L.W., Meier, J.R., Dosani, M.A. and
Herrmann, R.F. (1999) Degradation of polynuclear aromatic hydrocarbons under
bench-scale compost conditions. Environmental Science and Technology 33,
1717-1725.
Pravecek, T.L., Christman, R.F. and Pfaender, F.K. (2006) Microbial
bioavailability of pyrene in three laboratory-contaminated soils under aerobic
and anaerobic conditions. Journal of Contaminant Hydrology 86, 72-86.
Rapp, P. and Gabriel-Jiirgens, L.H.E. (2003) Degradation of alkanes and highly
chlorinated benzenes, and production of biosurfactants, by a psychrophilic
Rhodococcus sp. and genetic characterization of its chlorobenzene dioxygenase.
Microbiology 149, 2879-2890.
Raymond, J.W., Rogers, T.N., Shonnard, D.R. and Kline, A.A. (2001) A review
of structure-based biodegradation estimation methods. Journal of Hazardous
Materials B84, 189-215.
266
Reeve, R.N. (2002) Introduction to Environmental Analysis. Chichester: John
Wiley and Sons Ltd.
Reid, B.J., Jones, K.C. and Semple, K.T. (2000) Bioavailability of persistent
organic pollutants in soils and sediments- a perspective on mechanisms,
consequences and assessment. Environmental Pollution 108 103-112.
Rhykerd, R.L., Crews, B., McInnes, K.J. and Weaver, R.W. (1999) Impact of
bulking agents, forced aeration, and tillage on remediation of oil-contaminated
soil. Bioresource Technology 67, 279-285.
Ribeiro, F.A. and Ferreira, M.M. (2003) QSPR models of boiling point, octanol—
water partition coefficient and retention time index of polycyclic aromatic
hydrocarbons. Journal of Molecular Structure (Theochem) 663, 109-126.
Riffaldi, R., Levi-Minzi, R., Cardelli, R., Palumbo, S. and Saviozzi, A. (2006)
Soil biological activities in monitoring the bioremediation of diesel oil-
contaminated soil. Water, Air and Soil Pollution 170, 3-15.
Ritchie, N.J., Schutter, M.E., Dick, R.P. and Myrold, D.D. (2000) Use of length
heterogeneity PCR and fatty acid methyl ester profiles to characterize microbial
communities in soil. Applied and Environmental Microbiology 66, 1668-1675.
Rizzo, A.C.L., da Cunha, C.D., Santos, R.L.C., Santos, R.M., Magalhaes, H.M.,
Leiteb, S.G.F. and Sorianoc, A.U. (2008) Preliminary identification of the
267
bioremediation limiting factors of a clay bearing soil contaminated with crude
oil. Journal of the Brazilian Chemical Society 19, 169-174.
Romantschuk, M., Sarand, I., Petanen, T., Peltola, R., Jonsson-Vihanne, M.,
Koivula, T., Yrjala, K. and Haahtela, K. (2000) Means to improve the effect of in
situ bioremediation of contaminated soil: an overview of novel approaches.
Environmental Pollution 107, 179-185.
Ron, E.Z. and Rosenberg, E. (2002) Biosurfactants and oil bioremediation.
Current Opinion in Biotechnology 13, 249-252.
Rogues, D.E., Overton, E.B. and Henry, C.B. (1994) Using gas
chromatography/mass spectrometry fingerprint analyses to document progress of
oil degradation. Journal of Environmental Quality 23, 851-855.
Rosenberg, E. and Ron, E.Z. (1996) Bioremediation of Petroleum Hydrocarbons.
In Bioremediation: Principles and Applications eds. Crawford, R.L. and L., C.D.
pp.100-124. Cambridge: Cambridge University Press.
Rouse, T.O. (1998) Mineral insulating oil in transformers. IEEE Electrical
Insulation magazine 14, 6-16.
Rychen, G., Jurjanz, S., Toussaint, H. and Feidt, C. (2008) Dairy ruminant
exposure to persistent organic pollutants and excretion to milk. Animal 2, 312-
323.
268
Sabato, J., Villas, M. and Solanas, A.M. (2004) Laboratory-scale bioremediation
experiments on hydrocarbon-contaminated soils International Biodeterioration
and Biodegradation 54, 19-25.
SabljiO, A., Olsten, H., Verhaar, H. and Hermens, J. (1995) QSAR modelling of
soil sorption. Improvements and systematics of log 1(3c vs. log Ic>, correlations.
Chemosphere 31, 4489-1514.
Saenton, S., Illangasekare, T.H., Soga, K. and Saba, T.A. (2002) Effects of
source zone heterogeneity on surfactant-enhanced NAPL dissolution and
resulting remediation end-points. Journal of Contaminant Hydrology 59, 27- 44.
Saim, N., Dean, J.R., Abdullah, M.P. and Zakaria, Z. (1997) Extraction of
polycyclic aromatic hydrocarbons from contaminated soil using Soxhlet
extraction, pressurised and atmospheric microwave-assisted extraction,
supercritical fluid extraction and accelerated solvent extraction. Journal of
Chromatography A 701, 361-366.
Salanitro, J. (2001) Bioremediation of petroleum hydrocarbons in soil. Advances
in Agronomy 72, 53-105.
Samanta, S.K., Sing, O.V. and Jain, R.K. (2002) Polycyclic aromatic
hydrocarbons: environmental pollution and bioremediation. Trends in
Biotechnology 20, 243-248.
269
Sarkar, D., Ferguson, M., Datta, R. and Birnbaum, S. (2005) Bioremediation of
petroleum hydrocarbons in contaminated soils: Comparison of biosolids addition,
carbon supplementation, and monitored natural attenuation. Environmental
Pollution 136, 187-195.
§agek, V., Bhatt, M., Cajthaml, T., Malachova, K. and D., L. (2003) Compost-
mediated removal of polycyclic aromatic hydrocarbons from contaminated soil.
Archives of Environmental Contamination and Toxicology 44, 336-342.
Scelza, R., Rao, M.-A. and Gianfreda, L. (2007) Effects of compost and of
bacterial cells on the decontamination and the chemical and biological properties
of an agricultural soil artificially contaminated with phenanthrene. Soil Biology
and Biochemistry 39, 1303-1317.
Schelble, R.T., McDonald, G.D., Hall, J.A. and Nealson, K.H. (2005)
Community structure comparison using FAME analysis of desert varnish and
soil, Mojave Desert, California. Geomicrobiology Journal 22, 353-360.
Scherr, K., Aichberger, H., Braun, R. and Loibner, A.P. (2007) Influence of soil
fractions on microbial degradation behaviour of mineral hydrocarbons. European
Journal of Soil Biology 43, 341-350.
Scullion, J. (2006) Remediating polluted soils. Naturwissenschaften 93 51-65.
270
Semple, K.T., Morriss, A.W.J. and Paton, G.I. (2003) Bioavailability of
hydrophobic organic contaminants in soils: fundamental concepts and techniques
for analysis. European Journal of Soil Science 54, 809-818.
Semple, K.T., Reid, B.J. and Fermor, T.R. (2001) Impact of composting
strategies on the treatment of soils contaminated with organic pollutants.
Environmental Pollution 112, 269-283.
Sharer, M., Park, J.-H., Voice, T.C. and Boyd, S.A. (2003) Aging effects on the
sorption—desorption characteristics of anthropogenic organic compounds in soil.
Journal of Environmental Quality 32, 1385-1392.
Shell Lubricants. (2005) Electrical Oils. Available from: http://www.shell-
lubricants.com/Electrical/overview.html. Accessed: 4/4/05.
Short, J. (2003) Composting Training Manual London: Enviros Ltd and
Cleanaway.
Smith, M.J., Lethbridge, G. and Burns, R.G. (1997) Bioavailability and
biodegradation of polycyclic aromatic hydrocarbons in soils. FEMS
Microbiology Letters 152, 141-147.
Smith, P.N., Cobb, G.P., Godard-Codding, C., Hoff, D., McMurry, S.T.,
Rainwater, T.R. and Reynolds, K.D. (2007) Contaminant exposure in terrestrial
vertebrates. Environmental Pollution 150 41-64.
271
Soga, K., Page, J.W.E. and Illangasekare, T.H..(2004) A review of NAPL source
zone remediation efficiency and the mass flux approach. Journal of Hazardous
Materials 110, 13-27.
Staikova, M., Wania, F. and Donaldson, D.J. (2004) Molecular polarizability as a
single-parameter predictor of vapour pressures and octanol—air partitioning
coefficients of non-polar compounds: a priori approach and results. Atmospheric
Environment 38, 213-225.
Staples, C.A., Peterson, D.R., Parkerton, T.F. and Adams, W.J. (1997)
Environmental fate of phthalate esters: A literature review. Chemosphere 35,
667-749.
Steger, K., Jarvis, A., Vasara, T., Romantschuck, M. and Sundh, I. (2007) Effects
of differing temperature management on development of Actinobacteria
populations during composting. Research in Microbiology 158, 617-624.
Stelmachowski, M., Jastrzebska, M. and Zarzycki, R. (2003) In-vessel
composting for utilizing of municipal sewage-sludge. Applied Energy 75, 249-
256.
Stokes, J.D., Paton, G.I. and Semple, K.T. (2006) Behaviour and assessment of
bioavailability of organic contaminants in soil: relevance for risk assessment and
remediation. Soil Use and Management 21, 475-486.
272
Sun, H., Tatedaa, M., Ike, M. and Fujita, M. (2003) Short- and long-term
sorption/desorption of polycyclic aromatic hydrocarbons onto artificial solids:
effects of particle and pore sizes and organic matters. Water Research 37, 2960-
2968.
Svensson, C. (2005) Personal communication, Nitro 10GBN: Technical Services,
Nynas Naphthenics.
Syracuse Research Corporation. (1999) Interactive PhysProp Database.
Available from: http://www.syrres.com/esc/physdemo.htm. Accessed: 12/6/08.
Tellez, G.T., Niralakhandan, N. and Gardea-Torresdet, J.L. (2002) Performance
evaluation of an activated sludge system for removing petroleum hydrocarbons
from oilfield produced water. Advances in Environmental Research 6, 455-470.
Thomassin-Lacroix, E.J.M., Yu, Z., Eriksson, M., Reimer, K.J. and Mohn, W.W.
(2001) DNA-based and culture-based characterization of hydrocarbon-degrading
consortium enriched from Arctic soil. Canadian Journal of Microbiology 47,
1107-1115.
Tiquia, S.M., Tam, N.F.Y. and Hodgkiss, I.J. (1996) Microbial activities during
composting of spent pig-manure sawdust litter at different moisture contents.
Bioresource Technology 55, 201-206.
273
Tolls, J., van Dijk, J., Verbruggen, E.J.M., Hermens, J.L.M., Loeprecht, B. and
Schuermann, G. (2002) Aqueous solubility-molecular size relationships: A
mechanistic case study using C10- to C19-alkanes. Journal of Physical Chemistry
A 106, 2760-2765.
Toxnet. (2008) Toxicology Data Network. US National Library of Medicine.
Available from: http://toxnet.nlm.nih.gov/. Accessed: 12/6/08.
Trindade, P.V.O., Sobral, L.G., Rizzo, A.C.L., Leite, S.G.F. and Soriano, A.U.
(2005) Bioremediation of a weathered and a recently oil-contaminated soils from
Brazil: a comparison study. Chemosphere 58, 515-522.
Tuomela, M., Vikman, M., Hatakka, A. and Itavaara, M. (2000) Biodegradation
of lignin in a compost environment: a review. Bioresource Technology 72, 169-
183.
USGS. (2008) Octanol-Water Partition Coefficient (K00. Available from:
http://toxics.usgs.gov/definitions/kow.html. Accessed: 10/7/08.
Van Gestel, K., Mergaert, J., Swings, J., Coosemans, J. and Ryckeboer, J. (2003)
Bioremediation of diesel oil-contaminated soil by composting with biowaste.
Environmental Pollution 125, 361-368.
274
Van Hamme, J.D., Singh, A. and Ward, O.P. (2003) Recent advances in
petroleum microbiology. Microbiology and Molecular Biology Reviews 67, 503-
549.
VanderGheynst, J.S., Gossett, J.M. and Walker, L.P. (1997) High-solids aerobic
decomposition: pilot-scale reactor development and experimentation. Process
Biochemistry 32, 361-375.
Vasudevan, N. and Rajaram, P. (2001) Bioremediation of sludge-contaminated
soil. Environment International 26, 409-411.
Verbruggen, E.M.J., Hermens, J.L.M. and Tolls, J. (2000) Physicochemical
properties of higher nonaromatic hydrocarbons: A literature study. Journal of
Physical and Chemical Reference Data 29, 1435-1446.
Verschueren, K. (1996) Handbook of Environmental Data on Organic
Chemicals. New York: Van Nostrand Reinhold.
Wammer, K.H. and Peters, C.A. (2005) Polycyclic aromatic hydrocarbon
biodegradation rates: A structure based study. Environmental Science and
Technology 39, 2571-2578.
Wang, W., Zappi, M., Albritton, G., Crawley, A., Singletary, J., Hall, N. and
Karr, L. (2001) Using chemical priming as a means of enhancing performance of
biocells for the treating petroleum products containing recalcitrant chemical
275
species: Environmental Technology Research and Applications Laboratory,
Mississippi State University.
Wang, Z. and Fingas, M. (1995) Differentiation of the source of spilled oil and
monitoring of the oil weathering process using gas chromatography- mass
spectrometry. Journal of Chromatography 712, 321-343.
Ward, 0., Singh, A. and Van Hamme, J. (2003) Accelerated biodegradation of
petroleum hydrocarbon waste. Journal of Industrial Microbiology and
Biotechnology 30, 260-270.
Weber, W.J. and Kim, H.S. (2005) Optimizing contaminant desorption and
bioavailability in dense slurry systems. 1. Rheology, mechanical mixing, and
PAH desorption. Environmental Science and Technology 39, 2267-2273.
White, J.C., Quinones-Rivera, A. and Alexander, M. (1998) Effect of wetting
and drying cycles on the bioavailability of organic compounds sequestered in
soil. Environmental Toxicology and Chemistry 17, 2378-2382.
Wild, E., Dent, J., Thomas, G.O. and Jones, K.C. (2005) Direct observation of
organic contaminant uptake, storage, and metabolism within plant roots.
Environmental Science and Technology 39, 3695-3702.
276
Xu, R. and Obbard, J.P. (2003) Effect of nutrient amendments on indigenous
hydrocarbon biodegradation in oil-contaminated beach sediments. Journal of
Environmental Quality 32, 1234-1243.
Yaron, B., Calvet, R. and Prost, R. (1996) Soil Pollution: Processes and
Dynamics. Berlin: Springer Verlag.
Yerushahni, L., Rocheleau, S., Cimpoia, R., Sarrazin, M., Sunahara, G.,
Peisajovich, A., Leclair, G. and Guiot, S.R. (2003) Enhanced biodegradation of
petroleum hydrocarbons in contaminated soil. Bioremediation Journal 7, 37-51.
Zhou, Q. and Hua, T. (2004) Bioremediation: A review of applications and
problems to be resolved. Progress in Natural Science 14, 937-944.
Zucchi, M., Angiolini, L., Bonin, S., Brusetti, L., Dietrich, N., Gigliotti, C.,
Barbieri, P., Sorlini, C. and Daffonchio, D. (2003) Response of bacterial
community during bioremediation of an oil-polluted soil. Journal of Applied
Microbiology 94, 248-257.
277
Appendices
Appendix 4.1: List of DEFRA 17 PAHs
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benz[a]anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
Indeno(1,2,3,cd)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
Coronene
278