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whether published or unpublished.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


5.1.1. Hypotheses 118

9

5.1.2. Aims and Objectives 118


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


Chapter 6. Effect of Soil Properties on the Removal of TPH Contamination

During In-Vessel Composting 143


6.1.1. Hypothesis 145



6.3. Results 148

6.4. Discussion 152


Chapter 7. Effect of Contaminant Ageing on the Composting

Bioremediation of Electrical Insulating Oil-Contaminated Soils 157



10



7.3. Results 165

7.4. Discussion 170


Chapter 8. Effect of Contaminant Physico-Chemical Properties on Removal

of Contamination During In-Vessel Composting 174




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


Chapter 9. PLFA Analysis of Microbial Communities During Composting

Bioremediation 203





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


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


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



Figure 8.7. Relationship between increasing compound vapour pressure and loss

of compounds in; freshly contaminated a) gault clay, b) brick earth and aged


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


brick earth compost samples 222

Figure 9.15. Changes in microbial group ratios and loss of TPH during

composting under optimized conditions 226


composting in composted oil-contaminated gault clay 227


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


properties and compound loss for changes in soil to green waste ratio 185


properties and compound loss using different moisture contents 186


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.


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)



(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.


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.


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)




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.


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.


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 %)


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)




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


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


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


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)


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.


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.


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



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.


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 - - - • 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



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 - - - • 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 - - - • 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


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 • • - .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



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


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.


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


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


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


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)


gault clay compost samples

221

❑ Brick Fresh o Brick Naturally Aged


0 7 21 45 54 80 98

Time (days)


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)


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)


composting in composted oil-contaminated gault clay

0 7 21 45 54 80 98

Time (days)


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


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.


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.


Hassen, A., Belguith, K., Jedidi, N., Cherif, A., Cherif, M. and Boudabous, A.

(2001) Microbial characterization during composting of municipal solid waste.


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


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


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


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


Mohamad-Ghazali, F., Rahman, R.N., Bakar Salleh, A. and Basri, M. (2004)

Biodegradation of hydrocarbons in soil by microbial consortium. International


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


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.


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.


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.


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.


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


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.


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.


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.


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


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


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.


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


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

In-Vessel Composting Bioremediation of Electrical Insulating ...

Documents

Transcript of In-Vessel Composting Bioremediation of Electrical Insulating ...