Roles of organic-inorganic interactions in the generation of petroleum as exemplified by Lower Palaeozoic petroleum systems, Europe [original]

Roles of organic-inorganic interactions in the

generation of petroleum as exemplified by

Lower Palaeozoic petroleum systems, Europe

vorgelegt von

M.Sc. Geologe

Shengyu Yang

geb. in Liaoning, China

von der Fakultät VI - Planen Bauen Umwelt

der Technischen Universität Berlin

zur Erlangung des akademischen Grades

Doktor der Naturwissenschaften

Dr. rer. nat.

genehmigte Dissertation

Promotionsausschuss:

Vorsitzender: Prof. Dr. Wilhelm Dominik

Berichter: Prof. Dr. Brian Horsfield

Berichter: Prof. Dr. Reinhard Sachsenhofer

Tag der wissenschaftlichen Aussprache: 07.09.2017

Berlin 2017

Für meine Eltern

献给父母

“So lange Vater und Mutter auf der Welt sind, solltest Du nicht in die Ferne reisen –

tust Du es dennoch, so gib wenigstens Dein Reiseziel bekannt.”

Konfuzius (551 BC – 479 BC)

父母在，不远游，游必有方。

孔子

ACKNOWLEDGEMENTS

It was 2011 when I met Prof. Dr. Brian Horsfield and Dr. Hans-Martin Schulz in Beijing

for the first time. Their courteous and affable characters besides profound knowledge

attracted me here in Germany, 7000+ km away from home. My first gratitude goes to Brian

who opened the gate of geochemistry world to me and showed me the way (cited from a

self-made song Hotel Organic Geochemistry). His constructive suggestions on academic

topics and enlightenments on life always inspire me to be a good researcher and enjoy the

life at the same time. Hans-Martin is another mentor to be greatly acknowledged which is

not merely attributed to the fact that we are both Borussia Dortmund fans. His elaborate

attitude toward science and humility to people act as great examples for me. I also

benefited greatly from his abundant suggestions and efficient corrections on my

manuscripts, some of which were even finished in the weekend.

My sincere gratitude is extended to Dr. Nicolaj Mahlstedt, Dr. Niels Hemmingsen

Schovsbo, Prof. Dr. Rolando di Primio, Dr. Mareike Noah, and Dr. Stefanie Pötz for

fruitful discussions. It’s impossible to carry out my research without the professional and

reliable technical supports from Ferdinand Perssen, Cornelia Karger, Anke Kaminsky, and

Kristin Günther.

Chinese Scholarship Council is greatly acknowledged for sponsoring my Ph.D. study.

I am grateful to be a member of an NGO (non-governmental organization) called

“cyclopentane” which consist Janina Stapel, Sascha Kuske, Seyed Hossein Hosseini

Baghsangani, and Volker Ziegs, in addition to me. The five “carbon atoms” bonded

together ends in never-ending entertainment, friendship, and trust.

Gratitude to my beloved father Yang Desan and mother Song Yuying is beyond words.

More than 30 years of constant love and expectations from them provide me an infinite

power forward. My two sisters and brother are also gratefully acknowledged for their love

and encouragement. Due to limited space here, my endless thanks to my wife Zhang Yu

will be presented in detail in the rest of my life.

III

LIST OF PUBLICATIONS

Articles:

(1) S. Yang, B. Horsfield, N. Mahlstedt, M. H. Stephenson, and S. F. Könitzer, 2015, On

the primary and secondary petroleum generating characteristics of the Bowland Shale,

northern England: Journal of the Geological Society, v. 173, p. 292-305 (postprint),

doi: 10.1144/jgs2015-056.

(2) S. Yang, and B. Horsfield, 2016, Some predicted effects of minerals on the

generation of petroleum in nature: Energy & Fuels, v. 30, p. 6677-6687 (postprint),

doi: 10.1021/acs.energyfuels.6b00934.

(3) S. Yang, H.-M. Schulz, N. H. Schovsbo, and J. A. Bojesen-Koefoed, 2017, Oil-source

rock correlation of the Lower Palaeozoic petroleum system in the Baltic Basin

(northern Europe), (in press; preliminary version published online Ahead of Print May

22, 2017): AAPG Bulletin (postprint), doi: 10.1306/02071716194.

(4) S. Yang, H-M. Schulz, B. Horsfield, and N.H. Schovsbo, H. Rothe and K. Hahne

Impact of uranium irradiation on the petroleum potential of the Cambro-Ordovician

Alum Shale, Northern Europe (preprint). Submitted to Geochimica et Cosmochimica

Acta on 27 Jul 2017.

Poster and Presentation:

(1) S. Yang, B. Horsfield, and M. H. Stephenson, 2015, Kinetics of primary and

secondary petroleum generation of the Bowland Shale: IMOG, Czech Republic. Poster.

(2) S. Yang, H.-M. Schulz, N. H. Schovsbo, and J. A. Bojesen-Koefoed, 2016, Oil-source

rock correlation of the Lower Palaeozoic petroleum system in the Baltic Basin: AAPG

Geosciences Technology Workshop, Lithuania. Oral presentation.

ABSTRACT

The clay-rich Lower Carboniferous Bowland Shale (England) and the uranium-rich

Cambro-Ordovician Alum Shale (Northern Europe), both deposited in marine

environments, are characterized by dominating gaseous and aromatic compounds in

pyrolysates. Therefore, the kerogens in both of these shales can be classified as type III

from classical perspectives, which is not consistent with the depositional environment. This

dissertation aims to investigate the organic-inorganic interactions in changing the

petroleum formation and occurrence. Special efforts were made to validate the interactions

in geological environments, besides the findings derived from laboratory experiments.

Interdisciplinary geochemical techniques covering pyrolytic methodologies, e.g., Rock-Eval

pyrolysis, pyrolysis gas chromatography (Py-GC), bulk kinetics, micro-scale sealed vessel-

Py-GC (MSSV-Py-GC), and high resolution experiments based on extractions, e.g., GC-

Flame Ionization Detector (GC-FID), GC-mass spectrometry (GC-MS), and Fourier

Transform Ion Cyclotron Resonance MS (FT-ICR-MS) are employed in unravelling the

organic-inorganic interactions. Furthermore, mass-balances, PhaseKinetics, and 1D

modelling are also carried out to predict the occurrence and behaviour of petroleum.

The mineral matrix effect (MME) does not only influence basic geochemical parameters,

e.g. a decrease in HI and increases in Tmax, OI and gas generation, but also changes the

hydrocarbon generation kinetics and phase behaviour. Accordingly, significant errors can

be induced by MME when basin modelling approaches are applied to predict secondary gas

generation. The extent of MME in shale varies according to the mineralogical composition.

In contrast to the clay-rich Bowland Shale, the calcite-rich Toolebuc Oil Shale and the

quartz-rich Alum Shale are only slightly and not affected by MME, respectively.

Furthermore, the size of the interface area between clay minerals and organic matter could

be another factor that influences the extent of MME.

The MME is found heating rate dependent, i.e., a lower heating rate weakens the MME on

pyrolysate aromaticity and generation quantity. Therefore, the catalytic and retention effects

of clay minerals are speculated to only exist in the laboratory environment and not in a

geological maturation process. This is further supported by the fact that the kerogen

pyrolysates resemble natural products more than products derived from whole rock

pyrolysis as revealed by FT-ICR MS experiments.

Robust correlations can be established between uranium contents and pyrolysate GOR and

aromaticity on the immature Alum Shale samples, indicating a strong uranium irradiation

effect on petroleum generation. Also, the aromaticity and GOR of free hydrocarbons from

thermovaporization-GC are proportional to those in pyrolysates which validates that the

uranium irradiation does influence the petroleum generation in nature. The FT-ICR MS

data reveal that the macro-molecules in the uranium-rich Alum Shale samples are less

alkylated and could be responsible why such kind of samples tend to generate gaseous

products.

Basin modelling results reveal that most Alum Shale horizons reached peak oil generation

between Late Devonian to Early Carboniferous (360-385 Mya) which means kerogens have

had experienced about a quarter of irradiation until peak oil generation compared with

radiation measured from nowadays core samples which have received full irradiation since

their formation (478-500 Mya). Kerogen structures during the oil window time were

reconstructed based on the fact that both irradiation time and uranium contents are linearly

correlated with irradiation damages. The kerogens are much more oil-prone and tend to

generate more aliphatic hydrocarbons back to Devonian time compared with nowadays

samples. In addition, the gas sorption capacity of the Alum Shale is supposed to be less

strong during Palaeozoic time in contrast to sorption experiments.

The unique distribution of triaromatic steroids in uranium-rich Alum Shale samples, i.e.,

decreased concentrations of C26-C28 triaromatic steroids with increasing uranium contents,

was found in oil samples sourced from the Alum Shale. This parameter could act as an oil-

source-correlation proxy, especially when aliphatic biomarkers are normally destroyed in

the uranium-rich Alum Shale.

In summary, catalytic effects under high temperatures and hydrocarbon retention induce

the MME, and the uranium irradiation works in condensation and cross-linking reactions.

The MME is a laboratory induced artefact and does not exist in geological environments.

The kerogen structures are not changed by minerals during diagenesis. In contrast, uranium

irradiation significantly alters kerogen structures, and furthermore changes the quantity and

quality of petroleum generation in both laboratory and geological environments. The

removal of minerals by acidic dissolution is an efficient way to avoid misleading

information caused by MME. A reconstruction of kerogen structures back to the time of

petroleum generation is crucially important to correctly evaluate the hydrocarbon

generation characteristics of uranium-rich source rocks.

VII

ZUSAMMENFASSUNG

Die Ton-reichen Schichten des Unterkarbonischen Bowland-Schiefers (England) und des

Uran-reichen Kambroordovizischen Alaunschiefers (Nordeuropa), die unter marinen

Bedingungen abgelagert wurden, zeichnen sich durch gasförmige und aromatische

Komponenten in ihren Pyrolysaten aus. Somit wird das Kerogen nach klassischen

Gesichtspunkten als Typ III beschrieben, was nicht mit den Ablagerungsbedingungen im

Einklang steht. Diese Dissertation hat zum Ziel, die organisch-anorganischen Interaktionen

zu untersuchen, welche die Erdölgenese und -vorkommen beeinflussen. Besonderes

Augenmerk wurde auf die Validierung der Prozesse unter geologischen Bedingungen gelegt,

sowie auf deren Vergleich mit Laborexperimenten.

Zur detaillierten Erkundung organisch-anorganischer Interaktionen wurden

interdisziplinäre geochemische Verfahren angewandt, die von pyrolytischen Methoden

reichen, z.B. Rock-Eval-Pyrolyse, gekoppelte Pyrolyse-Gaschromatographie (Py-GC),

Bulk-Kinetik, micro-scale sealed vessel-Pyrolyse-Gaschromatographie (MSSV-Py-GC), und

hochauflösende Experimente an Muttergesteinsextrakten, z.B. Gaschromatographie mit

Flammenionisierungsdetektor (GC-FID), gekoppelte Gaschromatographie-

Massenspektrometrie (GC-MS), und Fourier Transform Ion Cyclotron Resonance MS (FT-

ICR-MS). Weiterhin wurden zur Vorhersage von Vorkommen und Phasenverhalten des

gebildeten Erdöls Massenbilanzierungsberechnungen, PhaseKinetics und 1D-Modellierung

durchgeführt.

Der Mineralmatrix-Effekt (MME) wirkt sich nicht nur auf grundlegende geochemische

Parameter aus, verursacht z.B. einen Abfall des Wasserstoff-Indizes (HI) und einen Anstieg

des Tmax, Sauerstoff-Indizes (OI) und der Gasgenese, sondern beeinflusst die Kinetik der

Kohlenwasserstoffgenese und deren Phasenverhalten. Daraus können sich signifikante

Fehler in der Beckenmodellierung zur Vorhersage der Sekundärgasgenese ergeben, welche

hauptsächlich zur thermogenen Schiefergasproduktion beiträgt. Das Ausmaß der MME in

Schiefergesteinen variiert mit der Mineralzusammensetzung. Im Gegensatz zum Ton-

reichen Bowland-Schiefer sind der Kalzit-reiche Toolebuc Ölschiefer und der Quarz-reiche

Alaunschiefer durch den MME nur wenig bzw. nicht beeinflusst. Ein weiterer Faktor, der

sich auf den MME auswirkt, können Grenzflächeninteraktionen zwischen Tonmineralen

und organischer Materie sein.

VIII

Der Mineralmatrixeffekt ist Heizraten-abhängig; eine niedrige Heizrate vermindert die

Aromatizität der Pyrolysate und die Menge der gebildeten Produkte. Hieraus wird

geschlussfolgert, dass die katalytischen Effekte von und Retention an Tonmineralen

lediglich im Labor entstehen, jedoch nicht unter geologischen Reifebedingungen. Eine

höhere Ähnlichkeit von Kerogenpyrolysaten mit Mutter-gesteinsextrakten als mit deren

Pyrolysaten in FT-ICR-MS-Experimenten unterstützt diese These.

Unreife Alaunschiefer-Proben zeigen robuste Korrelationen zwischen Uranium-Gehalten

und Gas-Öl-Verhältnis (GOR) bzw. Aromatizität der Pyrolysate und deuten auf einen

starken Strahlungseffekt von Uranium auf die Erdölgenese. Weiterhin sind Aromatizität

und GOR von freien Kohlenwasserstoffen aus Thermovaporisationsexperimenten

proportional zu denen in Pyrolysaten, und unterstützt den Strahlungseffekt auf die

Erdölgenese unter natürlichen Bedingungen. FT-ICR-MS-Daten zeigen, dass

Makromoleküle im Uranium-reichen Alaunschiefer weniger alkylisiert sind, was die erhöhte

Gasgenese derartiger Proben verursachen kann.

Die Beckenmodellierung zeigt, dass der Großteil der Alaunschieferschichten den

Höhepunkt der Ölgenese zwischen Spätem Devon und Frühem Karbon (360-385 Mill.

Jahren vor heute, Ma) erreichten. Dies bedeutet, dass das Kerogen zu jener Zeit nur

ungefähr ein Viertel der Strahlung erhalten hat als dies an heutigen Kernproben gemessen

wurde, welche die volle Strahlungsmenge seit ihrer Bildung (478-500 Ma) erhalten haben.

Die Kerogenstruktur zur Zeit der maximalen Erdölgenese wurde, basierend auf linearen

Korrelationen von Strahlungsschäden mit Strahlungsdauer bzw. Uranium-Gehalten,

rekonstruiert. Das modellierte Kerogen generiert wesentlich höhere Mengen an Öl, welches

aliphatischer ist, als heute vorzufindende Proben. Zusätzlich wird angenommen, dass die

Gassorptionskapazität des Alaunschiefers im Paläozoikum weniger stark war als heutzutage.

Die bseondere Verteilung von triaromatischen Steroiden in Uranium-reichen

Alaunschiefer-Proben, z.B. verringerte Konzentrationen von C26-C28 triaromatischen

Steroiden mit steigendem Uranium-Gehalt, wurde ebenfalls in genetisch verwandten Ölen

gefunden. Dieser Parameter könnte als Korrelationsproxy für Öle und Muttergesteine

dienen, besonders da aliphatische Biomarker im Alaunschiefer normalerweise zerstört sind.

Katalytische Effekte bei hohen Temperaturen und die Retention von Kohlenwasserstoffen

lösen MME aus, und die Uranium-Strahlung verursacht Kondensationsreaktionen und

Querverbindungen. Der MME ist ein Effekt, der sich auf Laborexperimente beschränkt,

und existiert nicht unter natürlichen Bedingungen. Kerogenstrukturen werden während der

Diagenese nicht durch Minerale verändert. Im Gegensatz dazu beeinflusst Uranium-

Strahlung Kerogenstrukturen signifikant und determiniert Menge und Beschaffenheit

generierter Erdöle, sowohl unter Labor- als auch natürlichen Bedingungen. Eine effektive

Methode die irreführenden Informationen durch den MME zu vermeiden ist die azide

Demineralisierung der Muttergesteine. Eine Rekonstruktion der originären

Kerogenstruktur zur Zeit der maximalen Erdölgenese ist entscheidend für die korrekte

Evaluation der Kohlenwasserstoff-genesecharakteristika Uranium-reicher Muttergesteine.

CONTENTS

ACKNOWLEDGEMENTS ....................................................................................................................... I

LIST OF PUBLICATIONS.................................................................................................................... III

ABSTRACT ................................................................................................................................................... V

ZUSAMMENFASSUNG ........................................................................................................................ VII

CONTENTS ................................................................................................................................................ XI

LIST OF FIGURES .................................................................................................................................. XV

LIST OF TABLES .................................................................................................................................. XIX

LIST OF ABBREVIATIONS ............................................................................................................. XXI

1. INTRODUCTION .................................................................................................................................1

1.1 Development of Petroleum Formation Theories .............................................. 1

1.1.1 Biogenic vs. Abiogenic .............................................................................................................................. 1

1.1.2 Deep vs. Shallow ........................................................................................................................................ 3

1.1.3 Fatty Acid vs. Kerogen.............................................................................................................................. 5

1.2 Characterisation of Kerogen Structure ............................................................... 6

1.2.1 Generation quality and quantity ............................................................................................................. 6

1.2.2 Kinetics ......................................................................................................................................................... 8

1.3 Influence from Inorganic Materials ...................................................................... 9

1.3.1 Minerals ........................................................................................................................................................ 9

1.3.2 Uranium ..................................................................................................................................................... 14

1.4 Research Perspectives and Objectives .............................................................. 19

1.4.1 MME ............................................................................................................................................................ 19

1.4.2 Uranium ..................................................................................................................................................... 21

1.5 The Structure of the Dissertation ...................................................................... 23

2. MINERAL MATRIX EFFECT (MME) ........................................................................................ 27

2.1 Abstract ................................................................................................................ 27

2.2 Introduction ......................................................................................................... 27

2.3 Samples and Analytical Procedure .................................................................... 29

2.3.1 Samples ...................................................................................................................................................... 29

2.3.2 Analytical procedure .............................................................................................................................. 30

2.4 Results and Discussion ......................................................................................... 32

2.4.1 Primary generation ................................................................................................................................. 32

2.4.2 Secondary cracking ................................................................................................................................. 41

2.5 Application ........................................................................................................... 43

2.5.1 Primary generation ................................................................................................................................. 43

2.5.2 Secondary cracking ................................................................................................................................. 45

2.6 Conclusions .......................................................................................................... 47

XII

2.7 Acknowledgements .............................................................................................. 48

3. HEATING RATE DEPENDENCY OF MME ......................................................................... 49

3.1 Abstract ................................................................................................................ 49

3.2 Introduction .......................................................................................................... 50

3.3 Samples and Analytical Methods ........................................................................ 52

3.3.1 Samples ...................................................................................................................................................... 52

3.3.2 Analytical Methods .................................................................................................................................. 54

3.4 Result and Discussion ........................................................................................... 56

3.4.1 The existence of MME ........................................................................................................................... 56

3.4.2 The heating rate dependence of MME ............................................................................................... 60

3.4.3 Geological Calibration ........................................................................................................................... 63

3.4.4 Insights into Hetero-element Geochemistry ................................................................................... 64

3.5 Conclusions ........................................................................................................... 66

3.6 Acknowledgments ................................................................................................ 66

4. URANIUM IRRADIATION ON PETROLEUM GENERATION ................................. 67

4.1 Abstract ................................................................................................................ 67

4.2 Introduction .......................................................................................................... 68

4.3 Study Area and Samples ..................................................................................... 70

4.4 Experimental Methods ........................................................................................ 72

4.4.1 Uranium measurement .......................................................................................................................... 72

4.4.2 Pyrolytic techniques ............................................................................................................................... 72

4.4.3 FT-ICR MS................................................................................................................................................. 72

4.5 Results ................................................................................................................... 73

4.5.1 Screening data .......................................................................................................................................... 73

4.5.2 Open pyrolysis-gas chromatography and thermovaporisation ................................................... 74

4.5.3 FT-ICR MS................................................................................................................................................. 78

4.6 Discussion.............................................................................................................. 80

4.6.1 The uranium enrichment ....................................................................................................................... 80

4.6.2 Hydrocarbon precursors and products ............................................................................................ 81

4.6.3 Heterocompounds .................................................................................................................................. 84

4.6.4 Kerogen structure reconstruction ..................................................................................................... 86

4.7 Conclusion ............................................................................................................ 91

4.8 Acknowledgment ................................................................................................. 91

5. URANIUM IRRADIATION ON BIOMARKERS ................................................................. 93

5.1 Abstract ................................................................................................................ 93

5.2 Introduction .......................................................................................................... 94

5.3 Regional Petroleum Geology .............................................................................. 95

5.3.1 Regional Geodynamics and Basin Evolution ..................................................................................... 95

5.3.2 Source Rocks............................................................................................................................................ 97

5.3.3 Reservoirs ................................................................................................................................................. 99

5.3.4 Maturation and Accumulation .............................................................................................................. 99

XIII

5.4 Samples and Methods ........................................................................................ 100

5.4.1 Samples .................................................................................................................................................... 100

5.4.2 Methods ................................................................................................................................................... 101

5.5 Results ................................................................................................................. 101

5.5.1 GC-FID .................................................................................................................................................... 101

5.5.2 Aliphatic Biomarkers ............................................................................................................................ 102

5.5.3 Aromatic and NSO Biomarkers ........................................................................................................ 103

5.5.4 Oil Family Assignments ........................................................................................................................ 104

5.6 Discussion ........................................................................................................... 106

5.6.1 Maturity ................................................................................................................................................... 106

5.6.2 Volcanic Intrusion Induced Maturation ........................................................................................... 107

5.6.3 Correlation ............................................................................................................................................. 107

5.6.4 Heterogeneity within Alum Shale ..................................................................................................... 111

5.6.5 Migration and Mixing ............................................................................................................................ 113

5.7 Conclusions ........................................................................................................ 114

5.8 Acknowledgement ............................................................................................. 114

6. SUMMARY AND PESPECTIVES ............................................................................................. 115

6.1 Summary ............................................................................................................ 115

6.1.1 MME .......................................................................................................................................................... 115

6.1.2 Uranium ................................................................................................................................................... 116

6.1.3 Comparison ............................................................................................................................................ 117

6.2 Perspectives ....................................................................................................... 118

6.2.1 MME .......................................................................................................................................................... 118

6.2.2 Uranium ................................................................................................................................................... 118

REFERENCES .......................................................................................................................................... 121

XIV

LIST OF FIGURES

Fig. 2.1. Locations of samples and 1 D basin modelling well. Namurian basin distribution after Fraser and

Gawthorpe (2003).

Fig. 2.2. Stratigraphy of the study area. Depth, lithology and bulk δ13Corg data of the Carsington C4

core samples (Könitzer et al., 2014) as well as the locations of the 3 shale samples investigated

in this paper.

Fig. 2.3. Rock-Eval and TOC diagrams for kerogen type and maturity identification.

Fig. 2.4. PyGC chromatograms of the 6 samples. Normal alkane and alkene peaks have been highlighted

and selectively numbered. Representative aromatic compounds are ethylbenzene (a), meta- and

para-xylenes (b), ortho-xylene (c), 1,2,4-trimethylbenzene (d), naphthalene (e) and 2-

methylnaphtalen (f).

Fig. 2.5. Pyrolysate chain length distribution and Petroleum Type Organofacies classification.

Fig. 2.6. Petroleum composition predictions from PyGC results according Larter (1984) and Eglinton et

al. (1990).

Fig. 2.7. Bulk kinetics models of the whole rock samples and kerogen concentrates.

Fig. 2.8. Transformation ratio variations in geological heating rate (3 ℃ /Ma).

Fig. 2.9. Compositional kinetic models of selected samples.

Fig. 2.10. Gas:oil ratio of the samples analyzed as a function of increasing transformation ratio.

Fig. 2.11. Phase envelopes of whole rock and kerogen of sample No.3 during artificial maturation.

Fig. 2.12. Measured MSSV pyrolysis data of kerogen No.3 for boiling ranges C1+, C6+ and C1-5 normalized

to the maximum C1+ yield and fitted spline curves for calculated primary and secondary gas

generation using the heating rates of 5.0℃/min, compared to normalized SRA TR curve.

Fig. 2.13. Kinetics models of primary oil, primary gas and secondary gas generation of kerogen No.3.

Fig. 2.14. Computed generation rate curves as a function of temperature at a geological heating rate of

3℃/ma and vitrinite reflectance for kerogen No.3.

Fig. 2.15. Transformation ratio and Ro evolution histories of well Grove 3 and phase envelopes of

primarily generated fluids according to the maturity for upper Bowland Shale. The well location

can be found in Fig. 2.1. C-N and C-W in stratigraphy part represent Namurian and

Westphalian in Carboniferous respectively. Red triangle in each of the phase envelops

represents reservoir condition in geological burial history respectively.

Fig. 2.16. A comparison of secondary gas generation per km2 of upper Bowland Shale in well Grove 3 if

(a) secondary cracking kinetics model in this research and (b) kinetics model from Quigley et al.

1987 are applied in the basin modelling respectively.

XVI

Fig. 2.17. The maximum secondary gas generation per km2 of upper Bowland Shale in well Grove 3

when secondary kinetics of this research was applied as well as the predictions from 9 other

default Kerogen-Oil-Gas kinetic models in the PetroMod 2013.

Fig. 3.1. Basic geochemical screening based on Rock-Eval & TOC of whole-rock and kerogen samples.

Fig. 3.2. Comparison of PyGC maps on whole-rock and kerogen pairs. Bowland Shale pyrolysate shows

obviously higher aromatic compounds concentration compared with its kerogen counterpart.

(benz: benzene, tol: toluene, m,p xyl: meta- and para-xylene, o-xyl: ortho-xylene.)

Fig. 3.3. Quick classification on the pyrolysates of both whole-rock and kerogen concentrates.

Fig. 3.4. Bulk kinetic parameters of whole-rock and kerogen samples. Negligible (Alum Shale), Small

(Toolebuc Oil Shale) and significant (Bowland Shale) differences can be figured out.

Fig. 3.5. Geological extrapolation (heating rate: 3K/million year) of bulk kinetics parameters and

comparison of comparison on whole-rock and kerogen samples.

Fig. 3.6. Pyrolysate GC traces of Bowland whole-rock sample at different heating rates when heated to

TR 50%. The data manifests that the slower the heating rate is, the more aliphatic the products

would be generated. benz: benzene, tol: toluene, EB: ethylbenzene, xyl: xylene, TMB:

tetramethylbenzidine.

Fig. 3.7. Aliphatic/aromatic compounds ratio, GOR and bulk hydrocarbon generation/TOC variations of

Bowland Shale and Toolebuc Oil Shale samples in all three heating rates. Aliphatic compounds

include all normal alkenes and alkanes from C1-C30. Aromatic compounds are composed of

benzene, toluene, ethyl benzene, xylenes, tetramethylbenzidines, naphthalene, and branched

naphthalenes.

Fig. 3.8. The variation trends of aliphatic/aromatic ratios and bulk generation according to changing

heating rates on Bowland Shale samples when they are heated to TR 50%. Natural bitumen

reference was shown in figure 8a.

Fig. 3.9. Elemental class distribution pie charts of pyrolysates and matured shale extract in the negative

ESI spectra assigned with molecular formulas.

Fig. 3.10. “Class” comparison of Bowland kerogen and whole-rock pyrolysates. The O1/O2, N1/N2 and

N1O1/N1O2 ratios of the reference matured Bowland Shale extract are 0.31, 0.30 and 0.59,

respectively, which are more resemble to kerogen rather than whole-rock sample.

Fig. 4.1. Geographical overview of the Alum Shale sample distribution. The grey dash line depicts the

boundary of the Baltic Basin, and the red dash lines represent the isolines of vitrinite-like

maceral reflectance of the Alum Shale modified after Buchardt et al. (1997). Ages of the samples

are given in coloured circles. The red star denotes the well for 1-D basin modeling.

Fig. 4.2. The correlations between uranium contents and key Rock-Eval parameters. (a) Different

kerogen types can be identified based on the pseudo-van Krevelen diagram (Espitalie et al.,

1977). HI and OI are poorly correlated with uranium contents. (b) Tmax values of samples

from two boreholes are inversely proportional to their uranium contents in general.

Fig. 4.3. The pyrolysis- and thermovaporisation- GC traces of three Alum Shale samples with very

different uranium contents. n-alkenes and n-alkanes are named by carbon numbers, and major

aromatic compounds are illustrated. (a) The pyrolysates show increasing gas/oil ratio and

aromaticity in the products with increasing uranium contents from the top to bottom. (b) The

Tvap products is featured by the absence of alkenes compared with Py-GC, but they still show

the same trend of compositional changes in response to uranium contents as revealed by

pyrolysates.

XVII

Fig. 4.4. Correlations between uranium contents with compositional information derived from pyrolysis-

GC (a and b) and Tvap-GC (c and d). The gas percentage in (a) and o-xylene percentage in (b)

are two end members of two classical ternary diagrams as shown in Fig. 5 which are instructive

to organic facies. Gas/oil ratio in (c) is calculated from gas/resolved oil in Tvap which reflects

the gas richness as (a) does. Since the 2,3-dimethylthiophene concentration in Tvap is low and

can’t be accurately interpreted, only o-xylene and n-nonane were used in (d). Nevertheless,

both (b) and (d) represent the aromaticity of the products.

Fig. 4.5. Ternaries of the pyrolysates showing interpretations of the organic facies and kerogen

structures of the shales (Eglinton et al., 1990; Horsfield, 1989).

Fig. 4.6. Elemental class (inner circle) and compound class (outer circle) distribution pie charts of four

representative Alum Shale samples derived from ESI(-) FT-ICR MS analyses. Uranium-poor

samples (LO-9 and MCm-2) have lower oxygen contents and uranium-rich samples (LO-6 and

UCm-1) are featured by the absence of N2 compounds.

Fig. 4.7. DBE against carbon number diagrams on the N1 class of two uranium-poor (a and b) samples

and two uranium-rich ones (c and d). The size of the circles denotes the relative abundance of

each compound and the dash lines in the right part of each diagram enable comparison of the

alkylation.

Fig. 4.8. The exponential decay curve of 238U. A zoom in on the geological time scale manifest that the

decay can be roughly viewed as linearly correlated with time.

Fig. 4.9. The back-calculation of products that could be generated from sample UCm-2. Calculated

vitrinite reflectance curves is based on basin modelling work on well A23-1/88 (location in Fig.

1) from Kosakowski et al. (2010). Oil window was estimated with Ro between 0.5-1.3 %. The

gas and o-xylene percentages curves are based on the correlation curves in Fig. 4a and b,

respectively. Pyrolysates from sample LO-9 were set as the left end members of these two

curves.

Fig. 5. 1. Simplified map of the Baltic Basin and its possible oil kitchen. The thermal maturity of the Alum

Shale measured on vitrinite-like macerals (Buchardt et al., 1997). Symbols indicate the source

rock and oil samples locations.

Fig. 5.2. Simplified stratigraphy of the Lower Palaeozoic petroleum system of the Baltic Basin.

Fig. 5.3. Diagram of pristane/n-C17 vs.phytane/n-C18 with two representative GC-FID traces.

Depositional environment and biodegradation can be evaluated accordingly (Peters et al., 1999).

Fig. 5.4. Comparison of terpane and sterane traces of three representative samples. Traces shown here

were provided by GC-MS for direct visual comparison purpose. Ts/Tm ratios and sterane

biomarkers presented in table 5.1 and figures were interpreted from GC-MS/MS to ensure

better data quality.

Fig. 5.5. The distributions of triaromatic steroids (m/z=231) in three Alum Shale extracts and two typical

oil samples. C20-C21 triaromatic steroids can be detected from all samples, but some samples

show no/very low C26-C28 responses.

Fig. 5.6. (A) Principal component analysis on oil samples based on representative biomarkers and (B)

hierarchical cluster result in grouping the oils into two groups.

Fig. 5.7. Cross plots of terpane and sterane biomarkers show the maturities of source rock and oil

samples. Yellow stripes in panel B manifest the thermodynamic equilibrium intervals of the

sterane isomer. Diasterane/sterane radio is calculated from [total C27 to C29 β, α 20S+20R

diasterane]/ [total C27 to C29 α,1β,1β and α,α,α 20S+20R] steranes (Peters et al., 1990). Ts/Tm is

the ratio of C27-trisnorneohopane over C27-trisnorhopane (Seifert and Moldowan, 1978).

Sterane maturity biomarkers are calculated from C29 sterane epimer ratios as described by

Mackenzie et al. (1980).

XVIII

Fig 5.8. The terpane biomarker cross-plot can efficiently separate Llandovery source rocks from the rest.

C24 TeT and C26 TT are abbreviations for C24 tetracyclic terpane and C26 tricyclic terpane

respectively. The extended tricyclic terpane ratio (ETR) (Holba et al., 2001) is calculated from

(C28+C29)/(C28+C29+Ts) in m/z191. The size of the dot demonstrates the maturity of each

sample to evaluate the maturity dependency of the biomarkers.

Fig. 5.9. Ternary diagram showing the relative distribution of C27, C28 and C29 iso-steranes [5α,14β,17β(H)

20S+20R]. The Llandovery shale extracts and oil sample Gec are featured by high

concentrations in C29 iso-sterane.

XIX

LIST OF TABLES

Table 1.1 Uranium Deposit Classification (OECD/NEA and IAEA, 2009)

Table 2.1. Rock-Eval and TOC data.

Table 2.2. Detailed information about default Kerogen-Oil-Gas kinetics models shown in Fig. 2.17.

Table 3.1. Generalized information and Rock-Eval & TOC data of the samples tested in the research.

Table 4.1. Background information, uranium contents, and Rock-Eval & TOC data of the Alum Shale

samples.

Table 4.2. Py-GC and Tvap data of 16 Alum Shale samples. The ternary end members are normalized as

described by Horsfield (1989) and Eglinton et al. (1990). GOR in Tvap was calculated from gas

over resolved oil fractions.

XXI

LIST OF ABBREVIATIONS

A Frequency factor

CPI Carbon Preference Index

Ea Activation energy

ETR Extended Tricyclic-terpane Ratios

FID Flame Ionization Detector

FT-ICR Fourier Transform Ion Cyclotron Resonance

GC Gas Chromatography

GOR Gas to Oil Ratio

HC Hydrocarbon

HCA Hierarchical Cluster Analysis

HI Hydrogen Index, HI = S2/TOC × 100 (mg HC/g TOC)

MPLC Medium Pressure Liquid Chromatography

MS Mass Spectrometry

MSSV Micro Scaled Sealed Vessel

m/z mass to charge ratio

NSO Nitrogen, Sulfur and Oxygen

OI Oxygen Index, OI = S3/TOC × 100 (mg CO2/g TOC)

OEP Odd to Even Predominance

PCA Principal Component Analysis

PI Production Index

Py Pyrolysis

Pr/Ph Pristane/ Phytane

S1 Amount of Volatized Hydrocarbons at Rock Eval

S2 Amount of Generated Hydrocarbons during Pyrolysis at Rock Eval

Tmax Temperature of maximum Pyrolysis Yield

TOC Total Organic Carbon

TR Transformation Ratio

Ts/Tm C27-trisnorneohopane/C27-trisnorhopane

Tvap Thermovaporisation Gas Chromatography

Ro Vitrinite Reflection

XXII

1. INTRODUCTION

1.1 Development of Petroleum Formation Theories

Petroleum is a collective term for any subsurface material that can be produced as oil or gas

(including some associated non-hydrocarbons) (Mackenzie et al., 1988). The history of

petroleum utilisation can be traced back four thousand years to when asphalt was used for

construction in Babylon (Al-Sammerrai et al., 1987). The earliest known oil well was drilled

in China using bamboo in 347 AD (Groysman, 2014). The modern history of petroleum

began in the 19th century with the drilling of exploration wells in Pennsylvania and Poland,

and refining of crude oil to obtain paraffins. Later on, the invention of the internal

combustion engine was the major influence in the rise of the petroleum industry. While the

enormous economic impact of petroleum is self-evident, its origin has been a topic of great

and ongoing debate.

1.1.1 Biogenic vs. Abiogenic

It was during the renaissance that the first reasonable theories about the origin of

petroleum were developed. In 1546, Georgius Agricola, a German physician who coined

the term “petroleum”, proposed that bitumen is condensed from sulphur (Walters, 2006).

Andreas Libavius, another German physician, hypothesised in 1597 that bitumen formed

from the resins of ancient trees. These early discussions mark the beginnings of one of the

longest running scientific debates: whether petroleum is formed by abiogenic processes

that occur deep within the Earth, or from sedimentary organic matter derived from once

living organisms.

As fossil evidences emerged during the 18th century that coals were derived from plant

remains, Mikhailo Lomonosov proposed petroleum was formed from coal through

underground heat and pressure in 1763 (Hedberg, 1964). Modern theories that petroleum

originated from organic-rich rocks, and not necessarily coal, emerged during the 19th

century (Hunt, 1863). Meanwhile, the famous Russian scientist Mendeleev (1877) proposed

that petroleum was created in the depths of the Earth from chemical reactions between

water and iron carbides in the hot upper mantle.

The beginning of the 20th century marks the development of modern petroleum geology.

The biogenic origin of petroleum gained major recognition among geologists (Arnold and

van Vleck Anderson, 1907; Pompeckj, 1901), especially after Treibs (1934) discovered

porphyrin pigments in petroleum with structures originating from chlorophylls in

petroleum. Meanwhile, Fischer and Tropsch (1930) successfully synthesised long-chain

hydrocarbons using inorganic reactants, i.e., carbon monoxide and hydrogen.

Starting in the 1950s, Kudryavtsev (1951) and other subsequent publications (Rudakov,

1967) from the former Soviet Union proposed a modernised version of Mendeleev’s theory,

relying on thermodynamic equilibrium for chemical reactions that only allows spontaneous

formation of methane at high temperature and pressure, comparable to those of the upper

mantle region. This abiogenic theory was appealing because it offered an explanation for

the presence of petroleum deposits in metamorphic rocks of the basement. However, with

the development of analytical techniques, overwhelming evidence strongly argues that

petroleum is actually originated from biologically derived organic source materials, and that

therefore the oil stored in basement rocks was allochthonous, ostensibly because it had

simply migrated from overlying shale layers. (1) Oakwood et al. (1952) showed that oils

retain fractions are optically active, just like biological matter; (2) stable carbon isotope

compositions in oil were found to be in line with a biogenic origin (Craig, 1953) as living

organisms favour certain carbon isotopes more than others; (3) biomarkers, e.g., steranes

and terpanes, found in the petroleum can be traced to their biological predecessors

(Eglinton and Calvin, 1967), some of which can even correlate oil to specific geological

ages (Hoffmann et al., 1987; Moldowan et al., 1994); (4) hydrocarbons resembling

petroleum can be formed through thermal cleavage of kerogen (Horsfield et al., 1989).

However, the fundamental dispute continues. The astronomer Gold (1985) believes that

there is a huge amount of primordial methane within the Earth since hydrocarbons were

found in chondrites and other planetary bodies, including asteroids, comets, and moons.

However, the utter failure of two deep wells drilled in the Siljan Ring (Sweden) where was

supposed to be an ideal place to gain inorganic sourced gas (Gold and Soter, 1982) made

the abiogenic origin theory remain unproven and less plausible. Even today, people who

believe that petroleum has migrated upward from the mantle have not been deterred by

this setback. Kenney et al. (2001) who was the drilling manager of the Sijan Ring wells

claimed that: “natural petroleum has no connection with biological matter”. High-

temperature and high-pressure experiments, simulating mantle environment, are still

running in Russian labs to synthesise inorganic origin petroleum (Belonoshko et al., 2015;

Kolesnikov et al., 2009).

In summary, the abiogenic origin theory is mainly based on theoretical studies and

laboratory verifications. It is ironic that this theory is advocating that the mass of

petroleum is essentially infinite on Earth (Kolesnikov et al., 2009), yet only traceable

amount of abiotic oil have been discovered (Walters, 2006). Geochemists do not deny the

existence of a very small amount of abiogenic hydrocarbons on Earth. Nevertheless, it is

irrefutable that over 99.99% of the petroleum has been found in sedimentary basins and all

commercial findings are guided by the modern biogenic origin theory.

1.1.2 Deep vs. Shallow

Although most geologists agree that petroleum was formed from organic matter, there

were debates within the framework of biogenic theory from the 1940s to 1960s: is the

petroleum formed at shallow depths and then preserved, or is burial an essential element

(roughly 1-5 km)? In other words, how and when are oil and gas generated from organic

matter?

During the middle and late 1940s, the idea of shallow origin and migration of oil began to

gain favour. Expansion of modern sediment studies reflected the importance attached to

depositional processes and early diagenesis in understanding geologic processes. Zobell

(1945) found that bacteria can produce methane and heavier hydrocarbons which resemble

petroleum and thus concluded petroleum was formed in shallow areas with low

temperatures. (Pratt, 1947) proposed that petroleum accumulation is completed soon after

deposition of the source beds based on the occurrence of oil in continental shelves. The

discovery of hydrocarbons in recent sediments from the Gulf of Mexico gave support to

the shallow generation theory (Smith, 1952). Corbett (1955) suggested that humic material

may be washed by meteoric water into shallow sandstone reservoirs where transformation

to oil take place. Baker (1959) and Meinschein (1961) proposed that there is no chemical

reaction involved in the petroleum formation, oil and gas being hydrocarbon-rich fluids

that were selectively washed out from modern sediments by solubilizers-containing water.

The counterargument was that petroleum was formed through the thermal cracking of

organic matter and thus requires a considerable depth to reach the temperature. (Gussow,

1954) emphasised the concept that oil globules could only migrate out of the source beds

into reservoirs once overburden pressures are sufficiently high as to overcome capillary

pressures. Emery and Hoggan (1958) realised that the hydrocarbons in modern sediment

are compositionally significantly different from those in oil fields by virtue of their very

high methane percentage and the absence of gasoline hydrocarbons. Bray and Evans (1961)

reported the Carbon Preference Index (CPI) in crude oil resembles that of ancient

sedimentary rocks and are far lower than CPI values in modern sediments. Shimoyama and

Johns (1971) further proved that the odd to even predominance (OEP) n-fatty acids

decreases from high values to near unity from modern to ancient sediments and petroleum

reservoir water. Philippi (1965) suggested that petroleum is generated at depths where the

subsurface temperature is above 115°C.

After the 1970s, the debate gradually settled down and a broad consensus was reached: in

shallow depth, dry gas can be generated through bacterial and microbial mechanisms, while

oil and gas would be formed in deeper areas by the thermal degradation of organic

precursor molecules in sediments. Carbon isotope acts an efficient tool to differentiate gas

generated through different mechanisms (Schoell, 1980; Stahl and Carey, 1975). The gas

formed by microbial activity was termed biogenic gas. Favourable conditions for biogenic gas

generation include (1) sequential elimination of oxygen and the electron acceptor sulphate

from the sediment, (2) moderate temperature (<75 oC), and (3) sufficient space available

for the bacterial bodies (Rice and Claypool, 1981). Most of the biogenic gas was generated

from dead biomass which contains degradable biomolecules, e.g., proteins and lipids, but

new research manifests that it can also be formed from long-chain alkanes by anaerobic

microorganisms (Zengler et al., 1999). Shale gas found in the Devonian Antrim Shale (USA)

acts as a successful example of exploiting biogenic gas (Curtis, 2002). In comparison,

thermogenic oil and gas require a much deeper burial history to achieve the temperatures

needed for cracking the organic matter. Thermal cracking of kerogen via free radicals (Rice,

1931) or the acidic catalyst related cracking (Whitmore, 1934) serve as the two main

mechanisms of the generation of thermogenic petroleum. Most of the petroleum found in

oil field reservoirs is formed by a thermogenic mechanism.

In addition to these two “orthodox” mechanisms, Mango (Mango, 1990, 1992, 1994, 1997,

2010) advocated the formation of light hydrocarbons as result of catalysis of transition

metals. He suggested that transition metals (e.g., V, Ni, Ti, Co, and Fe) are mainly captured

from sedimentary waters by the tetrapyrrole nucleus of chlorophyll and act in a steady-state

reaction to form light hydrocarbons. This theory has drawn the attentions of geologists for

at least two reasons. (1) The gas wetness of pyrolysates in all thermal cracking-based

pyrolysis experiments is much higher than reservoired gas (Behar et al., 1992; Horsfield et

al., 1989; Lewan et al., 1979), while the metal catalysed reaction is more consistent with the

observations of producing gas wetness. (2) Some light isoparaffins remain virtually

invariant through the course of oil generation (Mango, 1987) which can hardly be explained

by the routine cracking theory. The invariance was interpreted to be related to cyclopropyl

intermediates formed by metal catalysts (Mango, 1990) and was supported by kinetic

models based on these ring closure reactions (van Duin and Larter, 1997; Xiao and James,

1997). Even though the metal catalysis hypothesis appears to explain some “contradictions”

of the classical thermal cracking theory, counterarguments on this theory are based on

many aspects. (1) It is suggested that gas wetness in reservoirs described by Mango does

not necessarily represent the gas composition when the gas was generated because

fractionation during migration would change it significantly. Price and Schoell (1995)

reported that in the Bakken Shale, which serves as a closed system of both source and

reservoir, associated methane accounts for ∼45 wt.%, which is consistent with laboratory

simulations of thermal cleavage. Snowdon (2001) advocated that gas wetness of cutting

samples gives a better representative of the naturally generated gas wetness rather than

conventional reservoir production data. (2) Hydrous pyrolysis experiments delivered

evidence revealing that transition metals have no effect on methane enrichment or δ13C

changes (Lewan et al., 2008). Thus the metal catalysis theory would not be a realistic option

if the source rock is initially water-saturated. (3) The contact of kerogen with transition

metals in natural environments is very limited, and NiO normally used as a catalyst in

Mango’s experiments (Mango, 2007; Mango and Hightower, 1997) is very rare in source

rocks. (4) Furthermore, some assumptions in Mango’s theory appear to be questionable.

For examples, olefins are important intermediates during catalytic reactions (Mango, 1987;

Mango, 1994) and it was suggested that petroleum should be at least three orders of

magnitude more stable than the kerogen precursors under catagenic conditions (Mango,

1991), but olefins are rarely found in natural petroleum (Curiale and Frolov, 1998) and

petroleum in natural reservoirs can be secondarily cracked between 160 oC-190 oC

(Horsfield et al., 1992c). Nevertheless, the transition metal catalysis theory provides the

best explanation for the isoparaffin invariances and acts as an alternative petroleum

generation pathway despite being a general matter of debate.

1.1.3 Fatty Acid vs. Kerogen

Another matter of contention was related to what kind of organic matters are the

precursors of petroleum? Fatty acids and kerogen which is insoluble in common chemical

solvents are the two candidates.

Much early pyrolysis work was concentrated on molecules soluble in organic solvents,

especially fatty acids. In the 1960s, Cooper and Bray (1963) and Jurg and Eisma (1964)

generated petroleum-like hydrocarbons, with a CPI around unity, through heating fatty

acids. Kvenvolden (1966) speculated hydrocarbons were formed through decarboxylation

of fatty acids. Welte and Waples (1973) and Douglas et al. (1975) suggested dehydration

and reduction of fatty alcohols can also lead the formation of alkanes from fatty acids.

Based on these extensive experiments, it is no wonder that the fatty acids were considered

as an important precursor of petroleum.

Abelson (1963) pointed out fatty acids would disappear rapidly in geological environments

and only kerogen can be considered capable of quantitatively accounting for the petroleum

found in reservoirs. It comprises by far the largest organic matter pool on Earth, i.e., 1014

tons of carbon in kerogen compared to ca. 1012 tons in living biomass (Welte, 1970). Later

research found that the kerogen is formed from proteins, carbohydrates, lipids, and lignin

through degradation-recondensation (Tissot and Welte, 1984) or by selective preservation

(Tegelaar et al., 1989). It is now firmly established that petroleum is primarily formed from

the maturation, over time and with burial at elevated temperatures, of kerogen.

1.2 Characterisation of Kerogen Structure

The kerogen structure and the maturation process it experienced are the initial factors that

control the thermogenic petroleum composition, followed by secondary processes like

migration fractionation, phase separation, biodegradation, water washing, etc. The

structural features of kerogen describe the elemental constitution, chemical bond

connection, and the stability of the structures, and thus define the quality, quantity, and

kinetics of petroleum generation.

1.2.1 Generation quality and quantity

1.2.1.1 Elemental analysis and Rock-Eval

Pioneering studies by Down and Himus (1941) attributed kerogen compositional

differences to variation in plant sources, depositional environment and bacterial reworking.

Now it is widely accepted that four types of kerogen can be recognised by elemental

analysis (Durand and Espitalié, 1973; Van Krevelen, 1950) and Rock-Eval pyrolysis

(Espitalie et al., 1977). These kerogen types can provide information not only about past

environments and biota, but also petroleum generation characteristics.

Source rocks deposited in anoxic lakes or anoxic shallow marine basins tend to contain a

very hydrogen-rich kerogen derived from plankton (type I or II) and are oil-prone. Those

deposited in fluvial and deltaic facies usually contain hydrogen-leaner kerogen (type III)

derived from higher land plants and predominantly generate gas-rich products or high-wax

oils. Type IV kerogen which is dominated by inertinite has a very little petroleum

generation potential. Although such general correlations between depositional

environments, kerogen type and hydrocarbon generation are fulfilled in most scenarios,

exceptions exist. The generation potential of source rocks is reflected by the Hydrogen

Index (HI), this being the discriminator for kerogen type recognition. Type I (HI>600

mgHC/gTOC) and type II (200 mg HC/gTOC<HI<600 mg HC/gTOC) contain great

hydrocarbon generation abilities and sourced most of the commercial petroleum reservoirs

in the world. Type III kerogen which has low HI enabled only limited petroleum

discoveries. Nevertheless, gas fields in Mid-European Basin(Lutz et al., 1975), waxy crude

oils in Asia (Peters et al., 1999), and coal bed methane (Law, 1988) are examples of

petroleum generated by type III kerogen.

1.2.1.2 Pyrolysis-GC

Although the gross hydrocarbon generation ability and general oil vs. gas preference can be

estimated through calibrated kerogen typing, little exploration-oriented information can be

derived, e.g., GOR, aromaticity, and sulphur contents. Pyrolysis, defined as a chemical

degradation reaction that is induced by thermal energy alone (Ericsson and Lattimer, 1989),

when interfaced with GC, is able to provide compositional information on a molecular

level.

Kerogens degrade upon pyrolysis to yield many compound types including hydrocarbons,

ketones, alcohols, nitriles and thiols, as represented by cyclic and acyclic, saturated and

unsaturated carbon skeletons (Dembicki et al., 1983; Van de Meent et al., 1980). Of these

the most commonly occurring major identifiable components seen by pyrolysis-GC are

doublets of normal alk-1-enes and alkanes, alkylphenols, alkylbenzenes, alkylnaphthalenes

and alkylthiophenes. It was proven that the structural information contained within the

readily identifiable and major components of kerogen pyrolysis products are coincident

with aromaticity data gained by 13C-NMR which is a non-destructive technique (Horsfield,

1989). It’s now widely accepted that the abundances and distributions of resolved pyrolysis

products give clues as to the bulk compositions of natural petroleums, such as paraffinicity

and aromaticity (Damsté et al., 1993; Espitalie et al., 1988). In according, not only the

aromaticity (Larter, 1984), and sulphur contents (di Primio and Horsfield, 1996; Eglinton et

al., 1990) of the petroleum which could be generated from the source rock can be inferred

from pyrolysates, the type of petroleum can also be predicted. For example, five major

types of petroleum can be predicted according to the kerogen pyrolysate length

distributions, namely, Gas and Condensate, Paraffinic-Naphthenic-Aromatic oil with

high/low wax content, and Paraffinic Oil with high/low wax content (Horsfield, 1989). At

the same time, these kerogens can be related in a general way to terrestrial, deltaic, marine

and lacustrine depositional environments (Horsfield, 1997).

1.2.2 Kinetics

The kinetics of a source rock defines its thermal reaction to heating time, temperature

and/or pressure. Bulk kinetics which describes the generation of the primary products as a

whole are well established (Braun and Burnham, 1987; Quigley et al., 1987). More advanced

kinetic models are also available. For examples, secondary kinetics of source rocks (Behar

et al., 1997; Pepper and Dodd, 1995; Ungerer et al., 1988), secondary kinetics in reservoirs

(Horsfield et al., 1992c; Waples, 2000) are especially important in evaluating areas

experienced high maturation processes and in the exploration of shale gas. Compositional

kinetics (Behar et al., 1992; di Primio and Horsfield, 2006; Dieckmann et al., 2000a;

Düppenbecker and Horsfield, 1990) provide much more detailed information in simulating

the generation of petroleum.

Quasi-first-order reactions are generally applied for the investigation of kinetics and the

result is typically shown as a variable frequency factor and a distribution of activation

energy (Burnham et al., 1987; Quigley et al., 1987), in which the frequency factor is related

to the vibration frequency of the reaction. Although a compensation effect (Constable,

1925) occurs in the kinetics results which shows a clear correlation between frequency

factor and average activation energy (Lakshmanan et al., 1991; Stainforth, 2009) the most

intuitive way to compare kinetic features of different samples is to extrapolate the results to

a typical geological heating rate (2-4 oC/Ma) and to compare the transformation rate curves

(Schenk et al., 1997; Tegelaar and Noble, 1994).

The stabilities of chemical bonds and kerogen heterogeneities together define the

frequency factor and activation energy distributions. Thus the kinetic characteristics of

source rocks, especially the values and the shapes of the activation energy distributions, are

initially controlled by the organic facies (Pepper and Dodd, 1995). Typically, type I kerogen

tends to have a very concentrated activation energy distribution due to the homogenous

distribution of C-C bonds and the heterogeneous type III kerogen is normally featured by a

very broad activation energy distribution (Ungerer and Pelet, 1987). The marine type II

kerogen generally shows an intermediate distribution, but it can be variable according to

depositional environment and/or precursors. For examples, the enrichment of

Gloeocapsamorpha Prisca in an Ordovician marine shale would focus the activation energy

distribution as for a type I kerogen (Waples, 2010) and the kinetics can be very active when

the shale is sulphur-rich (Tegelaar and Noble, 1994). The kinetic features of marine shale

also change with sedimentary facies and geological age of the samples (Mahlstedt, 2012;

Waples and Marzi, 1998).

1.3 Influence from Inorganic Materials

As stated above, pyrolysis plays a crucially important role in evaluating kerogen structures

and providing practical information for petroleum exploration. The fact that some

inorganic materials also influence pyrolysate compositions, sometimes overprinting the

initial control of pyrolysates from kerogen structure, raises the question as to whether

organic-inorganic reactions control petroleum yields and compositions in active source

rocks. Similarly, the presence or absence of water influences pyrolysate characteristics. In

addition to these pyrolysis-inspired lines of research, the role of radioactive bombardment

on source rock generating characteristics is still under investigation.

1.3.1 Minerals

1.3.1.1 Mineral and petroleum occurrence

Quartz, clay, and calcite are generally the main minerals in shale (Shaw and Weaver, 1965).

Other minor constitutes include feldspar, pyrites, phosphate, mica, and siderite. For

example, the Mississippian Barnett Shale has average quartz, clay, and calcite contents of

34.3%, 24.2%, and 16.1%, respectively (Loucks and Ruppel, 2007), while the Cretaceous

Eagle Ford Shale is typically featured by a carbonate content over 60% and relatively low

quartz concentration (Mullen, 2010).

The significant concomitance of clay minerals with oil production has drawn the attention

of the importance of mineral in searching for petroleum (Frost, 1945; Grim, 1947). For

example, Weaver (1960) found oil production in different geological periods is statistically

correlated to the montmorillonite proportion of layers developed in the periods although

carbonate source rocks were not taken into consideration at that time. In general, clay

minerals have been believed to play important roles in kerogen preservation and the

catalytic formation of petroleum. Furthermore, the process of illitization is also related to

oil generation and expulsion in time and space (Burst, 1969).

Degradation-recondensation (Tissot and Welte, 1984) and selective preservation (Tegelaar

et al., 1989) are classical pathways in interpreting kerogen formation, while natural

sulphurization (Sinninghe Damsté et al., 1989) and clay sorptive protection (Salmon et al.,

2000) are also important supplementary mechanisms. Clay minerals, acting as natural

adsorbents in sedimentary systems, can adsorb and then protect amorphous, labile and

dissolved organic components (e.g., amino acids and simple sugars) from complete

microbial degradation. As a result, the presence of clay minerals might play a significant

role in organic matter accumulation and the subsequent concentration reactions to kerogen

in sedimentary rocks (Wu et al., 2012). This hypothesis is supported by the fact that TOC is

inversely proportional to the grain size in most continental shelf sediments (Mayer, 1994)

which was ascribed to a monolayer adsorption of organic compounds onto minerals

(Salmon et al., 2000).

Besides the thermal cracking of kerogen via free radical in generating petroleum (Rice,

1931), the acid catalytic cracking mechanism via carbonium intermediates is an alternative

(Whitmore, 1934). The carbonium-ions are primarily generated on the Brønsted and Lewis

acid sites of clay minerals. In chemistry, the Brønsted acid is viewed as a proton donor and

the Lewis acid can act as an electron acceptor. The acid catalytic mechanism affects

isomerization, polymerization, and disproportion of hydrogen (Frost, 1945; Kissin, 1987),

and might, therefore, be responsible for the presence of aromatic hydrocarbons and the

absence of olefins in natural petroleum (Brooks, 1948).

Smectite can be transformed into illite in the illite/smectite (I/S) mixed-layer during

diagenesis and catagenesis stages (Burst, 1969; Powers, 1967). The illitization process can

be related to petroleum generation and expulsion. For examples, smectite alters to illite at a

temperature between 80 to 120 oC (Burst, 1969) which corresponds to the oil generation

peak in the same temperature range. The illite percentage in the I/S layer has a good

correlation with the Tmax value (Burtner and Warner, 1986) and is applied in evaluating

petroleum generation stages (Foscolos et al., 1976; Lindgreen and Drits, 2000). Illitization

is accompanied with dehydration from interlayers and the release of pore water (Perry Jr

and Hower, 1972). The expelled water may create overpressure (Powers, 1967) and serves

as primary migration driving force (Bruce, 1984), although Osborne and Swarbrick (1997)

counterargued that the small water volume released during illitization is not sufficient to

create overpressure.

Petroleum generated from source rocks with different mineralogical compositions can vary

in many aspects. For example, oil originated from carbonate source rocks are generally

heavier, richer in sulphur more naphthenic, and with a lower Carbon Preference Index

(CPI) than petroleum generated from shale (Hughes, 1984; Jones, 1984). Furthermore, at a

molecular level, pristane/phytane, diasterane/sterane, and Ts/Tm ratios are higher in shale

samples compared with carbonate rocks at equivalent thermal maturities (Didyk, 1978;

McKirdy et al., 1981; Rubinstein et al., 1975). It has to be pointed out that although these

differences are superficially related to the mineralogical composition, the redox conditions,

precursors as well as other factors during deposition are also important to explain

compositional variations of oil sourced from carbonate and shale.

1.3.1.2 Mineral matrix effect (MME)

Espitalie et al. (1980) and Horsfield and Douglas (1980) firstly reported the influence of

minerals on Rock-Eval and pyrolysis gas chromatography (Py-GC) products, respectively.

The so-called mineral matrix effect can cause changes in HI, Oxygen Index (OI) and Tmax

in the Rock-Eval analysis, as well as gas-oil ratio and aromaticity in the case of Py-GC, and

even can influence the hydrocarbon generation kinetics (Dembicki, 1992).

Espitalie et al. (1980) found that illite and montmorillonite can adsorb the generated heavy

hydrocarbons in open system pyrolysis and hinder elution, thus leading to shifts of the Tmax

values toward high temperatures and decreases in S1, S2 and HI values. Tarafa et al. (1983)

further reported that the retaining mechanism could also delay some hydrocarbons from S1

into S2 and therefore cause an erroneous Production Index (PI). Besides the physical

adsorption, minerals also have selective catalytic impacts on the pyrolysates. For example,

CO2 generation could be enhanced by the existence of carbonate minerals and may lead to

an increase of OI values (Katz, 1983). In brief, when MME is an efficient factor, then the

whole rock is always lower in HI and higher in OI compared with its kerogen counterpart

after demineralisation. Calibrated by Van Krevelen diagram based on elemental data, it was

found that kerogen data plotted in the pseudo-Van Krevelen diagram are more reliable in

kerogen type identification than whole rock data (Katz, 1983).

Detailed compositional information revealed by Py-GC and Py-GC-MS also demonstrates

that the MME exists in laboratory pyrolytic experiments. In general, montmorillonite and

kaolinite are catalysts to generate more gas-rich and more aromatic hydrocarbons from

kerogen (Dembicki et al., 1983; Horsfield and Douglas, 1980), indicating the activity of

Lewis acids. Tannenbaum and Kaplan (1985b) further reported that branched, alicyclic, and

aromatic hydrocarbons in the pyrolysates are consistently higher in the presence of

montmorillonite compared with any other minerals can be attributed to the carbonium-ion

cracking mechanism. The release of aliphatic biomarkers (Lu et al., 1989) and diamondoids

(Wei et al., 2006) through pyrolysis can also be enhanced by the presence of

montmorillonite. By comparison, the catalytic effect of illite and calcite on hydrocarbon

generation are negligible (Tannenbaum et al., 1986a). Recently, Lewan et al. (2014) found

that bitumen would enter smectite interlayers before illitization and undergoes cross-linking

to form pyrobitumen instead of thermally cracking into oil, and hence leads to a decrease in

petroleum generation rates and expulsion efficiencies.

The MME on kinetic effects was discussed by Dembicki (1992) who determined bulk

kinetic data on mixtures of kerogens and different minerals. However, only activation

energy distributions were compared leaving out the variable frequency factors, this leads to

critical reviews by Pelet (1994) and Burnham (1994b). A reply to these two comments from

Dembicki (1994), after extrapolating the kinetic data to a simplified geological situation,

showed that quartz and calcite can hinder kerogen transformation in contrast to bentonite.

A similar trend was reported by Reynolds and Burnham (1995), but the degree of the

MME on kinetics is concluded to be minor when modelling petroleum generation.

In summary, clay minerals are especially important in simulating MME, and two main

mechanisms are discussed. (1) The high surface area of clay minerals (Sing, 1985) leads to

adsorption of heavy compounds generated during pyrolysis, and hence influences Tmax,

GOR, S2 and HI, as well as the kinetics of the reactions. The adsorption ability of different

minerals generally decreases in the following order: illite > montmorillonite > calcite > and

kaolinite (Espitalie et al., 1980). (2) The selective catalysis induced by the active acid cites

leads to a carbonium-ion cracking and influences the isomerization and aromatization

processes. Thus the hydrocarbon composition and biomarker generation can be changed.

The catalytic ability of minerals decreases in the order of montmorillonite > kaolinite >

calcite > illite (Hu et al., 2014).

It has to be pointed out that not every source rock necessarily undergoes MMEs in

pyrolysis experiments. Horsfield and Douglas (1980) and Katz (1983) concluded that the

MME varies according to the TOC content and mineralogical composition of the samples.

A very high TOC content (> 6%) or a low clay content can decrease or even avoid the

MME (Peters, 1986). The presence of water is supposed to decrease the contact between

organic matter and minerals, and thus significantly attenuates the activity of the clay

catalysing function (Huizinga et al., 1987; Lewan et al., 2014; Pan et al., 2007; Tannenbaum

and Kaplan, 1985a).

1.3.1.3 Water influence on MME

Tannenbaum and Kaplan (1985a) and Pan et al. (2007) emphasized that the presence of

liquid water in the pyrolysis system could significantly attenuate the activity of the clay-

catalyzing function, counterarguments from Eglinton et al. (1986) and Behar et al. (2010)

noted that the function of water in pyrolysis is very limited.

The importance of water in petroleum generation was highlighted by Jurg and Eisma (1964)

who achieved petroleum-like products by heating fatty acid with and without water. Lewan

et al. (1979) claimed that hydrous pyrolysates are more resemble to natural petroleum

compared with products from other pyrolytic techniques because they basically contain no

alkene, less aromatic and polar compounds and more liquid fractions than anhydrous

pyrolysates (Lewan et al., 1985). It is suggested that water acts as an exogenous source of

hydrogen (Hoering, 1984) and reduces the rate of decomposition, promotes thermal

cracking and inhibits carbon-carbon bond cross-linking (Lewan, 1997).

Hydrous pyrolysis was also applied in bulk kinetic research (Hunt et al., 1991; Lewan and

Ruble, 2002), but the results are very different from the well-established anhydrous kinetic

models, e.g., the hydrous-kinetic parameters result in very narrow oil windows with

significant heterogeneities among samples (Burnham, 2015; Lewan and Ruble, 2002). The

samples used in hydrous pyrolysis experiments are typically 0.2-2 cm in length (Lewan et al.,

1985; Lewan et al., 2014) which are significantly coarse-grained and larger than finely

powdered samples (micrometre level) used in anhydrous system kinetic tests. This would

increase the temperature errors caused by thermal transients and hinder the efficient

expulsion of generated products (Peters, 1986; Stainforth, 2009). Expulsion efficiency

research would benefit from this big samples size (Lewan et al., 2014), but great errors

would be induced in kinetic research which requires accurate temperature determinations.

Furthermore, the gas/oil ratios (GOR) of type I, II, II-S and III kerogens through hydrous

pyrolysis reveals that GOR initially decreases with increasing temperatures (Lewan and

Henry, 1999) which is inconsistent with geological observations (Hunt, 1996).

The influence of water on MME in petroleum generation is not further discussed in this

dissertation for the following reasons. (1) The excessive water applied in experiments does

not always match the geological situation as the water saturation in a shale layer can be very

low due to compaction and clay dehydration (Foscolos and Powell, 1979). On the other

hand, Michels et al. (1995) suggested that water is self-sufficient in shale through de-

oxygenation and can provide enough protons to saturate the generated hydrocarbons. (2)

Anhydrous close-system pyrolysis can also generate petroleum-like hydrocarbons without

alkenes (Horsfield et al., 1989; Michels et al., 1995) because a certain pressure plays a more

important role than water in removing the alkenes (Monthioux et al., 1985). (3) Most of the

routine pyrolysis related techniques are anhydrous systems, e.g., Rock-Eval (Espitalie et al.,

1977), open pyrolysis (Horsfield, 1989; Van de Meent et al., 1980), bulk kinetics (Braun and

Burnham, 1987), micro-scale sealed vessel (MSSV) (Horsfield et al., 1989), and some

golden tube pyrolysis (Behar et al., 1992). Thus, detailed investigations of MME in

anhydrous systems would make this contribution of more practical importance.

1.3.2 Uranium

1.3.2.1 The Uranium enrichment in shale

Uranium is one of the most common elements in the Earth's crust, being 40 times more

common than silver and 500 times more common than gold (Vine and Tourtelot, 1970). It

can be found almost everywhere in soil, rivers, and oceans, for example, the average

uranium contents in a river is 0.6 ppb and 3.3 ppb in sea water (Bloch, 1980). Sedimentary

rocks are estimated to have an average uranium content of four ppm (Alloway, 2013;

Swanson, 1960). In black shales the uranium concentration is higher, i.e., the average

contents are eight ppm for shales and 20 ppm for organic-rich marine shales (Swanson and

Swanson, 1961). However, high uranium concentrations can be found in some black shale,

ranging from 50 to 500 ppm. Black shale-related uranium mineralisation includes marine

and terrestrial shales, containing uranium adsorbed onto the organic material and clay

minerals. Examples include the uraniferous Alum Shale (Sweden and Estonia), the

Chattanooga shale (USA), the Chanziping shale (China), and the Gera-Ronneburg deposit

(Germany).

Uranium in black shale is generally proposed to be initially derived from sea water by

synsedimentary processes, although Anderson et al. (1989) suggested that uranium is

mainly precipitated within the sediments rather than the removal from water. Primary

mechanisms of uranium fixation and remobilization in shale are summarised here.

(1) Reducing Environment

Uranium in shale is believed to be precipitated in reducing environments, and the concepts

date back to the 1930s (Goldschmidt, 1937). This is furthermore validated by modern

sedimentology research which revealed that uranium was precipitated on the Baltic Sea

shelf where oxygen deficiency was induced by biological activity (Koczy et al., 1957).

Uranium in seawater is soluble in a oxidizing environment and its solubility is significantly

decreased in reducing environments (Durand, 2003). Therefore, a combination of uranium-

rich sea water and reducing environment can lead to uranium precipitation from the water

and enrichment in the sediments (Breger and Brown, 1962; Disnar and Sureau, 1990; Vine

and Tourtelot, 1970). Similarly, Leventhal (1991) emphasised the importance of euxinic

bottom water and slow sedimentation rate in controlling the uranium concentration in the

Alum Shale.

(2) Water Circulation

In contrast to underline the reducing environments as a controlling factor, Leckie et al.

(1990) found that the uranium-rich Shaftsbury Formation (Cretaceous) in Canada was

deposited in relatively shallow water based on palynological, micropalaeontological and

geochemical results. Schovsbo (2002) also reported that the uranium concentration in the

Alum Shale is inversely correlated to the layer thickness which implied that uranium is

more enriched in the inner-shelf instead of the outer-shelf facies. This type of uranium

concentration is explained by a higher degree of bottom water circulation resulting in

enhanced supply of uranium in the sediment/water interface where the uranium removal

from the sea water took place (Schovsbo, 2002).

(3) Humic Organic Matter.

Swanson (1960) found that humic organic matter contains far more uranium than

sapropelic type does in U.S.A shales. Thus, it was concluded that uranium in shales is

concentrated from sea water within, on, or near the humic organic matter by one or a

combination of the following processes: (1) direct precipitation of uranium, probably by

hydrogen sulfide; (2) removal of uranium ions from solution by adsorption and complexing

on solid humic material; and (3) adsorption or complexing of uranium by humic acids while

in solution. In contrast, Breger and Brown (1962) argued that uranium is only correlated

with TOC irrespective of the organic types. This is based on the fact that some marine

shales (e.g., Alum Shale) which contain little humic organic matter can also be enriched in

uranium.

(4) Hydrothermal Activity

Migration of hydrothermal fluid into sea water causes changes in temperature and pressure,

and thus leads uranium precipitation into the shale (Fisher and Bostrom, 1969; Oliver et al.,

1999). The abnormal metal composition of the Alum Shale (Oslo Region) may be a result

of interactions of submarine hydrothermal activity with an anoxic bottom water during

deposition (Berry et al., 1986). This local volcanic influence may be taken as an explanation

why shales isochronously deposited with Alum Shale in Wales, Northern America, and

South America lack such high uranium contents.

(5) Phosphorites

Phosphorite laminae or nodules in shale are normally rich in uranium, similar to the

hosting shale (Veeh et al., 1974). However, only some certain phosphorite minerals are

characterised by high uranium concentrations and point to early diagenetic and selective

fixation from the pore water (Schovsbo, 2002; Veeh et al., 1974).

(6) Remobilisation

After synsedimentary fixation in the shale, the uranium can be remobilized and locally

concentrated. For example, the Silurian graptolitic black shale in Gera-Ronneburg

(Germany) is rich in stockwork uranium which was further enriched by hydrothermal and

supergene processes (Dahlkamp, 2013; Lange and Freyhoff, 1991).

1.3.2.2 Uranium and petroleum generation

The most common isotopes in natural uranium are 238U (99.27%) and 235U (0.72%)

(Osmond and Cowart, 1976). Most of the radiation resulting from 238U decay in natural

systems will be emitted in the form of α-radiation, which has a shallow penetration depth

of <100 μm, followed by less intensive γ-radiation which penetrates several decimetres

(Jaraula et al., 2015). Alpha particles were suggested to be important in petroleum

generation for a long time (Lind and Bardwell, 1926). Experimental research has

manifested that fatty acids can be decarboxylated by alpha particle radiation at 130o to form

hydrocarbons (Sheppard and Burton, 1946). It was proposed that the radiolytic cracking of

kerogen can be an alternative maturation pathway besides the thermal and microbial

mechanisms (Jaraula et al., 2015).

The TOC content in shale is generally proportional to the uranium content (Bates, 1958;

Leventhal, 1981) and this correlation is applied while using gamma-ray spectral logging in

finding shales (Schmoker, 1981; Serra, 1983). A positive relation between oil yield and

uranium concentrations was suggested by Swanson (1960) for the Chattanooga shale (USA).

Furthermore, the effects of uranium irradiation on the organic matter are a matter of

debate for a long time, and are outlined as follows.

(1) Extractability and Biomarker

Intensive uranium radiolysis was suggested to cause hydrocarbon polymerization through a

free radical crosslinking mechanism (Charlesby, 1954), or by the gradual construction of

methane (Court et al., 2006). This process was considered to significantly change the

organic matter solubility and biomarker distribution.

Atomic pile radiation on petroleum revealed that paraffins would turn into an insoluble gel

after a certain dose of radiation is reached (Charlesby, 1954). In Alum Shale, the bitumen

extractability and high molecular (C26-C28) triaromatic biomarkers are reversely correlated

with the uranium content (Dahl et al., 1988a, b; Lewan and Buchardt, 1989). Hoering and

Navale (1987) speculated that the absence of biomarkers in some Alum Shale horizons was

caused by irradiation damage. Lewan and Buchardt (1989) concluded that the irradiation-

induced crosslinking would convert the single hydrocarbons into insoluble complexes and

that the longer side chains of the triaromatic steroids are more susceptible to the

polymerization.

(2) Pyrolysates

The ratio of aromatic compounds (toluene and naphthalene) to n-alkanes during pyrolysis

of the Chattanooga shale was reported to be positively correlated to the uranium content

by Leventhal (1981). Similar results from Horsfield et al. (1992a) revealed that the uranium-

rich Alum Shale tends to generate very gas-rich and aromatic products and that “dead

carbon ” get enhanced with increasing maturation, findings that were further confirmed by

Bharati et al. (1995) and Sanei et al. (2014). However, the Alum Shale is still a key target to

answering questions about the influence of uranium irradiation on organic matter in

sediments, and recent investigations showed that the uranium concentration in Alum Shale

is proportional to the quantity of gas generation and inversely correlated to oil generation

(Kotarba et al., 2014a). Other Py-GC-MS studies demonstrated that uranium-rich bitumen

tends to generate less diverse hydrocarbons with smaller and less alkylated PAHs compared

with samples with lower uranium contents (Court et al., 2006).

(3) Element Ratios

The atomic H/C and O/C ratios which are proxies to define the kerogen type and thermal

maturity (Durand and Espitalié, 1973) are also susceptible to uranium irradiation, i.e., H/C

is decreased and O/C is increased when the uranium content in the shale is high (Pierce

et al., 1958). This general trend revealed by elemental analyses is further supported by other

techniques including Rock-Eval (Landais, 1996; Leventhal et al., 1986) and 13C NMR

spectroscopy (Bharati et al., 1995). The liberation of hydrogen by uranium ionising

radiation (Colombo et al., 1964; Dole, 1958) is one explanation for the H/C decrease (Dahl

et al., 1988b). The increased oxygen content is presumably derived from hydroxyl radicals,

formed by the radiolysis of water (Court et al., 2006).

(4) Stable Carbon Isotopes

Leventhal and Threlkeld (1978) firstly reported that the 13C/12C ratios of Upper Jurassic

Morrison Formation samples are correlated to the log of the U concentrations. This was

attributed to a preferential loss of isotopically light volatiles, formed by radiolysis of the

organic matter, leaving a 13C-enriched residue. Further investigations of shale and bitumen

samples also confirmed this irradiation effect on stable carbon isotopes (Court et al., 2006;

Dahl et al., 1988b).

(5) Maturity Indicators

The vitrinite reflectance of humic coal can be enhanced by the crosslinking caused by

uranium irradiation irrespective to the geological heating (Breger, 1974). Tmax was also

reported to be shifted to higher values when uranium contents are high (Forbes et al., 1988;

Landais, 1996). In contrast, Dahl et al. (1988b) reported Tmax values decrease with

increasing uranium contents. Today, it is widely accepted that decarboxylation occurs

during irradiation at low temperatures leading to higher values for maturity indicators

(Jaraula et al., 2015; Landais, 1993).

(6) Olefins

The unsaturated olefins are thermally unstable in geological environments. They typically

contribute less than 1% to crude oil but can be as high as 10% in some Siberian oils

(Curiale and Frolov, 1998). The high temperature of hydrothermal activity can generate

olefins (Simoneit and Lonsdale, 1982) which is similar to anhydrous pyrolysis experiments

in the laboratory (Horsfield, 1989). Alpha radiolytic dehydration is another main process to

cause the occurrence of olefins in nature (Frolov and Smirnov, 1994). The hypothesis

behind this phenomenon is that all C-C n-alkane bonds are similarly sensitive for the

dehydration to n-alkenes in response to alpha particle radiation (Frolov et al., 1996).

1.4 Research Perspectives and Objectives

The fundamentals of petroleum geochemistry in exploration are now well established

(Hunt, 1996; Killops and Killops, 2013; Tissot and Welte, 1984) thanks to a basic

understanding of the processes controlling yield, quality and maturation characteristics, and

the development of modelling protocols for assessing rates of physical and chemical

change in time and space in sedimentary basins. Yet, there are still gaps in our level of

understanding of organic-inorganic interactions in source rocks, namely, the adsorptive and

catalytic effects of clay minerals and bombardment of organic matter by the radioactive

decay of uranium. This dissertation seeks to remedy these deficiencies. The clay-rich Lower

Carboniferous Bowland Shale of northern England was the focus of research on MME,

while the the Cambro-Ordovician Alum Shale of Northern Europe was studied to evaluate

the effects of uranium irradiation. Special efforts were made to examine whether laboratory

simulation resembles the catalytic effects of inorganic matter in nature and how these

organic-inorganic interactions influence petroleum exploration.

1.4.1 MME

The hydrocarbon retaining function and the catalytic effect of clay minerals in petroleum

generation have been well documented by previous works. Two main issues related to

MME are discussed in this dissertation.

(1). Most of the previous investigations of the MME focused on Rock-Eval and Py-GC

experiments to show its effects and mechanisms. However, the influence on petroleum

exploration concepts is not systematically reported. A comprehensive comparison of

isolated kerogen material and whole rock samples, covering organic type identification,

kinetics determination, and the application in petroleum system modelling, would unravel

scenarios how much uncertainty can be induced when samples w/o minerals were used in

laboratory experiments.

(2). The validity of MME in nature is still a matter of debate. On the one hand, the catalytic

effects of clay minerals serve as the premise of the carbonium-ion cracking theory (Frost,

1945; Grim, 1947), and Tannenbaum and Kaplan (1985b) suggested that the MME works

efficiently in nature and would change the composition of generated petroleum. Hill et al.

(2007) proposed that the secondary gas in Barnett Shale could be formed by the

combination of hydrocarbon retention on the mineral surface and subsequent catalytic

cracking. On the other hand, Welte (1965) pointed out that petroleum generated in clay

catalytic reactions has a tendency to show a thermodynamic equilibrium, which, however, is

not observed in crude oils with respect to the iso- and n-paraffin ratio. Espitalié et al. (1984)

suggested that the MME is induced by high temperatures in the absence of water and that a

less drastic effect is expected in geological reality. Vandenbroucke and Largeau (2007)

doubted the mineral catalytic theory in nature based on the fact that the primary cracking

mainly occurs in the organic network of kerogen, where minerals are absent.

To achieve these two goals, the Carboniferous Bowland Shale from Northern England was

a key target of the investigations of this dissertation. Taken as the most prospective shale

gas play in England, the Bowland Shale was deposited in a marine environment with

significant organic carbon content. This shale is widely distributed in several sedimentary

basins in Northern England. The conceptual research outline is as follows:

(1). This dissertation will not only investigate the MME on the routine Rock-Eval and Py-

GC basis but also its effect on the phase behaviour prediction and 1-D basin modelling.

Comparison of data derived from whole rock and their kerogen counterpart shall present

how different the petroleum type, quantity, and phase behaviour predictions would be.

Therefore, how the MME influences source rock evaluation and pre-drill petroleum

prediction can be systematically evaluated.

(2). To examine the validity of MME in nature, a series of pyrolysis experiments were

carried out applying different heating rates as one of the most significant differences

between laboratory pyrolysis and geological maturation lies in the heating rate. Geological

heating rates normally fall in the range 10-10 to 10-12 K/min, while a typical laboratory

pyrolysis heating rate is not slower than 10-1K/min. The heating rate dependency of MME

would allow interpretations whether the effect is ubiquitous or will be attenuate in the slow

geological heating environment. A naturally matured sample as calibration would also help

in determining whether the pyrolysis of kerogen or whole rock better simulate petroleum

generation and thus to conclude if the MME exists in nature or not.

1.4.2 Uranium

Although many pyrolysis, isotope, and biomarker studies have been carried out in

exploring how uranium changes the organic matter and influences the petroleum

occurrence, there are still several major puzzles which need to be solved.

(1). Some marine Alum Shale horizons tend to generate gas-rich and very aromatic

hydrocarbons (Bharati et al., 1995; Horsfield et al., 1992a; Lewan and Buchardt, 1989; Sanei

et al., 2014). High uranium concentration serve as an explanation for the atypical petroleum

production characteristics, but the influence of precursor still can not be excluded (Bharati

et al., 1995; Horsfield et al., 1992a). For example, some marine algae, e.g., Chlorella marina,

can also yield aromatic-rich hydrocarbons (Derenne et al., 1996). Furthermore, the trilobite

chitin which can produce aromatic pyrolysates (Arthur Stankiewicz et al., 1996) could also

serves as an explanation why such marine shales generate very aromatic products.

(2). Previous pyrolysis work to investigate uranium irradiation effect was mostly based on

laboratory tests on source rock samples. However, there is a lack of validation of these

experimental results by natural petroleum. Uncertainties still exist whether the high GOC

and aromaticity of the uranium-rich shale pyrolysates are merely induced by organic-

uranium interactions under the artificially high-temperature exposure or whether such

features are true in real natural products. Outcrop samples used in previous studies, which

were heavily weathered and hydrocarbon depleted, turned out to be non-relevant.

(3). The impact of uranium irradiation on kerogen stability is not clear. The maturity

indicator Tmax changes in the response of uranium contents are contradictory, as shown in

various studies (Dahl et al., 1988b; Forbes et al., 1988), and was not supported by kinetic

studies yet.

(4). Previous hydrocarbon generation property evaluations based on pyrolytic researches

are problematic. Because, when petroleum generation occurred in Palaeozoic time, the

kerogen structures of the source are expected to be very different from those in nowadays

samples due to different exposure times to irradiation. Therefore, a reconstruction of the

oil window time kerogen structure is necessary for predicting petroleum generation.

(5). Aliphatic biomarkers in the uranium-rich shale samples are normally not detectable

(Dahl et al., 1988a), thus, the oil-source rock correlation using established biomarkers is

challenging. Important to note is that the distribution of triaromatic steroids in uranium-

rich shales is unique (Dahl et al., 1988a; Lewan and Buchardt, 1989). However, whether

this feature can be observed in producing oil and whether it may serve as a correlation tool

has not been reported so far.

The Middle Cambrian – Lower Ordovician Alum Shale which holds the biggest low-grade

uranium resource in Europe was investigated in this dissertation. A freshly drilled well in

Saint Petersburg (Russia) provides unweathered shale samples with different uranium

contents. Results about scattered outcrop samples from the Scandinavian area are also

presented as supplements. Crude oils produced from Lower Palaeozoic petroleum systems

were collected from the Baltic Basin to enable biomarker correlations with the Alum Shale.

This research concept includes the application of a wide range of methodologies:

(1). The relationship between uranium content, pyrolysate composition, and

thermovaporisation composition was well examined using immature Dictyonema (Lower

Ordovician Alum Shale) core samples. Thus, influences from maturity, weathering, and

precursor can be safely excluded. Accordingly, the uranium irradiation effect leading to

differences in hydrocarbon generation in pyrolysis experiments, as well as in the natural

petroleum occurrence, can be examined.

(2). Bulk kinetic investigations were carried out on Alum Shale samples with various

uranium contents and facilitates a proper evaluation how different uranium contents

influence petroleum generation. This is important in further petroleum system studies

using basin modelling approaches to determine the timing of petroleum generation.

(3). The application of ESI mode FT-ICR MS on the Alum Shale bitumen is helpful in

revealing the molecular characteristics of nitrogen, sulphur, and oxygen (NSO) compounds

which can also reflect the kerogen structure. Based on the structural changes of kerogen

induced by uranium, the mechanisms of ionising irradiation in petroleum generation can be

elaborated.

(4). Since the irradiation damage to organic matter is linear to both uranium contents and

irradiation time, a back-calculation of the kerogen structure during the Palaeozoic time is

possible based on the experiment derived correlation between uranium content and

irradiation damage.

(5). Aromatic biomarkers in extracts gained from the Alum Shale and from producing oil

samples were compared. Possible aromatic steroids biomarkers can be applied in oil-

uranium-rich source rock correlation studies.

1.5 The Structure of the Dissertation

After this INTRODUCTION in CHAPTER 1, the mineral matrix effect in Bowland Shale

samples will be discussed in CHAPTER 2 and 3, and the uranium influence on petroleum

generation from the Alum Shale are presented in CHAPTER 4 and 5.

CHAPTER 2 reveals how clay minerals affect the quality, quantity, kinetics, and phase

behaviour of hydrocarbons generated by Bowland Shale samples based on laboratory

pyrolysis experiments.

CHAPTER 3 is focused on the heating rate dependency of MME, and thus speculated the

effectiveness of MME in geological environment.

CHAPTER 4 compiles results how uranium irradiation changes petroleum generation from

Alum Shale using natural samples for calibration. The original kerogen structures are also

reconstructed.

CHAPTER 5 describes how uranium irradiation modifies biomarker patterns in producing

oil and Alum Shale extracts. A comprehensive oil-source rock correlation was achieved in

the Baltic Basin.

SUMMARY and PESPECTIVES are given in the last CHAPTER.

This chapter has been published as: Yang, S., B. Horsfield, N. Mahlstedt, M. H. Stephenson, and S. F.

Könitzer, 2015, On the primary and secondary petroleum generating characteristics of the Bowland

Shale, northern England: Journal of the Geological Society, v. 173, p. 292-305 (postprint), doi:

10.1144/jgs2015-056.

2. MINERAL MATRIX EFFECT (MME)

2.1 Abstract

The Carboniferous Bowland Shale of northern England has drawn considerable attention

because it has been estimated to have 1329 trillion cubic feet hydrocarbons in-place (gas

and liquids) resource potential (Andrews, 2013). Here we report on the oil and gas

generation characteristics of three selected Bowland Shale whole rock samples taken from a

core and their respective kerogen concentrates. Compositional kinetics and phase

properties of the primary and secondary fluids were calculated through the PhaseKinetics

and GORfit approaches and PVT modelling software. The three Bowland Shale samples

contain immature, marine Type II kerogen. Pyrolysate compositions infer primary

generation of Paraffinic-Naphthenic-Aromatic (PNA) Oil with low contents of wax and

sulphur. Bulk kinetic parameters share many similarities to productive American Palaeozoic

marine shale plays. The secondary gas generation potential of Bowland Shale is bigger than

primary gas potential although it requires 14 kcal/mol activation energy higher to achieve

peak production. Primary oil, primary gas and secondary gas reach their maximum

generation at 137, 150 and 230°C respectively for a 3°C/Ma heating rate. Different driving

forces of expulsion including the generation of hydrocarbon and overpressure caused by

phase separation during sequential periods of subsidence and uplift could be inferred.

2.2 Introduction

It was in the nineteen eighties that Selley (1987) drew attention to the high shale gas

potential of organic-rich shales in the U.K. (Selley, 1987; Selley, 2005; Smith, 1995). He

suggested that the most prospective candidates were the Lower Carboniferous basins in

northern Britain. Concerted exploration with a view to exploitation was never seriously

considered until twenty or so years later, when horizontal drilling and hydraulic fracturing

revolutionized unconventional gas production in the USA and directly led to the rapid

emergence of the shale gas industry (Bowker, 2007; Curtis, 2002; Pollastro, 2007). Indeed,

it was after the UK’s 13th Onshore Licensing Round in 2008 that companies and

government made a concerted effort in shale gas assessment and exploration (DECC,

2011). EIA (2011) evaluated that the Bowland Shale system possesses a risked GIP of 95

trillion cubic feet (tcf) and made a risked recoverable resource estimate of 19 tcf. Gas-in-

place has been estimated by the British Geological Survey (BGS) to lie in the range of 822-

2281 (tcf) (Andrews, 2013). Known as a conventional source rock in the Bowland Basin

and elsewhere in northern England (Lawrence et al., 1987), the Carboniferous Bowland

Shale is also the prime shale gas target in UK (Selley, 2012; Smith et al., 2010). The first

U.K. shale gas well (Preese Hall No.1) drilled by Cuadrilla Resources in 2010 targeted the

play as having the highest shale gas potential in the country (Green et al., 2012).

The integration of outcrop, well and seismic data have shown that the Bowland Shale can

be divided into lower and upper parts. The upper Bowland Shale is thinner but possesses a

higher organic matter content and exhibits better lateral continuity than the lower part

(Andrews, 2013). The organic-rich upper part of the Bowland Shale is hemi-pelagic in

origin and is dominated by clay-rich mudstone intercalated with very thin calcareous

mudstones (Chisholm et al., 1988). The average thickness of this formation is 150 m

(locally reaches 890 m). The upper Bowland Shale, focus of the current investigation, was

deposited in several adjoining basins (Widmerpool Gulf, North Staffordshire Basin, Edale

Basin, etc.) separated by the emergent East Midlands Shelf (Fig. 2.1). The lower shale layer

(age from late Chadian to Brigantian), which was interbedded with mass-flow limestones

and sandstones (Waters et al., 2009), is considerably thicker, reaching 3000 m in its

depocentres (Andrews, 2013). The sedimentology and structural geology of the Bowland

Shale have been studied by many authors (Barrett, 1988; Fraser and Gawthorpe, 2003;

Lawrence et al., 1987; Leeder, 1988; Waters et al., 2009). Biomarkers and stable carbon

isotopes have established likely precursor biota of the organic matter (Armstrong et al.,

1997; Ewbank et al., 1993). Also, Konitzer et al. (2014) have used TOC and carbon isotope

composition to differentiate various depositional environments vertically. The mass of in-

place gas can be evaluated using a combination of forward (includes kinetic) and inverse

(includes mass balance) modelling. Because shales are extremely heterogeneous, both

laterally (tens to hundreds of kilometres) and vertically (metres to decimetres), the

exploration equation has to be applied at appropriate intervals for regional and reservoir

scale applications

Fig. 2.1. Locations of samples and 1 D basin modelling well. Namurian basin distribution after Fraser and

Gawthorpe (2003)

To date, mainly inverse modelling has been applied to the Bowland Shale; thus, its shale gas

potential (Raji et al., 2013; Selley, 2005; Selley, 2012; Smith et al., 2010) and total energy

resource potential (Andrews, 2013; DECC, 2011; EIA, 2011; Hough and Vane, 2014) have

been calculated using basic geochemical data, including TOC and Rock-Eval. Gross et al.

(2015) concluded that the high average TOC content and large thicknesses of the

mudstone lithofacies point to a significant shale gas/liquid potential in areas with

appropriate maturity (1.2-3.5 %Ro), but that a relatively low average HI and high clay

content may be seen as detrimental to shale gas potential. (Green et al., 2012), Imber et al.

(2014) and Pater and Baisch (2011) have recently documented the orientation of fractures

in the Bowland Shale and evaluated the formation’s frackability (ease with which the rock

can be fractured) using rock mechanical experiments and seismic data. The Bowland Shale

consists mainly of impermeable, brittle rock (varying according mineralogy), with many

faults and fractures. The maximum horizontal stress orientation of 8°NNW agrees with the

regional stress orientation.

Little or no kinetics-related work has been published on the Bowland Shale; that means

neither “conversion” using bulk kinetics nor “% gas” using compositional kinetics/physical

property prediction, as a function of organic matter type and/or facies. Kinetic models

basically describe the “ease” with which the substituents in the kerogen breakdown to form

hydrocarbons via assumed pseudoreactions (Braun and Burnham, 1987; Burnham et al.,

1988; Schenk et al., 1997). Utilising specific kinetic parameters for the target shale is

imperative whenever possible (Dieckmann and Keym, 2006; Peters et al., 2006).

Compositional kinetics provide compositional information on the generated fluids; surface

GOR and gas dryness prediction are of great importance, because both have big influences

on the quality of the produced oil and gas. The PhaseKinetics model characterizes the

compositional evolution of the fluids generated with increasing thermal stress, as well as

the phase behaviour of the petroleum at different maturity levels (di Primio and Horsfield,

2006). Predictions of bulk petroleum compositions and physical properties have already

been published for basins in South Africa (Hartwig et al., 2012), North Dakota (Kuhn et al.,

2012), Norway (Duran et al., 2013), Eastern Canada (Baur et al., 2010), Brazil (di Primio

and Horsfield, 2006) and China (Tan et al., 2013), in several cases with confirmation of

correct prediction using local calibration. In the current study we employed the

PhaseKinetics approach on both whole rock samples and kerogen concentrates to predict

the properties of fluids generated from Bowland Shale, also taking into consideration the

impact of rock-fluid interactions (mineral matrix effects, considered to be laboratory

artefects) on the results (Espitalie et al., 1980; Horsfield and Douglas, 1980).

The contribution of secondary gas at high levels of thermal stress, determined to be

dominant in the majority of shale gas “sweet spots” (Jarvie et al., 2007) also needs to be

qualitatively and quantitatively evaluated. The GOR-Fit model (Mahlstedt et al., 2013)

discriminates between primary gas, primary oil (both from kerogen breakdown) and

secondary gas (from the breakdown of primary oil) has been successfully applied to marine

and lacustrine shales in Germany (Ziegs, 2013). In the current study we employed the

GOR-Fit kinetic model to predict secondary gas formation in the Bowland Shale.

In the current contribution we have focussed on the specific kinetic parameters themselves,

utilising a 1D basin model to demonstrate how fluid properties change as a function of

organic maturity and reservoir conditions. We have then considered the evolution of

expulsion mechanisms during different geological times, and thereafter considered the

degree of secondary gas formation, comparing our model with default kinetics models

derived from published studies in Petromod©.

2.3 Samples and Analytical Procedure

2.3.1 Samples

The shallow Carsington Dam Reconstruction C4 borehole (SK 244 503) which targeted

upper Bowland Shale, was drilled in Derbyshire, Northern England. Both hemi-pelagic

marine shale and pro-deltaic turbidites deposited in deep-basin water were recognized using

micropetrography and TOC-bulk δ13Corg data (Konitzer et al., 2014). Thick intervals of

marine Bowland Shale are relatively homogeneous on a metre-scale, whereas intervals of

interbedded shales and turbiditic sandstones are relatively more heterogeneous, as revealed

by petrophysical (Hough and Vane, 2014), geochemical (Gross et al., 2015) and lithological

(Konitzer et al., 2014) properties. Three unweathered core shale samples as well as their

kerogen concentrates from the marine part of that well were tested in this study. Thin

section observation and mineralogy had previously shown that sample No.1 and No.3

represent thin-bedded carbonate-bearing clay-rich mudstones and sample No.2 is lenticular

clay-dominated mudstone. All the three samples have δ13Corg values between -28.0‰ and -

28.4 ‰ indicating that the kerogen is derived from marine planktonic algae, so the three

samples are representative of the thick hemi-pelagic mudstones, and these lithofacies

dominate the succession at Carsington. (Fig. 2.2) (Konitzer et al., 2014).

Fig. 2.2. Stratigraphy of the study area. Depth, lithology, TOC and bulk δ13Corg data of the Carsington C4 core

samples (Konitzer et al., 2014) as well as the locations of the 3 shale samples investigated in this paper.

2.3.2 Analytical procedure

2.3.2.1 Kerogen isolation and screening

Kerogen concentrates were obtained by (1) crushing the shale to millimetre size, (2)

treating with hydrofluoric acid for one week at room temperature and (3) sieving to 10-500

microns. TOC and Rock-Eval analyses were performed using a Leco SC-632 Analyser and

Rock-Eval 6 instrument respectively following established procedures (Espitalie et al.,

1977).

2.3.2.2 PhaseKinetics (compositional kinetics)

The PhaseKinetics approach of di Primio and Horsfield (2006) has four stages:

(1). Pyrolysis-gas chromatography (PyGC), providing a quick evaluation of the kerogen

structure characteristics in terms of Petroleum Type Organofacies, was performed using

the Quantum MSSV-2 Thermal Analysis System® interfaced with an Agilent GC-6890A

(Horsfield et al., 2014). Briefly stated, milligram quantities (whole rock: 14-16 mg, kerogen:

2-4 mg) of each sample were loaded into a small open glass tube and heated under flowing

helium; free hydrocarbons were vented for 3 minutes during an isothermal purge at 300°C,

after which the C2+ pyrolysis products generated during heating from 300℃ to 600℃ were

collected in a cryogenic trap (liquid nitrogen). Methane passed through the trap and passed

through the GC column to the Flame Ionisation Detector (FID). Trapped products were

liberated by removing the cooling agent and heating the trap to 300℃. AHP-Ultra 1

dimethylpolysiloxane capillary column (50 m length, inner diameter of 0.32 mm, film

thickness of 0.52 mm) connected to a FID was used to with helium as carrier gas.

Quantification of individual compounds and boiling range splits was conducted by external

standardisation with n-butane.

(2). Bulk kinetic parameters were assessed by subjecting samples to open-system, non-

isothermal pyrolysis at four different linear heating rates (0.7, 2, 5, 15 ℃/min) using a

Source Rock Analyser® (SRA) following established procedures (Braun and Burnham,

1987). The discrete activation-energy (Ea) distribution optimization with a single, variable

frequency factor (A) as well as geological extrapolation were performed using the

KINETICS 2000® and KMOD® programs.

(3). Non-isothermal closed-system micro scale sealed vessel (MSSV) pyrolysis (Horsfield et

al., 1989) is a micro-analytical method to artificially mature sedimentary organic matter to

different stages of conversion and to quantify the composition of generated products. It

provides the possibility of determining primary and secondary reaction kinetics of specific

compound groups and to extrapolate their generation to geological heating rates (Horsfield

et al., 2014). For each experiment, milligram quantities of samples were sealed in glass

capillaries and artificially matured at 0.7℃/min to temperatures corresponding to 10, 30, 50,

70 and 90 transformation ratio (TR) as defined by bulk kinetic results. The tubes were then

cracked open using a piston device coupled with the injector, and the released products

were swept into the GC using a flow of helium. Quantification was performed by external

standardisation using n-butane.

(4). Compositional kinetics and physical property modelling were fulfilled in the last step.

The hydrocarbons generated during MSSV are divided into 14 pseudo compositions. Seven

of them are in the gas fraction (C1, C2, C3, i-C4, n-C4, i-C5, n-C5) and the gas composition

was corrected based on a GOR- gas-wetness correlation from natural black oil. The other

seven compounds describe the liquid phase consisting of C6 and pseudo boiling ranges of

C7-15, C16-25, C26-35, C36-45, C46-55, C56-80. According to the weight percentage of the 14 pseudo

compositions, each bulk kinetic potential which has the same activation energy was

populated into 14 parts, afterward these compositional kinetics models are ready to be

applied to basin modelling software (especially the IES PetroMod® which has a module

for inputting these) which make this method very convenient. Physical property modelling

was carried out using PVT-Sim® based on the 14 pseudo compounds determined by

MSSV. Standard temperature and pressure (STP) GOR was calculated through separator

simulator module in the software and phase envelopes were also drawn.

2.3.2.3 GOR-Fit

The GOR-Fit model based on open-system SRA and closed-system MSSV pyrolysis

consists of three main steps (Mahlstedt et al., 2013). In the first one, the MSSV - generation

of C1-5, C6+ and the total C1+ boiling fractions are normalized to the maximum MSSV-yields.

Since the normalized C1+ MSSV yields curve and SRA-TR curves are identical (Schenk and

Horsfield, 1993) and only primary cracking takes place in the open-system SRA pyrolysis,

the primary oil and gas splines can be deduced from the SRA-TR curve by multiplying by

an oil and gas ratio assumed fixed and derived from pyrolysis gas chromatography after a

small temperature adjustment to fit the measured MSSV oil and gas generation curves

better. The second step is to calculate the secondary gas amount by subtracting primary gas

from measured MSSV oil yields at corresponding temperatures. A secondary gas spline is

again approximated by “factorising” the SRA-curve (factor derived from multiplication of

the C6+ spline factor by 0.7 assuming that 70% of C6+ compounds are degraded to gas and

30% to coke) which is then temperature shifted to match calculated secondary gas yields.

After obtaining the generation characteristics of primary oil, primary gas and secondary gas

in 3 heating rates, the kinetics models and geological extrapolations were achieved by using

KINETICS 2000® and KMOD® in the last step.

2.3.2.4 1-D Basin modelling

1-dimensional basin modelling was carried out on well Grove 3 from the East Midlands

Shelf using IES PetroMod®2013. Lithology and depth inputs in the modelling came from

the drilling , and stratigraphic ages were taken from International Commission on

Stratigraphy (Cohen et al., 2013) and Gradstein et al. (. A Kerogen-Oil-Gas kinetics model

with secondary reaction developed on kerogen No.3 in this research was used as kinetics

input. Two periods of uplift of the Bowland Shale in the late Carboniferous/early Permian

and after the Late Cretaceous have been recognised by (Leeder, 1988) and (Barrett, 1988),

and the heat flow model was modified after (Jarvis and McKenzie, 1980). Calibration on

vitrinite reflectance is after the report from BGS which also produced a 1 D model on well

Grove 3 (Andrews, 2013).

2.4 Results and Discussion

2.4.1 Primary generation

Here we compare the results for whole rock samples with those of kerogen concentrates.

Significant differences in composition are reported, the causes discussed, and the

ramifications for petroleum composition outlined.

2.4.1.1 Whole rock

All three whole rock samples have TOC contents higher than 2% (Table 2.1) fitting the

minimum TOC requirement for shale gas development (Curtis, 2002; Muntendam-Bos,

2009), but the S2 and HI values are low (Table 2.1). Rock-Eval crossplots confirm that the

three whole rock samples generate and release pyrolysates with a type Ⅲ composition (Fig.

2.3 a). Samples 1 and 3 are immature samples while sample 2 is marginally mature (Fig. 2.3

b).

In cases where analytical artefacts are excluded (e.g. mineral-organic interactions occurring

during pyrolysis;(Horsfield and Douglas, 1980) the proportion of resolved and identifiable

compounds in the GC trace reflects the kerogen structure as a whole (Horsfield et al., 1989;

Larter, 1984). Their GC traces show that the whole rock samples analysed here tend to

generate high percentages of low molecular weight compounds (C1 – C5) and high

concentrations of aromatic compositions like ethylbenzene, xylenes, trimethylbenzene, and

naphthalene (Fig. 2.4). The average alkyl chain length distribution of the pyrolysates from

whole rocks 1 and 3 are of the Gas and Condensate type whereas whole rock 2 falls in Low

Wax Paraffinic-Naphthenic-Aromatic (P-N-A) Crude Oil field (Fig. 2.5). Two additional

ternary diagrams were used to characterize the pyrolysate in terms of aromaticity,

paraffinicity and either sulphur content (Eglinton et al., 1990) or phenol content (Larter,

1984). They clearly show that the whole rock pyrolysate is very aromatic (Fig. 2.6 a and b);

sulphur-containing compounds and phenols are in low abundance (Fig. 2.6 a and b).

Table 2.1. Rock-Eval and TOC data.

Sample Number GFZ

Number Depth

(meter)

(mg/g)

Tmax

(°C)

(mg HC/g

TOC)

(mg CO2

TOC)

(S1/(S1+S2)) TOC

(%)

Whole rock 1 G013218 28.68 0.18 1.71 1.67 429 62 61 0.0952 2.75

Whole rock 2 G013219 28.42 0.16 4.88 0.7 438 188 27 0.0317 2.6

Whole rock 3 G013220 22.36 0.37 6.46 0.49 430 206 16 0.0542 3.14

Kerogen 1 G013688 28.68 1.36 60.89 1.20 430 329 6 0.0218 18.5

Kerogen 2 G013689 28.42 0.39 14.13 0.48 432 318 11 0.0269 4.45

Kerogen 3 G013690 22.36 1.18 33 0.83 426 324 8 0.0345 10.2

S1: quantity of free hydrocarbons (gas + oil). S2: quantity of thermally generated (cracked) hydrocarbons. S3: quantity of CO2 generated during

pyrolysis of the sample. HI (hydrogen index)=(S2*100)/TOC. OI (oxygen index)=(S3*100)/TOC. PI (production index) = S1/(S1+S2)

Fig. 2.3. Rock-Eval and TOC diagrams for kerogen type and maturity identification.

0246810 12 14 16 18 20

S2(mg HC/g TOC)

TOC(%)

HI 600

II/III

III

HI 350

HI 200

HI 50

100

200

300

400

500

600

700

800

900

400 420 440 460 480 500

Hydrogen Index(mg/g TOC)

Tmax(℃)

III

0.5% Ro

1.3% Ro

(b)

(a)

123

Whole rock

Kerogen

Fig. 2.4. PyGC chromatograms of the 6 samples. Normal alkane and alkene peaks have been highlighted and

selectively numbered. Representative aromatic compounds are ethylbenzene (a), meta- and paraxylenes (b),

orthoxylene (c), 1,2,4-trimethylbenzene (d), naphthalene (e) and 2-methylnaphtalen (f).

No. 1

Whole rock

n-alkane / -alkene doublets

Aromatics

12 14 16 18

20 22

No. 3

Whole rock

No. 2

Whole rock

10 12 14

16 18

20 22

12 14

16 18 20 22

24 26

10 12 14 16 20

24 26 28

cde

10 12 14 16 18 20 22

24 26 28

cde

No. 1

Kerogen

No. 2

Kerogen

No. 3

Kerogen

Fig.2.5. Pyrolysate chain length distribution and Petroleum Type Organofacies classification (Horsfield, 1989)

Fig. 2.6. Petroleum composition predictions from PyGC results according Larter (1984) and Eglinton et al. (1990)

As far as thermal response is concerned, whole rock Bowland Shales have peak activation

energies between 55-57 kcal/mol and frequency factors exceed 2.85×1014 (Fig. 2.7).

Applying these values to natural maturation using a typical geological heating rate (3℃/ma),

the whole rock samples 1 and 3 reach 50%TR at about 150℃ (Fig. 2.8), whereas whole

rock sample 2 needs around ten more degrees to reach that TR. These kinetics

characteristics of the whole rock samples are unusual, in that the samples require higher

temperatures for kerogen breakdown than most known Palaeozoic marine shales

(Mahlstedt, 2012).

123

Whole rock

Kerogen

123

Whole rock

Kerogen

(a)

Larter, 1984

(b)

Eglinton et al, 1990

Fig. 2.7. Bulk kinetics models of the whole rock samples and kerogen concentrates.

Fig. 2.8. Transformation ratio variations in geological heating rate (3o/ma).

Two compositional kinetics models for whole rock samples are shown in Fig. 2.9. These

were built by populating the bulk kinetic potentials with MSSV pyrolysis data (di Primio

and Horsfield, 2006). Cumulative GORs in surface environment of whole rock samples get

enhanced with increasing thermal maturation except 2 slight deviant values from sample

No.1 and No.2 at 50% TR (Fig. 2.10). The maximum GOR can be as high as 498Sm3/Sm3

of sample 1 at 90% percentage which is very similar to the GOR behaviour of Arang coal

(organic type Ⅲ) of Indonesia (di Primio and Horsfield, 2006). The pressure-temperature

phase envelope for multicomponent mixture gives the region of temperatures and

pressures at which the mixture forms two phases. Generally speaking, the envelope in

40 43 46 49 52 55 58 61 64 67

No.1No.2 No.3

Whole rock

Kerogen

40 43 46 49 52 55 58 61 64 67

A=2.8518E+14/s

A=1.54E+13/s

A=1.215E+15/s

A=1.88E+13/s

A=3.4457E+14/s

A=2.42E+13/s

Activation Energy(kcal/mol)

Fraction of total potential (%)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

50 70 90 110 130 150 170 190 210 230 250

Transformation ratio

Temperature (℃)

Whole rock

Kerogen

temperature axis direction is controlled by molecular weight, while GOR and gas wetness

control the pressure axis direction (Amyx et al., 1960). Phase envelopes of whole rock

sample 3 reflect the fact that the primary generated fluids are dominated by low molecular

weight compounds. For a hypothetical reservoir at 100℃ and 200 bar, the critical point of

hydrocarbons accumulating up to 90% TR is very close to that reservoir condition, and the

fluids can be termed volatile oil (McCain, 1990).

Fig. 2.9. Compositional kinetic models of selected samples.

Fig. 2.10. Gas:oil ratio of the samples analyzed as a function of increasing transformation ratio.

40 42 44 46 48 50 52 54 56 58 60 62 64 66 68

Activation Energy( kcal/mol)

Kerogen 1

A=1.54E+13/s

Kerogen 3

A=2.42E+13/s

40 42 44 46 48 50 52 54 56 58 60 62 64 66 68

Whole rock 1

A=2.8518E+14/s

Whole rock 3

A=3.4457E+14/s

40 42 44 46 48 50 52 54 56 58 60 62 64 66 68

C56-80 C46-55

C36-45 C26-35

C16-25 C7-15

n-C6 n-C5

i-C5 n-C4

i-C4 n-C3

n-C2 n-C1

Fraction of total potential (%)

TR(%)

GOR(Sm3/Sm3)

100

150

200

250

300

350

400

450

500

10 30 50 70 90

Whole rock 1

Whole rock 2

Whole rock 3

Kerogen 1

Kerogen 3

2.4.1.2 Kerogen

Although the three kerogen concentrates have quite different TOC contents and S2 values

(Table 2.1), they share very similar HIs. With the HI values range of 318-329

mgHC/gTOC the three kerogen concentrates are classified as type Ⅱ (Fig. 2.3a). The

maturity indicator Tmax implies that the three kerogen concentrates are immature (Fig. 2.3b).

The pyrolysates of kerogen are dominated by normal alkanes/enes, subsidiary aromatics

and other resolved peaks are generated (Fig. 2.4). All three concentrates fall in the Low

Wax P-N-A Oil area (Fig. 2.5), which manifests the typical characteristics of many marine

shales (Horsfield, 1997). Fig. 2.6 suggests that hydrocarbons generated by kerogen are

richer in paraffinic compounds than aromatic ones. In addition, the sulphur content is also

very low (Fig. 2.6a).

Peak activation energies of the three kerogen concentrates are 51 kcal/mol and their

frequency factors range from 1.54-2.42×1013 (Fig. 2.7). The geological extrapolation curves

fall closely together as regards TR variations according to temperature (Fig. 2.8). The

Bowland Shale kerogen generation kinetics are more stable than sulphur-rich marine shale

(Dieckmann, 2005) and closely resemble those reported for productive unconventional

shale plays from the US including Barnett shale (Jarvie et al., 2010), Bakken shale (Kuhn et

al., 2012) and Woodford shale (Mahlstedt, 2012).

Due to the limited amount of sample available MSSV experiments were not carried out on

kerogen No.2. Compositional kinetics results of sample No.1 and 3 show that about half of

the hydrocarbons in peak generation (activation energies ranges between 50-54 kcal/mol)

are contributed by compounds between C7-C25 and gases make up only small proportions

of the total products (Fig. 2.9).

Cumulative GOR variations as a function of increasing TR for kerogens 1 and 3 are closely

similar (Fig. 2.10). GOR increases steadily from less than 100 m3/Sm3 at 10% TR to about

200 m3/Sm3 at the highest TR and this GOR variation pattern is very similar as those of

the Woodford Shale and Kimmeridge Clay (di Primio and Horsfield, 2006). The cumulative

hydrocarbon phase envelopes imply that fluids generated by Bowland kerogen concentrates

are typical black oil (McCain, 1990) and the systematic decrease in cricondentherms and

increase in cricondenbars together with the shift of the critical point towards higher

pressures and lower temperatures with increasing TR consistent with critical points shift

pathways of fluids from Snorre Fields during maturation (di Primio et al., 1998).

2.4.1.3 Comparison and discussion

Clearly, there are significant compositional differences between the respective pyrolysates

of whole rock and kerogen concentrate pairs. The whole rock samples show low HI values

and are classified as containing type Ⅲ organic matter, while the kerogen concentrates have

higher HI and are classified as comprising type Ⅱ organic matter (Fig. 2.3). Whole rock

samples tend to generate higher percentages of low molecular weight compounds (C1-C5)

and alkylaromatics (Fig. 2.4), whereas the equivalent kerogen pyrolysate is dominated by

normal alkanes and alkenes (Fig. 2.4). Ternary plots also demonstrate differences of organic

facies and paraffinicity between these two materials (Fig. 2.5 and 2.6). From the kinetic

perspective, whole rock samples are more refractory and heterogeneous than the kerogen

(Fig. 2.7 and 2.8). More light compounds were generated during pyrolysis from whole rock

samples, which is responsible for that they have higher GORs (Fig. 2.10) and lower

cricondentherms in the phase envelopes (Fig. 2.11) than their kerogen counterparts.

Fig. 2.11. Phase envelopes of whole rock and kerogen of sample No.3 during artificial maturation.

The differences between kerogen and whole pyrolysates have been discussed over decades.

Saxby (1970) and Robl and Davis (1993) reported that the treatment of whole rock by

hydrofluoric acid during mineral dissolution and kerogen concentration does not change

kerogen structure significantly. Differences in pyrolysate compositions have been attributed

to the Mineral Matrix Effect. The effect occurs in many open- and closed system pyrolysis

experiments including Rock-Eval (Espitalie et al., 1980; Espitalié et al., 1984; Makadi, 1983),

pyrolysis GC (Horsfield and Douglas, 1980; Karabakan and Yürüm, 1998), bulk kinetics

100

150

200

250

-100 0 100 200 300 400 500 600

Whole rock

No.3

Kerogen

No.3

10%

90%

70%

30%

50%

Critical points

Temperature (℃)

Pressure (bar)

determination (Burnham, 1994a; Dembicki, 1992; Dessort et al., 1997; Pelet, 1994) and

hydrocarbon expulsion efficiency calculations (Lewan et al., 2014), and is brought about by

sorption followed by catalytic thermal degradation. Because of their high surface area (Sing,

1985), clay minerals (especially smectite) have the ability to strongly adsorb pyrolysate

(Espitalie et al., 1980; Espitalié et al., 1984), especially the heavy compounds (Katz, 1983).

Clay mineral tends to catalyze the kerogen to generate more CO2, light hydrocarbons and

aromatic compounds (Espitalié et al., 1984; Larter, 1984; Lu and Kaplan, 1989;

Tannenbaum et al., 1986b). A disproportionation of hydrogen occurs in the pyrolyser,

enhancing C1-C5 yield, while simultaneously depositing dead carbon, bringing about

diminished HIs and lower yields of heavy compounds in Py-GC data, and being more

refractory from a kinetic perspective. The overall outcome as far as bulk petroleum is

concerned, is that genetic potential is diminished (Fig. 2.3 and 2.6), inherent oil potential is

lowered relative to gas and in absolute terms (Fig. 2.5, 2.10) and phase envelopes change

their shape accordingly (Fig. 2.11).

Sedimentological, organic petrological and stable carbon isotope studies (Armstrong et al.,

1997; Fraser and Gawthorpe, 2003; Konitzer et al., 2014) have shown that the 3 upper

Bowland Shale in this research is marine shale from predominantly hemi-pelagic deposition

and organic matter is derived from planktonic phytoclasts. These attributes are better

represented by the isolated kerogen pyrolysis data (Rock-Eval, PyGC and bulk kinetic

results) than by the equivalent whole rock data. It should be pointed out that the mineral

matrix effects shown here are thought to only exist in artificial pyrolysis experiments and

not during natural catagenesis. Fast heating rates, high temperatures, a dry pyrolysis

environment and enhanced contact of the organic matter with minerals after grinding are

considered as the main reasons that lead the mineral matrix effect in laboratory pyrolysis

(Makadi, 1983; Vandenbroucke and Largeau, 2007). Another important thing is that not

every source rock necessarily has to suffer from this effect. Horsfield and Douglas (1980)

and Katz (1983) concluded that the matrix effect varies according to mineralogy and TOC

content of the rocks under investigation. A very high TOC or low clay content can

decrease or avoid the effect (Reynolds et al., 1995). Tannenbaum and Kaplan (1985a), and

Lewan et al. (2014) also reported that the existence of water in the pyrolysis experiments

can hinder the excessive formation of coke and catalyzing function of clay minerals.

However, since the TOC content of Bowland Shale ranges from 1.3% to 9.1% (Gross et al.,

2015), clay contents are considered to be medium/high (EIA, 2011) and all pyrolysis

experiments employed here are anhydrous systems, the mineral matrix effect is likely

inevitable in the Bowland Shale (except those samples with TOC content higher than 6%)

pyrolysis experiments. If whole rock samples are used in the Bowland Shale PhaseKinetics

research here, the cumulative GOR can be greatly over-estimated and leads to erroneous

conclusions in phase prediction and resource evaluation.

2.4.2 Secondary cracking

GOR-Fit was applied to kerogen 3 to explore the generation characteristics and kinetics of

primary oil, primary gas and secondary gas formation. As the 5℃/min heating rate MSSV

experiments shows, the C6+ fraction starts to decrease at around 460℃ with increasing

temperature (Fig. 2.12) as a result of secondary cracking. The decrease of C1+ generation

and C1-5 products after 510℃ and 540℃ (Fig. 2.12) manifests the formation of coke or

pyrobitumen (Dieckmann et al., 1998). Small scale secondary cracking occurs when

temperature reaches 420℃ in the MSSV and significantly more secondary gas was formed

after 460℃ where is the onset of the decrease of C6+ compounds in the MSSV (Fig. 2.12).

The excellent identical trend of secondary gas in MSSV and calculated secondary gas (Fig.

2.12) attest to the robustness of the GOR-Fit approach.

By combining the generation spline under three heating rates (0.7, 2.0, 5.0 ℃/min), the

kinetics of primary oil, primary gas and secondary cracking can be drawn (Fig. 2.13). The

activation energy distribution of primary oil (Fig. 2.13) is very similar to the bulk kinetic

distribution (Fig. 2.7). Compared with primary gas, the peak activation energy of secondary

gas is 14 kcal/mol higher, and frequency factor also increased 2 degrees (Fig. 2.13).

Fig. 2.12. Measured MSSV pyrolysis data of kerogen No.3 for boiling ranges C1+, C6+ and C1-5 normalized to

the maximum C1+ yield and fitted spline curves for calculated primary and secondary gas generation using the

heating rates of 5.0o/min, compared with normalized SRA TR curve.

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

300 350 400 450 500 550 600

Cumulative Yield (TR)

Temperature (℃)

C1+ MSSV

C6+ MSSV

C1-5 MSSV

Secondary Gas MSSV

SRA TR

Primary Oil

Primary Gas

Secondary Gas

℃

/min

Fig. 2.13. Kinetics models of primary oil, primary gas and secondary gas generation of kerogen No.3.

The generation rate of bulk primary hydrocarbons (SRA), primary oil, primary gas, and

secondary gas at a linear heating rate of 3K/Ma is shown in Fig. 2.14. Both the primary oil

and gas generation curve are within the SRA range. About 80% of the primary

hydrocarbon was contributed by primary oil, and this partly explains why both the SRA

and primary oil reach peak generation rates between 136-138℃ (Fig. 2.14). When the

temperature has reached 150℃ and the Ro is 1.2% the primary gas reaches its maximum

generation rate. The secondary gas generation capability of the sample a crucial important

in shale gas potential evaluation, as seen in e.g. the Fort Worth Basin (Jarvie et al.,

2007).Much higher maturity is required to achieve the secondary gas compared with the

primary gas, for the peak secondary generation temperature reaches 230℃ which is 80℃

higher than primary gas and vitrinite reflectance is as high as 2.85% (Fig. 2.14).

Fig. 2.14. Computed generation rate curves as a function of temperature at a geological heating rate of 3

℃

/ma

and vitrinite reflectance for kerogen No.3.

Activation Energy(kcal/mol)

Fraction of total potential (%)

40 42 44 46 48 50 52 54 56 58 60 62

41 43 45 47 49 51 53 55 57 59 61 63

51 53 55 57 59 61 63 65 67 69 71 73 75

Primary oil

A=8.75E+13/s

Primary gas

A=1.701E+14/s

Secondary gas

A=4.0736E+16/s

0.0

1.0

2.0

3.0

4.0

5.0

6.0

050 100 150 200 250 300

Vitrinite Reflectioin (%)

Gen. Rate (mg/g TOC)

Temperature (℃)

Bulk Generation

Primary Oil

Primary Gas

Secondary Gas

Vitr (%Ro)

℃/ma

2.5 Application

2.5.1 Primary generation

The upper Bowland Shale in well Grove 3 lies in the lower part of Namurian stage and was

overlaid by the Millstone Grit. Since the target layer is very thin in the borehole, here we

use Namurian layer to represent it in the 1 D basin modelling for illustrative purpose.

Modelling result demonstrates that the upper Bowland Shale had experienced a rapid burial

in the late Carboniferous and the kerogen attained about 26% TR at the end of

Carboniferous (roughly equivalent to Ro:0.6%) (Fig. 2.15 a) when the primary oil reached a

high generation level (Fig. 2.14). During this period, the main driving force for

hydrocarbon expulsion would be pressure-driven flow as kerogen degradation and rapid

compaction took place (Tissot and Welte, 1984). Fluids generated at TR 30% fall into the

one-phase field in reservoir conditions (Fig. 2.15 c) indicating that in-situ primary

hydrocarbons would exist as an undersaturated liquid in the source rock. The uplift that

happened during late Carboniferous to early Permian not only stopped the rapid organic

maturation but also changed the reservoir conditions significantly (point A has a

temperature of 90 ℃ and pressure of 186 bar, the temperature and pressure of point B are

51 ℃ and 72 bar respectively) (Fig. 2.15 a). If we use 30% TR fluids to roughly represent

the hydrocarbon generated at point A, it can be predicted that when petroleum in the

source rock was shifted from point A to point B the decrease in temperature and pressure

would cause a phase separation and gas composition would contribute 8% in volume of

the whole fluids (Fig. 2.19 c). This sudden gas-exsolution in the very tight shale reservoir

would cause an abnormal pressure in the source rock and increase the expulsion efficiency

greatly (Momper, 1979). Thus the driving force for expulsion in this period would be

changed to the abnormal pressure caused by volumetric expansion induced by phase

separation. After then the shale was deeply buried again between Jurassic and Cretaceous

time and the TR of the organic matter reached as high as 92% at a maximum burial of

2900m (Fig. 2.15 a). By the end of the Mesozoic was reached (roughly equivalent to

Ro:1.4%) primary generation entered its late stage (Fig. 2.14). Expulsion should have been

driven by continuous burial and vast hydrocarbon generation. A major uplift happened in

the Cenozoic which reduced the temperature and pressure of the Bowland Shale reservoir

again (point C has a temperature of 160 ℃ and pressure of 283 bar, the temperature and

pressure of point D are 73 ℃ and 187 bar respectively). However, this time phase

separation would not be likely to occurduring the uplift when reservoir conditions of TR

90% fluids were changed from point C to D (Fig. 2.15 f). Practically, if petroleum under

reservoir conditions (point D in Fig. 2.15 a and f) was produced to the surface (point E in

Fig. 2.15 a and f) a gas-exsolution would happen again, and the vapour phase would hold

about 25% of the total fluid.

Fig. 2.15. Transformation ratio and Ro evolution histories of well Grove 3 and phase envelopes of primarily

generated fluids in according maturities by upper Bowland Shale. The well location can be found in Fig. 2.1. C-N

and C-W in stratigraphy part represent Namurian and Westphalian in Carboniferous respectively. Red triangle in

each of the phase envelop represents reservoir condition in geological burial history respectively. Black diamonds

in 1D modelling map and phase envelops stand for the temperatures and pressures of point A, B, C, D, and E,

and the black dash lines linking the points simulate the processes of the fluids being migrated among different

reservoir conditions with the arrows imply the moving directions.

In the combination of burial history and phase properties of hydrocarbons generated in

geological time, different dominant expulsion driving forces can be proposed and the

surface GOR can be assessed. Different expulsion driving forces and mechanisms lead to

varying expulsion efficiency and define the amount and property of the unconventional

resource left in the source rock. Although only primary hydrocarbon is addressed here, any

further secondary cracking, migration or biodegradation would act upon this first-formed

composition. The produced GOR prediction is very important in oil field strategy making,

because different fluids varies in economic perspective and engineering requirement.

Unfortunately, there are no production data in this well to verify the phase prediction

results. Nevertheless, this systematic approach including hydrocarbon composition

simulation, phase variation prediction, basin modelling application could be a new thinking

in unconventional system production prediction and resource evaluation.

2.5.2 Secondary cracking

Kinetics results shown in Fig. 2.13 can be converted as a Kerogen-Oil-Gas kinetics input

model in the basin modelling software, thus gas generated under secondary cracking by

upper Bowland Shale in well Grove 3 can be simulated (Fig. 2.16 a). The maximum

generation of secondary gas is 151169 tons of secondary gas/km2 (Table 2.2) and about

90000 tons in the area (Fig. 2.16 a), however, if a default Kerogen-Oil-Gas kinetics model

developed by (Quigley et al., 1987) was applied, the secondary gas products are predicted to

be only 30492 ton/km2 (Table 2.2) and about 20000 tons in the area (Fig. 2.16 b). A more

comprehensive comparison of maximum secondary gas generation in Grove 3 predicted by

kinetics developed in this research and other 9 type Ⅱ source rock default kinetics models

(Abu-Ali et al., 1999; Behar et al., 1997; Dieckmann et al., 2000b; Dieckmann et al., 1998;

Pepper and Corvi, 1995; Quigley et al., 1987; Ungerer, 1990; Vandenbroucke et al., 1999;

Waples et al., 1992) manifests that the result can be very enormously different (Fig. 2.17.

and Table 2.2). A maximum secondary generations of 226920 ton/km2 predicted by

Waples et al. (1992) is more than 70 times bigger than the (Burnham and Sweeney, 1989)

model which is only 3182 ton/km2 (Fig. 2.17. and Table 2.2). Relatively speaking,

secondary kinetics model developed in this research provides a moderately high production

of secondary gas and share many similarities with the prediction of the (Pepper and Corvi,

1995) model (Fig. 2.17. and Table 2.2). It has to be pointed out that the vitrinite reflectance

of upper Bowland Shale in well Grove 3 is only 1.4%, which implies that the shale only

experienced a low degree of secondary cracking. Thus more significant differences in

secondary gas predictions must exist at higher maturity areas when different kinetics

models are applied in the basin modelling.

Kinetics parameters defines in which period of secondary cracking the shale is , for

example a certain maturity is in the beginning of secondary cracking in one kinetics model,

but might be in peak generation in another model. Every target source rock must be

approached with a unique secondary kinetics model, huge errors might be induced if

default models are selected which plays very important role in shale gas in-place assessment.

Fig. 2.16. A comparison of secondary gas generation per km2 of upper Bowland Shale in well Grove 3 if (a)

secondary cracking kinetics model in this research and (b) kinetics model from Quigley et al. (1987) are applied

in the basin modelling respectively.

Fig. 2.17. The maximum secondary gas generation per km2 of upper Bowland Shale in well Grove 3 when

secondary kinetics of this research was applied as well as the predictions from 9 other default Kerogen-Oil-Gas

kinetic models in the PetroMod 2013.

Table 2.2. Detailed information about default Kerogen-Oil-Gas kinetics models shown in Fig. 2.17.

Author and year Kerogen type Lithology Location of sample Age of sample Secondary gas/ton

Burnham and Sweeney. (1989) Ⅱ / / /

3182

Vandenbroucke et al. (1999) Ⅱ Shale North Sea Kimmeridge

5597

Behar et al. (1997) Ⅱ Shale Paris Basin Toarcian 8028

Dieckmann et al. (2000) Ⅱ Lime mudstones Western Canada Basin Upper Devonian 8226

Quigley et al. (1987) Ⅱ / / / 30492

Dieckmann et al. (1998) Ⅱ Shale Lower Saxonian Basin Toarcian 101124

Ungerer (1990) Ⅱ Shale North Sea Kimmeridge 130254

This research Ⅱ Shale Northern England Lower

Carboniferous 151169

Pepper&Corvi (1995) Ⅱ Siliciclastic mixed / 161430

Waples et al. (1992) Ⅱ Artificial / /

226920

2.6 Conclusions

Although the mineral matrix effect is a laboratory induced artefact and may not occur on

every shale, the three Bowland Shale researched here suffered from this effect severely. If

whole rock samples are used in pyrolysis experiments instead of kerogen concentrates, the

hydrocarbon generation potential as defined by HI is under-estimated, bulk kinetic

parameter indicate higher thermal stabilities, and the inferred natural GOR is over-

estimated.

The three upper Bowland Shale samples are immature (equivalent Ro less than 0.5%)

marine shales and comprise type Ⅱ kerogen. Kerogens generate pyrolysates diagnostic of

Paraffinic-Naphthenic-Aromatic Oil with low content of wax and sulphur. The bulk kinetic

frequency factors range from 1.54×1013 to 2.42×1013 and main activation energies range

from 50 to 53 kcal/mol. All these characteristics of Bowland Shale are quite similar to

productive Palaeozoic marine shale in the US such as Bakken, Barnett and Woodford shale.

The Bowland Shale possesses a high secondary gas generation potential and primary oil,

primary gas and secondary gas reach their maximum generation at 137, 150 and 230℃

respectively in geological time.

In the combination of phase properties and burial history, different driving forces of

expulsion can be concluded and produced GORs are predicted. Expulsion efficiency varies

with the organic maturity and reservoir conditions, and the amount and property of the

unconventional resource left in the source rock change accordingly.

Vast differences can be found in secondary gas amount prediction when varying default

kinetics models are chosen which emphasizes the significance of a targeted secondary

kinetics model in shale gas resource basin modelling evaluation.

2.7 Acknowledgements

The authors thank Ferdinand Perssen for technical assistance. We are also grateful to

editors and two anonymous reviewers for their constructive comments and suggestions.

This project was funded by China Scholarship Council.

Reprinted with permission from Yang, S., and B. Horsfield, 2016, Some predicted effects of

minerals on the generation of petroleum in nature: Energy & Fuels, v. 30, p. 6677-6687 (postprint),

doi: 10.1021/acs.energyfuels.6b00934. Copyright (2016) American Chemical Society.

3. HEATING RATE DEPENDENCY OF

MME

3.1 Abstract

The presence of some minerals can strongly influence the composition of laboratory

pyrolysates. The question is whether similar effects may also occur in nature, thereby

influencing the gas-oil ratio and other bulk compositional characteristics. A series of

experiments have been conducted at varying heating rates to examine this issue. Three

source rocks that vary significantly in mineralogy (a quartz-rich, a calcite-rich and a clay-

rich sample), namely the Alum Shale, Bowland Shale, and Toolebuc Shale, respectively,

were tested by Rock-Eval pyrolysis, open-system pyrolysis gas chromatography (PyGC)

and bulk kinetics parameters to check for the existence or otherwise of mineral matrix

effects (MME). Kerogen and whole-rock samples were then pyrolyzed at three heating

rates using closed-system pyrolysis to examine the heating rate dependency on

hydrocarbon aromaticity, gas-oil ratio and total yield. The solvent extract of one Bowland

Shale sample was used as a reference material for the natural system when extrapolating the

results from laboratory experiments to nature. A comparison of the natural reference

sample with kerogen- and whole-rock pyrolysates using Fourier Transform - Ion Cyclotron

Resonance Mass Spectrometry was also made, providing insights into NSO compounds in

laboratory and natural environments. The MME in Alum Shale, Toolebuc Oil Shale and

Bowland Shale has negligible, weak and strong influences on Rock-Eval, PyGC and bulk

kinetic results, respectively. MME on the hydrocarbon aromaticity total yield are heating

rate dependent, with decreasing heating rates, the effect is weakened. Bowland kerogen

pyrolysate resembles natural products more in certain NSO class ratios compared with its

whole-rock counterpart. The MME is speculated to be induced by the fast heating rates

and higher temperatures in the laboratory, and it is concluded that the effects do not occur

in the geological maturation process.

3.2 Introduction

Petroleum consists of an exceedingly complex mixture of hydrocarbons and non-

hydrocarbons, extending from methane to macromolecular aggregates. The relative

proportions of these components are quite variable and depend initially upon the nature of

the kerogen in the parent source rock and its level of maturity at the time of expulsion, and

subsequently upon the pressure and temperature conditions of the source-carrier-reservoir

system during expulsion, migration and accumulation. While the role played by diagenetic

minerals on reservoir quality is well known, for example the formation of clays from

feldspars or the conversion of smectite to illite (Bethke et al., 1986; Velde and Espitalié,

1989), it remains unclear as to whether bulk petroleum composition is strongly influenced

by retention and/or catalytic processes in source rocks. It has been suggested that clay

minerals can be used in the search for oil as a result of its importance in petroleum

generation and expulsion (Weaver, 1960). For instance, clay can effect isomerization,

disproportion of hydrogen and polymerization of unsaturated hydrocarbons (Frost, 1945;

Grim, 1947), and might, therefore, be responsible for the presence of aromatics and the

absence of olefins in petroleum (Brooks, 1948, 1952). Mango (1990) proposed that the

transition metals, captured from sedimentary waters by chlorophyll, are the catalytic agents

that convert n-alkane biolipids into the rearranged light hydrocarbons in petroleum. At the

molecular level, diasterane/sterane ratios for biomarkers are commonly related to clay

content (Rubinstein et al., 1975; van Kaam-Peters et al., 1998), because clays catalyse the

formation of diasterane precursors. The high heterogeneity of both organic matter and

minerals in gas shales argues that organic-inorganic interactions may be quite different

depending upon the depositional environment (Bernard et al., 2012).

The purported importance of organic-inorganic interactions during petroleum formation in

nature stems largely from pyrolysis experiments. For example, Jurg and Eisma (1964)

produced hydrocarbons by decarboxylating a fatty acid mixed with kaolinite in the presence

and absence of water; Shimoyama and Johns (1971) produced the same using

montmorillonite. β cleavage produced a hydrocarbon with two carbon atom atoms less

than the parent fatty acid when pyrolysis was conducted in the presence of calcium

carbonate (Johns and Shimoyama, 1972). The so called Mineral Matrix Effects (MME)

(Espitalie et al., 1980) can bring about changes in the gas-oil ratio, gas wetness and fluid

aromaticity pyrolysates, as reported from laboratory open- and closed-system experiments

including Rock-Eval (Espitalié et al., 1984; Katz, 1983), pyrolysis-GC (Horsfield and

Douglas, 1980), pyrolysis-FT IR (Öztaş and Yürüm, 2000), pyrolysis-GC/MS

(Tannenbaum et al., 1986b), bulk kinetics determination (Dembicki, 1992; Dessort et al.,

1997), PhaseKinetics modelling (Yang et al., 2015) and hydrocarbon expulsion simulation

(Lewan et al., 2014). For clay minerals, two main mechanisms are active: (1) due to the high

surface area (Sing, 1985), they tend to adsorb heavy compounds generated in pyrolysis,

hence influence the pyrolysis maturity indicator (Tmax), composition and quantity of the

pyrolysate (Horsfield et al., 1983), as well as the kinetics of the reaction, (2) a selective

catalytic feature of the clay minerals can change the gas-oil ratio (GOR), aromaticity,

oxygen index (OI) and other compositional features of the products (Dembicki, 1990; Wu

et al., 2012). It has to be pointed out that not every source rock in pyrolysis experiments

necessarily has to suffer from the MME. Horsfield and Douglas (1980), Horsfield et al.

(1983) and Katz (1983) concluded that MME varies according to the TOC content and

mineralogy of the samples under investigation. A very high TOC (>6%) or low clay

content can decrease even avoid the MME (Reynolds and Burnham, 1995). Tannenbaum

and Kaplan (1985a), Pan et al. (2010) and Lewan et al. (2014) emphasized that the

presence of liquid water in the pyrolysis system could significantly attenuate the activity of

the clay catalysing function , although Eglinton et al. (1986) and Behar et al. (2010) noted

that the function of water in pyrolysis is limited.

One of the most significant differences between laboratory pyrolysis and geological

maturation lies in the heating rate. Geological heating rates normally fall in the range 10-10

to 10-12 K/min, while a typical laboratory pyrolysis heating rate is not slower than 10-

1K/min. With that being said, the slowest experimental heating rate is 300°C over 6-years,

or 10-5K/min (Saxby and Riley, 1984) which is not realistic for a routine test. Retort yield

from Green River Shale is reported to decrease at lower heating rates (Burnham and

Singleton, 1983). From a petroleum generating kinetics perspective, during non-isothermal

pyrolysis of immature oil shale and coal, the onset and Tmax of organic transformation

reactions are shifted to higher temperatures with increasing rate of heating (Burnham et al.,

1987; Schenk et al., 1997). For the MSSV pyrolysis of Duvernay Shale, aromaticity and the

unresolvable GC “hump” decrease with decreasing heating rates (Dieckmann et al., 2000a).

Similarly, pyrolysates are always richer in polar and aromatic compounds than petroleums

(Horsfield, 1997) and are higher in gas wetness (di Primio and Horsfield, 2006) compared

with natural products irrespective of the type of pyrolysis being employed. Karabakan and

Yürüm (1998) reported the influence of heating rate to MME, but instead of using kinetic

derived temperatures they simply heated samples to certain fixed temperatures under

different heating rates which is less meaningful in a geological sense. The comparison of

these pyrolysates does not reflect the heating rate dependency in pyrolysis; it is mainly

controlled by maturity instead. Only the products that were generated at the same TR are

directly comparable.

It would represent a significant step forward if MME in pyrolysis experiments could be

extrapolated to geological systems by consideration of heating rate as a parameter. In this

paper, we evaluate the MME on three types of source rocks which have significantly

different mineralogies (a quartz-rich, a calcite-rich and a clay-rich sample) using Rock-Eval,

open pyrolysis and bulk kinetic modelling. Two types of samples with strong MME were

applied to a closed-system (MSSV) pyrolysis to investigate the heating rate dependency of

MME on aromaticity, GOR and bulk generation amounts. The projected changes in MME

as a function of heating rate were firstly extrapolated from laboratory to nature, and the

predictions compared with the results of thermovaporisation of a natural bitumen sample.

The ultrahigh-resolution Fourier transform ion cyclotron resonance mass spectrum (FT-

ICR MS) technique, which has been applied to identify acids and heteroatom compounds

in crude oil (Hughey et al., 2002), asphaltene (Klein et al., 2006), oil sand (Barrow et al.,

2004) and coal extracts (Wu et al., 2003), etc., was first induced into MME research here to

compare the major compound classes present in whole-rock and kerogen pyrolysates with

those in natural bitumen to elucidate whether MME occurs in nature.

3.3 Samples and Analytical Methods

3.3.1 Samples

Three immature whole-rock samples from the Alum Shale, Toolebuc Oil Shale and

Bowland Shale, as well as their kerogen concentrates were studied in this research using

pyrolysis (Table 3.1). The solvent extract of one Bowland Shale was used as a reference.

The Alum Shale sample taken from a core drilled in Central Sweden is of Early Ordovician

age. This marine shale was deposited in an inner shelf facies and is characterized by an

enrichment of uranium (>200ppm) which was diffused from seawater across the

sediment/water interface (Schovsbo, 2002). Quartz (80%) is a major mineralogical

component, together with K-feldspar (18%) (Schulz et al., 2015).

As one of the most important oil shale plays in Australia, the marine Toolebuc Formation

of Early Cretaceous age underlies about 484,000 km2 of the Eromanga and Carpenteria

Basins (Dyni, 2006). The outcrop Toolebuc Oil Shale samples studied here comes from

Julia Creek where the oil shale was deposited in an epicontinental sea environment

(Boreham and Powell, 1987) and has a calcite content over 45% (Patterson et al., 1986).

Table 3.1. Generalized information and Rock-Eval & TOC data of the samples tested in the research.

Sample

Name

Locatio

n Age Mineralogy Sample

type Tmax

(°C)

HI (mg HC/g

TOC)

OI (mg

CO2/g

TOC)

TOC

(%)

Alum

Shale

Middle

Sweden

Early

Ordovician

Quartz-rich

(80%)

whole-

rock 418 357 4 16.7

kerogen 420 356 6 41.1

Toolebuc

Oil Shale

Northea

stern

Australi

Early

Cretaceous

Calcite-rich

(+45%)

whole-

rock 417 457 24 13.1

kerogen 421 426 31 52.5

Bowland

Shale

Norther

England

Late

Carbonifer

ous

Clay-rich

(+50%)

whole-

rock 431 194 10 3.17

kerogen 426 302 4 24.2

The Namurian (Late Carboniferous) Bowland Shale sample is from 31.02m in the

Carsington Dam Reconstruction C4 borehole, which was drilled in the Widmerpool

Trough, Northern England. Thin section observation and carbon isotope research have

previously shown that this Bowland Shale sample is a thin-bedded carbonate-bearing clay-

rich mudstone, its bulk δ13Corg value of -28.8 ‰ indicating that the kerogen is derived

from marine planktonic algae (Konitzer et al., 2014). The mineralogy of the marine

Namurian Bowland Shale in Northern England is characterized by a high concentration of

clay (+50%) and a moderate content of quartz (22.1±3.9 %) (Spears and Amin, 1981).

Among the clay minerals, kaolinite is the most abundant, constituting 20.8±7.5 % of the

total minerals; montmorillonite and illite contents are low.

The reference Bowland Shale core sample was taken from 1,404 metres depth in Old

Dalby. The whole-rock sample has an HI of 188 mg/g TOC which is similar to the

pyrolyzed whole-rock (HI: 194 mg/g TOC; see results section). With a Tmax of 437°C, the

shale is considered to be an early oil-window matured sample which is suitable as a

reference for comparing artificial and natural products.

3.3.2 Analytical Methods

3.3.2.1 Kerogen isolation and screening

Kerogen concentrates were isolated from whole-rock samples by (1) crushing the shale

sample to sub-millimetre size, (2) treating with hydrochloric acid/6N hydrofluoric acid [2:1]

for one week at room temperature and (3) sieving to 10-500 microns.

Rock-Eval and TOC analyses were performed using a Rock-Eval 6 and Leco SC-632

Analyser respectively following established procedures.

3.3.2.2 Pyrolysis gas chromatography (PyGC)

PyGC was performed using the Quantum MSSV-2 Thermal Analysis System® interfaced

with an Agilent GC-6890A (Horsfield et al., 2014). Milligram quantities of each sample

were loaded into a small open glass tube and heated under flowing helium; free

hydrocarbons were vented for 3 minutes during an isothermal purge at 300°C, after which

the C2+ pyrolysis products generated during heating from 300°C to 600°C were collected in

a cryogenic trap (liquid nitrogen). Methane passed through the trap and passed through the

GC column to the Flame Ionisation Detector (FID). Trapped products were then liberated

by removing the cooling agent and heating the trap to 300°C. An HP-Ultra 1

dimethylpolysiloxane capillary column connected to the FID was employed using helium as

carrier gas. Quantification of individual compounds and boiling range splits was conducted

by external standardisation with n-butane.

3.3.2.3 Bulk kinetics

Bulk pyrolysis was performed using a Source Rock Analyser® (SRA) at three different

heating rates (0.7, 2 and 5 K/min) following established procedures (Burnham et al., 1987).

The discrete activation-energy (Ea) distribution optimization with a single frequency factor

(A) as well as geological extrapolation were performed using the KINETICS 2000® and

KMOD® programmes. The corresponding temperatures of Transformation Ratio (TR)

30%, 50% and 70% for each of the three different heating rates (0.7, 2 and 5 K/min) were

selected for MSSV pyrolysis and measurement of pyrolysate composition.

3.3.2.4 Micro scale sealed vessel (MSSV) pyrolysis-GC

As described by Horsfield et al. (2014), milligram quantities of samples were sealed in glass

capillaries and artificially matured to temperatures corresponding to 30, 50 and 70 TR for

each of the heating rates 0.7, 2 and 5 K/min respectively. The tubes were then cracked

open using a piston device coupled with the injector, and the released products were swept

into the GC using a flow of helium. Quantification was performed by external

standardisation using n-butane.

3.3.2.5 Thermovaporisation-GC

Around 10 mg of coarsely crushed reference shale was weighed into MSSV glass capillary

tubes, which were then sealed by an H2 flame after having reduced the internal volume

with pre-cleaned quartz sand. After introduction into the Quantum MSSV-2 Thermal

Analysis System, the external surfaces of the tube were purged for 5 min at 300°C, during

which time volatiles were mobilised within the tube; thereafter the tube was cracked open

by a piston device to transfer the products into a liquid nitrogen-cooled trap. The

composition of these volatiles was analysed as described under PyGC.

3.3.2.6 MSSV-FT-ICR MS

Aliquots of MSSV pyrolysates were extracted using dichloromethane and methanol (V/V

9:1). Mass analyses were performed in negative ion ESI mode with a 12 T FT-ICR mass

spectrometer equipped with an Apollo II ESI source, both from Bruker Daltonik GmbH.

Nitrogen was used as drying gas at a flow rate of 4.0 L/min and a temperature of 220 °C

and as nebulizing gas with 1.4 bars. The sample solutions were infused at a flow rate of 150

μL/h. The capillary voltage was set to 3000 V and an additional CID (collision-induced

dissociation) voltage of 70 V in the source was applied to avoid cluster and adduct

formation (Poetz et al., 2014).

3.3.2.7 Analysis of reference material

A Soxhlet extractor was used for extraction of the reference sample bitumen. Core sample

material (20 g) was filled in an extraction tube and extracted with a solvent mixture of

dichloromethane and methanol (v/v = 99:1) at 40 °C for 24 h. The bitumen was then

analysed by FT-ICR MS as described above.

3.4 Result and Discussion

3.4.1 The existence of MME

3.4.1.1 Rock-Eval and TOC

Significant differences in maturity parameters and organic matter type identification indices

can be found between Bowland and Toolebuc whole-rock-kerogen pairs while Alum Shale

sample pairs seem to be less influenced by MME (Table 3.1 and Fig. 3.1). The clay-rich

Bowland whole-rock sample manifests higher Tmax (Fig. 3.1a), higher OI (Fig. 3.1b) and

lower HI (Fig. 3.1b) than its kerogen counterpart and these make the Bowland whole-rock

samples seem to be more mature and more terrestrial in origin as when compared with the

kerogen. By comparison, the calcite-rich Toolebuc Oil Shale whole-rock sample was

influenced by a less significant MME, and in the opposite way to that of the Bowland

samples, namely the Toolebuc whole-rock signature looks less mature and has a better

organic matter type (Fig. 3.1). A negligible MME in Tmax and OI was noted for the quartz -

dominated Alum Shale samples.

The MME induced by clay minerals in these Rock-Eval tests resemble those reported by

Dembicki et al. (1983),Espitalié et al. (1984)and Heller-Kallai et al. (1984) who concluded

(1) the retention of relatively heavy compounds on clay minerals is responsible for the

increase of Tmax and decrease of HI in whole-rock samples and (2) the selective catalytic

effect of acid minerals in generating CO2 (Larsen and Hu, 2006) explains why the OI of

whole-rock sample was shifted to higher values. The impact of MME on the calcite-rich

Toolebuc Oil Shale samples supports the work of Katz (1983) who reported carbonate

mineral can enhance the HI and hinder the OI.

Fig. 3.1. Basic geochemical screening based on Rock-Eval & TOC of whole-rock and kerogen samples.

3.4.1.2 PyGC

The PyGC analysis of source rocks is a technique which can make a quick evaluation of

kerogen structure (Dembicki et al., 1983; Van de Meent et al., 1980) and relate that

structure to bulk petroleum composition by means of Petroleum Type Organofacies

(Horsfield, 1989). As a Lower Palaeozoic marine shale, it is not typical for the Alum Shale

to be rich in low molecular weight and aromatic pyrolysis products, e.g., xylene and toluene

(Fig.3. 2 and 3.3), this unique characteristic could be induced either by an unusual

precursor biota or by effects related to the presence of uranium (Horsfield et al., 1992b).

Negligible differences in the pyrolysates of Alum whole-rock and kerogen can be observed

which implies the quartz does not bring about MME because of its inert nature, possibly

combined with the already aromatic nature of the pyrolysate (Fig. 3.2 and 3.3).

Fig. 3.2. Comparison of PyGC maps on whole-rock and kerogen pairs. Bowland Shale pyrolysate shows an

obviously higher aromatic compounds concentration compared with its kerogen counterpart. (benz: benzene, tol:

toluene, m,p xyl: meta- and para-xylene, o xyl: ortho-xylene.)

Fig. 3.3. Quick classification on the pyrolysates of both whole-rock and kerogen concentrates.

The pyrolysate of Toolebuc Oil Shale belongs to the Low Wax P-N-A (Paraffinic-

Naphthenic-Aromatic) oil Petroleum Type Organofacies (Fig. 3.3. a)(Horsfield, 1989).

Thiophenic sulphur compounds generated from the carbonate oil shale are more abundant

than in either of the two clastic sediments (Fig. 3.3. b, d). Although the differences of

kerogen and whole-rock pyrolysates are not significant (Fig. 3.2), it still can be recognised

that the kerogen products are more aromatic than the whole-rock counterpart (Fig. 3.3).

In contrast, pronounced differences can be found between the clay-rich Bowland Shale

whole-rock pyrolysis products and its kerogen pyrolysate, i.e., aromatic compounds are

more abundant, and there is an increased complexity of the compound mixture in whole-

rock products, when compared with kerogen that was heated alone (Fig. 3.2). Different

from the calcite-rich Toolebuc Oil Shale, Bowland Shale kerogen generates more high

molecular (Fig. 3.3a) and aliphatic products (Fig. 3.3 b-d).

The preferential catalytic effect of clay minerals on low molecular weight aromatic and

branched hydrocarbons has been attributed to cracking via a carbonium-ion intermediate

which forms on the Lewis acid sites of the clay (Tannenbaum and Kaplan, 1985b). The

MME induced by carbonate minerals was considered as less significant and opposite to clay

minerals (Hu et al., 2014; Tannenbaum et al., 1986b).

3.4.1.3 Bulk kinetics

The activation energy distribution and frequency factor of Alum Shale kerogen degradation

seem not to be influenced by the intimate presence of minerals during pyrolysis (Fig. 3.4).

Toolebuc kerogen shows slightly more refractory characteristics compared with the whole-

rock, and the Bowland Shale kerogen’s activation energy distribution was shifted to

obviously lower values when MME was eliminated (Fig. 3.4). These preferential shift

features of illite, calcite and clay minerals on bulk kinetic parameters are in agreement with

the stated effects of mineral matrices when building kinetic models (Dembicki, 1992),

(Dessort et al., 1997).

An instructive way of comparing the differences in kinetic parameters noted above is to

apply them to a geological heating rate, here chosen to be 3K/million year (Fig. 3.5).

Considering a TR of 50% for example, the geological temperature shifts between whole-

rock-kerogen pairs of Toolebuc Oil Shale and Bowland Shale are 5°C and 17°C,

respectively (Fig.3. 5). It is clear that these variations can lead to huge differences in the

estimated timing of hydrocarbon generation, expulsion and accumulation.

Fig. 3.4. Bulk kinetic parameters of whole-rock and kerogen samples. Negligible (Alum Shale), Small (Toolebuc

Oil Shale) and significant (Bowland Shale) differences can be figured out.

Fig. 3.5. Geological extrapolation (heating rate: 3K/million year) of bulk kinetics parameters and comparison of

comparison on whole-rock and kerogen samples.

3.4.2 The heating rate dependence of MME

With decreasing heating rates, the GC “hump” (unresolved complex mixture) and

alkene/alkane ratios of both Toolebuc and Bowland pyrolysates decrease, which is in

agreement with previous studies on the heating rate dependency of pyrolysis products

(Dieckmann et al., 2000a; Williams et al., 1990). Here, we discuss how aromaticity, GOR

and total yield were influenced by heating rates with and without minerals.

3.4.2.1 Aromaticity

For the pyrolysates of samples artificially matured to TR 50%, the lower the heating rate is,

the lower is the aromaticity (benzene, toluene, xylenes and tetramethylbenzene compared

with nearby normal alkanes). The biggest change is seen for the Bowland Shale whole-rock

sample (Fig. 3.6). Changes of the Bowland whole-rock pyrolysates are bigger than those of

their kerogen counterparts, pointing to the stronger heating dependence on organic-

inorganic interactions than on the thermal degradation of kerogen. By way of contrast, the

aromaticities of Toolebuc whole-rock and kerogen samples seem to not be strongly

controlled by heating rates at all, ostensibly because MME are weak (Fig. 3.7a).

Fig. 3.6. Pyrolysate GC traces of Bowland whole-rock sample at different heating rates when heated to TR 50%.

The data manifests that the slower the heating rate is, the more aliphatic the products would be generated. benz:

benzene, tol: toluene, EB: ethylbenzene, xyl: xylene, TMB: tetramethylbenzidine.

Fig. 3.7. Aliphatic/aromatic compounds ratio, GOR and bulk hydrocarbon generation/TOC variations of Bowland

Shale and Toolebuc Oil Shale samples in all three heating rates. Aliphatic compounds include all normal alkenes

and alkanes from C1-C30. Aromatic compounds are composed of benzene, toluene, ethyl benzene, xylenes,

tetramethylbenzidines, naphthalene, and branched naphthalenes.

3.4.2.2 Gas-oil ratio

Differences in GOR between the whole-rock and kerogen samples is another important

compositional influence caused by MME (Horsfield and Douglas, 1980). The GOR of

both Toolebuc and Bowland samples increase with decreasing heating rate (Fig. 3.7b). The

volatile generation in closed system pyrolysis was contributed by kerogen and oil cracking,

a slower heating rate which allows longer time for cracking is possibly the reason why

GOR is elevated with decreasing heating rate (Gibbins-Matham and Kandiyoti, 1988).

GOR of the clay-rich Bowland whole-rock sample changes much more significantly with

varying heating rates compared with the other pairs (Fig. 3.7b) because of MME.

3.4.2.3 Generated Product Yield

Significant differences in bulk generation product yields (including unresolved complex

mixture in the GC “hump”, normalized by TOC) can be found between whole-rocks and

kerogen concentrates of Toolebuc and Bowland samples (Fig. 3.7c). Whole-rocks of the

Toolebuc Oil Shale possess higher bulk generation/TOC than their kerogen counterparts

while Bowland Shale samples show the opposite trend (Fig. 3.7c), which correlates with HI

changes caused by the MME (Fig. 3.1). Gross hydrocarbon generation by kerogen samples

is roughly independent of heating rate, which is consistent with the prerequisite of kinetic

theory (Braun and Burnham, 1987; Dieckmann et al., 2000a). In contrast, the quantity of

whole-rock product depends on heating rate especially in the case of the Bowland Shale

samples (Fig. 3.7c). It seems that with decreasing heating rate, the MME on bulk generation

amount can be diminished, and whole-rock generation yields can approach those of its

kerogen counterpart. It has to be pointed out that the “pseudo HI” in MSSV pyrolysis (Fig.

3.7c) is not directly comparable to Rock-Eval derived HI because: (1) the highest TR in

MSSV pyrolysis is 70% in this research, while Rock-Eval HI is achieved when TR has

reached 100%; (2) MSSV products were introduced to the GC column where part of the

pyrolysate is retained at the GC-interface, whereas very little pyrolysate is lost during direct

FID ignition in Rock-Eval (Horsfield, 1997); and it should be noted that (3), coke can be

formed in the low-pressure closed-system MSSV pyrolysis but only at high TR. The first

two of these factors explain why “pseudo HI” in MSSV is much lower than the HI

provided by Rock-Eval. Anyway, the gross hydrocarbon quantity variation induced by

MME in both of these systems should be similar.

3.4.3 Geological Calibration

The aliphaticity of thermal extracts from the natural bitumen in Bowland Shale is always

higher than artificial pyrolysis products generated from the immature starting material (Fig.

3.8a). With slower heating rates, aliphaticities of both whole-rock and kerogen pyrolysates

increase. In the logarithmic coordinates system, the changing rate of whole-rock products

is obviously faster than kerogen pyrolysates which implies that in a geological heating rate

the differences between these two materials become smaller, in another words, the

aliphaticity difference is very likely to have disappeared completely. Geological

extrapolation trend lines from artificial pyrolysis show that the prediction of aliphaticity

roughly matches that of the natural rock bitumen chosen here for calibration, although,

geometrically speaking, there are many possibilities for smoothing the trend lines from the

limited dataset shown here.

From the perspective of gross hydrocarbon yield, the heating rate independency of kerogen

breakdown implies that the quantity predicted in the laboratory using kerogen is

comparable to the geological situation (Fig. 3.8b). Obviously, this is not the case for direct

measurements on whole-rock samples; but using extrapolated data, the slower the heating

rate, the closer the generation amount of Bowland whole-rock is to its kerogen counterpart

(Fig. 3.8b). This observation could be explained by deducing that the MME on gross

hydrocarbon generation is quantitatively heating rate dependent and a slower pyrolysis

heating rate could reduce the effectiveness of MME on hydrocarbon retention.

Fig. 3.8. The variation trends of aliphatic/aromatic ratios and bulk generation according to changing heating rates

on Bowland Shale samples when they are heated to TR 50%. Natural bitumen reference was shown in Fig. 3.8a.

3.4.4 Insights into Hetero-element Geochemistry

The distribution of elemental classes in source rock extracts and crude oil revealed by FT-

ICR MS is largely determined by deposition environment (Chiaberge et al., 2013; Hughey et

al., 2002), maturity (Oldenburg et al., 2014; Poetz et al., 2014) and secondary alteration

effects such as migration fractionation (Liu et al., 2015), biodegradation (Kim et al., 2005;

Pan et al., 2013) and thermochemical sulphate reduction (Walters et al., 2015). Here we

examined pyrolysates to see if MME affects heteroelement distributions.

The pyrolysates of Bowland kerogen and whole-rock samples show a decrease in oxygen

and increase in nitrogen compounds with increasing TR (Fig. 3.9) thereby resembling

changes noted for natural solvent extracts of Type II source rocks with increasing thermal

maturity (Poetz et al., 2014). The gross oxygen, nitrogen and sulphur composition of

kerogen and whole-rock pyrolysates are rather similar which is surprising considering that

significant MMEs have described earlier. However, a detailed comparison of the

compound class ratios of O1/O2, N1/N2-4 and N1O1/N1O2 demonstrates significant

differences do indeed occur (Fig. 3.10). Kerogen pyrolysates show much higher ratios in

O1/O2, N1/N2-4 and a lower N1O1/N1O2 ratio. Although it was reported that O2 class

distribution can be changed by contamination and biodegradation (Kim et al., 2005; Pan et

al., 2013), all samples were tested by the sample instrument on the same day, these factors

can be safely excluded. Since the values of the compound class ratios are small and

relatively constant with increasing TR, it can be demonstrated that no significant heating

rate dependency occurs within either kerogen or whole-rock samples. More importantly,

the O1/O2, N1/N2-4 and N1O1/N1O2 ratios of the natural Bowland Shale extract being used

for calibration are 0.31, 0.30 and 0.59 respectively which are rather close to kerogen

products but not whole-rock pyrolysates (Fig. 3.10). Since the reference sample is in the

early oil window maturity and occurs within a thick shale layer (over 20m), very limited

expulsion is anticipated. The low ratios of O1/O2 and N1/N2-4 of whole-rock sample appear

to be caused by the catalytic effect of minerals which tends to bring about condensation of

oxygen and nitrogen atoms into aromatic ring systems.

The great resemblance of kerogen pyrolysates and shale extract in the three compound

class ratios implies that though kerogen MSSV pyrolysis can’t accurately reflect the

occurrence of NSO compounds in natural bitumens, it does provides better prediction

than whole-rock pyrolysis. The vast differences between artificially heated whole-rock and

naturally matured shale extract on both elemental gross compounds and compound class

ratios manifest that MME caused by clay minerals on NSO generation is heating rate

related, and that is why the extreme slow geological heating generates similar product as

kerogen pyrolysate although there are minerals in the shale. In brief, MME on both

hydrocarbon and NSO compounds generation are obviously heating rate dependent and

the slower the heating rate is, the weaker the effects of MME. Thus, it can be speculated

that the MME only exists in a laboratory environment, but is not significant in the

geological maturation process.

Fig. 3.9. Elemental class distribution pie charts of pyrolysates and matured shale extract in the negative ESI

spectra assigned with molecular formulas.

Fig. 3.10. “Class” comparison of Bowland kerogen and whole-rock pyrolysates. The O1/O2, N1/N2 and N1O1/N1O2

ratios of the reference matured Bowland Shale extract are 0.31, 0.30 and 0.59, respectively, which are more

resemble to kerogen rather than whole-rock sample.

3.5 Conclusions

1. The MME in the quartz-rich Alum Shale, calcite-rich Toolebuc Oil Shale and clay-

rich Bowland Shale has negligible, week and strong effluence on Rock-Eval, PyGC

and bulk kinetics result respectively. Kerogen type, maturity, organic facies and

kinetics can be influenced accordingly.

2. MME on the hydrocarbon aromaticity and generation amount are heating rate

dependent, with decreasing heating rate, the effect is weakened. The heating rate

dependency of whole-rock pyrolysate GOR is stronger than MME and gets

enhanced when heating rate is slowed down.

3. Neither the Bowland kerogen nor the whole-rock pyrolysates show similar gross

elemental composition as natural products. However, kerogen pyrolysate resembles

natural products more in certain NSO class ratios compared with whole-rock

pyrolysate, which implies MME on NSO compounds is diminished in a geological

situation.

4. The MME is speculated to only exist in the laboratory environment, not in a

geological maturation process. The MME doesn’t change kinetic of hydrocarbon

generation nor the composition of products in nature; it only affects laboratory

pyrolysis results in assessing kerogen type, maturity, organic facies or kinetics.

3.6 Acknowledgments

This study is financially supported by the Chinese Scholarship Council. The authors wish

to thank Prof. Michael Stephenson (BGS), Dr. Christopher Vane (BGS), Dr. Chris

Boreham (Geoscience Austrilia) and Grippen Oil & Gas, Sweden for providing shale

samples. Cornelia Karger and Ferdinand Perssen in GFZ are acknowledged for their

technical support. We also thank three anonymous referees for their insightful reviews of

the manuscript.

This chapter has been submitted to Geochimica et Cosmochimica Acta on 27 Jul 2017 (pretprint).

4. URANIUM IRRADIATION ON

PETROLEUM GENERATION

4.1 Abstract

An interdisciplinary geochemical study covering Rock-Eval pyrolysis, pyrolysis-gas

chromatography, thermovaporisation-gas chromatography and Fourier Transform Ion

Cyclotron Resonance mass spectrometry (FT-ICR MS) was carried out to unravel the

organic-inorganic interactions caused by radiogenic decay of uranium in immature organic-

rich Alum Shale (Middle Cambrian-Lower Ordovician).

The uranium content is correlated with the gas-oil ratios as well as the aromaticity of the

pyrolysates of immature samples, indicating a strong uranium irradiation effect on the

quantity and quality of the petroleum potential. Also, the gas-oil ratios and aromaticity of

thermovaporisation results resemble those in pyrolysates, and proves that the influence of

uranium irradiation on the petroleum generation is valid in nature. The FT-ICR MS data

reveal that the macro-molecules in the uranium-rich Alum Shale samples are less alkylated

and provide evidences for kerogen structures alteration by irradiation to a more gas- and

aromatic-prone though geological time.

The radiation dosage resulting from the decay of uranium is linearly correlated with

uranium content and impact time, whereas the kerogen structure changes exponentially

since labile structures react early and become stabilized in later stages. As a result, the gas

percentage and aromaticity of petroleum generated during the oil window time were

calculated pointing to more oil- and aliphatic-prone characteristics than those in pyrolysis

experiments. In addition, the gas sorption capacity of the Alum Shale is assumed to be less

developed during Palaeozoic times in contrast to result suggested by sorption experiment

preformed present day. The kerogen restructuring avoids over-estimation on gas

generation and gas retention in the Alum Shale and can reduce exploration risks

significantly.

4.2 Introduction

The potential role played by uranium (U) in petroleum generation was recognised a long

time ago. Alpha particles, which are the main products of uranium decay, were suggested to

be important (Lind and Bardwell, 1926). Experimental data showed that fatty acids can be

decarboxylated by alpha particle radiation at 130oC to form hydrocarbons (Sheppard and

Burton, 1946). A correlation between uranium concentration and oil yield in Chattanooga

Shale was presented by Swanson (1960). Later, it was found that the total organic carbon

(TOC) content is proportionally related to the uranium content (Leventhal, 1981; Swanson

and Swanson, 1961), and this finding is routinely applied in using gamma-ray spectral

logging to delineate the occurrence of organic rich shales (Schmoker, 1981; Serra, 1983).

Uranium irradiation is responsible for bringing about changes in organic matter

composition. The reflectance of vitrinite in humic coal can be enhanced by the crosslinking

of polymers caused by radiolytic effects from uranium (Breger, 1974). The atomic H/C and

O/C ratios which are used to define the kerogen type and thermal maturity (Durand and

Espitalié, 1973) were also considered susceptible to uranium irradiation (Pierce et al., 1958).

The uranium enrichment can lead to decreased aliphatic biomarker concentrations

(Hoering and Navale, 1987) and may furthermore influence the aromatic biomarker

distributions (Dahl et al., 1988a, b; Lewan and Buchardt, 1989). Leventhal and Threlkeld

(1978) showed that the 13C/12C ratios are correlated to the log of the uranium

concentrations, and this finding was confirmed by further investigations on shale (Dahl et

al., 1988b) and bitumen samples (Court et al., 2006). Marine shale samples with high

uranium contents produce atypical pyrolysates in that there is a significantly enhanced

percentage of gas and aromatic compounds relative to long chain aliphatics (Horsfield et al.,

1992a; Leventhal, 1981).

These comprehensive investigations played fundamental roles in understanding the

influence of uranium irradiation on petroleum generation. However, several key issues are

still under debate.

(1) The actual processes leading to changes of the Hydrogen Index (HI), Oxygen Index

(OI) and Tmax in uranium-rich samples which have undergone similar maturation during

burial are not clear. Forbes et al. (1988) and Landais (1996) reported that OI and Tmax

values of vitrinite in Akouta uranium deposit (Niger) tend to be increased by increasing

uranium contents. In contrast, very low OI values have been reported for the uranium-rich

Alum Samples (Schulz et al., 2015) and Dahl et al. (1988b) suggested that Tmax is inversely

proportional to uranium content.

(2) Due to limited sample amounts and uranium depletion in response to surface

weathering, no clear correlation could be found between uranium contents and the atypical

pyrolysate characteristics of the Alum Shale in a study by Horsfield et al. (1992a). This led

to the hypothesis that a unique algal biomolecule might be the precursor for the gas- and

aromatic-rich properties (Bharati et al., 1995). Importantly, the mechanism of uranium

irradiation in changing the petroleum generation potential is still a matter of debate. It is

not known which alteration processes change the kerogen structures in response to

irradiation and how these changes influence the petroleum generation and occurrence.

(3) Up until now most studies have only featured pyrolysis experiments (Court et al., 2006;

Dahl et al., 1988b; Horsfield et al., 1992a; Leventhal, 1981). Studies utilising natural

laboratories have not been employed to study the effect of uranium irradiation on

petroleum generation. It is not known whether the atypically generated hydrocarbons

represent structural moieties caused by uranium irradiation or the effects are secondary and

related to the high-temperatures and fast heating rates during pyrolysis experiments.

(4) In either case pyrolysis experiments based on black shale samples in their current state

could provide misleading information about petroleum generation because the kerogen

structures hundreds of million years ago at the very start of the irradiation history could

have been essentially the same as what we would consider typical for marine shales, with a

more paraffinic character (Horsfield, 1989).

The Alum Shale (Middle Cambrian – Lower Ordovician) holds the largest low-grade

uranium resource in Europe. Here, we present new interdisciplinary geochemical data from

investigations of the Alum Shale followed by a discussion of the implications regarding the

influence of uranium on petroleum generating potential. Fresh immature samples have

been systematically investigated using inductively coupled plasma-mass spectrometry (ICP-

MS), pyrolysis methods, thermovaporisation- gas chromatography (Tvap-GC), Fourier

Transform Ion Cyclotron Resonance mass spectrometry (FT-ICR MS), and a new method

then proposed for back-calculating the aromaticity and chain length distributions of the

original kerogen structures.

4.3 Study Area and Samples

The Baltic Basin covers parts of the southern Baltic Sea, the Kaliningrad Oblast, Northern

Poland and the western parts of the Baltic States (Fig. 4.1), and contains sediments ranging

in age from the Early Cambrian to the present day. The basin fill is thin in the north-

eastern part and thickens toward the Teisseyre-Tornquist (Ulmishek, 1990).

The Alum Shale is considered as one of the most important source rocks for oil and gas in

the Baltic Basin because of its wide occurrence, considerable thicknesses and high TOC

contents (Buchardt, 1999; Kotarba et al., 2014b). Named after the hydrated potassium and

aluminium-bearing sulphate [KAl(SO4)2·12H2O], the Alum Shale Formation is a formal

name for the collective of Middle Cambrian, Upper Cambrian (Furongian) and Lower

Ordovician (Tremadocian) shale (Andersson, 1985; Thickpenny, 1984). This shale is widely

distributed within and around the Baltic Basin, and can be as thick as 180 metres in

offshore Denmark (Nielsen and Schovsbo, 2006) and 90 metres in southern Sweden (Pool

et al., 2012). The TOC content of Alum shales is typically higher than 2 wt.% (Schovsbo,

2003) and up to 22 wt.% in Middle Sweden (Kosakowski et al., 2016).

34 Alum Shale samples, covering different ages, were analysed from Sweden, Estonia and

Russia (Fig. 4.1). Most of the samples are from boreholes, five samples were carefully

selected from outcrops (Table 4.1). These samples are of low maturity according to

previous maturity assessments based on the reflectance of vitrinite-like macerals (Buchardt

et al., 1997; Petersen et al., 2013), and are thus suitable for studies of petroleum potential.

Fig. 4.1. Geographical overview of the Alum Shale sample distribution. The grey dash line depicts the boundary of

the Baltic Basin, and the red dash lines represent the isolines of vitrinite-like maceral reflectance of the Alum

Shale modified after Buchardt et al. (1997). Ages of the samples are given in coloured circles.

Table 4.1. Background information, uranium contents, and Rock-Eval & TOC data of the Alum Shale samples.

Age Name Type Well/Place Uranium

(ppm)

TOC

(%)

Rock-Eval

S1 (mg/g)

(mg/g)

Tmax (°C)

HI (mg HC/g

TOC) OI (mg CO2/g

TOC)

Lower

Ordovician

LO-1

borehole

Saint Petersburg

11.1

0.3

23.7

409

214

LO-2

borehole

Saint Petersburg

190

9.0

0.3

17.6

412

195

LO-3

borehole

Saint Petersburg

244

13.6

0.4

21.9

411

161

LO-4

borehole

Saint Petersburg

274

6.2

0.2

5.9

414

LO-5

borehole

Saint Petersburg

110

9.4

0.1

8.4

419

LO-6

borehole

NA-3

136

14.1

0.8

53.6

419

380

LO-7

borehole

F-342

107

8.1

0.3

33.8

406

416

LO-8

borehole

P-1949

119

12.4

2.5

47.6

405

385

LO-9

outcrop

Ottenby

8.1

0.6

30.1

441

374

Upper

Cambrian

UCm-1

borehole

OA-1

155

16.7

1.0

59.6

417

387

UCm-2

borehole

GH-2B

413

21.7

1.3

73.4

426

338

UCm-3

borehole

KN-1A

135

13.1

1.9

50.5

425

385

UCm-4

outcrop

Kakeled

186

21.7

2.2

83.4

416

384

UCm-5

outcrop

Kakeled

194

11.1

0.6

42.2

418

381

UCm-6

borehole

Hällekis-1

201

11.6

1.4

57.4

413

497

UCm-7

borehole

Hällekis-1

177

14.6

1.0

53.1

417

365

UCm-8

borehole

Hällekis-1

4.1

0.2

11.6

420

284

UCm-9

borehole

Hällekis-1

142

3.0

0.5

8.8

419

298

UCm-10

borehole

Hällekis-1

130

14.0

0.9

60.9

420

435

UCm-11

borehole

Hällekis-1

109

12.6

0.9

59.0

420

467

UCm-12

borehole

Hällekis-1

13.4

0.9

64.1

421

479

UCm-13

borehole

Hällekis-1

22.1

1.9

138.1

426

624

UCm-14

borehole

Hällekis-1

10.7

1.3

52.2

424

490

Middle

Cambrian

MCm-1

outcrop

N. Djupvik

9.9

1.8

33.4

421

338

MCm-2

outcrop

Kakeled

13.5

2.0

47.7

418

353

MCm-3

borehole

Hällekis-1

11.1

2.5

59.8

423

537

MCm-4

borehole

Hällekis-1

11.1

3.9

56.6

420

511

MCm-5

borehole

Hällekis-1

10.8

2.1

21.1

423

195

MCm-6

borehole

Hällekis-1

11.3

3.2

77.3

426

686

MCm-7

borehole

Hällekis-1

22.0

2.7

128.4

423

583

MCm-8

borehole

Hällekis-1

10.4

2.9

52.8

416

508

MCm-9

borehole

Hällekis-1

12.2

4.3

55.0

418

451

MCm-10

borehole

Hällekis-1

3.3

1.4

16.3

426

498

MCm-11

borehole

Hällekis-1

5.9

1.8

31.3

422

529

4.4 Experimental Methods

4.4.1 Uranium measurement

Uranium contents were measured by inductively ICP-MS as described by Romer and

Hahne (2010). About 250 mg of rock powder, which had been dried at 105°C, was weighed

into 15 ml teflon vials (Savillex®) and decomposed using HF, Aqua Regia (3:1 mixture of

37% HCl and 63% HNO3), and perchloric acid (HClO4). In a first step, 4 ml HF and 4 ml

Aqua Regia were added to the samples. The tightly closed vials were placed into a heating

block (160°C) for 14 hours. After cooling, 1 ml HClO4 (70%) was added to destroy the

organic material and fluorides. This solution was evaporated at 180°C to incipient dryness.

The samples were re-dissolved in 1 ml 7N HNO3 and dried. Then, the HClO4 step was

repeated twice. The samples were re-dissolved in 7N HNO3 and kept at 100°C for 14

hours. This solution was brought to a volume of 50 ml for analysis. Data were acquired in

peak jumping mode using a Galileo 4870 in pulse counting mode.

4.4.2 Pyrolytic techniques

Rock-Eval pyrolysis and TOC measurement were performed using Rock-Eval 6 and Leco

SC-632 analysers, respectively, following established procedures. Pyrolysis-GC and

thermovaporisation were performed using the Quantum MSSV-2 Thermal Analysis System

interfaced with an Agilent GC-6890A (Horsfield et al., 2014). (1) For Py-GC, about 10 mg

of coarsely crushed reference shale was filled into a small open glass tube and heated at 300

oC for three minutes to vent the free hydrocarbons. Hydrocarbons generated between 300

to 600 oC were collected and measured. Quantification of individual compounds was

conducted by external standardisation with n-butane. (2) For thermovaporisation-GC,

around 10 mg of sample was weighed into MSSV glass capillary tubes, which were then

sealed by a hydrogen flame. After introduction into the Quantum MSSV-2 Thermal

Analysis System, the external surfaces of the tube were purged for five minutes at 300°C,

during which time volatiles were mobilised within the tube; thereafter the tube was cracked

open by a piston device to transfer the products into a liquid nitrogen-cooled trap. The

composition of these volatiles was quantified as described under Py-GC.

4.4.3 FT-ICR MS

Based on screening data, four representative Alum Shale samples, were Soxhlet-extracted

using a mixture of dichloromethane and methanol (v/v = 99:1). Mass analysis of the

resulting bitumen samples was performed in negative ion Electospray Ionisation (ESI)

mode with a 12 T FT-ICR mass spectrometer equipped with an Apollo II ESI source, both

from Bruker Daltonik GmbH. Nitrogen was used as drying gas at a flow rate of 4.0 L/min

and at a temperature of 220 °C, and acting as nebulizing gas with 1.4 bars. The sample

solutions were infused at a flow rate of 150 μL/h. The capillary voltage was set to 3000 V

and an additional collision-induced dissociation voltage of 70 V in the source was applied

to avoid cluster and adduct formation. Ions were accumulated in the collision cell for 0.05 s

and transferred to the ICR cell within 1 ms. Spectra were recorded in broadband mode

using 4 megaword data sets. For each mass spectrum, 200 scans were accumulated in a

mass range from m/z 147 to 1000.

An external calibration was done using an in-house calibration mixture for ESI negative

mode containing fatty acids and modified polyethylene glycols. Subsequently, each mass

spectrum was internally recalibrated using known homologous series. A quadratic

calibration mode was chosen for all samples. The RMS errors of the calibrations were

between 0.001 and 0.031 ppm. Elemental formulas were assigned to the recalibrated m/z

values with a maximal error of 0.5 ppm.

4.5 Results

4.5.1 Screening data

The uranium contents in the Alum Shale are highly variable, ranging from 11 ppm to 413

ppm (Table 4.1). The Middle Cambrian samples are generally lower in uranium content

compared with Upper Cambrian and Lower Ordovician Alum Shale samples (Table 4.1).

With total organic carbon (TOC) contents greater than 4.0 %, except for sample UCm-9

which contains 3.0 % (Table 4.1), most of the analysed Alum Shale samples can thus be

viewed as “excellent” source rocks from the perspective of organic richness (Hunt, 1961;

Peters and Cassa, 1994). The correlation between uranium and TOC contents is poor

(Table 4.1), as also found in a more comprehensive study with more than 300 Alum Shale

samples presented in Schovsbo (2002).

The Tmax value of sample LO-9 from Ottenby, Southern Sweden (Fig. 4.1) is 441 oC which

suggests that the organic matter is early mature with respect to oil generation whereas all

other samples are immature with Tmax values lower than 430 oC (Table 4.1). In the pseudo-

van Krevelen diagram, seemingly Types I, II, III, and even type IV kerogens occur in the

Alum (Fig. 4.2a). They are, however, in conflict with the marine depositional environment

(Thickpenny, 1984). Samples with low HI and high OI are mostly from a shallow well close

to St. Petersburg (Russia) while the Swedish and Estonian samples are characterized by

Types I and II kerogen. Low OI (<10 mgCO2/gTOC) and moderate HI values (between

300-500 mgHC/gTOC) of the Alum Shale from South-Central Sweden were also reported

by Sanei et al. (2014) and Schulz et al. (2015). Samples from Hällekis-1 and Saint

Petersburg boreholes demonstrate inverse relationships in general between Tmax values and

uranium contents, especially those with uranium contents higher than 100 ppm (Fig. 4.2b).

Fig. 4.2. The correlations between uranium contents and key Rock-Eval parameters. (a) Different kerogen types

can be identified based on the pseudo-van Krevelen diagram (Espitalie et al., 1977). HI and OI are poorly

correlated with uranium contents. (b) Tmax values of samples from two boreholes are inversely proportional to

their uranium contents in general.

4.5.2 Open pyrolysis-gas chromatography and thermovaporisation

The Py-GC and Tvap-GC data unravel detailed compositional information equivalent to

the Rock-Eval derived S2 and S1 components, respectively. Py-GC provides an overview

of the structural characteristics of kerogen (Horsfield, 1989; Van de Meent et al., 1980),

while the Tvap-GC manifests the hydrocarbons that have been generated and retained in

the source rock over its geological history, minus volatile losses that have occurred during

sample acquisition and storage.

16 representative samples were selected for pyrolytic analyses and significantly different

pyrolysates occur among these samples (Table 4.2). The uranium-poor sample (MCm-1,

U=14 ppm) which can be viewed as “typical” marine shale from the perspective of

uranium concentration generates short (C1-C5), middle- (C6-C14) and long-chained (C15+)

hydrocarbons dominated by normal alk-1-enes and alkanes (Fig. 4.3a). With increasing

uranium contents, the pyrolysates are less aliphatic and characterized by increasing content

of aromatic compounds. This can be seen for sample LO-3 (U=244 ppm) where the oil

range pyrolysates are almost exclusively consist of aromatic compounds (Fig. 4.3). For all

samples, the gas percentages and aromaticity of the pyrolysates appear to be exponentially

controlled by the uranium content (Fig. 4.4a and b).

Table 4.2. Py-GC and Tvap data of 16 Alum Shale samples. The ternary end members are normalized as

described by Horsfield (1989) and Eglinton et al. (1990). GOR in Tvap was calculated from gas over resolved oil.

Name

Uranium

content

(ppm)

PyGC

Tvap

Horsfield, 1989

Eglinton et al., 1990

GOR o-Xyl/C9

C1-5 Bulk

(%)

nC6-14

Res.(%)

nC15+

Res. (%)

2,3-

dmThiophene

(%)

nonene;

9:1 (%)

o-xylene

(%)

Lower

Ordovician

LO-1

87.5

12.5

0.0

14.2

18.8

67.0

3.2

2.6

LO-2

190

94.6

5.4

0.0

11.6

10.1

78.3

6.1

5.2

LO-3

244

96.5

3.5

0.0

12.9

6.9

80.2

5.4

9.1

LO-4

274

97.4

2.6

0.0

10.5

8.1

81.4

6.9

12.7

LO-5

110

94.0

6.0

0.0

9.0

17.2

73.8

6.3

6.1

LO-6

136

91.3

8.5

0.2

8.0

18.6

73.4

5.8

3.5

LO-7

107

93.3

6.7

0.0

9.0

17.6

73.4

5.9

3.7

LO-8

119

93.5

6.5

0.0

9.5

14.4

76.2

5.6

5.1

LO-9

78.1

20.4

1.8

6.2

54.3

39.4

0.4

0.5

Upper

Cambrian

UCm-1

155

92.5

7.3

0.2

8.5

13.4

78.1

6.8

3.2

UCm-2

413

98.0

2.0

0.0

9.1

5.1

85.8

6.4

12.5

UCm-3

135

92.0

7.8

0.2

9.3

14.9

75.8

6.6

4.4

UCm-4

186

92.4

7.4

0.2

7.3

13.2

79.5

5.5

7.5

UCm-5

194

94.2

5.8

0.0

10.0

10.5

79.5

4.1

8.9

Middle

Cambrian

MCm-1

77.0

21.0

2.0

9.5

45.3

45.2

0.8

1.8

MCm-2

79.2

18.8

2.0

13.4

39.2

47.4

0.6

1.1

Fig. 4.3. The pyrolysis- and thermovaporisation- GC traces of three Alum Shale samples with very different

uranium contents. n-alkenes and n-alkanes are named by carbon numbers, and major aromatic compounds are

illustrated. (a) The pyrolysates show increasing gas/oil ratio and aromaticity in the products with increasing

uranium contents from the top to bottom. (b) The Tvap products is featured by the absence of alkenes

compared with Py-GC, but they still show the same trend of compositional changes in response to uranium

contents as revealed by pyrolysates.

At least two groups can be identified among the samples when plotting the pyrolysate data

in the classical discrimination ternary diagrams (Fig. 4.5). The organic matter in all Middle

Cambrian samples and in one Lower Ordovician sample (LO-9), lean in uranium and

characterised as Type II kerogen based on Rock-Eval data, have the potential to generate

Paraffinic-Naphthenic-Aromatic oil with low wax contents which is similar to “classical”

marine shales (Fig. 4.5a). On the other hand, the pyrolysates generated from uranium-rich

Alum Shale samples are dominated by gaseous and aromatic compounds and fall in the

Type III field of the pseudo-van Krevelen diagram (Fig. 4.5). Phenol and cresol which are

dominant compounds in terrestrial organic matter pyrolysates (Larter, 1984; Van de Meent

et al., 1980) are nearly absent when pyrolysing Alum Shale (Fig. 4.3a).

Very interestingly, the gas to oil ratio (GOR) and the o/xylene ratio resulting from Tvap

analyses are also proportional to uranium content (Fig. 4.3b and Fig. 4.4c and d). The high

sensitivity of the Tvap derived GOR to weathering and storage conditions might partly

explain some inconsistencies of the correlations. Furthermore, it seems that a GOR

threshold between 5.5 and 6.9 is reached in Tvap experiments and it might define the

maximum gas storage capacity of finely powdered samples.

Fig. 4.4. Correlations between uranium contents with compositional information derived from pyrolysis-GC (a and

b) and Tvap-GC (c and d). The gas percentage in (a) and o-xylene percentage in (b) are two end members of

two classical ternary diagrams as shown in Fig. 5 which are instructive to organic facies. Gas/oil ratio in (c) is

calculated from gas/resolved oil in Tvap which reflects the gas richness as (a) does. Since the 2,3-

dimethylthiophene concentration in Tvap is low and can’t be accurately interpreted, only o-xylene and n-nonane

were used in (d). Nevertheless, both (b) and (d) represent the aromaticity of the products.

Fig. 4.5. Ternaries of the pyrolysates showing interpretations of the organic facies and kerogen structures of the

shales (Eglinton et al., 1990; Horsfield, 1989).

4.5.3 FT-ICR MS

4.5.3.1 General elemental composition

The FT-ICR MS technique offers the ultra-high resolution detection of petroleum

constituents (Hughey et al., 2001; Marshall et al., 1998) and is fundamental to the concept

of “Petroleomics” (Marshall and Rodgers, 2004). FT-ICR MS run in ESI negative mode

can identify up to 30,000 NSO-containing compounds in crude oil (Bae et al., 2010), and

has been widely applied in petroleum science, e.g., oil typing (Hughey et al., 2002),

biodegradation (Kim et al., 2005), maturity (Oldenburg et al., 2014; Poetz et al., 2014),

thermochemical sulphate reduction (Walters et al., 2015), migration fractionation (Liu et al.,

2015; Mahlstedt et al., 2016), and mineral-organic interactions (Yang and Horsfield, 2016).

In this study, total extracts from four representative Alum Shale samples were analysed.

The relative total monoisotopic ion abundance (TMIA) of each NSO class was calculated

by normalizing the peak area to the total. Oxygen-containing compounds are clearly more

enriched in the uranium-rich samples (Fig. 4.6) although interestingly it is these samples

that are exclusively low in OI (Table 4.1). Significant differences can also be found in the

nitrogen class distributions (Fig. 4.6), i.e., the nitrogen-containing group exclusively consists

of the N1 class in the uranium-rich samples (LO-6 and UCm-1), while the N2 classes only

occur in the uranium-poor samples (LO-9 and MCm-2).

Fig. 4.6. Elemental class (inner circle) and compound class (outer circle) distribution pie charts of four

representative Alum Shale samples derived from ESI(-) FT-ICR MS analyses. Uranium-poor samples (LO-9 and

MCm-2) have lower oxygen contents and uranium-rich samples (LO-6 and UCm-1) are featured by the absence

of N2 compounds.

4.5.3.2 Detailed N1 class characterisations

Since the oxygen compounds detectable by FT-ICR MS are highly sensitive to

biodegradation (Kim et al., 2005; Liao et al., 2012) and contamination (Teräväinen et al.,

2007), the N1 class which is one of the dominating classes in each sample (Fig. 4.6) offers

further investigation potential. Organic compounds containing one nitrogen atom

measurable in ESI(-) FT-ICRMS are most commonly pyrrolic and indolic belonging to

carbazoles (Hughey et al., 2002; Pakarinen et al., 2007).

In the vertical direction (Fig. 4.7), the DBE distribution of sample LO-9 is generally higher

especially no DBE < 9 is detected. Much more pronounced differences are reflected by the

carbon number distribution (Fig. 4.7) which denotes the alkylation degree of the core

structures. It can be concluded that the carbon numbers range to higher values in the

uranium-poor samples (Fig. 4.7a and b) compared with samples with higher uranium

contents (Fig. 4.7c and d). In the case of DBE 9, sample UCm-1 (U=155 ppm) contains up

to 27 carbon numbers (Fig. 4.7d) which means that maximum 15 saturated carbon atoms

are attached to the core structure of carbazole (C12H9N). While the alkylation degree on

carbazoles in sample LO-9 (U=33 ppm) is much more pronounced, the carbon number

can be as high as 36 (Fig. 4.7a).

Fig. 4.7. DBE against carbon number diagrams on the N1 class of two uranium-poor (a and b) samples and two

uranium-rich ones (c and d). The size of the circles denotes the relative abundance of each compound and the

dash lines in the right part of each diagram enable comparison of the alkylation.

4.6 Discussion

4.6.1 The uranium enrichment

Organic-lean grey shales are estimated to have an average uranium content of four ppm

(Alloway, 2013; Swanson, 1960). In organic-rich marine black shales the uranium contents

are much higher averaging 20 ppm (Swanson and Swanson, 1961). For example, the

Mississippian Barnett Shale has a maximum uranium content of 14 ppm (Abouelresh and

Slatt, 2012; Tian, 2010), while the famous Silurian “hot shale” in the Middle East with high

TOC contents (6-10%) and high gamma ray logging responses contains uranium contents

of up to 25 ppm (Loydell et al., 2009; Lüning et al., 2005). Obviously, uranium

concentrations in the Alum Shale, with the exceptions of Middle Cambrian samples and

sample LO-9, are much higher than typical marine shale and most of other black organic

rich shales (Table 4.1).

Uranium can be immobilised in sediment under a reducing depositional environment

(Goldschmidt, 1937), for it is soluble in oxic seawater and the solubility significantly

decreases in reducing environments (Durand, 2003). Therefore, uranium precipitation from

the aqueous phase requires both uranium-rich sea water and reducing conditions which

finally lead to an enrichment in sediments (Breger and Brown, 1962; Disnar and Sureau,

1990; Vine and Tourtelot, 1970).

Swanson (1960) found that humic organic matter in North American shales contains far

more uranium than the sapropelic type. In contrast, Breger and Brown (1962) argued that

uranium is correlated with the TOC content irrespective of the organic matter type. The

poor correlation between uranium and TOC contents in the Alum Shale (Table 4.1) and in

previous data sets (Schovsbo (2002) emphasises an uranium enrichment mechanism

different from classical pathways. So far, there are three representative hypotheses available:

(1) Fisher and Bostrom (1969) and Oliver et al. (1999) found that hydrothermal fluids

change the temperature and pressure of sea water causing uranium precipitation in fine-

grained sediments. The unusual metal composition of the Alum Shale in the Oslo Graben

may be a result of such interactions (Berry et al., 1986). This local volcanic influence may

be taken as an explanation as to why shales contemporaneously deposited with the Alum

Shale in Wales, North America, and South America lack such high uranium contents.

(2) The comparison with the Devonian Chattanooga Shale led Leventhal (1991) to

conclude that the Alum shale lacks humic material. The high uranium contents in Alum

Shale were attributed to a slow sedimentation rate and the long-term persistence of euxinic

bottom water which may have preserved the accumulating organic matter and its associated

metals.

(3) In contrast to a reducing depositional environment as a controlling factor, Leckie et

al. (1990) found that the uranium-rich Shaftsbury Formation (Cretaceous) in Canada was

deposited in relatively shallow water based on palynological, micropalaeontological and

geochemical results. Schovsbo (2002) reported that the uranium contents of the Alum

Shale are inversely correlated with the layer thickness which implied that uranium was

preferentially enriched in the inner-shelf area rather than in the outer-shelf facies.

Accordingly, an intensified bottom water circulation may have resulted in an enhanced

supply of uranium at the sediment/water interface where the uranium extraction from the

sea water took place.

In summary, during deposition of the Alum Shale, uranium was not captured by humic

organic matter which developed at post-Silurian times (Kenrick and Crane, 1997). Instead,

the physicochemical conditions of the depositional environment were the controlling

factors.

4.6.2 Hydrocarbon precursors and products

4.6.2.1 Kerogen structure

The atomic H/C and O/C ratios which define kerogen type and thermal maturity (Durand

and Espitalié, 1973) are influenced by uranium irradiation, with H/C decreasing and O/C

increasing when the uranium content of the shale is high (Pierce et al., 1958). This general

trend is also manifested in equivalent Rock-Eval parameters (Landais, 1996; Leventhal et al.,

1986). However, uranium content is probably not the controlling factor of the HI and OI

values in the current sample set (Fig. 4.2a) at least the OI is not necessarily high when the

sample is rich in uranium (Table 4.1). Seemingly Types III and IV kerogen of the St.

Petersburg samples could be the result of a unique depositional environment (Dronov and

Holmer, 1999) or, more likely, due to biodegradation of the organic matter (Tolmacheva et

al., 2001). It could be possible that the hydrocarbon potential was reduced due to the

liberation of hydrogen by uranium ionising radiation (Colombo et al., 1964; Dole, 1958).

However, since the original HI values of the sample are variable, a relationship between

uranium content and the current HI values is not detectable.

Middle Cambrian samples and sample LO-9 with relatively low uranium contents are

predicted to produce Paraffinic-Naphthenic-Aromatic hydrocarbons which are features of

classical marine shales (Horsfield, 1989). In according, it is atypical of marine shales in

general that the pyrolysates of the rest samples are extremely rich in gas and aromatic

compounds (Fig. 4.5). The abundance of short chain aliphatics and alkylbenzenes can be

indicative of terrestrial originated organic matter (Eglinton et al., 1990; Horsfield, 1989)

which, however, does not apply here. Another possible explanation could be the mineral

matrix catalytic effect during pyrolysis (Espitalie et al., 1980; Horsfield and Douglas, 1980).

For example, Yang et al. (2016) showed that the whole rock pyrolysate of the argillaceous

marine Bowland Shale (Mississippian, UK) generates a much gassier and more aromatic

pyrolysate than its kerogen concentrate does. However, the pyrolysates of a uranium-rich

Alum Shale samples before and after demineralization are identical, thereby ruling out this

explanation.

Using the analogue that Ordovician seas were globally and uniquely populated by the alga

Gloeocapsamorpha Prisca (Fowler and Douglas, 1984), Horsfield et al. (1992a) suggested

that the Cambrian might have had its own unique biota whose residues on pyrolysis yielded

the unusual pyrolysates. Bharati et al. (1995) and Sanei et al. (2014) expanded the precursor

hypothesis, as algae, e.g., Chlorella marina, can yield aromatic-rich hydrocarbons (Derenne et

al., 1996). Furthermore, pyrolysis of trilobite chitin results in aromatic hydrocarbons

(Arthur Stankiewicz et al., 1996) and could also serve as an explanation. Concerning the

current dataset, Lower Ordovician samples with similar palaeo-biota yield very different

pyrolysates (Fig. 4.5) depending on the uranium contents (Fig. 4.4a and b). The robust

correlations between uranium concentrations and key pyrolysate parameters substantiate

the hypothesis that irradiation is a major control on the Alum Shale products. Furthermore,

with decreasing uranium contents, the pyrolysates are similar (Fig. 4.4a and b) pointing to

the fact that the original kerogen structures of all Alum Shale samples are essentially

uniform. It is the increasing irradiation dosage (dependent on the uranium content) that

changes the basically uniform structures to be more gas-prone and aromatic.

The vitrinite reflectance of humic coal were reported to be enhanced by crosslinking of

polymers caused by uranium irradiation irrespectively of the geological heating (Breger,

1974). Similarly, (Forbes et al., 1988) and (Landais, 1996) suggested the Tmax of vitrinite in

Akouta uranium deposit (Niger) increases when the uranium contents are high. However,

the reverse correlation between Tmax and uranium contents in Alum Shale is in accordance

with previous findings by Dahl et al. (1988b). Since the correlation between Tmax and OI is

poor (Table 4.1), the O-C bonds are generally stronger than C-C bonds don’t serve as

explanations for this. Based on molecular modelling, Claxton et al. (1993) found that the

presence of aromatic rings in kerogen results in weakening the bond energies of carbon

chains except for the α bond (the first carbon bond attached to the ring structure). The

aromatic products-rich features of samples with high uranium contents (Table 4.2) imply

that the breaking of carbon chains attached to aromatic structures occurred extensively

during pyrolysis. Since such kind of bonds attached to the aromatic structures (except α

bond) are generally weaker than those of the same position in a ring structure-free system,

this could explain why uranium-rich samples present relatively low Tmax values. Another

possible explanation could be related to the generation and polymerisation of hydrocarbons

by uranium irradiation (Court et al., 2006). These complex hydrocarbons would not be

activated under the S1 temperature (up to 300 oC) and their releases during S2 temperatures

(300-650 oC) would decrease the peak temperature of S2 since the vaporisation of them are

anticipated to be easier than the breaking of kerogens.

4.6.2.2 Free hydrocarbons

The free hydrocarbons stored in the immature (low Tmax) Alum Shale samples and detected

by Tvap may have been formed by incipient thermal cleavage reactions or by irradiation

induced maturation. Jaraula et al. (2015) reported that the odd/even carbon preference of

n-alkanes in an immature deposit (Ro=0.26%) decreases with increasing uranium contents,

and thus proposed an alternative radiolytic cracking pathway besides the thermal and

microbial mechanisms of petroleum generation. It needs to be pointed out that the free

hydrocarbons are quantitatively limited, especially from the uranium-rich samples, as

revealed by small S1 values (Table 4.1). Nevertheless, the fit between pyrolysis results and

Tvap products (Fig. 4.5) manifests that the uranium irradiation effect on compositional

variation does not only work in laboratory experiments but also is effective in geological

environments.

4.6.3 Heterocompounds

The gross elemental composition and detailed DBE distributions in each class of the NSO

compounds of the bitumen allow further insights into changes of the organic matter

structure in response to uranium irradiation.

4.6.3.1 Oxygen-containing moieties

The oxygen compounds in petroleum or source rock extracts are influenced by the

depositional environment, biodegradation, and maturity. Hughey et al. (2002) found that

crude oils generated from lacustrine source rocks are richer in acids (O2 compounds) than

crude oils generated from marine source rocks. Biodegradation of oils normally raises the

concentrations of oxygen compounds, especially the O2 class (Kim et al., 2005).

Furthermore, extracts from Posidonia Shale (Lower Jurassic) gets more depleted in oxygen

compounds during thermal maturation (Poetz et al., 2014).

Actually, the aforementioned factors play only a very minor role, if any: 1) The Alum Shale

was deposited in a fully marine environment that predates the evolution of terrestrial plants.

This implies that the source type variation played a very minor role for oxygen compounds

variations. 2) Sample LO-6 and UCm-1 are borehole samples (Table 4.1) less prone to be

biodegraded, and the outcrop samples (LO-9 and MCm-2) that a most likely to degraded

show low oxygen contents (Fig. 4.6) which argue against biodegradation; (3) Sample LO-9

which is more mature among the presented four samples in Fig. 4.6 contains fewer oxygen

compounds, which also contradict a possible maturation effect. Instead, the O2 content in

the Alum Shale is related to the uranium content (Fig. 4.6) and we argue that the uranium

irradiation may have induced the relative oxygen enrichment in the Alum Shale bitumen as

indicated by data from elemental analyses (Pierce et al., 1958). Purportedly the radiolytic

cleavage of water yields highly reactive OH- radicals which quickly react with the in-situ

organic matter during diagenesis (Court et al., 2006; Jaraula et al., 2015) leading to

enhanced oxygen contents in the kerogen structure . However, these high oxygen moieties

do not appear to degrade into low molecular weight volatiles upon pyrolysis as shown by

the low OI of many uranium-rich samples (Table 4.1 and Fig. 4.2a). Data gained by nuclear

magnetic resonance spectroscopical analyses of the Alum Shale revealed that the oxygen-

bearing functional groups are still intact within the kerogen macro-molecule through

catagenesis stages (Bharati et al., 1995), i.e., some highly mature Alum Shale kerogens are

still rich in oxygen.

N2 class is typically present in source rocks, but is rare in oil (Bae et al., 2010), because it is

preferentially retained in the source rock during expulsion as heavier compounds compared

with the N1 class (Mahlstedt et al., 2016). However, no hydrocarbon expulsion is

anticipated from these immature Alum Shale samples. The absence of the N2 compounds

in uranium-rich samples (LO-6 and MCm-1) may thus be also induced by uranium

irradiation.

4.6.3.2 Pyrrolic nitrogen-containing moieties

Generally, the average DBE in the N1 class shifts toward higher values with increasing

maturity of the shale (Poetz et al., 2014) or crude oil (Oldenburg et al., 2014) due to

annulation and aromatisation. The DBE distribution in sample LO-9, which has the

highest Tmax value among the four samples, could be enhanced by thermal maturation. In

contrast, the other three samples have comparable maturities (Tmax values between 417 oC

to 419 oC) and the influence of maturity can be excluded. Although the uranium-rich

samples tend to generate more aromatic products during pyrolysis, the bitumen aromaticity

is not correlated with the uranium content (Fig. 4.7c and d) and thus no induced effect here

is anticipated to have happened.

Atomic pile radiation on petroleum revealed that paraffins would turn into an insoluble gel

after a certain dose of radiation is reached (Charlesby, 1954). In the Alum Shale, the

bitumen extractability and aliphatic biomarker concentrations are inversely correlated with

the uranium content (Dahl et al., 1988a, b; Lewan and Buchardt, 1989) and were attributed

to cross-linking and aromatisation by uranium irradiation (Hoering and Navale, 1987).

Similar aromatisation mechanisms of the kerogen structures in response to irradiation were

proposed by Forbes et al. (1988) and Kribek et al. (1999). However, the pyrrolic nitrogen

compounds in the Alum Shale extracts are not obviously aromatised in the uranium rich

samples (Fig. 4.7c and d). This could imply that cross-linking, rather than aromatisation, is

the primary reaction path responsible for the low bitumen extractability in uranium-rich

Alum Shale samples.

Mahlstedt et al. (2016) reported that alkylation of the N1 class decreases with increasing

maturity in solvent extracts of the Lower Jurassic Posidonia Shale during thermal

maturation. This maturity control does not serve as an explanation for the strong alkylation

reduction of the uranium-rich Alum Shale samples (Fig. 4.7c and d) since both are

immature. We deduce that it is irradiation that causes side-chain cracking of alkyl chains

attached to the aromatic core structures, leaving lower levels of alkylation behind as a result.

Similar damaging effects have been observed on Alum Shale biomarker distributions (Dahl

et al., 1988a; Lewan and Buchardt, 1989), i.e., high molecular triaromatic steroids (C26-C28)

are always absent in uranium-rich samples, this being atypical for extracts of immature

marine shales in general (McKenzie et al., 1983). The low alkylation degree of the uranium-

rich samples does not only depict the kerogen structure but may also explain why such kind

of shale tends to generate gaseous products instead of long-chained oil (Fig. 4.5a).

4.6.4 Kerogen structure reconstruction

4.6.4.1 Why necessary?

The Alum Shale may hold a huge potential unconventional gas potential based on its wide

occurrence, significant thickness and high TOC contents (EIA, 2015). Furthermore, the

Alum Shale was suggested to have a high storage capacity, based on methane adsorption

measurements in the laboratory of as much as 3.5 m3/t (Gasparik et al., 2014). However,

shale gas exploration activities in southern Sweden and northern Denmark were not

successful, ostensibly due to very low gas saturation (Pool et al., 2012) and possibly gas

leakage (Schovsbo et al., 2014). As far as liquid hydrocarbon potential is concerned in both

conventional and unconventional “plays”, the shale gas potential of the Alum Shale is said

to be high, and its oil potential is anticipated to be very limited based on pyrolysis

experiments (Kotarba et al., 2014a; Sanei et al., 2014). Importantly, however, petrological

(Schleicher et al., 1998), carbon isotope (Więcław et al., 2010b) and biomarker (Yang et al.,

2017) investigations indicate that the crude oil in Middle Cambrian sandstone reservoirs in

the Baltic Basin was sourced by the Alum Shale. Thus, the composition of organic matter

in the uranium-rich Alum Shale has to be evaluated carefully, with due consideration of the

source rock in its present state versus that of the same shale prior to extensive radiation

damage. Specifically, the kerogen in the investigated Alum Shale samples has undergone

irradiation for 478-500 Myr, and must be compositionally and structurally different from

that which underwent major petroleum generation in the basin centre beginning in the

Early Devonian. The kerogen entering the oil window would have been less gas-prone and

aromatic than current predictions suggest.

4.6.4.2 Radiation dosage

The most common isotopes of natural uranium are 238U (99.27%) and 235U (0.72%)

(Osmond and Cowart, 1976). Most of the radiation resulting from 238U decay in natural

systems is being emitted in the form of α-radiation followed by less intensive γ-radiation

(Jaraula et al., 2015). With a half-life of 4.5 billion years, the 238U decays exponentially (Fig.

4.8) and is independent of either temperature or pressure (Goodwin et al., 2009). However,

the time span since the Cambrian is very short compared with the half-life making the

curvature of the exponential curve extremely small (Fig. 4.8). Therefore, the decay can be

roughly viewed as linearly correlated with time, i.e., the activity of decay in Alum Shale is

considered constant.

Fig. 4.8. The exponential decay curve of 238U. A zoom in on the geological time scale manifest that the decay

can be roughly viewed as linearly correlated with time.

Whyte (1973) proposed that the radiation dosage (D) from a point source to a detector

over times (t) can be calculated through:

where E denotes the energy per decay (MeV) which is constant for 238U; C describes the

activity (Bq) of decay and can also be viewed as a fixed rate as described above; S

represents the area of the detector and 4πr2 is the area of a sphere with radius r, thus the

S/4𝜋𝑟2 shows the probability that the radiation will reach the detector; μ is the mass

energy-absorption coefficient.

Although E, C, and μ in the formula can be constant values, it is impossible to accurately

measure the distance between a uranium atom and organic matter, especially the

interactions must be tortuous. Nevertheless, since the uranium in Alum Shale is primarily

accumulated in organic matter and marine phosphates (Lecomte et al., 2017) and the spatial

relationship of uranium and organic matter is fixed through geological time, it can be

concluded that the radiation dosage from one uranium atom is proportional to time.

Furthermore, the gross uranium irradiation on kerogen structures within a shale sample is

linearly correlated with both uranium content and time since shale deposition.

4.6.4.3 Kerogen reconstruction

The exponential relationships between uranium and pyrolysate parameters (Fig. 4.4a and b)

imply that the response of kerogen structures to irradiation is not linear. Labile kerogen

structures can be easily changed in the early stages; thereafter the altered structures would

become less sensitive to radiation and will reach equilibrium in the end. The response

curves in Fig. 4.4a and b describe a scenario that samples with different uranium contents

have experienced a similar time of irradiation. These pathways should also work when the

uranium content is fixed and only radiation times vary since the uranium content and

radiation time are linearly complementary and the loss of 238U is negligible.

In the case of sample UCm-2 (U=413 ppm), the current pyrolysate is featured by a gas

content of 98% and an o-xylene content of 86% (Table 4.2). With decreasing radiation

dosage, i.e., less radiation time, the products from this shale are less rich in gas and

aromatic compounds. One-dimensional burial reconstructions presented by Kosakowski et

al. (2010) indicate that the Alum Shale in the Baltic Basin centre started to generate and

expel petroleum from the Early Devonian, e.g., Alum Shale in well A23-1/88 (Fig. 4.1) was

in in the oil window between 420-340 Myr ago (Fig. 4.9). Accordingly, the compositional

information of hydrocarbons generated during that time can be back-calculated assuming

the kerogen structures in sample LO-9 (U=33 ppm) represent the primitive ones of sample

UCm-2. The gas and o-xylene contents are calculated as 88-91% and 63-72 % (Fig. 4.9),

respectively, which are still high for typical marine shales (Fig. 4.5). When the petroleum

generation started in Devonian times, the Alum Shale had experienced about 1/5 time of

the total irradiation compared with nowadays sample, but the kerogen alteration has

fulfilled approximately half due to the exponential response of kerogen structures to

irradiation dosage. This method can also be applied to samples with lower uranium

contents, and the products dated back to Palaeozoic times must be more oil-rich and

aliphatic than those from sample UCm-2.

In summary, the investigated Alum Shale samples had initially a lower gas potential and a

higher oil potential than previously assumed based on analysis of present day state samples.

The reconstructions of the kerogen structure back to the time when oil window maturity

prevailed are crucial in predicting the petroleum characteristics and thereby in reducing the

hydrocarbon exploration risks.

Fig. 4.9. The back-calculation of products that could be generated from sample UCm-2. Calculated vitrinite

reflectance curves is based on basin modelling work on well A23-1/88 (location in Fig. 1) from Kosakowski et al.

(2010). Oil window was estimated with Ro between 0.5-1.3 %. The gas and o-xylene percentages curves are

based on the correlation curves in Fig. 4a and b, respectively. Pyrolysates from sample LO-9 were set as the left

end members of these two curves.

4.6.4.4 Further implication

Ziegs et al. (2017) reported that the light hydrocarbon retaining ability of shale is

proportional to the aromaticity of its labile kerogen, therefore, the gas adsorption capacity

of the Alum Shale over geological time could also have been overestimated. The expected

highest risks of gas leaks in Alum Shale are assumed to have occurred during late

Caledonian uplift prior to Zechstein deposition and during Neogene uplift (Pool et al.,

2012; Schovsbo et al., 2014). During the late Caledonian uplift when the aromaticity of

pyrolysates were half irradiated (Fig. 4.9), the maximum gas retention capacity by the Alum

Shale kerogen should be much smaller than the values suggested by adsorption

experiments (Gasparik et al., 2014). Therefore, the experimental data based on nowadays

Alum Shale samples should only be used as the upper bound of gas adsorption capacity in

resource estimates.

The uranium irradiation effects in changing triaromatic steroids biomarker distribution

(Dahl et al., 1988a; Lewan and Buchardt, 1989) can be applied in oil-source rock

correlation. Yang et al. (2017) proposed reservoir oil with low C26–C28 triaromatic steroids

to be sourced from uranium-rich Alum Shale. As revealed by pyrolysis result, the

compositions of Alum Shale products are related to uranium content (Fig 4.4 a and b). In

this way, the kerogen structures of UCm-4 (U=231 ppm) are partly altered by uranium

irradiation. However, the C26–C28 triaromatic steroids are already completely removed

(Yang et al., 2017). Similarly, Dahl et al. (1988a) documented that an Alum Shale sample

with 163 ppm uranium (sample B26 in Fig. 4.9) was totally depleted in such high-molecular

triaromatic steroids. These together imply that the aromatic biomarkers are much more

sensitive to uranium irradiation than kerogen structures. Therefore, when the Alum Shale

entered oil window during Devonian times, most of the high-molecular triaromatic

biomarkers might be removed, in comparison to half of the kerogen structure was changed

(Fig. 4.9). In brief, the application of triaromatic steroid biomarker distribution induced by

irradiation is an important correlation tool in researching the petroleum system in the Baltic

Basin, especially aliphatic biomarkers are typically absent when high-uranium content Alum

Shale is involved.

In addition, the organic matter changes in response to irradiation revealed here can also be

instructive to extra-terrestrial research in which strong radiation environments are

ubiquitous (Allen et al., 1998) and the evaluation of the long-term impact on environment

resulted from uranium waste disposal in shale (Gautschi, 2001).

4.7 Conclusion

1. Neither the HI nor the OI of the immature Alum Shale is correlated with the

uranium contents. The HI could be decreased by uranium irradiation, but the

diverse original HI values prevent the correlation. Oxygen compounds are enriched

by the existence of uranium, but they are mainly intact in the kerogen macro-

molecules.

2. Most of the Alum Shale samples generate gas-rich and aromatic products through

pyrolysis experiments which are atypical for marine shale. Gas and o-xylene

percentages in pyrolysates and natural products are both proportional to uranium

contents pointing to an irradiation control on the kerogen structures.

3. The irradiation does not significantly increase the aromatisation of macro-

molecules, instead, the alkylation to aromatic structures was shortened by the

irradiation bombardment and this is responsible for the high production of gas.

4. Both uranium content and radiation time are linearly correlated with uranium

radiation dosage. However, the response of kerogen structure to radiation is

exponential; for the labial structures are altered in the early stage of irradiation.

5. About half of the irradiation induced kerogen changes occurred when the Alum

Shale was in oil-window maturity. The back-calculation of kerogen structure can

efficiently avoid the over-estimations of gas generation and gas retaining ability in

the Alum Shale.

4.8 Acknowledgment

We thank Gripen Oil & Gas AB and the Geological Survey of Sweden for providing the

samples. Dr. Nicolaj Mahlstedt is acknowledged for fruitful discussion. Kind thanks are

extended to Ferdinand Perssen and Cornelia Karger for their technical support. This study

is financially supported by the Chinese Scholarship Council.

This chapter has been published as: Yang, S., H.-M. Schulz, N. H. Schovsbo, and J. A. Bojesen-

Koefoed, 2017, Oil-source rock correlation of the Lower Palaeozoic petroleum system in the Baltic

Basin (northern Europe), (in press; preliminary version published online Ahead of Print May 22,

2017): AAPG Bulletin (postprint), doi: 10.1306/02071716194.

5. URANIUM IRRADIATION ON

BIOMARKERS

5.1 Abstract

The correlation of Lower Palaeozoic marine source rocks with reservoired oils by

biomarkers is complex due to the uniform Early Phanerozoic biomass (bacteria and algae)

and the lack of land plant and animal input. Accordingly, the main source rocks for the

most prolific oil province in the Baltic Basin are still a matter of debate.

Ten source rocks and 15 oil samples from five north European countries bordering the

Baltic Sea Basin were analysed by gas chromatography (GC) with flame ionization detector,

GC-MS (mass spectrometry), and GC-MS/MS to detect acyclic isoprenoids, and aliphatic,

aromatic, and NSO (nitrogen, sulphur and oxygen) biomarkers. Chemometric tools were

applied to screen for meaningful source- and age-related biomarkers and to highlight

genetics. Extended tricyclic terpane ratios, C24 tetracyclic terpane/C26 tricyclic terpane ratios,

and relative C29 sterane concentrations are considered the most promising biomarkers in

differentiating Llandovery shales from Cambrian to Ordovician Alum Shale and for

correlation with expelled oil. The uranium irradiation related C26-C28 triaromatic steroid

concentrations provides possible distinguishing criteria for the source potential of the

different Alum Shale units. Enhanced oil maturation by volcanic intrusion is highlighted by

sterane biomarkers and polycyclic aromatic hydrocarbons.

The Alum Shale is here considered the main source rock for oil accumulations in Lower

Palaeozoic reservoirs of the Baltic basin. Oil seepage occurring in Ordovician limestone

was mainly generated by the Middle Cambrian Alum Shale, and Middle Cambrian

sandstone reservoirs were mainly sourced by Upper Cambrian and Lower Ordovician

Alum Shale with higher maturity. Considerations about the assessment of migration

distance are based on carbazole concentrations and C29 sterane isomerization. Advanced

studies to unravel Lower Palaeozoic oil-source rock correlations are based on

meaningful biomarkers; offer approaches to significantly reduce the exploration risk in this

area, and could be applied to similar Early Palaeozoic petroleum systems in other basins.

5.2 Introduction

Lower Palaeozoic successions in the Baltic Basin (northern Europe) host both

conventional oil and gas accumulations, and unconventional hydrocarbons in the source

rocks itself. Petroleum accumulations occur in Middle Cambrian sandstones onshore and

offshore Poland (Karnkowski et al., 2010; Więcław et al., 2010b), in the Kaliningrad Oblast

(Brangulis et al., 1993), and in Lithuania (Zdanavičiūtė, 2012). Scattered Upper Ordovician

limestone oil reservoirs were also discovered in Latvia (Kanev et al., 1994) and around the

Swedish Gotland island (Sivhed, 2004). The two most important source rocks of Lower

Palaeozoic age in the Baltic Basin and surrounding areas are the Alum Shale (Middle

Cambrian to Lower Ordovician) and the Llandovery (Lower Silurian) black shale which are

both promising targets for shale gas exploration (Schovsbo et al., 2011; Schulz et al., 2010;

Zdanavičiūtė and Lazauskienė, 2009).

Although oil is produced from Middle Cambrian sandstone and Upper Ordovician

limestone in the Baltoscandian countries for decades, there is a lack of convincing evidence

about the main source rock(s) since these are very alike with respect to most conventional

geochemical and organic petrographic tools. All potential source rocks are marine shales

containing type II kerogen (algae and bacteria originated) (Kanev et al., 1994; Nielsen and

Schovsbo, 2006; Więcław et al., 2010a) and the analyses of established proxies like the

pristane/phytane (Pr/Ph) ratio or routine biomarkers delivered less sufficient evidence in

clearly unravelling the differences between the different marine shaly source rocks. The

main complicating issue is the uniformity of shale due to the lack of higher terrigenous land

plant input during the Lower Palaeozoic time and a relatively monotonous depositional

environment. Although using data about light hydrocarbon distributions, selected

biomarkers, and carbon isotope data then Kanev et al. (1994) and Więcław et al. (2010b)

were unable to separate the shales and concluded that all shales of Middle Cambrian to

Lower Silurian age are potential source rock candidates for Lower Palaeozoic petroleums

reservoired in the Baltica basin. In contrast, Zdanavikiute and Bojesen-Koefoed (1997) and

Pedersen et al. (2007) suggested that Llandovery shales are the main contributors for oil in

Middle Cambrian sandstones according to Pr/Ph ratios and sterane data, while Schleicher

et al. (1998) and Kotarba and Lewan (2013) argued that the petroleum was mainly

generated by the Alum Shale based on biomarker and carbon isotope data. Dahl et al.

(1989) assumed that the oil in limestone reservoirs on Gotland was sourced from the Alum

Shale which presumably was heated locally by volcanic activity, and referred to

characteristic sterane, tricyclic terpane and homohopane distributions. However, the lack of

Alum Shale occurrence in the near subsurface is a reasonable counterargument (Sivhed,

2004). In summary, great uncertainties exist despite previous investigations.

In this contribution we present analytical data of samples taken on a broader regional scale

than previous studies in the Baltic Basin. The samples are from five Baltoscandian

countries and comprise Cambrian to Silurian source rocks and oils from Middle Cambrian

and Ordovician reservoirs. Conceptually, we correlated results about maturity, source, and

age-specific information. Chemometric tools were applied to classify the oil samples into

clusters. Specific biomarker combinations considered effectively in differentiating source

rock extracts and oils from different ages in this area were introduced first. In addition, oils

matured by regional deep burial versus locally from volcanic heating are differentiated. This

contribution offers a practical solution for further exploration activities in the Baltic Basin

by providing the best tools to differentiate source rocks and oils of Lower Palaeozoic age

in general that are applicable not only in Baltic Basin but also elsewhere.

5.3 Regional Petroleum Geology

5.3.1 Regional Geodynamics and Basin Evolution

The Baltic Basin covers part of the Baltic Sea, Kaliningrad Oblast, Northern Poland and

the western parts of the Baltic States (Fig. 5.1). It overlies the western margin of the East

European Craton with its Precambrian crystalline basement and comprises of sediments

deposited from the Early Cambrian to present day. The basin fill is thin in the north-

eastern part (less than 100 m [300 ft] thickness in Estonia) and increasingly thickens south-

westward towards the Teisseyre-Tornquist Zone (over 4000 m [13000 ft] thickness in

northern Poland) which forms the southern boundary of the basin (Ulmishek, 1990).

Since its formation, the Baltic Basin underwent four main geodynamic evolution stages

each with different burial dynamics until the Late Palaeozoic. These different episodes were

the main controlling factors for the tectonic and sedimentary features of the basin (Šliaupa

and Hoth, 2011).

(1). During the passive continental margin stage (Vendian-Late Ordovician) the basin

started to develop as a passive margin basin from latest Vendian to Early Cambrian times

in response to the break-up of the Rodinia supercontinent (Kosakowski et al., 2010;

Poprawa et al., 1999). This initial basin stage gave rise to coarse-grained siliciclastics of

Ediacaran age. With progressing continental separation, marine transgressions flooded the

basin, and marine shale, sandstone and carbonate sedimentation occurred from Cambrian

to Ordovician periods (Nielsen and Schovsbo, 2006).

(2). The foreland basin stage (Late Ordovician-Early Devonian) developed due to the

docking of the Avalonian plate to the western margin of the Baltica plate, mainly during the

Late Ordovician-Silurian times (Poprawa et al., 1999; Vejbæk et al., 1994). The subsidence

increased during the Silurian when up to 4500 m [14700 ft] of siliciclastic and calcareous

sediments were deposited.

(3). A more continuous but slower subsidence took place during the Devonian and

characterises the intra-craton basin stage (Šliaupa and Hoth, 2011). Lagoonal and deltaic

deposits covered the area during that time (Ūsaityt, 2000).

(4). A thermal doming and sag basin stage developed during the Carboniferous-Permian

transition. The basin subsidence ceased in the Mississippian time, thereafter, significant

uplift and erosion affected the margin of the basin (Sopher et al., 2016). In response to the

thinning of the lithosphere, diabase sills and dykes intruded the Baltic Basin (Motuza et al.,

1994) and Scandia (Obst, 2000; Priem et al., 1968).

Fig. 5.1. Simplified map of the Baltic Basin and its possible oil kitchen. The thermal maturity of the Alum Shale

measured on vitrinite-like macerals (Buchardt et al., 1997).

5.3.2 Source Rocks

Several marine shales with a considerable petroleum generation potential are of Early

Palaeozoic age (Fig. 5.2), while carbonate deposits contain total organic carbon (TOC)

contents less than 0.2 wt.% which denote them as non-source rocks (Šliaupa and Hoth,

2011). The Middle Cambrian-Lower Ordovician Alum Shale and the Llandovery shale are

considered as the two most important and potential source rocks because of their wide

occurrence, considerable thicknesses and high TOC contents (Kanev et al., 1994).

Named after the hydrated potassium and aluminium-bearing sulphate [KAl(SO4)2·12H2O],

the Alum Shale consists of fine-grained, blackish mudstone and shale of Middle Cambrian,

Upper Cambrian (Furongian) and Lower Ordovician (Tremadocian) age (Nielsen and

Schovsbo, 2006) (Fig. 5.2). The thickness of the Alum Shale is regionally variable. To the

west of the Baltic Basin, an approx. 180 m (590 ft) thick Alum Shale was found in the

Danish offshore (Nielsen and Schovsbo, 2006). Within the basin, the Alum Shale reaches

its maximum thickness of ca. 100 m [300 ft] in the Teisseyre-Tornquist Zone, and wedges

out eastward and northward due to erosion (Buchardt et al., 1997). The Alum Shale is not

present to the east and north of Kaliningrad-South Gotland line (Fig. 5.1) except for a

maximum seven meter (23 ft) thick Tremadocian Alum Shale in Northern Estonia (Loog et

al., 2001). Covering a time span of 23 million years (500-477 Ma), the three Alum Shale

sub-stages (Middle Cambrian, Upper Cambrian and Lower Ordovician) differ in their

organic and inorganic geochemical composition (Thickpenny, 1984). High TOC contents,

typically up to 10 wt.% and high sulphur contents dominate in the Lower Ordovician and

Middle Cambrian Alum Shale whereas the Upper Cambrian Alum Shale is featured by

TOC contents of up to 22 wt.% and high silica contents (Kosakowski et al., 2016;

Schovsbo et al., 2015).

Formed in a deep water environment, the Llandovery Shale is a dark grey to black shale or

marlstone. This shale is rich in TOC (up to 16 %wt.) and its hydrocarbon generation

potential is known in the southern and eastern part of the basin (Zdanavikiute and

Bojesen-Koefoed, 1997). A typical thickness of the layer in the eastern part of the basin is

5-25 m (16-82 ft) (Kanev et al., 1994), but increases to 100 m (300 ft) in offshore Poland

(Modliński and Podhalańska, 2010).

Besides these two layers, the locally distributed Caradocian age Upper Ordovician shales

can also be considered as possible source rocks (Kanev et al., 1994). In Sweden and

Lithuania, the Upper Ordovcian shales are named as Fjäcka Shale and the Mossen Shale

with TOC contents of up to 6 wt.%. However, these two shales have maximum thickness

of 10 m (32 ft) (Högström and Ebbestad, 2004; Zdanavičiūtė and Lazauskienė, 2004),

typically less than five m (16 ft) thick (Šliaupa and Hoth, 2011), and occur very locally. The

equivalent shale in Poland is called Sasino Shale (Modliński and Szymański, 1997; Stouge

and Nielsen, 2003). It has typical TOC between 1-2.5 wt. % with variable hydrogen index

values, ranging from 11 to 359 mg hydrocarbon/gTOC, and is considered to be deposited

in a sub-oxic environment (Więcław et al., 2010a). The limited distribution and

hydrocarbon generation ability of the Upper Ordovician shales make them possibly

regional source rocks rather than main contributors of the oil accumulations found all

around the basin.

Fig. 5.2. Simplified stratigraphy of the Lower Palaeozoic petroleum system in the Polish part of the Baltic Basin.

5.3.3 Reservoirs

The Middle Cambrian sandstones and the Upper Ordovician limestone reefs are the

reservoir rocks for most of the petroleums found in the Lower Palaeozoic strata in the

Baltic Basin (Fig. 5.2). Petroleums trapped in sandstone occur in Poland, Russia

(Kaliningrad) and Lithuania while oil reservoired in limestone reefs occur on the Swedish

island of Gotland and in Latvia (offshore and onshore) (Fig. 5.1).

The trap style in the sandstone reservoirs are mosly asymmetric brachy-anticlinal domes as

a result of the Caledonian orogeny. The effective porosity and permeability in the

southwestern part of the Baltic Basin are generally low with values around 5-14% and 5-24

mD caused by intensive quartz cementation (Schleicher et al., 1998; Semyrka et al., 2010),

whereas higher porosities characterise the eastern basin part. For instance, reservoir

porosity can be as high as 20% in western Lithuania and 34% in Kaliningrad (Zdanavičiūtė,

2012).

In the northern part of the Baltic Basin, oil is produced from about 100 limestone reef

structures situated in the Ordovician limestone sequence (Sivhed, 2004). In general, the

reefs are located at shallow depth of only a few hundreds of metres. The onshore reefs are

generally fairly small with a maximum diameter of 800 meters (2600 ft) and an height of up

to 50 meters (160 ft), but can be up to two km in diameter offshore (Tuuling and Flodén,

2009). The oil is produced from fractures, cavities, and vuggy porosity of the limestone

(Chatzis, 2014).

5.3.4 Maturation and Accumulation

Subsidence during the Caledonian orogeny is the main maturation control, especially in the

southwestern part of the Baltic basin. Volcanic intrusions during Permo-Caboniferous

times led to a localized high maturation of the Palaeozoic shales offshore Kaliningrad

(Motuza et al., 1994; Motuzaa et al., 2015), island Bornholm in the Baltic Basin (Obst,

2000), and in central Sweden (Dahl et al., 1989), but only slightly contributed to the overall

thermal maturation (Karnkowski et al., 2010). The 1-D burial reconstruction carried out by

Kosakowski et al. (2010) and the 2-D basin modelling by Wróbel and Kosakowski (2010)

indicate that the onset of petroleum generation from the lower Palaeozoic source rocks

occurred from the Early Devonian through the Mississippian period. The peak of

hydrocarbon generation took place from the Late Devonian to the Mississippian time and

100

main expulsion almost concurrently happened. During Permian and Mesozoic times

tectonic activities led to remobilization with following adjustment of accumulations.

During the passive continental margin stage, normal fractures were widely developed in the

basin and served as petroleum migration paths. For example, oil produced in the

southwestern Baltic basin migrated eastward and northward up to shallower Ordovician

limestone reservoirs through fractures, sandstones, and unconformity connections.

Migration from younger source rock to older reservoirs in the Baltic Basin and into

surrounding areas can also be observed. Examples are pyrobitumens found in Lower

Cambrian sandstones on Bornholm island (Denmark) (Møller and Friis, 1999) as well as

asphaltite in central Swedish bedrock fractures (Sandström et al., 2006) where there is no

effective source rock beneath. This is caused either by downward migration due to the

overpressure and abundance of fractures, or more likely due to the upward charging from

younger source rocks into the tectonically uplifted older strata as suggested by Wróbel and

Kosakowski (2010).

5.4 Samples and Methods

5.4.1 Samples

In the frame of this study, 15 Llandovery shales and Alum Shale samples were investigated

(Table 5.1). Of these five Upper Cambrian Alum Shale samples from Sweden were also

available, but did not deliver sufficient extraction amounts for geochemical investigations

and were excluded.

15 oil samples from five countries have been analysed (Table 5.1). Two oil samples from

the Brattefors quarry were taken directly from a fresh oil seepage after the mining company

had cut a Middle Ordovician limestone which are similar to Fig. 20b as described by

Buchardt et al. (1997). The other samples are from boreholes in the basin area.

The oil samples cover most of the oil producing areas in the Baltic Basin (Fig. 5.1).

Previous geochemical results of analysed oil samples from the Baltic States were reported

by Bojesen-Koefoed et al. (2001). This basic data set provides a meaningful geographic and

stratigraphic distribution, and is used in this study to build on with further conceptual

approaches and techniques to unravel the oil-source correlation of the Early Palaeozoic

petroleum system in the Baltic Basin.

101

5.4.2 Methods

Soxhlet extraction was carried out on powdered shale and solid bitumen samples for 24 hrs

at 50 °C. A dichloromethane and methanol mixture (99:1) was used as solvent. In a next

step, asphaltenes were precipitated from the extracted bitumen as well as from the crude oil

samples before medium pressure liquid chromatography (MPLC) fractionation. Aliphatic

and aromatic hydrocarbons, and NSO (nitrogen, sulphur and oxygen) compounds were

separated from the maltene fraction after MPLC as described by Radke et al. (1980).

Gas chromatography with flame ionization detector (GC-FID) was carried out on the

aliphatic hydrocarbon fractions of all samples. The instrument was equipped with a HP

Ultra 1 capillary column. The oven temperature was programmed from 40 °C to 300 °C

with a 5 °C/min heating rate.

Biomarkers from aliphatic, aromatic, and NSO compounds were detected by gas

chromatography-mass spectrometry (GC-MS). Gas chromatography-mass spectrometry-

mass spectrometry (GC-MS/MS) was applied to detect aliphatic biomarkers for better peak

separation. GC-MS test runs were based on a Trace GC Ultra system coupled to a DSQ

mass spectrometer. The GC was equipped with a programmed temperature vaporizer

(PTV) injection system and a fused silica capillary column, and was heated from 50 °C to

310 °C at a rate of 3 °C/min. GC-MS/MS measurements were performed on a Finnigan

MAT 95XL mass spectrometer coupled to a HP 6890A gas chromatograph with a PTV

injection system. The GC oven temperature was programmed from 50 °C to 310 °C with a

heating rate of 3 °C/min, followed by an isothermal phase of 30 min.

5.5 Results

5.5.1 GC-FID

All samples show very high signal/noise ratios for normal alkanes and acyclic isoprenoids,

and pristane/nC17 and phytane/nC18 ratios are generally low (Table 5.1 and Fig. 5.3). The

biodegradation scale by Peters and Moldowan (1991) suggests that all samples are non- to

slightly biodegraded (Table 5.1). Thus, a very weak impact from biodegradation on the

following oil-source correlation is expected.

No significant odd or even carbon number predominance of n-alkanes pattern was

observed in the samples. Most of them have Pr/Ph ratios between 1.0-3.0, except the three

Llandovery Shale extracts from the B3 well with Pr/Ph ratios higher than 3.0 (Table 5.1).

102

Fig. 5.3. Diagram of pristane/n-C17 vs.phytane/n-C18 with two representative GC-FID traces. Depositional

environment and biodegradation can be evaluated accordingly (Peters et al., 1999).

5.5.2 Aliphatic Biomarkers

Based on the m/z 191 and m/z 217 traces of aliphatic biomarkers, at least two endmember

types can be identified. For example, the Llandovery shale extracts are characterized by

higher concentrations of C24 tetracyclic terpane and C28-C29 tricyclic terpane peaks are

significantly smaller than Ts and Tm (Fig. 5.4A). The sterane and diasterane biomarkers of

the Llandovery shale samples are dominated by C27 and C29 homologues (Fig. 5.4B). In

contrast, the C24 tetracyclic terpane contents in the Alum Shale extracts and reservoir oil are

very low, and Ts and Tm peaks are less predominant. Furthermore, these samples show

obviously higher C28/C29 steranes ratio in comparison with the Llandovery shale extracts

(Fig. 5.4B).

All samples have low concentrations in gammacerane and C35 homohopane. Accordingly,

confident interpretations of these biomarkers can only be achieved in a few samples (Table

5.1). Other redox-, precursor-, and age-related biomarkers, like 28,30-Bisnorhopane,

dinosteranes, and 24-norcholestanes were not identified applying through GC-MS/MS

analysis from any of the samples. 24-iso-/n-propylcholestanes were only recognized in

sample “Deb”.

103

Fig. 5.4. Comparison of terpane and sterane traces of three representative samples. Traces shown here were

provided by GC-MS for direct visual comparison purpose. Ts/Tm ratios and sterane biomarkers presented in table

5.1 and figures were interpreted from GC-MS/MS to ensure better data quality.

5.5.3 Aromatic and NSO Biomarkers

Naphthalenes, phenanthrenes, and dibenzothiophenes can be clearly identified in all

samples, while the aromatic steroid concentrations differ significantly (Fig. 5.5). The Lower

Ordovician and Upper Cambrian Alum Shale samples show no or very little C26-C28

triaromatic steroid distribution patterns and all oils from Middle Cambrian reservoirs are

also depleted in these compounds (Table 5.1 and Fig. 5.5). When comparing the polycyclic

aromatic hydrocarbons (PAHs), most samples are featured by very low

pyrene/phenanthrene ratios, only oil samples from the Brattefors quarry are characterized

by pyrene concentrations higher than phenanthrene.

In the investigated NSO fractions, the carbazole concentrations vary among the samples.

In general, shale extracts are always rich whereas only four oil samples from Middle

Cambrian sandstone and oil seepage in the Brattefors quarry show identifiable carbazoles

signals (Table 5.1).

104

Fig. 5.5. The distributions of triaromatic steroids (m/z=231) in three Alum Shale extracts and two typical oil

samples. C20-C21 triaromatic steroids can be detected from all samples, but some samples show no/very low C26-

C28 responses.

5.5.4 Oil Family Assignments

Chemometric tools which use multivariate statistics to remove noise and to show affinities

among samples are very helpful in oil-source correlations (Peters et al., 2013; Peters et al.,

2016). Depositional environment-, precursor-, and age-related biomarkers were chosen to

as meaningful input parameters for principal component analysis (PCA) and hierarchical

cluster analysis (HCA), instead maturity dependent parameters were not integrated (Table

5.1).

The gained PCA results point to differences between Llandovery shale extracts and the rest

of the samples. The key signals for correlation are extended tricyclic terpane ratios (ETR),

105

C24TeT/C26TT, and C29 sterane distributions, whereas other ratios are relatively less

significant (Fig. 5.6A). Two main clusters can be grouped by HCA data (Fig. 5.6B), and the

Llandovery shale extracts are clearly separated from the Alum Shale extracts and the oils.

However, additional sub-clusters can be identified within cluster I: the Alum Shale extracts

and two oil samples from the Brattefors quarry are closely related, and the oil sample Gec

appears to be a mixture of the two end members (Fig. 5.6).

It has to be pointed out that almost none of the source-related and age-related biomarkers

used here are totally maturity independent and vertical and horizontal heterogeneities of

the source rock can also significantly influence the result of the oil-source correlation. Thus,

detailed discussions on the results are necessary.

Fig. 5.6. (A) Principal component analysis on oil samples based on representative biomarkers and (B) hierarchical

cluster result in grouping the oils into two groups.

106

5.6 Discussion

5.6.1 Maturity

The maturity assessment is not only important in evaluating the influence of local

hydrothermal overprints, but also helpful in unravelling the impact of maturity on oil-

source correlations. A good correlation can be found between the maturity biomarker

ratios diasterane/sterane and Ts/Tm (Fig. 5.7A). It has to be pointed out that both of

these two maturity parameters can be influenced by mineralogy (McKirdy et al., 1981;

Rubinstein et al., 1975), for example, carbonate-rich source rocks are characterized by small

values of these ratios. However, the mineralogical impact on the maturity seems unlikely,

since all samples are marine siliciclastic sediments. The C29 sterane epimer ratios are also

considered as important maturity parameters and most of the samples show maturities

within their thermodynamic equilibrium thresholds (Fig. 5.7B).

In general, the four maturity parameters show that two Llandovery shale samples (Gen and

Ruk) and one Alum Shale (B7) sample are in the “oil window” whereas all other shale

samples are immature or in the early stage of oil generation (Fig. 5.7).

The maturity parameters show that the oils in the Ordovician reefs were mainly charged

from source rocks during the early “oil window” except for the two samples from

Brattefors. Oils in Middle Cambrian reservoirs are generally more mature. Of these the C9

and Gec oils are suggested to represent in peak “oil window” (Fig. 5.7B).

Fig. 5.7. Cross plots of terpane and sterane biomarkers show the maturities of source rock and oil samples.

Yellow stripes in panel B manifest the thermodynamic equilibrium intervals of the sterane isomer.

107

5.6.2 Volcanic Intrusion Induced Maturation

The Bra/a and Bra/b oil samples taken from Middle Ordovician limestones in the

Brattefors quarry present 20S/(20S+20R) sterane ratios which are significantly higher than

the thermodynamic equilibrium threshold (Fig. 5.7B). This implies that these samples are

likely to have experienced abnormally fast heating during maturation (Strachan et al., 1989).

Furthermore, only these two samples have high pyrene/phenanthrene ratios. The

pyrene/phenanthrene ratio in petroleum is generally low, and elevated values for this ratio

are suggested to be caused by pyrolytic maturation (Budzinski et al., 1997). Because rapid

heating and high temperatures can cause the breaking of bonds in organic matter leading to

small molecular fragments, mostly free radicals recombine to form PAHs by pyrosynthesis

as the temperature drops (Huang et al., 2015b). For instance, PAHs are enriched in artificial

pyrolysates (Brocks et al., 2003), hydrothermally generated oil (Simoneit and Lonsdale,

1982), sediments influenced by forest fires (Venkatesan and Dahl, 1989), or oils generated

by near volcanic intrusions (Huang et al., 2015b).

The regional maturity pattern reveals that potential source rocks are immature in south

central Sweden (Fig. 5.1), and that there is a lack of pathways for the oil to migrate from

the basin centre. Due to the widespread occurrence of Permo-Carboniferous diabases in

this area, oil samples from the Middle Ordovician Brattefors limestone are genetically

related to this volcanic intrusion period (Dahl et al., 1989). Sterane maturity and PAHs

characteristics similar to the Brattefors oil is missing in the other investigated samples, so

that oil formation in the Baltic State area is the result of geological burial.

5.6.3 Correlation

5.6.3.1 Acyclic Isoprenoids

Traditionally Pr/Ph was applied to differ oxic from anoxic depositional environments

(Didyk, 1978). However, palaeosalinity (ten Haven et al., 1987), maturity (Dzou et al., 1995),

and organic matter type input (Goossens et al., 1984) were also considered as influencing

factors. Thus, Pr/Ph values between 0.8-3.0 offer a broad interpretation window, and are

not exclusively indicative for a depositional environment assignment (Peters et al., 2005).

In our sample set, three Llandovery shale samples from well B3 have Pr/Ph ratios higher

than 3.0 (Table 5.1 and Fig. 5.3), and, despite other considerations, could refer to terrestrial

organic matter input under oxic conditions (Peters et al., 2005). However, advanced

108

terrestrial plant community is widely believed not to flourish before the Devonian time

(Kenrick and Crane, 1997), but recent studies reveal the occurrence of first

eutracheophytes in the Late Ordovician (Gerrienne et al., 2016), and that sediments across

the Late Ordovician-Early Silurian boundary in Sweden host spores (Badawy et al., 2014).

Accordingly, some terrestrial organic matter can have been admixed into marine

depositional environments during the foreland basin stage. In summary, the high Pr/Ph

ratios in some Llandovery shale samples are reasonable indications that the depositional

environment during the early Silurian time was stronger oxidized than during deposition of

the Alum Shale.

5.6.3.2 Terpanes

The routine BNH/H ratio, Gammacerane index and C35 homohopane index are not

indicative in differentiating the Lower Palaeozoic samples in this study, because most of

these biomarkers were not detected by GC-MS and GC-MS/MS analyses. As suggested by

PCA, ETR and the C24 tetracyclic/C26 tricyclic terpane (C24TeT/C26TT) ratio were

compared and a nice correlation can be found (Fig. 5.8). The Llandovery shale extracts

show very low ETR values and enrichments in C24 tetracyclic terpanes, whereas the Alum

Shale extracts and oil samples from Upper Ordovician limestone reservoirs plot in the

lower right part of the template (Fig. 5.8).

ETR (ETR=[C28+ C29]/ [C28+ C29+Ts]) was taken as an age-related biomarker ratio in

differentiating Jurassic reservoired oil from Triassic (Hanson et al., 2007; Holba et al., 2001)

and is relatively resistant to thermal maturation and biodegradation (Holba et al., 2002).

The ETR values correlate well with the Gammacerane Index and Pr/Ph ratio (De Grande

et al., 1993; Hao et al., 2011), and are thus considered indicators for salinity and alkalinity.

The C24 tetracyclic terpane is considered as a biomarker in evaluating depositional

environments (Tao et al., 2015; Volk et al., 2005) and can be enriched in carbonate rocks

(Palacas et al., 1984), evaporites (Clark and Philp, 1989), or rocks with terrestrial organic

matter input (Philp and Gilbert, 1986). Furthermore, the tetracyclic terpanes are regarded

to be more resistant against biodegradation than tricyclic terpanes due to the more stable

stereoisomerism (Aquino Neto et al., 1981; Williams et al., 1986). Thus, the C24TeT/C26TT

ratio would increase with intensified biodegradation level. However, the application of the

PM scale (Peters and Moldowan, 1991) shows that the bitumen in the Llandovery shale has

undergone very slight biodegradation (Table 5.1), i.e. only a few n-alkanes have been

109

depleted (see GC-FID traces in Fig. 5.3). The more biodegraded oil samples, e.g. “Adz”

and “Ber” exhibit only very low C24TeT/C26TT ratios (Table 5.1 and Fig. 5.8). In summary,

the biodegradation of these samples is not as pronounced as to control the

tetracyclic/tricyclic terpane distribution.

The scattered distribution of oil samples and the Alum Shale extracts in Fig. 5.8 can be

explained by the combination of facies variations and maturity changes. Deposited in a vast

area the Alum Shale is characterized by vertical and horizontal variations during deposition

across the area. With increasing maturity, the ETR is anticipated to be decreased, because

of the relative Ts enrichment in the denominator of the ratio. Also, Farrimond et al. (1999)

and Huang et al. (2015a) pointed out that the C24TeT/C26TT ratio would slightly increase

especially after peak oil generation although this is not the case in our sample set. These

maturation controls can partly explain why the more mature oil samples in the Middle

Cambrian sandstone reservoir have lower ETR and higher C24TeT/C26TT ratios than the

oil from the Upper Ordovician limestone. Summarising the above, the oil samples can be

regarded as genetically closer to the Alum Shale than to the Llandovery shale.

Fig 5.8. The terpane biomarker cross-plot can efficiently separate Llandovery source rocks from the rest. C24 TeT

and C26 TT are abbreviations for C24 tetracyclic terpane and C26 tricyclic terpane respectively. The extended

tricyclic terpane ratio (ETR) (Holba et al., 2001) is calculated from (C28+C29)/(C28+C29+Ts) in m/z 191. The size

of the dot demonstrates the maturity of each sample to evaluate the maturity dependency of the biomarkers.

110

Although both were deposited below wave base under dysoxic to anoxic conditions, the

Alum Shale and Llandovery shales differ in depositional features. The Alum Shale was

deposited in a large and shallow epicontinental sea (Schovsbo, 2002; Thickpenny, 1984),

while the Llandovery shale was formed in a foreland basin with open circulation to the

ocean. The salinity differences indicated from biomarkers might reflect differences in water

column stratification between the two settings, and the Alum Shale was deposited in water

column with stratification. In general, organic lean shale, sandstone and carbonate

intercalations can be found more frequently in Lower Silurian shales than in the Alum

Shale (Lazauskiene et al., 2003; Modliński and Podhalańska, 2010) which implies a less

stratified, more unstable, and shallower-water prevailed environment during deposition.

5.6.3.3 Steranes

The ternary diagram about the C27, C28, and C29 distribution is meant to enable the

distinction of source rocks or oils from different ecosystems or depositional settings

(Huang and Meinschein, 1979).

The Llandovery shale samples can be clearly separated from the rest of the samples due to

their elevated C29 proportions. The Alum Shale extracts and the oil samples overlap (Fig.

5.9) and thus confirm marine depositional conditions (Moldowan et al., 1985). The oil

sample Gec could be partly co-sourced by the Llandovery shale which is indicated by very

similar sterane distributions (Fig. 5.9). With an elevated C29 concentration, the C9 oil could

also be influenced by Llandovery shale (Fig. 5.9). Samples with C29 sterane concentrations

larger than 60% are typically referred to an organic material dominated by terrestrial

precursors (Huang and Meinschein, 1979), and may be related to the early advanced plant

development in this area. Alternatively, the C29 steranes can also be retraced to green algae,

e.g. Chlorophyceae (Grantham and Wakefield, 1988; Katz and Everett, 2016; Kelly et al.,

2011). Consequently, the precursor of the organic matter in the Llandovery shale can be

clearly differentiated from those organisms which delivered the organic matter preserved in

the Alum Shale and most of the oil samples.

The Deb oil is the only sample with identifiable 24-iso-/n-propylcholestanes. This C30

sterane is interpreted as an indicator of marine algae (Moldowan et al., 1990) or

demosponges (Love et al., 2009). C30 steranes were not found in the Alum Shale and in oils

in Cambrian reservoirs (Moldowan et al., 1990). These data are based on the analyses of a

limited sample set and lead to the conclusion that the precursor sterols for the formation of

111

C30 steranes appeared between the Early Ordovician and the Devonian. However, the

detection of C30 steranes is dependent on the instrumental sensitivity and often due to

contamination from lubricating oil during drilling (Antcliffe and Stouge, 2013). Since the

C30 sterane signals in the “Deb” oil sample are generally weak, it is hard to explicit the

geological meaning of these compounds in this sample so far.

To briefly summarize the above findings, the Lower Palaeozoic oil found in the Baltic

Basin is primarily generated from the Alum Shale. The Llandovery shale could add only a

subordinate contribution to these oils in the area.

Fig. 5.9. Ternary diagram showing the relative distribution of C27, C28 and C29 iso-steranes [5α,14β,17β(H)

20S+20R]. The Llandovery shale extracts and oil sample Gec are featured by high concentrations in C29 iso-

sterane.

5.6.4 Heterogeneity within Alum Shale

The Alum Shale is vertically and horizontally highly heterogenous in terms of lithology

(Thickpenny, 1984), depositional environment (Schovsbo, 2001), rock types (Schovsbo et

al., 2015), mineralogy (Snäll, 1988), artificial pyrolysates (Bharati et al., 1995; Horsfield et al.,

1992a), and uranium concentrations (Kosakowski et al., 2016; Schovsbo, 2002). For

example, the Upper Cambrian and Lower Ordovician Alum Shale in the northern part of

112

the basin are rich in uranium (> 100ppm) (Schovsbo, 2002) which was either induced by

hydrothermal activity (Mossman et al., 1993; Oliver et al., 1999) or controlled by the

depositional environments (Schovsbo, 2002), while the Middle Cambrian Alum Shale is

generally poor in uranium. However, only minor variations in source-related and age-

related aliphatic biomarker compositions can be found among the three sub-stages.

Nevertheless, triaromatic steroids distribution could be helpful in differentiating them.

Two Alum Shale extracts lack in or have very low C26-C28 triaromatic steroid concentrations

although C20-C21 triaromatic steroids can be identified in all samples (Table 5.1 and Fig. 5.5)

and the strong resistance of triaromatic steroids against biodegradation (Larter et al., 2012;

Peters and Moldowan, 1991) argues against a removal by biodegradation. In general,

triaromatic steroids are considered as aromatization products from monoaromatic steroids.

During thermal maturation the C26-C28 / C20-C21 triaromatic steroid concentrations will

decrease due to the preferential degradation of the long-chain triaromatic steroids (Beach et

al., 1989; Mackenzie et al., 1981). The C26-C28 triaromatic steroids are expected to be

entirely cracked to shorter chained compounds during condensate or wet gas generation

stages (Peters et al., 2005). The Baltic samples without C26-C28 triaromatic steroids are either

immature or in the peak oil generation stage which implies that those compounds were not

thermally cracked. Dahl et al. (1988a) and Lewan and Buchardt (1989) found that the C26-

C28 triaromatic steroid concentrations in the Alum Shale are inversely proportional to its

uranium content and concluded that the uranium irradiation initiates the cleavage of the

aliphatic side chains of high molecular weight steroids to form lower molecular weight

steroids. This theory fits the results of our work as the two Alum Shale extracts without

C26-C28 triaromatic steroids are from the uranium-rich Upper Cambrian and Lower

Ordovician Alum Shale (Table 5.1), and as the Middle Cambrian Alum Shale samples with

significantly higher C26-C28 triaromatic steroid responses are generally lean in uranium

concentrations (Schovsbo, 2002). Importantly, the oil sample from the Middle Cambrian

sandstone reservoirs contains no C26-C28 triaromatic steroids and is probably sourced from

the uranium-rich Alum Shale interval. In contrast, the oil samples in the Upper Ordovician

reef reservoirs all have high triaromatic steroid compounds and thus can be genetically

linked with the Middle Cambrian Alum Shale. However, further geological or geochemical

evidence is required to support this hypothesis. The weak response of C26-C28 triaromatic

steroids in the Brattefors oil samples is either an original one or caused by the high

temperature influence during volcanic intrusions.

113

5.6.5 Migration and Mixing

The distribution of pyrrolic nitrogen compounds is considered to reflect migration

distances, the abundance of carbazoles and the benzo[a]carbazole/benzo[c]carbazole

isomer ratio both decrease with increasing migration distance (Larter et al., 1996; Li et al.,

1995). However, Horsfield et al. (1998) and Bakr and Wilkes (2002) have pointed out that

the benzocarbazole isomer ratio is also influenced by maturity and depositional

environment. Nevertheless, the carbazole concentration seems to reflect the migration

distance because the NSO compounds undergo retention due to geochromatographic

effects and their less affinity to hydrocarbons (Silliman et al., 2002; van Duin and Larter,

1998). All oil samples from Upper Ordovician limestone reservoirs are depleted in

carbazole compounds (Table 5.1). Only four oil samples from Middle Cambrian reservoirs

(C9, B8, Deb, and Gec) have clear carbazole signal responses (Table 5.1), and indicate

relatively short migration distances from the source rock. This conclusion is supported by

the Biomarker Migration Index theory suggested by Seifert and Moldowan (1981) and

Carlson and Chamberlain (1986)Carlson and Chamberlain (1986) as these four oil samples

plot along the source sample maturity trend line in the C29 sterane isomer maturity diagram

(Fig. 5.7B). This feature corresponds to the First Order Kinetic Conversion lines and oil

samples Sak and Gus are shifted to the right of the trend line due to strong fractionation

during migration (Fig. 5.7B). In other words, the source rock for the oil in the reservoirs is

“close by” and this phenomenon fits the regional maturity trend (Fig. 5.1).

Although the Alum Shale is regarded to be the main source rock for the Lower Palaeozoic

petroleum, contributions from the Llandovery shales to some reservoirs is possible. For

example, the Gec oil sample is generally grouped between the Alum Shale and the

Llandovery shale in the chemometric analysis (Fig. 5.6), and is characterized by a similar

terpane distribution as the Alum Shale (Fig. 5.8) while the C29 sterane ratio resembles the

Llandovery shale (Fig. 5.9). The Deb oil sample is among the least mature samples

considering Ts/Tm and Diasterane/sterane ratios (Fig. 5.7A), but C29 sterane isomers

indicate a higher maturity than most of the oil in Upper Ordovician reef reservoirs. This

inconsistency in maturity could also be caused by multi-charging processes from the same

source through time, since carbonate-rich source rocks are not involved here. Most likely

the oil in the Middle Cambrian sandstone reservoirs is exclusively sourced by the uranium-

rich Alum Shale, because C26-C28 triaromatic steroids would occur in this oil if other source

rock were involved.

114

5.7 Conclusions

1. Chemometric tools are efficient in screening decisive oil-source rock correlation

biomarkers and showing genetic affinity relationship of samples.

2. Oil samples from Brattefors quarry in south central Sweden were heated by volcanic

intrusions with high pyrene/phenanthrene ratios and have C29 S/R sterane ratios

significantly beyond the normal thermodynamic threshold. Oil reservoired in basinal

areas is generated by normal geological burial processes.

3. ETR, C24 tetracyclic terpane/C26 tricyclic terpane, and relative C29 sterane

concentrations are the most useful biomarkers in differentiating Llandovery shale from

Alum Shale. The uranium irradiation induced decrease of C26-C28 triaromatic steroids

in Alum Shale can be a practical key signal to link oil to the specific sub-stages of the

Alums Shale.

4. Alum Shale is the main source rock for the oil in the Lower Palaeozoic sandstone and

reef reservoirs in the Baltic Basin. A contribution from the Middle Cambrian Alum

Shale to the oil in Upper Ordovician limestone reservoirs is noticeable. In contrast, oil

in Middle Cambrian siliciclastic reservoirs was mainly sourced from more mature

Upper Cambrian and Lower Ordovician Alum Shale.

5. The evaluation of migration distances can help in understanding the accumulation

processes. Possible oil mixing occurs in some Lithuanian wells where the Llandovery

shale occurs with good source rock properties.

5.8 Acknowledgement

The authors thank Gripen Oil & Gas AB, the Geological Survey of Sweden and Dr. Jurga

Lazauskiene for providing rock and oil samples. Cornelia Karger and Anke Kaminsky

(both from GFZ Potsdam) are acknowledged for their technical support during laboratory

work.

Reviews and constructive comments from Pawel Kosakowski and Jon H. Pedersen are

much appreciated. Kind thanks are extended to AAPG Bulletin Editor Barry Katz for his

comments and suggestions.

115

6. SUMMARY AND PESPECTIVES

6.1 Summary

6.1.1 MME

The extent of the MME on the Bowland Shale is significant and can be attributed to its

high clay content. Previous findings of sedimentological features, biomarkers, and stable

carbon isotope data revealed that the Bowland Shale was deposited in a marine

environment, and these results fit the Rock-Eval data of kerogen samples rather than

whole rock samples in this dissertation. Compared with data of kerogen concentrates, the

whole rock samples have lower HI and higher OI, and thus, point to type III kerogen. Also,

the GOR and aromaticity data gained from analyses of the pyrolysates are significantly high

in the whole rock samples suggesting a gas and a condensate potential; this is furthermore

confirmed by phase predictions based on MSSV data. The bulk kinetic and secondary

kinetic features of the Bowland Shale were also influenced by the complex mineral matrix

effects. As a result, questionable predictions of the secondary gas generation in basin

modelling studies can be induced by the application of such kinetic data. Thus, the MME

strongly impacts the quantity, quality, and timing of the petroleum generation of the

Bowland Shale, and especially regarding the quantitative modelling of secondary gas

generation. It is important to note that secondary gas generation contributes the major part

of thermogenic shale gas production.

The extent of MME in shale varies according to the mineralogical composition. In contrast

to the Bowland Shale, the calcite-rich Toolebuc Oil Shale and the quartz-rich Alum Shale

are not or only slightly affected by MME. Furthermore, the interface area between clay

minerals and organic matter could be another factor that influences the extent of MME.

The heating rate dependency of MME on the Bowland Shale is obvious as revealed by

MSSV experiments. A lower heating rate weakens the MME on aromaticity and generation

amount. A calibration sequence with natural samples manifests that neither the kerogen

nor the whole-rock pyrolysates show a similar gross elemental composition as natural

products. However, kerogen pyrolysates resemble natural products more in certain NSO

class ratios compared with whole-rock pyrolysates, which implies that MME on NSO

116

compounds is diminished under geological situations. In brief, the MME is speculated to

only exist in the laboratory environment and not during a geological maturation process.

6.1.2 Uranium

Immature Alum Shale samples generally contain higher uranium concentrations, up to 20

times more than other typical marine shales. Pyrolysates, solvent extracts, and kerogen

structures of the Alum Shale were all changed by the uranium irradiation.

Atypical pyrolysis products can be generated from the marine Alum Shale. GOR,

aromaticity, and gas quantity are significantly higher compared with typical marine shales.

The HI and OI are not well correlated with uranium contents compared with data from

previous publications. HI seems to be controlled by original depositional rates and OI is

typically low (<5 mgHC/gTOC). In contrast, GOR and aromaticity of pyrolysates are

exponentially correlated with U contents in the source rock. These findings imply that the

original kerogen structures are uniform and irradiation caused a higher aromaticity and gas-

prone character.

The o-xylene/n-C9 alkane ratios gained by thermovaporisation resemble laboratory

pyrolysates. Also, the GORs from thermovaporisation can be generally correlated with

pyrolysates. Some discrepancies are likely caused by the random decrease of GOR through

weathering, sampling and storage. The hydrocarbons in the immature shale samples reflect

early generated bitumen or were formed by the bombardment of uranium radiation. These

findings confirm that uranium irradiation effects do influence the petroleum generation in

natural environments.

The FT-ICR MS data reveal that the NSO compounds in the uranium-rich Alum Shale

samples are less alkylated, and reflect the kerogen structures. This could be the reason why

a significant amount of gas can be generated from such kind of samples. Basin modelling

investigations suggest that most of the Alum Shale reached peak petroleum generation

from Late Devonian to Early Carboniferous (360-385 Myr ago) in the basin centre

(Kosakowski et al., 2010), thus, the source rocks at that time have experienced 20%-30%

dosage of irradiation compared with nowadays samples which undergone irradiation for

478-500 Myr. Based on the correlations between irradiation damage, irradiation time and

uranium contents, the Alum Shale hydrocarbon generation properties, e.g., GOR and

aromaticity are calculated for the oil window time. This reconstruction plays a crucial role

in predicting the quality and quantity of petroleum that has been generated.

117

The unique distribution of triaromatic steroids in uranium-rich Alum Shale samples was

found in its petroleum. The absence of the high molecular weight (C26-C28) compounds is

not a matter of thermal maturity, biodegradation, or migration fractionation, but is more

likely due to uranium irradiation.

To summarise, uranium irradiation decreases the alkylation of kerogen structures and also

plays an important catalytic role in converting the kerogen into gaseous and aromatic

hydrocarbons in pyrolysis experiments through condensation and cross-linking

mechanisms.

6.1.3 Comparison

Marine source rock samples with high contents of clay minerals or uranium may present

similar features in pyrolysis experiments, i.e., high GOR and high aromaticity in the

pyrolysates which are similar to type III kerogens. But they still have fundamental

differences. For example, phenol and cresol which are dominant compounds in the

pyrolysates of terrestrial organic matter are nearly absent in pyrolysates of the clay-rich and

uranium-rich marine shale.

Excluding those similarities, clay minerals and uranium have intrinsic differences in

influencing petroleum generation and occurrence through different processes. The

interactions between solids and organic matter during diagenesis are limited due to few

interface reactions, and the catalytic effect of clay minerals is only effective under very high

temperatures. However, the irradiation is penetrative and the spontaneous fission of

uranium is independent of temperature and pressure. Therefore, the kerogen structure of

clay-rich shale would not significantly be changed during diagenesis, but uranium

irradiation can severely alter the molecular structure of kerogen during geological time

scales. A removal of minerals by acidic dissolution can easily avoid the MME in the

laboratory, while the influence of uranium during pyrolysis is impossible to be eliminated.

Both clay minerals and uranium have certain impacts on biomarkers used for maturity

assessments. Reactive clay mineral entities mainly increase aliphatic biomarkers applied as

maturity indicators, e.g., Ts/Tm or diasteranes/steranes ratios. By contrast, the influence of

uranium on biomarkers is mainly focused on aromatic biomarkers, e.g., TA(I)/TA(I + II).

118

6.2 Perspectives

6.2.1 MME

Although MME result in an artefact during laboratory pyrolysis, possible false appearances

of HI, OI, kinetics, and open-pyrolysis results, which are crucially important in evaluating

source rock type and maturity, can finally cause significant risks in petroleum exploration.

The preparartion of kerogen concentrates from each sample before pyrolysis for a

screening purpose is time consuming, but shale samples poor in organic matter or rich in

clay contents must be re-examined. Special attention has to be paid on kinetic data

regarding the MME, because the timing of petroleum generation in basin modelling is very

sensitive to the kinetic result. For example, a 1 oC error caused during laboratory work

would result in a 3 oC uncertainty in the geological extrapolation (Burnham, 1994b).

Results gained from the application of pyrolysis techniques allow an estimate about the

petroleum generation and reflect the kerogen structure, whereas the heating rate

dependency of MME emphasises the differences between laboratory heating and geological

maturation. For example, the NSO compound concentrations in pyrolysates are

significantly higher than those in natural products (Horsfield, 1997). High temperatures

(>300 oC) and fast heating rates used in the laboratory would enable certain reactions

which would not happen efficiently through geological environments. Therefore, some

data gained from pyrolytic techniques should not be over-interpreted, but critically re-

evaluated.

6.2.2 Uranium

Based on the application of pyrolytic methodologies, evidence for the shale gas potential of

the Alum Shale was highlighted and that the oil potential is anticipated to be very limited

(Kotarba et al., 2014a; Sanei et al., 2014). In addition, the gas sorption capacity of the Alum

Shale was also expected to be very high due to the high aromatic kerogen structures.

However, the shale gas explorations in Sweden and Denmark were unsuccessful which can

be partly attributed to gas leakages in the response of tectonic movements (Pool et al.,

2012). Furthermore, oil discoveries in the Baltic Basin can be traced back to the Alum

Shale as a source rock by meaningful biomarkers and isotope characteristics which

counteract the predictions based on pyrolysis experiments.

119

The reconstruction of the kerogen structure during the time span of petroleum generation

could avoid an over-estimation of the gas potential and gas sorption capacity, and an

under-estimation of the oil potential, and would thus reduce the exploration risks

significantly.

The unique distribution of triaromatic biomarkers in both Alum Shale extracts and

petroleum caused by irradiation can serve as a new parameter for correlation, especially

when aliphatic biomarkers are absent due to irradiation.

120

121

REFERENCES

Abelson, P.H. (1963) Organic geochemistry and the formation of petroleum,

6th World Petroleum Congress. World Petroleum Congress.

Abouelresh, M.O. and Slatt, R.M. (2012) Lithofacies and sequence

stratigraphy of the Barnett Shale in east-central Fort Worth Basin,

Texas. AAPG bulletin 96, 1-22.

Abu-Ali, M., Rudkiewicz, J., McGillivray, J. and Behar, F. (1999) Paleozoic

petroleum system of central Saudi Arabia. GeoArabia 4, 321-336.

Al-Sammerrai, F., Al-Sammerrai, D. and Al-Rawi, J. (1987) The use of

thermogravimetry and NMR spectroscopy in the attempted

identification of the source of babylonian building asphalt.

Thermochimica Acta 115, 181-188.

Allen, R.G., Pereira, L.S., Raes, D. and Smith, M. (1998) Crop

evapotranspiration-Guidelines for computing crop water

requirements-FAO Irrigation and drainage paper 56. FAO, Rome 300,

D05109.

Alloway, B.J. (2013) Bioavailability of elements in soil, Essentials of medical

geology. Springer, pp. 351-373.

Amyx, J.W., Bass, D.M. and Whiting, R.L. (1960) Petroleum reservoir

engineering: physical properties. McGraw-Hill College.

Anderson, R.F., Fleisher, M.Q. and LeHuray, A.P. (1989) Concentration,

oxidation state, and particulate flux of uranium in the Black Sea.

Geochimica et Cosmochimica Acta 53, 2215-2224.

Andersson, A. (1985) The Scandinavian alum shales. Sveriges geologiska

undersökning.

Andrews, I. (2013) The Carboniferous Bowland Shale gas study: geology

and resource estimation.

Antcliffe, J.B. and Stouge, S. (2013) Questioning the evidence of organic

compounds called sponge biomarkers. Palaeontology 56, 917-925.

Aquino Neto, F.d., Trendel, J., Restle, A., Connan, J. and Albrecht, P. (1981)

Occurrence and formation of tricyclic and tetracyclic terpanes in

sediments and petroleums. Advances in organic geochemistry 10,

659-676.

Armstrong, J.P., Smith, J., D’Elia, V.A. and Trueblood, S.P. (1997) The

occurrence and correlation of oils and Namurian source rocks in the

Liverpool Bay-North Wales area. Geological Society, London, Special

Publications 124, 195-211.

Arnold, R. and van Vleck Anderson, R. (1907) Geology and oil resources of

the Santa Maria oil district, Santa Barbara County, California. US

Government Printing Office.

Arthur Stankiewicz, B., van Bergen, P.F., Duncan, I.J., Carter, J.F., Briggs,

D.E. and Evershed, R.P. (1996) Recognition of chitin and proteins in

invertebrate cuticles using analytical pyrolysis/gas chromatography

122

and pyrolysis/gas chromatography/mass spectrometry. Rapid

Commun Mass Sp 10, 1747-1757.

Badawy, A.S., Mehlqvist, K., Vajda, V., Ahlberg, P. and Calner, M. (2014)

Late Ordovician (Katian) spores in Sweden: oldest land plant

remains from Baltica. GFF 136, 16-21.

Bae, E., Na, J.-G., Chung, S.H., Kim, H.S. and Kim, S. (2010) Identification

of about 30 000 Chemical Components in Shale Oils by Electrospray

Ionization (ESI) and Atmospheric Pressure Photoionization (APPI)

Coupled with 15 T Fourier Transform Ion Cyclotron Resonance Mass

Spectrometry (FT-ICR MS) and a Comparison to. Energy & Fuels 24,

2563-2569.

Baker, E. (1959) Origin and migration of oil. Science 129, 871-874.

Bakr, M.M. and Wilkes, H. (2002) The influence of facies and depositional

environment on the occurrence and distribution of carbazoles and

benzocarbazoles in crude oils: a case study from the Gulf of Suez,

Egypt. Organic Geochemistry 33, 561-580.

Barrett, P.A. (1988) Early Carboniferous of the Solway Basin: a

tectonostratigraphic model and its bearing on hydrocarbon potential.

Marine and petroleum geology 5, 271-281.

Barrow, M.P., Headley, J.V., Peru, K.M. and Derrick, P.J. (2004) Fourier

transform ion cyclotron resonance mass spectrometry of principal

components in oilsands naphthenic acids. J Chromatogr A 1058, 51-

59.

Bates, T.F. (1958) An investigation of the mineralogy, petrography and

paleobotany of uranium-bearing shales and lignites. United States

Atomic Energy Commission, Technical Information Service., Oak

Ridge, Tennessee, p. volume.

Baur, F., Littke, R., Wielens, H., Lampe, C. and Fuchs, T. (2010) Basin

modeling meets rift analysis – A numerical modeling study from the

Jeanne d'Arc basin, offshore Newfoundland, Canada. Marine and

Petroleum Geology 27, 585-599.

Beach, F., Peakman, T., Abbott, G., Sleeman, R. and Maxwell, J. (1989)

Laboratory thermal alteration of triaromatic steroid hydrocarbons.

Organic geochemistry 14, 109-111.

Behar, F., Kressmann, S., Rudkiewicz, J.L. and Vandenbroucke, M. (1992)

Experimental simulation in a confined system and kinetic modelling

of kerogen and oil cracking. Organic Geochemistry 19, 173-189.

Behar, F., Roy, S. and Jarvie, D. (2010) Artificial maturation of a Type I

kerogen in closed system: Mass balance and kinetic modelling.

Organic Geochemistry 41, 1235-1247.

Behar, F., Vandenbroucke, M., Tang, Y., Marquis, F. and Espitalie, J. (1997)

Thermal cracking of kerogen in open and closed systems :

determination of kinetic parameters and stoichiometric coefficients

for oil and gas generation. Organic Geochemistry 26, 321-339.

123

Belonoshko, A.B., Lukinov, T., Rosengren, A., Bryk, T. and Litasov, K.D.

(2015) Synthesis of heavy hydrocarbons at the core-mantle

boundary. Sci Rep-Uk 5.

Bernard, S., Horsfield, B., Schulz, H.-M., Wirth, R., Schreiber, A. and

Sherwood, N. (2012) Geochemical evolution of organic-rich shales

with increasing maturity: A STXM and TEM study of the Posidonia

Shale (Lower Toarcian, northern Germany). Marine and Petroleum

Geology 31, 70-89.

Berry, W.B., Wilde, P., Quinby-Hunt, M.S. and Orth, C.J. (1986) Trace

element signatures in Dictyonema Shales and their geochemical and

stratigraphic significance. Norsk Geol. Tidsskr 66, 45-51.

Bethke, C.M., Vergo, N. and ALTANER, S.P. (1986) Pathways of smectite

illitization. CLAYS CLAY MINER. Clays Clay Miner. 34, 125.

Bharati, S., Patience, R.L. and Larter, S.R. (1995) Elucidation of the Alum

Shale kerogen structure using a multi-disciplinary approach. Organic

Geochemistry 23, 1043-1058.

Bloch, S. (1980) Some factors controlling the concentration of uranium in

the world ocean. Geochimica et Cosmochimica Acta 44, 373-377.

Bojesen-Koefoed, J.A., N., H.P., R., P., B., B., Weiss, H., Kanev, K.,

Zdanavičiūtė, O., Khubldikov, A. and Pokorski, J. (2001) Petroleum

in the Baltic syneclise and the peribaltic region composition and

variability. 20th International Meeting on Organic Geochemistry 1,

419-420.

Boreham, C. and Powell, T. (1987) Sources and preservation of organic

matter in the Cretaceous Toolebuc Formation, eastern Australia.

Organic Geochemistry 11, 433-449.

Bowker, K.a. (2007) Barnett Shale gas production, Fort Worth Basin: Issues

and discussion. AAPG Bulletin 91, 523-533.

Brangulis, A., Kanev, S., Margulis, L. and Pomerantseva, R. (1993) Geology

and hydrocarbon prospects of the Paleozoic in the Baltic region.

Geological Society, London, Petroleum Geology Conference series 4,

651-656.

Braun, R. and Burnham, A. (1987) Analysis of chemical reaction kinetics

using a distribution of activation energies and simpler models.

Energy & Fuels 1, 153-161.

Bray, E. and Evans, E. (1961) Distribution of n-paraffins as a clue to

recognition of source beds. Geochimica et Cosmochimica Acta 22, 2-

15.

Breger, I. (1974) The role of organic matter in the accumulation of

uranium: the organic geochemistry of the coal-uranium association.

Formation of uranium ore deposits, 99-124.

Breger, I.A. and Brown, A. (1962) Kerogen in the Chattanooga shale.

Science 137, 221-224.

Brocks, J.J., Love, G.D., Snape, C.E., Logan, G.A., Summons, R.E. and

Buick, R. (2003) Release of bound aromatic hydrocarbons from late

124

Archean and Mesoproterozoic kerogens via hydropyrolysis.

Geochimica et Cosmochimica Acta 67, 1521-1530.

Brooks, B.T. (1948) Active-surface catalysts in formation of petroleum.

AAPG Bulletin 32, 2269-2286.

Brooks, B.T. (1952) Evidence of catalytic action in petroleum formation.

Industrial & Engineering Chemistry 44, 2570-2577.

Bruce, C.H. (1984) Smectite dehydration--its relation to structural

development and hydrocarbon accumulation in northern Gulf of

Mexico basin. AAPG Bulletin 68, 673-683.

Buchardt, B. (1999) Gas Potential of the Cambro-Ordovician Alum Shale in

Southern Scandinavia and the Baltic Region. The Search for Deep

Gas.

Buchardt, B., Nielsen, A.T., Schovsbo, N.H. and Bojesen-Kofoed, J. (1997)

Alun skiferen i Skandinavien. Geologisk Tidsskrift 1997, 1-30.

Budzinski, H., Jones, I., Bellocq, J., Pierard, C. and Garrigues, P. (1997)

Evaluation of sediment contamination by polycyclic aromatic

hydrocarbons in the Gironde estuary. Marine chemistry 58, 85-97.

Burnham, A.K. (1994a) Comments on “The effects of the mineral matrix on

the determination of kinetic parameters using modified Rock-Eval

pyrolysis” by H. Dembicki Jr, and the resulting comment by R. Pelet.

Organic Geochemistry 21, 985-986.

Burnham, A.K. (1994b) Comments on The effects of the mineral matrix on

the determination of kinetic parameters using modified Rock-Eval

pyrolysis” by H. Dembicki Jr, and the resulting comment. Organic

Geochemistry 21, 985-986.

Burnham, A.K. (2015) A Simple Kinetic Model of Oil Generation,

Vaporization, Coking, and Cracking. Energy & Fuels.

Burnham, A.K., Braun, R.L., Gregg, H.R. and Samoun, A.M. (1987)

Comparison of methods for measuring kerogen pyrolysis rates and

fitting kinetic parameters. Energy & Fuels 1, 452-458.

Burnham, A.K., Braun, R.L. and Samoun, A.M. (1988) Further comparison

of methods for measuring kerogen pyrolysis rates and fitting kinetic

parameters. Organic geochemistry 13, 839-845.

Burnham, A.K. and Singleton, M.F. (1983) High-pressure pyrolysis of Green

River oil shale, ACS Symp. Ser.;(United States). Lawrence

Livermore National Lab., CA.

Burnham, A.K. and Sweeney, J.J. (1989) A chemical kinetic model of

vitrinite maturation and reflectance. Geochimica et Cosmochimica

Acta 53, 2649-2657.

Burst, J.F. (1969) Diagenesis of Gulf Coast clayey sediments and its

possible relation to petroleum migration. AAPG bulletin 53, 73-93.

Burtner, R.L. and Warner, M.A. (1986) Relationship between illite/smectite

diagenesis and hydrocarbon generation in Lower Cretaceous Mowry

and Skull Creek shales of the northern Rocky Mountain area. Clays

and Clay Minerals 34, 390-402.

125

Carlson, R. and Chamberlain, D. (1986) Steroid biomarker-clay mineral

adsorption free energies: implications to petroleum migration

indices. Organic geochemistry 10, 163-180.

Charlesby, A. (1954) The cross-linking and degradation of paraffin chains by

high-energy radiation, Proceedings of the Royal Society of London

A: Mathematical, Physical and Engineering Sciences. The Royal

Society, pp. 60-74.

Chatzis, C. (2014) Description of pore systems from a vuggy Palaeozoic

carbonate, Gotland, Sweden. The University of Manchester, p. 54.

Chiaberge, S., Fiorani, T., Savoini, A., Bionda, A., Ramello, S., Pastori, M.

and Cesti, P. (2013) Classification of crude oil samples through

statistical analysis of APPI FTICR mass spectra. Fuel Processing

Technology 106, 181-185.

Chisholm, J.I., Charsley, T.J. and Aitkenhead, N. (1988) Geology of the

country around Ashbourne and Cheadle. HMSO.

Clark, J. and Philp, R. (1989) Geochemical characterization of evaporite and

carbonate depositional environments and correlation of associated

crude oils in the Black Creek Basin, Alberta. Bulletin of Canadian

Petroleum Geology 37, 401-416.

Claxton, M., Patience, R. and Park, P. (1993) Molecular modelling of bond

energies in potential kerogen sub-units. Organic Geochemistry, 198-

201.

Cohen, K., Finney, S., Gibbard, P. and Fan, J.-X. (2013) The ICS

international chronostratigraphic chart. Episodes 36, 199-204.

Colombo, U., Sironi, G. and Denti, E. (1964) A geochemical investigation

upon the effects of ionizing radiation on hydrocarbons. J. Inst. Petrol.

50.

Constable, F. (1925) The mechanism of catalytic decomposition.

Proceedings of the Royal Society of London. Series A, Containing

Papers of a Mathematical and Physical Character 108, 355-378.

Cooper, J. and Bray, E. (1963) A postulated role of fatty acids in petroleum

formation. Geochimica et Cosmochimica Acta 27, 1113-1127.

Corbett, C.S. (1955) In situ origin of McMurray oil of northeastern Alberta

and its relevance to general problem of origin of oil. AAPG Bulletin

39, 1601-1621.

Court, R.W., Sephton, M.A., Parnell, J. and Gilmour, I. (2006) The alteration

of organic matter in response to ionising irradiation: Chemical

trends and implications for extraterrestrial sample analysis.

Geochimica et cosmochimica acta 70, 1020-1039.

Craig, H. (1953) The geochemistry of the stable carbon isotopes.

Geochimica et Cosmochimica Acta 3, 53-92.

Curiale, J.A. and Frolov, E.B. (1998) Occurrence and origin of olefins in

crude oils. A critical review. Organic geochemistry 29, 397-408.

Curtis, J.B. (2002) Fractured shale-gas systems. AAPG bulletin 11, 1921-

1938.

126

Dahl, J., Chen, R. and Kaplan, I. (1989) Alum Shale bitumen maturation

and migration: implications for Gotland's oil. Journal of Petroleum

Geology 12, 465-476.

Dahl, J., Hallberg, R. and Kaplan, I. (1988a) Effects of irradiation from

uranium decay on extractable organic matter in the Alum Shales of

Sweden. Organic geochemistry 12, 559-571.

Dahl, J., Hallberg, R. and Kaplan, I. (1988b) The effects of radioactive

decay of uranium on elemental and isotopic ratios of Alum Shale

kerogen. Applied geochemistry 3, 583-589.

Dahlkamp, F.J. (2013) Uranium ore deposits. Springer Science & Business

Media.

Damsté, J.S.S., de las Heras, F.X.C., van Bergen, P.F. and de Leeuw, J.W.

(1993) Characterization of Tertiary Catalan lacustrine oil shales:

discovery of extremely organic sulphur-rich Type I kerogens.

Geochimica et Cosmochimica Acta 57, 389-415.

De Grande, S., Neto, F.A. and Mello, M. (1993) Extended tricyclic terpanes

in sediments and petroleums. Organic geochemistry 20, 1039-1047.

DECC (2011) The unconventional hydrocarbon resources of Britain’s

onshore basins - shale gas. DECC Promote website, December 2010.

Dembicki, H. (1990) Mineral matrix effect during analytical pyrolysis of

source rocks. Association of Petroleum Geochemist and Explorer

Bulletin 6, 78-105.

Dembicki, H. (1992) The effects of the mineral matrix on the determination

of kinetic parameters using modified Rock Eval pyrolysis. Organic

Geochemistry 18, 531-539.

Dembicki, H. (1994) Reply to comments by R. Pelet on the paper “The

effects of the mineral matrix on the determination of kinetic

parameters using modified Rock-Eval pyrolysis” by H. Dembicki Jr,

Org. Geochem., 18, 531–539 (1992). Organic Geochemistry 21,

982-984.

Dembicki, H., Horsfield, B. and Ho, T.T. (1983) Source rock evaluation by

pyrolysis-gas chromatography. AAPG Bulletin 67, 1094-1103.

Derenne, S., Largeau, C. and Berkaloff, C. (1996) First example of an

algaenan yielding an aromatic-rich pyrolysate. Possible geochemical

implications on marine kerogen formation. Organic Geochemistry 24,

617-627.

Dessort, D., Connan, J., Derenne, S. and Largeau, C. (1997) Comparative

studies of the kinetic parameters of various algaenans and kerogens

via open-system pyrolyses. Organic Geochemistry 26, 705-720.

di Primio, R., Dieckmann, V. and Mills, N. (1998) PVT and phase behaviour

analysis in petroleum exploration. Organic Geochemistry 29, 207-

222.

di Primio, R. and Horsfield, B. (1996) Predicting the generation of heavy oils

in carbonate/evaporitic environments using pyrolysis methods.

Organic Geochemistry 6380.

127

di Primio, R. and Horsfield, B. (2006) From petroleum-type organofacies to

hydrocarbon phase prediction. AAPG Bulletin 90, 1031-1058.

Didyk, B. (1978) Organic geochemical indicators of palaeoenvnonmental

conditions of sedimentation. Nature 272, 216-222.

Dieckmann, V. (2005) Modelling petroleum formation from heterogeneous

source rocks: the influence of frequency factors on activation energy

distribution and geological prediction. Marine and Petroleum Geology

22, 375-390.

Dieckmann, V., Horsfield, B. and Schenk, H.J. (2000a) Heating rate

dependency of petroleum-forming reactions: implications for

compositional kinetic predictions. Organic Geochemistry 31.

Dieckmann, V. and Keym, M. (2006) A new approach to bridge the effect of

organofacies variations on kinetic modelling and geological

extrapolations. Organic Geochemistry 37, 728-739.

Dieckmann, V., Schenk, H.J. and Horsfield, B. (2000b) Assessing the

overlap of primary and secondary reactions by closed- versus open-

system pyrolysis of marine kerogens. Journal of Analytical and

Applied Pyrolysis 56, 33-46.

Dieckmann, V., Schenk, H.J., Horsfield, B. and Welte, D.H. (1998) Kinetics

of petroleum generation and cracking by programmed-temperature

closed-system pyrolysis of Toarcian Shales. Fuel 77, 23-31.

Disnar, J. and Sureau, J. (1990) Organic matter in ore genesis: Progress

and perspectives. Organic Geochemistry 16, 577-599.

Dole, M. (1958) The Effects of Ionizing Radiation on Natural and Synthetic

High Polymers. J Am Chem Soc 80, 6694-6694.

Douglas, A., Coates, R., Bowler, B. and Hall, K. (1975) Alkanes from the

pyrolysis of recent sediments. Advances in Organic Geochemistry,

357-374.

Down, A. and Himus, G. (1941) A preliminary study of the chemical

constitution of kerogen. Journal of the Institute of Petroleum 27,

426-445.

Dronov, A. and Holmer, L. (1999) Depositional sequences in the Ordovician

of Baltoscandia. Acta-Universitatis Carolinae Geologica, 133-136.

Düppenbecker, S. and Horsfield, B. (1990) Compositional information for

kinetic modelling and petroleum type prediction. Organic

Geochemistry 16, 259-266.

Duran, E.R., di Primio, R., Anka, Z., Stoddart, D. and Horsfield, B. (2013)

3D-basin modelling of the Hammerfest Basin (southwestern Barents

Sea): A quantitative assessment of petroleum generation, migration

and leakage. Marine and petroleum geology 45, 281-303.

Durand, B. (2003) A history of organic geochemistry. Oil & gas science and

technology 58, 203-231.

Durand, B. and Espitalié, J. (1973) Evolution de la matière organique au

cours de l'enfouissement des sédiments. Compte rendus de

l’Académie des Sciences (Paris) 276, 2253-2256.

128

Dyni, J. (2006) Geology and resources of some world oil shale deposits:

scientific investigations report 2005–5294, US Department of the

Interior. US Geological Survey, 49.

Dzou, L., Noble, R. and Senftle, J. (1995) Maturation effects on absolute

biomarker concentration in a suite of coals and associated vitrinite

concentrates. Organic Geochemistry 23, 681-697.

Eglinton, G. and Calvin, M. (1967) Chemical fossils. Scientific American 216,

32-43.

Eglinton, T., Rowland, S., Curtis, C. and Douglas, A. (1986) Kerogen-

mineral reactions at raised temperatures in the presence of water.

Organic Geochemistry 10, 1041-1052.

Eglinton, T.I., Damsté, J.S.S., Kohnen, M.E. and de Leeuw, J.W. (1990)

Rapid estimation of the organic sulphur content of kerogens, coals

and asphaltenes by pyrolysis-gas chromatography. Fuel 69, 1394-

1404.

EIA (2015) Technically Recoverable Shale Oil and Shale Gas Resources.

EIA, U. (2011) World shale gas resources: an initial assessment of 14

regions outside the United States.

Emery, K.O. and Hoggan, D. (1958) Gases in marine sediments. AAPG

Bulletin 42, 2174-2188.

Ericsson, I. and Lattimer, R. (1989) Pyrolysis nomenclature. Journal of

analytical and applied pyrolysis 14, 219-221.

Espitalie, J., Madec, M. and Tissot, B. (1980) Role of mineral matrix in

kerogen pyrolysis: influence on petroleum generation and migration.

AAPG Bulletin 64, 59-66.

Espitalie, J., Madec, M., Tissot, B., Mennig, J. and Leplat, P. (1977) Source

rock characterization method for petroleum exploration, Offshore

Technology Conference. Offshore Technology Conference.

Espitalié, J., Makadi, K.S. and Trichet, J. (1984) Role of the mineral matrix

during kerogen pyrolysis. Organic Geochemistry 6, 365-382.

Espitalie, J., Ungerer, P., Irwin, I. and Marquis, F. (1988) Primary cracking

of kerogens. Experimenting and modelling C 1, C 2–C 5, C 6–C 15

and C 15+ classes of hydrocarbons formed. Organic Geochemistry

13, 893-899.

Ewbank, G., Manning, D.A.C. and Abbott, G.D. (1993) An organic

geochemical study of bitumens and their potential source rocks from

the South Pennine Orefield, Central England. Organic geochemistry

20, 579-598.

Farrimond, P., Bevan, J.C. and Bishop, A.N. (1999) Tricyclic terpane

maturity parameters: response to heating by an igneous intrusion.

Organic Geochemistry 30, 1011-1019.

Fischer, F. and Tropsch, H. (1930) Process for the production of paraffin-

hydrocarbons with more than one carbon atom. Google Patents.

Fisher, D.E. and Bostrom, K. (1969) Uranium rich sediments on the East

Pacific Rise. Nature 224, 64-65.

129

Forbes, P., Landais, P., Bertrand, P., Brosse, E., Espitalie, J. and Yahaya, M.

(1988) Chemical transformations of type-III organic matter

associated with the Akouta uranium deposit (Niger): geological

implications. Chemical geology 71, 267-282.

Foscolos, A. and Powell, T. (1979) Catagenesis in shales and occurrence of

authigenic clays in sandstones, North Sabine H-49 well, Canadian

Arctic Islands. Canadian Journal of Earth Sciences 16, 1309-1314.

Foscolos, A., Powell, T. and Gunther, P. (1976) The use of clay minerals and

inorganic and organic geochemical indicators for evaluating the

degree of diagenesis and oil generating potential of shales.

Geochimica et cosmochimica acta 40, 953-966.

Fowler, M. and Douglas, A. (1984) Distribution and structure of

hydrocarbons in four organic-rich Ordovician rocks. Organic

Geochemistry 6, 105-114.

Fraser, A.J. and Gawthorpe, R.L. (2003) An atlas of Carboniferous basin

evolution in Northern England. Geological Society, London, Memoirs

28, NP-NP.

Frolov, E., Melikhov, V. and Smirnov, M. (1996) Radiolytic nature of n-

alkene/n-alkane distributions in Russian Precambrian and Palaeozoic

oils. Organic geochemistry 24, 1061-1064.

Frolov, E. and Smirnov, M. (1994) Unsaturated hydrocarbons in crude oils.

Organic geochemistry 21, 189-208.

Frost, A. (1945) The role of clays in the formation of petroleum. Uspekhi.

Khim 14, 501.

Gasparik, M., Bertier, P., Gensterblum, Y., Ghanizadeh, A., Krooss, B.M. and

Littke, R. (2014) Geological controls on the methane storage

capacity in organic-rich shales. International Journal of Coal Geology

123, 34-51.

Gautschi, A. (2001) Hydrogeology of a fractured shale (Opalinus Clay):

Implications for deep geological disposal of radioactive wastes.

Hydrogeol J 9, 97-107.

Gerrienne, P., Servais, T. and Vecoli, M. (2016) Plant evolution and

terrestrialization during Palaeozoic times—the phylogenetic context.

Review of Palaeobotany and Palynology 227, 4-18.

Gibbins-Matham, J. and Kandiyoti, R. (1988) Coal pyrolysis yields from fast

and slow heating in a wire-mesh apparatus with a gas sweep.

Energy & fuels 2, 505-511.

Gold, T. (1985) The origin of natural gas and petroleum, and the prognosis

for future supplies. Annual Review of Energy 10, 53-77.

Gold, T. and Soter, S. (1982) Abiogenic methane and the origin of

petroleum. Energ Explor Exploit 1, 89-104.

Goldschmidt, V.M. (1937) The principles of distribution of chemical elements

in minerals and rocks. The seventh Hugo Müller Lecture, delivered

before the Chemical Society on March 17th, 1937. Journal of the

Chemical Society (Resumed), 655-673.

130

Goodwin, J., Golovko, V., Iacob, V. and Hardy, J.C. (2009) Half-life of the

electron-capture decay of Ru 97: Precision measurement shows no

temperature dependence. Phys Rev C 80, 045501.

Goossens, H.d., De Leeuw, J., Schenck, P. and Brassell, S. (1984)

Tocopherols as likely precursors of pristane in ancient sediments

and crude oils. Nature 312, 440-442.

Gradstein, F., Ogg, J. and Smith, A. A geologic time scale 2004. Cambridge,

Cambridge University Press.

Grantham, P. and Wakefield, L. (1988) Variations in the sterane carbon

number distributions of marine source rock derived crude oils

through geological time. Organic Geochemistry 12, 61-73.

Green, C.A., Styles, P. and Baptie, B.J. (2012) Preese Hall shale gas

fracturing review and recommendations for induced seismic

mitigation. London, UK: Department of Energy and Climate Change.

Grim, R.E. (1947) Relation of clay mineralogy to origin and recovery of

petroleum. AAPG Bulletin 31, 1491-1499.

Gross, D., Sachsenhofer, R., Bechtel, A., Pytlak, L., Rupprecht, B. and

Wegerer, E. (2015) Organic geochemistry of Mississippian shales

(Bowland Shale Formation) in central Britain: Implications for

depositional environment, source rock and gas shale potential.

Marine and Petroleum Geology 59, 1-21.

Groysman, A. (2014) History of Crude Oil and Petroleum Products,

Corrosion in Systems for Storage and Transportation of Petroleum

Products and Biofuels. Springer, pp. 221-226.

Gussow, W.C. (1954) Differential entrapment of oil and gas: a fundamental

principle. AAPG bulletin 38, 816-853.

Hanson, A.D., Ritts, B.D. and Moldowan, J.M. (2007) Organic Geochemistry

of oil and source rock strata of the Ordos Basin, north-central China.

AAPG bulletin 91, 1273-1293.

Hao, F., Zhou, X., Zhu, Y. and Yang, Y. (2011) Lacustrine source rock

deposition in response to co-evolution of environments and

organisms controlled by tectonic subsidence and climate, Bohai Bay

Basin, China. Organic Geochemistry 42, 323-339.

Hartwig, A., di Primio, R., Anka, Z. and Horsfield, B. (2012) Source rock

characteristics and compositional kinetic models of Cretaceous

organic rich black shales offshore southwestern Africa. Organic

Geochemistry 51, 17-34.

Hedberg, H.D. (1964) Geologic aspects of origin of petroleum. AAPG Bulletin

48, 1755-1803.

Heller-Kallai, L., Aizenshtat, Z. and Miloslavski, I. (1984) The effect of

various clay minerals on the thermal decomposition of stearic acid

under ‘bulk flow’conditions. Clay Miner 19, 779-788.

Hill, R.J., Zhang, E., Katz, B.J. and Tang, Y. (2007) Modeling of gas

generation from the Barnett Shale, Fort Worth Basin, Texas. AAPG

Bulletin 91, 501-521.

131

Hoering, T. (1984) Thermal reactions of kerogen with added water, heavy

water and pure organic substances. Organic Geochemistry 5, 267-

278.

Hoering, T.C. and Navale, V. (1987) A search for molecular fossils in the

kerogen of Precambrian sedimentary rocks. Precambrian Res 34,

247-267.

Hoffmann, C., Foster, C., Powell, T. and Summons, R. (1987) Hydrocarbon

biomarkers from Ordovician sediments and the fossil alga

Gloeocapsomorpha prisca Zalessky 1917. Geochimica et

Cosmochimica Acta 51, 2681-2697.

Högström, A.E.S. and Ebbestad, J.O.R. (2004) Preservational aspects of the

Upper Ordovician Fjäcka Shale, central Sweden–exemplified by a

medusoid problematicum. WOGOGOB-2004 8th Meeting on the

Working Group on the Ordovician Geology of Baltoscandia 1, 45-46.

Holba, A., Ellis, L., Dzou, I., Hallam, A., Masterson, W., Francu, J. and

Fincannon, A. (2001) Extended tricyclic terpanes as age

discriminators between Triassic, Early Jurassic and Middle-Late

Jurassic oils. 20th International Meeting on Organic Geochemistry 1,

464.

Holba, A., Masterson, W., Ellis, L. and Dzou, L. (2002) Triassic source facies

in high paleo-latitude petroleum systems, Pacific Section

AAPG/Western Region SPE Conference, Anchorage Alaska, p. 1146.

Horsfield, B. (1989) Practical criteria for classifying kerogens: some

observations from pyrolysis-gas chromatography. Geochimica et

Cosmochimica Acta 53, 891-901.

Horsfield, B. (1997) The bulk composition of first-formed petroleum in

source rocks. Petroleum and basin evolution.

Horsfield, B., Bharati, S. and Larter, S.R. (1992a) On the atypical

petroleum-generating characteristics of alginite in the Cambrian

Alum Shale, in: Schidlowski, M., Golubic, S., Kimberley, M.M., Sr,

D.M.M., Trudinger, P.A. (Eds.), Early Organic Evolution. Springer

Berlin Heidelberg, pp. 257-266.

Horsfield, B., Bharati, S. and Larter, S.R. (1992b) On the atypical

petroleum-generating characteristics of alginite in the Cambrian

Alum Shale. Early Organic Evolution.

Horsfield, B., Clegg, H., Wilkes, H. and Santamaria-Orozco, D. (1998) Effect

of maturity on carbazole distributions in petroleum systems: new

insights from the Sonda de Campeche, Mexico, and Hils Syncline,

Germany. The Science of Nature 85, 233-237.

Horsfield, B., Dembicki, H. and Ho, T. (1983) Some potential applications of

pyrolysis to basin studies. Journal of the Geological Society 140,

431-443.

Horsfield, B., Disko, U. and Leistner, F. (1989) The micro-scale simulation

of maturation: outline of a new technique and its potential

applications. Geologische Rundschau 78, 361-373.

132

Horsfield, B. and Douglas, A. (1980) The influence of minerals on the

pyrolysis of kerogens. Geochimica et Cosmochimica Acta 44, 1119-

1131.

Horsfield, B., Leistner, F. and Hall, K. (2014) Microscale sealed vessel

pyrolysis. Principles and Practice of Analytical Techniques in

Geosciences 4, 209.

Horsfield, B., Schenk, H.J., Mills, N. and Welte, D.H. (1992c) An

investigation of the in-reservoir conversion of oil to gas:

compositional and kinetic findings from closed-system programmed-

temperature pyrolysis. Organic Geochemistry 19, 191-204.

Hough, E. and Vane, C.H. (2014) The Bowland Shale in the Roosecote

Borehole of the Lancaster Fells Sub-Basin Craven Basin UK: A

Potential UK Shale gas Play? SPE/EAGE European …, 25-27.

Hu, M., Cheng, Z., Zhang, M., Liu, M., Song, L., Zhang, Y. and Li, J. (2014)

Effect of Calcite, Kaolinite, Gypsum, and Montmorillonite on Huadian

Oil Shale Kerogen Pyrolysis. Energy & Fuels 28, 1860-1867.

Huang, H., Zhang, S. and Su, J. (2015a) Geochemistry of Tri- and

Tetracyclic Terpanes in the Palaeozoic Oils from the Tarim Basin,

Northwest China. Energy & Fuels 29, 7014-7025.

Huang, H., Zhang, S. and Su, J. (2015b) Pyrolytically derived polycyclic

aromatic hydrocarbons in marine oils from the Tarim Basin, NW

China. Energy & Fuels 29, 5578-5586.

Huang, W.-Y. and Meinschein, W. (1979) Sterols as ecological indicators.

Geochimica et Cosmochimica Acta 43, 739-745.

Hughes, W.B. (1984) Use of thiophenic organosulfur compounds in

characterizing crude oils derived from carbonate versus siliciclastic

sources.

Hughey, C.A., Hendrickson, C.L., Rodgers, R.P., Marshall, A.G. and Qian, K.

(2001) Kendrick mass defect spectrum: a compact visual analysis

for ultrahigh-resolution broadband mass spectra. Analytical

Chemistry 73, 4676-4681.

Hughey, C.A., Rodgers, R.P., Marshall, A.G., Qian, K. and Robbins, W.K.

(2002) Identification of acidic NSO compounds in crude oils of

different geochemical origins by negative ion electrospray Fourier

transform ion cyclotron resonance mass spectrometry. Organic

Geochemistry 33, 743-759.

Huizinga, B.J., Tannenbaum, E. and Kaplan, I. (1987) The role of minerals

in the thermal alteration of organic matter—III. Generation of

bitumen in laboratory experiments. Organic Geochemistry 11, 591-

604.

Hunt, J.M. (1961) Distribution of hydrocarbons in sedimentary rocks.

Geochimica et cosmochimica Acta 22, 37-49.

Hunt, J.M. (1996) Petroleum geochemistry and geology. WH Freeman New

York.

133

Hunt, J.M., Lewan, M. and Hennet, R.J. (1991) Modeling Oil Generation with

Time-Temperature Index Graphs Based on the Arrhenius Equation

(1). AAPG Bulletin 75, 795-807.

Hunt, T.S. (1863) Report on the geology of Canada. Canadian Geological

Survey Report: Progress to 1863. Canadian Geological Survey.

Imber, J., Armstrong, H. and Clancy, S. (2014) Natural fractures in a UK

shale reservoir analogue, Cleveland Basin, NE England. AAPG

Bulletin.

Jaraula, C.M., Schwark, L., Moreau, X., Pickel, W., Bagas, L. and Grice, K.

(2015) Radiolytic alteration of biopolymers in the Mulga Rock

(Australia) uranium deposit. Applied Geochemistry 52, 97-108.

Jarvie, D.M., Behar, F. and Mazéas, L. (2010) Decomposition of Organic

Matter and Impact on Shale Resource Play Assessments. 11-14.

Jarvie, D.M., Hill, R.J., Ruble, T.E. and Pollastro, R.M. (2007)

Unconventional shale-gas systems: The Mississippian Barnett Shale

of north-central Texas as one model for thermogenic shale-gas

assessment. AAPG Bulletin 91, 475-499.

Jarvis, G.T. and McKenzie, D.P. (1980) Sedimentary basin formation with

finite extension rates. Earth and Planetary Science Letters 48, 42-52.

Johns, W.D. and Shimoyama, A. (1972) Clay minerals and petroleum-

forming reactions during burial and diagenesis. AAPG Bulletin 56,

2160-2167.

Jones, R. (1984) Comparison of carbonate and shale source rocks.

Jurg, J. and Eisma, E. (1964) Petroleum hydrocarbons: generation from

fatty acid. Science 144, 1451-1452.

Kanev, S., Margulis, L., Bojesen-Koefoed, J., Weil, W., Merta, H. and

Zdanavičiūtė, O. (1994) Oils and hydrocarbon source rocks of the

Baltic syneclise. Oil and Gas Journal 92, 1-4.

Karabakan, A. and Yürüm, Y. (1998) Effect of the mineral matrix in the

reactions of oil shales: 1. Pyrolysis reactions of Turkish Göynük and

US Green River oil shales. Fuel 77, 1303-1309.

Karnkowski, P.H., Pikulski, L. and Wolnowski, T. (2010) Petroleum geology

of the Polish part of the Baltic region-an overview. Geological

Quarterly 54, 143-158.

Katz, B.J. (1983) Limitations of ‘Rock-Eval’pyrolysis for typing organic

matter. Organic Geochemistry 4, 195-199.

Katz, B.J. and Everett, M.A. (2016) An overview of pre-Devonian petroleum

systems–Unique characteristics and elevated risks. Marine and

Petroleum Geology 73, 492-516.

Kelly, A.E., Love, G.D., Zumberge, J.E. and Summons, R.E. (2011)

Hydrocarbon biomarkers of Neoproterozoic to Lower Cambrian oils

from eastern Siberia. Organic Geochemistry 42, 640-654.

Kenney, J., Shnyukov, Y.F., Krayishkin, V., Tchebanenko, I. and Klochko, V.

(2001) Dismissal of claims of a biological connection for natural

petroleum. Energia 22, 26-34.

134

Kenrick, P. and Crane, P.R. (1997) The origin and early evolution of plants

on land. Nature 389, 33-39.

Killops, S.D. and Killops, V.J. (2013) Introduction to organic geochemistry.

John Wiley & Sons.

Kim, S., Stanford, L.A., Rodgers, R.P., Marshall, A.G., Walters, C.C., Qian,

K., Wenger, L.M. and Mankiewicz, P. (2005) Microbial alteration of

the acidic and neutral polar NSO compounds revealed by Fourier

transform ion cyclotron resonance mass spectrometry. Organic

Geochemistry 36, 1117-1134.

Kissin, Y. (1987) Catagenesis and composition of petroleum: origin of n-

alkanes and isoalkanes in petroleum crudes. Geochimica et

Cosmochimica Acta 51, 2445-2457.

Klein, G.C., Kim, S., Rodgers, R.P., Marshall, A.G. and Yen, A. (2006) Mass

spectral analysis of asphaltenes. II. Detailed compositional

comparison of asphaltenes deposit to its crude oil counterpart for

two geographically different crude oils by ESI FT-ICR MS. Energy &

fuels 20, 1973-1979.

Koczy, F.F., Tomic, E. and Hecht, F. (1957) Zur geochemie des urans im

ostseebecken. Geochimica et Cosmochimica Acta 11, 86-102.

Kolesnikov, A., Kutcherov, V.G. and Goncharov, A.F. (2009) Methane-

derived hydrocarbons produced under upper-mantle conditions.

Nature Geoscience 2, 566-570.

Konitzer, S.F., Davies, S.J., Stephenson, M.H. and Leng, M.J. (2014)

Depositional Controls On Mudstone Lithofacies In A Basinal Setting:

Implications for the Delivery of Sedimentary Organic Matter. Journal

of Sedimentary Research 84, 198-214.

Kosakowski, P., Kotarba, M.J., Piestrzyński, A., Shogenova, A. and Więcław,

D. (2016) Petroleum source rock evaluation of the Alum and

Dictyonema Shales (Upper Cambrian–Lower Ordovician) in the Baltic

Basin and Podlasie Depression (eastern Poland). International

Journal of Earth Sciences 105, 1-19.

Kosakowski, P., Wróbel, M. and Poprawa, P. (2010) Hydrocarbon generation

and expulsion modelling of the lower Paleozoic source rocks in the

Polish part of the Baltic region. Geological Quarterly 54, 241-256.

Kotarba, M. and Lewan, M. (2013) Sources of natural gases in Middle

Cambrian reservoirs in Polish and Lithuanian Baltic Basin as

determined by stable isotopes and hydrous pyrolysis of Lower

Palaeozoic source rocks. Chemical Geology 345, 62-76.

Kotarba, M., Lewan, M. and Więcław, D. (2014a) Shale Gas and Oil Potential

of Lower Palaeozoic Strata in the Polish Baltic Basin by Hydrous

Pyrolysis. Conference: Fourth EAGE Shale Workshop.

Kotarba, M., Więcław, D., Dziadzio, P., Kowalski, A., Kosakowski, P. and

Bilkiewicz, E. (2014b) Organic geochemical study of source rocks

and natural gas and their genetic correlation in the eastern part of

the Polish Outer Carpathians and Palaeozoic–Mesozoic basement.

Marine and Petroleum Geology 56, 97-122.

135

Kribek, B., Zak, K., Spangenberg, J.E., Jehlicka, J., Prokes, S. and Kominek,

J. (1999) Bitumens in the late Variscan hydrothermal vein-type

uranium deposit of Pribram, Czech Republic; sources, radiation-

induced alteration, and relation to mineralization. Economic Geology

94, 1093-1114.

Kudryavtsev, N. (1951) Against the organic hypothesis of the origin of

petroleum. Neftianoye Khozyaistvo 9, 17-29.

Kuhn, P.P., di Primio, R., Hill, R., Lawrence, J.R. and Horsfield, B. (2012)

Three-dimensional modeling study of the low-permeability

petroleum system of the Bakken Formation. AAPG Bulletin 96, 1867-

1897.

Kvenvolden, K.A. (1966) Evidence for transformations of normal fatty acids

in sediments. Advances in organic geochemistry 1968, 355-366.

Lakshmanan, C.C., Bennett, M.L. and White, N. (1991) Implications of

multiplicity in kinetic parameters to petroleum exploration:

distributed activation energy models. Energy & Fuels 5, 110-117.

Landais, P. (1993) Bitumens in uranium deposits, Bitumens in ore deposits.

Springer, pp. 213-238.

Landais, P. (1996) Organic geochemistry of sedimentary uranium ore

deposits. Ore Geol Rev 11, 33-51.

Lange, G. and Freyhoff, G. (1991) GEOLOGIE UND BERGBAU IN DER

URANLAGERSTATTE RONNEBURG/THURINGEN. Erzmetall 44, 264-

269.

Larsen, J.W. and Hu, J. (2006) Kerogen chemistry. 8. Hydrous pyrolysis of

rundle kerogen: Source of the oxygen in CO2 and mineral catalysis.

Energy & fuels 20, 278-280.

Larter, S. (1984) Application of analytical pyrolysis techniques to kerogen

characterization and fossil fuel exploration/exploitation, Analytical

pyrolysis. Butterworths London, pp. 212-272.

Larter, S., Huang, H., Adams, J., Bennett, B. and Snowdon, L.R. (2012) A

practical biodegradation scale for use in reservoir geochemical

studies of biodegraded oils. Organic Geochemistry 45, 66-76.

Larter, S.R., Bowler, B.F.J., Li, M. and Chen, M. (1996) Molecular indicators

of secondary oil migration distances. Nature 383, 593-597.

Law, B.E. (1988) Coal-bed methane. Petroleum Systems of the United

States, 52.

Lawrence, S., Coster, P. and Ireland, R. (1987) Structural development and

petroleum potential of the northern flanks of the Bowland Basin

(Carboniferous) Northwest England. Petroleum geology of north-

west Europe 1, 225-233.

Lazauskiene, J., Sliaupa, S., Brazauskas, A. and Musteikis, P. (2003)

Sequence stratigraphy of the Baltic Silurian succession: tectonic

control on the foreland infill. Geological Society, London, Special

Publications 208, 95-115.

136

Leckie, D.A., Singh, C., Goodarzi, F. and Wall, J.H. (1990) Organic-rich,

radioactive marine shale: a case study of a shallow-water

condensed section, Cretaceous Shaftesbury Formation, Alberta,

Canada. Journal of Sedimentary Research 60.

Lecomte, A., Cathelineau, M., Michels, R., Peiffert, C. and Brouand, M.

(2017) Uranium mineralization in the Alum Shale Formation

(Sweden): Evolution of a U-rich marine black shale from

sedimentation to metamorphism. Ore Geol Rev 88, 71-98.

Leeder, M. (1988) Recent developments in Carboniferous geology: a critical

review with implications for the British Isles and NW Europe.

Proceedings of the Geologists' Association 99, 73-100.

Leventhal, J. (1991) Comparison of organic geochemistry and metal

enrichment in two black shales: Cambrian Alum Shale of Sweden

and Devonian Chattanooga Shale of United States. Mineralium

Deposita 26, 104-112.

Leventhal, J.S. (1981) Pyrolysis gas chromatography-mass spectrometry to

characterize organic matter and its relationship to uranium content

of Appalachian Devonian black shales. Geochimica et Cosmochimica

Acta 45, 883-889.

Leventhal, J.S., Daws, T.A. and Frye, J.S. (1986) Organic geochemical

analysis of sedimentary organic matter associated with uranium.

Applied geochemistry 1, 241-247.

Leventhal, J.S. and Threlkeld, C.N. (1978) Carbon-13/carbon-12 isotope

fractionation of organic matter associated with uranium ores induced

by alpha irradiation. Science 202, 430-432.

Lewan, B.M.D. and Henry, A.A. (1999) Gas : Oil Ratios for Source Rocks

Containing Type-I , -II , -IIS , and -III Kerogens as Determined by

Hydrous Pyrolysis. USGS report.

Lewan, M. and Buchardt, B. (1989) Irradiation of organic matter by uranium

decay in the Alum Shale, Sweden. Geochimica et Cosmochimica

Acta 53, 1307-1322.

Lewan, M., Kotarba, M., Więcław, D. and Piestrzyński, A. (2008) Evaluating

transition-metal catalysis in gas generation from the Permian

Kupferschiefer by hydrous pyrolysis. Geochimica et Cosmochimica

Acta 72, 4069-4093.

Lewan, M., Spiro, B., Illich, H., Raiswell, R., Mackenzie, A., Durand, B.,

Manning, D., Comet, P., Berner, R. and De Leeuw, J. (1985)

Evaluation of petroleum generation by hydrous pyrolysis

experimentation [and discussion]. Philosophical Transactions of the

Royal Society of London A: Mathematical, Physical and Engineering

Sciences 315, 123-134.

Lewan, M., Winters, J. and McDonald, J. (1979) Generation of oil-like

pyrolyzates from organic-rich shales. Science 203, 897-899.

Lewan, M.D. (1997) Experiments on the role of water in petroleum

formation. Geochimica et Cosmochimica Acta 61, 3691-3723.

137

Lewan, M.D., Dolan, M.P. and Curtis, J.B. (2014) Effects of smectite on the

oil-expulsion efficiency of the Kreyenhagen Shale, San Joaquin Basin,

California, based on hydrous-pyrolysis experiments. AAPG Bulletin

98, 1091-1109.

Lewan, M.D. and Ruble, T.E. (2002) Comparison of petroleum generation

kinetics by isothermal hydrous and nonisothermal open-system

pyrolysis. Organic Geochemistry 33, 1457-1475.

Li, M., Larter, S., Stoddart, D. and Bjorøy, M. (1995) Fractionation of

pyrrolic nitrogen compounds in petroleum during migration:

derivation of migration-related geochemical parameters. Geological

Society, London, Special Publications 86, 103-123.

Liao, Y., Shi, Q., Hsu, C.S., Pan, Y. and Zhang, Y. (2012) Distribution of

acids and nitrogen-containing compounds in biodegraded oils of the

Liaohe Basin by negative ion ESI FT-ICR MS. Organic Geochemistry

47, 51-65.

Lind, S. and Bardwell, D. (1926) The chemical action of gaseous ions

produced by alpha particles. IX. Saturated hydrocarbons. J Am

Chem Soc 48, 2335-2351.

Lindgreen, H. and Drits, V.A. (2000) Illite-smectite structural changes

during metamorphism in black Cambrian Alum shales from the Baltic

area. American Mineralogist 85, 1223-1238.

Liu, P., Li, M., Jiang, Q., Cao, T. and Sun, Y. (2015) Effect of secondary oil

migration distance on composition of acidic NSO compounds in

crude oils determined by negative-ion electrospray Fourier

transform ion cyclotron resonance mass spectrometry. Organic

Geochemistry 78, 23-31.

Loog, A., Kurvits, T., Aruvali, J. and Petersell, V. (2001) Grain size analysis

and mineralogy of the Tremadocian Dictyonema shale in Estonia. Oil

Shale 18, 281-297.

Loucks, R.G. and Ruppel, S.C. (2007) Mississippian Barnett Shale:

Lithofacies and depositional setting of a deep-water shale-gas

succession in the Fort Worth Basin, Texas. AAPG Bulletin 91, 579-

601.

Love, G.D., Grosjean, E., Stalvies, C., Fike, D.A., Grotzinger, J.P., Bradley,

A.S., Kelly, A.E., Bhatia, M., Meredith, W., Snape, C.E., Bowring,

S.A., Condon, D.J. and Summons, R.E. (2009) Fossil steroids record

the appearance of Demospongiae during the Cryogenian period.

Nature 457, 718-721.

Loydell, D., Butcher, A., Frýda, J., Lüning, S. and Fowler, M. (2009) Lower

Silurian “hot shales” in Jordan: a new depositional model. Journal of

Petroleum Geology 32, 261-270.

Lu, S.-T. and Kaplan, I.R. (1989) Pyrolysis of kerogens in the absence and

presence of montmorillonite—II. Aromatic hydrocarbons generated

at 200 and 300° C. Organic geochemistry 14, 501-510.

Lu, S.-T., Ruth, E. and Kaplan, I.R. (1989) Pyrolysis of kerogens in the

absence and presence of montmorillonite—I. The generation,

138

degradation and isomerization of steranes and triterpanes at 200

and 300° C. Organic Geochemistry 14, 491-499.

Lüning, S., Shahin, Y.M., Loydell, D., Al-Rabi, H.T., Masri, A., Tarawneh, B.

and Kolonic, S. (2005) Anatomy of a world-class source rock:

Distribution and depositional model of Silurian organic-rich shales in

Jordan and implications for hydrocarbon potential. AAPG Bulletin 89,

1397-1427.

Lutz, M., Kaasschieter, P. and van Wijhe, H. (1975) PD 2 (3) geological

factors controlling Rotliegend gas accumulations in the Mid-

European Basin, 9th World Petroleum Congress. World Petroleum

Congress.

Mackenzie, A., Hoffmann, C. and Maxwell, J. (1981) Molecular parameters

of maturation in the Toarcian shales, Paris Basin, France—III.

Changes in aromatic steroid hydrocarbons. Geochimica et

Cosmochimica Acta 45, 1345-1355.

Mackenzie, A., Leythaeuser, D., Muller, P., Quigley, T. and Radke, M.

(1988) The movement of hydrocarbons in shales. Nature.

Mahlstedt, N., di Primio, R. and Horsfield, B. (2013) GOR-Fit-From Liquids

to Late Gas: Deconvoluting Primary from Secondary Gas Generation

Kinetics. Book of Abstracts of the Communications, 26th

International Meeting on Organic Geochemistry - IMOG (Costa Adeje,

Tenerife Spain 2013), pp. 193-194.

Mahlstedt, N., Horsfield, B., Wilkes, H. and Poetz, S. (2016) Tracing the

Impact of Fluid Retention on Bulk Petroleum Properties Using

Nitrogen-Containing Compounds. Energy & Fuels 30, 6290-6305.

Mahlstedt, N.L. (2012) Evaluating the late gas potential of source rocks

stemming from different sedimentary environments genehmigte

Dissertation, pp. 370pp-370pp.

Makadi, K.S. (1983) Étude expérimentale des interactions entre matière

organique-matrice minérale au cours de la pyrolyse. IFP.

Mango, F.D. (1987) An invariance in the isoheptanes of petroleum. Science

237, 514-517.

Mango, F.D. (1990) The origin of light hydrocarbons in petroleum: a kinetic

test of the steady-state catalytic hypothesis. Geochimica et

Cosmochimica Acta 54, 1315-1323.

Mango, F.D. (1991) The stability of hydrocarbons under the time–

temperature conditions of petroleum genesis. Nature.

Mango, F.D. (1992) Transition metal catalysis in the generation of

petroleum: A genetic anomaly in Ordovician oils. Geochimica et

Cosmochimica Acta 56, 3851-3854.

Mango, F.D. (1994) The origin of light hydrocarbons in petroleum: Ring

preference in the closure of carbocyclic rings. Geochimica et

Cosmochimica Acta 58, 895-901.

Mango, F.D. (1997) Review paper: The light hydrocarbons in petroleum: a

critical review. Organic Geochemistry 26, 417-440.

139

Mango, F.D. (2007) In situ conversion of heavy hydrocarbons to catalytic

gas. US Patent App. 11/856,566 2.

Mango, F.D. (2010) “Distinguishing gases derived from oil cracking and

kerogen maturation: Insights from laboratory pyrolysis experiments.”

Guo Liguo, Xiao Xianming, Tian Hui, Song Zhiguang, 2009, Organic

Geochemistry 40, 1074–1084. Organic Geochemistry 41, 719-720.

Mango, F.D. and Hightower, J. (1997) The catalytic decomposition of

petroleum into natural gas. Geochimica et Cosmochimica Acta 61,

5347-5350.

Marshall, A.G., Hendrickson, C.L. and Jackson, G.S. (1998) Fourier

transform ion cyclotron resonance mass spectrometry: a primer.

Mass spectrometry reviews 17, 1-35.

Marshall, A.G. and Rodgers, R.P. (2004) Petroleomics: The next grand

challenge for chemical analysis. Accounts of Chemical Research 37,

53-59.

Mayer, L.M. (1994) Surface area control of organic carbon accumulation in

continental shelf sediments. Geochimica et Cosmochimica Acta 58,

1271-1284.

McCain, W.D. (1990) The properties of petroleum fluids. PennWell Books.

McKenzie, D., Mackenzie, A.S., Maxwell, J. and Sajgo, C. (1983)

Isomerization and aromatization of hydrocarbons in stretched

sedimentary basins.

McKirdy, D., Aldridge, A. and Ypma, P. (1981) A geochemical comparison of

some crude oils from pre-Ordovician carbonate rocks, in: Bjorøy, M.

(Ed.), Advances in organic geochemistry. John Wiley & Sons Ltd,

University of Bergen, Norway, pp. 99-107.

Meinschein, W. (1961) Significance of hydrocarbons in sediments and

petroleum. Geochimica et Cosmochimica Acta 22, 58-64.

Mendeleev, D. (1877) L’origine du petrole. Revue Scientifique 8, 409-416.

Michels, R., Landais, P., Torkelson, B.E. and Philp, R.P. (1995) Effects of

effluents and water pressure on oil generation during confined

pyrolysis and high-pressure hydrous pyrolysis. Geochimica et

Cosmochimica Acta 59, 1589-1604.

Modliński, Z. and Podhalańska, T. (2010) Outline of the lithology and

depositional features of the lower Paleozoic strata in the Polish part

of the Baltic region. Geological Quarterly 54, 109-121.

Modliński, Z. and Szymański, B. (1997) The Ordovician lithostratigraphy of

the Peribaitic Depression (NE Poland). Geological Quarterly 41, 273-

288.

Moldowan, J.M., Dahl, J., Huizinga, B.J., Fago, F.J., Hickey, L.J., Peakman,

T.M. and Taylor, D.W. (1994) The molecular fossil record of

oleanane and its relation to angiosperms. Science, 768-771.

Moldowan, J.M., Fago, F.J., Lee, C.Y., Jacobson, S.R., Watt, D.S., Slougui,

N.-E., Jeganathan, A. and Young, D.C. (1990) Sedimentary 24-n-

140

propylcholestanes, molecular fossils diagnostic of marine algae.

Science 247, 309-312.

Moldowan, J.M., Seifert, W.K. and Gallegos, E.J. (1985) Relationship

between petroleum composition and depositional environment of

petroleum source rocks. AAPG bulletin 69, 1255-1268.

Møller, L.N. and Friis, H. (1999) Petrographic evidence for hydrocarbon

migration in Lower Cambrian sandstones, Bornholm, Denmark.

Bulletin of the Geological Society of Denmark 45, 117-127.

Momper, J.A. (1979) Domestic Oil Reserves Forecasting Method and

Assessment of Regional Potentials: ABSTRACT. AAPG Bulletin 63,

497-498.

Monthioux, M., Landais, P. and Monin, J.-C. (1985) Comparison between

natural and artificial maturation series of humic coals from the

Mahakam delta, Indonesia. Organic Geochemistry 8, 275-292.

Mossman, D.J., Nagy, B. and Davis, D.W. (1993) Hydrothermal alteration of

organic matter in uranium ores, Elliot Lake, Canada: implications for

selected organic-rich deposits. Geochimica et cosmochimica acta 57,

3251-3259.

Motuza, G., Kepezinskas, P. and Šliaupa, S. (1994) Diabases from the well

D-1 in the Baltic Sea. Geologija 16, 16-20.

Motuzaa, G., Šliaupab, S. and Timmermanc, M.J. (2015) Geochemistry and

40Ar/39Ar age of Early Carboniferous dolerite sills in the southern

Baltic Sea. Estonian Journal of Earth Sciences 64, 3.

Mullen, J. (2010) Petrophysical characterization of the Eagle Ford Shale in

south Texas, Canadian Unconventional Resources and International

Petroleum Conference. Society of Petroleum Engineers.

Muntendam-Bos, A.G. (2009) Inventory non-conventional gas. TNO Built

Environment and Geosciences.

Nielsen, A.T. and Schovsbo, N.H. (2006) Cambrian to basal Ordovician

lithostratigraphy in southern Scandinavia. Bulletin of the Geological

Society of Denmark 53, 47-92.

Oakwood, T., Shriver, D., Fall, H., McAleer, W. and Wunz, P. (1952) Optical

activity of petroleum. Industrial & Engineering Chemistry 44, 2568-

2570.

Obst, K. (2000) Permo-Carboniferous dyke magmatism on the Danish island

Bornholm. Neues Jahrbuch Fur Geologie Und Palaontologie-

Abhandlungen 218, 243-266.

Oldenburg, T.B.P., Brown, M., Bennett, B. and Larter, S.R. (2014) The

impact of thermal maturity level on the composition of crude oils,

assessed using ultra-high resolution mass spectrometry. Organic

Geochemistry 75, 151-168.

Oliver, N., Pearson, P., Holcombe, R. and Ord, A. (1999) Mary Kathleen

metamorphic–hydrothermal uranium–rare‐earth element deposit:

ore genesis and numerical model of coupled deformation and fluid

flow. Australian Journal of Earth Sciences 46, 467-484.

141

Osborne, M.J. and Swarbrick, R.E. (1997) Mechanisms for generating

overpressure in sedimentary basins: a reevaluation. AAPG bulletin

81, 1023-1041.

Osmond, J. and Cowart, J. (1976) The theory and uses of natural uranium

isotopic variations in hydrology. Atomic Energy Review 14, 621-679.

Öztaş, N. and Yürüm, Y. (2000) Pyrolysis of Turkish Zonguldak bituminous

coal. Part 1. Effect of mineral matter. Fuel 79, 1221-1227.

Pakarinen, J.M., Teräväinen, M.J., Pirskanen, A., Wickström, K. and

Vainiotalo, P. (2007) A positive-ion electrospray ionization Fourier

transform ion cyclotron resonance mass spectrometry study of

Russian and North Sea crude oils and their six distillation fractions.

Energy & Fuels 21, 3369-3374.

Palacas, J.G., Anders, D.E. and King, J.D. (1984) South Florida basin--a

prime example of carbonate source rocks of petroleum, in: Palacas,

J.G. (Ed.), Petroleum Geochemistry and Source Rock Potential of

Carbonate Rocks. American Association of Petroleum Geologists,

Tulsa, Oklahoma, pp. 71-96.

Pan, C., Geng, A., Zhong, N. and Liu, J. (2010) Kerogen pyrolysis in the

presence and absence of water and minerals: Steranes and

triterpenoids. Fuel 89, 336-345.

Pan, C., Geng, A., Zhong, N., Liu, J. and Yu, L. (2007) Kerogen pyrolysis in

the presence and absence of water and minerals. 1. Gas

components. Energy & Fuels, 416-427.

Pan, Y., Liao, Y., Shi, Q. and Hsu, C.S. (2013) Acidic and Neutral Polar NSO

Compounds in Heavily Biodegraded Oils Characterized by Negative-

Ion ESI FT-ICR MS. Energy & Fuels 27, 2960-2973.

Pater, C.J.D. and Baisch, S. (2011) Geomechanical study of Bowland Shale

seismicity. Synthesis Report.

Patterson, J., Ramsden, A., Dale, L. and Fardy, J. (1986) Geochemistry and

mineralogical residences of trace elements in oil shales from Julia

Creek, Queensland, Australia. Chemical geology 55, 1-16.

Pedersen, J.H., Karlsen, D.A., Spjeldnæs, N., Backer-Owe, K., Lie, J.E. and

Brunstad, H. (2007) Lower Paleozoic petroleum from southern

Scandinavia: Implications to a Paleozoic petroleum system offshore

southern Norway. AAPG Bulletin 91, 1189-1212.

Pelet, R. (1994) Comments on the paper “The effects of the mineral matrix

on the determination of kinetic parameters using modified Rock-Eval

pyrolysis” by H. Dembicki Jr, Org. Organic geochemistry 21, 979-

981.

Pepper, A.S. and Corvi, P.J. (1995) Simple kinetic models of petroleum

formation.Part I:oil and gas generation from kerogen. Marine and

Petroleum Geology 12, 291-319.

Pepper, A.S. and Dodd, T.a. (1995) Simple kinetic models of petroleum

formation. Part II: oil-gas cracking. Marine and Petroleum Geology

12, 321-340.

142

Perry Jr, E.A. and Hower, J. (1972) Late-stage dehydration in deeply buried

pelitic sediments. AAPG Bulletin 56, 2013-2021.

Peters, K. (1986) Guidelines for evaluating petroleum source rock using

programmed pyrolysis. AAPG Bulletin 70, 318-329.

Peters, K. and Moldowan, J. (1991) Effects of source, thermal maturity, and

biodegradation on the distribution and isomerization of

homohopanes in petroleum. Organic geochemistry 17, 47-61.

Peters, K.E. and Cassa, M.R. (1994) Applied source rock geochemistry.

AAPG Memoir.

Peters, K.E., Coutrot, D., Nouvelle, X., Ramos, L.S., Rohrback, B.G.,

Magoon, L.B. and Zumberge, J.E. (2013) Chemometric

differentiation of crude oil families in the San Joaquin Basin,

California. AAPG bulletin 97, 103-143.

Peters, K.E., Fraser, T.H., Amris, W., Rustanto, B. and Hermanto, E. (1999)

Geochemistry of crude oils from eastern Indonesia. AAPG Bulletin 83,

1927-1942.

Peters, K.E., Walters, C.C. and Mankiewicz, P.J. (2006) Evaluation of kinetic

uncertainty in numerical models of petroleum generation. AAPG

Bulletin 90, 387-403.

Peters, K.E., Walters, C.C. and Moldowan, J.M. (2005) The biomarker guide:

biomarkers and isotopes in the environment and human history.

Cambridge University Press, New York.

Peters, K.E., Wright, T.L., Ramos, L.S., Zumberge, J.E. and Magoon, L.B.

(2016) Chemometric recognition of genetically distinct oil families in

the Los Angeles basin, California. AAPG Bulletin 100, 115-135.

Petersen, H.I., Schovsbo, N.H. and Nielsen, A.T. (2013) Reflectance

measurements of zooclasts and solid bitumen in Lower Paleozoic

shales, southern Scandinavia: Correlation to vitrinite reflectance.

International Journal of Coal Geology 114, 1-18.

Philippi, G. (1965) On the depth, time and mechanism of petroleum

generation. Geochimica et Cosmochimica Acta 29, 1021-1049.

Philp, R.t. and Gilbert, T. (1986) Biomarker distributions in Australian oils

predominantly derived from terrigenous source material. Organic

Geochemistry 10, 73-84.

Pierce, A.P., Mytton, J.W. and Barnett, P.R. (1958) Geochemistry of

uranium in organic substances in petroliferous rocks. Geological

Survey, Washington, DC.

Poetz, S., Horsfield, B. and Wilkes, H. (2014) Maturity-Driven Generation

and Transformation of Acidic Compounds in the Organic-Rich

Posidonia Shale as Revealed by Electrospray Ionization Fourier

Transform Ion Cyclotron Resonance Mass Spectrometry. Energy &

Fuels 28, 4877-4888.

Pollastro, R.M. (2007) Total petroleum system assessment of undiscovered

resources in the giant Barnett Shale continuous (unconventional)

gas accumulation, Fort Worth Basin, Texas. AAPG Bulletin 91, 551-

578.

143

Pompeckj, J. (1901) Der Jura zwischen Regensburg und Regenstauf:

Geognostische Jahreshefte, v. 14.

Pool, W., Geluk, M., Abels, J. and Tiley, G. (2012) Assessment of an

unusual European shale gas play: the Cambro-Ordovician alum

shale, southern Sweden. SPE/EAGE European ….

Poprawa, P., Šliaupa, S., Stephenson, R. and Lazauskien, J. (1999) Late

Vendian–Early Palaeozoic tectonic evolution of the Baltic Basin:

regional tectonic implications from subsidence analysis.

Tectonophysics 314, 219-239.

Powers, M.C. (1967) Fluid-release mechanisms in compacting marine

mudrocks and their importance in oil exploration. AAPG bulletin 51,

1240-1254.

Pratt, W.E. (1947) Petroleum on continental shelves. AAPG Bulletin 31, 657-

672.

Price, L.C. and Schoell, M. (1995) Constraints on the origins of hydrocarbon

gas from compositions of gases at their site of origin. Nature 378,

368.

Priem, H., Mulder, F., Boelrijk, N., Hebeda, E., Verschure, R. and

Verdurmen, E.T. (1968) Geochronological and palaeomagnetic

reconnaissance survey in parts of central and southern Sweden.

Physics of the Earth and Planetary Interiors 1, 373-380.

Quigley, T., Mackenzie, A. and Gray, J. (1987) Kinetic theory of petroleum

generation. Collection colloques et séminaires-Institut français du

pétrole, 649-665.

Radke, M., Willsch, H. and Welte, D.H. (1980) Preparative hydrocarbon

group type determination by automated medium pressure liquid

chromatography. Analytical Chemistry 52, 406-411.

Raji, M., Cornford, C., Pather, A. and Satterfield, D. (2013) Unconventional

oil and gas potential of the Widmerpool Gulf in the East Midlands

Province of England. Unconventional Resources Technology

Conference (URTEC).

Reynolds, J.G. and Burnham, A.K. (1995) Comparison of kinetic analysis of

source rocks and kerogen concentrates. Organic Geochemistry.

Reynolds, J.G., Burnham, A.K. and Mitchell, T.O. (1995) Kinetic analysis of

California petroleum source rocks by programmed temperature

micropyrolysis. Organic geochemistry 23, 109-120.

Rice, D.D. and Claypool, G.E. (1981) Generation, accumulation, and

resource potential of biogenic gas. AAPG Bulletin 65, 5-25.

Rice, F. (1931) The thermal decomposition of organic compounds from the

standpoint of free radicals. I. Saturated hydrocarbons. J Am Chem

Soc 53, 1959-1972.

Robl, T.L. and Davis, B.H. (1993) Comparison of the HF-HCl and HF-BF 3

maceration techniques and the chemistry of resultant organic

concentrates. Organic Geochemistry 20, 249-255.

144

Romer, R.L. and Hahne, K. (2010) Life of the Rheic Ocean: scrolling through

the shale record. Gondwana Res 17, 236-253.

Rubinstein, I., Sieskind, O. and Albrecht, P. (1975) Rearranged sterenes in

a shale: occurrence and simulated formation. Journal of the

Chemical Society, Perkin Transactions 1 1, 1833-1836.

Rudakov, G. (1967) Recent developments in the theory of the non-biogenic

origin of petroleum. Chemical Geology 2, 179-185.

Salmon, V., Derenne, S., Lallier-Verges, E., Largeau, C. and Beaudoin, B.

(2000) Protection of organic matter by mineral matrix in a

Cenomanian black shale. Organic Geochemistry 31, 463-474.

Sandström, B., Tullborg, E.L., Torres, T.D. and Ortiz, J.E. (2006) The

occurrence and potential origin of asphaltite in bedrock fractures,

Forsmark, central Sweden. GFF 128, 233-242.

Sanei, H., Petersen, H.I., Schovsbo, N.H., Jiang, C. and Goodsite, M.E.

(2014) Petrographic and geochemical composition of kerogen in the

Furongian (U. Cambrian) Alum Shale, central Sweden: Reflections

on the petroleum generation potential. International Journal of Coal

Geology 132, 158-169.

Saxby, J. (1970) Isolation of kerogen in sediments by chemical methods.

Chemical Geology 6, 173-184.

Saxby, J. and Riley, K. (1984) Petroleum generation by laboratory-scale

pyrolysis over six years simulating conditions in a subsiding basin.

Nature.

Schenk, H.J. and Horsfield, B. (1993) Kinetics of petroleum generation by

programmed-temperature closed-versus open-system pyrolysis.

Geochimica et Cosmochimica Acta 57, 623-630.

Schenk, H.J., Horsfield, B. and Krooss, B. (1997) Kinetics of petroleum

formation and cracking. Petroleum and basin evolution.

Schleicher, M., Koster, J., Kulke, H. and Weil, W. (1998) Reservoir and

source-rock characterisation of the early Palaeozoic interval in the

peribaltic syneclise, northern Poland. Journal of Petroleum Geology

21, 33-56.

Schmoker, J.W. (1981) Determination of organic-matter content of

Appalachian Devonian shales from gamma-ray logs. AAPG Bulletin

65, 1285-1298.

Schoell, M. (1980) The hydrogen and carbon isotopic composition of

methane from natural gases of various origins. Geochimica et

Cosmochimica Acta 44, 649-661.

Schovsbo, N.H. (2001) Why barren intervals? A taphonomic case study of

the Scandinavian Alum Shale and its faunas. Lethaia 34, 271-285.

Schovsbo, N.H. (2002) Uranium enrichment shorewards in black shales: A

case study from the Scandinavian Alum Shale. GFF 124, 107-115.

Schovsbo, N.H. (2003) The geochemistry of Lower Palaeozoic sediments

deposited on the margins of Baltica. Bulletin of the Geological

Society of Denmark 50, 11-27.

145

Schovsbo, N.H., Esbensen, K.H., Nielsen, A.T., Derbez, E., Gaucher, E.C.,

Poirier-Coutansais, X., Riou, A., Tallone, P. and Milton-Taylor, D.

(2015) Rock Types in the Scandinavian Alum Shale Resource Play:

Definitions and Predictions, 77th EAGE Conference & Exhibition 2015,

IFEMA, Madrid, Spain, pp. 1-5.

Schovsbo, N.H., Nielsen, A.T. and Gautier, D.L. (2014) The Lower

Palaeozoic shale gas play in Denmark. Geological Survey of

Denmark and Greenland Bulletin 31, 19-22.

Schovsbo, N.H., Nielsen, A.T., Klitten, K., Mathiesen, A. and Rasmussen, P.

(2011) Shale gas investigations in Denmark: Lower Palaeozoic

shales on Bornholm. Geological Survey of Denmark and Greenland

Bulletin 23, 9-12.

Schulz, H.-M., Biermann, S., van Berk, W., Krüger, M., Straaten, N., Bechtel,

A., Wirth, R., Lüders, V., Schovsbo, N.H. and Crabtree, S. (2015)

From shale oil to biogenic shale gas: Retracing organic–inorganic

interactions in the Alum Shale (Furongian–Lower Ordovician) in

southern Sweden. AAPG Bulletin 99, 927-956.

Schulz, H.-M., Horsfield, B. and Sachsenhofer, R. (2010) Shale gas in

Europe: a regional overview and current research activities, in:

Vining, B.A., Pickering, S.C. (Eds.), Geological Society, London,

Petroleum Geology Conference series. Geological Society of London,

London, pp. 1079-1085.

Seifert, W.K. and Moldowan, J.M. (1981) Paleoreconstruction by biological

markers. Geochimica et Cosmochimica Acta 45, 783-794.

Selley, R. (1987) British shale gas potential scrutinized. Oil Gas J 85, 62-64.

Selley, R.C. (2005) UK shale-gas resources. Geological Society, London,

Petroleum Geology Conference series 6, 707-714.

Selley, R.C. (2012) UK shale gas: the story so far. Marine and petroleum

geology 31, 100-109.

Semyrka, G., Jarzyna, J., Semyrka, R., Kaźmierczuk, M. and Pikulski, L.

(2010) Reservoir parameters of lithostratigraphic successions of the

lower Paleozoic strata in the Polish part of the Baltic region based on

laboratory studies and well logs. Geological Quarterly 54, 227-240.

Serra, O. (1983) Fundamentals of well-log interpretation.

Shaw, D.B. and Weaver, C.E. (1965) The mineralogical composition of

shales. Journal of Sedimentary Research 35.

Sheppard, C.W. and Burton, V.L. (1946) The Effects of Radioactivity on

Fatty Acids1, 1a. J Am Chem Soc 68, 1636-1639.

Shimoyama, A. and Johns, W.D. (1971) Catalytic conversion of fatty acids

to petroleum-like paraffins and their maturation. Nature 232, 140-

144.

Silliman, J.E., Li, M., Yao, H. and Hwang, R. (2002) Molecular distributions

and geochemical implications of pyrrolic nitrogen compounds in the

Permian Phosphoria Formation derived oils of Wyoming. Organic

Geochemistry 33, 527-544.

146

Simoneit, B.R. and Lonsdale, P.F. (1982) Hydrothermal petroleum in

mineralized mounds at the seabed of Guaymas Basin. Nature 295,

198-202.

Sing, K.S. (1985) Reporting physisorption data for gas/solid systems with

special reference to the determination of surface area and porosity

(Recommendations 1984). Pure and applied chemistry 57, 603-619.

Sinninghe Damsté, J.S., Rijpstra, W.I.C., Kock-van Dalen, A., De Leeuw,

J.W. and Schenck, P. (1989) Quenching of labile functionalised lipids

by inorganic sulphur species: evidence for the formation of

sedimentary organic sulphur compounds at the early stages of

diagenesis. Geochimica et Cosmochimica Acta 53, 1343-1355.

Sivhed, U. (2004) Upper Ordovician carbonate mounds on Gotland, central

Baltic Sea: distribution, composition and reservoir characteristics.

Journal of Petroleum Geology 27, 115-140.

Šliaupa, S. and Hoth, P. (2011) Geological evolution and resources of the

Baltic Sea area from the Precambrian to the Quaternary, in: Harff, J.,

Björck, S., Hoth, P. (Eds.), The Baltic Sea Basin. Springer Berlin

Heidelberg, pp. 13-51.

Smith, N. (1995) Unconventional hydrocarbons: changing exploration

strategies. Earthwise, 14-15.

Smith, N., Turner, P. and Williams, G. (2010) UK data and analysis for shale

gas prospectivity. Geological Society, London, Petroleum Geology

Conference series.

Smith, P.V. (1952) The occurrence of hydrocarbons in recent sediments

from the Gulf of Mexico. Science 116, 437-439.

Snäll, S. (1988) Mineralogy and maturity of the alum shales of south-

central Jamtland, Sweden. Sveriges geologiska undersökning,

Uppsala.

Snowdon, L.R. (2001) Natural gas composition in a geological environment

and the implications for the processes of generation and

preservation. Organic Geochemistry 32, 913-931.

Sopher, D., Erlström, M., Bell, N. and Juhlin, C. (2016) The structure and

stratigraphy of the sedimentary succession in the Swedish sector of

the Baltic Basin: New insights from vintage 2D marine seismic data.

Tectonophysics 676, 90-111.

Spears, D. and Amin, M. (1981) Geochemistry and mineralogy of marine

and non‐marine Namurian black shales from the Tansley Borehole,

Derbyshire. Sedimentology 28, 407-417.

Stahl, W.J. and Carey, B.D. (1975) Source-rock identification by isotope

analyses of natural gases from fields in the Val Verde and Delaware

basins, west Texas. Chemical Geology 16, 257-267.

Stainforth, J.G. (2009) Practical kinetic modeling of petroleum generation

and expulsion. Marine and Petroleum Geology 26, 552-572.

Stouge, S. and Nielsen, A.T. (2003) An integrated biostratigraphical analysis

of the Volkhov-Kunda (Lower Ordovician) succession at Fågelsång,

147

Scania, Sweden. Bulletin of the Geological Society of Denmark 50,

75-94.

Strachan, M., Alexander, R., Van Bronswijk, W. and Kagi, R. (1989) Source

and heating rate effects upon maturity parameters based on ratios

of 24-ethylcholestane diastereomers. Journal of Geochemical

Exploration 31, 285-294.

Swanson, V.E. (1960) Oil yield and uranium content of black shales.

Geological Survey, Washington, DC (USA).

Swanson, V.E. and Swanson, V.E. (1961) Geology and geochemistry of

uranium in marine black shales: a review. US Government Printing

Office Washington, DC.

Tan, J., Horsfield, B., Mahlstedt, N., Zhang, J., di Primio, R., Vu, T.A.T.,

Boreham, C.J., van Graas, G. and Tocher, B.A. (2013) Physical

properties of petroleum formed during maturation of Lower

Cambrian shale in the upper Yangtze Platform, South China, as

inferred from PhaseKinetics modelling. Marine and Petroleum

Geology 48, 47-56.

Tannenbaum, E., Huizinga, B.J. and Kaplan, I. (1986a) Role of minerals in

thermal alteration of organic matter--II: a material balance. AAPG

bulletin 70, 1156-1165.

Tannenbaum, E. and Kaplan, I.R. (1985a) Low-Mr hydrocarbons generated

during hydrous and dry pyrolysis of kerogen.

Tannenbaum, E. and Kaplan, I.R. (1985b) Role of minerals in the thermal

alteration of organic matter—I: Generation of gases and

condensates under dry condition. Geochimica et Cosmochimica Acta

49, 2589-2604.

Tannenbaum, E., Ruth, E. and Kaplan, I.R. (1986b) Steranes and

triterpanes generated from kerogen pyrolysis in the absence and

presence of minerals. Geochimica et cosmochimica acta 50, 805-812.

Tao, S., Wang, C., Du, J., Liu, L. and Chen, Z. (2015) Geochemical

application of tricyclic and tetracyclic terpanes biomarkers in crude

oils of NW China. Marine and Petroleum Geology 67, 460-467.

Tarafa, M.E., Hunt, J.M. and Ericsson, I. (1983) Effect of hydrocarbon

volatility and adsorption on source-rock pyrolysis. Journal of

Geochemical Exploration 18, 75-85.

Tegelaar, E., De Leeuw, J., Derenne, S. and Largeau, C. (1989) A

reappraisal of kerogen formation. Geochimica et Cosmochimica Acta

53, 3103-3106.

Tegelaar, E.W. and Noble, R.a. (1994) Kinetics of hydrocarbon generation

as a function of the molecular structure of kerogen as revealed by

pyrolysis-gas chromatography. Organic Geochemistry 22, 543-574.

ten Haven, H., de Leeuw, J., Rullkotter, J. and Sinninghe Damsté, J. (1987)

Restricted utility of the pristane/phytane ratio as a

palaeoenvironmental indicator? Nature 330, 641-643.

Teräväinen, M.J., Pakarinen, J.M., Wickström, K. and Vainiotalo, P. (2007)

Comparison of the composition of Russian and North Sea crude oils

148

and their eight distillation fractions studied by negative-ion

electrospray ionization Fourier transform ion cyclotron resonance

mass spectrometry: The effect of suppression. Energy & fuels 21,

266-273.

Thickpenny, A. (1984) The sedimentology of the Swedish Alum Shales.

Geological Society, London, Special Publications 15, 511-525.

Tian, Y. (2010) An Investigation of regional variations of Barnett shale

reservoir properties, and resulting variability of hydrocarbon

composition and well performance. Texas A&M University.

Tissot, B.P. and Welte, D.H. (1984) Petroleum formation and occurrence.

Tolmacheva, T.J., Koren, T.N., Holmer, L.E., Popov, L.E. and Raevskaya, E.

(2001) The Hunneberg Stage (Ordovician) in the area east of St.

Petersburg, north-western Russia. Paläontologische Zeitschrift 74,

543-561.

Treibs, A. (1934) Chlorophyll‐und Häminderivate in bituminösen Gesteinen,

Erdölen, Erdwachsen und Asphalten. Ein Beitrag zur Entstehung des

Erdöls. Eur J Org Chem 510, 42-62.

Tuuling, I. and Flodén, T. (2009) The Llandovery–lowermost Wenlock

sequence in the Baltic Sea between Saaremaa and Gotland;

subdivision, thicknesses and correlation, based on marine seismic

studies. Mar Geol 267, 55-70.

Ulmishek, G. (1990) Geologic evolution and petroleum resources of the

Baltic Basin, in: Leighton, M.W., Kolata, D.R., Oltz, D.F., Eidel, J.J.

(Eds.), Interior Cratonic Basins. American Association of Petroleum

Geologists, Tulsa, Oklahoma, pp. 603-632.

Ungerer, P. (1990) State of the art of research in kinetic modelling of oil

formation and expulsion. Organic Geochemistry 16, 1-25.

Ungerer, P., Behar, F., Villalba, M., Heum, O.R. and Audibert, A. (1988)

Kinetic modelling of oil cracking. Organic Geochemistry 13, 857-868.

Ungerer, P. and Pelet, R. (1987) Extrapolation of the kinetics of oil and gas

formation from laboratory experiments to sedimentary basins.

Nature 327, 52-54.

Ūsaityt, D. (2000) The geology of the southeastern Baltic Sea: a review.

Earth-Science Reviews 50, 137-225.

Van de Meent, D., Brown, S.C., Philp, R.P. and Simoneit, B.R. (1980)

Pyrolysis-high resolution gas chromatography and pyrolysis gas

chromatography-mass spectrometry of kerogens and kerogen

precursors. Geochimica et Cosmochimica Acta 44, 999-1013.

van Duin, A.C. and Larter, S.R. (1997) Unravelling Mango's mysteries: a

kinetic scheme describing the diagenetic fate of C 7-alkanes in

petroleum systems. Organic geochemistry 27, 597-599.

van Duin, A.C. and Larter, S.R. (1998) Application of molecular dynamics

calculations in the prediction of dynamical molecular properties.

Organic Geochemistry 29, 1043-1050.

149

van Kaam-Peters, H.M., Köster, J., van der Gaast, S.J., Dekker, M., de

Leeuw, J.W. and Damsté, J.S.S. (1998) The effect of clay minerals

on diasterane/sterane ratios. Geochimica et Cosmochimica Acta 62,

2923-2929.

Van Krevelen, D. (1950) Graphical-statistical method for the study of

structure and reaction processes of coal. Fuel 29, 269-284.

Vandenbroucke, M., Behar, F. and Rudkiewicz, J.L. (1999) Kinetic modelling

of petroleum formation and cracking: implications from the high

pressure/high temperature Elgin Field (UK, North Sea). Organic

Geochemistry 30, 1105-1125.

Vandenbroucke, M. and Largeau, C. (2007) Kerogen origin, evolution and

structure. Organic Geochemistry 38, 719-833.

Veeh, H., Calvert, S. and Price, N. (1974) Accumulation of uranium in

sediments and phosphorites on the South West African shelf. Marine

Chemistry 2, 189-202.

Vejbæk, G., Stouge, S. and Poulsen, K.D. (1994) Palaeozoic tectonic and

sedimentary evolution and hydrocarbon prospectivity in the

Bornholm area. Geological Survey of Denmark.

Velde, B. and Espitalié, J. (1989) Comparison of kerogen maturation and

illite/smectite composition in diagenesis. Journal of Petroleum

Geology 12, 103-110.

Venkatesan, M. and Dahl, J. (1989) Organic geochemical evidence for global

fires at the Cretaceous/Tertiary boundary. Nature 338, 57-60.

Vine, J.D. and Tourtelot, E.B. (1970) Geochemistry of black shale deposits;

a summary report. Economic Geology 65, 253-272.

Volk, H., George, S.C., Middleton, H. and Schofield, S. (2005) Geochemical

comparison of fluid inclusion and present-day oil accumulations in

the Papuan Foreland–evidence for previously unrecognised

petroleum source rocks. Organic Geochemistry 36, 29-51.

Walters, C. (2006) The origin of petroleum. Practical Advances in Petroleum

Processing, 79-101.

Walters, C.C., Wang, F.C., Qian, K., Wu, C., Mennito, A.S. and Wei, Z.

(2015) Petroleum alteration by thermochemical sulfate reduction–A

comprehensive molecular study of aromatic hydrocarbons and polar

compounds. Geochimica et Cosmochimica Acta 153, 37-71.

Waples, D.W. (2000) The kinetics of in-reservoir oil destruction and gas

formation: constraints from experimental and empirical data, and

from thermodynamics. Organic Geochemistry 31, 553-575.

Waples, D.W. (2010) SOURCE-ROCK KINETICS. Unpublished.

Waples, D.W. and Marzi, R.W. (1998) The universality of the relationship

between vitrinite reflectance and transformation ratio. Organic

Geochemistry 28, 383-388.

Waples, D.W., Suizu, M. and Kamata, H. (1992) The Art of Maturity

Modeling. Part 2: Alternative Models and Sensitivity Analysis. AAPG

Bulletin.

150

Waters, C.N., Waters, R.A., Barclay, W.J. and Davies, J.R. (2009) A

lithostratigraphical framework for the Carboniferous successions of

southern Great Britain (Onshore). BGS Report.

Weaver, C.E. (1960) Possible uses of clay minerals in search for oil. AAPG

Bulletin 44, 1505-1518.

Wei, Z., Moldowan, J.M., Dahl, J., Goldstein, T.P. and Jarvie, D.M. (2006)

The catalytic effects of minerals on the formation of diamondoids

from kerogen macromolecules. Organic Geochemistry 37, 1421-

1436.

Welte, D. and Waples, D. (1973) Über die Bevorzugung geradzahliger n-

Alkane in Sedimentgesteinen. Naturwissenschaften 60, 516-517.

Welte, D.H. (1965) Relation between petroleum and source rock. AAPG

Bulletin 49, 2246-2268.

Welte, D.H. (1970) Organischer Kohlenstoff und die Entwicklung der

Photosynthese auf der Erde. Naturwissenschaften 57, 17-23.

Whitmore, F.C. (1934) Mechanism of the polymerization of olefins by acid

catalysts. Industrial & Engineering Chemistry 26, 94-95.

Whyte, G.N. (1973) Principles of radiation dosimetry. John Wiley and Sons,

Inc., New York.

Więcław, D., Kotarba, M.J., Kosakowski, P., Kowalski, A. and Grotek, I.

(2010a) Habitat and hydrocarbon potential of the lower Paleozoic

source rocks in the Polish part of the Baltic region. Geological

Quarterly 54, 159-182.

Więcław, D., Kotarba, M.J. and Kowalski, A. (2010b) Origin of oils

accumulated in the Middle Cambrian reservoirs of the Polish part of

the Baltic region. Geological Quarterly 54, 205-216.

Williams, J., Bjorøy, M., Dolcater, D. and Winters, J. (1986) Biodegradation

in South Texas Eocene oils—effects on aromatics and biomarkers.

Organic Geochemistry 10, 451-461.

Williams, P.T., Besler, S. and Taylor, D.T. (1990) The pyrolysis of scrap

automotive tyres: The influence of temperature and heating rate on

product composition. Fuel 69, 1474-1482.

Wróbel, M. and Kosakowski, P. (2010) 2-D modelling of petroleum

processes of the lower Paleozoic strata in the Polish part of the

Baltic region. Geol. Quart 54, 257-266.

Wu, L.M., Zhou, C.H., Keeling, J., Tong, D.S. and Yu, W.H. (2012) Towards

an understanding of the role of clay minerals in crude oil formation,

migration and accumulation. Earth-Science Reviews 115, 373-386.

Wu, Z., Jernström, S., Hughey, C.A., Rodgers, R.P. and Marshall, A.G.

(2003) Resolution of 10 000 compositionally distinct components in

polar coal extracts by negative-ion electrospray ionization Fourier

transform ion cyclotron resonance mass spectrometry. Energy &

fuels 17, 946-953.

Xiao, Y. and James, A. (1997) Is acid catalyzed isomerization responsible

for the invariance in isoheptanes of petroleum, The 18th

151

International Meeting on Organic Geochemistry, Book of Abstracts,

pp. 769-770.

Yang, S. and Horsfield, B. (2016) Some predicted effects of minerals on the

generation of petroleum in nature. Energy & Fuels 30, 6677-6687.

Yang, S., Horsfield, B., Mahlstedt, N., Stephenson, M. and Könitzer, S.

(2015) On the primary and secondary petroleum generating

characteristics of the Bowland Shale, northern England. Journal of

the Geological Society.

Yang, S., Horsfield, B., Mahlstedt, N., Stephenson, M.H. and Könitzer, S.F.

(2016) On the primary and secondary petroleum generating

characteristics of the Bowland Shale, northern England. Journal of

the Geological Society 173, 292-305.

Yang, S., Schulz, H.-M., Schovsbo, N.H. and Bojesen-Koefoed, J.A. (2017)

Oil-source rock correlation of the Lower Palaeozoic petroleum

system in the Baltic Basin (northern Europe). AAPG Bulletin

preliminary version published online Ahead of Print May 22, 2017.

Zdanavičiūtė, O. (2012) The Middle Cambrian Succession In The Central

Baltic Basin: Geochemistry Of Oils And Sandstone Reservoir

Characteristics. Journal of Petroleum Geology 35, 237-254.

Zdanavičiūtė, O. and Lazauskienė, J. (2004) Hydrocarbon migration and

entrapment in the Baltic Syneclise. Organic Geochemistry 35, 517-

527.

Zdanavičiūtė, O. and Lazauskienė, J. (2009) Organic matter of Early Silurian

succession–the potential source of unconventional gas in the Baltic

Basin (Lithuania). Baltica 22, 89-99.

Zdanavikiute, O. and Bojesen-Koefoed, J.A. (1997) Geochemistry of

Lithuanian oils and source rocks: a preliminary assessment. Journal

of Petroleum Geology 20, 381-402.

Zengler, K., Richnow, H.H., Rosselló-Mora, R., Michaelis, W. and Widdel, F.

(1999) Methane formation from long-chain alkanes by anaerobic

microorganisms. Nature 401, 266-269.

Ziegs, V. (2013) Development of a compositional kinetic model for primary

and secondary petroleum generation from Lower Cretaceous

Wealden Shales , Lower Saxony Basin , Northern Germany.

Ziegs, V., Horsfield, B., Skeie, J.E. and Rinna, J. (2017) Petroleum retention

in the Mandal Formation, Central Graben, Norway. Marine and

Petroleum Geology 83, 195-214.

Zobell, C.E. (1945) The role of bacteria in the formation and transformation

of petroleum hydrocarbons. Science (New York, NY) 102, 364-369.