Coal and Coal-bearing Strata as Oil-prone Source Rocks?
Geological Society Special Publications Series Editor A. J. F...
150 downloads
1454 Views
20MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Coal and Coal-bearing Strata as Oil-prone Source Rocks?
Geological Society Special Publications Series Editor A. J. FLEET
GEOLOGICAL SOCIETY SPECIAL PUBLICATION NO 77
Coal and Coal-bearing Strata as Oil-prone Source Rocks? EDITED BY
A N D R E W C. S C O T T Geology Department, Royal Holloway University of London AND
A N D R E W J. F L E E T BP Exploration, Sunbury-on-Thames
1994 Published by The Geological Society London
THE GEOLOGICAL SOCIETY The Society was founded in 1807 as The Geological Society of London and is the oldest geological society in the world. It received its Royal Charter in 1825 for the purpose of 'investigating the mineral structure of the Earth'. The Society is Britain's national society for geology with a membership of 7500 (1992). It has countrywide coverage and approximately 1000 members reside overseas. The Society is responsible for all aspects of the geological sciences including professional matters. The Society has its own publishing house which produces the Society's international journals, books and maps, and which acts as the European distributor for publications of the American Association of Petroleum Geologists and the Geological Society of America. Fellowship is open to those holding a recognized honours degree in geology or cognate subject and who have at least two years relevant postgraduate experience, or who have not less than six years relevant experience in geology or a cognate subject. A Fellow who has not less than five years relevant postgraduate experience in the practice of geology may apply for validation and, subject to approval, may be able to use the designatory letters C. Geol (Chartered Geologist). Further information about the Society is available from the Membership Manager, The Geological Society, Burlington House, Piccadilly, London W1V 0JU, UK. Published by The Geological Society from: The Geological Society Publishing House Unit 7 Brassmill Enterprise Centre Brassmill Lane Bath BA1 3JN UK (Orders: Tel. 0225 445046 Fax 0225 442836)
Distributors USA A A P G Bookstore PO Box 979 Tulsa Oklahoma 74101-0979 USA (Orders: Tel. (918)584--2555 Fax (918)584-0469)
First published 1994
Australia Australian Mineral Foundation 63 Conyngham St Glenside South Australia 5065 Australia (Orders: Tel. (08)379-0444 Fax (08)379-4634)
(~) The Geological Society 1994. Apart from fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publisher or in the case of reprographic reproduction in accordance with the terms of licences issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W l P 9HE, UK. Users registered with Copyright Clearance Center, 27 Congress St., Salem, MA 01970, USA: the item-fee code for this publication is 0305-8719/94 $07.00. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 0-903317-99-0
Typeset by EJS Chemical Composition, Midsomer Norton, Bath, Avon Printed by Alden Press, Oxford, UK
India Affiliated East-West Press PVT Ltd G-1/16 Ansari Road New Delhi 110 002 India (Orders: Tel. (11)327-9113 Fax (11)326--0538) Japan Kanda Book Trading Co. Tanikawa Building 3-2 Kanda Surugadai Chiyoda-Ku Tokyo 101 Japan (Orders: Tel. (03)3255-3497 Fax (03)3255-3495)
Contents FLEET,A. J. & SCOTT, A. C. Coal and coal-bearing strata as oil-prone source rocks: an overview General studies
POWELL, T. G. & BOREHAM, C. J. Terrestrially sourced oils: where do they exist and what are our limits of knowledge? COLLINSON, M. E., VAN BERGEN, P. F., ScoTT, A. C. & DE LEEUW, J. W. The oilgenerating potential of plants from coal and coal-bearing strata through time: a review with new evidence from Carboniferous plants PHILP, P. R. Geochemical characteristics of oils derived predominantly from terrigenous source materials STOUT, S. A. Chemical heterogeneity among adjacent coal microlithotypes implications for oil generation and primary migration from humic coal MACGREGOR, D. S. Coal-bearing strata as source rocks - a global overview
11
31 71 93 107
Case histories
THOMPSON, S., COOPER, B. S. & BARNARD,P. C. Some examples and possible explanations for oil generation from coals and coaly sequences MATCHETTE-DOWNES, C. J., FALLICK, A. E., KARMAJAYA, & ROWLAND, S. A maturity and palaeoenvironmental assessment of condensates and oils from the North Sumatra Basin, Indonesia CURRY, D. J., EMMET'r, J. K. & HUNT, J. W. Geochemistry of aliphatic-rich coals in the Cooper Basin, Australia, and Taranaki Basin, New Zealand: implications for the occurrence of potentially oil-generative coals BAGGE, M. A. & KEELY, M. L. The oil potential of Mid-Jurassic coals in northern Egypt ScoTT, A. C. & FLEET, A. J. Coal and coal-bearing strata as oil-prone source rocks: current problems and future directions
119
201
Index
207
139
149 183
Coal and coal-bearing strata as oil-prone source rocks: an overview ANDREW
J. F L E E T 1 & A N D R E W
C. S C O T T 2
1Bp Exploration, BP Research and Engineering Centre, Chertsey Road, Sunbury-on-Thames, Middlesex, TW16 7LN, UK 2Department of Geology, Royal Holloway University of London, Egham, Surrey TW20 OEX, UK Abstract: Despite many advances in our knowledge over the last decade, understanding of why some coals and coal-bearing strata are oil prone, and our ability to predict such sequences ahead of drilling in petroleum exploration, is relatively poor. Here we review the current status of the knowledge, highlighting contributions made in this volume. Oil-prone coals are hydrogen rich. Those which are generally acknowledged to have given rise to significant oil accumulations occur either as low latitude Tertiary deposits or within late Jurassic-Palaeogene sequences of the Australian region. Oils derived solely from coals and other terrigenous kerogens can be recognized using, geochemical criteria. Recognition of oil-prone coals and associated mudrocks visually or geochemically, however, is problematical. What controls the expulsion of petroleum in the liquid phase from coalbearing sequences is probably the critical factor. Our knowledge of the botanical, depositional and diagenetic controls which determine the formation of oil-prone coaly sequences, and hence our ability to predict their presence, is currently empirical and lacks understanding of the inputs and processes involved.
This book sets out to review the current status of our understanding of the formation of oil accumulations from coals. It is not concerned just with coals sensu stricto but also with the organic-rich mudrocks found in coal-bearing strata. Similarly although the focus is very much on kerogen derived from terrigenous material it is impossible to exclude some consideration of sequences of coals and mudrocks containing algal kerogen because coal accumulation and lacustrine sedimentation often sit at opposite ends of a depositional continuum in space and/or time (e.g. Powell & Boreham). Microbial biomass is a third, and very probably often significant, contributor to terrigenous kerogen (e.g. Curry e t M.). Non-marine source rocks can be estimated to account for less than 10% of world oil (Fleet & Brooks 1987) and much of this non-marine contribution is from lacustrine source rocks, accounting for 85-95% of the oil in areas such as Brazil, China and Indonesia (Katz et al. 1991). Despite being a poor relation as an oil source in global terms, oil-prone coal sequences are recognized as key oil source rocks in at least Southeast Asia, Australia and New Zealand. Even in these regions the critical factors which make coals and their associated mudrocks oil prone is debated. Elsewhere, for instance in Arctic North America and the North Sea, the
ability of coal sequences to source oil is contenious. Stratigraphically oil-prone coals are mainly limited to late Jurassic and younger sequences (e.g. Macgregor) but some workers question any implied causal link (e.g. D u r a n d & Paratte 1983). Understanding why coal-beating sequences are oil prone can unlock a predictive capability for petroleum exploration and so help to reduce exploration risk.
Background The link between oils and terrigenous organic matter was first made by Hedberg (1968). He recognized that high-wax, low sulphur oils were commonly associated with non-marine, often coal-bearing, strata, though he correctly suggested that both terrigenous organic matter and non-marine aquatic matter could be the source materials of these oils. Since then, various workers have reviewed the question of which factors make coal-bearing strata oil prone or, at least, which control the expulsion of liquid-phase petroleum from these strata (e.g. D u r a n d & Parratte 1983; Murchison 1987). Most recently a symposium of the American Chemical Society reviewed the topic (Hunt 1991 et seq.) through a number of case studies and laboratory approaches. As geochemical and related thinking on oil-prone coals has edged
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coaland Coal-bearingStrata as Oil-prone Source Rocks? Geological Society Special Publication No. 77, pp. 1-8.
2
A. J. FLEET & A. C. SCOq~F
forward since Hedberg's (1968) observations, thinking on coal formation and palaeobotany have evolved rapidly (e.g. Scott 1987; Bertrand 1991). There is now a need to try and fully integrate the contributions which different disciplines can make to understanding oil-prone coals. This book represents an attempt to begin to draw these disciplines together. T h e issues
The underlying question behind this book is what factor(s) make(s) coals or coal-bearing strata capable of expelling petroleum in the liquid phase? Implicit in this question is understanding the relative contributions of liquids and gases which any coal-bearing sequence can expel in response to a particular thermal history. The reason for the question is that petroleum explorers need to be able to predict ahead of drilling both sourcing systems and the phases and composition of petroleum in prospects. In this book we attempt to review our current understanding of at least the major issues which need to be resolved if we are to approach answering this underlying question. • How do we recognize oil-prone coals and associated mudrocks if we have samples for analysis and how do we characterize oils derived from coal-bearing sequences? Which parameters, or combination of parameters, do we use for these tasks (e.g. visual, geochemical, palaeobotanical)? • What botanical input or depositional environment or early diagenetic conditions, or combination of these, governs the formation of oil-prone coal-bearing sequences? Can a predictive model, which can be used prior to drilling, be derived from understanding these factors? • How do variations in generation, expulsion behaviour and secondary migration determine whether or not coal-bearing sequences give rise to oil? Characterization Microscopy and geochemistry have been used alone and in combination in attempts to try and distinguish oil-prone coals from other coals (e.g. Murchison 1987). The key overriding factor is that the coals are rich in hydrogen relative to carbon. Hunt (1991) has suggested that this translates into approximate thresholds for the following parameters of: • H/C ratios > 0.9
• Hydrogen Indices > 200 • liptinite contents > 15% Powell & Boreham argue that none of these can be used in isolation: 'the overall petrographic composition of coal is a poor guide to its petroleum potential' and 'there is not a simple relationship between the elemental composition of terrigenous kerogen, the gross pyrolysis yield by Rock-Eval and the yield of normal hydrocarbons in pyrolysis gas chromatography'. They suggest that a combination of a mass balance approach and pyrolysis is the best indicator of expelled liquid and gaseous petroleum from coals. The problems with using microscopy are partly those of the mismatches between the terminologies of coal petrology and kerogen typing and the actual biological components from which individual macerals or kerogen constituents come (Collinson et al.), but they also relate to the often imperfect characterization of organic matter which can be made microscopically. For instance, Curry et M. suggest that microbial biomass is a significant contributor to the Permian coals of the Cooper Basin, Australia, which are the source of oils found in Jurassic and older reservoirs of the basin (Powell & Boreham). Under the microscope these coals are inertinite rich and very low in liptinites. Another factor which may be significant in militating against using coal macerals for recognizing oil proneness is that it is possibly the association of macerals, the microlithotype, rather than the macerals themselves, which controls the expulsion of generated liquid petroleum from coals (Stout). The recognition of oils derived from coals can be as important as characterizing coals themselves. Recognition that produced or seeped oil in a basin is from coals will either open up new plays if a coal source has not previously been identified in the basin or allow typing of the sourcing system operating in the basin (e.g. Matchette-Downes e t al.). The 'correlation' will not, of course be to coals sensu stricto but to sequences containing terrigenous organic matter. As Hedberg (1968) pointed out, the clues to this type of source will be high wax and low sulphur; what will confirm the diagnosis will be high pristane/phytane ratios (e.g. > 3 - 4 Powell & Boreham). Biomarker molecules in the oils, the origins of which can be linked back to land plant communities, can add further evidence to the 'correlation' (e.g. Philp). Biomarker evidence alone may, however, be equivocal since biomarkers may be derived from terrigenous organic matter deposited in a marine or lacustrine environment. The Niger Delta
AN OVERVIEW offers a good example of this dilemma. The oils of the delta are cited as classic examples of oils derived from terrigenous kerogen as they contain the biomarker oleanane which is believed to be derived from flowering plants (Hills & Whitehead 1966; Peters & Moldowan 1993). No source rock, though, has been unequivocally identified in the literature. The delta top sequences which have been sampled through drilling seem to contain no rich source rock but may make up in thickness what they lack in quality and richness (Bustin 1988). Alternatively, or in addition, the pro- and (?)pre-delta Akata shales, which are unsampled in potential source kitchens, may be the source (Weber & Daukoru 1975): these mudrocks could contain abundant terrigenous organic matter which was • deposited beyond the delta front. Isotopes provide another way of characterizing coals and their petroleum products. To date they have generally been used for characterizing the total carbon or bulk fraction of a kerogen, oil or gas. Isotope analysis of individual molecules has largely been restricted to simple C1-Cs hydrocarbon constituents of gases but over recent years gas chromatograph-mass spectrometers capable of compound-specific isotope analysis have come into use. Nevertheless isotope analysis is currently probably at its most useful for characterizing gases and for oiloil and oil-source correlations. Isotopes offer a principal line of evidence for the origin of gases, potentially distinguishing biogenic gas, oilassociated gas and thermogenic gas which is not associated with oil (e.g. Schoell 1983; Clayton 1991). They are, therefore, important in studying the petroleum generated from coal-bearing strata which, even when containing liquid products, also contain high proportions of gas (see below). In contrast, the use of carbon isotopes to characterize kerogen as non-marine or marine is not as straightforward as it was once considered. Based on recent organic matter, non-marine kerogens have been characterized as having lighter carbon isotope compositions (81aC of about - 27%o) relative to marine kerogens (d13C of about - 20%0) (e.g. Tissot & Welte 1984). However, recent studies (e.g. Clayton in press) suggest that this has only been the situation since Miocene times; previously the relationship was reversed, the relative difference varying through the Phanerozoic.
Prediction: input, depositional environment, diagenesis In potentially coal-bearing basins there is a need in petroleum exploration to predict ahead of
3
drilling whether any coal-bearing strata present are oil prone. Similarly, if the presence of oil-prone coal-bearing sequences has been identified from sediment samples or deduced from oil samples there is a need to predict the distribution of the oil-prone coaly facies. This need then becomes one of assessing the probable gas : oil ratios (GOR) of the petroleum expelled from these facies so as to predict GORs of individual prospects. Any of these predictions requires an ability to identify or understand the factors which determine whether a coal is oil prone. In terms of facies distribution the botanical origin of the organic input to the sediments, the depositional environment and early diagenesis may all play a part, very possibly in a close interrelationship. (They may also ultimately determine expulsion behaviour from the coals and, therefore, oil proneness: see next section.) The botanical input to coal depositional environments will have varied through geological time as plant communities have evolved and will have also been governed by climate and other environmental factors (Collinson & Scott 1987). Thomas (1982) drew attention to these likely controls and argued that, at least in Australian basins, the dominance of conifers in swamp floras since the Jurassic, and the evolution of the angiosperms from Late Cretaceous times, meant that there was a relatively abundant input of potentially oil-prone detritus which became preserved as exinite. Thompson et al. (1985 and this volume) added another dimension to this kind of reasoning when considering the Tertiary oil-prone coal sequences of Southeast Asia. They suggested that the oil-prone coals in these basins resulted from preferential preservation of potentially oilprone detritus in coastal plain environments under everwet, tropical conditions. Cawley & Fleet (1987) suggested these strands of reasoning might be knitted together and point to Cretaceous and younger tropical and subtropical lower delta-plain deposits as being the prime sites of oil-prone coal development. The papers in this volume suggest that at least two other types of organic input need to be taken into account when considering the prediction of oil-prone coals. They are microbial biomass added during diagenesis and resinite. As discussed above, microbial biomass probably accounts for the oil-prone nature of the Permian coals from the Cooper Basin, Australia. What might control the development of microbial biomass in abundance and whether such biomass is a significant contributor to all oil-prone coals is currently an open question. The contribution
4
A. J. FLEET & A. C. SCOTT
of resinite to the oil potential of coals has been much debated (see discussion in Powell & Boreham). In general its presence does not seem critical to the potential but in one area, the Beaufort-Mackenzie Delta region of northern Canada, resinite has been argued to be the key terrigenous kerogen component present (Snowden 1991). Macgregor's statistical analysis of the distribution of oil-prone coals bears out the hypothesis that oil-prone coals are Jurassic and younger phenomena at least as far as significant accumulations ( > 50 million barrels of recoverable oil) of coal-sourced oil are concerned. He find~ that significant oil-prone coal-bearing sequences are restricted to two 'fairways': • Tertiary basins within 20 ° of the palaeoequator; • Late Jurassic-Eocene basins formed on the Australian and associated plates. Macgregor's analysis only relates to extant oil accumulations and is subject to the observation that coals may have given rise to significant oil accumulations in early times which have subsequently been destroyed by tectonic disruption, oil to gas cracking etc. Certainly this is one view Durand & Paratte (1983) took of previous observations of the stratigraphic distribution of coalsourced oils. Their alternative view was that all coals are capable of generating oil so expulsion must be the critical factor for the petroleum potential of coals. The innovative studies of Collinson e t al. support the latter view. Their studies of the components of Carboniferous coals suggest that such materials would have been sources of the constituent compounds of oil. Whether they gave rise to expelled liquids, though, is an open question. Overall our ability to predict oil-prone coal-bearing sequences remains a matter of debate. Such strata may either generally be restricted to Cretaceous and younger sequences, because of botanical input or some as yet unrecognized control on oil expulsion, or they may be expected throughout post-Devonian sequences but in association with fewer oil accumulations the older they are.
Generation, expulsion and migration For any given thermal history oil-prone coal and other terrigenous kerogens generally generate oil (Cs+) at higher temperatures than the aquatic kerogens of marine and lacustrine source rocks (e.g. Tissot et al. 1987; Horsfield et al. 1988; Noble et al. 1991). An exception is probably resinite, or at least diterpenoid resinite which, because of its labile nature, may be responsible
for causing earlier significant generation in some sequences (see discussion by Snowden 1991 and Powell & Boreham). As discussed above, the critical factor which determines whether or not coals and associated sediments give rise to oil may be expulsion. Controls on expulsion are debated. Broadly they range from generated oil having to exceed some threshold value to fracturing (e.g. Duppenbecker et al. 1991 and Pepper 1991 and references therein). The threshold which must be exceeded is either some percentage of the source-rock pore network, above which it is assumed interconnected pores form pathways out of the source rock, or some value above which adsorption by the source rock, on, say, the kerogen matrix, is exceeded. The dominance of a particular control on oil expulsion may depend on the source rock: for instance adsorption onto kerogen may only become critical in source rocks with relatively moderate or low hydrogen indices (e.g. of <300-400), such as coals (England & Fleet 1991). A further aspect of migration may be the overall nature of the coalbearing sequence which may dictate how easily petroleum escapes from the sequence or migrates to traps within the sequences. Net: gross sand ratios, faulting etc. may be key factors; for example Cornford et al. (1986) have suggested that the high proportions of sand in the Brent coal-bearing sequences of the North Sea may have led to easy migration of petroleum through and out of the sequence. Generated oil retained in a source rock may, of course, undergo cracking to gas if maturity increases sufficiently. Oil components (Cs+ hydrocarbons) can then be expelled in the gas phase to form condensates under lower temperature and pressure conditions when the expelled petroleum has migrated to shallower depths. The composition of the oil dissolved in the gas phase seems to be enhanced by increased pressure and temperature in the source rock (Leythaeuser & Poelchau 1991). Predicting the likely gas : oil ratio (GOR) in a prospect ahead of drilling requires assessing the total oil and gas charge expelled from the source kitchen of the prospect. A coal-bearing sequence may expel liquid phase products (i.e. Cs+ hydrocarbon oil components with dissolved Cl_s hydrocarbon gas components) and will, with increased maturity, expel gas phase products (i.e. Cl_s hydrocarbons which may have dissolved C5÷ hydrocarbons). The phase(s) in which these products occur in the prospect is determined by the composition of the total charge reaching the prospect and the pressuretemperature conditions of the prospect (e.g.
AN OVERVIEW England et al. 1991). Coal-bearing strata are, of course, very significant sources of thermogenic dry gas (e.g. Macgregor), but even those which give rise to liquid accumulations yield large amounts of gas (e.g. Powell & Boreham). For example, the overall G O R of the Mahakam Delta is 10 000 scf bb1-1, equivalent to a gas mass fraction of approximately three quarters (Pepper 1991). This fact must influence the production and economics of any petroleum province sourced by coal-bearing strata.
Regional variations Southeast Asia and Australia are the main regions where significant coal-sourced oil provinces are recognized (e.g. Macgregor). The source sequences are generally Tertiary (e.g. Indonesian Basins: Matchette-Downes et al., Stout, Thompson et al.) or Jurassic/Cretaceous (e.g. Gippsland: Powell & Boreham) in age, the Permian coals of the Cooper Basin, Australia being notable exceptions (see discussion above and Powell & Boreham and Curry et al.). Elsewhere, Late Cretaceous or Palaeogene coalbearing sequences are, or may be, the sources of oil in the Assam province of India, Central Myanmar and the Taranaki Basin of New Zealand (Macgregor). While, apparently uniquely, allochthonous Tertiary terrigenous kerogen containing significant amounts of resinite has sourced oils within deltaic sediments of the Beaufort-Mackenzie Basin of northern Canada (see discussion in Powell & Boreham). A question mark hangs over the relative contribution that terrigenous detritus makes to the oils of other deltaic systems such as that of the Niger. Certainly geochemical evidence links some constituents of the oils with terrigenous kerogen but this may represent land-plant material deposited in the marine environment (see discussion above). Indeed, reworking of coastal-plain peats into pro-delta sediments is the source of the oil in the Carboniferous deltas of northern England according to Thompson et al.. Whether such reworking is sufficient to give rise to an oil-prone kerogen without accompanying preservation of marine organic matter must remain open to question. There seems no evidence from the pro-delta sediments of Southeast Asia for preservation of oil-prone reworked material (e.g. Duval et al. 1992); reworking seems to have, at best, left some residual potential for gas. In contrast, the prodelta depositional settings of both the Carboniferous of northern England (e.g. Fraser et al. 1990) and of the Cretaceous/Palaeogene Niger (e.g. Whiteman 1982) were likely to have been
5
restricted and allowed the preservation of potentially oil-prone marine organic matter along with allochthonous terrigenous detritus. The Jurassic coals of the Western Desert of Egypt offer another example of oil-prone coals where reworking may have concentrated oilprone material of higher plant or algal origin (Bagge & Keeley). They were deposited in a coastal environment subject to marine incursions. Thompson et al. suggest that either the coals contain a significant non-marine algal component or that, during a marine transgression, degradation of the coal by sulphatereducing bacterial led to the conversion of higher plant detritus into microbial biomass. The coals of the North Sea are generally recognized to be gas prone. Carboniferous coals sourced the gas fields of the Southern North Sea (e.g. Cornford e t al. 1986) but, while oil-prone lacustrine algal coals, similar to those of the oil shales of the Midland Valley of Scotland (e.g. Parnell 1988), may occur, they are not recognized to have contributed significantly to the oil reserves of the area. Jurassic coals of the North Sea and Norwegian continental margin are generally believed to be gas prone (e.g. Thomas et al. 1985) though Lower Jurassic coals may give rise to condensate from expelled gas-phase products (e.g. Elvsborg et al. 1985). Indeed, on the basis of geochemical data, Pittion & Gouadian (1985) argued that the Lower Jurassic 'Coal Unit' of the Haltenbanken area may have had the potential to source oil and Thompson et al. (1985) speculated that the Late Triassicearly Jurassic offshore Central and Northern Norway may contain coaly oil-prone source rocks similar to those of Southeast Asia. Evidence in support of these ideas is equivocal. Heum et al. (1986), using a predominantly modelling approach, concluded that the Lower Jurassic coal sequence of Haltenbanken is the most important source rock for liquids as well as for gases in the area. Forbes et al. (1991) used data from the SmCrbukk SCr field, Haltenbanken, to test a model of petroleum generation and expulsion. They found that Lower Jurassic coal-bearing sequence could have sourced petroleum, mainly oil, principally over the last 5 Ma. The amount of petroleum expelled is modelled to have been several times greater than that needed to fill the recognized accumulations, suggesting spillage from structure; the composition of the Sm0rbukk S0r field compares well with the modelled composition of the petroleum expelled from the source over the last 0.5 Ma. In contrast, others (e.g. Cohen & Dunn 1987), considering both regional maturity and expulsion and geochemical correlation para-
6
A. J. FLEET & A. C. SCOTF
meters, have concluded that the Lower Jurassic coal sequences of Mid-Norway have given rise to gas and condensate but are not the source of the oil in the region. In comparison to the Lower Jurassic coals of the North Sea, Middle Jurassic Brent coals are believed to have limited potential for gas generation (e.g. Cornford et al. 1986). Cornford et al. (1986) ascribe this to the presence of large amounts of inertinite in these coals, possibly as a result of a fluctuating seasonal water table leading to oxidation of accumulated peat. Alternatively the inertinite represents fossil charcoal resulting from wild fires, c o m m o n in many Jurassic sequences (Scott 1989). Variations in potential do occur, though, and minor oil-prone coals have been identified. For example, Ducazeaux et al. (1991) describe two coals from the Brent Group: one oil prone, the other gas prone. The former is dominated by pteridophyte spores and has a Hydrogen Index of 459. Petrographically it contains mainly vitrinite with little inertinite and some exinite. It is interpreted as a spore accumulation deposited close to source in a back-barrier swamp. The other coal contains mainly pollen, is dominated by inertinite and has a Hydrogen Index of 151. In contrast to the components of the oil-prone coal, its constituent pollen is believed to have undergone considerable transport and been deposited in a tidal flat environment.
Conclusions While our knowledge of oil-prone coals and the oils to which they give rise has grown considerably over the last decade or so, our understanding of why a coal is oil prone and how we predict the presence of oil-prone coals ahead of drilling is poor. In summary we know that: • oil-prone coals are hydrogen rich (e.g. with Hydrogen Indices > 200-300); • oil-prone coals which are generally acknowledged to have given rise to significant oil accumulations occur either as low latitutde Tertiary deposits or within late JurassicPalaeogene sequences of the Australian region; • oils derived solely from coals and other terrigenous kerogen have pristane/phytane ratios > 3-4. The key areas which we need to address in order to advance our understanding further, and how we might progress, are summarized at the end of this volume by Scott & Fleet. This volume arises from a joint meeting of the Coal and Petroleum Groups of the Geological Society held in London in June 1992. We thank all the contributors and participants at the meeting for the lively discussion which, although not providing sudden revelations, did clarify current wisdom and outstanding issues. We would like to thank the staff of the Geological Society, particularly Heide Gould and Helen Softley, without whom the meeting would not have taken place. Finally we thank the contributors to this volume, the reviewers and Angharad Hills and Joanna Cooke of the Geological Society Publishing House for ensuring this volume appeared in a reasonable time frame.
References BERTRAND, P. (ed.) 1991. Coal: Formation, occurrence and related properties. Bulletin de la Socidtd Gdologique de France, 162, 137-448. BUSTIN, R. M. 1988. Sedimentology and characteristics and dispersed organic matter in Tertiary Niger Delta: origin of source rocks in a deltaic environment. American Association of Petroleum Geologists Bulletin, 72,277-298. CAWLEY,S. J. & FLEET, A. J. 1987. Deltaic petroleum source rocks: a review. In: Abstracts for Deltas: sites and traps for fossil fuels. Meeting of the Geological Society of London. CLAYTON, C. J. 1991. Carbon isotope fractionation during natural gas generation from kerogen. Marine and Petroleum Geology, 8,232-240. in press. Microbial and organic processes. In: SELLWOOD,B. & PARKER, A. J. (eds) Quantiative diagenesis. NATO Advances Studies Series. COHEN, M. J. & DUNN, M. E. 1987. The hydrocarbon habitat of the Haltenbanken-Traenabank area, offshore Norway. In; BROOKS, J. & GLENNIE,
K. W. (eds) Petroleum Geology of North West Europe. Graham & Trotman, London, 10911104. COLLmSON, M. E. & ScoTt, A. C. 1987. Implications of vegetational change through the geological record on models for coal-forming environments. In: ScoTt, A. C. (ed.) Coal and Coalbearing Strata: Recent Advances. Geological Society, London, Special Publication, 32, 67-85. CORNFORD, C., NEEDHAM, C. E. J. ~£ DE WALQUE, L. 1986. Geochemical habitat of North Sea oils and gases. In: SPENCER,A. M. et al. (eds) Habitat of hydrocarbons on the Norwegian Continental Shelf. Graham & Trotman, London, 39-54. DUCAZEAUX, J., LE TRAN, K. & NICOLAS, G. 1991. Brent coal typing by combined optical and geochemical studies. Bulletin des Centres de Recherches Exploration-Production, ElfAquitaine, 15,369-381. DUPPENBECKER, S. J., DOHMEN, L. • WELTE, D. H. 1991. Numerical modelling of petroleum expul-
AN OVERVIEW sion in two areas of the Lower Saxony Basin, Northern Germany. In: ENGLAND, W. A. & FLEET, A. J. (eds) Petroleum Migration. Geological Society, London, Special Publication, 59, 47-64. DURAND, B. & PARATrE, M. 1983. Oil potential of coals: a geochemical approach. In: BROOKS, J. (ed.) Petroleum Geochemistry and Exploration of Europe. Geological Society, London, Special Publication, 12,255-265. DUVAL, B. C., CHOPPINDE JANVRY, G. & LOIRET, B. 1992. Detailed geoscience reinterpretation of Indonesia's Mahakam delta scores. Oil and Gas Journal, 90 (32), 67-72. ELVSBORG, A., HAGEVANG,T. & THRONDSEN,T. 1985. Origin of gas-condensate of the Midgard field at Haltenbanken. In: THOMAS, B. M. et al. (eds) Petroleum Geochemistry in Exploration of the Norwegian Shelf. Graham & Trotman, London, 213-219. ENGLAND, W. A. & FLEET, A. J. (eds) 1991. Introduction. In: ENGLAND, W. A. & FLEET, A. J. (eds) Petroleum Migration. Geological Society, London, Special Publication, 59. , MANN, A. L. & MANN, D. M. 1991. Migration from source to trap. In: MERRILL, R. K. (ed.) Source and Migration Processes and Evaluation Techniques. American Association of Petroleum Geologists Treatise of Petroleum Geology, 23-46. FLEET, A. J. & BROOKS, J. 1987. Introduction. In: BROOKS, J. & FLEET, A. J. (eds) Marine Petroleum Source Rocks. Geological Society, London, Special Publication, 26, 1-14. FORBES, P. L., UNGERER,P. M., KUHFUSS,A. B., FIIS, F. & EGGEN, S. 1991. Compositional modelling of petroleum generation and expulsion: trial application to a local mass balance in the Smorbukk Sor Field, Haltenbanken. American Association of Petroleum Geologists Bulletin, 75, 873-893. FRASER, A. J., NASH, D. F., STEELE, R. P. & EBDON, C. C. 1990. A regional assessment of the intraCarboniferous play of northern England. In: BROOKS, J. (ed.) Classic Petroleum Provinces. Geological Society, London, Special Publication, 50, 417-440. HEDBER6, H. D. 1968. Significance of high-wax oils with respect to genesis of petroleum. American Association of Petroleum Geologists Bulletin, 52, 736--750. HEUM, O. R., DALLAND,A. & MEISINGSET,K. K. 1986. Habitat of hydrocarbons at Haltenbanken (PVTmodelling as a predictive tool in hydrocarbon exploration). In: SPENCER,et al. (eds) Habitat of Hydrocarbons on the Norwegian Continental Shelf. Graham & Trotman, London, 259-274. HILLS, I. R. & WHITEHEAD,E. V. 1966. Triterpanes in optically active petroleum distillates. Nature, 209, 977-979. HORSFIELD,B., YORDY, K. L. & CRELLING,J. C. 1988. Determining the petroleum generating potential of coals using organic geochemistry and organic petrology. Organic Geochemistry, 13, 121-129.
7
HUNT, J. 1991. Generation of gas and oil from coal and other terrestrial organic matter. Organic Geochemistry, 17,673-680. KATZ, B. J., KELLEY, P. A., ROYLE, R. A. & JORJORIAN, T. 1991. Hydrocarbon products of coals as revealed by pyrolysis-gas chromatography. Organic Geochemistry, 17,711-722. LEYTHAEUSER,D. & POELCHAU,H. S. 1991. Expulsion of petroleum from type III kerogen source rocks in gaseous solution: modelling of solubility fractionation. In: ENGLAND, W. A. • FLEET, A. J. (eds) Petroleum Migration. Geological Society, London, Special Publication, 59, 33-46. MURCHISON, D. G. 1987. Recent advances in organic petrology and organic geochemistry: an overview with some reference to 'oil from coal'. In: SCOTT, A. C. (ed.) Coal and Coal-bearing Strata: Recent Advances. Geological Society, London, Special Publication, 32,257-302. NOBLE, R. A., Wu, C. H. & ATKINSON,C. D. 1991. Petroleum generation and migration from Talang Akar coals and shales offshore N.W. Java, Indonesia. Organic Geochemistry, 17, 363-374. PARNELL, J. 1988. Lacustrine petroleum source rocks in the Dinantian Oil Shale Group, Scotland: a review. In: FLEET, A. J., KELTS, K. & TALBOT, M. R. (eds) Lacustrine Petroleum Source Rocks. Geological Society, London, Special Publication, 40,205-217. PEPPER, A. S. 1991. Estimating the petroleum expulsion behaviour of source rocks: a novel quantiative approach. In: ENGLAND, W. A. 8Z FLEET, A. J. (eds) Petroleum Migration. Geological Society, London, Special Publication, 59, 9-32. PETERS, K. E. & MOLDOWAN, J. M. 1993. The biomarker guide: interpreting molecular fossils in petroleum and ancient sediments. Prentice Hall, New Jersey. PITTION, J. L. & GOUADIAN,J. 1985. Maturity studies of the Jurassic 'coal unit' in the three wells from the Haltenbanken area. In: THOMAS,B. M. et al. (eds) Petroleum Geochemistry in Exploration of the Norwegian Shelf. Graham & Trotman, London, 205-212. SCHOELL,M. 1983. Genetic characterization of natural gases. American Association of Petroleum Geologists Bulletin, 67, 2225-2238. ScoTT, A. C. (ed.) 1987. Coaland Coal-bearing Strata: Recent Advances. Geological Society, London, Special Publication, 32. 1989. Observations on the nature and origin of fusain. International Journal of Coal Geology, 12,443-475. SNOWDEN, L. R. 1991. Oil from type III organic matter; resinite revisited. Organic Geochemistry, 17,743-747. THOMAS, B. M. 1982. Land-plant source rocks for oil and their significance in Australian basins. Australian Petroleum Exploration Association Journal, 22, 164-178. , MOLLER-PEDERSEN, P., WHITAKER, M. F. & SHAW, N. D. 1985. Organic facies and hydro-
-
8
A. J. FLEET & A. C. SCOTT carbon distributions in the Norwegian North Sea. In: THOMAS, B. M. et al. (eds) Petroleum
Geochemistry in Exploration of the Norwegian Shelf. Graham & Trotman, London, 3-26. THOMPSON, S., MORLEY, R. J., BARNARD,P. C. & COOPER, B. S. 1985. Facies recognition of some Tertiary coals applied to prediction of oil source rock occurrence. Marine and Petroleum Geology, 2,288-297. TISSOT, B. P., PELET, R. & UNGERER, Ph. 1987. Thermal history of sedimentary basins,
maturation indices, and kinetics of oil and gas generation. American Association of Petroleum Geologists Bulletin, 71, 1445-1466. -& WELTE, D. H. 1984. Petroleum Formation and Occurrence (2nd edn). Springer Verlag. WEBER, K. J. & DAUKORU, E. 1975. Petroleum geology of the Niger delta. Ninth World
Petroleum Congress Proceedings, 2,209-221. WHITEMAN, A. 1982. Nigeria: its petroleum geology, resources and potential. Graham & Trotman, London.
Terrestrially sourced oils: where do they exist and what are our limits of knowledge? - a geochemical perspective T. G. POWELL
& C. J. B O R E H A M
Australian Geological Survey Organization, GPO Box 378, Canberra, A. C. T., Australia 2601 Abstract: There are relatively few well-documented cases where significant amounts of oil
(> 500 million barrels) have their origin from coals or their associated carbonaceous shales. The best-documented cases are from Australia and Indonesia and possibly the North Sea. The extensive debate as to whether coals as such or their associated carbonaceous shales are the source for oil has been essentially sterile: it has failed to take into account the heterogeneous nature of coals, the variety of their floral origins and the continuity of mineral dilution from pure coal to carbonaceous shale. Each case must be considered on its merits. The overall petrographic composition of coal is a poor guide to its petroleum potential. Sub-microscopic lipids of bacterial and also plant origin occur in vitrinite and inertinite and contribute significantly to petroleum potential. Suberinite and cutinite are the maj or source of waxy hydrocarbons. Hydrogen Indices measured on the immature versions of documented terrigenous source rocks are surprisingly low falling in the range 200-350 and occasionally lower. There is not a simple relationship between the elemental composition of terrigenous kerogen, the gross pyrolysis yield by Rock-Eval and the yield of normal hydrocarbons in pyrolysis gaschromatography. Our assessment of liquid potential in marginal source rocks, wherein most terrigenous sediments lie, is far from perfect. A combination of the mass balance approach in conjunction with pyrolysis studies has been shown to be the most effective way of demonstrating source potential and hydrocarbon products (gas vs. oil) expected from these marginal source rocks. There is clear evidence that significant amounts of gas are generated in the conventional oil window, and that high gas to oil ratios facilitate migration from the leaner source rocks. Comparative maturation studies on oils and gas-condensates show that they can be generated at similar maturation levels. The majority of terrigenous oils and condensates are paraffinic and the oils may have a high wax content. Exceptions occur when resin is a significant source in which case there is an increased proportion of naphthenic and aromatic hydrocarbons. Oils derived from coalbearing sequences have high pristane to phytane ratios (> 4.0); they have a high proportion of iso- and anteiso-alkanes derived from bacterial sources; hopanes predominate over steranes which are dominated by the Cz9 member. A variety of diterpenoid hydrocarbons may be found which relate to the nature of the parent flora and increasingly can be used for oil-source correlation in non-marine sequences.
H e d b e r g (1968) was the first to recognize that a significant n u m b e r of oils on a w o r l d w i d e basis had a terrestrial origin. His criteria w e r e b a s e d on the wax c o n t e n t and the paraffinic n a t u r e of the oils implying an origin f r o m the cuticular waxes of land plants. As an a p p r e c i a t i o n of the i m p o r t a n c e of lacustrine source rocks grew, it also b e c a m e a p p a r e n t that f r e s h w a t e r algae and possibly b a c t e r i a (Powell 1986 and r e f e r e n c e s t h e r e i n ) could also be a source for waxes and paraffinic h y d r o c a r b o n s . This, in effect, r e d u c e d the n u m b e r of examples w h e r e i n oils of waxy origin could be a t t r i b u t e d to a land plant source. Thus the n u m b e r of d o c u m e n t e d cases w h e r e significant v o l u m e s of oil can u n e q u i v o c a l l y be
d e m o n s t r a t e d to have an origin in coals or their associated c a r b o n a c e o u s shales r e m a i n s relatively few (Table 1). R e c e n t reviews ( M u r c h i s o n 1987; Powell & B o r e h a m 1991; B o r e h a m & Powell, in press) have discussed in detail the g e o c h e m i c a l and organic petrological issues r e l a t e d to the n a t u r e of, and recognition of, oil-prone source rocks in terrestrial s e q u e n c e s , the m o l e c u l a r c o m p o sition of n o n - m a r i n e oils and their precursors and the issue of migration. This p a p e r concentrates on a general o v e r v i e w of terrestrially d e r i v e d oils in the context of the e v a l u a t i o n of terrestrial s e q u e n c e s for p e t r o l e u m , concentrating on the e v i d e n c e available f r o m the case
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coal and Coal-bearing Strata as Oil-prone Source Rocks? Geological Society Special Publication No. 77, pp. 11-29.
11
12
T . G . POWELL & C. J. BOREHAM
Table 1. Terrigenous oil provinces - documented case studies Source rock
Country Basin/Province
Age
Hydrogen Indices
Reference
Late CretaceousTertiary Permian Jurassic Permian
200-350
Moore etal. 1992
150-300 200-400 150-250
Vincent etal. 1985
Eocene-Palaeocene
130-250
Issler & Snowdon 1990
Jurassic
?
Huang etal. 1991
Late Oligocene Middle Miocene
250-400 200-350
Noble etal. 1991 Durand & Oudin 1979
Late Cretaceous Tertiary
?
Czochanska etal. 1988
Late CretaceousTertiary
< 200
Bustin 1988
Jurassic
275
Forbes et al. 1991
Australia Gippsland Basin Cooper/Eromanga Basin Bowen/Surat Basin
Boreham (unpublished)
Canada Beaufort-Mackenzie Basin
China Turpan Basin
Indonesia Ardjuna Sub-basin, Java Mahakam DeltaKutei Basin, Kalimantan
New Zealand Taranaki Basin
Nigeria Niger Delta
Norway Haltenbanken area, N. Sea
Note: Hydrogen Indices are measured on immature samples.
histories and the uncertainties remaining in our understanding of generation of petroleum in land plant-dominated terrestrial sequences.
Examples of documented terrigenous petroleum systems Table 1 lists the best-documented cases of terrigenous petroleum systems in which coal or terrestrial organic matter in associated carbonaceous shales is deemed to be the source for oils and associated gas condensates. Examples where there is deemed to be a significant algal component to the source, such as occurs in many lacustrine sequences, are specifically excluded. The decision is an arbitrary one since there is a continuum between deposition of terrestrial organic matter in swamps through flood plain lakes to large lake systems developed in tectonic depressions (Powell 1986). At any point in this continuum algae may become a contributing source and, of course, in large lake systems may become the dominant source material (Powell 1986). This largely excludes the Chinese non-marine basins
where the preponderance of high-wax oil appears to originate from algal-dominated lacustrine source rocks, although there are references in the literature to coals being the source for some oils (e.g. Huang et al. 1988). The Turpan Basin may be a case in point (Huang et al. 1991). It was suggested early in the development of modern petroleum geochemistry in Australia that a significant proportion of Australian oil and gas reserves in the Gippsland, C o o p e r Eromanga and Bowen-Surat basins were derived from terrigenous source rocks in which coal was a significant if not the dominant component (Brooks & Smith 1967, 1969; Brooks et al. 1971; Powell & McKirdy 1972). The Gippsland Basin in Australia is by any standards a major hydrocarbon province with initial reserves estimated at 4.2 billion barrels of oil and condensate and 9.1 trillion cubic feet (TCF) of gas (Moore et al. 1992). Early in its exploration history it was recognized that the hydrocarbons were probably of terrigenous origin, but it is only more recently that there has been widespread acceptance of the terrigenous nature of the Gippsland Basin source rocks. They consist of a mixture of coals
TERRESTRIALLY SOURCED OILS and carbonaceous shales of late Cretaceous age deposited in a lower delta plain setting (Moore et al. 1992) (Table 1). The organic matter was derived from cool temperate coniferous forest (Shanmugan 1985). The Taranaki Basin in New Zealand is also oil and gas bearing with many characteristics similar to the Gippsland Basin. Again coals and associated shales of late Cretaceous to early Tertiary age are the identified source rocks (Czochanska et al. 1988; Weston et al. 1989). The Cooper-Eromanga Basin sequence in central Australia is another classic sequence in which the terrigenous nature of the source was suggested early in its exploration (Brooks et al. 1971; Powell & McKirdy 1972). The Permian fluviatile source sequence is dominated by coals and is the predominant source for the oil and gas-condensate in Permian and Triassic reservoirs (Vincent et al. 1985). The source rocks were deposited in a cold temperate climate in an inter-montane setting (Hobday 1987). The overlying Eromanga Basin is oil prone and contains the majority of the recoverable oil reserves in the basin complex. Perhydrous Jurassic coals were deposited in a fluviatile-lacustrine environment under warm humid conditions and have the potential to generate oil. However, they are thin and lack lateral continuity. Oil source rock correlations based on plant biomarkers (Alexander et al. 1988) and volumetric and migration considerations (Vincent et al. 1985) demonstrate that most oils reservoired in Jurassic and older rocks originate from the Permian coal sequences. Some contribution from Jurassic sources has been recognized in a few instances (Alexander et al. 1988). Small oil pools occur in Cretaceous lacustrine sands and have been demonstrated to originate in adjacent lacustrine carbonaceous shales containing terrestrial organic matter (Powell et al. 1989). The Bowen-Surat basins in southern Queensland have only minor reserves, but again the oils originate from Permian terrigenous source rocks showing some degree of marine influence (Philp & Gilbert 1986). A very large proportion of oil in Indonesia either has a lacustrine (algal) or terrigenous source (Robinson 1987). In the case of Sumatran fields it is often difficult to determine the relative importance of a purely terrigenous source (Katz & Mertani 1989). However, there are two published cases of major reserves associated with terrigenous sources in Indonesia. In the Mahakam Delta (Kutei Basin) Kalimantan, terrigenous source rocks of Miocene Age are located in the delta plain facies and consist of interbedded coals and shales and give rise to
13
major oil and gas reserves (Durand & Oudin 1979). Similarly in the Ardjuna Sub-basin of NW Java, coals and associated carbonaceous shales in the Talang Akar Formation of late Oligocene age have been documented as the source for the very significant reserves in that basin (Gordon 1985; Horsefield et al. 1988; Pramono et al. 1990; Noble et al. 1991). In both cases the organic matter is derived from deltaic facies in a tropical setting. The Tertiary section of the Niger Delta has long been considered to be an example of a terrigenous oil province (Hedberg 1968; Reed 1969). No identified rich source rocks have been identified - aUochthonous terrigenous organic matter is widely dispersed in the finer-grained facies. The highest concentration of organic matter and the most oil-prone organic matter occurs in the floodplain deposits. The poor quality of the source rocks has been compensated by their great volume (Bustin 1988). The Beaufort-Mackenzie Basin, Canada is a significant hydrocarbon province and, like the Niger Delta, consists largely of a Tertiary Delta in which allochthonous terrigenous organic matter appears to be the source (Snowdon 1978). Unlike the other examples, resin compounds appear to be a major component of the source material and the oils are naphthenicaromatic rather than paraffinic. Oil-source correlations have shown that the EocenePalaeocene Richards Formation is the source (Brooks 1986a, b). Once again the source organic matter is derived from coniferous species growing at high latitudes. This particular example is discussed in more detail below. It has been proposed that Jurassic coals are a major source for liquid hydrocarbons in the Haltenbanken region in the Norwegian Sector of the North Sea and recent volumetric studies have suggested that that source is adequate to account for the known reserves (Forbes et al. 1991). Other examples are emerging of significant oil resources being derived from autochthonous and/or allochthonous terrestrial organic matter, e.g. Brunei (Scherer & Hitam 1992) and Sarawak (Woodroof & Carr 1992), but neither have been extensively documented in the literature.
Implications of depositionai environment and flora for characteristics of source rocks in terrigenous sequences Over the last 10 years there has been considerable debate as to whether coals as such can be a source for oil or whether it is the dispersed
14
T. G. POWELL & C. J. BOREHAM
organic matter in associated shales which is the primary source for petroleum in coal-bearing sequences. The authors' view is that this debate has been essentially sterile, because it has failed to take into account the heterogeneous nature of coals and the variety of depositional conditions under which they are formed and the continuum in mineral dilution from pure coal to carbonaceous shale. Much of our knowledge of the physicochemical structure of coals is essentially confined to Euramerican Carboniferous coals which are economically important as thermal coals or for coke manufacture. By definition these coals are of low ash - otherwise they would not be economic - and they are derived from a particular flora (lycopsids) that dominated in the Carboniferous (Collinson & Scott 1987). They were probably deposited in a particular depositional environment to account for their low ash content (McCabe 1984). It is noteworthy that none of our case histories (Table 1) relate to Carboniferous coals. Indeed, with the notable exception of the Permian coals of the Cooper Basin, all examples of non-marine oils are derived from coal-bearing sequences which are much younger and from source materials derived from more advanced plants than those represented in the Carboniferous. In addition, the range of coal depositional environments represented by the case histories, both in terms of climate and depositional setting, is far broader than is represented by the Carboniferous coals. A particular case is the Permian coals of the Cooper Basin, Australia which were deposited in a cool temperate setting far removed from marine influence (Hobday 1987). Differing flora and depositional settings impose their own structure on coals seams. Ash contents are continuously variable from carbonaceous shale through to pure coals which may be allochthonous or autochthonous in origin. From these observations it is clear that the particular model of coal structure and physico-chemical properties represented by Euramerican Carboniferous coals is an inappropriate basis to argue the general merits of in situ generation and migration of oil from coal. Indeed many nonmarine source rocks identified as 'coals' would not be economically exploitable coals from the standpoint of their ash content. Reiteration of some key quotes and conclusions from a paper by McCabe (1984) on the depositional environments of coal illustrate the points made above: 'Care should be exercised in using coal as a paleoclimatic indicator. It is certainly not an indicator of warm temperatures or a high rainfall.'
'Swamps vary in their morphology, vegetation type, degree and type of degradation of organic matter and the input of clastic sediment. All these factors are important in determining coal's quality. Raised swamps and floating peats, may be important in the formation of low ash coals.' 'The composition of peat/coal is as variable as any other sedimentary rock. Facies analysis of coal seams can provide important indications relative to the nature of the depositional environment. The composition of a coal is also important in determining its utilisation.' 'Peat can accumulate in any depositional setting. The importance of deltas as sites for peat accumulation has probably been greatly over-emphasized in the literature.' Quite apart from the wide variety of depositional environments in which coal or carbonaceous sediments can accumulate, the original peat-forming vegetation has undergone major changes through the geological record (Collinson & Scott 1987). Particular features which affect peat formation and composition include (Collinson & Scott 1987): • nature and timing of leaf fall; • proportions of lignin and cellulose; • presence/absence of open water and marginal aquatic plants; • overall community diversity and the interactions between members of the community. These factors have clearly changed in response to plant evolution. Indeed the evolution of the depositional conditions within the coal-forming environment will in itself inevitably lead to changes in the plant community and affect the composition of peats that are formed. A very significant factor that must be taken into account in considering the variability in composition of potential coal source rocks is the dramatic degradation and reduction in plant biomass that accompanies peat accumulation (Clymo 1987). The mass of peat ultimately preserved represents only a very small fraction (a few percent) of the plant mass originally deposited. This volume reduction occurs as a result of dehydration, extensive microbiological activity and ongoing chemical reactions not mediated by biological activity. The resulting organic accumulation consists of the remnant plant tissues impregnated with, and set in, a matrix of homogenized decay products. Given the variation in depositional conditions of peat preservation this process is a source of very significant heterogeneity in the preserved coals and carbonaceous shales. A direct implication of
TERRESTRIALLY SOURCED OILS this process is that bacterial organic matter would be expected to be a very significant component of the preserved organic material. Estimates of bacterial biomass in peats are of the order of 15% (Stach et al. 1982). It is clear from this discussion that the physicochemical properties of organic-rich shales, high ash coals and pure coals will be extremely variable and that in turn their ability to generate hydrocarbons and yield them for migration will be extremely variable and each case must be considered on its merits. Furthermore it has been demonstrated in several of the case examples cited in Table 1 that the overall geochemical composition of coals is identical with that of the organic matter in the associated shales (Durand & Oudin 1979; Vincent et al. 1985). The question of the ultimate source depends on arguments related to the efficiency of migration out of shales compared with coals. The relative masses of organic matter in coal compared with shale is also a factor in attributing source (Moore et al. 1992). Terrigenous organic matter as a source for hydrocarbons
Coal consists of a complex mixture of microscopic components, termed macerals, which are distinguishable one from another on the basis of their morphology and physical properties. Three major maceral groups are identifiable (Table 2). In most instances coal macerals are derived Table 2. Coal maceral nomenclature (after Stach et al. 1982) and inferred hydrocarbon potential
Liptinite (Exinite)
Vitrinite
Inertinite
Maceral Sporinite Cutinite Resinite Suberinite Lipodetrinite Exsudatinite Telinite Collinite Vitrodetrinite Semifusinite Fusinite Macrinite Sclerotinite Inertodetrinite Micrinite
ultimately from extant plant components. The inertinite and vitrinite macerals are derived predominantly from the structural parts of plants. Sporinite, cutinite, resinite and suberinite are the primary macerals of the liptinite group and are derived from spores, cutin, resin and suberin respectively. The various forms of alginite (algae), which are the predominant maceral in most marine and lacustrine source rocks, are absent or only occur in trace amounts in most coals with the notable exception of boghead coal. The elemental composition of isolated macerals reflects to some degree the original composition of the plant precursors (Fig. la) although they are substantially modified during the biochemical stages and early thermal stages of coalification (see above). Lipinite is richer in hydrogen than vitrinite which in turn is richer than inertinite (van Krevelen 1961). The maceral groups therefore occupy different pathways or coalification tracks with thermal maturation. These pathways are similar to the pathways for Types I - I V organic matter defined for sedimentary rocks in general (Tissot et al. 1974; Harwood 1977). Geochemical measurements and oil potential
Overall c o m p o s i t i o n
Maceral group
15
Hydrocarbon potential Hydrogen-rich; precursors for oil and gas
Hydrogen-poor, mainly precursors forgas Very hydrogenpoor; largely inert, but may produce gas at very late stages of maturation
Since hydrogen is the limiting factor in the generation of hydrocarbons from sedimentary organic matter, theoretical considerations suggest that the hydrocarbon potential of terrigenous organic matter may be explained in terms of the relative proportions of hydrogenpoor and hydrogen-rich components. Thus the associations of the microscopic components of coal with particular hydrogen content forms an empirical basis for the use of petrography in source rock assessment. However, quantitative knowledge of the hydrocarbon-generating capacity of individual macerals is lacking, but observations on organic matter types in general (Powell 1978, 1988a) suggest that in order for a source rock to be effective, 10-20% of its organic matter must equate with Type I organic matter, or 20-30% must equate with Type II organic matter. The bulk atomic H/C ratios would therefore fall in the range 0.8-0.9 (Powell 1988a) or Hydrogen Indices in Rock-Eval analysis above 220-300 before any oil expulsion is implied (Powel11988a; Hunt 1991; Noble et al. 1991). There are, however, many uncertainties in translating these percentages to proportions of maceral components to enable an organic petrographic means of quantifying oil source potential (see below).
16
T. G. POWELL & C. J. BOREHAM 2.0
A
2.0-
B
evolution path of Type I
RESIN ii
oE 1.0
~ :
1.s
T ~ ~
II !
Mean evolution path of
.
.
.
.
-
1.o -
/j
....
/ //
0.5
~
t
1
/
j
.i"
-
le evolution path o f Type IV
O.5-
oi,
;.2
o'.3
o:,
O/C (Atomic Ratio)
o'.~ o o
o'.t
o12
o'.3
oi,
01s
O/C ( Atomic Ratio)
Fig. 1. Comparison of kerogen types, evolution paths and petrographic components of coal based on atomic ratios. (A) Types I-III (Tissot et al. 1974); Type IV (Harwood 1977). Shaded area represents field of kerogen composition. (B) Petrographic components of coal (van Krevelen 1961).
More than any other type of source rock, the critical issue for the evaluation of potential terrigenous source rocks is the threshold concentration of hydrocarbons required to allow migration out of the source rock. This is not a simple question since it relates to rate of generation, gas-oil ratios and phase behaviour. These can only be evaluated properly by quantitative modelling of hydrocarbon generation and migration in each particular case. However, it is possible to make some general observations related to these theoretical concepts. Firstly, the range of Hydrogen Indices derived from Rock-Eval analysis of the immature source rocks in the case histories (Table 1) varies considerably. They range from 150 to 300 for coals and shales in the Cooper Basin, Niger Dalta, Mahakam Delta, and Beaufort-Mackenzie Basin through to over 400 for some coals and shales in the Ardjuna Sub-basin. Values around 200 appear to predominate in source rocks of the Niger Delta, Mahakam Delta and Beaufort-
Mackenzie Basin. Any source rock model must account for an apparent wide variation in labile kerogen content and for the low values observed, although it is not clear if the effect of the mineral matrix has been taken into effect in the Rock-Eval measurements in all cases (cf. Powell et al. 1989). Failure to do so may account for some low Hydrogen Indices reported for shales. Powell et al. (1991) have clearly shown that the correlation between Hydrogen Index and elemental composition of kerogen breaks down for coals with atomic H/C ratios in the range 0.8-1.0. They also have clearly shown that for a given Hydrogen Index there is a wide range in the pyrolysable yield of straight chain hydrocarbons and vice versa. Since most non-marine oils are paraffinic (greater than about 50% paraffins, Tissot & Welte 1984) the yield of paraffins on pyrolysis is in the most non-marine source rocks a suitable indicator of their ability to yield oil for migration. An exception is where
TERRESTRIALLY SOURCED OILS resin sources predominate (Snowdon 1980). It follows from these observations that the precursors of non-marine oil are set in a matrix which itself has varying degrees of lability under pyrolysis conditions. This labile matrix apparently contributes very little to hydrocarbon generation under natural conditions. This concept is perfectly understandable in that terrigenous organic matter, which is for the most part aromatic in structure, produces amongst the most hydrogen-rich oils found in nature. These observations re-emphasize the heterogeneous nature of coal and place limitations on the use of elemental analysis and the empirical interpret-. ation of bulk pyrolysis results in the determination of source potential in marginal quality terrigenous source rocks.
Petrographic considerations Thomas (1982) has suggested that differences in hydrocarbon source potential between Palaeozoic, Mesozoic and Tertiary coals may be related to the evolution of successive floras and climatic controls. Our studies (Powell et al. 1991) of Australian coals and carbonaceous shales from different ages and environments have shown that they have overlapping compositions with samples of a particular age tending to one extreme or another. They could be ordered with decreasing oil potential as follows Jurassic> Cretaceous = Upper Cretaceous - Tertiary > Permian=Triassic. Whereas this sequence follows the broad trends in petrographic composition (i.e. relative proportions of vitrinite, liptinite and inertinite) of Australian coals (Cook 1975; Thomas 1982) there is only a poor correlation between geochemical and petrographic compositions at the maceral group level on individual samples. Analysis of the petrographic and geochemical data (elemental ratios, Rock-Eval and quantiative pyrolysis gaschromatography) showed a better relationships between maceral composition and the ability to yield normal hydrocarbons on pyrolysis. The principal results are as follows: • Liptinite-poor (<10%) samples may yield significant amounts of normal hydrocarbons but typically normal hydrocarbons with carbon numbers below C17 predominate. • Liptinite-rich samples with abundant sporinite give only low to intermediate yields of normal hydrocarbons and these are predominantly of lower molecular weight (below
C17).
• Sporinite appears not to be a significant source of normal hydrocarbons.
17
• Suberinite is associated with high yields of waxy hydrocarbons, but some samples with high yields do not contain this maceral. • Cutinite is associated with pyrolysates with a high proportion of waxy hydrocarbons but there was no obvious correlation with high yield. • No clear association exists between high yields of waxy hydrocarbons and liptinite or individual liptinite macerals in many Permian and Tertiary samples. It is now clear that the vitrinite and inertinite which are the predominant coal macerals can contain sub-microscopic amounts of lipidic material that are not readily recognized under the petrographic microscope except as general background fluorescence of varying degrees of intensity. The electron microscopic evidence of the existence of sub-microscopic bacterial or possibly algal (Taylor et al. 1988) remains in inertinites in Cooper Basin coals is one explanation for their ability to yield aliphatic hydrocarbons on pyrolysis and to yield light oils and gas-condensates (Table 1). As stated above, micro-organisms may contribute 10-15% of the biomass in peats (Stach et al. 1982). Cork tissue is the precursor for suberinite and is associated with woody plant tissues which normally give rise to vitrinite and/or inertinite. If cork tissues are in low concentration and extensive homogenization has occurred during the biochemical stages of coalification, suberinite may not be readily recognized in coal as a discrete maceral. Similarly resin may be dispersed and may not be recognized as such, although it may cause vitrinite to fluoresce. Vitrinite may, therefore, contain variable amounts of hydrocarbon precursors. This may account for some differences in the yield and distribution of normal hydrocarbons from vitrinite (see also Collinson et al., this volume). Clearly petrographic observations can only be interpreted in a broad qualitative sense for the purposes of assessing petroleum potential. Detailed studies and refinements in classification of macerals and maceral assembleges are of assistance but the basic ambiguity in the interpretation of the results will always remain (Boreham & Powell in press).
The role o f resinite The proposal that plant resin could be a major source for petroleum hydrocarbons at relatively low levels of maturation arose out of studies conducted in the Beaufort-Mackenzie Basin in the late 1970s and early 1980s (Snowdon 1978;
18
T. G. POWELL & C. J. BOREHAM
Snowdon & Powell 1982). Snowdon (1991) has recently summarized the evidence for the model. Briefly, the resinite hypothesis evolved as a means of explaining a number of phenomena observed in the Tertiary deltaic sediments of the Beaufort-Mackenzie Basin. These observations included the following (after Snowdon 1991 with additions): • the occurrence of crude oils and condensates with low maturity chemical properties including biomarker ratios, odd/even n-alkane predominance, high pristane/ n-Ca7 ratios and low gasoline range maturity ratios consistent with maturity levels in adjacent sediments; • crude oils of unusual naphthenic-aromatic composition, but with low asphaltene contents; low amounts of wax, but significant amounts of discrete tricyclic diterpenoid hydrocarbon compounds and unresolved hydrocarbons chromatographically in the nC13 to nC2o range (sesquiterpane and diterpane humps) and indicators of a terrestrial source including the sterane distribution and specific terpane compounds such as oleanane and lupanes and high pristane to phytane ratios; these biomarkers being the basis for correlation to deltaic shales containing Type III organic matter of terrigenous origin; • a lower gas/oil ratio for discoveries than is typically observed for other Tertiary deltas; • up to 10 km of Tertiary clastic deltaic sediments which contain essentially only terrestrial (Type III or higher land plant) organic debris; • unusually low maturation gradients and commensurate low absolute maturity (0.7% vitrinite reflectance) at depths down to 5 km through the nearshore and offshore sections. A more recent study of the hydrocarbon generation kinetics and thermal modelling in the Beaufort-Mackenzie Basin (Issler & Snowdon 1990) has shown rapid burial in the PlioPleistocene in this region and that the hydrocarbon generation model is consistent with hydrocarbon generation at conventional maturities, but that the distribution of activation energies for the resinite-rich Type III organic matter was weighted towards the lower activation energies than for classical Type II organic matter. The apparent 'lack of maturity' of the generated hydrocarbons is apparently a consequence of late rapid heating experienced by the source rocks and the differing kinetic controls on maturation indicators as distinct from hydrocarbon generation (Tissot et al. 1987).
Nonetheless the unusual composition of the Beaufort-Mackenzie oils, the presence of resinderived biomarkers and the emerging literature on resin-derived biomarkers in terrigenous oils from elsewhere, all point to plant resins as being a significant source for petroleum hydrocarbons. Only in the Beaufort-Mackenzie Basin are the resin compounds in sufficient concentration in the source rock to generate naphthenic aromatic oils in contrast to the predominantly paraffinic nature of most terrigenous oils and condensates. Gas p o t e n t i a l
There is relatively little information on the quantitative aspects of gas generation as a function of maturation in coals and carbonaceous shales. Monnier et al. (1983) have accumulated a mass of cutting gas data that clearly show significant gas generation occurs from Type III kerogens throughout the mature zone (vitrinite reflectance range 0.7-1.3% Ro) and that the gas to oil ratios were significantly higher than for hydrocarbons derived from Type II kerogen over the same maturation range. Harwood (1977) found similar results based on laboratory pyrolysis experiments. More recent experimental work by Higgs (1986) indicate that significant gas generation from terrigenous organic matter commences in the conventional oil window. Higgs' work demonstrated that peak gas generation from coals occurs at c. 1.0% vitrinite reflectance and that the intensity of gas generation at any particular level of maturation can vary with coal type and age. In the case of the Jurassic Walloon Coal Measures, Australia the remaining hydrocarbon potential of kerogen has been examined as a function of increasing maturation (Fig. 2) (Boreham & Powell, 1991; Powell & Boreham 1991). There is little change in the gas fraction of the remaining hydrocarbon potential of the kerogen up to 0.9% reflectance whereupon there is a rapid increase. These results show that the Walloon Coal Measures generate oil and gas in a fixed ratio through the main phase of oil generation, and this reflects the distribution of gas and oil precursors in the immature kerogen. If oil was preferentially generated, the gas fraction of the remaining hydrocarbon potential should increase. Only at relatively high rank (approximately 1.0% reflectance) when the liquid precursors are being exhausted or extensive cracking occurs does the gas to oil ratio of the generated hydrocarbons change. In contrast to the Jurassic Walloon Coals, pyrolysates of low rank Permian coals initially have a higher gas fraction and this remains
TERRESTRIALLY SOURCED OILS 1.0-
0.9-
/////
0,5,~ 0
0
0
0.8-
S°
0.7-
I I / o---o4~ato"aw
/Porma
P
Selected Permian Co
//..
,on./
/ Walloon
....
/
110
'
Coa, M . . . . . .
J . . . . . ,0
0.8-
0.5
016
'
114
'
;'6
'
212
Vitrinite Reflectance % R0
Fig. 2. Evolution of Gas Fraction (Ca-Cs/Ca-C27) from pyrolysis gas chromatography of selected terrigenous kerogens from Walloon Coal Measures, Clarence-Moreton Basin and Patchawarra Formation in the Cooper Basin with maturation (Powell & Boreham 1991). constant up to 0.7% vitrinite reflectance (Powell & Boreham 1991) (Fig. 2). At this point the liquid hydrocarbon potential is relatively exhausted and the gas fraction climbs rapidly as hydrocarbon generation continues. The high oil to gas ratios observed in the mature zone for terrigenous source rocks may well provide for an efficient early expulsion mechanism for gas and condensate in the conventional oil window. A practical
approach
terrigenous
source
to assessment
of
potential
Source rock evaluation
It is clear from the discussion above that care must be exercised in applying routine interpretations of conventional petrographic and geochemical source rock analyses. Petrographic analyses remain a useful tool for screening purposes but particular attention must be paid to the type of liptinite maceral present. Particularly in samples with relatively low liptinite contents the presence of background fluorescence may indicate hydrocarbon potential in vitrinite and inertinite. Rock-Eval analysis is an essential prerequisite for source rock analysis in non-marine sequences despite some difficulties in assessing the significance of the results. At
19
Hydrogen Index values above 300 and particularly above 350 the evidence to date suggests that conventional interpretations of source rock quality are appropriate in terrigenous source rocks. Below Rock-Eval values of 300 and extending down to 150 one would be tempted to be dismissive of oil source potential. The case histories identified in Table 1 clearly show that this may be misleading, particularly where large volumes of mudrock or coal are involved. In these instances it is clearly well worth undertaking investigations by quantiative pyrolysis gas chromatography (Larter & Sentfle 1985; Horsfield et al. 1988; Horsfield 1989; Powell et al. 1991) to assess the potential for generation of the normal paraffins, which form such a large proportion of non-marine oils, and to obtain estimates of possible gas to oil ratios based upon the relative distribution of light hydrocarbons to heavier hydrocarbons at different pyrolysis temperatures (Horsfield 1989). The authors do not consider it practical to introduce nonstandard methods of assessment as advocated by Smith et al. (1987) only to assess the potential of non-marine source rocks. The Permian coals and carbonaceous shales of the Cooper Basin (Table 1) are probably close to the lower limit of an effective terrigenous source rock. The yields of normal hydrocarbons (> C7) of a substantial proportion of the samples studied fall in the range 10-20mg per gram organic carbon (Powell et al. 1991). Although assessment of normal hydrocarbon yield is an appropriate measure of source potential for paraffinic oils, it would not be effective for source rocks in which resin is the predominant source material since the dominant hydrocarbon products would be naphthenic and aromatic in composition. A judicious combination of Rock-Eval pyrolysis and quantitative pyrolysis gas chromatography makes it possible to measure the hydrocarbon potential of most terrigenous source rocks and to infer the proportion of that potential which is attributable to oil and gas. However, these procedures do not quantify the potential for formation of commercial petroleum pools: here we need a mass balance approach described below. Furthermore, there is an increasing weight of observational evidence that the gas to oil ratio of generated hydrocarbons in combination with the temperature history is a major determinant of the timing and expulsion efficiency of petroleum from coals and associated carbonaceous shales. Modelling in each specific case is the only satisfactory way of addressing these issues (e.g. Noble et al. 1991).
20
T. G. POWELL & C. J. BOREHAM
M a s s balance a p p r o a c h to p e t r o l e u m potential Obviously while there is scope for future developments in empirical approaches to delineating source richness and maturation, only a mass balance approach is likely to yield a quantitative assessment of petroleum plays. The approach adopted here is that adopted by Cooles et al. (1986). Kerogen is classified into three types. Labile kerogen is capable of producing oil and gas, refractory kerogen generates primarily gas, while inert kerogen has no hydrocarbon generative potential. To represent petroleum generation and migration, a simple algebraic scheme based on organic carbon mass balance was devised by Cooles et al. (1986). A carbon deficit in a mature source rock compared with its immature analogue, implies that the carbon has been lost by migration from the source rock. The carbon content of an immature source rock is divided between free-oil carbon, reactive carbon (labile plus refractory kerogen) and inert carbon. The free oil content can be determined either from measurement of RockEval S1 or from solvent extractable organic matter and corrected for heavy-end and lightend hydrocarbon losses respectively (Cooles et al. 1986). The reactive carbon can be determined from Rock-Eval $2 while the inert carbon is the difference between TOC content and the sum of the free oil and the reactive carbon. A similar proportioning of carbon is obtained from the mature correlative. However, in order to compare the mature and immature correlatives and calculate the volumes of petroleum generated and migrated, it is assumed that there is a common starting point for the mature and the immature rock and there is no loss or additions to the inert carbon content as the sediment matures. Since sediments have variable TOC contents, the inert carbon content in the mature analogue is normalized to the inert carbon content of the immature analogue. Applying this normalization factor to correct free-oil and reactive carbon in the immature analogue enables the calculation of the Petroleum Generation Index (PGI) and the Petroleum Expulsion Efficiency (PEE) which are defined as follows: PGI =
Petroleum generated + Initial petroleum Total petroleum potential
PEE =
Petroleum expelled Petroleum generated + Initial petroleum
PGI defines the proportion of the total petroleum potential (free oil plus reactive
kerogen in the immature rock) that has been generated from the source rock at various stages of maturity. PEE is a measure of the extent of expulsion and relates to the difference between the calculated amount of petroleum that is liberated at a given PGI value and the amount of petroleum actually retained in the source rock. The data for the Westphalian coals only cover the maturation range equivalent to vitrinite reflectance levels of 1.0-3.0% R0. These data show that gas generation from refractory kerogen is essentially complete by a maturation level corresponding to 2.0% R0 vitrinite reflectance. This observation is consistent with the occurrence of coal-sourced gas in the Southern North Sea where the gas pools are associated with vitrinite reflectance levels of between 1 and 2% R0 in the Carboniferous (Barnard & Cooper 1983). Laboratory studies on US Carboniferous coals (Hiss 1986) indicate that significant gas generation occurs at lower maturation levels (<1% Ro vitrinite reflectance). Petroleum generation also commences at lower maturation levels in the Patchawarra coals from the Cooper Basin and the Walloon Coals from the Clarence-Moreton Basin. From the plot of PGI versus vitrinite reflectance (Fig. 3) it is clear that the Walloon Coals begin to generate hydrocarbons at 0.75% vitrinite reflectance level. The efficiency of petroleum expulsion from these is not quantitatively significant during the initial stages of hydrocarbon generation, but increases rapidly as the PGI increases to 0.5. The bulk of oil generation occurs in the reflectance range 0.81.0% vitrinite reflectance. Petroleum Expulsion Efficiencies are greater than 60% when the PGI increases to above 0.4 and are at a maximum between 0.9 and 1.0%. This conclusion is broadly consistent with the maturation levels observed for migration of oil from coals in the Gippsland Basin (Stainforth 1984; Burns et al. 1984; Powell & Boreham 1991) and from the Talang Akar coals in the Ardjuna Basin (Noble et al. 1991) and the coals and shales of the Mahakam Delta (Durand & Oudin 1979). By comparison, the humic coals of the Patchawarra Formation in the Cooper Basin have a lower liquid potential (Powell et al. 1991). Hydrocarbon generation patterns are initially similar to the Walloon Coals, but at a PGI ratio of about 0.4, the petroleum generation curve trends to the gas generation line defined by the Westphalian coals. However, at the same PGI, the PEE and gas to oil ratios of the Patchawarra coals are consistently higher than for the Walloon Coals. In this case, oil generation and expulsion occurs relatively early in the oil
TERRESTRIALLY SOURCED OILS 3.0- (a)
From the examples above it can be seen that a mass balance approach avoids many of the problems associated with empirical interpretation of classical techniques and leads to a quantitative assessment of oil and gas potential. Care must be taken to adequately characterize the source rock interval particularly in respect of gas-to-oil ratios derived from pyrolysis data. Also this mass-balance scheme does not consider losses from the overall pool of migratable hydrocarbons through thermal coking of retained hydrocarbons which increases the proportion of inert kerogen in the process. During the main phase of oil generation and labile kerogen cracking, coking reactions are of minor importance (Ungerer et al. 1988). The quoted expulsion efficiencies remain valid in the lower maturity range, but at high maturation levels the expulsion efficiencies may be overestimated.
(3 © Westphalian coals, E~ro7
2.0-
~ 1.0- O ~
0 I 0
0
= 0.1
xX.- ~ ° .~alloon
~ Patchawarrxa F ~ x "
~ 0,2
i 0.3
i 0.4
/I~"
i 0.5
i 0.6
i 0.7
~ 0.8
Coals
i 0.9
Petroleum Generation Index
(b)
1.0-
21
0.90.8~
~u.i "
~ e
Composition
0.70.6-
O.5-
0.4-
~" 0.30.2010
o
01,
i
o.2
o13
0:4
o'.5
i
i
o.~ oI~ o18 o19 ,.o
Petroleum Generation Index
Fig. 3. Evolution of Petroleum Generation Index and Petroleum Expulsion Efficiency with maturation for Walloon Coal Measures, Clarence-Moreton Basin and Patchawarra Formation in the Cooper Basin (Powell & Boreham 1991). Data for Westphalian coals of Europe are from Cooles et al. 1986.
window as measured against the vitrinite reflectance scale and the reservoir of liquid potential is rapidly depleted. The values for the Methyl Phenanthrene Index measured on samples from the Cooper Eromanga Basin reveal a wide maturation range for hydrocarbon generation/ migration (0.6-0.95 calculated vitrinite reflectance) (Alexander et al. 1988; Boreham et al. 1988; Tupper & Burkhardt 1990) for both oils and condensates. Here a gas-phase migration may be appropriate. The high gas-to-oil ratios of the generated petroleum from humic organic matter promotes early migration and may result in phase separation to form oil and gas pools in response to local conditions.
o f oils
Figure 4 illustrates the overall composition of terrestrial oils from Australia and Canada. All are low in asphaltenes and in nitrogen, sulphur and oxygen (NSO) compounds. Hedberg (1968) formalized the concept that high-wax oil was derived from terrestrial organic matter and defined oil with an excess of 5% C22+ paraffins as a high-wax oil. However, an extremely wide variation in the wax content is common in oils of terrigenous origin; this variation can be attributed to any of several geochemical processes (Powell 1988a). The nature of the source matter determines the initial composition of the oil, but the richness of the source or gas to oil potential and the migration history will determine the amount of was in the reservoired oil. If the content of liquid is relatively low, the reservoired petroleum will be light oil dominated by low boiling hydrocarbon and gas-condensate. The preferential migration of light ends from terrigenous sources has been observed (Leythaeuser et al. 1984). In the Mahakam Delta, Indonesia, gas and light hydrocarbons are the preferred major migration products from overpressured zones, even though the source rock potential is favourable for crude oil migration. Where the source rock occurs above the overpressured zone, crude oil is the migration product (Durand & Oudin 1979). Oils derived from terrestrial organic matter are commonly dominated by paraffinic hydrocarbons (Fig. 5 - Fortescue). Notable exceptions are the naphthenic-aromatic oils and condensates found in the Beaufort-Mackenzie Basin (Snowdon & Powell 1979) (Fig. 5). Abundant
22
T. G. POWELL & C. J. B O R E H A M
A. Australian Oils Eromanga Basin Cooper Basin Gippsland Basin Perth Basin
B. Canadian Oils Beaufort- Mackenzie Basin
/~
~,..
10 80 10
70/
x°
x
xxx
7
,/
/~
,0 :o " x 50/
3[ o
OO
ou
i, °
O O
/
o
O O
Oo
60
o \
6O
50
//~,~\ ~ ~k~k ~
C. Canadian Oils Scotian Shelf
/
Labrador Shelf
/ /
7,
/,0 /
60
~ //
/ -
~ro~a,,cs
5O
o
Lowwaxoils(<5%
wax)
•
High wax oils ( > 5% wax)
....,,~
/50/ /
Fig. 4. Compositional variation (> 210°C fraction) among non-marine oils from Australia and Canada (after Powell 1988a).
TERRESTRIALLY SOURCED OILS FORTESCUE FIELD GIPPSLAND BASIN, AUSTRALIA / n C25 .C2o\ /nClO
I
I
nC15~ PRISTANE
, k]l
J
/he30
CRUDE OIL BEAUFORT-MACKENZIE BAS N,CANADA
o
nC25
-,~-~.
Fig. 5. Whole oil chromatograms of selected oils from Gippsland Basin and Beaufort-Mackenzie Basin (after Powell 1988). Note contrast between paraffinic oil and naphthenic-aromatic oil and high pristane to phytane ratios in both cases.
pristane and high pristane to phytane ratios are characteristic of most terrigenous oils and are reliable indicators of terrigenous oils. Bituminous coals and oils originating from higher plants have high Pr/Ph ratios (5-10) in comparison with oils and sediments containing marine organic matter (<1-3) (Brooks et al. 1969; Powell & McKirdy 1973). Further Pr/Ph ratios increase from 1-2 in brown coals to a maximum of 4-10 in bituminous coals, and significant pristane generation occurs at the threshold of oil generation. Similar changes are seen in sediments containing terrigenous organic matter (Powell et al. 1978; Snowdon 1978). Although the ratio of the two isoprenoids has been linked to the redox depositional history of the parent organic matter, the Pr/Ph cannot be used as a palaeoenvironmental indicator at low maturity level (Brooks et al. 1969; Powell & McKirdy 1973; Volkman and Maxwell 1986). At higher maturation levels, high Pr/Ph ratios
23
(>3.0) are reliable indicators of input of terrestrial organic matter (Powell 1988b). However, substantial variations can occur in the isoprenoid to n-alkane ratio independently of maturation control. Amongst terrigenous oils there is considerable variation in pristane to nC~7 ratio at an essentially constant maturation level. Lijmbach (1975) has offered an explanation for this observation. In environments with extensive bacterial activity the proportion of the preserved biomass derived from photosynthetic organisms decreases relative to the proportion of bacterial biomass preserved. This is reflected in a lower proportion of acyclic isoprenoids in the saturate fractions of some terrigenous sediments and oils. Support for the role of microbial processes in the formation of source organic matter for many terrigenous oils is the presence in the oils of significant concentrations of hydrocarbons derived from bacterial precursors. Iso- and anteiso-alkanes, cyclohexyl-alkanes of microbial origin and of drimanes and rearranged drimanes, also considered to be of microbial origin (Alexander et al. 1984) are common constituents of Australian terrigenous oils (Fig. 6). Thus a substantial proportion of the hydrocarbons in non-marine oils are derived from organic matter of bacterial origin. A further indication of the role of bacterial organic matter in forming the source for nonmarine oils is the relatively high proportion of hopanes relative to steranes in the oils and source rocks. Steranes are dominated by the C29 homologues reflecting the land plant source. A distinguishing feature of the New Zealand Taranaki Basin oils is a much lower relative triterpane concentration and a relative abundance of diasteranes (Czochanska et al. 1988). This combination of biomarkers in the New Zealand oils compared to the Australian and Indonesian crudes suggest that paralic shales are the source for the former oils while the latter crudes are derived from peat-swamp (mire) environments (Czochanska et al. 1988). Increasingly specific naphthenic and aromatic hydrocarbons in terrigenous oils are traced to diterpenoids associated with conifer wood resins and leaf oils (Noble et al. 1986; Alexander et al. 1987; Alexander et al. 1988) and have application for specific oil-source correlations in terrestrial oils. Diterpane structures specific to the conifer tree resins of the Araucariaceae family (trees of the extant New Zealand Kauri pine group), are prominent in Southern Hemisphere organic matter (Alexander et al. 1988). The Cretaceous-Jurassic Eromanga Basin oils contain Araucariaceae-related biomarkers whereas sediments from the Triassic-
24
T. G. POWELL & C. J. BOREHAM IP-19
Acyc/ic isoprenoids (Cls-C21) = I P - 1 6 to I P - 2 1 A l k y l c y c l o h e x a n e s (C 4--Cz~ ) = C~4 CH toC21 CH
PRISTAN E Iso - + anteiso - alkanes (Cr5-C2o) = C~51 + A
to C2o I + A
Bicylanes I
Rearranged drimane
II
8 fl ( H ) - O r i m a n e
III
8 fl ( H ) - H o r n o d r i m a n e
I P - 1 8 + Cle CH C22 I + A + C21 CH C2o I + A + C ~ a C H
C21 I -I- A IP-16 Czo C H - ' ~ CrsCH I
i
M
rl
I-t-A ~
II
c l ; , CH
IP-20 I PHY'TANE
q I
L, .111 ,.Ati :It
CH
800
"C
CleI+A
Cle I + A
II
I 9O0
"'
1 1000
I 1! O0
I 1200
I 1300 Scan number
I 1400
I 1500
I 1600
17/00
I 1800 14-1/24
Fig. 6. Reconstructed ion chromatogram from gas chromatographic-mass spectrometric analysis of branched and cyclic hydrocarbons from Jackson oil, Eromanga Basin. Note importance of hydrocarbons derived from bacterial precursors - iso- and anteiso-alkanes, drimanes and rearranged drimanes.
Permian Cooper Basin, which pre-date the proliferation of the Araucariacean flora during the early to mid--Jurassic, have different diterpane biomarker distributions (Alexander et al. 1988). Conifer resin-derived diterpanes are also characteristic of sediments and oils from the Cretaceous-Tertiary Gippsland Basin, Australia (Alexander et al. 1987 and the Tertiary oils of New Zealand (Weston et al. 1989). Distinctive biomarkers termed bicadinanes are found in Tertiary organic matter and their derived oils (Grantham et al. 1983; Cox et al. 1986). They are thought to arise from the catagenetic depolymerization of resin material (van Aarssen et al. 1990). Dammar resin, found only today in the hardwood tropical forests of southeast Asia, is a possible source of these bicadinanes (van Aarssen et al. 1990). It is not surprising that terrestrially derived oils from this region are rich in bicadinanes. With the evolution of the flowering plants in the late Cretaceous there appears in coals and oils derived from terrestrial sediments specific angiosperm biomarkers - the isomeric C3otriterpane, 18a(H) oleanane (geological isomer) and 18fl(H) oleanane (biological isomer) (Riva
et al. 1988; Ekweozor & Telnaes 1990) - whilst
in some terrestrial Tertiary oils bisnorlupanes are present (Brooks 1986a). The Tertiary oils of southeast Asia and New Zealand are characterized by the relatively high abundances of angiosperm-related triterpanes and are correlated with sediments containing abundant microfossil angiosperm remains. Indeed the smooth lateral variation in the relative concentrations of conifer versus angiosperm biomarkers in some New Zealand oils faithfully follows the relative abundances of the two plant communities in the associated terrestrial sediments. (Weston et al. 1989). Despite this wealth of specific biomarker evidence it has still proved difficult to pin the source of non-marine oils down to specific stratigraphic intervals or particular lithofacies in terrestrial sequences. This is not surprising given the heterogeneity in organic matter content and composition in terrestrial sequences. In such sequences sampling becomes a critical issue. Thus the composition of the oil represents the average of the hydrocarbons derived from the drainage area of the structure which may not be represented by a single sample or group of
TERRESTRIALLY SOURCED OILS samples. Whilst this is an issue for oil-source correlation in general, marine source rocks tend to be more homogeneous in the composition of the originally deposited organic matter than is commonly found in terrigenous sequences.
Composition of terrigenous gas The isotopic composition of natural gas derived from terrestrial organic matter is extremely variable. Methane derived from coaly source rocks in northwest Germany ranges in isotopic composition from - 25 to - 35%0 relative to the PDB standard, the former being the most mature and the latter the least mature (Stahl 1974, 1977; Stahl et al. 1977). These represent amongst the most 13C-enriched commercial gases encountered. Within the broader Northwest European gas province, in which Carboniferous coals are considered to be the gas source, the isotopic ratios are more variable (Barnard & Cooper 1983). 'Heavy' values similar to those obtained from the German examples have been found in the Cooper Basin, Australia (Rigby & Smith 1981) - these originate from Permian coals. They differ from gas extracted from Permian coals in the Sydney Basin (Rigby & Smith 1981, 1982) and elsewhere (Schoell 1983) which are isotopically much lighter and more like those derived from marine sources. Thus values for terrestrially sourced gases in the Mahakam Delta, Indonesia are in the vicinity of -45%0 (Schoell 1983). The isotopic composition of methane reflects both the nature of the source material and the level of maturation. Notwithstanding the maturation control the range of isotopic compositions of methane derived from terrestrial sources is surprisingly large. Similarly James (1983, 1990) and Schoell (1983, 1984) have shown that the isotopic composition of the wet gas components reflect both the maturation level and the nature of the gas's source. The relative weighting of maturity compared to source control depends on the type of source. For humic-coal and Type III kerogen there is a stronger source control than for lipidrich type II kerogen source. The variation in isotopic composition of gases derived from coals probably reflects the heterogeneity in coal composition outlined above in which the hydrocarbon potential can be considered to be represented by a mixture of hydrogen-rich and hydrogen-poor components. In marine gases, with increasing molecular weight, the isotopic composition of these components appears to approach the isotopic composition of the source (James 1990). For
25
gases derived from humic coals the isotopic compositions of the heavier components show smaller isotopic separations than for gases derived from other organic matter types at the same maturation level (James 1990). In fact, gases from humic coals commonly depart from the normal trend to heavier isotopic values for propane, isobutane and butane. In this case n-butane may be isotopically lighter than propane which in turn is lighter than isobutane. For all types of organic matter, source control is lost above the maturity level Ro = 1.5% corresponding to thermal destruction of wet gas components (James 1990). The content of inorganic gas in terrestrially derived gases is also variable. Large quantities of nitrogen are associated with much of the gas in the Northwest European gas province (Barnard & Cooper 1983) and substantial quantities of carbon dioxide are associated with gas in the Cooper Basin. Both inorganic and organic sources have been proposed for the gas in the latter case. Large quantities of carbon dioxide are also associated with certain gases in the Gippsland Basin and are considered to be organic in origin (Stainforth 1984). Summary Coal and disseminated terrigenous organic matter are sources for commercial reservoirs for oil and gas. The number of documented cases are, however, relatively few. Terrigenous organic matter can accumulate as peats or dispersed in mudrocks over a wide variety of climatic and sedimentary conditions which may vary rapidly both laterally and through time. The extent of microbial degradation prior to preservation is also very variable. As a consequence, the terrigenous source facies tend to be extremely heterogeneous both at the microscopic and macroscopic levels. No clear picture of the floral or depositional controls on terrigenous source rock has emerged to date. Most large terrigenous oil provinces have source rocks of late Cretaceous or Tertiary age. Clearly the SE Asian and Australasian regions are important in this respect. However, the occurrence of substantial terrigenous amounts of terrigenous oils and gas-condensate originating from Permian sediments in central Australian basins, suggests that opportunities for the occurrence of terrigenous oils may be more widespread than the currently documented cases suggest. The process of defining a potential terrigenous source rock using classical techniques and criteria is at best ambiguous and is depen-
26
T. G. POWELL & C. J. BOREHAM
dent on empirical rules. For instance, a threshold H/C ratio of 0.9 has been used as the cut-off limit for liquid potential in immature terrestrial sequences (Hunt 1991 ; Powel11988a). This roughly translates to a Hydrogen Index between 200 and 300. However, closer examination reveals a wide range in composition of the evolved petroleum for organic matter close to these limits. In these circumstances analysis of the pyrolysate is required to obtain an accurate picture of source rock quality. Furthermore for the majority of terrestrial sequences, there is no simple correlation between hydrogen content of organic matter, composition and yield of hydrocarbon products and the visual analysis of organic matter (Powell et al. 1991). Finely dispersed sub-microscopic lipids of either plant of bacterial origin are important sources of liquid hydrocarbons in terrigenous organic matter. Our assessment of liquid potential in marginal source rocks, wherein most terrigenous sediments lie, is far from perfect and is usually only applicable to well-defined end members. A mass balance approach, combined with pyrolysis studies to ascertain the gas-oil mix of the generated hydrocarbons, has been shown to partially overcome these inadequacies and is essential to quantify the amounts of petroleum generated and expelled from a source rock. In general, maximum oil generation occurs over the maturation range 0.8-1.0% vitrinite reflectance. Gas generation becomes important above
0.65% vitrinite reflectance and gas-condensate can be formed at any stage of maturation depending on the relative proportions of oil and gas precursors in the parent kerogen. In combination with kinetic and physico-chemical modelling of the expelled products it becomes possible to predict oil-gas ratios and migration volumes (Noble et al. 1991; Forbes et al. 1991). A high-wax content is characteristic of oil formed from terrestrial organic matter although increasingly specific cyclic and aromatic components are being recognized as unique biomarkers of the plant source material that can be used for oil-source correlation. In the specific case of the Beaufort-Mackenzie Basin, resin components are of sufficient concentration in the source to yield naphthenic-aromatic oils. The wax content in the derived oils can be extremely variable depending on the precise nature of the source material and conditions of migration, i.e. oil or gas phase. High pristane to phytane ratios are characteristic of oils derived from land plant sources. The occurrence in terrigenous oil of substantial quantitities of of iso- and anteisoalkanes and drimanes and the high proportion of hopanes relative to steranes all indicate the importance of bacterial contribution to the source material. The isotopic composition of terrestrially derived gas is extremely variable and reflects the wide range of depositional conditions in which terrigenous organic matter accumulates as well as the well-known maturation controls.
References VAN AARSSEN,B. G. K., Cox, H. C., HOOGENDOORN, P. & DE LEEUW, J. W. 1990. A canidene biopolymer in fossil and extant dammer resins as a source for cadinanes and bicadinanes in crude oils from South east Asia. Geochimica et Cosmochimica Acta, 54, 3021-3031. ALEXANDER,R., KAGI, R. I., NOBLE, R. & VOLKMAN, J. K. 1984. Identification of some bicyclic alkanes in petroleum. In: SHENCK, P. A., DE LEEUW, J. W. & LUMBACH,G. W. M. (eds) Advances in Organic Geochemistry, 1983. Wheaton, Exeter, 63-70. - - , LARCHER,A. V. & KAGI, R. I. 1988. The use of plant derived biomarkers for the correlation of oils with source rocks in the Cooper/Eromanga Basin system, Australia. Australian Petroleum Exploration Association Journal, 28, 310--324. - - , NOBLE, R. A. & KAGI, R. I. 1987. Fossil resin biomarkers and their application in oil source rock correlation, Gippsland Basin, Australia. Australian Petroleum Exploration Association Journal, 27, 63-72. BARNARD, P. & COOPER, B. S. 1983. A review of the geochemical data related to the Northwest
European Gas Province. In: BROOKS, J. (ed.) Petroleum Geochemistry and Exploration in Europe. Blackwell, London, 19-34. BOREHAM, C. J., CRICK,I. H. & POWELL,T. G. 1988. Alternative calibration of the Methylphenanthrene Index against vitrinite reflectance: Application to maturity measurements in oils and sediments. Organic Geochemistry, 12, 289-294. -& POWELL,T. G. 1991. Variation in pyrolysate composition of sediments from the Jurassic Walloon Coal measures, eastern Australia as a function of thermal maturation. Organic Geochemistry, 17,723-734. & in press. Petroleum source rock potential of coal and associated sediments: Qualitative aspects. In: RICE, D. D. & LAW, B. E. (eds) Hydrocarbons from Coal. American Association of Petroleum Geologists Special Publication. BROOKS, J. D., GOULD, K. & SMITH, J. W. 1969. Isoprenoid hydrocarbons in coal and petroleum. Nature, 222,257-259. - - , HESP, W. R. & RIGBY, D. 1971. The natural conversion of oil to gas in sediments of the
TERRESTRIALLY SOURCED OILS Cooper Basin. Australian Petroluem Exploration Association Journal, 11,121-125. & SMITH, J. W. 1967. The diagenesis of plant lipids during the formation of coal, petroleum and natural gas-I - Changes in the n-paraffin hydrocarbons. Geochimica et Cosmochimica Acta, 31, 2389-2397. & 1969. The diagenesis of plant lipids during the formation of coal, petroleum and natural gas-II - Coalification and the formation of oil and gas in the Gippsland Basin. Geochimica et Cosmochimica Acta, 33, 1183-1194. BROOKS, P. 1986a. Unusual biological marker geochemistry of oils and possible source rocks, offshore Beaufort-Mackenzie Delta, Canada. Organic Geochemistry, 10, 401-406. 1986b. Biological marker geochemistry of crude oils and condensate from the Beaufort/ Mackenzie Basin. Bulletin of Canadian Petroleum Geology, 34, 490-505. BURNS, B. J., JAMES, A. T. & EMMETT,J. K. 1984. The use of gas isotopes in determining the source of some Gippsland Basin oils. Australian Petroleum Exploration Association Journal, 24,217-21. BUSTIN, R. M. 1988. Sedimentology and characteristics of dispersed organic matter in Tertiary Niger Delta: Origin of source rocks in a deltaic environment. American Association of Petroleum Geologists Bulletin, 72,277-298. CLVMO, R. S. 1987. Rainwater fed peat as a precursor for coal. In: SCOTT, A. C. (ed.) Coal and CoalBearing Strata: Recent Advances. Geological Society, London, Special Publication, 32, 17-24. COLLINSON,M. E. & Scorr, A. C. 1987. Implications of vegetational change through the geological record on models for coal forming environments. In: ScoTt, A. C. (ed.) Coal and Coal-Bearing Strata: Recent Advances. Geological Society, London, Special Publication, 32, 67-86. COOK, A. C. 1975. The spatial and temporal variation in the type and rank of Australian coals. In: CooK, A. C. (ed.) Australian Black Coal. Australasian Institute of Mining & Metallurgy, Illawarra Branch, Wollongong, 63-84. COOLES, G. P., MACKENZIE, A. S. & QUIGLEY, T. M. 1986. Calculation of petroleum masses generated and expalled from source rocks. In: LEVTHAEUSER, D. & RULLKOTrER, J. (eds) Advances in Organic Geochemistry 1985. Pergamon Journals, 235-245. Cox, H. C. DE LEEUW, J. W., SCHENCK,P. A., VAN KONINGSVELD,H., JANSEN,J. C., VAN DE GRAAF, B., GEERESTEIN,J., KANTERS,J. A., KRUK, C. & JANS, A. W. H. 1986. Bicanidane, a C30 pentacyclic isoprenoid hydrocarbon found in crude oil. Nature, 319, 112-118. CZOCHANSKA, Z., GILBERT, D., PHILP, R. P., SHEPPARD, C. M., WESTON, R. J., WOOD, T. A. & WOOLHOUSE,A. D. 1988. Geochemical application of sterane and triterpane biomarkers to a description of oils from the Taranaki Basin of New Zealand. Organic Geochemistry, 12, 123-135. DURAND, B. & OUDIN, J. L. 1979. Exemple de migration des hydrocarbures dans une serie
27
deltaique - La delta Mahakham, Kalimantan, Indonesie. Proceedings of the Tenth World Petroleum Congress, 2, 3-12. EKWEOZOR, C. M. & TELNAES, N. 1990. Oleanane parameter: verification by quantiative study of the biomarker occurrence in sediments of the Niger delta. Organic Geochemistry, 16,401--413. FORBES, P. L., UNGERER,P. M., KUHFUSS,A. B., Rns, F. & ECGEN, S. 1991. Compositional modelling of ptetroleum generation and expulsion: trial application to a local mass balance in the Smorbukk Sor field, Haltenbanken area, Norway. American Association of Petroleum Geologists Bulletin, 75,873-893. GORDON, T. L. 1985. Talang Akar Coals - Ardjuna Sub-basin oil source. Proceedings of the Indonesian Petroleum Association Fourteenth Annual Convention, October 1985, 91-120. GRANTHAM, P. J., POSTHUMA, J. & BAAK, A. 1983. Triterpanes in a number of far-eastern crude oils. In: BJoROY et al. (eds) Advances in Organic Geochemistry 1981. J. Wiley & Sons, Chichester, 675-683. HARWOOD, R. J. 1977. Oil and gas generation by laboratory pyrolysis. American Association of Petroleum Geologists Bulletin, 61, 2082-2102. HEDBERG, H. D. 1968. Significance of high wax-oils with respect to genesis of petroleum. American Association of Petroleum Geologists Bulletin, 52, 736-750. HIGGS, M. D. 1986. Laboratory studies into the generation of natural gas from coals. In: BROOKS, J. et al. (eds) Habitat of Palaeozoic Gas in N. W. Europe - Geological Society, London, Special Publication, 23, 113-120. HOBDAY, D. K. 1987. Gondwana coal basins of Australia and South Africa: tectonic setting, depositional systems and resources. In: SCOTT, A. C. (ed.) Coal and Coal-Bearing Strata: Recent Advances. Geological Society, London, Special Publication, 32,219-233. HORSFIELD, B. 1989. Practical criteria for classifying kerogen; Some observations from pyrolysis-gas chromatography. Geochimica et Cosmochimica Acta, 53, 891-901. --, YORDY, K. L. & CRELLING,J. C. 1988. Determining the petroleum generating potential of coals using organic geochemistry and organic petrology. Organic Geochemistry, 13,121-129. HUANG, D., SHANG,H. & LI, J. 1988. Latest advances in research on on-marine oil generation in China. In: WAGNER, H. C., WAGNER, L. C., WANG, F. H. & WONG, F. L. (eds) Petroleum Resources of China and Related Subjects. Circum-Pacific Council for Energy and Mineral Resources Earth Science Series 10, 63-88. --, ZHANG, D., LI, J. • HUANG, X. 1991. Hydrocarbon genesis of Jurassic coal measures in the Turpan Basin, China. Organic Geochemistry, 17,827-838. HUNT, J. M. 1991. Generation of gas from coal and other terrestrial organic matter. Organic Geochemistry, 17, 673-680. ISSLER, D. R. & SNOWDON,L. R. 1990. Hydrocarbon generation kinetics and thermal modelling,
28
T. G. P O W E L L & C. J. B O R E H A M
Bulletin Basin. Canadian Petroleum Geology, 38, 1-16.
Beaufort-Mackenzie
of
JAMES, A. T. 1983. Correlation of natural gas by use of carbon isotope distribution between hydrocarbon components. American Association of Petroleum Geologists Bulletin, 67, 1176-1191. 1990. Correlation of reservoired gases using the carbon isotopic compositions of wet gas components. American Association of Petroleum Geologists Bulletin, 74, 1441-1458. KATZ, B. J. & MERTANI, B. 1989. Central S u m a t r a A geochemical paradox. Proceedings of the
-
Indonesian Petroleum Association Eighteenth Annual Convention, October 1989, 403-424. VAN KREVELEN, D. W. 1961. C o a l - Typology, Chemistry, Physics, Constitution. Elesevier.
-
--
LARTER, S. R. & SENTFLE, J. T. 1985. Improved kerogen typing for petroleum source rock analysis. Nature, 318,277-280. LEYTHAEUSER, D., RADKE, M., SCHAEFER, R. G. BJOROY, M. 1984. A novel approach for recognition and quantitation of hydrocarbon migration effects in shale-sandstone sequences.
American Association of Petroleum Geologists Bulletin, 68, 196-219. LIJMBACH, G. W. M. 1975. On the origin of petroleum.
Proceedings of the Ninth World Petroleum Congress, 2,357-369. MCCABE, P. J. 1984. Depositional environments of coal and coal-bearing strata. In; RAHMAN1,R. A. & FLORES, R. M. (eds) Sedimentology of Coalbearing Sequences. International Association of Sedimentologists Special Publication, 7, 13-42. MONNIFER, F., POWELL,Z. G. & SNOWDON,L. R. 1983. Qualitative and quantitative aspects of gas generation. Examples from Canadian frontier basins. In: BJOROY, M. et al. (eds) Advances in Organic Geochemistry 1981. John Wiley & Sons, Chichester, 487-495. MOORE, P. S., BURNS, B. J., EMMETI-, K. K. & GUTHRIE, D. A. 1992. Integrated source, maturation and migration analysis, Gippsland Basin, Australia. Australian Petroleum Exploration Association Journal, 32,313-324. MURCHISON, D. G. 1987. Recent advances in organic petrology and organic geochemistry; an overview with some reference to 'oil from coal'. In: ScoTr, A. C. (ed.) Coal and Coal-bearing Strata: Recent Advances. Geological Society, London, Special Publication, 32,257-302. NOBLE, R. A., ALEXANDER,R. I., KAG1, R. I. & KNOX, J. 1986. Identification of some dieterpenoid hydrocarbons in petroleum. Organic Geochemistry, 10,363-374. --, Wu, C. H. & ATKINSON,C. D. 1991. Petroleum generation and migration from Talang Akar coals and shales offshore N.W. Java, Indonesia. Organic Geochemistry, 17,363-374. PHILP, R. P. t~ GILBERT, T. D. 1986. Biomarker distributions in Australian oils predominantly derived from terrigenous source material. Organic Geochemistry, 9, 73-84. POWELL, T. G. 1978. An assessment of the hydrocarbon source rock potential of the Canadian
--,
Arctic Islands. Geological Survey of Canada Paper, 78--12. 1986. Petroleum geochemistry and depositional setting of lacustrine source rocks. Marine and Petroleum Geology, 3, 200-219. 1988a. Developments in concepts of hydrocarbon generation from terrestrial organic matter. In: WAGNER, H. C., WAGNER, L. C., WANG, F. F. H. & WONG, F. L. (eds) Petroleum resource of China and Related Subjects. Circum-Pacific Council for Energy and Mineral resources Earth Science Series, 10,807-824. 1988b. Pristane/phytane ratio as environmental indicator. Nature, 333,604. c~; BOREHAM, C. J. 1991. Petroleum generation and source rock assessment in terrigenous sequences: an update. Australian Petroleum Exploration Association Journal, 29,114-129. - - , MCKIRDY, D. M., MICHAELSEN,B. H. & SUMMONS, R. E. 1989. Petroleum geochemistry of the Murta Member, Mooga Formation, and associated oils, Eromanga Basin. Australian Petroleum Exploration Association Journal, 29, 114-129. --, SMYTH, M., RUSSEL, N. & CooK, A. C. 1991. Petroleum source rock assessment in nonmarine sequences: Pyrolysis and petrographic analysis of Australian coals and carbonaceous shales. Organic Geochemistry, 17, 375-394. FOSCOLOS,A. E., GUNTHER, P. R. & SNOWDON, L. R. 1978. Diagenesis of organic matter and fine clay minerals: a comparative study. Geochimica et Cosmochimica Acta, 42, 1181-1197. & MCKIRDY, D. M. 1972. The geochemical characterisation of Australian crude oils.
Australian Petroleum Exploration Association Journal, 12, 125-131. --
& -1973. Relationships between ratio of pristane to phytane, crude oil composition and geological environment in Australia. Nature (Physical Sciences), 243, 37-39. PRAMANO, H., Wu, C. H. C. & NOBLE, R. A. 1990. A new oil kitchen and petroleum bearing subbasin in the offshore northwest Java Basin.
Proceedings of the Indonesian Petroleum Association Nineteenth Annual Convention October, 1990, 253-277. REED, K. J. 1969. Environment of deposition of source beds of high wax oil. American Association of Petroleum Geologists Bulletin, 53, 1502-1506. R1GBY, D. & SMITH, J. W. 1981. An isotopic study of gas and hydrocarbons from the Cooper Basin.
Australian Petroleum Exploration Association Journal, 21,222-229. -
& -1982. A reassessment of stable carbon isotopes in hydrocarbon exploration. Erdoel und Kohl Erdgas Petrochemie, 35,415-417. RIVA, A. P., CACCIALANZA,P. G. & QUAGLIAROLI,F. 1988. Recognition of 18 (H)oleanane in several crudes and Tertiary-Upper Cretaceous sediments: Definition of a new maturity parameter. Organic Geochemistry, 13,671-675. ROBINSON, K. M. 1987. An overview of source rocks and oils in Indonesia. Proceedings of the -
T E R R E S T R I A L L Y S O U R C E D OILS
Indonesian Petroleum Association Sixteenth Annual Convention, October 1987, 97-122. SCHERER, M. & HITAM, R. 1992. Distribution and maturation of source rocks in Brunei Darussalem. Exploration Frontiers in Asia and the Western Pacific. A A P G International Conference and Exhibition Sydney, August 2-5, 1992. Abstract. SCHOELLM. 1983. Genetic characterisation of natural gases. American Association of Petroleum Geologists Bulletin, 67, 2225-2238. 1984. Recent advances in petroleum isotope geochemistry. Organic Geochemistry, 6, 645663. SHANMUGAN,G. 1985. Significance of coniferous rainforests and related organic matter in generating commercial quantities of oil, Gippsland Basin, Australia. American Association of Petroleum Geologists Bulletin, 69, 1241-1254. SMITH, J. W., BATTS, B. W. d~ GILBERT, T. D. 1987. A quest for a new parameter in petroleum exploration geochemistry. Australian Petroleum Exploration Association Journal, 27, 98-105. SNOWDON, L. R. 1978. Organic geochemistry of the
-
-
Upper Cretaceous-Tertiary delta complexes of the Beaufort-Mackenzie sedimentary basin, northern Canada. Ph.D. thesis, Rice University, Houston, -
-
-
-
-
-
-
Texas. 1980. Resinite - a potential source in the Upper Cretaceous/Tertiary of the B e a u f o r t - Mackenzie Basin. In: MIALL, A. D. (ed.) Facts and Principles of World Oil Occurrence. Canadian Society of Petroleum Geologists Memoir, 6, 421-446. 1991. Oil from Type III organic matter: resinite revisisted. Organic Geochemistry, 17,743-747. & POWELL,T. G. 1979. Families of crude oils and condensates in the Beaufort-Mackenzie Basin. Bulletin of Canadian Petroleum Geology, 27, 453-466. & -1982. Immature oil and condensate modification of hydrocarbon generation model for terrestrial organic matter. American
Association of Petroleum Geologists Bulletin, 66,775-788. STACH, E., MACKOWSKY,M. Th., TEICHMULLER,M., TAYLOR, G. H., CHANDRA, D. & TEICHMULLER, R. 1982. Textbook of Coal Petrology, 3rd edn. Borntraeger. STArtL, W. J. 1974. Carbon isotope fractionations in natural gases. Nature, 251,134-135. 1977. Carbon and nitrogen isotopes in hydrocarbon research and exploration. Chemical Geology, 20,121-149. --, WOLLANKE, G. & BOI~K, H. 1977. Carbon and nitrogen isotope data of Upper Carboniferous and Rotliegende natural gases from north Germany and their relationship to the majority
29
of the organic source material. In: CAMPOS, R. 8¢ GONI, J. (eds) Advances in Organic Geochemistry, 1975. Enadisma, Madrid, 539-559. STAINFORTH, J. G. 1984. Gippsland h y d r o c a r b o n s perspective from the basin edge. Australian
Petroleum Exploration Association Journal, 24, 91-100. TAYLOR, G. H., LIU, S. Y. & SMYTH, M. 1988. New light on the origin of Cooper Basin oil. Australian Petroleum Exploration Association Journal, 28, 303-308. THOMAS, B. M. 1982. Land plant source rocks for oil and their significance in Australian Basins.
Australian Petroleum Exploration Association Journal, 22, 164-178. TISSOT, B., DURAND, B., ESPITALIE,J. • COMBAZ, A. 1974. Influence of nature and diagenesis of organic matter in formation of petroleum.
~,
American Association of Petroleum Geologists Bulletin, 58,499-506. PELET, R. & UNGERER, P. H. 1987. Thermal
history of sedimentary basins, maturation indices and kinetics of oil and gas generation. American Association of Petroleum Geologists Bulletin, 71, 1445-1466. & WELTE, D. H. 1984. Petroleum Formation and Occurrence (2nd edn). Springer-Verlag, New York. TUPPER, N. P. & BUR~:HARDT, D. M. 1990. Use of the methylphenanthrene index to characterise expulsion of Cooper and Eromanga Basin oils.
Australian Petroleum Exploration Association Journal, 30,373-85. UNGERER, P., BEHAR, F., VILLALBA,M., HEUM, 0 . R. & AUDIBERT, A. 1988. Kinetic modelling of oil cracking. Organic Geochemistry, 13, 857--868. VINCENT, P. W., MORTIMORE,I. R. & MCKIRDY, D. M. 1985. Hydrocarbon generation, migration and entrapment in the Jackson-Naccowlah area, ATP 259P, South Western Queensland.
Australian Petroleum Exploration Association Journal, 25, 62-84. VOL~:MAN, J. K. & MAXWELL, J. R. 1986. Acyclic isoprenoids as biological markers. In: JOHNS, B. (ed.) Methods in Geochemistry and Geophysics, 24. Elsevier, 1-42. WESTON, R. J., PHILP, R. P., SHEPPARD, C. M. & WOOLHOUSE, A. D. 1989. Sesquiterpanes, diterpanes and other higher terpanes in oils from the Taranaki Basin of New Zealand. Organic Geochemistry, 14,405-421. WOODROOF, P. B. & CARR, A. D. 1992. Source rock and hydrocarbon geochemistry, offshore N. W. Sabah, Malaysia. Exploration Frontiers in Asia and the Western Pacific. A A P G International Conference and Exhibition Sydney, August 2-5, 1992. Abstract.
The oil-generating potential of plants from coal and coal-bearing strata through time: a review with new evidence from Carboniferous plants M A R G A R E T E. C O L L I N S O N 1, P I M F. V A N B E R G E N 1'2'3 A N D R E W C. S C O T T 1 & J A N W. DE L E E U W 2'3
IDepartment of Geology, Royal Holloway University of London, Egham, Surrey TW20 OEX, UK 2Organic Geochemistry Unit, Delft University of Technology, Faculty of Chemical Technology and Materials Science, De Vries van Heystplantsoen 2, 2628 RZ Delft, The Netherlands 3present address: Division of Marine Biogeochemistry, Netherlands Institute for Sea Research (NIOZ), PO Box 59, 1790AB, Den Burg, Texel, The Netherlands Abstract: A combination of microscopy and chemistry is used to characterize high
molecular weight components of plant material which contributed to Carboniferous coals and coal-bearing strata. Material was selected from coals, paper coals, coal balls, oil-shales and shales and included cuticles from stems and leaves, periderm from stems and rooting systems, spore walls, algal cell walls and 'resin' rodlets. Chemical analyses were undertaken using Curie-point pyrolysis-gas chromatography (-mass spectrometry) (Py-GC-(MS)) and 13C solid state Nuclear Magnetic Resonance (NMR). Light microscopy and scanning electron microscopy were used to ensure a detailed understanding of the plant material which was analysed chemically. This study has emphasized those plants, plant tissues and organs which are known to have been dominant or major contributors to Carboniferous coals and coal-bearing sequences: arborescent lycophyte periderm (Diaphorodendron stems and Stigmaria rhizomorphs); pteridosperm cuticles (medullosan (e.g. Alethopteris) and Karinopteris), lycophyte stem cuticle (Eskdalia) and arborescent lycophyte spores (megaspores and microspores). Several algal cell walls (Tetraedron, Tasmanites and Gloeocapsomorpha) and the 'resin' rodlets derived from medullosan pteridosperm petioles have also been analysed. Results show that all of these elements (except resins) contain (or are dominated by) highly resistant, highly aliphatic macromolecules which have the potential to yield, upon catagenesis, n-alkanes which are found in crude oils. Resins potentially contribute cyclic hydrocarbons to crude oils. Combining all this evidence it is concluded that Carboniferous coals are oil prone and that explanations for the absence of oil-pools derived from such coals must be sought in geological or exploration factors and not in the nature of the coals themselves. A review of coal-forming floras through time, and of suggested terrestrially sourced oils and their parent coals, leads us to conclude that high molecular weight components of higher plant materials have potentially made a major contribution to oils sourced from coals and coal-bearing strata over a long period of geological time (Devonian onwards). Cell walls of freshwater algae have made a comparable contribution since the Ordovician. The oil-generating potential of each coal or organic-rich sediment will depend upon a combination of depositional environment and the floristic and chemical composition of the source vegetation and its component plants and plant parts.
C r u d e oils generally contain relatively high a m o u n t s of aliphatic and a r o m a t i c hydrocarbons. In m a n y cases n-alkanes are the most a b u n d a n t h y d r o c a r b o n s . As a result of r e c e n t w o r k it has b e c o m e clear that the n - a l k a n e s are mostly g e n e r a t e d f r o m selectively p r e s e r v e d , highly aliphatic, resistant b i o m a c r o m o l e c u l e s .
T h e s e occur in algal cell walls (algaenan); in higher plant leaf cuticles and the t e g m e n layer of seed coats (cutan and o t h e r u n n a m e d comp o u n d s ) and in the bark of w o o d y plants ( T e g e l a a r et al. 1989a; D e L e e u w et al. 1991; D e L e e u w & L a r g e a u in press; van B e r g e n et aI. 1993). In addition, the class of c o m p o u n d s
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coal and Coal-bearing Strata as Oil-prone Source Rocks? Geological Society Special Publication No. 77, pp. 31-70.
31
32
M. E. COLLINSON E T AL.
informally termed 'sporopollenins' present in spores and pollen, probably consists of an aromatic and an aliphatic component (van Bergen et al. in press a; Hemsley et al. in press). Suites of cyclic hydrocarbons can occur in high abundance in some oils and are thought to be generated in part from dammar (polycadinene) and gymnosperm resins (Snowdon & Powell 1982; van Aarssen et al. 1990, 1992; van Aarssen & De Leeuw 1992a & b). The evolution of appropriate plant groups, bearing the requisite organs and tissues, controls the occurrence of these n-alkane and other hydrocarbon precursors. Some are widespread in the plant kingdom and have a long fossil history. Because of their high stability these materials are not fully recycled in the biosphere and, as a consequence, they become incorporated in the geosphere. One particular situation where recycling is strongly inhibited is in peatforming environments which may, upon burial, yield coals and associated organic-rich sediments (Scott 1987). In this paper we consider the evolution of plants and selected plant compounds relevant to the generation of oil from coal and coal-bearing strata. There are various definitions of coal but we follow that in the ICCP handbook (1963) 'Coal is a combustible sedimentary rock formed from plant remains in various stages of preservation by processes which involved the compaction of the material buried in basins, initially of moderate depth' - combined with that of Schopf (1956) - ' a readily combustible rock containing more than 50% by weight and more than 70% by volume of carbonaceous material formed from compaction or induration of variously altered plant remains similar to those of peaty deposits.' We include consideration of algal-dominated coals as well as those dominated by higher plants. Although extensive coals and associated organic-rich strata were deposited in the Carboniferous (Hower et al. in press), terrestrially sourced oils are mainly derived from Mesozoic or Tertiary strata (Murchison 1987). The absence of Carboniferous terrestrially sourced oils requires some explanation. We document here new results on the chemical composition of a range of well-characterized higher plant materials (spores, cuticles, bark and resin-like secretions) from plants which grew in Carboniferous peat-forming communities. Our attention is focused on the oil-generating potential of these higher plant materials but we also consider the role of freshwater algae which, in certain circumstances, can be significant contributors to coals and organic-rich sediments.
We briefly review the fossil history of peatforming floras through time and survey the major examples in the literature which suggest that some of the resultant coals have yielded terrestrially sourced oils. These reviews enable us to incorporate the new results from Carboniferous plants into an overview of the oilgenerating potential of higher plant and algal material in coals and coal-bearing strata through time.
Review of peat (coal)-forming floras through time There are several recent reviews of coal-forming plants through time, namely Cross & Phillips (1990) which deals with North America, Lapo & Drozdova (1989), which emphasizes former USSR and DiMichele et al. (1987) and Collinson & Scott (1987) which attempt a global coverage although neither claim to be exhaustic reviews. In this paper we present only a brief summary of the major events in the evolution of coalforming floras drawn from these review articles, without repetition of references, but with occassional additional references where necessary. The earliest coals were apparently derived from cyanobacteria in the late Precambrian. Ordovician coals include those formed from the alga G l o e o c a p s o m o r p h a (Derenne et al. 1992) which is closely related to Botryococcus, an alga which is found in Middle Devonian and later Boghead coals. Algal-dominated coals (e.g. torbanites, Boghead coals etc.) subsequently occur, in selected settings, throughout the geological record. Middle Devonian coals in China are cuticle rich, formed from a flora of pteridophytes including mainly small herbaceous psilophytes, lycophytes and zosterophylls with some taller lycophytes (Dexin 1989; Sheng et al. 1992). A puzzling occurrence of tetracyclic terpanes, more usually associated with gymnosperms, requires further study (Sheng et al. 1992). Another Devonian cuticle-rich coal in the Kuznetsk Basin (former USSR) is dominated by Orestovia devonida Ergolskyaya, an enigmatic herbaceous leafless vascular plant (Lapo & Drozdova 1989). One example of a Late Devonian coal in North America is dominated by a scrambling fern-like shrub R h a c o p h y t o n . Many Lower Carboniferous coals were dominated by herbaceous and arborescent lycophytes but ferns were also important in some cases. Upper Carboniferous coals were dominated by arborescent lycophytes with a subordinate
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW element of cordaites and pteridosperms (see also later in this paper). The arborescent lycophytes were spore-bearing and many were heterosporous (producing both megaspores and microspores). They produced little secondary xylem (wood) but large amounts of bark. They were generally small trees, short lived and with determinate growth, resulting in a small canopy and low leaf production. Taxonomic and community diversity were generally low in these peat-swamps. In Euramerica, the latest Carboniferous (Stephanian) and Permian coal-forming floras were dominated by tree ferns or tree ferns with Calamites respectively. In contrast, in the Cathaysian Province (e.g. in China), arborescent lycophytes continued to thrive until the end of the Permian in association with tree ferns and cordaites. In places the seed plants replaced spore-bearing plants during the Permian. Permian coals tend to be dominated by wood and leaf materials in contrast to bark which dominated many Carboniferous coals (Lapo & Drozdova 1989). In the Permian of Gondwanaland (the southern continent, e.g. Australia, Antarctica, S. Africa, India and South America) distinctive seed plants, including glossopterid trees, formed coals. E. L. Taylor et al. (1989) have studied these coalforming floras using permineralized Permian & Triassic peats from Antartica. Another important recent study is that of GuerraSommer et al. (1991) who document floras from coals and associated shales in Brazilian Permian sequences. Unfortunately, there are no coal balls in this sequence so the evidence for the coal-forming plants is based on palynology. This suggests three peat-forming associations, one dominated by herbaceous lycophytes with algae, one dominated by lycophytes and ferns and one by gymnosperms, mainly glossopterids. Early Mesozoic floras in coal-forming sequences included conifers with cycadophytes and rarely lycophytes, but this period of coal formation is poorly understood with most evidence coming from coal petrology and palynology rather than plant macrofossils. It is unclear which plants were dominant in the actual coal-forming areas themselves and hence whether or not the coals were mainly wood dominated. Lapo & Drozdova (1989) consider that bark tissues from conifers and ginkgoaleans were the dominant components of Jurassic and Lower Cretaceous coals in former USSR and they note that, even though some would term this time 'the Cycadales epoch', few cycadophytes occur in these coals. These authors also note paper coals dominated by ginkgoalean and czekonowskialean cuticles in
33
many coal deposits of the Jurassic and Lower Cretaceous of former USSR. In a recent detailed ecological and successional survey of a Jurassic coal basin in China, Miao et al. (1989) used palynofacies evidence and fusainized macrofossils from coal themselves to document various peat-forming communities as the sources of different seams. One of these was dominated by dwarf conifers, another by tree ferns and another by cycads and arborescent conifers. Each of these elements tended to form a subordinate component where not dominant. The Cretaceous and Tertiary mark a major new phase in coal-forming floras. Many coalswamps (at least in the northern hemisphere) were dominated by taxodiaceous conifers similar to those which grow in peat-forming swamps today. These are long-lived trees and produce very large amounts of secondary xylem (wood) in trunks and branches. They produce an extensive canopy. Many were deciduous and shed large numbers of leafy shoots. Reproduction was by seeds. In the Lower Cretaceous the understorey vegetation included tree ferns and herbaceous ferns. In the Late Cretaceous and Tertiary a diverse array of flowering plant trees came to be associated with, or successional to, the conifers in the peat-swamps. Herbaceous flowering plant communitites became established, especially in open-water areas and at the margins of open water (reed-marsh). Reed marshes potentially covered large areas given particular conditions of the water table. In the warmest climates, peats were dominated by flowering plant communities, as they are today in tropical areas. The diversity of flowering plants contributed to an increased diversity of peat-forming communities and thus to a greater variety of peat types. Boulter et at. (1993) used traditional palaeobotanical interpretations and objective statistical analyses and documented two coalforming communities (both dominated by Glyptostrobus Endl.-Taxodiaceae), with a variety of surrounding plant associations, in the Miocene of Czechoslovakia. Although mossdominated peats are common today, and spore evidence suggests their presence in some Tertiary peat-forming floras, we know of no preQuaternary examples of moss peats like those dominated by S p h a g n u m at the present day. A general summary of major events in the evolution of coal-forming floras is as follows: (1) an algal phase (possibly late Precambrian to Middle Devonian as the only coal-forming flora and subsequently to present day as a subordinate flora in distinctive habitats);
34
M. E. COLLINSON ET AL.
(2) a phase of herbaceous spore-bearing plants (Middle to Late Devonian); (3) a phase of arborescent, bark-rich but non-woody, spore-bearing plants (Late Devonian to Carboniferous/Permian in places)' (4) a phase with extinct groups of seed-bearing (gymnospermous) trees with distinctive taxa in different areas of the globe (Permian); (5) a phase of coniferous trees with fern understorey (early Mesozoic); (6) a phase with communities becoming similar to those of modern peat-forming environments, especially those dominated by large woody trees of taxodiaceous conifers. Flowering plant trees are associated with conifers in peat-swamps and herbaceous flowering plants produce distinct peatforming floras leading to a greatly increased variety of potential peat types (late Cretaceous to Recent).
R e v i e w of s o u r c e plants and floras for terrestrially s o u r c e d oils
With few exceptions (e.g. kukersites, tasmanites - see below) most oil-shales and coal and coal-
bearing sequences were deposited in terrestrial settings. (We use the term terrestrial to include all non-marine depositional environments: lacustrine, fluvial, floodplain, mire (bog & swamp) and landward deltaic areas.) Whilst oilshales have long been known as oil-source rocks (reviews in Yen & Chilingarian 1976; Stach et al. 1982; Hutton 1986, 1987; Fleet et al. 1988) it is only relatively recently that the oil potential of other terrestrial sources has been considered (Bertrand 1984). Murchison (1987) provided a review in which oils from various Australian basins; the Mahakam Delta, Indonesia; the Niger Delta, Nigeria; the Beaufort-Mackenzie Basin, Canada; various basins in China and a few other areas are all considered to be terrestrially sourced. Vascular p l a n t sources (a) Australian basins Philp & Gilbert (1986) provided biomarker evidence for a higher plant source for oils from the Gippsland Basin, the Sydney Basin, the Surat & Bowen Basin, and the CooperEreomanga Basin with other basins having a mixed marine/higher plant source. Thomas (1982) and Shanmugam (1985) have considered possible source plants.
In the Gippsland Basin the coal-beating, fluvio-deltaic Latrobe Group (Upper Cretaceous to Tertiary) is both reservoir and source for extensive hydrocarbon resources. On the basis of palynology, presence of large amounts of resin (some of which contains leaves of Agathis Salisb.) and preservation of trunks, leaves, cones and seeds, the dominant plants of the Latrobe Valley coals (= Victorian brown coals) are considered to be coniferous trees of the family Araucariaceae. Other coniferous fossils include members of the Podocarpaceae. Flowering plants seem to have contributed little to the sediments. Pollen from the flowering plant Nothofagus Blume is abundant in the Yallourn brown coal but this coal also contains thick lenses composed entirely of cuticle, pollen and resin, the latter, at least, apparently produced by Araucarian conifers (summary of data provided by Thomas 1982 & Shanmugam 1985). Anderson & Mackay (1990) reinterpreted the depositional environments of the Victorian brown coals but did not report specifically on fossil plant communities. Van Aarssen & De Leeuw (1992a & b) investigated the chemical composition of resins from the Yallourn brown coal (there termed lignite). Their results support an interpretation of the macromolecular component of these resins as polycommunic acid and this is consistent with a possible origin from Araucariaceae (Anderson & Winans 1991). Heating of these resins yielded bi- and tricyclic aromatic hydrocarbons frequently encountered in oils (van Aarssen & De Leeuw 1992a & b). The Walloon Coal (Middle Jurassic - Surat Basin) is composed mainly of conifers (Gould 1981; Gould & Shibaoka 1980). Other plants included woody cycadeoids and seed ferns along with herbaceous ferns, lycophytes and equisetaleans. As far as we are aware, there are no studies of permineralized plant material and very few studies clearly relating plant macrofossils to parent plants in these Australian coal-bearing sequences. Many of the sediments have not been investigated palaeobotanically. We, therefore, consider it unwise to assume that all terrestrially sourced oils in Australia were derived from Araucarian-rich coals. There is a major opportunity here for further study of these oils, their parent coals and the vegetation which produced them. (b) Mahakam Delta A delta, which coincided with the present Mahakam Delta, has existed since the Middle Miocene and prior to that clastic sedimentation
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW occurred in this area. This has resulted in a thick organic-rich sequence in which the kinds of organic material are thought to have changed little since the Miocene and are of terrestrial origin (Garrigues et al. 1988; Durand & Oudin 1980). Peat-forming environments and vegetation similar to those of the present day in the area are usually invoked as the depositional settings for the coals in this sequence (Thompson et al. 1985; Cole 1987). This is largely based on palynomorph evidence from Miocene coals in Borneo which yielded a reconstructed vegetation zonation comparable with that observed in modern ombrogenous (raised) mires from Borneo (Anderson & Muller 1975). Anderson (1983) has reviewed South East Asian peatforming floras, whilst Anderson & Muller (1975), Caratini & Tissot (1988) and Morley (1981) present evidence for the nature of the Late Pliocene, Quaternary & Holocene development of the floras. The modern peat-forming floras of the Mahakam Delta have never been fully documented (Gastaldo 1990). A coastal mangrove fringe includes Avicennia and Sonneratia, which is replaced, in slightly more stable settings, by extensive Nipa-dominated mangrove, in turn replaced by a hardwood swamp forest with a diversity of flowering plants (Gastaldo 1990), including Dipterocarpaceae trees, a major resinproducing group (van Aarssen et al. in press). SE Asian crude oils frequently contain polycyclic sesqui-, tri- and oligoterpenoids and chemical evidence shows that these have been derived from a polymeric cadinene found in resins (van Aarssen et al. 1990, 1992). Resin samples from SE Asian coals have also been shown to contain the biomacromolecule polycadinene (van Aarssen et al. 1990). Hence it is reasonable to argue that SE Asian crude oils containing distinctive di-, tri- and polycadinines were sourced either by resin-containing coals or by relatively pure resin layers (such as can be seen accumulating in deltaic sites today in Indonesia, A. Y. Hucs pers. comm. 1993). This distinctive resin chemistry has now been found in two quite unrelated groups of flowering plants (Dipterocarpaceae (dammar resin) and mastixioid Cornaceae - van Aarssen et al. in press) so it is no longer correct to assume Dipterocarpaceae trees as the only source plants (e.g. Langenheim 1990; Anderson & Winans 1991). Future work may well extend further the range of plants producing resins with this distinctive chemical composition. Cole (1987) using evidence from palynofacies analysis, inferred five categories of former peatforming environments which produced Tertiary
35
coals in Southeast Asia. One of these was considered aUochthonous and to have formed perhaps in a similar manner to peats forming on beach ridges of the present-day Mahakam Delta. The other four were considered autochthonous and to represent a lakeside peat swamp, alluvial braid plains, interdistributary bays and mangroves. Palynomorphs were not identified to specific source plants but general categories were recorded. The coal considered to represent a lakeside peat contained abundant pollen and spores from freshwater swamp plants along with abundant resinite, whilst algae were almost absent. The lacustrine coal contained abundant algae (Botryococcus and Pediastrum) along with palynomorphs from freshwater swamp elements. Botryococcus is almost absent from the fluvio-lacustrine coals although Pediastrum was still present in rather smaller numbers. Spores of the aquatic fern Ceratopteris were common in this setting. The interdistributary bays were characterized by small numbers of algae and mangrove taxa along with abundant freshwater swamp elements. The mangrove settings included abundant mangrove taxa, with marine elements and freshwater spores. The allochthonous coal and the mangrove coal were not considered oil prone, but all the others were on the basis of the presence of high proportions of cutinites, resinites and palynomorphs including algae. Thompson et al. (1985) did consider allochthonous coals as possible oil-source rocks postulating a selective enrichment of oil-prone liptinitic kerogen during water transport. Cole (1987), in contrast, favoured autochthonous coals as oil sources (see above). Taphonomic studies by Gastaldo (1990) documented high TOC (up to 40%) in plant debris accumulations up to 2 m thick, forming today in beaches along the Mahakam Delta coastline. In addition to woody tissues, these accumulations contained leaf fragments, cuticles and resin and thus would seem to be potentially oil prone (see results section). Whilst Cole (1987) considered the mangrove-derived coals and shales as poor oil source rocks being thin and largely vitrinitic, Brown (1989) argued that mangrove shales should be considered potentially oil prone, drawing in part on a model based on modern mangroves (Risk & Rhodes 1985). The enormous sequence of coal-bearing strata and the variety of sub-environments in the deltaic setting makes it difficult to generalize from the limited available evidence. The inference that peat-forming environments like those of the present covered this area since the
36
M. E. COLLINSON ET AL.
Miocene rests primarily on one detailed palynological investigation, that of Anderson & Muller (1975). Many more, detailed, palynological studies as well as studies of plant macrofossils are necessary before the nature of peat-forming floras can be established with any certainity. Taphonomic studies, such as those of Gastaldo (1990), are also necessary to improve our knowledge of the origins of peat accumulations in tropical settings. (c) Canada Snowdon & Powell (1982) summarized evidence for a major contribution from resinite to the terrestrially sourced oils in the Tertiary Beaufort-Mackenzie Basin. However, we know of no evidence pertaining to the botanical origin of the resin or for the flora of the coal-bearing sequence in general. The diterpenoid nature of the biomarkers indicates a class 1 resin of Anderson & Winans (1991) for which the most abundant possible sources are conifers such as Araucariaceae or Cupressaceae or the caesalpinioid legumes amongst the flowering plants (Langenheim 1990; Anderson & Winans 1991; van Aarssen 1992).
posed of this alga as seen in preparations studied using UV fluorescence microscopy. These were from the Sydney Basin, Permian of Australia; the Evander area, Permian of South Africa; and one unlocalized Australian sample ref BJ248. Sinninghe Damst6 et al. (1993), using microscopical and pyrolysis techniques, concluded that the Ribesalbes oil shale (Miocene, Catalonia, Spain) was dominated by Botryococcus in spite of the fact that very few complete cell walls of the alga could be observed by microscopy. They convincingly argued that the amorphous organic material, which dominated the kerogen, was derived from Botryococcus but had been altered during diagenesis. A high concentration of another green alga, Pediastrum, was noted from the lacustrine Rundle oil shale, Tertiary, Queensland, Australia (Powell 1986). Hutton (1986, 1987) referred to these and similar oil-shales (Hutton 1988) as lamosites after their often welllaminated nature. Sinninghe-Damst6 et al. (1993) showed that the Cerdanya oil shale (Miocene, Catalonia, Spain) also contained a high concentration (60%) of Pediastrum. Algal~higher p l a n t s o u r c e s
A l g a l sources
According to Moldowan et al. (1985) 'there is no need to invoke large contributions of higher plant material to explain normal paraffinic distributions typically found in non-marine crude oils'. Non-marine algae such as Botryococcus are regarded as equally likely sources and 'with a few exceptions, an explanation based on distributions of algal types and bacterial reworking can explain the majority of observed biologic marker distributions in crude oils'. We would argue, in contrast, that the presence of algal biomarkers e.g. botryococcane, merely indicates the presence of algae e.g. Botryococcus in the depositional environment of the source rock (see Derenne et al. 1988). We require microscopic evidence of an algaldominated kerogen in the source rock to invoke a purely algal source for the oil. In the absence of this evidence, and anticipating our results (see later), we consider that a higher land plant contribution cannot be dismissed. Torbanites and Boghead coals (dominated by Botryococcus) are well-known oil-source rocks (review in Murchison 1987; see also Stach et al. 1982; Hutton 1986, 1987; numerous papers in Fleet et al. 1988). Derenne et al. (1988) emphasized that many so-called torbanites actually contain only a minor amount of microscopically recognizable Botryococcus and they expressly studied only examples chiefly com-
Some oils from the Lower Cretaceous of Brazil are sourced from freshwater lacustrine sediments (Mello et al. 1988). However, the organic matter is said to be mainly amorphous and herbaceous with no mention of microscopically recognizable algae. Mixed algal and terrestrial organic matter is typical of the oil-shales of the Uinta Basin (Tertiary - Green River Formation USA) (Powell 1986). Talbot (1988) noted a mixed accumulation of algal and higher plant debris in Holocene sediments of tropical African lakes and considered these as potential oilsource rocks with Lake Victoria as a possible modern analogue for the Green River oil-shales. Katz (1988) also drew comparisons between the Green River Formation oil-shales and modern East African lake sediments. Yang et al. (1985) discussed the Songliao Basin in China. This is a large lake basin with Mesozoic to Cenozoic sediments and producing the oils of the Daquing oil field. The kerogen is referred to as sapropelic and lake conditions are said to have been eutrophic with abundant plankton forming the main source for the kerogen (p. 1121). However, the microscopic evidence from thin sections shows a small amount of algal fossils and some plant debris, leading to a conclusion that the sapropelic kerogen had two original organic sources, algae and higher terrestrial plants extensively reworked by bacteria (Yang et al. 1985, p. 1112-
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW 1113). A map of the distribution of kerogen types in this lake produced by Powell (1986, fig. 4) indicates type I kerogens in the centre of the lake grading into type III at the margin. The type I kerogens are described as having a high content of degraded terrestrial organic matter as well as algal remains, a large proportion of which appear to be derived from Pediastrum (Powell 1986). Several papers in Fleet et al. (1988) consider terrestrially sourced oils from China. Generally these do not give details of microscopic observations of the organic material but rely on studies of bulk kerogens and/or on biomarkers. In view of the documented contributions of resins to crude oils (see Canada & Mahakam Delta above) and of terrestrially sourced oils from China (Yang et al. 1985; Murchison 1987) we consider it appropriate to mention resin-rich coals in China, even though these have not yet been invoked as oil sources. The Fushun coalfield (Early Tertiary, Liaoning Province) includes a resin-rich unit within the coal (Guchengzi Formation, Eocene). The flora of the Fushun Group (Palaeocene and Eocene) has been studied but the amber is attributed to an unknown possible unique plant association within the swamp environment (Johnson 1990, p. 226-227). Chemical analysis of this resin (Van Aarssen 1992) revealed a polymerized bicyclic diterpenoid, probably a communic acid.
Stratigraphic distribution of terrestrially sourced oils Almost all of the terrestrially sourced oils, referenced above, are from Mesozoic or Cenozoic source rocks. A few of the Australian examples are Permian. In view of the extensive coals in Palaeozoic strata worldwide (Hower et al. in press; Given et al. 1980, see also section on coal-forming floras through time), the apparent absence of terrestrially sourced oils from this era requires an explanation. We have examined a range of Carboniferous plant material (see below) with a view to establishing the oil source potential of Palaeozoic coals and coal-bearing strata.
New chemical analyses of coal-forming plants Plant fossil material: source, nature, significance a n d pr e par at i on Plant fossil material for this study has been obtained from a number of different coal-
37
bearing sequences and a range of lithologies including coals, coal balls and organic-rich shales. A variety of plants (pteridosperms, lycophytes, algae) and plant parts (leaf and stem cuticles, periderm of stems and rootstocks, secretions, spores and algal cell wails) have been investigated. Full descriptions and anatomical details of the plant materials, their sources, significance and preparation methods are given below for each different plant or location which we have studied. In providing these rather lengthy details we intentionally emphasize the need for full documentation so that it is clear exactly what material has been analysed and what factors might have influenced the results and to enable analyses to be repeated by other workers. A summary of the material studied is also provided in Table 1. We are including herein our studies from selected algae, inspite of a slight overlap with other literature, because the algae are frequently important contributors to coals and organic-rich sediments and so the chemical composition of algal cell walls is very relevant to this paper. We also felt that it is helpful to illustrate the algal chemical composition for direct comparison with that of the various higher plant materials. In addition, as algae have been studied in several laboratories, the results enable comparison of our work with that of others. Two early analyses of modern plant material performed in the Delft laboratory, cutan from Agave (Fig. 3) and sporopollenin from Salvinia (Fig. 18) are included for comparative purposes. Details of these materials can be found in Nip et al. (1986a & b) and van Bergen etal. (in press a) respectively.
(a) Eskdalia, lycophyte stem cuticle: Russian 'paper' coal This material comes from the collections of the British Geological Survey, Keyworth, Nottingham, England specimen number 1299. The paper coal is from the Lower Carboniferous of the Moscow Basin where it occurs in thin seams ranging from l c m to l m in thickness in several localities (Walton 1926; Thomas & Meyen 1984). The BGS specimen 1299 is from the Tovakova locality. The paper coal (Fig. la) is dominated by one type of cuticle from the stems of a small lycophyte named Eskdalia Kidston (see Thomas 1968; Thomas & Meyen 1984). These cuticles are perforate (Fig. lb) and the apical margins of the perforations carry small tubes which are the linings of ligule pits. The perforations represent
Tissue/organ Stem cuticle Leaf cuticle Leaf cuticle Leaf cuticle Leaf cuticle Leaf cuticle Periderm Periderm Secretion Resinrodlet Microspore Megaspore Cell wall Cell wall Cell wall
Botanical affinity
Lycophyte Pteridosperm Pteridosperm Pteridosperm Pteridosperm Pteridosperm Lycophyte Lycophyte Pteridosperm Pteridosperm Lycophyte Lycophyte Alga Alga Alga
Genus
Eskdalia Karinopteris Neuropteris Reticulopteris Alethopteris Alethopteris Diaphorodendron Stigmaria Myeloxylon ?Medullosan Lycospora Tuberculatisporites Tasmanites Gloeocapsomorpha Tetraedron
L. Carboniferous U. Carboniferous U. Carboniferous U. Carboniferous U. Carboniferous U. Carboniferous U. Carboniferous U. Carboniferous U. Carboniferous U. Carboniferous L. Carboniferous U. Carboniferous Permian Ordovician Eocene
Age Tovakova, Russia Indiana, USA Indiana,USA Indiana, USA Kansas, USA Swillington, Yorkshire Dearplay, Yorkshire Dearplay, Yorkshire Indiana, USA Swillington, Yorkshire Pettycur, Scotland Swillington, Yorkshire Tasmania, Australia Estonia Messel, Germany
Source Papercoal Papercoal Coalball Coalball Coalball Shale Cgalball Coalball Coalball Shale Limestone Shale Coal Kukersite Oil-shale
Lithology
Sub-bituminous B (Ro 0.42) High Vol. Bituminous C High Vol. Bituminous C (Ro 0.49) High Vol. Bituminous C (Ro 0.49) High Vol. Bituminous C High Vol. Bituminous C (Ro 0.62) High Vol. Bituminous C (Ro 0.90) High Vol. Bituminous C (Ro 0.90) High Vol. Bituminous C (Ro 0.49) High Vol. Bituminous C (Ro 0.62) High Vol. Bituminous C (Ro 0.52) High Vol. Bituminous C (Ro 0.62) ?High Vol. Bituminous ?Bituminous Lignite (Ro 0.30)
Rank/maturity
Table 1. Summary of plant material used for chemical analyses. Rank/maturity measurements' based on our own work (R o = mean reflectance of vitrinite under oil) except for Karinopteris from Indiana and Alethopteris from Kansas (see Damberger 1991), Eskdalia from Russia (from A. Golitsyn, pers. comm.) and Tetraedron from Messel (see RullkOtter et al. 1988). For further details see text
o¢
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW the areas at which leaves were attached. The stem cuticles lack stomata and the compressed cuticle sheets reach widths up to 6 cm as seen in our material. Details of the cuticle anatomy are given by Walton (1926), Wilson (1931) and Thomas & Meyen (1984). One of the most completely known species of Eskdalia is described by Rowe (1988) but the true stature of entire plants of the genus remains unclear. Following a reinvestigation of compression fossils of Eskdalia from shales, Thomas & Meyen (1984, p. 712) concluded that the absence of leaves in the Eskdalia material from the Russian paper coals was most likely due to the leaves having a very thin cuticle which has not been preserved. This idea finds support in the observations of Wilson (1931) who described small fragments of thinner cuticle with stomata, attached to the perforations in the stem cuticles. Anticipating our results and bearing in mind published work on cuticle chemistry (Nip et al. 1986a, b; Telelaar et al. 1991) it is possible that the chemical composition of the leaf and stem cuticles also differed. Lycophyte leaf cuticles are notoriously difficult to prepare from compression fossils and it seems that many stem compressions may have been described as leafless because leaves have simply not been preserved (Thomas & Masarati 1982). Pieces of lycophyte cuticle are not readily recognizable in coal ball macerations and very few authors have described lycophyte leaf cuticle obtained from coal ball material (e.g. Graham 1935; Reed 1941) inspite of the huge literature on coal balls (Scott & Rex 1985). The lack of any leaf cuticle characters in recent major cladistic analyses of lycophytes serves to emphasize this situation (Bateman et al. 1992). In contrast, leaf cuticles from pteridosperms are frequently preserved and well documented (see sections on Indiana coal balls & Kansas coal balls). Stem cuticles (or more accurately leaf base or leaf cushion cuticles because arborescent lycophytes were clothed in persistent leaf bases), are well documented from compression fossils and coal balls for a variety of coal-forming lycophytes (Thomas & Masarati 1982, p. 371-374; references in Bateman et al. 1992) although those of the sigillarians are rarely preserved (Thomas & Masarati 1982). The determinate growth and 'pole-like' tree of the 'lycopsid tree habit' (Phillips & DiMichele 1992; see also reconstructions in Bateman et al. 1992, fig. 1) with a tall, unbranched trunk topped by a generally sparsely branching crown, would have resulted in limited leaf and leafy shoot contribution to underlying sediments. In contrast to a deciduous flowering plant tree e.g. Quercus,
39
the oak, shedding large numbers of leaves every year from an extensive canopy, a typical arborescent lycophyte would shed very few leaves from the trunk during early growth and then produce only one set of canopy leaves, either shed gradually during its mature state or lost suddently upon death of the plant. Arborescent lycophytes dominated the Upper Carboniferous coal-forming floras of Euramerica (see section on British coal balls). We are using the Eskdalia material as representative of the lycophyte stem cuticle. Based on evidence from preservation and the 'lycopsid tree habit' we argue that stem cuticle, rather than leaf cuticle, was the major cuticle from arborescent lycophytes preserved in coalbearing sequences. The Eskdalia cuticles were released from the matrix by treatment with cold 40% hydrofluoric acid (HF) and the residues were sorted in water using a low power microscope. No other treatments were used. Figure lb represents a typical example of the material which was selected for analysis. This consists only of the stem cuticle.
(b) Karinopteris, pteridosperm cuticle: Indiana 'paper' coal Our material (Fig. lc) comes from an outcrop of the Upper Block Coal Member of the Brazil Formation (Pennsylvanian, Upper Carboniferous) along Sugar Creek near Coke Oven Hollow, Indiana, USA; for details see Eggert & Phillips (1982) and DiMichele et al. (1984). The cuticle-rich shale (and cuticular coals from nearby sites) are dominated by frond cuticles of Karinopteris (species not identified), a vine-like pteridosperm. Although individual layers up to a few millimetres thick contain exceptionally abundant Karinopteris, a more diverse associated flora does occur including lycophytes, ferns and sphenophytes. Foliage of the latter is, however, poorly preserved; the Karinopteris cuticle, which is thick and seems to have been more resistant, has been concentrated in the cuticle-rich shales and coals (DiMichele et al., 1984; Eggert & Phillips 1982). The term 'paper coal' is used when the cuticle-rich coal has been modified by weathering which has removed the vitrinite leaving a residue of partially separated, flexible cuticular material (Neavel & Guennel 1960). The overall flora of this coal is distinct from that of typical Euramerican coal-swamps (see section on British coal balls) having more similarity to shale-derived compression floras. The clastic content of the cuticular coal ranges from 11-50%. These factors probably reflect a
i,i~
~ ,
.....
~%1:i~¸¸
Ill~:
~ ~,~i~iii~ °~
~!~i!!ii: ~iii~i!iiiiiii~iiiiiiiiii'i,!!i~
i~J •
,,? 'i '¸ !.....
~, i :fill¸I
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW depositional setting in a fluvially influenced upper delta plain rather than in coastal swamps of the lower delta plain (DiMichele et al. 1984). The Karinopteris material was isolated from the shale by short treatment with cold 40% HF. After neutralization, pieces were sorted in water using a low power microscope. Axes, rachides with pinnae and individual lobed pinnules were selected. A sample containing only axes and another containing only pinnae and pinnules (Fig. ld) were used for chemical analyses. Our results concur with those of DiMichele et al. (1984) who observe that 'pinnule venation is difficult to detect because the natural maceration process has removed most of the tissue of the leaf lamina, leaving the cuticle'. Thus, although the complete pinnules were included, we consider that the material we analysed consisted almost entirely of cuticle (Fig. ld). The axis material probably included rachides, petioles and stems and these did contain some coalified material in addition to the cuticle. A full description of the Karinopteris material is provided by DiMichele et al. (1984). Taylor et al. (1989) studied the ultrastructure of the cuticle (cuticular membrane) which ranged from 3-6/~m in thickness and consisted of three regions. An outermost polylamellate region up to 1.0/~m thick included convoluted lucent lamellae of 8-10 nm and opaque lamellae ranging from 15-125 nm thick. The bulk of the cuticle consisted of a fairly homogeneous region. The innermost region was a narrow fibrillar zone. W. A. Taylor et al. (1989) assigned this cuticle to type 1 of Holloway (1982) and noted that in extant plants the fibrillae are believed to consist of cellulose, hemicellulose and pectin. W. A. Taylor et al. (1989) noted that of the three
41
fossils they studied, fibrillae were best preserved in the Karinopteris. In their detailed ultrastructural survey of two modern cuticular membranes using SEM, Hill & Dilcher (1990) considered that the highly aliphatic biopolymer (recognized, and termed cutan, by Nip et al. 1986a, b; see also Tegelaar et al. 1991) might be found as part of the outer lamellate region, and as part of the framework and the matrix in the central region (which they showed to be composed of a small-scale complex meshwork). Spencer & DiMichele (1987) give a brief report on solvent extractable compounds from these Karinopteris cuticles. They record alkanes, triterpenoids, fatty acids and ketones and consider that these indicate a high degree of preservation of cuticular surface and soluble components. Nip et al. (1989, 1992) have studied the chemical structure of the fresh cuticle-rich coal, the weathered paper coal and the cutinite maceral fraction from the coal concentrated by density gradient centrifugation. Their results are discussed in our results section. (c) Medullosan pteridosperms: Indiana coal balls This material was collected from the roof of the Springfield No. 5 Coal at two closely linked localities, the Eby Pit and the Lemmon Pit, both part of the Lynville Mine in the Illinois Basin coal field. The material is Middle Pennsylvanian (Upper Carboniferous) in age, for details see Eggert et al. (1983). Middle Pennsylvanian coal swamps were dominated by lycophytes but medullosans generally contributed 10-15% of coal-swamp peat biomass (DiMichele et al. 1985). In the Upper Pennsylvanian (Stephanian), medullosans
Fig. 1. Sources and specimens of cuticles examined in this study. For further details and details of localities see text. (a) Block of Lower Carboniferous paper coal from the Moscow Basin containing the lycophyte stem cuticles of Eskdalia Kidston. BGS Keyworth specimen 1299, x 1. (b) Piece of stem cuticle isolated from (a) showing perforations where leaves were formerly attached and occasional ligule linings in these perforations, x 3. (c) Block of Upper Carboniferous paper coal, Indiana, USA, containing cuticles of the pteridosperm Karinopteris. x 0.5. (d) Lobed pinnule of Karinopteris isolated from (c). x 6 (e) Portion of pinnule of Reticulopteris type released from (g). × 15. (f) Coal ball with pinna of pteridosperm foliage Alethopteris lesquereuxi Wagner on the surface. Specimen 11940, Baxter Collection, University of Kansas, Upper Carboniferous, USA; figured by Baxter & Wilhite (1969), fig. 4 p. 772. x 0.75. (g) Coal ball with pinnules of pteridosperm foliage on the surface, Springfield Number 5 coal, Upper Carboniferous, Indiana, USA. x 3. (h) Portion of cuticle of neuropteroid type released from (g). x 50. (i) Portion of pinnule from (f) dissected to open out the cuticular envelope and display lower cuticle (with grooves, ridges and hairs) to left and upper cuticle to right, x 10. (j) detail of lower cuticle, with grooves (pale), ridges (dark) and hairs, from (i). × 50. (k) detail of upper cuticle from (i). × 50. (a, c, f, g): light macrophotographs of hand specimens; (b, d, e, h, i, j, k): transmitted light micrographs of specimens mounted in glycerine jelly on slides. Light micrographs produced using a Zeiss SV8 stereo microscope or a Zeiss standard laboratory microscope each fitted with a Zeiss MC 63 photographic apparatus.
42
M . E . COLLINSON ET AL.
contributed up to 20% (Phillips 1981). The peatforming flora of the Springfied No. 5 coal was dominated mainly by lycophytes but, in certain areas, especially along and near palaeochannels, medullosan pteridosperms were dominant or codominant (Phillips 1981; Phillips et al. 1985, p. 77-80). Leaves are usually the most abundant pteridosperm organs in peat biomass, contributing 75-95% of pteridosperm material. In some cases, stems and leaves may be equally important and sometimes rooting systems are also significant. Medullosan-dominated pteridosperm assemblages are largely composed of foliage (73-96% biomass). Medullosans were important in levee communites (and probably other areas with higher ground levels or lower water tables than the swamps) and their remains are frequently encountered in clastic sediments (e.g. those deposited in interdistributary bays) from coal-bearing sequences (Phillips 1981; Cleal & Zodrow 1989; Gastaldo et al. 1989). DiMichele et al. (1991) showed that, in a lower Desmoinesian, late Middle Pennsylvanian age coal-bearing sequence, medullosan pteridosperms accounted for only 4% of the coal-ball peat biomass but that medullosan foliage accounted for 45% of the counts in the compression flora of the roof shale. Thus medullosan pteridosperms are major contributors to coalbearing sequences. Coal balls (Fig. lf, g) are limestone concretions encountered in coal seams which represent peat that was permineralized by calcium and magnesium carbonate, before appreciable compaction or alteration took place (Scott & Rex 1985). Most coal balls are preserved in place in
the coal seam (Scott & Rex 1985) as can be demonstrated by tracing plant tissues from the coal ball into the enclosing coal (Winston 1986). The coal ball thus contains the peat-forming flora and the fine anatomical preservation enables constituent plants and plant organs/ tissues to be readily identified. Studies of coal balls have shown that the plant material is coalified to a rank approximately equal to that of the surrounding coal (Hatcher et al. 1982; Lyons et al. 1985). Coal balls have formed the basis for major ecological surveys of Carboniferous coalforming floras (e.g. Phillips et al. 1985; DiMichele et al. 1985; Scott & Rex, 1985; Cross & Phillips 1990; Phillips & DiMichele 1992). Our Springfield coal balls (Fig. lg) are medullosan dominated and have provided foliage and axial material pertinent to this paper. The coal balls were cut and peeled using standard techniques (Joy etal. 1956; Scott & Rex 1985). Peels were mounted on glass slides with elvacite for photomicroscopy. After peeling, relevant pieces of plant material were isolated from the main coal ball block using a trim saw. After checking to ensure that only the required plant material was included in the sub-sample, it was placed into dilute hydrochloric acid (HCI) to release the organic material. This material was used for chemical analysis. In some cases, bulk coal ball material (Fig. lg) was dissolved in dilute HCI and the residue searched using a low power microscope, in order to select discrete pieces of plant tissue which could be readily identified. These individual pieces were then used for chemical analysis (Fig. l e , h). The most frequently figured medullosan
Fig. 2. Various organs and tissues used for chemical analysis. For further details and details of localities see text. (a) Transverse section of small stem of Diaphorodendron, Peel from Coal ball, Union Seam, Upper Carboniferous, Britain. x 2. (b) Detail from (a) (lower centre) showing, at left, phelloderm of the periderm with cells in radial files, x 10. (c) Radial longitudinal section of a portion of an isolated piece of Stigmaria periderm, (see also figs 2d, f, g). Peel from Coal ball, Union Seam, Upper Carboniferous, Britain. x 2. (d) Detail from (c). x 10. (e) Tuberculatisporites, a sigillarian megaspore, from shales, Upper Carboniferous Swillington, England. x 22. (f) Transverse section of the same piece of Stigmaria periderm as in (c). x 4. (g) Detail from (f). x 10. (h) Detail from (1) showing resin rodlets in situ. x 50. (i) Resin rodlet isolated from shales, Upper Carboniferous, Swillington, England. × 40. (j) Detail of (i) showing impressions of epithelial cells on external surface. × 75. (k) Transverse section of Myeloxylon, a medullosan petiole, Peel from Coal ball, Springfield No 5 coal, Upper Carboniferous, Indiana, USA. × 4.5. (1) Detail from (k) showing distribution of resin rodlets (dark bodies). × 10. (m) Tasrnanites, prasinophyte algal cyst, mounted directly in glycerine jelly from rock sample with no treatment, Permian, Tasmania. × 50. (n) Portion of sporangium from the lepidodendrid cone of Flemingites scottii Jongmans showing microspores of the Lycospora type. Deeply etched block of Pettycur Limestone, Lower Carboniferous, Fife, Scotland. × 500. (o) Detail of single spore from (n). x 1500. (p) Detail of (m) showing radial pore canals. × 200. (a, b, c, d, f, g, h, k, 1): transmitted light micrographs of coal ball peels mounted in elvacite; (f, g): using dark field illumination, others bright field; equipment as in Fig. 1. (e, i, j): scanning electron micrographs of specimens isolated from shales by HC1 and HF treatment. (n, o): scanning electron micrographs of deeply etched surfaces of a limestone block. (m, p): transmitted light micrographs, details as in Fig. 1. SEMs were taken on a Cambridge S-100 scanning electron microscope.
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW
ii~! ,~i ¸¸
43
44
M. E. COLLINSON ET AL.
pteridosperm, Medullosa noei Steidtmann, is reconstructed as a small tree-fern-like plant, up to 5 m high, with a short stem and a crown of large, much dissected fern-like leaves (Stewart & Delevoryas 1956). Recent work has, however, suggested that many medullosan pteridosperms were not self-supporting but were twining, relying on each other or on other stems for support (Wnuk & Pfefferkorn 1984). Stems in these lax plants reached at least 10 m in length. The name Medullosa pertains to one form of stem anatomy. The laminar parts of the large leaves (fronds) include forms named Alethopteris Sternberg and Neuropteris (Brongniart) Sternberg sensu lato. (The genus Neuropteris has recently been revised and species reclassified into four genera (Cleal et al. 1990)). Leaves reached up to 7 m in length (Laveine 1986). Species of both these foliage genera have been found attached to twining Medullosa stems (Wnuk & Pfefferkorn 1984), and a species of Neuropteris (N. ovata Hoffmann) has been found in connection with M. noei stems (Zodrow & Clea11988; Cleal & Schute 1991). A species of Linopteris has been found attached to Sutcliffia, another pteridosperm stem (Meyen 1987). Less direct evidence for the interpretation that all (or most) Alethopteris, Neuropteris, Linopteris and other foliage genera such as Reticulopteris, Odontopteris and Lonchopteris were borne by medullosan pteridosperms comes from the similar anatomy of the leaf stalks (petioles) preserved both on stems and on leaves; from other distinctive anatomical features (hairs, canals, etc.) and from association of stems and foliage in numerous fossil assemblages (Stewart & Delevoryas 1956; Baxter & Willhite 1969; Ramanujam et al. 1974; Wnuk & Pfefferkorn 1984). Furthermore, the form-genera can be placed in couplets which have the same frond form but where one genus has open dichotomous venation but the other has reticulate venation, i.e. Neuropteris/Reticulopteris; Alethopteris/Lonchopteris and Paripteris/ Linopteris (Zodrow & Cleal 1993). There is strong evidence for a phylogenetic link between the Neuropteris obliqua (Brongniart) Zeiller and Reticulopteris frond types and one possible explanation for the evolution of reticulateveined forms is the drier climate in the middle Westphalian (Zodrow & Cleal 1993). For the purposes of this paper we shall consider all of these foliage form-genera as having been produced by medullosan pteridosperms (Trigonocarpales sensu Meyen 1987). We have analysed two types of medullosan material from the Springfield coal balls: resin
rodlets and leaf cuticles. Individual petioles (Myeloxylon Brongniart) from an unidentified species of medullosan pteridosperm (Fig. 2k) were used as the source of resin rodlets (Fig. 21, h). The rodlets were handpicked from petiole residues for SEM and for chemical analysis. These rodlets have been described by Schopf (1939) in Medullosa distellica Schopf, who demonstrated that they were formed in secretory canals with a clear epithellial layer. Schopf noted that 'their resinous nature is inadequately proved' but that 'they are essentially similar to resin in their mode of occurrence within the plant'. The resin rodlets are essentially opaque (Fig. 2h) and those we studied showed no autofluorescence. Schopf (1939) noted that these rodlets can often be isolated from coals and that they are chemically very inert, withstanding drastic chemical oxidation. Resin rodlets are a feature of Myeloxylon petioles (Stewart & Delevoryas 1956) and thus may well have occurred in all medullosan pteridosperms. Lyons et al. (1982, 1984) studied resin rodlets isolated from Upper Carboniferous and Lower Cretaceous shales and coals and Lyons et al. (1986) proposed the new maceral name Secretinite for these objects. We have also analysed isolated resin rodlets (Fig. 2i, j) derived from shales (see Swillington below). As far as we are aware these rodlets are the oldest putative resins and the only well-documented example of a secretory product in Palaeozoic plant material. The medullosan leaf cuticles (Fig. le, h) were picked from residues of bulk macerations. Individual pinnules (smallest frond portions) were identified, on the basis of venetation and cuticle characteristics to one of two types: neuropteroid-type (see Cleal & Zodrow 1989; Cleal et al. 1990) probably Neuropteris rarinervis Bunbury (now Laveineopteris see Cleal et al. 1990) which was described from Springfield coal balls by Oestry-Stidd (1979) and Reticulopteristype, (see Zodrow & Cleal 1993) probably Reticulopteris muensteri (Eichwald) Gothan emend Zodrow & Cleal 1993 which was described from Iowa coal balls by Reihman & Schabilion (1978). The fragments which were analysed consisted either only of an upper or lower cuticle (Fig. lh) or the entire cuticular envelope (Fig. le). In the latter, a small amount of other leaf tissues are assumed to have been present within the envelope.
(d) Medullosan pteridosperms, leaf cuticle: Kansas coal balls This material was provided from the Baxter collection in the University of Kansas,
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW Lawrence, Kansas USA and is from a coal ball from the Fleming Coal of southeast Kansas, Middle Pennsylvanian (Upper Carboniferous) in age. The block was numbered 11940 in that collection. Although the original publication (Baxter & Willhite 1969) did not give specimen numbers, we have recognized the specimen of Alethopteris which we have studied (Fig. lf) as one figured by Baxter & Willhite (1969, fig. 4 p. 772). We have removed a portion of this specimen for study, the original block and our slide preparations will be returned to the Kansas collections. This material was identified (Baxter & Willhite 1969) as Alethopteris lesquereuxi Wagner, a species originally described from compression floras. Permineralized material of the species was further discussed by Reihman & Schabilion (1976a, b) and Mickle & Rothwell (1982). This species has xeromorphic features including stomata sunken in grooves between ridges overlying veins on the lower surface. These can be seen in our preparation (Fig. li, j, k) which displays both the lower and upper cuticle. The specimens which we selected for chemical analysis were fragments of pinnules (Fig. li) (lacking the midrib). We assume that some additional leaf tissues were present within the cuticle envelope, however, the contribution from vascular tissues would have been minimal because lateral veinlets included only 3-5 tracheids (Baxter & Willhite 1969, p. 775). For further details of the significance of coal ball material, methods of study and significance of Alethopteris and medullosan pteridosperms see the section on Indiana coal balls. This material from a well-characterized species of Alethopteris complements our studies from less clearly identified portions of pteridosperm foliage (Indiana Coal Balls and Swillington Shales). (e) Arborescent lycophytes, pteriderm f r o m stem and rooting systems: British coal balls This material is from the tip heap at Deerplay near Burnley, Lancashire. The Deerplay tip comprises coal balls derived from the Union seam, Westphalian A (Upper Carboniferous) in age (Scott & Rex 1985). We have used these coal balls as a source of arborescent lycophyte stems and rooting systems for analyses. Floral lists for British coal balls are given in Phillips (1980) and the arborescent lycophytes contributed 91-95% of the peat biomass in coal balls (over 500 coal balls studied) from the Union Seam (Phillips et al. 1985). Euramerican Upper Carboniferous coal-
45
forming floras (Westphalian/Lower & Middle Pennsylvanian) were dominated by arborescent lycophytes which formed the basic swamp vegetation for some 30 million years (Phillips 1980; DiMichele & Phillips 1985; Phillips et al. 1985; DiMichele et al. 1985, 1986, 1991; Phillips & DiMichele 1992). They often contributed over 80% and rarely less than 60% of the peatbiomass (Phillips et al. 1985; DiMichele & Phillips 1985). DiMichele & Phillips (1985) and Phillips & DiMichele (1992) discuss the ecology and life history of arborescent lycophytes including assessments of the different niches in the swamps occupied by various species. Gastaldo (1987) and Gastaldo et al. (1989) have documented the dominance of arborescent lycophytes in swamps with a clastic substrate, thus emphasizing the importance of the potential lycophyte contribution to coal-bearing sequences in general and not only to coals themselves. Although it would be ideal to have analysed all the different tissues in the axial systems of this extremely important group - the arborescent lycophytes- this has not been possible within the time-scale available for this project. We selected periderm (the term here used to include the phellogen and all tissues it produces, i.e. phellem and phelloderm), which in a lycophyte is dominated by phelloderm, because this may have had an analagous role to that of the periderm (protective bark) of flowering plants in which a highly aliphatic resistant biomacromolecule has recently been recognized (Tegelaar et al. 1990). Phillips & DiMichele (1992) noted that lycophyte periderm was quite different from 'traditional bark' and that the chemical composition of lycophyte periderm cell walls was unknown. The tissue is thought to have consisted of living cells, potentially meristematic, and possibly photosynthetic if exposed to light. The tissue was also probably quite rigid and dense (somewhat 'wood-like'), being the main contributor to self-support in the lycophyte trunk. The abundant occurrence of the tissue in a variety of preservational settings suggests that it was decay resistant and probably relatively impervious when exposed (Phillips & DiMichele 1992). Periderm is the most abundant tissue type produced by arborescent lycophytes (which had very little secondary xylem or true wood) and it is found abundantly as loose pieces in coal ball floras (Phillips & DiMichele 1981; DiMichele et al. 1986). Bark (mainly from lycopod trees) contributed 20-45% of the biomass in Lower
46
M. E. COLLINSON ET AL.
and Middle Pennsylvanian coals (DiMichele et al. 1986). In the Herrin coal (Middle Pennsylvanian, USA) where lycophytes contributed 69-75% of the total peat volume, Stigmaria rooting systems contributed 11-36% of this and aerial vegetative tissues, mostly bark (i.e. periderm) contributed 20-56% (DiMichele & Phillips 1985). In the Hapton Valley Mine (Westphalian A, Britain) where lycophytes contributed 91% of the peat biomass, 46% of the peat biomass consisted of periderm (Phillips et al. 1985). Winston (1986) traced the periderm tissues of lycopod stems and Stigmaria periderm from coal balls into vitrinite in the enclosing coal. It is, therefore, clear that lycophyte periderm made a major contribution to Upper Carboniferous coals and hence the chemical composition of lycophyte periderm is of considerable significance in the context of this paper. The stem material for analysis was taken from the periderm of Diaphorodendron vasculare (Binney) DiMichele (formerly Lepidodendronsee DiMichele & Bateman 1992) which is the dominant lycophyte in many coal balls from the Union Seam (Phillips et al. 1985) and from the coal-swamps in North America during the wetter phases (DiMichele et al. 1985). The specimen we studied is illustrated in Fig. 2a. A detailed account of the anatomy of this species is given by DiMichele (1981). The specimen we analysed shows no products of vascular cambial activity and is from a small diameter stem where the periderm consists of homogeneous phelloderm (Fig. 2b). The periderm does not show the pronounced bifacial differentiation (diagnostic of Diaphorodendron - DiMichele & Bateman 1992; Phillips & DiMichele 1992; Bateman et al. 1992) into an outer homogeneous phellem and an inner, much thicker, phelloderm with alternating tangential bands of thick - and thinwalled cells, which is found in large diameter stems. It is obviously necessary to analyse these two distinct periderm tissues and we intend to do this as part of our future research on the chemical composition of well-characterized plant fossils. The material we analysed from the rooting system, the Stigmaria ficoides Brongniart rhizomorph, is illustrated in Fig. 2c, d, f & g. A detailed account of the anatomy of this species is given by Frankenberg & Eggert (1969). S. ficoides is the rhizomorph of arborescent lycophytes of the families Lepidodendraceae and Diaphorodendraceae (Bateman et al. 1992). Eggert (1961, 1972) also discusses the anatomy of Stigmaria, the latter paper docu-
menting a distinct species representing the rhizomorph of Sigillaria. Details of preparation techniques and general signficance of coal ball material are given in the section on Indiana coal balls. In both Diaphorodendron and Stigmaria special care was taken to include only periderm (by ensuring other tissues including leaf bases were cut away) in the blocks which were taken for analysis.
(f) Lycophyte spores: Pettycur Limestone This material is from a loose block of the true peat (sensu Rex & Scott 1987) from Pettycur Limestone, a permineralized peat of Asbian, late Visean, Lower Carboniferous age from Pettycur, Fife, Scotland. This permineralized peat provides high quality anatomical preservation similar to that in coal balls and preparation techniques are identical (see section on Indiana coal balls). Lycophytes were an important component of this Lower Carboniferous peatforming flora (Rex & Scott 1987) but ferns were also a significant component of some Lower Carboniferous peats (Scott et al. 1984). Specimens of the arborescent lycophyte (lepidodendrid) sporangia, Flemingites scottii Jongmans, which yield the microspore Lycospora (Fig. 2n, o) were used for analysis (Scott & Hemsley 1993). The micrographs show that the material consisted entirely or almost entirely of the spores themselves. In combination, the Tuberculatisporites megaspore material from shales at Swillington (see Swillington shale flora) and this Lycospora material from Pettycur provide a representative sample of the extensive lycophyte spore rain which became incorporated into Carboniferous peats. Whilst these spores may be of relatively little significance in terms of overall peat biomass (e.g. sporophyll units only 1-2% of total lycopod biomass in coal balls (DiMichele & Phillips 1985) they are of major importance in coals (Phillips et al. 1985; Bartram 1987). Spore and pollen walls make up the maceral sporinite of the liptinite maceral group considered to be important for the generation of terrestrially sourced oils (see review in Murchison 1987). (g) Shale flora: Upper Carboniferous, Swiltington, England This material is from bed 20f unit 9 of Scott (1978), a light grey floodplain shale within the coal-bearing sequence at Swillington Brickpit east of Leeds, Yorkshire, England. Full details are given in Scott (1978). The material is Westphalian B, Upper Carboniferous in age. This shale contains a varied flora from which
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW three kinds of material were selected in order that their chemical composition could be compared with that of similar (or identical) material which had been permineralized and preserved in coal balls. The shale was treated in cold concentrated HF to release the plant fossils which were then sorted using a low power microscope and individual items picked for analysis. Resin rodlets (for details see Medullosan pteridosperms: Indiana coal balls) are present in these residues although they were not specifically mentioned by Scott (1978). A SEM illustration of one of these rodlets from Swillington (provided by K. Bartram) is reproduced in Lyons et al. (1986, fig. 5). It shows very clear casts of the epithelial cells (possibly the cells themselves are preserved) which lined the secretory cavity. Our material (Fig. 2i, j) only shows impressions of these cells as slight modifications of the surface topography of the rodlet. On the basis of the structure, size and shape of the rodlets and their known abundance in medullosan petioles, but rarity elsewhere in Carboniferous plants (see Lyons et al. 1982), we assume that these rodlets are secreted products derived from medullosan pteridosperms. Megaspores of Tuberculatisporites cf. mammilarius (Bartlett) Potoni6 & Kremp/T. cf. brevispiculus Potoni6 & Kremp type (Fig. 2e), an arborescent lycophyte (sigillarian) megaspore described in detail by Hemsley & Scott (1991), were selected as an example of spore material. The actual specimens we analysed were not subjected to electron microscopy so we cannot say which of the two distinct exine morphologies figured by Hemsley & Scott (1991) were present in our material. For comments on the significance of spore material see section on Pettycur Limestone. Isolated pinnules of the pteridosperm foliage Alethopteris were selected as an example of cuticle from shale. For details of Alethopteris and the significance of medullosan pteridosperms in coals, see sections on Indiana coal balls and Kansas coal balls. (h) Tasmanites, prasinophyte algal cell walls: Permian, Australia This material comes from a tasmanite-rich coal, La Trobe, Mersey River, northern Tasmania (Newton 1875) and is represented in Royal Holloway College Geology Department rock collections by sample R 1657. A sample of the coal, without any treatment, was used for the chemical analyses. A 13C NMR trace from material taken from the same rock sample is
47
given in Hemsley et al. (in prep.). An almost identical 13C NMR trace from a Tasmanian tasmanite (locality not stated) is given in Wilson et al. (1988), who also figure a very similar trace from an Alaskan tasmanite. De Leeuw et al. (1991) comment briefly on the chemical composition of tasmanites. The coal is composed almost entirely of the disc- like, perforate- walled Tasmanites (T. punctatus Newton), non-motile cysts of a prasinophyte alga (Prasinophyta) comparable with the recent Pachysphaerapelagica Ostenfeld (Tappan 1980). A thin section of the coal is illustrated by Tappan (1980, p. 816) taken from Newton (1875) and this shows that some parts have very closely packed Tasmanites whilst others have intervening material of unknown nature. The coal ('white coal' or 'Tasmanite') accumulated in deposits several feet thick extending laterally for several miles (Newton 1875, cited in Tappan 1980). The depositional environment is not stated but modern prasinophytes are plankton mainly from shallow seas and Murchison (1987) gives a "shallow-marine' depositional setting for Tasmanites oil-shales. The Australian Permian Tasmanites (Fig. 2m, p) are thick-walled (up to 20 ~m) and the walls show perforations typical of prasinophyte cysts (Tappan 1980). The cysts reach about 500/~m in diameter (Tappan 1980, personal observations) and were originally spherical (Tappan 1980). There are various similar tasmanite-rich coals occurring widely in the Palaeozoic and sporadically in the Mesozoic. An early Mesozoic example from Alaska yields 500 litres of oil per tonne (150 gallons per ton) (Tappan 1980) thus demonstrating the significance of this material for oil generation. Hutton (1986, 1987) and Stach et al. (1982) give further details on tasmanite-rich coals. (/) Tetraedron, green algal cell walls: Messel oil-shale, Germany This material comes from the middle Eocene oilshale deposits of lake Messel, Messel near Darmstadt, Germany (Goth et al. 1988; Goth 1990; Schaal & Ziegler 1992). The oil-shales consist of distinct paired laminae (about 100200 ktm thick) with the organic-rich layer of the couplet composed almost entirely of the cell walls of Tetraedron, a single-celled green alga (Chlorococcaceae) (Goth et al. 1988; Goth 1990). The fossil alga is well illustrated in these two cited papers. Pyrolysis GC-MS (770°C Curie temperature) traces of modern and Messel Tetraedron have already been published (Goth et al. 1988; Goth 1990) and are not repro-
48
M. E. COLLINSON ET AL.
duced here. We do, however, take this opportunity to document the 13C NMR traces for both modern and Messel Tetraedron obtained by one of us (JDL). The modern material was taken from a cell culture from the collections of the Institute of Plant Physiology, University of G6ttingen, Germany (strain SAG 44.81, Tetraedron minimum (A. Braun) Hansgirg) (see Goth et al. 1988; Goth 1990) and the material is extensively illustrated by excellent SEMs in Goth (1990). Tetraedron is recorded as a component in palynological organic matter (POM) from freshwater sediments (e.g. Van Geel et al. 1981; Batten & Lister 1988) although its small size (<15/~m) may tend to result in loss if preparations are sieved to remove fine material. Cell walls of other freshwater chlorococcalean green algae including Scenedesmus, Pediastrum and Tetrastrum (Tappan 1980; Collinson 1983; Batten & Lister 1988; Hutton 1988; Fleming 1989) and Botryococcus (see next section) are also found in POM. Therefore, these algae make a general contribution to organic matter in sediments as well as occurring in concentrates in oilshales.
(]) Gloeocapsomorpha, algal cell walls, Ordovician kukersite, Estonia This material comes from the Estonian kukersites (K2 & K3) which were studied by Derenne et al. (1992). These kukersites are oil-shales in which the bulk of the organic matter is made up of colonies of the marine microorganism Gloeocapsomorpha prisca Zalessky which is now considered to be a planktonic green microalga related to Botryococcus braunii KiJtzing (Botryococcaceae, Chlorococcales) but which could adapt to greater variations in salinity (Derenne et al. 1992). The Estonian kukersites cover an area over 50000km 2 with a cumulative thickness of up to 20m; similar deposits occur in the Ordovician of North America and Australia (Derenne et al. 1992). The Gloeocapsomorpha morphology is illustrated using SEM and TEM by Derenne et al. (1992) who demonstrated morphological and chemical variation ('closed form phenol-rich and open form phenol-poor') between material from different sites. Similar morphological changes were shown to mirror salinity changes in cultured Botryococcus braunii with increasing salinity leading to a transition from 'open to closed' forms. Botryococcus-rich oil-shales are also widespread and include examples from the Permian of Autun, France and the Carboniferous of Scot-
land, known as torbanites or boghead coals. The names Pila and Reinschia are sometimes applied to Botryococcus-like fossils in Palaeozoic material. Botryococcus generally occurs in freshwaters today although concentrates are known from modern brackish settings in Lake Balkash (former USSR) and the Coorong area of South Australia, known as balkashites and coorongites respectively. Botryococcus and Botryococcus-like fossils also occur in a wide range of other sediments from the Precambrian onwards. Further details on Botryococcus and Botryococcus-dominated oil shales, may be found in numerous papers (Tappan 1980; Stach et al. 1982; Collinson 1983; Hutton 1986, 1987; Batten & Lister 1988; Wilson et al. 1988; several papers in Fleet et al. 1988; Derenne et al. 1988, 1992; De Leeuw etal. 1991; Sinninghe Damst6 et al. 1993). Thus the cell walls of Gloeocapsomorpha, Botryococcus and Botryococcus-like algal colonies are widespread in sediments and oilshales ranging from marine to freshwater settings. The chemical composition of these cell walls is, therefore, important in the context of this paper. An overview of current evidence is provided by De Leeuw etal. (1991) and Derenne et al. (1992).
Analytical m e t h o d s The samples for chemical analysis were ultrasonically extracted with methanol (3x) and dichloromethane (3 x). The residues were dried in a vacuum stove at 30°C.
(a) Pyrolysis Curie-point pyrolysis-gas chromatography (Py-GC) analyses were performed with a Hewlett-Packard gas chromatograph using a FOM-3LX unit for pyrolysis. The samples were applied to a ferromagnetic wire. The Curie temperature was 770°C or rarely 610°C (see figures). The gas chromatograph, equipped with a cryogenic unit, was programmed from 0°C (5 min.) to 300°C (10 min.) at a rate of 3°C/min. Separation was achieved using a fused-silica capillary column (25 m × 0.32 mm) coated with CP Sil-5 (film thickness 0.4/~m). Helium was used as the carrier gas. Curie-point pyrolysis-gas chromatographymass spectrometry (Py-GC-MS) analyses were performed using the same equipment and conditions as described above for Py-GC. The gas chromatograph was connected to a VG 70S mass spectrometer operated at 70eV with a mass range m/z 40-800 and a cycle time of 1.8 s.
49
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW
(b) 13C NMR 13C solid-state N M R spectra were acquired in London and in the Netherlands. Both were obtained with a Bruker MSL-300 N M R spectrometer using cross-polarization (CP) high-power proton decoupling and magic-angle spinning. All London spectra were acquired under identical conditions, namely using 1 ms contact time and a 1 s recycle delay at spinning speeds of 4.5-5 KHz. The H 90 ° pulse was 4/xs. In order to alleviate the problem of overlapping spinning sidebands the TOSS sideband suppression sequence was used. It is recognized that both CP and TOSS are not necessarily fully quantitative but the choice of experimental conditions is limited to a large extent by the relatively small amounts of sample available. The Dutch spectra (Tetraedron only) were acquired using a Hartman-Hahn contact time of 1 ms and a relaxation delay time of 4 s. The samples were rotated at a frequency of 3 KHz where the axis of rotation was tilted by an angle of 54044 ' with respect to the direction of the magnetic field. The system operated at a laC frequency of 75.468 MHz. The radio frequency spin lock field amplitude was 72KHz. The
spectra were recorded using varying numbers of accumulations in order to increase the signal to noise ratio to an acceptable level. The spectrometer was connected to an Aspect 3000 computer.
Results and potentialfor oil generation The discussion of our results will be restricted to major features i.e. to interpretation of the dominant biomacromolecules. We are aware of potentially significant variations (e.g. in alkane/alkene ratios and distributions) between material but we consider it premature to discuss these in this paper. Interpretation of these variations must await the results of a wider range of analyses. The differences in Curie temperature (770°C or 610°C) do not affect our interpretations of the dominant macromolecules in our materials.
(a) Cuticles The pyrolysates of all the cuticles examined (Figs 4, 6, 9) are dominated by homologous series of n-alk-l-enes and n-alkanes ranging from C6 up to C32, which indicate the presence of the highly aliphatic biomacromolecule cutan
Agave americana cutan PY (770°C)-GC • n-alkanes x n-alk-1-enes
x •
"E
C15
I
.>_
%
G) =_
C25
L.
L
L retention time
Fig. 3. Gas chromatogram of pyrolysate (Curie temperature 770°C) of the resistant, highly aliphatic, biomacromolecule cutan, isolated from modern leaves of Agave americana (See Nip etal. 1986a & b). Cls refers to n-pentadecane and n-pentadec-l-ene.
50
M.E.
COLLINSON ET AL.
x
I
. I
PY
O15
x
×
x
x
Eskdalia Cuticle (770°C)-GC-FID
x
× n-AIk-l-enes
0 9 x
x
• n-AIkanes x
Methylketones
,I
>,
x
•
¢._ G)
II x
.>__
x
II} ¢r
J
+z~
x
z~
I~,~-T/o.oRW
Retention time
Fig. 4. Gas chromatogram of pyrolysate (Curie temperature 770°C) of stem cuticle of the lycophyte Eskdalia, paper coal, Lower Carboniferous, Moscow Basin. Cls refers to n-pentadecane and n-pentadec-l-ene. (See Figs la, b & 5.)
176.510
127.621
30.643
CUTICLE OF ESKDALIA LOWER CARBONIFEROUS RUSSIA
i
2OO
/
100
/
0
PPM
Fig. 5.13C NMR spectrum of stem cuticle of the lycophyte Eskdalia, paper coal, Lower Carboniferous, Moscow Basin. (See Figs la, b & 4.)
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW (cf. Fig. 3) (Nip et al. 1986a & b; Tegelaar et al. 1991; De Leeuw et al. 1991). The pyrolysates of the three medullosan pteridosperms (only Alethopteris figured here) showed slightly different distribution patterns (see van Bergen et al. in press b). The material from the shale samples showed scarcely any differences from that obtained from coal balls. The 13C NMR traces of Eskdalia (Fig. 5) and Karinopteris (Fig. 7) are both dominated by chemical shifts of aliphatic carbons. A slightly lower overall contribution from aliphatic carbons is suggested by the additional resonance at 127 ppm in the Karinopteris NMR spectrum, indicating presence of aromatic carbons. This shift is almost absent in Eskdalia. Microscopic observations (see earlier) also showed that the Eskdalia was the purest cuticle sample we examined, being a sheet of stem cuticle rather than a leaf cuticular envelope. The Karinopteris axial material gave a similar NMR spectrum to that of the pinnules but with a slightly higher contribution of aromatic carbon as would be expected from the microscopic evidence of material enclosed within the cuticles. Nip et al. (1989, 1992) examined the Karinopteris-containing cuticle-rich coal from Indiana using fresh coal, weathered paper coal and their cutinite fractions separated by density gradient centirugation. Their findings, reproduced here in Fig. 8, show an increasing dominance of the alkene/alkane doublets in the pyrolysates from the fresh coal to the paper coal to the cutinite. These results are entirely consistent with the loss of vitrinite from the fresh coal resulting in selective preservation of cuticle in the paper coal (Neavel & GuenneI 1960). They also show that concentration by density gradient centrifugation produces a relatively pure cuticle-rich sample similar to that obtained by selecting pieces of cuticle using microscopy. Tegelaar et al. (1989b) showed, using a simulation experiment, that, upon maturation, cutan yielded n-alkane distributions indistinguishable from those found in naturally occurring high-wax crude oils. We, therefore, conclude that all the cuticles studied here could be major sources of n-alkanes in oils. (b) Algal cell walls
The pyrolystate of the Tasmanites (Fig. 10) is dominated by homologous series of n-alk-l-enes and n-alkanes ranging from C6 to C26, indicating the presence of a highly aliphatic biopolymer, an algaenan. The a3C NMR trac e s from modern and fossil Tetraedron (Figs 11 & 12) are both dominated by shifts reflecting aliphatic carbons. These results agree with those given elsewhere
51
from both Curie-point pyrolysis and a3C NMR on these and other algae (Philp et al. 1982; Derenne et al. 1988, 1992; Goth et al. 1988; Wilson et al. 1988; Zelibor et al. 1988; Goth 1990; De Leeuw et al. 1991; Sinninghe Damst6 et al. 1993; Hemsley et al. in prep.). Recent results suggest that the detailed composition of algaenans varies according to the algal species and/or population or race (Derenne et al. 1992; De Leeuw et al. in progress). This has important implications for distinguishing depositional environments and algal precursors, a possibility which was tentatively suggested by Fukushima & Ishiwatari (1988) following an investigation of modern and oil-shale kerogens (the latter including a torbanite, a tasmanite and a Green River sample each from a distinctive palaeoenvironment). Several papers in Fleet et al. (1988) also addressed this possibility, although based mainly on biomarker evidence rather than data from high molecular weight materials. Sinninge Damst6 et al. (1993) have shown that biomarkers may be unreliable indicators of the major sources of organic matter in lacustrine environments. The pyrolysate of the Gloeocapsomorpha (Fig. 13) also shows homologous series of n-alk1-enes and n-alkanes. However, a large number of peaks eluting before the alkene/alkane doublets reflect the presence of numerous other n-alkenes (with double bonds in different positions and with different stereochemistries) as well as a number of linear chain ketones. For further details of Gloeocapsomorpha chemistry see Derenne et al. (1992). The combined evidence shows that cell walls of the algae Gloeocapsomorpha (Ordovician), Botryococcus (Palaeozoic to Recent), Tetraedron (Cretaceous to Recent), Pediastrum (Cretaceous to Recent), Scenedesmus (Cretaceous to Recent) and Tasrnanites (Palaeozoic; with probable relatives through to Recent) all contain a highly aliphatic resistant biomacromolecule (algaenan). Powell (1986) considered that this highly aliphatic molecule would yield paraffinic hydrocarbons on maturation. Tegelaar et al. (1989b) showed, by means of simulation experiments, that the similar highly aliphatic biomacromolecule, cutan, yielded n-alkanes on artificial maturation. We conclude that the algaenans from the cell walls of all these algae can be major sources of n-alkanes in crude oils. (c) Periderm
The pyrolysates obtained from the lycophyte periderm (Fig. 14) from stems of Diaphorodendron and rooting stocks of Stigmaria are
52
M. E. COLLINSON E T A L .
X
I Kari nopterisi Cuticle i
X
PY (770°C)-GC-FIDi
•
n-AIk-l-enes n-Alkanes
,
Contaminants
×
07
C Q)
rr
+ x
x
x
x
x •
x •
x•
. •
x °
• *
•
I
W.AR-~O.OBW
Retention time
Fig. 6. Gas chromatogram of pyrolsate (Curie temperature 770°C) of pinnules of the pteridosperm Karinopteris, paper coal, Upper Carboniferous, Indiana, USA. Cls refers to n-pentadecane and n-pentadec-l-ene. (See Figs ld and 7.)
30,511
127.084
A
C U T I C L E O F KARINOPTERIS INDIANA PAPER COAL
-
'
'
2"00
'
rl
'
'
'
100~
~
'
'
'
O'
'
~
-
PPM
Fig. 7. t3C NMR spectrum of pinnules of the pteridosperm Karinopteris, paper coal, Upper Carboniferous, Indiana, USA. (See Figs ld & 6.)
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW
53
Freshcoal l Py (770°C)-GC-FID
x.cb~
C20
C1o
coal ]
Paper Py (770°C)-GC-FID
I .l.Jt._q L J.~,.,~,.
C10
C20
Cutinite (by DGC) Py (770°C)-GC-FID
t
*
L,.~tJ_~,jJ_j_J_CLL~_____
J C10
C20
Fig. 8. Gas chromatograms of pyrolysates (Curie temperature 770°C) of the fresh coal, paper coal, and cutinite fraction of the Karinopteris-rich coal, Upper Carboniferous, Indiana, USA. The cutinite fraction was separated by density gradient centrifugation (DGC). Redrawn after Nip et al. (1989, 1992) which should be consulted for further details. C10 refers to n-decane and n-dec-l-ene.
54
M.E.
COLLINSON
ET AL.
× 010
]=
x
'
Alethopteris
PY
x•
Cuticle (770°C)-GC-FID
× n-AIk-l-enes
X
• n-Alkanes * Contaminants
>
02o
II
• )
.4.-* ¢.13 .i-, r"
x
13
x
._> 13 n"
x
,,i
Retention time
Fig. 9. Gas chromatogram of pyrolysate (Curie temperature 770°C) of pinnule fragments of the medullosan pteridosperm cuticle Alethopteris lesquereuxi Wagner isolated from a coal ball, Upper Carboniferous, Kansas, USA. Clo refers to n-decane and n-dec-l-ene. (See Fig. li-k.)
lpy Tasmanite whole rock (61 O°C)-G C- FI D
09 ×
I
,, n-AIk-l-enes
• n-Alkanes x
* Contaminants x × 015
._z-
x
t13 .i-., t(1)
'% x, x" •
. >B
4..,
• ו
13
•
,
nr"
Retention time
T,~.~.~
Fig. 10. Gas chromatogram of pyrolysate (Curie temperature 610°C) of whole-rock rich in Tasmanites a prasinophyte algal cyst, Permian, Australia. Cxs refers to n-pentadecane and n-pentadec-l-ene. (See Fig. 2m &p.)
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW
55
30.766
ALGAENAN ISOLATED FROM
TETRAEDRON MINIMUM
i
~oo
.
.
.
.
loo
'
'
i
i
i
0
PPM
Fig. 11.13C NMR spectrum of algaenan isolated from cell walls of a culture of the modern green alga Tetraedron minimum (A. Braun) Hansgirg. (For details see text.)
30.766
ISOLATED ALGAL-RICH LAYERS MESSEL OIL SHALE
200
100 PPM
0
Fig. 12.13C NMR spectrum of the Tetraedron-rich layers of the Messel oil-shale, Eocene, Germany, (For details see text.)
56
M. E. COLLINSON ET AL. Kukersite
/
Gloeocapsomorphaprisca PY (610*C)-GC-FID
x
If)
E
C9 x
,, n-AIk-l-enes • n-Alkanes
×
×
×
Retention time
~"~'
Fig. 13. Gas chromatogram of pyrolysate (Curie temperature 610°C) of kukersite whole rock, dominated by Gloeocapsomorpha algal colonies, Ordovician, Estonia. (For details see text.) C9 refers to n-nonane and n-non-l-ene.
virtually identical. They show homologous series of n-alk-l-enes and n-alkanes indicating the presence of a highly aliphatic biomacromolecule. A similar macromolecule (informally termed 'suberan') has recently been recognized in the bark of the modern flowering plant Betula (Tegelaar et al. 1990). Using cupric oxide oxidation Logan & Thomas (1987) recorded lignin derivatives from compression fossils of the outermost layers of the stem in several Carboniferous arborescent lycophytes. They recognized very small amounts (0.01 mg g_l or less) ofp-hydroxybenzaldehyde, vanillin and acetovanillone in Lepidodendron Sternberg and Lepidophloios aculeatum laricinus Sternberg. Much larger quantities (13-57mgg -1) of these oxidation products were produced by Sigillaria ovata Sauveur and this was attributed to the fact that 'the decorticated tissues of Sigillaria contain an inner layer of periderm whose thickened walls must be liberating the oxidation products' (Logan & Thomas 1987, p. 166). Apart from the alkene/alkane doublets the pyrolysates of the lycophyte periderms we have studied also include very significant contri-
butions from alkylbenzene and naphthalene derivatives. Although it might be suggested that these compounds originated from lignin, the structures of these aromatic pyrolysates are very difficult to relate to those of lignins which have been transformed by diagenesis (van Bergen et al. in prep.). These compounds may represent an additional substance of unknown origin contained within periderm or possibly a diagenetic alteration product of a purely aliphatic macromolecule (aliphatic moieties having undergone internal cyclization and subsequent aromatization). Our results contrast with those of Logan & Thomas (1987) but we are confident that our pyrolysis techniques would reveal even low amounts of phenols (see e.g. van Bergen et al. in prep.). The tissues we examined are precisely identified microscopically as periderm (Fig. 2a-d, f, g) but the exact tissues present in compression fossils are difficult to ascertain. In these lycophytes they would certainly have included more material than periderm alone, e.g. leaf cushions and leaf scar strips (Logan & Thomas 1987). Furthermore, only the Sigillaria compression fossils yielded significant lignin deriva-
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW
57
Diaphorodendron i
C2
Periderm
PY (770°C)-GC-FID
× n-AIk-l-enes • n-Alkanes • Contaminants
010 x t
• ~
cl
•
.c
:
I1)
.> 4-.'
To.' 0 ×
(9
X x °
¢r
•
•
x
•
•
•
•
*
LP..VA'nO.m~
Retention time
Fig. 14. Gas chromatogram of pyrolysate (Curie temperature 770°C) of periderm from the stem of the arborescent lycophyte Diaphorodendron vasculare (Binney) DiMichele, isolated from a coal ball, Upper Carboniferous, Britain. Cls refers to n-pentadecane and n-pentadec-l-ene. (See Fig. 2a & b.)
tives and we have not yet studied this genus using pyrolysis. (d) Spores
The pyrolysates of the spores we have studied (Figs 15-17) contain homologous series ofn-alk1-enes and n-alkanes characteristic of highly aliphatic moieties. However, in contrast to the cuticles, periderm and algal cell walls, these are not the dominant pyrolysis products and they are only revealed clearly by mass chromatography (Fig. 16). The dominant pyrolysis products of the spore materials are derived from aromatic compounds. These products fall into three classes - alkylbenzenes, alkylphenols, naphthalenes and phenanthrenes. The relative intensity of each of these classes varies between spores but the basic constituents remain the same. Comparison of these pyrolysis data with those obtained from extant (Fig. 18) (Schenk et al. 1981; van Bergen et al. in press a) and fossil spores (van Bergen et al. in press a) reveals considerable differences. However, it should be noted that the chemical structure of resistant spore wall material, sporopollenin, is still far from understood. Hence, the pyrolysis results
presented here can be explained in two ways. Firstly, the pyrolysis products reflect the original, or slightly diagenetically altered, biomacromolecule of the Carboniferous spore walls. This would imply a previously unrecognized sporopollenin type. Secondly, as an alternative explanation, the suites of products are derived from a sporopollenin similar to that recently recognized by van Bergen et al. (in press a). In that case, the different aromatic products are thought to be due to significant diagenetic alteration causing loss of functionalities and increased cyclization leading to more alkylbenzene and naphthalene/phenanthrene moieties. The aliphatic moiety present may be selectively enriched during diagenesis (van Bergen et al. in press a). However, it should be noted that Nip et al. (1992) suggested that aliphatic constituents could be introduced by migration from other materials. (e) Resins
The pyrolysates of the two resin rodlet samples are rather different from one another. In the sample derived from isolated rodlets dispersed in shales (Fig. 19), most of the n-alkanes are probably not pyrolysis products, because their
58
ET AL.
M. E. C O L L I N S O N
j
ITuberculatisp°rites
.....
/ Megaspore / TIC ~ Y (7700C) -GC'Ms
OH
@c~
2
!
× n-AIk-l-enes ~
cl
• n-Alkanes
H
* Contaminants
p: 0
E
=>
×
•
_
1,000
2,000
Scan number
3,000
Fig. 15. Total ion current (TIC) of pyrolysate (Curie temperature 770°C), of the megaspore wall of the sigillarian spore isolated from a shale, Upper Carboniferous, Brtain. C8 refers to n-octane and n-oct-l-ene. (See Figs 16 & 2e.)
Tuberculatisporites,
Tuberculatisporites Megaspore 55+57 PY (770°C)-GC-MS
m/z
010
x,o
x n-AIk-l-enes • n-Alkanes
i
, Contaminants
x
E
I
/
I
xl
:
•
•
x
]
C 20
•
•
x
1,000
•
•
•
•
•
x
~
i
I
2,000
3,000
x
•
~i
•
•
Scan number
m/z Tuberculatisporites,
Fig. 16. Summed mass chromatogram of 55 + 57 indicative of n-alk-l-enes and n-alkanes from the megaspore wall of the sigillarian spore isolated from a shale, Upper Carboniferous, Britain. Czo refers to n-icosane and n-icos-l-ene. (See Figs 15 & 2e.)
59
O I L - G E N E R A T I N G P O T E N T I A L OF PLANTS: A R E V I E W OH PY OH
(770°C)-GC-FID
o~Cl I~
Lycospora
Microspores
OH
x n-AIk-l-enes • n-Alkanes • Contaminants
+ Prist-" -ene I1)
C 8 x
x
x x
• x
•
"
• ×
• ×
• × ,
*
*
Retention time
Fig. 17. Gas chromatogram of pyrolysate (Curie temperature 770°C) of a bulk sample of spore walls of
Lycospora type isolated from a sporangium of the lycophyte cone Flemingites scotii Jongmans, Lower Carboniferous, isolated from Limestone block, Pettycur, Scotland. C8 refers to n-octane and n-oct-l-ene. (See Fig. 2n & o.)
~O
Saivinia (Megaspores) Extant
OH
Py (610°C)-GC-FID • Contaminants
OH OH
4;0 ¢-
~-
q6 FA %FA'~
Retention time
Fig. 18. Gas chromatogram of pyrolysate (Curie temperature 610°C) of megaspores of the water fern Salvinia, modern material. (For details see Van Bergen etal. in press a). C16 refers to hexadecanoic acid and C18 refers to saturated and unsaturated octadeconoic acid. (Note the gas chromatogram is normalized on acetophenone.)
60
M. E. COLLINSON E T A L .
'Resin' rodlets Shale I~g (610°C)-GC-FID • n-Alkanes • Contaminants
~
OH
t'-
¢
"
'
"
._>
ix" •
•
o,
•
~-aT.anw
Retention time
Fig. 19. Gas chromatogram of pyrolysate (Curie temperature 610°C) of 'resin' rodlet isolated from shales, Upper Carboniferous, England. (See Fig. 2i & j.) OH
~cl
Medullosan 'Resin' TIC PY (610°C)-GC-MS
x n-AIk-l-enes • n-Alkanes • Contaminants
OH
6
¢
OH
i ~
tt-
.>_
C2
x
1,000
•
Scan number
•
x
2,000
x~
x
•
•
•
•
•
•
*
•
•
3,000
Fig. 20. Total ion current (TIC) of pyrolysate (Curie temperature 610°C), of 'resin' obtained from petioles of a meduUosan pteridosperm, isolated from a coal ball, Springfield No. 5 coal, Upper Carboniferous, Indiana, USA. (See Fig. 2h, k & 1.)
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW unsaturated counterparts, the n-alkenes, are absent or relatively unimportant. Although the sample was extracted extensively, the series of n-alkanes remains and so must be present as such, but not extractable for reasons which are not understood. One possibility is that low molecular weight material is occluded within the rodlet but an impenetrable exterior prevents solvent penetration. This suggestion finds support in several examples illustrated by Lyons et al. (1982, 1986) showing vesciculate and fractured internal features in rodlets surrounded by a continuous outer zone. It seems that this sample cannot be used to understand the nature of the original macromolecule. Much more work is needed to discover if any of the pyrolysis products represent resistant, high molecular weight material and if the rodlet internal structure would enable occlusion of low molecular weight material. The pyrolysate of the sample of resin rodlets derived from the medullosan petiole (Fig. 20), contains homologous series of n-alk-l-enes and n-alkanes suggesting the presence of a highly aliphatic moiety (unlike most resins Anderson & Winans 1991). The presence of alkylphenols is also unlike that of any known resin. It is rather remarkable that the pyrolysate of these resin rodlets shows a striking resemblance to the pyrolysate of a sub-bituminous coalified wood recorded by Hatcher et al. (1992). This puzzling resemblance requires further investigation. The medullosan resin rodlets may represent a novel resin chemistry or they may be formed from a distinct molecule of phenolic origin. Internally occluded products may have contributed n-alkanes to oils, as may the aliphatic biomacromolecule upon diagenesis, however, only in very small amounts.
Discussion and conclusions Most of the terrestrially sourced oils described in the literature (see review above) have source coals and coal-bearing strata for which the parent plants are poorly understood or unknown. Thus the inferred contribution of terrestrial plants to oils is based on biomarker evidence (especially in the case of resins and certain algae) or isotopic evidence (e.g. McKirdy et al. 1986; van Aarssen et al. 1990, 1992) rather than on the known potential for oil generation of the plant tissues and organs which produced the source coals and other organic-rich sediments. Tertiary, Mesozoic and Palaeozoic coals are known to yield oil upon pyrolysis in simulation experiments (Bertrand et al. 1986; Hetrnyi &
61
Sajg6 1990; Mukhopadhyay et al. 1989, 1991) but it has been noted that their potential yield does not readily correlate with composition as reflected in conventional maceral terms (Bertrand 1984; Bertrand et al. 1986). Very few oils which are thought to have terrestrial sources are derived from Palaeozoic source rocks (see earlier). In order to address these problems we have undertaken chemical analyses of wellcharacterized fossils of the plant tissues and organs which are known to have contributed extensively to Carboniferous coals and coalbearing sequences. We have also analysed some other materials and drawn on the available literature for appropriate results obtained elsewhere. Our attention has been focused on the resistant, high molecular weight, plant material which becomes selectively enriched in the geosphere. All of our analyses are based on plant material which can be characterized and identified (for example: as a Tetraedron (algal) cell wall, as an Alethopteris (pteridosperm) leaf cuticle, as a Diaphorodendron (lycophyte) periderm) using microscopy. We have shown that these materials are either dominantly highly aliphatic in nature or include highly aliphatic macromolecules. The plant tissues and organs which contain these macromolecules and their distribution through time is summarized in Table 2. All the highly aliphatic biomacromolecules have the potential to generate n-alkanes which are found in crude oils. In addition, the resins are capable of generating certain cyclic hydrocarbons occassionally found in crude oils. Given et al. (1980) stressed the importance of knowledge of the nature of parent plant material which contributed to coals and emphasized that this changed through time structurally and chemically. We earlier briefly reviewed the major variations seen during the evolution of coal-forming floras. We documented the relative contribution and significance in coals and associated sediments of each of the materials for which we undertook chemical analyses. Combining the structural and chemical evidence shows that Carboniferous coals include high amounts of oil-prone plant tissues, especially arborescent lycophyte periderm, lycophyte stem cuticles and pteridosperm leaf and axial cuticles. In addition, spore walls and resin rodlets would also contribute. Thus, overall, these coals are highly oil prone. The contribution of secondary xylem tissues (not yet studied chemically) was minimal. We tentatively suggest here, based on preservational, not direct
62
M. E. COLLINSON E T A L . Table 2. Spot occurrences proven by chemical analyses, and inferred ranges in time, o f resistant aliphatic biomacromolecules in various plant tissues and organs. Information taken from results presented here and papers cited in the appropriate sections o f the text
Resistant aliphatic biomacromolecules in: Time-scale
Quaternary --~
"
"
"
"
!
q~
~
Tertiary
Cretaceous
• 100Ma
Jurassic '200 Ma Triassic
Permian
qb ib
i
• 300Ma
Carboniferous .400 Ma
~
,p 9 ,b
i
Devonian Silurian Ordovician
chemical evidence (work in progress), that lycophyte leaves were unimportant as sources of potential n-alkanes in oils. Our results for arborescent lycophyte periderm show that it is not possible to merely extrapolate chemical composition from anatomy 'e.g. thick-walled cells were lignified' but indicate that all tissues of extinct plant groups need to be analysed chemically. Similarly, Tegelaar et al. (1991) have clearly shown that the presence of a cuticle does not necessarily imply the presence of cutan. (Work is in progress by our group on other coal ball plants and on material from other permineralizations and compression fossils in order to expand our understanding of the chemical composition of ancient plants.) These observations make it dangerous to speculate on the oil-generating potential of coals for which chemical analyses like those presented here are lacking. However, it does seem likely that cuticle-rich coals, by virtue of their mere presence in the fossil record, are dominated by cuticles which are selectively preserved and are thus likely to be composed of the highly resistant aliphatic biomacromolecule cutan. Thus the Devonian cuticle coals in China (Dexin 1989; Sheng et al. 1992) might well be oil prone. In view of the fossil record of cuticular sheets (Selden & Edwards 1989) it is at least possible
that 'cutan-sourced' oil might be generated in Ordovician strata, though we stress that until appropriate chemical analyses are undertaken the earliest evidence for cutan is the Lower Carboniferous material investigated herein. Coals dominated by algal cell walls are clearly oil prone and have been produced since at least the Ordovician. There are two major differences in parent plant biology between Carboniferous coals and late Cretaceous and Tertiary coals. These are the presence in the latter of large volumes of secondary xylem in trunks and branches and reproduction by seeds rather than spores. On the basis of chemical analyses of fossil examples (Triassic-Recent from many areas and including material of taxodiaceous and araucarian conifers) the dominant composition of secondary xylem was lignin which becomes transformed during diagenesis (Sigleo 1978; Stout et al. 1989; Stout & Spackman 1989; Hatcher 1988, 1989), is aromatic in nature and is, therefore, not oil prone. Seeds have scarcely been analysed chemically, but our own work in progress shows that fossil seed coats of flowering plants can include two distinct layers, an outer layer with transformed lignin chemistry (van Bergen et al. in prep.) and an inner layer with 'cutan-like' chemistry (van Bergen et al. 1993). Megaspore membranes in gymnosperm seeds
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW exhibit a 'sporopollenin-like' chemistry (Hemsley et al. in prep). We know of no chemical analyses of the high molecular weight fraction of gymnosperm sclerotesta (thickwalled seed coat). Various leaves from Permian, Mesozoic and Tertiary gymnosperms and flowering plants have been shown to contain cutan (Tegelaar et al. 1991). Resins from gymnosperm and flowering plant trees were clearly important contributors to Mesozoic and Tertiary coals, organic-rich sediments and terrestrially sourced oils. Therefore, late Cretaceous and Tertiary coal-forming plants clearly produced oil-prone tissues and organs but there are no studies of late Cretaceous or Tertiary coal-forming floras which combine a chemical and microscopical approach such as that which we have undertaken here for the Carboniferous examples. Such studies will be necessary to demonstrate the exact nature of the dominant elements in coals and coal-bearing strata and hence assess their oil-generating potential. Bertrand (1984) and Bertrand et al. (1986) suggested that the chemical diversity of vitrinites might explain the lack of clear correlation between maceral composition and simulated oil yield in coals. We have seen that periderm tissues, which would probably be included in the vitrinite maceral group, are infact oil prone. Equally, spore walls (placed in the maceral sporinite in the liptinite maceral group, often considered as highly oil prone alongside cutinite, resinite, alginite, etc.) have a much lower potential for oil generation than the cuticles, algal cell walls and periderms which we have studied. Mukhopadhyay et al. (1989, 1991) suggested that depositional environment is an important parameter in distinguishing oil-prone and nonoil-prone coals. Based on a variety of experimental results from Tertiary and Carboniferous coals these authors argued that 'mixinitic' coals, formed in mixed swamp/marsh settings with mainly herbaceous vegetation, have a higher hydrocarbon potential than humic coal (formed from mainly arboreal [sic] swamp vegetation). Our results show that the biology of the source plants is equally important in determining hydrocarbon potential because a Carboniferous coal, formed under an arborescent lycophyte swamp, would be expected to contain a high proportion of lycophyte periderm which is oil prone. In contrast, a Tertiary coal like that said to be formed under a N y s s a / T a x o d i u m swamp (Mukhopadhyay et al. 1989, 1991) should contain a large proportion of secondary xylem, a lignified tissue which is not oil prone. We also
63
predict considerable additional variations according to the amount of cutan-containing leaves shed by the plants. We conclude that a combination of depositional environment and parent plant community will control the oil-generating potential of coals. Whilst the range of the former might be regarded as relatively constant through time the latter most certainly are not. We have shown that Carboniferous plant communities, which formed peats and contributed to organic-rich shales, produced a variety of organs and tissues in abundance which could yield n-alkanes in crude oils. Therefore, the reasons for the absence of Carboniferous terrestrially sourced oils must be sought in geological or exploration factors, not in the absence of oil potential in the coals themselves. Larter (1990) argued that a high expulsion efficiency from source beds was a prerequisite for oil sources. Tertiary and Mesozoic coalbearing strata are thought to generate oils (see review above). Durand et al. (1983) suggested that oil had been generated from Palaeozoic coals but had been dissipated by migration and hence had not accumulated as oil pools. An alternative explanation is that oil was generated in Palaeozoic coals but was not expelled. Subsequent catagenesis would lead to cracking and generation of gas capable of expulsion in rocks of lower expulsion efficiency (Larter 1990). The gas-prone nature of Carboniferous coals would thus be due, not to their dominantly vitrinite- and sporinite-rich nature (Horsfield 1990), but to the generation of oil, from cutan, algaenan, periderm, etc., which was not expelled as oil but, after further cracking of the n-alkanes during catagenesis, yielded gas which was then expelled. The general low abundance of secondary xylem (wood) tissues in Carboniferous peat-forming plants might result in different oil expulsion efficiencies in Carboniferous versus younger coals. However, it is possible that expulsion efficiency in all coals is too low to yield oil pools whatever the chemistry of the dominant plant materials. If this is the case, the sources for Mesozoic and Tertiary terrestrially sourced oils may be from organicrich sediments rather than coal themselves. An extended chemical survey of coals and organic-rich strata and their constituent plants through time is necessary in order to develop widely applicable models concerning oil potential from coal and coal-bearing strata. This study was funded by the NERC Biomolecular Palaeontology special topic grant number GR/02/536 (to MEC & JDL) which is gratefully acknowledged.
64
M. E. COLLINSON ET AL.
We thank Patrick Barrie of the University of London IRS solid state NMR facilitylat University College London for recording the C NMR spectra of Eskdalia and Karinopteris and D O W chemicals (Terneuzen, the Netherlands) for recording the 13C NMR spectra of the Tetraedron and Tetraedron-rich Messel oil-shale samples. In the Geology Department, Royal Holloway University of London, we thank the
photographers for macrophotography of selected specimens and Deborah Parsons for the SEM studies. We thank Ben Van Aarssen, Alan Hemsley and Erik Tegelaar for valuable discussions. This work was undertaken by MEC whilst in receipt of a Royal Society 1983 University Research Fellowship which is gratefully acknowledged. This is Delft Organic Geochemistry Unit Contribution 289.
R e f e r e n c e s
VAN AARSSEN, B. G. K. 1992. Structural assignment
BATEMAN, R. M., DIMICHELE, W. A. & WILLARD,
and catagenetic behaviour of macromolecular constituents of recent and fossil natural resins. PhD thesis, University of Delft, The Netherlands. --, Cox, H. C., HOOGENDOORN, P. & DE LEEUW, J. W. 1990. A cadinene biopolymer in fossil and extant dammar resins as a source for cadinenes and bicadinenes in crude oils from South East Asia. Geochimica et Cosmochimica Acta, 54, 3021-3031. - & DE LEEUW. 1992a. Structural elucidation of polymeric diterpenoids in fossil gymnosperm resins by means of pyrolysis combined with GC-MS. Preprints of the American Chemical Society, Division of Fuel Chemistry, 36,774-780. - & -1992b. High-molecular-mass substances in resinites as possible precursors of specific hydrocarbons in fossil fuels. Organic Geochemistry, 19, 315-326. , --, COLLINSON,M. E., BOON, J. J. & GOTH, K. in press. Occurrence of polycadinene in fossil and recent resins. Geochimica et Cosmochima Acta. , HESSELS, J. K. C., ABBINK, O. A. & DE LEEUW, J. W. 1992. The occurrence of polycyclic sesqui-, tri-, and oligoterpenoids derived from a resinous polymeric cadinene in crude oils from southeast Asia. Geochimica et Cosmochimica Acta, 56, 1231-1246. ANDERSON, J. A. R. 1983. The tropical peat swamps of western Malesia. In: GORE, A. J. P. (ed.) Mires: swamp, bog, fen and moor. Ecosystems of the World (editor-in-chief: D. W. Goodall) Volume 4B. Elsevier, Amsterdam, 181-200. -& MULLER, J. 1975. Palynological study of a Holocene peat and a Miocene coal deposit from N. W. Borneo. Review of Palaeobotany and Palynology, 19,291-351. ANDERSON, K. B. & MACKAY, G. 1990. A review and reinterpretation of evidence concerning the origin of Victorian Brown Coal. International Journal of Coal Geology, 16,327-347. -t~ WINANS, R. E. 1991. Structure and structural diversity in resinites as determined by pyrolysisgas chromatography-mass spectrometry. Preprints of the American Chemical Society, Division of Fuel Chemistry, 36,765-773. BARTRAM, K. M. 1987. Lycopod succession in coals: an example from the Low Barnsley Seam (Westphalian B), Yorkshire, England. In: Sco'rr, A. C. (ed.) Coal and Coal-bearing Strata: Recent Advances. Geological Society, London, Special Publication, 32, 187-199.
D. A. 1992. Experimental cladistic analysis of anatomically preserved arborescent lycopsids from the Carboniferous of Euramerica: an essay on paleobotanical phylogenetics. Annals of the Missouri Botanical Gardens, 79,500-559. BATTEN, D. J. & LISTER,J. K. 1988. Early Cretaceous dinoflagellate cysts and chlrococcalean algae from freshwater and low salinity palynofacies in the English Wealden. Cretaceous Research, 9, 337-367. BAXTER, R. W. & WILLHITE, M. R. 1969. The morphology and anatomy of Alethopteris lesquereuxi Wagner. The University of Kansas Science Bulletin, 48,767-783. VAN BERrEN, P. F., COLLINSON,M. E. & DE LEEUW, J. W. in press a. Chemical composition and ultrastructure of fossil and extant salvinialean microspore massulae and megaspores. Grana. --, SINNINGHEDAMSTI~, J. S. & DE LEEUW, J. D. 1993. Chemical and microscopical characterization of inner seed coats of fossil water plants. Geochimica et Cosmochimica Acta, 57, 5327-5335. --, GONI, M., COLLINSON, M. E., BARRIE, P. & DE LEEUW, J. D. in prep. Chemical characterization of the outer seed coats in fossil and extant water plants. Geochimica et Cosmochimica Acta. --, ScoTt, A. C., BARRIE, P. J., DE LEEUW, J. W. & COLLINSON, M. E. in press b. The chemical composition of Upper Carboniferous pteridosperm cuticles. Organic Geochemistry. BERTRAND, P. 1984. Geochemical and petrographic characterization of humic coals considered as possible oil source rocks. Organic Geochemistry, 6,481-488. --, BEHAR, F. • DURAND, B. 1986. Composition of potential oil from humic coals in relation to their petrographic nature. Organic Geochemistry, 10, 601-608. BOULTER, M. C., HUBBARD,R. N. L. B. & KVACEK,Z. 1993. A comparison of intuitive and objective interpretations of Miocene plant assemblages from north Bohemia. Palaeogeography, Paleoclimatology, Palaeoecology , 101, 81-96. BROWN, S. 1989. The "mangrove model", can it be applied to hydrocarbon exploration in Indonesia? Proceedings of the Indonesian Petroleum Association, 1989, 13.04,385-401. CARATINI, C. & TtSSOT, C. 1988. Paleogeographical evolution of the Mahakam delta in Kalimantan, Indonesia during the Quaternary and late Pliocene. Review of Palaeobotany and Palynology, 55,217-228.
OIL-GENERATING POTENTIAL OF PLANTS: A REVIEW CLEAL, C. J. & SHUTE, C. H. 1991. The Carboni-
ferous pteridosperm frond Neuropteris heterophylla (Brongniart) Sternberg. Bulletin of the British Museum Natural History, (Geology), 46, 153-174. --, & ZOOROW, E. L. 1990. A revised taxonomy for Palaeozoic neuropteroid foliage. Taxon, 39, 486-529. -& ZODROW, E. L. 1989. Epidermal structure of some medullosan Neuropteris foliage from the Middle and Upper Carboniferous of Canada and Germany. Palaeontology, 32,837-882. COLE, J. M. 1987. Some fresh/brackish water depositional environments in the S. E. Asian Tertiary with emphasis on coal bearing & lacustrine deposits and their source rock potential. Proceedings of the Indonesian Petroleum Association 1987, 11/05,429-449. COLLINSON, M. E. 1983. Palaeofloristic assemblages and palaeoecology of the Lower Oligocene Bembridge Marls, Hamstead Ledge, Isle of Wight. Botanical Journal of the Linnean Society, 86, 177-225. - - & ScoTr, A. C. 1987. Implications of vegetational change through the geological record on models for coal-forming environments. In: ScoTt, A. C. (ed.) Coal and Coal-bearing Strata: Recent Advances. Geological Society, London, Special Publication, 32, 67-85. CROSS, A. T. t~ PHILLIPS, T. L. 1990. Coal-forming plants through time in North America. International Journal of Coal Geology, 16, 1-46. DAMBERGER, H. H. 1991. Coalification in North American Coalfields. In: GLUSKOTER, H. J., RICE, D. D. & TAYLOR, R. B. (eds) Economic Geology, US The Geology of North America, P-2, Geological Society of America, Boulder, Colorado, 503-522. DE LEEUW,J. W. & LARGEAU,C. in press. A review of the major classes of organic compounds that comprise living organisms and the pathways for their incorporation into sediments. In: ENGEL, M. H. & MACKO, S. A. (eds) Organic Geochemistry. Plenum Publishing Corporation. --, VAN BERGEN, P. F., VAN AARSSEN, B. G. K., GATELLIER, J-P. L. A., SINNINGHEDAMSTE, J. S. & COLLtNSON, M. E. 1991. Resistant biomacromolecules as major contributors to kerogen. Philosphical Transactions of the Royal Society of London, B333, 329-337. DERENNE, S., LARGEAU, C., CASADEVALL, E. & CONNAN, J. 1988. Comparison of torbanites of various origins and evolutionary stages. Bacterial contribution to their formation. Cause of the lack of botryococcane in bitumens. Organic Geochemistry, 12, 43-59. --,
METZGER, P., LARGEAU,C., VAN BERGEN, P. F., GATELLIER, J. P., SINNINGHE DAMSTt~, J. S.,
DE LEEUW,J. W. & BERKALOFF,C. 1992. Similar morphological and chemical variations of Gloeocapsomorpha prisca in Ordovician sediments and cultured Botryococcus braunii as a response to changes in salinity. Organic Geochemistry, 19, 299-313.
65
DEXIN, HAN. 1989. The features of Devonian coalbearing deposits in South China, the People's republic of China. International Journal of Coal Geology, 12,209-223. DIMICHELE, W. A. 1981. Arborescent lycopods of Pennsylvanian age coals. Lepidodendron, with description of a new species. Palaeontographica Abteilung, B175, 85-125. -& BATEMAN, R. M. 1992. Diaphorodendronaceae, Fam. Nov. (Lycopsida: Carboniferous): systematics and evolutionary relationships of Diaphorodendron and Synchysidendron, Gen. Nov. American Journal of Botany, 79,605-617. -& PHILLIPS, T. L. 1985. Arborescent lycopod reproduction and paleoecology in a coal-swamp environment of late middle Pennsylvanian age (Herrin Coal, Illinois, USA). Review of Palaeobotany and Palynology, 44, 1-26. & MCBRINN, G. E. 1991. Quantitative analysis and paleoecology of the Secor Coal and roof-shale foras (Middle Pennsylvanian, Oklahoma). Palaios, 6,390-409. -• OLMSTEAD, R. G. 1987. Opportunistic 'evolution: abiotic environmental stress and the fossil record of plants. Review of Palaeobotany and Palynology, 50, 151-178. -& PEPPERS, R. A. 1985. The influence of climate and depositional environment on the distribution and evolution of Pennsylvanian coalswamp plants. In: TIFFNEY, B. H. (ed.) Geological factors and the evolution of plants. Yale University Press, Yale, 223-256. -& WILLARD, D. a . 1986. Morphology and paleoecology of Pennsylvanian-Age coal-swamp plants. In: GASTALDO, R. A. (ed.) Land Plants notes for a short course. University of Tennessee Department of Geological Sciences Studies in Geology Number 15 (edited by T. W. Broadhead). University of Tennessee, Knoxville, 97-114. RISCHBIETER, M. O., EGGERT, D. L. & GASTALDO, R. A. 1984. Stem and leaf cuticle of Karinopteris: source of cuticles from the Indiana 'paper' coal. American Journal of Botany, 71, 626-637. DURAND, B. & OUDIN, J. L. 1980. Exemple de migration des hydrocarbures darts une series deltaique; le delta de la Mahakam, Kalimantan, Indonesie. Proceedings of the lOth Worm Petroleum Congress, Bucarest, Volume 2, Heiden, London, 3-11. --, PARRATrE, M. & BERTRAND, M. 1983. Le potential en huile des charbons: une approche g6ochimique. Revue de l'Institute Francais de P~trole, 38,709-721. EGGERT, O. A. 1961. The ontogeny of Carboniferous arborescent Lycopsida. Palaeontographica Abteilung, B108, 43-92. 1972. Petrified Stigmaria of sigillarian origin from North America. Review of Palaeobotany and Palynology, 14, 85-99. EGGERT, D. L., CHou, C.-L., MAPLES, C. G., PEPPERS, -
-
R. A. PHILLIPS,T. L. & REXROAD, C. B. 1983.
Origin and economic geology of the Springfield
66
M. E. COLLINSON ET AL.
Coal Member in the Illinois Basin (Field trip 9). In: SHAVER, R. H. & SUNDERMAN,J. A. (eds) Field trips in Midwestern Geology. Indiana University Bloomington, 2, 121-146. & PHILLIPS, T. L. 1982. Environments of deposition - coal balls, cuticular shale and grayshale floras in Fountain and Parke counties, Indiana. State of Indiana Department of Natural Resources Special Report, 30. Bloomington, Indiana, 1-43. FLEET, A. J., KELTS, K. & TALBOT, M. R. (eds) 1988. Lacustrine Petroleum Source Rocks. Geological Society, London, Special Publication, 40. FLEMING, R. F. 1989. Fossil Scenedesmus (Chlorococcales) from the Raton Formation, Colorado and New Mexico, USA. Review of Palaeobotany and Palynology, 59, 1-6. FRANKENBERG,J. M. & EGGERT, D. A. 1969. Petrified Stigmaria from North America, I. Stigmaria ficoides, the underground portions of Lepidodendraceae. Palaeonotgraphica Abteilung, B128, 1-47. FU~USHIMA, K. & ISHIWATARI,R. 1988. Geochemical significance of lipids and lipid-derived substructures interlaced in kerogen. Organic Geochemistry, 12,509-518. GARRIGUES, P., SURY, R. DE, ANGELIN, M. L., BELLOCO, J., OUDIN, J. L. & EWALD, M. 1988. Relation of the methylated aromatic hydrocarbon distribution pattern to the maturity of organic matter in ancient sediments from the Mahakam delta. Geochimica et Cosmochimica Acta, 52,375-384. GASTALDO, R. A. 1987. Confirmation of Carboniferous clastic swamp communities. Nature, 326, 869-871. - 1990. Phytotaphocoenoses in late Quaternary temperate and tropical coastal deltaic regimes. In: KNOBLOCH, E. & KVACEK, Z. (eds) Pro-
ceedings of the Symposium Paleofloristic and paleoclimatic changes in the Cretaceous and Tertiary. Geological Survey, Prague, 285-289. , GIBSON, M. A. & GRAY, T. D. 1989. An Appalachian-sourced deltaic sequence, northeastern Alabama, USA: biofacies-lithofacies relationships and interpreted community patterns. International Journal of Coal Geology, 12,225-257. VAN GEEL, B., BOHNCKE, S. J. P. & DEE, H. 1981. A palaeoecological study of an upper Late Glacial and Holocene sequence from 'De Borchert' The Netherlands. Review of Palaeobotany and Palaeoecology, 31,367-448. GIVEN, P. H., SPACKMAN,W., DAVIS, A. & JENKINS, R. G. 1980. Some proved and unproved effects of coal geochemistry on liquefaction behaviour with emphasis on US coals. In; WHITEHURST, D. D. (ed.) Coal liquefaction fundamentals. American Chemical Society Symposium Series, 139, American Chemical Society, Washington DC, 3-34. GOTH, K. 1990. Der Messeler Olschiefer - ein Algenlaminit. Courier Forschungsinstitut Senekenberg, 131, 1-143.
, DE, LEEUW, J. W., POTTMANN,W. • TEGELAAR, E. W. 1988. Origin of Messel Oil Shale Kerogen. Nature, 336, 759-761. GOULD, R. 1981. The coal-forming flora of the Walloon Coal Measures. Coal Geology, Journal
of the Coal geology Group of the Geological Society of Australia, 1, 83-105. --
& SHIBAOKA, M. 1980. Some aspects of the formation and petrographic features of coal members in Australia, with special reference to the Tasman orogenic zone. Coal Geology,
Journal of the Coal geology Group of the Geological Society of Australia, 2, 1-29. GRAHAM, R. 1935. An anatomical study of the leaves of the Carboniferous arborescent lycopods. Annals of Botany, 49,587-608. GUERRA-SOMMER,M., MARQUES-TOIGO,M. & CORRI~A DA SILVA, Z. 1991. Original biomass and coal deposition in southern Brazil (Lower Permian, Paran~i basin). Bulletin de la Socidtd Gdologique de France, 162, 227-237. HATCHER, P. G. 1988. Dipolar-Dephasing 13C NMR studies of decomposed wood and coalified xylem tissue: evidence for chemical and structural changes associated with defunctionalization of lignin structural units during coalification. Energy & Fuels, 2, 40-58. - 1989. Chemical structural models for coalified wood (vitrinite) in low rank coal. Organic Geochemistry, 16, 959-968. --, FAULON,J. -L., WENZEL, K. A. & CODY, G. D. 1992. A structural model for lignin-derived vitrinite from high-volatile bituminous coal (coalified wood). Energy & Fuels, 6,813-820. - - , LYONS,P. C., THOMPSON,C. L., BROWN,F, W. & MACIEL, G. E. 1982. Organic Matter in a Coal ball: Peat or Coal. Science, 217,831-833. HEMSLEY, A. R., BARRIE, P. J., CHALONER, W. G. & SCOTT, A. C. in press. The composition of sporopollenin and its use in living and fossil plant systematics. Grana. - - & Sco'rr, A. C. 1991. Ultrastructure and relationships of Upper Carboniferous spores from Thorpe Brickworks, West Yorkshire, UK. Review of Palaeobotany and Palynology, 69, 337-351. HETI~NYI, M. & SAJGO, CS. 1990. Hydrocarbon generation potential of some Hungarian low-rank coals. Organic Geochemistry, 16, 907-916. HILL, C. R. & DILCHER, D. L. 1990. Scanning electron microscopy of the internal ultrastructure of plant cuticle. In: CLAU~HER, D. (ed.) Scanning
Electron Microscopy in Taxonomy and Functional Morphology. Sytematics Association Special volume No. 41, Clarendon Press, Oxford, 95-124. HOLLOWAY, P. J. 1982. Structure and histochemistry of plant cuticular membranes: an overview. In: CUTLER, D. F., ALVIN, K. L. & PRICE, C. E. (eds) The Plant Cuticle. The Linnean Society Symposium Series No 10, Academic Press, London, 1-32. HORSFIELD, B. 1990. Evaluating kerogen type
O I L - G E N E R A T I N G P O T E N T I A L OF PLANTS: A R E V I E W according to source quality, compositional heterogeneity and thermal lability. Review of Palaeobotany and Palynology, 65,357-365. HOWER, J., SCOTT, A. C., Hua-rON, A. and PAREKH, B. K. in press. Coal availability, mining and preparation. In: BIslo, A. & BOOTS, S. G. (eds)
Encyclopedia of Energy Technology and the Environment. J. Wiley & Sons Ltd, New York. HUTrON, A. C. 1986. Classification of Australian oil shales. Energy Exploration and Exploitation, 4, 81-93. -1987. Petrographic classification of oil shales.
International Journal of Coal Geology, 8, 203-231. 1988. The lacustrine Condor oil shale sequence. In: FLEET, A. J., KELTS, K. & TALBOT, M. R. (eds) Lacustrine Petroleum Source Rocks Geological Society, London, Special Publication, 40,329-340. ICCP 1963. International Handbook of Coal Petrology (2nd edn). Centre National de la Recherche Scientifique, Paris, France. JOHNSON, E. A. 1990. Geology of the Fushun coalfield, Liaoning Province, People's Republic of China. International Journal of Coal Geology, 14, 217-236. JoY, K. W., WILLIS, A. J. & LACEY, W. S. 1956. A rapid cellulose peel technique in palaeobotany. Annals of Botany, 20, 635-637. KATZ, B. J. 1988. Clastic and carbonate lacustrine systems; an organic geochemical comparison (Green River Formation and East African lake sediments). In: FLEET, A. J., KELTS, K. & TALBOT, M. R. (eds) Lacustrine Petroleum Source Rocks. Geological Society, London, Special Publication, 40, 81-90. LANGENHEIM, J. H. 1990. Plant resins. American Scientist, 78, 16-24. LAPO, A. V. t~¢ DROZDOVA, I. N. 1989. Phyterals of humic coals in the USSR. International Journal of Coal Geology, 12, 477-510. LARTER, S. R. 1990. Molecular characterisation of k e r o g e n - Applications to primary and secondary migration studies and to maturation modelling.
- -
Review of Palaeobotany and Palynology,
65,
379-391. LAVEINE, J-P. 1986. The size of the frond in the genus Alethopteris Sternberg (Pteridospermopsida, Carboniferous). Geobios, 19, 49-56. LOGAN, K. J. & THOMAS,B. A. 1987. The distribution of lignin derivatives in fossil plants. New Phytologist, 105, 157-173. LYONS, P. C., FINKELMAN,R. B., THOMPSON, C. L., BROWN, F. W. & HATCHER, P. G. 1982. Properties, origin and nomenclature of rodlets of the inertinite maceral group in coals of the central Appalachian Basin, U.S.A. International Journal of Coal Geology, 1,313-346. , HATCHER, P. G. & BROWN, F. W. 1986. Secretinite: a proposed new maceral of the inertinite maceral group. Fuel, 65, 1094-1098. --, THOMPSON,C. L. & MILLAY, M. A. 1985. Coalification of organic matter in a coal ball from the Calhoun coal bed (Upper Pennsyl-
67
vanian), Illinois Basin, United States of America. Compte Rendu du Dixidme Congrds
International de stratigraphie et de Geologie du Carbonifdre, Madrid, 12-17 September 1983, Volume 1,155-159. --, MINKIN, J. A., THOMPSON,C. L. LARSON, R. R. BROWN, Z. A. & PHEIFER, R. N. 1984. Resin rodlets in shale and coal (Lower Cretaceous), Baltimore Canyon Trough. International Journal of Coal Geology, 3,257-278. MCKIRDY, D. M., Cox, R. E., VOLKMAN, J. K. & HOWELL, V. J. 1986. Botryococcane in a new class of Australian non-marine crude oils. Nature, 320, 57-59. MELLO, M. R., TELNAES, N., GAGLIANONE, P. C., CHICARELLI,M. I., BRASSELL,S. C. • MAXWELL, J. R. 1988. Organic geochemical characterisation of depositional palaeoenvironments of source rocks and oils in Brazilian marginal basins. Organic Geochemistry, 13, 31-45. MEYEN, S. V. 1987. Fundamentals of Palaeobotany. Chapman & Hall, London. MIAO, F., QIAN, L. & ZHANG, X. 1989. Peat-forming materials and evolution of swamp sequences case analysis of a Jurassic inland coal basin in China. International Journal of Coal Geology, 12,733-765. MICKLE, J. E. & ROTHWELL, G. W. 1982. Permineralized Alethopteris from the upper Pennsylvanian of Ohio and Illinois. Journal of
Paleontology,
56,392-402.
MOLDOWAN,J. M., SEIFERT,W. K. & GALLEGOS,E. J. 1985. Relationship between petroleum composition and depositional environment of petroleum source rocks. American Association of Petroleum Geologists Bulletin, 69, 1255-1268. MORLEY, R. J. 1981. Development and vegetation dynamics of a lowland ombrogenous peat swamp in Kalimantan Tengah, Indonesia. Journal of Biogeography, 8, 383-404. MUKHOPADHYAY,P. K., GORMLEY,J. R. t~ ZUMBERGE, J. E. 1989. Generation of hydrocarbons from the Tertiary coals of Texas - Coal as a potential source rock for liquid hydrocarbons in a deltaic basin? Organic Geochemistry, 14, 351-352. --, HATCHER,P. G. & CALDER,J. H. 1991. Hydrocarbon generation from deltaic and intermontane fluviodeltaic coal and coaly shale from the Tertiary of Texas and Carboniferous of Nova Scotia. Organic Geochemistry, 17, 765-783. MURCHISON, D. G. 1987. Recent advances in organic petrology and organic geochemistry: an overview with some reference to 'oil from coal'. In: SCOTT, A. C. (ed.) Coal and Coal-bearing Strata: Recent Advances. Geological Society, London, Special Publication, 32,257-302. NEAVEL, R. C. & GUENNEL, G. K. 1960. Indiana paper coal: composition and deposition. Journal of Sedimentary Petrology, 30,241-248. NEWTON, E. T. 1875. On tasmanite and Australian white coal. Geological Magazine, 2,337-342. NIP, M., DE LEEUW, J. W. & CRELLING, J. C. 1992. Chemical structure of bituminous coal and its constituing maceral fractions as revealed by flash pyrolysis. Energy & Fuels, 6, 125-136.
68
M. E. COLLINSON ET AL.
, - - , SCHENCK,P. A., WINDIG, W., MEUZELAAR, H. L. C. & CRELLING, J. C. 1989. A flash pyrolysis and petrographic study of cutinite from the Indiana paper coal. Geochimica et Cosmochimica Acta, 53,671--683. , TEGELAAR, E. W., BRINKHUIS, H., DE LEEUW, J. W., SCHENCK,P. A. • HOLLOWAY,P. J. 1986b. Analysis of modern and fossil plant cuticles by Curie point Py-GC and Curie point Py-GC-MS: Recognition of a new, highly aliphatic and resistant biopolymer. Organic Geochemistry, 10, 769-778. , , DE LEEUW, J. W., SCHENCK, P. A. & HOLLOWAY, P. J. 1986a. A new non-saponifiable highly aliphatic and resistant bioploymer in plant cuticles. Naturwissenschafien, 73,579-585. OESTRY-STIDD, L. L. 1979. Anatomically preserved Neuropteris rarinervis from American coal balls. Journal of Paleontology, 53, 37-43. PHILLIPS, T. L. 1980. Stratigraphic and geographic occurrences of permineralized coal-swamp plants - Upper Carboniferous of North America and Europe. In: DILCHER, D. L. & TAYLOR,T. N. (eds) Biostratigraphy of Fossil Plants. Dowden, Hutchinson & Ross, Stroudsburg, Pennsylvania, 25-92. 1981. Stratigraphic occurrences and vegetational patterns of Pennsylvanian pteridosperms in Euramerican coal swamps. Review of Palaeobotany and Palynology, 32, 5-26. ,~z DIMICHELE, W. A. 1981. Palaeoecology of Middle Pennsylvanian age coal swamps in southern Illinois: Herrin Coal Member at Sahara Mine No. 6. In: NIKLAS, K. J. (ed.) Paleobotany, Paleoecology and Evolution. Prager, New York, 231-284. & ~ 1992. Comparative ecology and lifehistory biology of arborescent lycopsids in late Carboniferous swamps of Euramerica. Annals of the Missouri Botanical Gardens, 79, 560-588. --, PEPPERS, R. A. & DIMICHELE, W. A. 1985. Stratigraphic and interregional changes in Pennsylvanian coal-swamp vegetation: Environmental inferences. International Journal of Coal Geology, 5, 43-109. PmLP, R. P. & GILBERT, T. D. 1986. Biomarker distributions in Australian oils predominantly derived from terrigenous source material. Organic Geochemistry, 10, 73--84. & RUSSELL, N. J. 1982. Characterization 'by pyrolysis-gas chromatography-mass spectrometry of the insoluble organic residues derived from the hydrogenation of Tasmanites sp oil shale. Fuel 61,221-226. POWELL, T. G. 1986. Petroleum geochemistry and depositional setting of lacustrine source rocks. Marine and Petroleum Geology, 3, 200-219. RAMANUJAM,C. G. K., ROTHWELL,G. W. & STEWART, W. N. 1974. Probable attachment of the Dolerotheca campanulum to a Myeloxylon-Alethopteris type frond. American Journal of Botany, 61, 1057-1066. REED, F. D. 1941. Coal flora studies: Lepidodendrales. Botanical Gazette, 102,663--683. -
-
-
-
-
-
REIHMAN, M. A. & SCHABILION,J. T. 1976a. Cuticles of two species of Alethopteris. American Journal of Botany, 63, 1039-1046. -& -1976b. Two species of Alethopteris from Iowa coal balls. Proceedings of the Iowa Academy of Sciences, 83, 10-19. 1978. Petrified neuropterid foliage from a Middle Pennsylvanian coal ball. American &
-
-
Journal of Botany,
6 5 , 8 3 4 - 8 4 4 .
REX, G. M. & SCOTT, A. C. 1987. The sedimentology, palaeoecology and preservation of the Lower Carboniferous plant deposits at Pettycur, Fife, Scotland. Geological Magazine, 124, 43-456. RISK, M. J. & RHODES, E. G. 1985. From mangroves to petroleum precursors: an example from Tropical Northeast Australia. American Association of Petroleum Geologists Bulletin, 69, 1230-1240. ROWE, N. P. 1988. Two species of the lycophyte genus Eskdalia Kidston from the Drybrook sandstone (Visean) of Great Britain. Palaeontographica Abteilung, B208, 81-103. RULLKOTTER, J., LITTKE, R., HAGEDORN-GOTZI, I. & JANOWSKI, B. 1988. Vorlaufige Ergebnisse der organisch-geochemischen und organischpetrographischen Untersuchungen an Kernproben des Messeler Olschiefers. Courier Forschungsinstitut Senckenberg, 107, 37-51. SCHAAL, S. t~ ZIEGLER, W. (eds) 1992. M e s s e l -
an insight into the history of life and of the earth. Clarendon Press, Oxford. SCHENCK, P. A., DE LEEUW, J. W., VAN GRAAS, G., HAVERKAMP,J. & BOUMAN,M. 1981. Analysis of recent spores and pollen and of thermally altered sporopollenin by flash pyrolysis-mass spectrometry and flash pyrolysis-gas chromatographymass spectrometry. In: BROOKS,J. (ed.) Organic
Maturation Studies and Fossil Fuel Exploration. Academic Press, London, 225-237. SCHOPF, J. M. 1939. Medullosa distellica, a new species of the Anglica group of Medullosa. American Journal of Botany, 26, 196-207. -1956. A definition of coal. Economic Geology, 51,521-527. ScoTr, A. C. 1978. Sedimentological and ecological control of Westphalian B plant assemblages from West Yorkshire. Proceedings of the Yorkshire Geological Society, 41,461-508. -1987 (ed.) Coal and Coal-bearing Strata: Recent Advances. Geological Society London, Special Publication, 32. ~, GALTIER,J. & CLAYTON,G. 1984. Distribution of anatomically preserved floras in the Lower Carboniferous in Western Europe. Transactions
of the Royal Society of Edinburgh: Earth Sciences, 75, 311-340. -
-
& HEMSLEY, A. R. 1993. The spores of the Dinantian lycopsid cone Flemingites scotti from Pettycur, Fife, Scotland. Special Papers in Palaeontology, 49, 31-41. & REX, G. 1985. The formation and significance of Carboniferous coal balls. Philosophical transactions of the Royal Society of London, B311, 123-137.
O I L - G E N E R A T I N G P O T E N T I A L OF PLANTS: A R E V I E W
SELDEN, P. A. & EDWARDS,D. 1989. Colonisation of the land. In: ALLEN, K. & BRIGGS, D. (eds) Evolution and the Fossil Record. Belhaven Press, London, 122-152. SHANMU6AM, G. 1985. Significance of coniferous rain forests and related organic matter in generating commercial quantities of oil, Gippsland Basin, Australia. American Association of Petroleum Geologists Bulletin, 69, 1241-1254. SHENG, G., SIMONEIT,B. R. T., LEIF, R. N., CHEN, X. & Fu, J. 1992. Tetracyclic terpanes enriched in Devonian cuticle humic coals. Fuel, 71,523-532. SIGLEO, A. G. 1978. Degraded lignin compounds identified in silicified wood 200 million years old. Science, 200, 1054-1056. SINNINGHEDAMSTI~,J., DE LAS HERAS, F. X. C., VAN BER6EN, P. F. & DE LEEUW, J. W. 1993, Characterization of Tertiary Catalan lacustrine oil shales: Discovery of extremely organic sulphurrich Type I kerogens. Geochimica et Cosmochimica Acta, 57,389-415. SNOWDON,L. R. & POWELL, T. G. 1982. Immature oil and condensate - modification of hydrocarbon generation model for terrestrial organic matter. American Association of Petroleum Geologists Bulletin, 66,775-788. SPENCER, K. C. & DIMICHELE, W. A. 1987. Paleochemotaxonomic analysis of cuticle from the Pennslyvanian-age gymnosperm Karinopteris. American Journal of Botany, 74,688. STACH, E., TAYLOR, G. H., MACKOWS~:Y, M.-TH., CHANDRA, D., TEICHM~LLER, M. & TEICHMI]LLER, R. 1982. (Translation by Murchison, D. G., Taylor, G. H. & Zierke, F.) Stach's Textbook of Coal Petrology. Gebriider Borntraeger, Berlin. STEWART, W. N. t~ DELEVORYAS, T. 1956. The Medullosan pteridosperms. The Botanical Review, 22, 45-80. STOUT, S. A. & SPACKMAN,W. 1989. Peatification and coalification of wood as deduced by quantitative microscopic methods. Organic Geochemistry, 14, 285-298. --, BooN, J. J., KISTEMAKER, P. G. & BENSLEY, D. F. 1989. Correlations between the microscopic and chemical changes in wood during peatification and early coalification: a canonical variant study. International Journal of Coal Geology, 13, 41-64. TALBOT, M. R. 1988. The origins of lacustrine source rocks: evidence from the lakes of tropical Africa. In: FLEET, A. J., KELT, K. & TALBOT,M. R. (eds) Lacustrine Petroleum Source Rocks, Geological Society, London, Special Publication, 40, 29-44. TAPPAN, H. 1980. The Paleobiology of Plant Protists. W. H. Freeman & Co., San Francisco. TAYLOR, E. L., TAYLOR, T. N. & COLLINSON,J. W. 1989. Depositional setting and paleobotany of Permian and Triassic permineralized peat from the central Transantarctic Mountains, Antarctica. International Journal of Coal Geology, 12,657-679. TAYLOR, W. A., TAYLOR, T. N. & ARCHANGELSKY,S. 1989. Comparative ultrastructure of fossil and ,
69
living gymnosperm cuticles. Review of Palaeobotany and Palynology, 59, 145-151. TEGELAAR, E. W., DE LEEUW, J. W., DERENNE, S. & LARGEAU, C. 1989a. A reappraisal of kerogen formation. Geochimica et Cosmochimica Acta, 53, 3103-3106. --, HOLLMAN, G., VAN DER VEGT, P., DE LEEUW, J. W. & HOLLOWAY,P. J. 1990. Chemical characterisation of the outer bark tissue of some angiosperm species: recognition of an insoluble, nonhydrolyzable highly aliphatic biopolymer (suberan). In: TEGELAAR, E. W. (ed.) Resistant
biomacromolecules in morphologically characterized constituents of kerogen A key to the relationship between biomass and fossil fuels. PhD thesis, University of Delft, Netherlands, 115-134. - - ' , KERP, H., VISSCHER, H., SCHENCK, P. A. & DE LEEUW, J. W. 1991. Bias of the paleobotanical record as a consequence of variations in the chemical composition of higher vascular plant cuticles. Paleobiology, 17, 133-144. , MATTHEZING, R. M., JANSEN, J. B. H., HORSFIELD, B. & DE LEEUW, J. W. 1989b. Possible origin of n-alkanes in high-wax crude oils. Nature, 342,529-531. THOMAS, B. A. 1968. A revision of the Carboniferous lycopod genus Eskdalia Kidston. Palaeontology, 11,439-444. & MASARATI,D. L. 1982. Cuticle and epidermal studies in fossil and living lycophytes. Linnean Society Symposium Series, 10,363-378. & MEYEN, S. V. 1984. A reappraisal of the Lower Carboniferous lepidophyte Eskdalia Kidston. Palaeontology, 27,707-718. THOMAS, B. M. 1982. Land-plant source rocks for oil and their significance in Australian Basins.
-
-
Journal of the Australian Petroleum Exploration Association, 22, 164-178. THOMPSON, S., COOPER, B. S., MORLEY, R. J. BARNARD, P. C. 1985. Oil-generating coals. In: THOMAS, B. M. et al. (eds) Petroleum Geochemistry in Exploration of the Norwegian Shelf. Graham & Trotman, London, 59-73. WALTON, J. 1926. A note on the structure of the plant cuticles in the paper-coal from Toula in Central Russia. Manchester Memoirs, 70, 119-123. WILSON, J. A. R. 1931. Some new facts about the structure of the cuticles in the Russian paper-coal and their bearing on the systematic position of some fossil Lycopodiales. Proceedings of the Royal Society of Edinburgh, 51,104-114. WILSON, M. A., BATrS, B. D. & HATCHER,P. G. 1988. Molecular composition and mobility of torbanite precursors: Implications for the structure of coal. Energy and Fuels, 2,668-672. WINSTON, R. B. 1986. Characteristic features and compaction of plant tissues traced from permineralized peat to coal in Pennsylvanian coals (Desmoinesian) from the illinois Basin. International Journal of Coal Geology, 6, 21-41. WNUK, C. & PFEFFERKORN,H. W. 1984. The life habits and paleoecology of Middle Pennsylvanian medullosan pteridosperms based on an in situ
70
M. E. COLLINSON ET AL.
assemblage from the Bernice Basin (Sullivan County, Pennsylvania, USA). Review of Palaeobomny and Palynology, 41,329-351. YANG, WANLI, LI YONGKANG & GAO RUIQI. 1985. Formation and evolution of nonmarine petroleum in Songliao Basin, China. American Association of Petroleum Geologists Bulletin, 69, 1112-1122. YEN, T. F. • CHILINGARIAN, G. V. 1976 (eds). Oil Shale. Elsevier, Amsterdam. ZELIBOR, J. L., ROMANKIW, L., HATCHER, P. G. & COLWELL, R. R. 1988. Comparative analysis of
the chemical composition of mixed and pure cultures of the green algae and their decomposed residues by 13C Nuclear Magnetic Resonance spectroscopy. Applied Environmental Biology, 54, 1051-1060. ZODROW, E. L. & CLEAL, C. J. 1988. The structure of the Carboniferous pteridosperm frond Neuropteris ovata Hoffmann. Palaeontographica Abteilung, B208, 105-124. -& -1993. The epidermal structure of the Carboniferous gymnosperm frond Reticulopteris. Palaeontology, 36, 65-79.
Geochemical characteristics of oils derived predominantly from terrigenous source materials R. P. P H I L P School o f Geology and Geophysics, University o f Oklahoma, Norman, OK. 73019, USA Abstract: It is well established that there are many major accumulations of crude oils that are derived predominantly from higher plant source materials. All of these oils have a number of geochemical characteristics in common. These features include n-alkane distributions, presence of a variety of sesqui-, di-, tri-, tetra-, and pentacyclic terpanes and isoprenoids. Variations in the distribution of these compounds reflect variations in the nature of the source materials and the original depositional environment. The purpose of this paper is to review these characteristic features so that they can be used to determine whether or not new oil samples are derived predominantly from higher plant source materials. The geochemical approach has been used successfully in the past to characterize samples derived from hypersaline and carbonate environments.
The association of specific geochemical characteristics of oils with particular types of source materials, or depositional environments, can greatly assist any exploration effort. Perusal of the geochemical literature reveals that such an approach has been successful in a number of situations, particularly with samples obtained from hypersaline environments (ten Haven et al. 1988) or carbonate source rocks (Palacas et al. 1984). The purpose of this paper is to review the geochemical characteristics of oils derived from source rocks known to contain a predominance of terrigenous source material, particularly higher plant material. In an earlier paper (Philp & Gilbert 1986) a number of geochemical characteristics were described for oils known to be derived from higher plant materials and the summary from this paper is reproduced as Table 1. Since the publication of that paper, new data have become available and some of the original interpretations found to be invalid. In the present paper it is proposed to note changes that have occurred to some of these earlier interpretations and, using a wider sample base, devise a more comprehensive set of characteristics that can be associated with oils from terrigenous source materials. It is also interesting to note that many geochemical parameters originally thought to be source specific for these oils are, in all probability, more likely related to the nature of the depositional environment. It has been established that oils produced in certain areas of the world are derived from source materials that are non-marine and contain significant amounts of higher plant material. Many of these oil accumulations occur in the Southern Hemisphere particularly in countries such as Australia, New Zealand,
Indonesia, and Taiwan. Oils associated with higher plant materials are also found in countries such as Nigeria, Greenland, Egypt and Beaufort-Mackenzie Basin, Canada. The results described in this paper will use examples from many of these areas to demonstrate that there are a number of geochemical characteristics common to this type of oil which permit them to be readily recognized. Furthermore, with the evolution of various plant types, particularly gymnosperms and angiosperms, individual compounds from different plant types can be used to estimate approximate ages of many oils derived from terrigenous source materials (Alexander et al. 1988). As is the case with so many geochemical studies, it does not necessarily follow that all oils from such sources will possess all of the geochemical characteristics described herein and hence, as always, some caution needs to be exercised in the application of this approach.
Bulk characteristics Since Hedburg (1968) published his classic paper on high-wax crude oils, almost all papers describing the characterization of oils from terrigenous source materials make reference to the fact that this type of oil will be very waxy. It is of interest to note that the bulk of the waxy crude oils described by Hedburg (1968) in his study are from the Cretaceous or early Tertiary. This parallels the evolution of the higher plants, such as conifers, which may be expected to make a significant waxy contribution to the source materials. The waxy hydrocarbons commonly found in the crude oils from terrigenous source materials are assumed to be derived from the
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coal and Coal-bearing Strata as Oil-prone Source Rocks? Geological Society Special Publication No. 77, pp. 71-91.
71
x
x
?
?
x
x
?
?
?
?
---
x
--
x
x
10: I
3:1 2: I 8:1 10:1
--
--~ --
x
?t
1:1 3: ! 2.5:1
x
Di-
x
x*
Sesqui-
5:1
--
--
--
--
--
--
--
--
x
--
--
Tricyclic
?
x
x
x
x
x
x
x
x
x
x
Sester(C24)
x
x
x
x
x
x
x
x
x
--
x
Hopanes
--
x
?
?
x
x
x
x
--
--
x
Czs-Bisnorhopane
*x means a particular class of biomarkers was analysed for and found to be present. t ? m e a n s this c o m p o u n d class was not analysed for and thus it is not k n o w n whether it is present or absent. , - - m e a n s c o m p o u n d class was analysed for and found to be absent. ~ o m p o u n d Y is an unidentified C27 pentacyclic terpane. IICompound X is an unidentified C30 pentacyclic terpane.
Tertiary
Tertiary
Late Cretaceous/ Tertiary Permian I) Permian 2) Jurassic and Triassic I) Permian 2) Jurassic 3) Triassic Tertiary
Age of reservoir
Terpanes Steranes
Terpanes
x
x
?
--
x
x
x
x
x
x
x
x
x
?
--
x
x
x
x
x
x
x
Compound Y§ Xll
x
x
x
x
--
--
--
--
--
--
tr
18a(H)oleananes
?
C~
?
C29
C~
C~
C29
C~ C27 and C~
C29
C~
Predominant steranes
x
?
x
?
?
--
--
--
--
--
x
Head/tail long chain isoprenoids
Resinite/higher plant
Higher plant
Higher plant
Higher plant Marine/higher plant Higher plant Higher plant/marine Higher p l a n t Higher plant Higher plant
Higher plant
Postulated predominant source material
1. A summary of biomarker characteristics for Australian and other oils derived predominantly from terrigenous source material (after Philp & Gilbert 1986)
Cooper/ Eromanga Basin M a h a k a m Delta, Indonesia Niger Delta, Nigeria Taranaki, New Zealand Beaufort/ MacKenzie, Canada
Gippsland Basin Sydney Basin Surat/Bowen Basin
Location of reservoir
Table
t~
OILS DERIVED FROM TERRIGENOUS SOURCE MATERIALS cuticular waxes of higher plants. However, it needs to be pointed out that there are also waxes associated with oils derived from marine source materials (Del Rio & Philp 1992a, b, c) and waxy oils associated with lacustrine deposits which may or may not be associated with high inputs of higher plant material. McKirdy et al. (1986) have shown that in certain stages of its life cycle, the green alga Botryoccocus braunii can produce significant quantities of waxy n-alkanes. In summary, all oils from terrigenous materials may be waxy but not all waxy oils are from terrigenous source materials. Another common bulk property of oils from terrigenous source materials reflects the composition of the oils in terms of various proportions of saturate/aromatic/polar/ asphaltene fractions. For the most part, their waxy nature means that the major fraction of these oils is the saturate fraction. In many cases this may correspond to 80 or 90% of the total oil. As a consequence the majority of the discussion in this paper will centre on the saturate fraction of the oils. The abundance of saturate fractions in these oils means that the relative concentrations of the asphaltene and polar fractions are low and for many of these oils it is virtually impossible to isolate any significant asphaltene fraction. This is a feature that clearly distinguishes these oils from the marine-derived oils which, for the most part, have abundant asphaltene fractions. The aromatic fractions are typically present in 5-10% relative abundance for most oils derived from terrigenous source materials. Finally, the organic sulphur content of oils from terrigenous source materials is generally very low and in many cases virtually zero. Both the bulk analyses for total sulphur and analyses of the aromatic fractions for individual sulphur compounds rarely, if ever, show any thiocompounds to be present. Exceptions to this observation that have been reported recently include the identification of sulphur-containing oleananes by Adam et al. (1991) and the presence of element sulphur in samples from Baffin Bay which contained some terrestrial organic matter (ten Haven et al. 1992). However, to date, this is an exception rather than a characteristic feature and probably more a reflection of the depositional environment than source material. The original environments in which the higher plant materials were deposited probably would have been fairly oxic, for example peat swamps or deltaic environments, compared to the more highly reducing marine environments. In the more oxic environment, the level of sulphate reduction will be minimal
73
and any hydrogen sulphide produced will be readily scavenged by the abundant metal ions in the environment. Following this brief summary of bulk characteristics, a review of individual classes of compounds that can be associated with oils from terrigenous sources will be discussed. As mentioned above the bulk of the organic material with these oils occurs in the saturate fraction and hence the following sections will be concerned mainly with the various classes of saturated biomarkers.
n-Alkanes The most widely cited characteristic of oils derived from terrigenous source materials is their waxy nature (Hedburg 1968). This characteristic results from the fact that higher plants have a high wax content and the waxes are fairly resistant to degradation and hence well preserved. Furthermore, the discovery of material called cutan in various plants (Telegaar 1990) introduces another possible source of n-alkanes upon thermal breakdown of this material (see Collinson et al., this volume). The n-alkane distributions of many waxy crude oils maximize in the range C20-C40 (Fig. la). It is important to point out here, however, that the distribution of n-alkanes maximizing in this region of the chromatogram is somewhat misleading and is as much a reflection of analytical limitations and isolation procedures as it is the source material. The upper temperature limit of most GC columns used for analysis of these samples is in the range 300-350°C. This is sufficient to permit alkanes with carbon numbers up to about C40 to elute from the column. However, in the past couple of years chromatographic columns have become available with upper temperature limits of 480°C. With the advent of these columns the possibility exists to analyse oils and waxes where the carbon numbers of individual compounds extends to at least C70 and probably higher (Fig. lb). As these columns are used more frequently to analyse samples the average value for carbon number maxima of the n-alkanes observed in crude oils from terrigenous sources will shift to higher carbon numbers. In a recent paper it was observed that waxes associated with oils derived from marine source rocks can also have n-alkane distributions which extend to higher carbon numbers than previously observed (Fig. 2). Hence, although oils from terrigenous sources are waxy, not all waxy samples are necessarily derived from terrigenous source materials (Del Rio & Philp 1992a, b, c).
74
R.P.
PHILP
C20 POOLOWANNA SATURATE FRACTION
C30
i
(a)
o
BO
4o
M%nutes
OZOCERITE SATURATE FRACTION
Cl'• °
1 III c
;5o
I
(b)
2
i 24
M~nutO~
c6°
l
! 40
OILS DERIVED FROM TERRIGENOUS SOURCE MATERIALS
75
011H
m
Wax H
. . . . . .
time
-
~
>
Fig. 2. n-Alkane distributions for an oil derived from marine source rocks and its associated wax which precipitated out during production.
Isoprenoids The ratio of the isoprenoids, pristane and phytane, has long been associated with the nature of depositional environments following the pioneering work of the Australian geochemists Brooks & Smith (1969) and Powell & McKirdy (1973). Their concepts were later reviewed and modified by Didyk et al. (1978) and ten Haven et al. (1987) who showed that in certain environments this ratio may also be affected by alternative sources for both pristane and phytane. In the earlier work it was assumed that both pristane and phytane were derived from the phytol side-chain of chlorophyll. More recently it has been shown that pristane can be
derived from a-tocopherol and phytane from the bis-phytanyl ethers which occur in archeabacteria. The recent development of the combined gas chromatographic-isotope ratio mass spectrometric technique has given u s the opportunity to determine whether or not in a particular sample, the pristane and the phytane are indeed derived from the same source material. For example, Freeman et aL (1990) have shown that in the case of the Messel oilshale these isoprenoids do indeed have different sources. However, for two of the Gippsland Basin oils, namely the Turrum and Barracouta oils, the isotopic values for the pristane and phytane were virtually identical as shown in
Fig. 1. n-Alkane distributions for (a) an oil derived from terrigenous source materials from the Cooper Basin, Australia; (b) a sample of Ozocerite extending to at least C70 and probably to C10oor higher in some samples when determined by High Temperature Gas Chromatography.
76
R P. PHILP 2. 6Z3C isotopic compositions (relative to PDB) of pristane and phytane for two Gippsland crude oils
Table
Sample
Pristane
Phytane
Barracouta Turrum
28.1 26.6
28.6 26.9
Table 2. Differences in the values between the two oils probably reflect differences in the maturity level of the samples. Following the early work, and in light of the more recent modifications, it would appear that rather than trying to assign a specific value of the Pr/Ph ratio to indicate the nature of the depositional environment it is better to say that where this ratio is greater than 4, or possibly even 5, then the original depositional environment was probably oxidizing. This was illustrated in an earlier paper by Powell (1988) who plotted worldwide distributions for Pr/Ph values for oils derived from terrigenous source materials. The histogram maximized in specific values of the region around values of 3 to 5. Pr/Ph values below this value should be treated with some degree of caution and care taken to see if alternative sources may also be responsible for the lower values rather than simply differ-
ences in the nature of the depositional environment. In addition to Pr and Ph, many of the oils often contain high concentrations of long-chain isoprenoids ranging up to at least C40 or C45 (Fig. 3). The majority of these compounds are regular isoprenoids and not the more unusual head-to-head compounds commonly associated with archeabacteria. Whilst the presence of these compounds may reflect the enhanced levels of microbial activity in the original oxic depositional environments, the possibility that they are also derived from polyprenols known to be associated with higher plant waxes cannot be excluded. For example, solanesol (C4s) was isolated from tobacco leaves where it was found partly as fatty acid esters (Rowland et al. 1956); Lingren (1965) isolated a series of C30 to C4s terpenolds, betualprenols from birch wood and in 1964 Burgos et al. identified a Cllo isoprenoid substance as a metabolic product of Aspergillum fumigatus. It is anticipated that detailed studies of the isotopic composition of these long chain compounds will provide additional information on the origin of these compounds. Sesquiterpanes
Abundant concentrations of sesquiterpanes were unequivocally identified for the first time in a number of Australian crude oils known to be
loo% ! m/z
183
33 28 3B I I
"
r "
"
I '
' "
"
t
" " "
"
I
"
'
"
"
I
. . . .
I
. . . .
t
. . . .
i'~
•
•
,
i
. . . .
1
. . . .
'
. . . .
I N C R E A S I N G R E T E N T I O N TIME - - - >
Fig. 3. Long-chain head-to-tail isoprenoids determined by SIM of the ion at m/z 183 in an oil from the Taranaki Basin, New Zealand.
OILS DERIVED FROM TERRIGENOUS SOURCE MATERIALS sourced from terrestrial source material (Fig. 4). Earlier reports by Bendoraitis (1974) had tentatively identified the presence of various bicyclic compounds in a highly degraded seep of unspecified origin. The early work of Alexander et al. (1983, 1984) identified the major sesquiterpanes previously detected by Philp et al. (1981), to be based on the drimane and eudesmane structures. In view of the apparent absence of eudesmanes in samples older than Devonian, it was concluded that the eudesmanes were associated with an input from higher plants. The abundance of the drimanes in these older sediments led to the conclusion that they were associated with a microbial input. However, these conclusions may need to be reconsidered since drimane precursors have also been found in higher plants. In all probability these sesquiterpanes are more of an indication of the bacterial activity in the more oxic type of depositional environments responsible for the accumulation of the higher plant material. Indeed Alexander et al. (1984) have also shown that in certain cases there was a similarity between the distributions of the hopanes and the drimanes supporting the idea that the sesquiterpanes may indeed be derived from the hopanes and formed via a microbial mechanism. Early papers also proposed that the distribution of sesquiterpanes ranged from C14 to C~6 but a recent paper by Wang et al. (1990)
Ioo~
77
has shown that these compounds extend to C24 in a Chinese coal sample. Various other types of sesquiterpanes have also been found in oils derived from terrigenous source materials. These include cadalanes known to be derived from resin precursors particularly in oils from the Far East (van Aarsen et al. 1992) and cedrene and cuparene derivatives have also been found in Chinese oils known to contain a predominance of higher plant source material (Yongsong et al. 1991). In summary, sesquiterpanes such as the drimanes in particular, and possibly the eudesmanes, reflect the degree of microbial activity during the original episode of deposition as much as the terrigenous source input. The possibility of using these compounds as maturity indicators has not been investigated in detail but is something that may be potentially useful (Noble et al. 1987). Care also needs to be taken in sample preparation if one is to look at these compounds in quantitative detail since they are relatively volatile and can be easily lost during sample preparation.
Diterpanes Tri- and tetracyclic terpanes in the C19 and C2o range have long been known to be associated with resins from higher plants (Thomas 1969). The presence of their hydrocarbon analogues in
HOMODRIMANE
DRIMANE
m / z 123
] ' ' '
' / '
" '
'
I
'
'
~
'
1
~
'
'
'
I
'
'
'
'
I
'
INCREASING RETENTION TIME - - - >
Fig. 4. Sesquiterpanes in an oil derived from terrigenous source material as determined by SIM of the ion at m/z 123.
78
R . P . PHILP Vl] loo%
vI
III
,,,e. ~4,--,,',,,' ~ '
'
'
IV
IX
Iv,,-'S.~,,- - ~
I
'
~
'
'
I
'
'
'
'
I
'
~
'
'
I
'
INCREASING RETENTION TIME
'
~
'
I
'
~
'
'
--->
Fig. 5. Typical diterpane distribution in a crude oil from the Gippsland Basin. Major components (peaks I-IX) are identified in Table 3.
oils and source rocks has been used to establish a higher plant input to the source rocks. Diterpanes are abundant in many oils known to contain higher plant materials from countries such as Australia, New Zealand, Taiwan, Papua New Guinea, and Indonesia in the Southern Hemisphere (Philp et al. 1981; Alexander et al. 1988; Weston et al. 1989). An example of the diterpane distribution in a crude oil is shown in Fig. 5 and their identities listed in Table 3. Other areas such as the Beaufort-Mackenzie Delta region in Canada where the source rocks contain vast quantities of resinite also contain high
Table 3. Major diterpanes identified in oils derived f r o m terrigenous source materials Compound Peak number
Name
I
8fl(H)-labdane 4fl(H)- 19-norisopimarane rimuane 17-nortetracyclane
II III IV V VI VII VIII IX
ent-beyerane
iso-pimarane 16fl(H)-phyllocladane ent- 16ct(H)-kaurane 16a(H)-phyllocladane
concentrations of diterpanes (Snowdon & Powell 1982). Perhaps the most widely studied oils in terms of their diterpane distributions are those from the Gippsland and Taranaki basins of Australia and New Zealand respectively (Fig. 5; Philp et al. 1981; Weston et al. 1989). After an early report of these compounds in the Gippsland Basin oils by Philp et al. (1981), Alexander and coworkers published a series of papers detailing the identity of many of these compounds, changes to their stereochemistry with increasing maturation and assignment of specific structures to different plant types (R. Alexander et al. 1987, 1988; Noble et al. 1985a, b, 1986). The 12 diterpanes most commonly found in crude oils and sediments can be classified into 6 families on the basis of structural similarities, namely the labdane, abietane, pimarane, beyerane, kaurane and phyllocladane families (Fig. 6a and b). A table summarizing the frequency of published reports of the occurrence of these compounds in various plant types was published by R. Alexander et al. (1987) and Noble et al. (1985a) and is reproduced as Table 4. Diterpenoids based on the labdane, abietane, pimarane, beyerane, kaurane and phyllocladane structures typically occur in gymnosperms. Kauranoid-type diterpenoids are probably an ubiquitous component of all angiosperms. In contempary resins, pimarane-
OILS DERIVED FROM TERRIGENOUS SOURCE MATERIALS
79
Podocarpaceae
ent-Beyerane
17-Nortetracyclane
16J3(H)-Phyllocladane
($TEREOCHEMISTRYUNASSIGNED)
(a)
16c~(H)-ent-Kaurane
Araucariaceae
(Agathis - Kauri Pine)
15,19-Bisnorlabdane
(b)
16c~(H)-Phyllocladane
Fichtelite
19-Norlabdane
Rimuane
Labdane
19-Norisopimarane
Isopimarane
Fig. 6. Typical diterpenoid structures present in (a) Podocarpacean (gymnosperm) resins; (b) Araucariacean (gymnosperm) resins. type diterpenoids occur in conifers of the Pinaceae and Cupressaceae families and the southern conifer families of Podocarpaceae and Araucariaceae. Precursors of phyllocladane occur widely in the Podocarpaceae and kauranes in the Podocarpaceae, Araucariaceae, and Taxodiaceae. Labdane-type diterpanes were identified in the Athabasca tar sands and considered to have a microbial origin. However, labdane precursors are also present in southern conifers and if the labdanes occur in the absence of their higher homologues it is likely they are derived from conifers and not of a microbial origin.
Whilst most of the characteristic properties of oils from terrigenous source materials are associated with the saturate fractions of the crude oils there are a number of interesting and associated studies related to the aromatic fractions particularly with the diterpenoids. The most significant study in this area is by Alexander et al. (1988), who used aromatic biomarkers derived from various diterpenoids to differentiate oils derived from Permian sediments in the Australian Cooper Basin versus those derived from the younger Jurassic sediments of the overlying Eromanga Basin. The rationale for this is rather simple. The Arau-
+ + ---
++
+
+
+
+
+ ++
++
Pimarane a
--
--
++
+
Beyerane b
Tetracyclic
+ +
+ +
+ ++
+ + +
Kaurane
--
+ +
Phyllocladane
a P i m a r a n e s include i s o p i m a r a n e s a n d r i m u a n e s ; b includes s t a c h e n e s a n d h i b a e n e s ; + + + : w i d e s p r e a d ; + + : c o m m o n ; + : i n f r e q u e n t ; - - : r a r e or u n r e p o r t e d .
+
++
Angiosperms ( f l o w e r i n g plants) Gymnosperms ( m a i n l y conifers) Pteridophytes (ferns) Bryophytes ( m o s s e s a n d liverworts)
Abietane
Bicyclic l a b d a n e
Tricyclic
Occurrence of diterpenoids in land plants (reproduced from R. Alexander et al. 1987)
Plant t y p e
T a b l e 4.
oo
OILS DERIVED FROM TERRIGENOUS SOURCE MATERIALS cariaceae (kauri pines) became prominent in the early to mid-Jurassic and were absent from the older sediments. When the sediments from the Eromanga Basin were examined they were found to contain diterpanes characteristic of the Araucariaceae, whereas sediments from the Permo-Triassic Cooper Basin had different distributions of the saturate diterpanes and particularly the aromatic hydrocarbons. Aromatization of the resin-type diterpanes typically found in the Araucariaceae will lead to four major products, namely 1,2,5-trimethylnaphthalene, 1,7-dimethylnaphthalene, retene, and 1-methylphenanthrene (Fig. 7). If these compounds are not present in similar concentration or one or more is absent, then it can be said with a high degree of certainty that the sediments from which they are derived did not contain any resins from the Araucariaceae as was the case in the Cooper Basin sediments. In addition, it has been observed that isomerization reactions which occur to the diterpanes, occur before the onset of oil generation. Hence these compounds provide a useful indication of maturity in a region where the more commonly used vitrinite reflectance scale is inoperable. The two compounds used
81
here are kaurane and phyllocladane which can exist as both 16a(H) and 16fl(H) isomers. The 16fl(H) isomer is formed, from a precursor compound, in preference to the thermodynamically more stable 16a(H) isomer by a kinetically favoured reaction pathway. As the maturity level increases, the equilibrium distribution of the two isomers reaches a ratio of approximately 4 : 1 (Fig. 8; G. Alexander et al. 1987; R. Alexander et al. 1987). Finally, it has also been proposed by R. Alexander et al. (1987) that ratio of diterpanes/hopanes can be used as an indicator of oxicity. As the environment becomes more oxic the relative amounts of hopanes will increase due to higher levels of microbial activity hence lowering this ratio. (Resin compounds from angiosperms and found in oils from Indonesia will be discussed below.)
Sesterterpanes Sesterterpanes are terpanes based on a C 2 4 hydrocarbon skeleton and are generally tetracyclic compounds and are particularly abundant in certain oils derived from terrigenous source
NATURAL PRODUCTS
R=H, CH 3 ROOC d " N)
R.
Agathicacidand Methylagathate
.
.
Communolsand Communieacid
AROMATIC BIOMARKERS
1,2,5-Trimethylnaphthalene
H2OH , COOH . .
~
R=CH2OH , COOH R Sandaracopimaradienoland Sandaracopimaricacid
SATURATED BIOMARKERS
1,7-Dimethylnaphthalene 15,19-Bisnorlabdane 91
Retene
Abieticacid
1-Methylphenanthmae
19-Norlabdane
Labda~
MATURATION
19-Norisopimarane
Fig. 7. Diagenetic pathways of diterpenoids proposed by Alexander et al. (1988).
Fichtelite
82
R . P . PHILP
~H
~ -
"~H
types, hopanes and non-hopanoid-type terpanes. It is easiest to deal with the two groups separately since the distribution of the latter is source related, whereas the former tend to reflect the oxicity of the depositional environment.
1613(H(4) )-Phyol cal dane 16ct(H)(1) -Phyol cal dane Fig. 8. Proposed isomerization reactions between the 16fl(H) and 16a(H)-isomers ofphyllocladane (shown) and kauranes.
materials. In an earlier paper it was proposed that significant concentrations of the 17,21-C24secohopane were particularly characteristic of oils derived from terrigenous source materials, particularly in the case of the Gippsland Basin, Australia (Philp & Gilbert 1986). Initially it was proposed that this was a source-related phenomenon, but more recently it has become apparent that it is not as much a source-related phenomena but rather a reflection on the nature of the original depositional environment. The nearshore deltaic environment or mire environment in which the source materials were deposited provided a relatively oxic-type environment. This, in turn, would lead to enhanced levels of bacterial activity and production of high levels of hopanoid precursors. The high levels of bacterial activity would subsequently lead to degradation of the hopanes and production of the 17,21-secohopanes, the lower end member of which is the C24 compound. Following the initial identification of the C24 secohopanes more recent work has shown the presence of a variety of degraded compounds in the oils derived from terrigenous source materials, including New Zealand, Nigeria, Indonesia and Taiwan (Woolhouse et al. 1992) (Fig. 9). Compounds tentatively identified include the C24 des-A-ring analogues of the oleananes, lupanes, and ursanes which are typically very abundant in many of these oils (Fig. 10). Whilst it may be proposed that these des-A-ring compounds form via some method of A-ring degradation, it should be noted that a number of ring-A fissioned derivatives of pentacyclic triterpen-3-ols have been reported (Baas 1985; Baas & van Berkec 1991; Baas et al. 1992).
Pentacyclic terpanes The pentacyclic terpanes found in oils from terrigenous sources can be divided into two
Hopanes
The distribution of hopanes in oils derived from predominantly terrigenous source materials is very simple. For the most part the distribution is dominated by the regular hopanes, maximizing at C3o with the concentration of the extended hopanes above C31 decreasing exponentially, characteristic of the distribution associated with more oxic-type environments. Some of the oils from the Gippsland Basin have been observed to contain high concentrations of the C31 homohopanes (Fig. l l a ; Philp et al. 1981). In those oils, this probably arises from a source-related phenomenon since there are reports in the literature of C3a hopanoid precursors occurring in Recent peat deposits (Quirke et al. 1984) and brown coal deposits. In another unusual example from Gippsland, the Turrum oil was found to have a value for the 22S/22R epimers of the C31 homohopane epimers of approximately 1 : 9 compared to the more commonly observed ratio of 3:2. This anomaly was proposed by Philp & Gilbert (1982) to represent a case of an oil dissolving biomarkers from an immature coal sequence during migration (Fig. llb). Methylhopanes and the various series of norhopanes identified in sample sets from other environments and source materials are generally absent from oils derived from terrigenous sources although ten Haven et al. (1992) did report the occurrence of methylhopanes in coal samples. The presence of a compound referred to as compound X eluting just after the Cz9 hopane was described in an earlier paper (Philp & Gilbert 1986). This compound appeared to be a characteristic feature of oils derived from terrigenous source materials, particularly those from the Gippsland Basin. Although it was tentatively suggested that this compound was source related it was not until recently that compound X was actually identified by Moldowan and co-workers (1991) as a diahopane and found not to be a source-related parameter. Indeed Moldowan et al. (1991) proposed that this compound is formed during diagenesis from rearrangement of regular hopanoid natural products and thus in turn is more of an environmental indicator than a source indicator. It also appears that the relative concentration of this compound increases
OILS D E R I V E D F R O M T E R R I G E N O U S S O U R C E M A T E R I A L S 1E,e4
330-PAR 191 24
t24* 24
e6e
IE~04 .349
344-PAR 191
25
25*
,~,
,,;,
,,'.
. . . . .
~2,,
'
'
"
' ^',-,
'
=E+83 t3. ? 5 ?
358-PAR 191 26
ono
,
le0e
900
i tree
26*
.
.
.
.
I gee
.
.
.
.
i leee
.
,
•
q
~^, 13oo
•
JE*04 ,422
27
372-PAR 191
ooe
! 12oe
.
.
.
i I!oe
.
.
.
.
A,, i t2oo
.
.
.
.
i t3ee
INCREASING RETENTION TIME Fig. 9. MS/MS data showing distributions of sesterterpanes in an oil.
,
,
83
84
R. P. PHILP i
|11111
(1) 1013(H)-des-A-oleanane (R=H,R'=CH3) (3) 1013(H)-des-A-ursane (R'=H, R=CH 3)
/1|11'"
(2) 1013(H)-des-A-lupane
(4) C24-17,21-secohopane
Fig. 10. Tetracyclic terpanes present in oils from terrestrial source materials.
with increasing maturity demonstrating the fact that it is more thermally stable than other members of the hopane series.
N o n - h o p a n o i d terpanes
Non-hopanoid terpanes derived from higher plant sources are far more abundant than the hopanoids in many oils from terrigenous source materials. Higher plants contain a wide variety of triterpenoids and many of their hydrocarbon derivatives can be found in the corresponding oils and source rocks (Whitehead 1974). This paper is not the place to go into great detail about the wide variety of non-hopanoid triterpanes that have been identified in these oils. However, in brief, there are the compounds based on pentacyclic structures such as oleananes, lupanes and ursanes, along with their demethylated analogues (Fig. 12). These compounds are abundant in the oils of New Zealand (Czochanska et al. 1988), Taiwan (Philp & Oung 1991), Beaufort-Mackenzie Basin, Canada (Curiale 1991), and Nigeria (Woolhouse et al. 1992). The other major type of triterpanes is based on the cadalane structures (Fig. 13), derived from precursors present in the dammar
resins, and particularly abundant in oils from Indonesia (van Aarssen et al. 1992). One particular feature to note is that the oils of the Gippsland Basin contain copious quantities of the diterpanes indicative of the higher plant input but, because of their pre-Tertiary origin, contain none of the complex mixtures of non-hopanoid terpanes that the oils from the other regions of the world contain. Clearly, variations in the triterpenoid distributions reflect variations in the nature of the plant material present in the source rocks and in turn the age of the oil itself. Oleananes may be derived from a number of naturally occurring precursors, including taraxer-14-en-3fl-ol and olean-12-en-3fl-ol, which through a series of complex reduction and isomerization reactions can produce two isomers of oleanane (ten Haven & Rullk6tter 1988). Initially it was thought that the oleananes existed as only one isomer in both oils and source rocks. However, with a good high efficiency capillary column, it is possible to show that there are two isomers present in many of these samples, namely the 18a(H) and 18fl(H)-isomers. It has also been shown that the ratio of these compounds will change with increasing maturity until an apparent equilibrium ratio is reached
OILS DERIVED FROM TERRIGENOUS SOURCE MATERIALS
85
lOO%
m / z 191 C31
C30 C29 Tm
J, ' ~ ....... r--'-'r--'--[
~
I
i
(a)
I
t
J
I
i
INCREASING RETENTION TIME - - - >
100%" m/z 191 c
, ''
(b)
'
'
'
I
'
'
'
'
,Jl I
'
'
'
'
I
'
'
2~
It '
'
I
'
'
~
'
I
'~
'
'
~
I
~
'
'
'
I
'
'
'
'
I
'
'
'
'
I
'
'
I
,
INCREASING RETENTION TIME - - - >
Fig. 11. (a) Hopane distribution (m/z 191) in a crude oil dominated by an abundance of Ca1 homohopanes. (b) An unusual 22S/22R Ca1 hopane distribution in the Turrum oil from the Gippsland Basin.
(Riva et al. 1988). It has been proposed that this ratio is particularly useful in the early stages of oil generation. However, whether its potential as a maturity indicator will ever be realized is unclear due to the problems of resolving the two isomers in a reproducible fashion. A variety of bisnorlupanes, norlupanes, nor-
oleananes, 28,30-bisnorhopane, de-A-lupane and dehydroabietane have been recently identified in oils from the Beaufort-Mackenzie region by Curiale (1991; Fig. 14). Peakman et al. (1991) identified 24-nor-urs-12-ene and 24-nor-olean12-ene in a Beaufort-Mackenzie oil as well as 24-nor-lupane and a series of 24-nor-oleananes.
86
R. P. PHILP
-2 23
24
24,27-Dinoroleanane
Oleanane
~hl,
24,28-Dinor-18oc-oleanane
i,,
23 24 Lupane
24-Norlupane
24-Norarborane
(Trendel eta./., 1991 - Egyptian oil)
CH
3
H
24-Nor-urs-12-ene
CH
3
H
24-Nor-olean-12-ene
(Peakman et aL, 1991. Beaufort~Mackenzie Delta Oil)
Fig. 12. Non-hopanoid type triterpanes present in oils from terrestrial source materials.
Trendel et al. (1991) reported the presence of 24-norlupane, 20-noraborane, 24,27-dinoroleanane and 24,28-dinor-18a-oleanane in an Egyptian oil. In both of these cases the presence of the compounds correlates with the contribution of higher plant material in the source rocks for these oils. In a 1983 paper Grantham et al. (1983) tentatively assigned the labels W, X, and T to three prominent compounds in oils from the Far East. Subsequent work by Cox, van Aarssen and others has shown that these compounds are bicadinane isomers derived from precursors that
are present in dammar resins (van Aarssen et al. 1992). Although the predominant members of the series are in the C3o range it has been shown that analogues of these compounds are present in the higher molecular weight region with components up to C60 and C75 and possibly higher. These bicadinane structures have also been reported in samples from Utah, the Messel shale and more recently in oils from New Zealand by Curry et al. (this volume). The presence of the saturated hydrocarbons is often accompanied by 1,6-dimethylnaphthalene and cadalene in the aromatic fraction.
OILS DERIVED FROM TERRIGENOUS SOURCE MATERIALS
87
,IIIIL~ -
=
Compound
T -
=
(trans-trans-trans)
1,6-DMN
=
Compound W - (cis-cis-trans)
Cadalene
Fig. 13. Angiosperm resin terpenoid structures based on the cadalane-type structure (van Aarssen et al. 1992).
Steranes The sterane distributions in most oils derived from terrigenous source materials are very simple and dominated by the C29 steranes. In the situation where there is a mixed input with
higher plant material as well as algal material the problem is more complex. In addition to the C27 and C28 steranes, some of the C29 steranes may also be derived from algal material. The other characteristic feature of the sterane distribution
1OO%-
24-Norlupane 17B(H)-24,28-Bisnorlupane
17a0I)-24,28-Bisnorlupane
m / z 177
Hopane L Oleananes~ll
INCREASING RETENTION TIME
--->
Fig, 14. M / Z 177 chromatogram for a sample from the Canadian Beaufort area (Curiale 1991) containing significant quantities of various demethylated lupane derivatives plus oleanane.
88
R. P. PHILP Table 5. Summary of quantification data for steranes/terpanes in crude oils Sample location
C27"
C2s*
C29"
C3o*
217/191
Offshore Taiwan Marine-sourced oils
8.09 43.43
7.83 30.75
25.33 34.75
1.89 13.36
0.80 3.23
* Expressed in ng//~l using d 2 a s internal standard.
for these types of samples is the presence of diasteranes. It has long been proposed that the relative concentration of the diasteranes reflected the presence of clay minerals and their ability to catalyse sterene rearrangement reactions. However, it has become clear in recent years that the oxicity of the depositional environment provides another clue to the fate of the sterols. Highly anoxic environments will rapidly reduce the sterenes, hence reducing the amount of sterenes available for the rearrangement reaction. However, a more oxic-type environment will leave more of the sterenes unaffected and hence available for rearrangement. It is possible, therefore, that the concentration of the rearranged steranes in these oils more closely reflects the oxicity of environment rather than the presence or absence of the clay minerals (Moldowan et al. 1986). The ratio of hopanes/steranes in oils, based on the m/z 191 and rn/z 217 chromatogram respectively, derived from terrigenous source materials is typically greater than 5 and as large as 10 in the case of Indonesian oils. This, again, is a characteristic of these oils and results from the high level of bacterial activity in the depositional environment and it is not a direct source indicator. The absolute concentrations of steranes in oils derived from terrigenous source materials is also much lower than in oils from marine source materials. A brief summary of results from a quantitation study performed on oils from Taiwan and compared with oils known to be derived from marine source materials is summarized in Table 5. In this summary average values are given from the analysis of several samples but a more comprehensive set of results can be found in the paper by Philp & Oung (1991).
Summary In the preceding sections a number of the important geochemical characteristics of oils derived from terrigenous source materials have been described. It is true to say that not all of the features will be common to all such oils, due to
variations in age and nature of plant types responsible for the organic matter in the source rocks. However, when reviewed collectively, any oil containing a number of these characteristics is probably sourced partially, if not entirely, from abundant higher plant material. It is also important to remember that due to the nature of the environment in which this organic matter is typically deposited, mire or nearshore coastal or fluvial-deltaic, the level of microbial activity will be relatively high due to high levels of oxicity. The contribution of organic matter from the microbial biomass will be significant, therefore, and reflected in terms of the contribution from the hopanes and other classes of terpanes. In the following list the most important geochemical characteristics of oils from terrigenous source materials are summarized along with an indication as to whether they more closely reflect the depositional environment or source input. • High concentration of saturated hydrocarbons relative to aromatic, polars and asphaltenes (Source). • Low, or negligible, S content (Depositional environment). • n-Alkane distributions typically maximize in C2o-C40 region (Source). • Pr/Ph ratios typically greater than 4 (Depositional environment). • Drimanes and e u d e s m a n e s important sesquiterpanes reflect both source and depositional environment. • Diterpanes - reflect source and can be used to differentiate plant types and age of oils. • Sesterpanes - derived from pentacyclic analogues, probably as a result of microbial action, and can be source and environment related. • Hopanes, for the most part, reflect oxic nature of depositional environments generally associated with the source of this type of oils. Concentration of the extended hopanes decrease exponentially with increasing carbon number.
OILS D E R I V E D FROM TERRIGENOUS SOURCE MATERIALS • M e t h y l h o p a n e s and n o r h o p a n e s not comm o n l y associated with these oils. • C o m p o u n d X c o m m o n in m a n y of these oils has n o w b e e n identified as a d i a h o p a n e . • N o n - h o p a n o i d t e r p a n e s associated with these oils include o l e a n a n e s , lupanes, ursanes, p o l y c a d i n a n e s , All reflect inputs of various plant types. U n s a t u r a t e d triterpanes also
89
p r e s e n t in s o m e oils derived f r o m t e r r i g e n o u s source material. • Steranes are d o m i n a t e d by C29 derivatives, truly indicative of higher plant input, in oils d e r i v e d exclusively from h i g h e r plant material. P r e s e n c e of C27 and C28 i n t r o d u c e s ambiguity as to w h e t h e r C29 is h i g h e r plant or algal.
References
VAN AARSSEN, B. G. K., HESSELS, J. K. C., ABBINK, O. A. & DE LEEUW, J. W. 1992. The occurrence of polycyclic sesqui- , tri , and oligoterpenoids derived from a resinous polymeric cadinene in crude oils from southeast Asia. Geochimica et Cosmochimica Acta, 56, 1231-1246. ADAM, P., TRENDEL, J. M. & ALBRECHT, P. 1991. Novel thiophene derived from higher plant triterpenes in sediments. Tetrahedron Letters, 56, 4179-4182. ALEXANDER, G., HAZAI, I., GRIMALT,J. & ALBAIGES, J. 1987. Occurrence and transformation of phyllocladanes in brown coals from Nograd Basin, Hungary. Geochimica et Cosmochimica Acta, 51, 2065-2073. ALEXANDER, R., KAGI, R. & NOBLE, R. 1983. Identification of the bicyclic sesquiterpenes drimane and eudesmane in petroleum. Journal of the
Chemical Society Chemical Communications, 226-228. , , & VOLKMAN, J. K. 1984. Identification of some bicyclic alkanes in petroleum. Organic Geochemistry, 6, 63-70. , LARCHER, A. V., KAGI, R. I. & PRICE, P. L. 1988. The use of plant-derived biomarkers for correlation of oils with source rocks in the Cooper/Eromanga basin system, Australia. Australian Petroleum Exploration Association Journal, 28, 310-324. --, NOBLE, R. A. & KA6I, R. J. 1987. Fossil resin biomarkers and their application in oil to sourcerock correlation, Gippsland basin, Australia. Australian Petroleum Exploration Association Journal, 27, 63-72. BAAS, W. J. 1985. Naturally occurring seco-ringA-triterpenoids and their possible biological significance. Phytochemistry, 24, 1875-1889. --& VAN BERKEL, I. E. M. 1991. 3,4-Secotriterpenoid acids and other constituents of the leaf wax of Hoya Naumanii. Phytochemistry, 30, 1625-1628. --, --, VERSLUIS, C., HEERMA, W. & KREYENBROEK, M. N. 1992. Ring-A fissioned 3,4-seco3-nor-triterpene-2-aldehydes and related pentacyclic triterpenoids from the leaf wax of Hoya Australia. Phytochemistry, 31, 2073-2078. BENDORAITIS, J. G. 1974. Hydrocarbons of biogenic origin in petroleum - aromatic triterpenes and bicyclic sesquiterpenes. In: TIssoT, B. & BIENNER, F. (eds) Advances in Organic Geochemistry 1973. Editions Technip, Paris, 209224.
BROOKS, J. D. & SMITH, J. W. 1969. The diagenesis of plant lipids during the formation of coal, petroleum and natural gas- I I - Coalification and the formation of oil and gas in the Gippsland Basin. Geochimica et Cosmochimica Acta, 33, 1183-1194. BURGOS, J., BUTTERWORTH,P. H. W., HEMMING,F. W. & MORTON,R. A. 1964. The biosynthesis of longchain polyisoprenoid alcohol by Aspergillus fumigatus, Fresenius. The Biochemical Journal, 91. CURIALE, J. 1991. The petroleum geochemistry of Canadian Beaufort Tertiary 'non-marine' oils. Chemical Geology, 93, 21-46. CZOCHANSKA, Z., GILBERT, T. D., PHILP, R. P., SHEPPARD,C. M., WESTON, R. J., WOOD, T. A. & WOOLHOUSE, A. D. 1988. Geochemical application of sterane and triterpane biomarkers to a description of oils from the Taranaki Basin in New Zealand. Organic Geochemistry, 12, 123-135. DEL RIO, J. C. & PHILP, R. P. 1992a. Nature and Geochemistry of High Molecular Weight Hydrocarbons (Above C4o) in Oils and Solid Bitumens. Organic Geochemistry, 18,541-553. -& -1992b. High Molecular Weight Hydrocarbons: A New Frontier in Organic Geochemistry. Trends in Analytical Chemistry, 11, 187-193. -& -1992c. Oligomerization of Fatty Acids as a Possible Source for High Molecular Weight Hydrocarbons and Sulphur-Containing Compounds in Sediments. Organic Geochemistry, 18, 869-880. DIDYK, B. M., SIMONEIT,B. R. T., BRASSELL,S. C. t~ EGLINTON, G. 1978. Organic geochemical indicators of paleoenvironmental conditions of sedimentation. Nature, 271,216-222. FREEMAN, K. H., HAYES, J. M., TRENDEL, J.-M. & ALBRECHT, P. 1989. Evidence from GC-MS carbon-isotopic measurements for multiple origins of sedimentary hydrocarbons. Nature, 353,254-256. GRANTHAM, P. J., POSTHUMA, J. • BAAK, A. 1983. Triterpanes in a number of Far-Eastern crude oils. In: BJOROY, M. et al. (eds) Advances in Organic Geochemistry 1981. J. Wiley and Sons, Chichester, 675-683. TEN HAVEN, H. L., DE LEEUW, J. W., SINNINGHE DAMSTI~, J., SCHENCK, P. A., PALMER, S. E. & ZUMBERCE, J. 1988. Application of biological markers in the recognition of palaeo-hypersaline
90
R.P.
environments. In: FLEET, A. J., KELTS, K. & TALBART, M. R. (eds) Lacustrine Petroleum Source Rocks. Geological Society, London, Special Publication, 40,123-130. , - - , RULLKOTTER,J. & SINNINGHEDAMSTI~,J. S. 1987. Restricted utility of the pristane/phytane ratio as a palaeoenvironmental indicator. Nature, 330, 641-643. , PEAKMAN,T. M. & RULLKOLLTER,J. 1992. Early diagenetic transformation of higher-plant triterpenoids in deep-sea sediments from Baffin Bay. Geochimica et Cosmochimica Acta, 56, 2001-2024. - & RULLKOTrER, J. 1988. The diagenetic fate of taraxer-14-ene and oleanene isomers. Geochimica et Cosmochimica Acta, 52, 2543-2548. HEDBURG, H. D. 1968. Significance of high wax oils with respect to genesis of petroleum. American Association of Petroleum Geologists Bulletin, 52, 736-750. LINGREN, B. O. 1965. Homologous aliphatic Cao-C4s terpenols in birch wood. Acta Chemica Scandinavica, 19, 1317-1326. MCKIRDY, D. M., Cox, R. E., VOLKMAN, J. K. & HOWELL, V. J. 1986. Botryococcane in a new class of Australian nonmarine crude oils. Nature, 320, 57-59. MOLDOWAN, J. M., FAGO, F. J., CARLSON, R. M. K., YOUNG, D. C., VAN DUYNE, G., CLARDY, J., SCHOELL, M., PILLINGER, C. T. & WATT, D. S. 1991. Rearranged hopanes in sediments and petroleum. Geochimica et Cosmochimica Acta, 55, 3333-3353. & SEIFERT, W. K. 1979. Head to head linked isoprenoid hydrocarbons in petroleum. Science, 204, 169-171. , SUNDARARAMAN,P. & SCHOELL,M. 1986. Sensitivity of biomarker properties to depositional environment and/or source input in the Lower Toarcian of SW Germany. In: LEYTHAUSER,D. t~ RULLHOTTER, J. (eds) Advances in Organic Chemistry 1985. Pergamon Press, London, 915-926. NOBLE, R. A., ALEXANDER, R. 8z KAGI, R. J. 1987. Configurational isomerization in sedimentary bicyclic alkanes. Organic Geochemistry, 11, 151-156. --, ., & KNOX, J. 1985b. Tetracyclic diterpenoid hydrocarbons in some Australia coals, sediments and crude oils. Geochimica et Cosmochimica A cta, 49, 2141-2147. - & -1986. Identification of some diterpenoid hydrocarbons in petroleum. Organic Geochemistry, 10, 825-829. ~, KNox, J., ALEXANDER,R. & KAGI, R. I. 1985a. Identification of tetracyclic diterpene hydrocarbons in Australian crude oils and sediments. Journal of the Chemical Society, Chemical Communications, 32-33. PALACAS J. G., ANDERS, D. E. & KING, J. D. 1984. South Florida Basin - A Prime example of carbonate source rocks in petroleum. In: PALACAS, J. G. (ed.) Petroleum Geochemistry and Source Rock Potential of Carbonate Rocks.
PHILP American Association of Petroleum Geologists, Studies in Geology, 18, 71-96. PEAKMAN,T. M., TEN HAVEN, H. L. & RULLKOTTER,J. 1991. Characterisation of 24-nor-triterpenoids occurring in sediments and crude oils by comparison with synthesized standards. Tetrahedron, 47(23), 3779-3786. PHILP, R. P. ~ GILBERT, T. D. 1982. Unusual distribution of biological markers in Australian crude oil. Nature, 229,245-247. --& 1986. Biomarker distributions in Australian oils predominantly derived from terrigenous source material. In: LEYTHAUSER,D. & RULLKOTq'ER, J. (eds) Advances in Organic Geochemistry 1985. Pergamon Press, London, 73-84. --, t~Z FRIEDRICH, J. 1981. Bicyclic sesquiterpenoids and diterpenoids in Australian crude oils. Geochimica et Cosmochimica Acta, 45, 1173-1180. - - & OUNG, J.-N. 1991. Biomarker distribution in crude oils as determined by tandem mass spectrometry. In: MOLDOWAN,J. M., ALBRECHT, P. & PHILP, R. P. (eds) Biological Markers in Sediments and Petroleum. Prentice Hall, Englewood Cliffs, New Jersey, 106-123. POWELL, T. G. 1988. Developments in concepts of hydrocarbon generation from terrestrial organic matter. In: WAGNER, H. C., WAGNER, L. C., WANG, F. F. H. & WONG, F. L. (eds) Petroleum resources of China and related subjects. CircumPacific Council for Energy and Mineral Resources Earth Science Series, 10,807-824. - & McKIRDY, D. M. 1973. Relationship between ratio of pristane to phytane, crude oil composition and geological environment in Australia. Nature, 243, 37-39. QUIRKE, M. M., WARDROPER, A. M. K., WHEATLEY, R. E. & MAXWELL, J. R. 1984. Extended hopanoids in peat environments. Chemical Geology, 42, 25-43. RIVA, A., CACCIALANZA,P. G. & QUAGLIAROLI,C. F. 1988. 18fl(H)-Oleanane in crudes and in Tertiary-Upper Cretaceous sediments. Definition of a new maturity parameters. In: MATTAVELI, L. & NOVELLI, L. (eds) Advances in Organic Geochemistry 1987. Pergamon Press, Oxford. ROWLAND, R. L., LATIMER,P. H. & GILES, J. A. 1956. Flue-cured tobacco I. Isolation of solanesol, an unsaturated alcohol. Journal of the American Chemical Society, 78, 4680-4683. SNOWDON, L. R. & POWELL, T. G. 1982. Immature oil and condensate-Modification of hydrocarbon generation model for terrestrial organic matter. American Association of Petroleum Geologists Bulletin, 66,775-788. TELEGAAR, E. 1990. Resistant biomacromolecules in morphologically characterized constituents of kerogen: A key to the relationship between biomass and fossil fuels. PhD thesis, Technical University of Delft, Delft. THOMAS, B. R. 1969. Kauri resins- modern and fossil. In: EGLINTON, G. & MURPHY, M. T. J. (eds)
OILS DERIVED FROM TERRIGENOUS SOURCE MATERIALS Organic Geochemistry - Methods and Results. Springer-Verlag, Berlin, 599-618. TRENDEL, J. M., ALBRECHT,P., RIVA, A. & GHISELLI, C. 1991. Novel demethylated higher plant triterpanes in petroleum. In: MANNING,D. A. C. (ed.) Organic Geochemistry. Advances and applications in the natural environment. Manchester University Press, Manchester, 212-214. WANG, T-G., SIMONEIT, B. R., PHILP, R. P. & Yu, C. P. 1990. Extended 8fl(H)-drimane and 8,14secohopane series in a Chinese boghead coal. Journal of Energy and Fuels, 4, 177-183. WESTON, R. J., PHILP, R. P., SHEPPARD, C. M. & WOOLHOUSE, A. D. 1989. Sesquiterpanes, diterpanes and other higher terpanes in oils from the Taranaki Basin of New Zealand. Organic Geochemistry, 14, 405--421.
91
WHITEHEAD, E. V. 1974. The structure of petroleum pentacyclanes. In: TISSOT,B. & BIENNER,F. (eds) Advances in Organic Geochemistry 1973. Editions Technip, Paris, 225-243. WOOLHOUSE, A. D., OUNG, J-N., PHILP, R. P. & WESarON, R. L 1992. Triterpanes and ring-A degraded triterpanes as biomarkers characteristic of Tertiary oils derived from predominantly higher plant sources. Organic Geochemistry, 18, 23-31. YONGSONG,H., ANSONG,G., Fu, J., SHENG,G., ZHAO, B., CHENG, Y. • LI, M. 1991. A comparative study of biomarker assemblage, oil precursor and depositional environment of Damintun extra high wax oils. In: ECKARDT, C. B., MAXWELL, J. R., LARTER,S. R. & MANNING,D. A. C. (eds) Advances in Organic Geochemistry 1991, Pergamon Press, London, 29-39.
Chemical heterogeneity among adjacent coal microlithotypesimplications for oil generation and primary migration from humic coal S C O T T A. S T O U T Unocal Energy Resources Division, P O B o x 76, Brea, C A 92621, U S A Abstract: Laser radiation focused onto a coal polished perpendicular to bedding has been
used to thermally extract and pyrolyse the organic matter from three adjacent microlithotypes in a Tertiary, autochthonous, hydrogen-rich, sub-bituminous (early catagenesis/ bituminization stage) humic coal from Indonesia. The distribution of hydrocarbons released during laser irradiation are shown to vary on a millimetre scale between adjacent humotelinite-, liptodetrinite-, and resinite-dominated microlithotypes. This variability is not due to varying thermal maturity but rather to (1) the inherent chemical variability of the macerals comprising each microlithotype; and/or (2) the presence of migrated, or nonindigenous, bitumen/oil. Hydrocarbon yields from each microlithotype suggest that primary migration between microlithotypes has occurred. Specifically, long chain n-alkanes and isoprenoids, which were most likely generated in the cutinite-derived liptodetrinitedominated microlithotype (liptoclarite), are much more abundant in the overlying liptinitefree, humotelinite-dominated microlithotype (telite) and in the underlying resinitedominated microlithotype (resinoclarite). These observations suggest that while the type of macerals plays an important role in the generation of oil within coals, it is the physical association between macerals (i.e. the microlithotypes present) which significantly influences the primary migration of oil within, and expulsion of oil from, coal seams. Thus since coals are extremely variable, both chemically and physically, generalizations suggesting all liptinite-rich humic coals can act as oil source rocks may be misleading. Liptinite-rich humic coals with high concentrations of finely-admixed liptinite and huminite macerals (e.g. liptodetrinite in a humodetrinite and mineral matrix, as would occur in microlithotypes with a detrital origin) may permit both oil generation and expulsion. Characteristics of the telite suggest, due to their high absorptive capacities and (micro)porosity throughout most of bituminization/catagenesis, coals which contain an abundance of massive huminite macerals (e.g. humotelinite or gelinite) may tend to trap hydrocarbons generated from nearby or encased liptinites within their networks. Such macerals (the precursors to desmocoUinites in bituminous coals) may become relatively hydrogen rich and often fluoresce under uv light and may not expel hydrocarbons until they are subsequently cracked to gas.
Coal is comprised of heterogeneous mixtures of macerals and lesser amounts of minerals. Coal macerals, however, seldom occur isolated from one another but instead are intimately associated with other macerals. These associations of macerals are termed microlithotypes (Stach et al. 1982) and are considered the 'organic facies' of coals (Jones 1987). When coal is polished perpendicular to its bedding plane and examined under the microscope, microlithotypes can be recognized as individual layers parallel or sub-parallel to the bedding planes. The International Commission for Coal Petrology (ICCP) dictates that these layers be at least 50,um thick before one microlithotype should be distinguished from another. Microlithotypes are classified according to the predominant type of maceral(s) present within each layer (Table 1). The classi-
fication of microlithotypes, thus, is essentially a classification of (1) the type of precursor organic matter; and (2) its degree of early diagenesis experienced in the original mire (Stach et al. 1982). It is well known that the chemical and physical properties of the different microlithotypes play a vital role in determining the technological properties (e.g. coking power) of a coal (Stach et al. 1982). This, in turn, arises from the chemical and physical properties of the individual macerals comprising microlithotypes. Thus, it follows that a coal's potential to generate oil will also be determined by the microlithotypes present. For example, there is general agreement that sapropelic coals, i.e. those with an abundance of hydrogen-rich, liptinite macerals, as would occur in liptites and clarites (Table 1), can generate liquid hydrocarbons (see
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coal and Coal-bearing Strata as Oil-prone Source Rocks? Geological Society Special Publication No. 77, pp. 93-106.
93
94
S. A. STOUT Table 1. Summary of microlithotype classification of bituminous coals modified from Stach et al. (1982) Dominant maceral(s)
Microlithotype
Group name
Collite Telite* Sporite Cutite Resite Algite Liptodetrite Semifusite Fusite Sclerotite Macroite
Vitrite
Monomaceralic
Collinite Telinite Sporinite Cutinite Resinite Alginite Liptodetrinite Semifusinite Fusite Sclerotinite Macrinite
Liptite
Inertite
Bimaceralie
Sporinite/Vitrinite Cutinite/Vitrinite Resinite/Vitrinite Liptodetrinite/Vitrinite Vitrinite/Inertinite Sporinite/Inertinite Cutinite/Inertinite Resinite/Inertinite Liptodetrinite/Inertinite
Sporoclarite Cutinoclarite Resinoclarite* Liptoclarite* Vitrinertite Sporodurite Cutinodurite Resinodurite Liptodetrinodurite
Clarite
Vitrinertite Durite
Trimaeeralie
Vitrinite-dominated mix Liptinite-dominated mix Inertinite-dominated mix
Duroclarite Vitrinertinoliptite Clarodurite
Trimacerite
* Precursors of these microlithotypes were studied herein.
Hunt 1991 and references therein). Sapropelic coals are, however, geologically rare. Alternatively, the geologically much more common humic coals, comprised of predominantly hydrogen-poor vitrite microlithotypes (Table 1), are considered to generate primarily gas (mostly methane and CO2). However, many humic coals, especially those occurring in Tertiary deltaic settings, contain 10-30% liptinites. It is this type of liptinite-rich humic coal that is the main subject of any 'oil-from-coal' debate. It is generally accepted that certain humic coals can generate oil. The primary question remaining surrounds the potential to expel liquid hydrocarbons from the coal network (prior to their being cracked to gas). Does primary migration occur in humic coals (Durand & Paratte 1983; Huc et al. 1986), or does the microporosity and plasticity of the coal network enhance absorption and inhibit primary migration (Saxby & Shibaoka 1986; Landais & Monthioux 1988)? While cracks and fissures in coal are probably important in this regard (Hvoslef et al. 1988), the answer to this question, at least to some degree, must depend on the
physical arrangement of, or association among, macerals (i.e. a coal's 'microtexture'). Recently, this physical property has been suggested to control the primary migration process in certain liptinite-rich humic coals (Bertrand 1989; Mukhopadhyay 1989). Thus in addition to maceral type, the maceral associations, i.e. microlithotypes, are likely to be of utmost importance in determining a coal's potential to generate and expel oil. The objective of this study is to document and compare the petrographic and chemical character of the organic matter within three adjacent microlithotypes in a Tertiary subbituminous, liptinite-rich humic coal from Indonesia. A novel, laser-based technique which permits a collective analysis of the thermal extract and pyrolysate from selected microlithotypes in situ on polished blocks has been employed. Hydrocarbon distributions from the individual microlithotypes are compared to conventional solvent extract and flash pyrolysate from the blended mixture of the three microlithotypes. The results have implications for determining: (1) small-scale variation in organic
COAL MICROLITHOTYPE HETEROGENEITY matter composition within coal seams; (2) the variability in chemistry among different microlithotypes; and (3) the capacity for primary migration in hydrogen-rich humic coals. A subbituminous rank coal was selected over higher or lower ranks as it represents the early stages of bituminization/catagenesis in coals which is generally reported to occur between 0.5 and 1.3% Rm (Stach et al. 1982). S a m p l e description and analytical m e t h o d s
Sample A fresh outcrop sample of a Tertiary, autochthonous, sub-bituminous coal from the Bekoso seam in the Pasir Sub-basin, East Kalimantan (several miles southwest of Balikpapan) was collected. A small (4 cm 3) specimen of unextracted coal was embedded in epoxy resin and polished perpendicular to bedding. Huminite/vitrinite reflectance (random), RockEval pyrolysis yields, and total organic carbon content were determined for an aliquot of the selected coal using standard techniques. These numerical data are shown in Table 2.
Organic petrology Petrographic examination (200 point counts) of the crushed (+ 20 mesh) blended coal revealed an abundance of liptinite macerals (mostly liptodetrinite and resinite) in varying concentrations in a huminite matrix. Inertinite was relatively minor (Table 2). This concentration of liptinite (28%) is usually considered sufficient to generate oil in humic coal (Hunt 1991). Micro-
Table 2. Characteristicsof the blended microlithotypes from the Bekoso coal Maceral composition ( % vol.) Huminite Liptinite Inertinite Huminite reflectance TOC Rock-Eval S1 (mg HC g-1 rock) $2 (rag HC g-1 rock) $3 (mg CO2 g-a rock) Tmax HI (mg HC g-1TOC) OI (mg CO2 g-1TOC) Solvent extract yield (% wt)
68 28 4 0.61 65.8 10.2 209.0 9.5 433 318 14 4.94
95
lithotype analysis of the uncrushed coal polished perpendicular to bedding revealed three petrographically distinct but adjacent microlithotypes selected for this study (Fig. 1). The maceral terminology employed throughout the paper is according to the ICCP (1973) classification for low rank coals. The ICCP has not formally adopted a microlithotype terminology specific to low rank coals. As a result, the terminology employed throughout this paper will be adapted from that used for bituminous coals (Table 1). One might think of the microlithotypes studied as being precursors to those listed in Table 1. For example, the upper microlithotype was dominated by a mixture of ulminite and textinite which was, no doubt, derived from ligno-cellulosic cell wall material (Fig. la). The macerals present would be considered precursors to telinite, thus this microlithotype would be considered a telite precursor, and is thus termed telite herein. Similarly, the middle microlithotype was considered a liptoclarite precursor dominated by a mixture of humodetrinite (attrinite) and liptodetrinite (mostly spore and cuticle fragments; Fig. lb). This microlithotype also contained the highest proportion of mineral matter of the three microlithotypes and could be considered to have a detrital origin. The bottom microlithotype was considered a resinoelarite precursor being dominated by resinite mixed with ulminite (Fig. lc).
Laser extraction and micropyrolysis-GCMS A 4.0 W continuous wave Nd-YAG (CVI Model C-95) near-IR (1064 nm) laser beam was focused through a microscope (with special IR-coated optics) and glass window of a small Teflon sample chamber onto the polished surface of each of the microlithotypes studied. The sample chamber was continually flushed with helium and had a slight vacuum (c. 75 kPa) pulled from the opposite side of the chamber through a heated (290°C) nickel transfer line (Fig. 2). The heat from the laser beam was used to volatilize and pyrolyse the targeted organic matter. The effluent was immediately swept/sucked from the sample chamber into the heated exit tube where it was subsequently trapped using liquid nitrogen (LN2). Quenching and condensation of volatiles on the sample chamber walls or glass window was minimal due to the close proximity of the heated exit tube to the targeted area. Upon release from LN2, the effluent was analysed on-line using gas
96
s . A . STOUT
A
~
~
/
~
~
~
i
~
....
~ f i s s u r e
,
...........
..,
:'-" ,. '-2.i'".~.~""~"h:7.~liptodetrinite
:"~_ ~".-'7:.. ~
":.."$"
:;J:.~ : ' - .--' - ~~"7.-, ~ .:4 [ ' :-..-'.~,.~ . ~ . " ....~'~.sporinite ~ attrinite
/-~~_~~.~,~--~ -~ .... " '. . . . . ". . . . '~:'~-Z,~
r..'.--- -',w-"'~_. ~ . . . . . ~.;',.i.~ ~.-~..- =. ~,,~i,..'-,,~ l
I, ~
I--
- . . . . _
0.5cm--I
° ~
II
e
"
-~
4
)
~ulminite
~
resinite
1--50/am~ I Fig. 1. Generaldistributionand microscopicappearance of the microlithotypesstudied.
LASER BEAM
¥
VIDEO CAMERA --
MICROSCOPE
I! v
TRANSFER
GCMS ~
LINE
' :
SAMPLE
=
<
CHAMBER
He
SAMPLE PELLET STAGE
Fig. 2. Schematic diagram of the laser-based microscope system used to thermally extract and pyrolyse microscopic regions on targeted coal microlithotypes.
chromatography, mass spectrometry (GCMS) as described previously (Stout & Hall 1991). Laser extraction/pyrolysis of each microlithotypes required that the laser beam be slightly defocused to a spot size of approximately 50/tm. The duration of irradiation for each microlithotype was 60 seconds during which time the sample was slowly moved so that a small region within each microlithotype was extracted/ pyrolysed. The target was observed during irradiation through the aid of a video camera/ monitor system. The total area irradiated per microlithotype averaged about 0.25cm 2. The laser power reaching the sample chamber was recorded with a hand-held power meter to be c. 2.8W. Sample surface temperatures were not directly measured but are expected (based on previous results; Stout & Hall 1991) to be of the order of 600°C. Selected ion monitoring (SIM) conditions were employed and 20 ions were monitored. Only those masses representing n-alkanes (m/z 57), n-alkenes (m/z 55), acyclic isoprenoids (m/z 113), sesquiterpenoids (m/z 183,198,109, 123), pentacyclic triterpenoids (m/z 191) are presented herein. Caution must be used when comparing the concentration of compounds between mass chromatograms. Thus, all the comparisons made herein are relative and should not be compared to full scan or GC data. Compound identi-
COAL MICROLITHOTYPE HETEROGENEITY fications were based on experience with full scan G C - M S data. For comparison, a conventional solvent extract of an aliquot of the blended coal was analysed as described by Curiale (1991). In addition, a solvent extracted aliquot of the blended coal was analysed using quantitative pyrolysis-gas chromatograph (Py-GC) at 600°C (for 20 s) as described by Stout (1991). Results and discussion
Distributions of n-aliphatics A comparison of intensity among the ions monitored revealed that the dominant products released from each microlithotype were n-aliphatics. Considering the overall liptinite-
TELITE
97
rich (28%) and hydrogen-rich (HI = 318 mg g-a) nature of this coal (Table 2) this is not surprising. Figure 3a-c compares the distribution of n-alkenes and n-alkanes, as determined from the m/z 55 plus 57 mass chromatograms, for the three microlithotypes studied. Due to the fact that the areas irradiated were only approximately the same for the three microlithotypes, the results are considered qualitative. Thus the histograms are normalized to the most abundant n-aliphatic and absolute quantitative comparisons between microlithotypes yields cannot be made. The distribution of n-aliphatics from conventional flash P y - G C (600°C) of the blended microlithotypes is shown for comparison (Fig. 3d). Keep in mind that the microlithotype distributions represent combinations of thermally extracted and pyrolysis products while the P y - G C distribution of the extracted
a
lOO
Ii
z~
b
LIPTOCLARITE
~
n-alkanes
50-
I 5
7
9
11
13
15
17
19
2t
23
25
27
29
31
33
J
35
o 1-5
7
I
,1~ 9
11
13
15
CARBON NUMBER
19
21
23
25
27
29
31
33
35
CARBON NUMBER
C
RESINOCLARITE
~
17
WHOLE COAL
'°°If
iIIIIi, i ,jtlllll t II1 olltJIIilI11111 ]111
so
1-5
7
9
11
13
15
17
19
21
23
CARBON NUMBER
25
27
29
31
33
35
1-5
7
9
11
13
15
17
19
21
23
25
27
29
31
33
38
CARBON NUMBER
Fig. 3. Relative distributions of n-aliphatics released from (a) telite, (b) liptoclarite and (c) resinoclarite microlithotypes during laser irradiation and (d) from the solvent extracted microlithotype mixture following conventional flash pyrolysis at 600°C. (a--c) based on m/z 55 + 57 mass chromatograms and (d) on FID response.
98
S . A . STOUT
blend represents pyrolysates only. Also, the microlithotype distributions are based on the abundance of fragment masses whereas the P y - G C distribution is based on an FID response. Thus, some caution is necessary when comparing the microlithotypes to the blended coal's pyrogram. The differences in the naliphatic distributions between the three microlithotypes are, however, real and indicate some of the chemical differences between them. Each microlithotype produced n-aliphatic distributions that are generally trimodal. The first maximum includes the n-Cl_s gases released and/or produced from each microlithotype. These were most (relative) abundant in the liptoclarite but testify to the overall gas-prone character of each. The second maximum occurs around n-C12_ls while a third maximum occurs at n-C27_29.
The telite and resinoclarite yielded mostly long-chain n-alkanes ( > n-C2s) which displayed a strong odd-carbon dominance. Collectively considering (1) the similar distribution and prominence of long-chain n-alkanes in the conventional solvent extract (n-alkenes are absent of course, Fig. 4); (2) the lack of a strong
odd-carbon dominance in the extracted coal's pyrolysate (Fig. 3d); and (3) the predominance of n-alkane over n-alkenes released from the telite and resinoclarite (Fig. 3a and c), suggests the telite and resinoclarite contain a significant amount of 'unbound' or extractable n-alkanes. This proposal is further supported by the similarity of the distribution of n-alkenes for all three microlithotype. That is to say that the pyrolysis of the different microlithotypes seems to have produced similar n-alkene distributions and that the differences among the n-aliphatic distributions lie primarily with n-alkane (some of which are produced on pyrolysis and some of which are present in the microlithotypes, especially the telite and resinoctarite, as unbound, or perhaps physically trapped, hydrocarbons or bitumen). Thus, it is clear that the 'unbound' long chain n-aliphatics (such as occur in the solvent extract; Fig. 4) are not evenly distributed in microlithotypes only a few millimetres apart. Long-chain n-hydrocarbons with an oddcarbon dominance are presumably generated from terrestrial plant waxes such as occur in plant cuticles (Nip et al. 1986; Tissot & Welte
100 90
-
n-alkanes
~
70
60
50
40
30
20
1-5
7
9
11
13
15
17
19
21
23
25
27
29
31
CARBON NUMBER
Fig. 4. Relative distribution of n-alkanes in the solvent extract of the microlithotype mixture.
33
35
COAL MICROLITHOTYPE HETEROGENEITY 1984). Long-chain n-hydrocarbons are also thought to be preferentially generated in coals relatively early in bituminization/catagenesis (Clayton et al. 1991). Surprisingly, the liptoclarite (which contained the most 'waxy' macerals; Fig. lb) yielded the least (relative) long-chain n-aliphatics and displayed no odd-carbon dominance. Their relative abundance in the liptinite-free (ligno-cellulose-derived) telite (Fig. la) implies, if one considers 'waxy' macerals (e.g. cutinite and cutinite-derived liptodetrinite) as the primary progenitors to these long-chain n-aliphatics, that these hydrocarbons are not indigenous, i.e. they must have migrated into the telite (perhaps, from the adjacent litoclarite; see below). The results shown in Fig. 3 indicate that the telite, in spite of its huminitic composition and ligno-cellulosic origin, is hydrogen rich (i.e. a perhydrous vitrinite, vitrinite B, or desmocollinite precursor). While the textinite and ulminite comprising the telite did not initially fluoresce when irradiated with uv light, they did exhibit a positive fluorescence alteration to reveal an orange-brown fluorescence after prolonged uv exposure. This fact tends to confirm the hydrogen-rich character of the telite (Teichmuller & Wolf 1977). Horsfield et al. (1988) and Thompson et al. (1985) also observed fluorescing vitrinites in other Tertiary coals from Indonesia. Such vitrinites may form during the original deposition due to incorporation of sub-microscopic liptinites (which invoke the fluorescence; Taylor 1966). While I cannot dismiss the occurrence of an indigenous fluorescing material being present, the results above indicate that hydrogen-rich vitrinites (desmocollinites) might also form during early bituminization/ catagenesis when they become impregnated with migrated fluorescing materials generated from nearby liptinites (rather than due to an original presence of distinct, sub-microscopic liptinites). Admittedly, it is unlikely that the fluorescence of the telite macerals is due to the n-hydrocarbons (which should not fluoresce due to a lack of conjugated double bonds) described above. Rather, it is likely to be due to small aromatic molecules that must also be present in the migrated fraction. In any case, fluorescence microscopy is of considerable importance in better recognizing the hydrogen richness and oil proneness of vitrinite macerals (Teichmiiller & Wolf 1977). The relative abundance of n-alkenes (pyrolysis products) over n-alkanes in the liptoclarite suggests that (unlike in the telite and resinoclarite) relatively few 'unbound' n-alkanes
99
were present therein. However, as mentioned above, since the 'waxy' macerals in the liptoclarite are considered precursors of long-chain n-alkanes in oils (Tissot & Welte 1984), their relative lack (compared to n-alkenes) might suggest either (1) the n-alkanes were not yet generated or, (2) as was suggested above, they were generated and have migrated from the liptoclarite microlithotype. The first explanation seems unlikely considering (1) the abundance of extractable organic matter contained in the other microlithotypes; (2) the maturity (early bituminization/catagenesis) of the sample (VR = 0.61; Table 2); and (3) the occurrence of exsudatinite in some voids within the coal (see below). The second explanation, thus, seems most likely. Considering the aforementioned argument for the presence of allochthonous n-hydrocarbons in the telite and resinoclarite, primary migration out of the liptoclarite into the surrounding microlithotypes is indicated. The distributions and relative amounts of long-chain n-hydrocarbons, as compared to the macerals they are thought to be generated from, also would suggest that primary migration has occurred in this coal. In having established the migration of n-hydrocarbon between microlithotypes, one could speculate that the intimate physical association of the macerals in the liptoclarite (Fig. lb), i.e. liptodetrinite dispersed among humodetrinite (or small bits of oil-prone macerals among small bits of non-oil-prone macerals), may promote the primary migration process. That is to say that the physical association between macerals may be as important to oil expulsion from coals as the type of macerals is to oil generation in coals. Thus, generation and expulsion may be most effective in coals with a high proportion of liptoclarite-like microlithotypes (i.e. abundant liptinite particles among finely divided vitrinite macerals and minerals). These results support the contention by Mukhopadhyay (1989) and Mukhopadhyay et al. (1991) that liptodetrinite/humodetrinite associations would (not only have a sufficient oil proneness but also) have a greater porosity and permeability, and therefore greater explusion capacity, than other more massive macerals. While this association between expulsion and porosity/permeability would intuitively seem correct, it is probably the microporosity of the macerals present (and not the macroporosity) which has the greatest influence on oil expulsion. The results described above also support the intuitive discussion by Bertrand (1989) concerning the importance of coal microtexture on oil expulsion.
100
s . A . STOUT
Any discussion on oil expulsion in coals must include some mention of cleats, cracks and fissures, which would seemingly aid the primary migration and expulsion process (Saxby & Shibaoka 1986; Hvoslef et al. 1988). In this particular coal there were cracks present in each microlithotype. However, the role of the cracks in the primary migration process was not obvious since exsudatinite was only rarely observed. However, the occurrence of any exsudatinite in the coal is testimony to the fact that oil generation and mobilization has occurred within the coal (Stach et al. 1982). This lends further support to the conclusion that primary migration of n-hydrocarbons has occurred between microlithotypes. Since exsudatinite is essentially a polymerized residual oil (solid bitumen), the low concentration of exsudatinite may merely reflect the fact that hydrocarbons which once moved through these cracks and fissures (prior to uplift) have been lost or absorbed into the coal network. In this particular coal, certain macerals, presumably the more massive huminites (i.e. textinites and ulminites, which are the precursor to most vitrinites) in the adjacent telite and resinoclarite microlithotypes (Fig. 1) appear to have trapped n-hydrocarbons which were apparently generated in the liptoclarite. Such 'trapping' would make sense in light of: (1) the tremendous microporosity and absorption capacity which vitrinite precursors normally exhibit at this rank (Thomas & Damberger 1976); and (2) the progressive increase in n-alkane yields from vitrinites between the high volatile b and low volatile bituminous ranks (Allan & Douglas 1977). These phenomena seem to be related since it is currently thought that porosity is reduced through bituminization/ catagenesis as hydrocarbons displace water and fill pores within the vitrinite network (Levine 1991). Thus, while rank and the amount of oil generated relative to the absorptive capacity/ microporosity of the coal are intimately related, oil expulsion from humic coals dominated by massive vitrite-type lithotypes (and only minor liptoclarite-type lithotypes) is probably retarded. This would further support the contention that absorptive sites within humic coals must first become 'saturated' or 'de-activated' before hydrocarbons can be expelled freely (Durand et al. 1987; Horsfield et al. 1988). This particular sub-bituminous, liptinite-rich humic coal, while having initiated generation and primary migration of hydrocarbons, has probably not yet been able to expel beyond the seam due to the high absorptive capacity of the massive huminite macerals (textinite, ulminite,
gelinite) which make up 68% of the coal (Table 2). The unusual abundance of n-Cll_a3 released from the resinoclarite microlithotype (Fig. lc) is interesting and currently unexplained. However, it may be due to a contribution to the rn/z 57 mass chromatogram from other co-eluting compounds (alkyl-naphthalenes?) in this region. Distributions o f acyclic isoprenoids
Figure 5 shows the relative distributions (m/z 113 mass chromatograms) of acyclic isoprenoids released during irradiation of the three microlithotypes. Each produced a series of C15, C16, C18, Ca9 (pristane = Pr) and C2o (phytane = Ph) acyclic isoprenoids (Fig. 5). The Pr/Ph and Pr/ n-C~7 ratios (as calculated from the m/z 113 mass chromatograms) for the three microlithotypes reveal differences in the relative distributions of isoprenoids over only a few millimetres distance. Since Pr/Ph and Pr/n-fa7 ratios are typically calculated from the FID response for solvent extracts, the ratios reported herein are only comparable internally. Nonetheless, the high Pr/Ph ratios distinguish the telite (13.4) and resinoclarite (13.2) from the liptoclarite (5.0). Indonesian non-marine oils usually have Pr/Ph ratios above 5.0 which would be in accordance to those ratios observed herein. The Pr/n-C17 ratios show that the telite (3.7) and resinoclarite (8.2) released more (relative) acyclic isoprenoids than n-alkanes while the opposite was true for the liptoclarite (0.5). Since isoprenoids are generated relatively early in catagenesis (Van Graas et al. 1981), this difference could further indicate that hydrocarbons generated in the liptinite-rich litoclarite have migrated into the surrounding microlithotypes. The insets to Fig. 5 show the m/z 57 mass chromatogram in the vicinity of n-C17_18.As can be seen, the isoprenoid pyrolysis products, prist1-ene and prist-2-ene, were also produced from each microlithotype in similar proportions (i.e. prist-l-ene>>>prist-2-ene). Prist-2-ene is thought to form from the mineral-catalyzed isomerisation of prist-l-ene (Regtop et al. 1986), therefore these compounds are thought to have the same precursor(s). The calculated pristane formation indices (PFI; after Goosens et al. 1988) indicate that the liptoclarite produced a greater proportion of pristene(s), as compared to pristane, than the other microlithotypes. This suggests that a greater proportion of pristane precursor(s) still exist bound to macerals in the liptoclarite than in the telite or resinoclarite. Assuming that all pristane derives from such a precursor(s), the greater relative concentration
101
COAL MICROLITHOTYPE HETEROGENEITY 100 Pr/Ph
Pr
13.4
Pr/nC17
a) TELITE
3.7
looI Lr PFI = 0.90
100
5.0
Pr/Ph
b) LIPTOCLARITE
Pr/nC17 0.5 100~
I-Z LU
Pr
0
IZ LU
18
16
•
1 IrPr-, I
15
PFi = 0 . 3 9
~.~ 100
13.2
Pr/Ph
Pr/nC17
Pr
8.2
C) R E S I N O C L A R I T E
1001
1Pr
16 PFI = 0 . 8 6
TIME
Fig. 5. M/z 113 partial mass chromatograms showing the relative distributions of acyclic isoprenoids released from (a) telite, (b) liptoclarite, and (c) resinoclarite during laser irradiation. Insets show distribution of pristenes determined from m/z 57 mass chromatograms for the same samples. PFI = pristane formation index (see text).
of pristane (over pristene) in the telite and resinoclarite could further indicate that pristane generated in the liptoclarite has also migrated into these surrounding microlithotypes. Thus, these results support those previously described for the n-hydrocarbons. Current views indicate that many isoprenoids, including the pyrolysis pristenes, form from tocopherols (Goosens et al. 1984). Since the difference in relative yields of pristenes cannot be due to differences in thermal maturity, the
fact that there seems to be a greater relative amount of pristenes released from the liptoclarite may indicate a relative abundance of these precursor compounds occurring in or among the liptodetrinite/humodetrinite macerals dominating this microlithotype (Fig. lb). The utility of various isoprenoid ratios and isoprenoid/n-aliphatic ratios to imply the degree of thermal maturation (Van Graas et al. 1981; Curry & Simpler 1988; Goosens et al. 1988)
102
S.A. STOUT 29
would seem to be in question considering the variability observed over a distance of only a few millimetres. Instead the differences observed among these microlithotypes could suggest that (1) the pristane precursor(s) is distributed unequally among the macerals and/or (2) the migration of isoprenoids (specifically, pristane) may accompany that of n-aliphatics (e.g. from the liptoclarite into the other microlithotypes, see above).
a) TELITE
27
3O
ttlz~
I II~°'°~ '
,,
~
'
Distributions of pentacyclic triterpenoids Figure 6 shows the distribution of pentacyclic triterpenoids (via m/z 191) for the microlithotypes studied. These traces represent a mixture of the 'unbound' (or physically-trapped) hopanoids along with hopanoids cleaved from maceral polymers during pyrolysis. Figure 7 shows the distribution (via m/z 191) of solvent extractable hopanoids for the blended microlithotypes. Each trace is dominated by the 17~(H), 21fl(H)30- norhopane (C29) and 17a(H)-22,29,30-trisnorhopane (C27; Figs 6 and 7). The relative distribution of these and the 17a(H),21fl(H) hopane (C3o) are shown in the ternary inset to Fig. 7. The inset shows a slight enrichment of hopane in the solvent extract but an overall similarity among the C27,29,30hopane distributions. Similarly, the distribution of the C3o+ hopanes and moretanes in the telite (Fig. 6a), liptoclarite (Fig. 6b), and the solvent extract (Fig. 7) are comparable. Relatively few hopanoids were released from the resinoclarite as indicated by the poor signal to noise ratio (Fig. 6c). There were six unidentified hopenes (one C27, two C29, two C3o, and one C31) recognized in the telite and liptoclarite traces that were absent in the solvent extract (Fig. 6a, b and Fig. 7). These hopenes are considered to be pyrolysates cleaved from macerals within the telite and liptoclarite. The relatively large C27 hopene has been recognized previously in conventional flash pyrolysates of kerogens (Seifert 1978; Larter 1978; Philp & Gilbert 1984; Van Graas 1986; Li & Johns 1990a) and its yield, relative to the 17a(H) trisnorhopane (17a(H)C27 hopane/ hopene ratio), has been reported to decrease with maturity (Philp & Gilbert 1984, 1985; Van Graas 1986). In this case, the ratios are comparable and when compared to previous studies, the low 17ot(H)C27 hopane/hopene ratios for the telite (1.57) and liptoclarite (1.75) testify to the relatively low maturity of the sample (Table 2). The slight variation among the telite and liptoclarite ratios is probably negligible.
. .-..i........i....,-w-.-i-.
r¸ ,-w ''''w'r''r'r '-- I ........W"' 29 I
" "'11 .¸ ¸1¸¸¸''¸¸¸¸=¸ --' "1 b) LIPTOCLARITE
27
3O r-----n 3 i11/ 1 1
29 100 -
'
r i
1 r 32 1
C) RESINOCLARITE '
27 30
0 ~I,IZ~TIII~T11.],, i rzz,zzzzlz. ,,,, z~T1r~rrrr~.` izzzllzzzzlzzz.l,,,,, ,. ~ .Tl,Z,~1],,,llzzzzl,zzT~1..z~zll,llll.l,,` j "1 l'llll''lz''r' 'Irl
TIME
Fig. 6. M/z 191 partial mass chromatograms showing the relative distributions of pentacyclic triterpenoids released from (a) telite, (b) liptoclarite, and (c) resinoclarite during laser irradiation. Numbers correspond to carbon numbers. Shaded peaks are unidentified hopenes.
Although it has been suggested that hopanoid precursors are likely to be bound to coal macerals via functional groups on the side chain (Li & Johns 1990b), additional work is needed
COAL MICROLITHOTYPE HETEROGENEITY KEY 100
27 '
A - TELITE
f 1T~
103
29
~
B - LIPTOCLARITE
30
C - RESINOCLARITE D - SOLVENT
!
EXTRACT
27
31
r~
s+
29
30
I
I
i
i
39:00
42:00
45:00
48:00
51 ':00
54100
i
i
57:00
1:00:00
TIME
Fig. 7. M/z 191 partial mass chromatogram of the solvent extract of the microlithotype mixture. Inset shows relative distributions of C27, C29, and C30 17a(H),21fl(H) hopanes.
to recognize the position of the double bond for the hopenes shown here. The fact that significant m / z 191 ions are produced suggests the double bond is somewhere on the right side of the molecule (rings C, D, or E). Furthermore, the prominent yield of C27 hopene (over other hopenes; Fig. 6a, b) would seem to suggest a complete loss of the side chain during cleavage from a maceral. Previous workers have used hopanoid ratios (e.g. C22 epimer ratios of the C31_33 hopanes, moretane/hopane ratios, and the presence/ absence of 17fl(H),21/~(H) hopanes) from the conventional pyrolysis of kerogens to imply thermal maturity (Seifert 1978; Seifert & Moldowan 1980; Philp & Gilbert 1984). The 22R/(22R + 22S) homohopane ratios for the telite, liptoclarite and solvent extract are 0.43, 0.42, and 0.40 respectively. This and other ratios testify to the equitable and early catagenesis rank of the coal. Furthermore, these molecular maturity parameters appear to be consistent over very small distances. However, some maturity parameters, for example Ts/Tm, vary considerably between the solvent extract (45.0) and the laser extract/ pyrolysate (5.1-telite and 7.7-1iptoclarite). This particular discrepancy is due to the near absence of 18a(H) trisnorhopane in the solvent extract as compared to the laser extracts/pyrolysates.
The cause for this marked difference is unknown. Since the hopanoids discussed above are derived from prokaryotic bacteria (Mycke et al. 1987) their clear presence in the telite and liptoclarite suggests that: (1) bacteria made direct contributions during the accumulation of the maceral precursors in each microlithotype and/ or (2) hopanoid hydrocarbons migrated into these microlithotypes. Discussions above have suggested that primary migration from the liptoclarite to the other microlithotypes has occurred. However, the relative lack of hopanoids in the resinoclarite (Fig. 6c) would seem to indicate that the former is more likely. After all, if migration of hopanoids accompanied the migration of n-aliphatic and isoprenoids one would expect them to be relatively enriched in the resinoclarite (and depleted in the liptoclarite). Why then are hopanoids depleted in the resinoclarite? One possibility is that bacterial activity was greater in the telite and liptoclarite precursor sediments than it was in the resinoclarite precursor. The fact that plant resins (such as formed the resinite in this microlithotype; Fig. lc) often display bactericidal properties (Sj6strom 1981) would tend to support this contention. This interesting observation requires further investigation. No oleanoids were observed in the mass
104
s . A . STOUT
chromatograms studied. This might be considered unusual in light of the generally high contribution of angiosperms in Tertiary basins. However, gymnosperm vegetation (e.g. Araucaria) may have dominated the coal facies studied.
Other biomarkers None of the microlithotypes yielded discernible quantities of steranes (as detected by m/z 217, 218, 231,253 mass chromatograms). However, steranes were present in small concentration in the solvent extract. Detection of the small concentration of steranes following laser irradiation may be due to the limited dynamic range of the mass spectrometer used in the experiments. However, a similar discrepancy was observed by other workers (Seifert 1978; Van Graas 1986). Van Graas (1986) suggested that sterane precursors may 'degrade' upon cleavage and produce fragments other than rn/z 217. That is, the lack of steranes in the laser extract/ pyrolysate may be due to the mechanism by which steranes are released from macerals and not by their absence. Alternatively, the absence of steranes in the pyrolysate could simply suggest that their precursors were not incorporated into the macerals irradiated. Other products of interest include the similar yield of a series of Cls diaromatic sesquiterpenoids from each microlithotype (e.g. rn/z 198 and 183; not shown). These compounds were dominated by cadalene which was a significant product of each microlithotype. Similarly, a series of resin-derived C~4_16 bicyclic sesquiterpenoids was produced from each microlithotype (m/z 109 and 123). Sesquiterpenoids are reported to be contained in angiosperm fossil resins (Grantham & Douglas 1980), thus their association with the resinite-bearing resinoclarite and liptoclarite microlithotypes (Fig. lb, c) is not unexpected. The presence of these compounds in the resinitebarren telite (Fig. la) could indicate a migration of these compounds and/or that the lignocellulosic tissues which formed the telite originally contained microscopicallyunrecognized resinous material.
Conclusions The complexity of hydrocarbon distributions within coals is only beginning to be understood. While the results from the novel laserbased extraction/pyrolysis technique described herein are only qualitative, the technique provides, for the first time, a geochemical finger-
print of the hydrocarbons generated from, or contained within, microscopically identified maceral associations in situ. This technique can be applied to conventional petroleum source rocks in the same manner and with refinement may be applied to single, selected coal macerals (and shale phytoclasts). This study has shown that the distribution of hydrocarbons in an autochthonous liptinite-rich humic coal varies on a very fine scale (millimetres). The variability between microlithotypes was not due to variable maturity but instead to heterogeneity in the inherent chemistry of the macerals comprising the microlithotypes and the presence/absence of migrated hydrocarbons. The results support the occurrence of primary migration of bitumen between microlithotypes. Specifically, the relative depletion of long-chain n-alkanes and isoprenoids in the liptinite-rich (mostly cutinitederived liptodetrinite in an attrinite matrix) liptoclarite and their relative abundance in the overlying telite (textinite/ulminite) and underlying resinoclarite (ulminite/resinite) suggests that normal and isoprenoid hydrocarbons generated in the liptoclarite have migrated into the adjacent microlithotypes. It is suggested that the intimate physical association of oil-prone, hydrogen-rich macerals (such as liptodetrinite) among fine-grained humodetrinite macerals (e.g. attrinite) will favour both oil generation and primary migration in coals. Thus, while the type of macerals present may dictate the potential to generate oil, the physical association between macerals (i.e. the microlithotypes present or microtexture of the coal) will play an important role in the potential for oil to migrate within, and ultimately expel from, coal. The absorptive capacity and microporosity (which generally decreases with rank) of some huminitic/vitrinitic macerals will also influence the efficiency of the primary migration and expulsion process in coals. Expulsion of oil from liptinite-rich (c. 10-30%) humic coals would seem to be favoured in liptodetrinite/humodetrinite-rich coals (i.e. coals with a high proportion of microlithotypes of a detrital origin) rather than in coals where the liptinites are encased within or bounded by more massive forms of huminites/vitrinites. Thus, in addition to conventional maceral analysis, any source rock assessment on coals could probably benefit from some form of microlithotype analysis. With the exception of several hopenes that are present in the laser extract/pyrolysate, the hopanoid biomarker distributions in each microlithotype are similar to that of the solvent extract from the whole coal. This probably indicates
COAL MICROLITHOTYPE H E T E R O G E N E I T Y
that microbial flora originally present during peat accumulation were generally similar. The relatively low abundance of bacterial hopanoids in the resinite-rich microlithotype could be due to the resin's antibiotic capacity. Using the novel laser-based thermal extraction/pyrolysis technique direct comparison between petrographic maturity parameters (e.g. VR) and molecular maturity parameters can be made on a single, very small sample. Future studies should improve our understanding of the following: (1) primary migration processes in all types of oil source rocks; (2) maturation versus source controls on hydrocarbon and non-hydrocarbon composition; (3) chemical variability a n d source potential a m o n g p e t r o g r a p h i c a l l y similar m a c e r a l s in
105
coals (e.g. different vitrinite types) a n d m o r e c o n v e n t i o n a l source rocks (e.g. amorphous kerogen). W i t h the c u r r e n t apparatus, c o m p o u n d s that occur as ' u n b o u n d ' or free b i t u m e n , physically t r a p p e d within the coal n e t w o r k , and released by t h e r m a l b o n d cleavage during pyrolysis are analysed collectively. F u t u r e configurations will a t t e m p t to regulate the release of these materials so that they m i g h t be analysed in series, i.e. t h e r m a l extraction followed by pyrolysis. The author would like to thank Dr Richard Armin (UNOCAL) for collecting and supplying the sample, Drs John R. Fox, Rui Lin, Joe Curiale and Jeffrey R. Levine for helpful discussions and comments, UNOCAL management for permission to publish, and Dr Andrew C. Scott for his editorial handling of the manuscript.
References ALLAN, J. & DOUGLAS, A. G. 1977. Variations in the content and distribution of n-alkanes in a series of Carboniferous vitrinites and sporinites of bituminous rank. Geochimica et Cosmochimica Acta, 41, 1223-1230. BERTRAND, P. R. 1989. Microfacies and petroleum properties of coals as revealed by a study of North Sea Jurassic coals. International Journal of Coal Geology, 13,575-595. CLAYTON, J. L., RICE, D. D. • MICHAEL,G. E. 1991. Oil-generating coals of the San Juan Basin, New Mexico and Colorado, USA. Organic Geochemistry, 17, 735-742. CURIALE,J. A. 1991. Molecular maturity parameters within a single oil family - A case study from the Sverdrup Basin, Artic Canada. In: MOLDOWAN, J. M., ALBRACHT, P. & PHILP, R. P. (eds) Biological Markers in Sediments and Petroleum. Prentice-Hall, London, 295-300. CURRY,D. J. & SIMPLER,T. K. 1988. Isoprenoid constituents in kerogens as a function of depositional environment and catagenesis. Organic Geochemistry, 13,995-1001. DURAND,B., Hvc, A. V. & OUDIN,J. L. 1987. Oil saturation and primary migration: Observations in shales and coals from the Kerbau wells, Mahakam delta, Indonesia. In: DOLIGEZ, B. (ed.) 2nd IFP Research Conference on Exploration: Migration of Hydrocarbons in Sedimentary Basins, Carcans Maubuisson, 173-195. & PARATTE, M. 1983. Oil potential of coals: A geochemical approach. In: BROOKS,J. (ed.) Petroleum Geochemistry and Exploration of Europe. Blackwell, Boston, 255-265. GOOSENS, H., DE LEEUW, J. W., SCHENCK,P. A. & BRASSELL, S. C. 1984. Tocopherols as likely precursors of pristane in ancient sediments. Nature, 312, 440-442. -
-
--,
DEU, A., DE LEEUW,J. W. VANDEGRAAF,B. & SCHENCK, P. A. 1988. The pristane formation
index, a new molecular maturity parameter. A simple method to assess maturity by pyrolysis/ evaporation-gas chromatography of unextracted samples. Geochimica et Cosmochimica Acta, 52, 1189-1193. GRANTHAM, P. J. & DOUGLAS,A. G. 1980. The nature and origin of sesquiterpenoids in some Tertiary fossil resins. Geochimica et Cosmochimica Acta, 44, 1801-1810. HORSFIELD,B., YORDY, K. L. & CRELLING,J. C. 1988. Determining the petroleum-generating potential of coal using organic geochemistry and organic petrology. Organic Geochemistry, 13, 121-129. Hvc, A. Y., DURAND, B., ROUCACHET, J., VANDERBROUKE,M. t~; PITTION,J. L. 1986. Comparison of three series of organic matter of continental origin. Organic Geochemistry, 10, 65-72. HUNT, J. M. 1991. Generation of gas and oil from coal and other terrestrial organic matter. Organic Geochemistry, 17,673-680. HVOSLEF, S., LARTER, S. R. & LEYTHAEUSER,D. 1988. Aspects of generation and migration of hydrocarbons from coal-bearing strata of the Hitra Fm., Haltenbanken area, offshore Norway. Organic Geochemistry, 13,525-536. INTERNATIONALCOMMITTEEOF COAL PETROLOGY1973. International Handbook of Coal Petrology. Centre National de la Recherche Scientifique, Pads. JONES, R. W. 1987. Organic facies. In: BROOKS, J. & WELTE, D. (eds) Advances in Petroleum Geochemistry. Vol. 2. Academic Press, New York, 5. LANDAIS, P. & MONTHIOUX, M. 1988. Closed system pyrolysis: An efficient technique for simulating natural coal maturation. Fuel Processing Technology, 20, 123-132. LARTER,S. R. 1978. A geochemical study of kerogen
106
s . A . STOUT
Geochimica et Cosmochimica Acta, 42,473-484.
and related materials. PhD thesis, University of Newcastle-upon-Tyne. LEVINE, J. R. 1991. The impact of oil formed during coalification on coal bed reservoir systems.
Proceedings of the 1991 Coalbed Methane Symposium. University of Alabama, Tuscaloosa, Alabama, 1-9. LI, M. & JOHNS, R. B. 1990a. Thermal desorption as an analytical technique in biomarker analysis of immature coals. Journal Analytical Applied Pyrolysis, 18, 41-58. - - & -1990b. Kerogen extract interrelationships of terpenoid biomarkers from a Jilin brown coal. Organic Geochemistry, 15, 109-121. MUKHOPADHYAY,P. K. 1989. Organic petrography and organic geochemistry of Texas Tertiary coals in relation to depositional environment and hydrocarbon generation. Bureau of Economic
--,
Geology, University of Texas, Austin Report of Investigation, 188. HATCHER,P. G. • CALDER,J. H. 1991. Hydro-
carbon generation from deltaic and intermontane fluviodeltaic coal and coaly shale from the Tertiary of Texas and Carboniferous of Nova Scotia. Organic Geochemistry, 17, 765-783. MYCKE, B., NARJES, F. & MICHAELIS, W. 1987. Bacteriohopanetetrol from chemical degradation of kerogen. Nature, 326, 179-181. NIP, M., TEGELAAR, E. W., BFINKUS, H., DE LEEUW, J. W., SCHENCK,P. A. & HOLLOWAY, P. J. 1986. Analysis of modern and fossil plant cuticles by Curie-point Py-GC and Py-GC,MS: Recognition of a new, highly aliphatic and resistant biopolymer. Organic Geochemistry, 10, 76%778. PHILP, R. P. & GILBERT, T. D. 1984. Characterization of petroleum source rocks and shales by Py-GC-MS multiple ion detection. Organic Geochemistry, 6,489-501. & - 1985. Source rock and asphaltene biomarker characterization by Py-GC-MS multiple ion detection. Geochimica et Cosmochimica Acta, 49, 1421-1432. REGTOP, R. A., CRISP, P. T., ELLIS, J. & FOOKES, C. J. R. 1986. 1-Pristene as a precursor for 2-pristene in pyrolysates of oil shale from Condor, Australia. Organic Geochemistry, 9,233-236. SAXBY, J. D. & SHIBAOKA,M. 1986. Coal and coal macerals as source rocks for oil and gas. Applied Geochemistry, 1, 25-36. SEIFERT,W. K. 1978. Steranes and terpanes in kerogen pyrolysis of correlation of oils and source rocks.
--
& MOLDOWAN,J. M. 1980. The effect of thermal stress on source rock quality as measured by hopane stereochemistry. In: DOUGLAS, A. G. & MAXWELL, J. R. (eds) Advances in Organic Geochemistry. Pergamon Press, Oxford, 591-596. SJOSTROM, E. 1981. Wood Chemistry, Fundamentals and Applications. Academic Press, New York. STACH, E., MACKOWSKY, M.-Th., TEICHMULLER,M., TAYLOR, G. H., CHANDRA,D. & TEICHMULLER, R. 1982. Stach's Textbook of Coal Petrology (3rd ed.) Gebruder Borntraeger, Berlin. STOUT, S. A. 1991. Principal components analysis of quantitative pyrolysis-gas chromatography and organic petrolgraphic data of kerogens. Journal of Analytical and Applied Pyrolysis, 18, 277-292. -& HALL, K. 1991. Laser pyrolysis-gas chromatography/mass spectrometry of two synthetic organic polymers. Journal of Analytical and Applied Pyrolysis, 21,195-205. TAYLOR, G. H. 1966. The electron microscopy of vitrinites. Coal Science, Advanced Chemistry Series, 5 5 . American Chemical Society Publishers, Washington, D.C., 274-283. TEICHMULLER, M. & WOLF, M. 1977. Application of fluorecence microscopy in coal petrology and oil exploration. Journal of Microscopy, 109, 49-73. THOMAS, J. Jr. & DAMBERGER, H. H. 1976. Internal surface area, moisture content and porosity of Illinois coals: Variations with coal rank. Illinois State Geological Survey, Circular, 493. THOMPSON, S., COOPER, B., MORLEY, R. J. & BARNARD, P. C. 1985. Oil-generating coals. In: THOMAS, B. M. et al. (eds) Exploration of the Norwegian Shelf. Graham & Trotman, London, 59-73. TISSOT, B. P. & WELTE, D. H. 1984. Geochemical fossils and their significance in petroleum exploration. In: TISSOT, B. P. & WELTE, D. H. (eds) Petroleum Formation and Occurrence (2nd edn). Springer-Verlag, Berlin, 93-130. VAN GRAAS, G. 1986. Biomarker distributions in asphaltenes and kerogens analyzed by flash Py-GC-MS. Organic Geochemistry, 10, 1127-1135. --, DE LEEUW, J. W., SCHENCK, P. A. & HAVERKAMP, J. 1981. Kerogen in Toarcian shales of the Paris Basin. A study of its maturation by flash pyrolysis techniques. Geochimica et Cosmochimica Acta, 45, 2465-2474.
C o a l - b e a r i n g s t r a t a as s o u r c e r o c k s - a g l o b a l o v e r v i e w DUNCAN
S. M A C G R E G O R
BP Exploration Operating Company, 4/5 Long Walk, Stockley Park, Uxbridge, UK Abstract: Compilation of a global source rock database has shown that coal-bearing sequences are significant oil generators only in very specific and relatively uncommon geological settings. Coals and associated carbonaceous shales are thought to be the primary oil-prone source facies in Australasia and an important secondary source facies in Southeast Asia. In other regions of the world, there is, however, no evidence that they have expelled major quantities of oil. In terms of their contribution to the world's petroleum reserves, coal measures are the origin of relatively minor proportions of oil, but of significant amounts of gas. Coal sequences that are believed to have expelled significant amounts of liquid hydrocarbons seem to be restricted to two palaeoclimatic and palaeobotanical 'fairways'. 1. Tertiary angiosperm assemblages within 20° of the palaeo-equator. 2. Late Jurassic-Eocene gymnosperm assemblages formed on the Australian and associated plates. The reasons for this are not well understood. Multidisciplinary research work is recommended to compare these oil-prone coals with the much greater volumes of apparently gas-prone coals deposited outside these regions.
This paper summarizes the results of an analysis of coal measure source rocks performed on a database of the world's major source rocks. This database was compiled jointly by BP and SimonRobertson, as part of a project aimed at identifying and classifying the world's major productive source rocks. Only source beds deposited in coastal plain, delta top, or marginal lacustrine environments and containing discrete beds of coal are considered here. Source rocks deposited in open-marine environments containing significant amounts of reworked terrigenous kerogen (e.g. Beaufort-Mackenzie) are not covered by this paper. The database was compiled from a variety of sources. Published data, usually the most recent available on any basin, were compared with internal BP analyses to assess the likely origin of hydrocarbons in each basin. In most cases, a consensus was found to exist on the identity of the source rock, though this is less frequently the case in many coal-bearing basins. Where controversy was encountered, the internal BP view is usually preferred (e.g. Haltenbanken - Cohen & D u n n 1987). The accuracy of this study, therefore, relies in turn on the accuracy of the interpretations in the references listed at the end of the paper. The aims of this contribution are to take a wider view of coal measures as source rocks than is possible in most published papers, to outline empirical trends, and to put forward hypotheses for testing by more detailed research. Study on this scale has obvious limitations. Little can be
contributed to some of the more detailed and controversial questions pertaining to coal source rocks, e.g. the debate as to the relative oil contributions from coals and interbedded shales. These questions are not, therefore, addressed in this paper. The compilation covers source rocks that have been tied to at least 50 million barrels of recoverable oil or to at least 3 trillion cubic feet of gas reserves. A number of small d o c u m e n t e d examples of small oil reserves tied to coals are excluded on this basis (e.g. the Surat-Bowen Basin in Australia). Basins that m e e t the criteria are subdivided into those where the overwhelmingly predominant hydrocarbon phase is gas (Table 1), and those in which significant quantities of oil also occur (Table 2). These listings form the basis for the following discussions.
Oils tied to coal-bearing sequences The term 'oil-prone' will be used for convenience for those provinces listed in Table 1 although it should be noted that in many cases, the predominant phase in volumetric terms is gas. It can be seen from the table that the only coal-bearing sequences tied to major oil reserves lie either in Australasia or Southeast Asia. The Kutei and Gippsland Basins are the largest, most fully d o c u m e n t e d and least disputed cases of coal measures-sourced oil provinces. Largely because of the significance of the Gippsland province, some 80% of Australasian
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coaland Coal-bearing Strata as Oil-prone Source Rocks? Geological Society Special Publication No. 77, pp. 107-116.
107
Location
Indonesia Indonesia Indonesia
India Malaysia
Indonesia
Indonesia Vietnam
Venezuela
Kutei NW Java S. Sumatra
Assam Luconia
E. Java
Tarakan Saigon
Maracaibo
Australia Australia China
China Bangladesh
Eromanga Cooper Junggar
Qaidam Surma
Possible (Secy) Possible
Probable Probable Possible (Secy)
Possible Probable
Possible
Probable Possible
Possible
Possible Probable
Probable Prob. (Mixed)* Prob. (Mixed)*
Probable
* Mixed sapropelic coal and lacustrine source.
Egypt New Zealand
W. Desert Taranaki
(b) Oil reserves 5 0 - 2 0 0 M M B O
Australia
Gippsland
(a) Oil reserves > 200 M M B O
Province
Status of interpreted source
Mid-Jurassic Late CretaceousEocene Late Jurassic Permian Early-MidJurassic Mid-Jurassic EoceneOligocene
Late CretaceousEocene Early Miocene Oiigocene OligoceneEarly Miocene Oligocene OligoceneEarly Miocene PalaeoceneEocene Oligocene OligoceneEarly Miocene Eocene
Source age
Table 1. Oil provinces tied to coal-bearing sequence source rocks
< 0.2? < 0.1
0.06 0.04 < 0.1
0.1 0.1
< 1
< 1 <1
<1
2 0.5
2.5 2 2
3
Oil reserves (bill. barr.)
Hoffmann et al. 1984 Atkinson etal. 1993 Robinson & Asril Kamal 1988 Saikia & Dutta 1980 Scherer & Ramlee Hitam 1992 Phillips etal. 1991
I0N 0 0
<1 1
0.8 ?
9
1
0
>10:1
c. 20:1
55N 60N
55S 60S 60N
10N 60S
Huang Difan et al. 1991 Grunau & Gruner 1978
Kantsleretal. 1984 Kantsleretal. 1984 Huang Difan et al. 1991
Bagge (this vol.) Pilaar & Wakefield 1984
5N
c. 10:1
Blaser & White 1984
5N 10N
<1 <1 <1
Samuel 1980
0
I0N 10N
Burns & E m m e t t 1992
References
45S
P/Lat.
<1
c. 30:1
>2:1
6 1.5 1 0.5 15
c. 1
Free gas to oil ratio
2.5
Gas reserves (bill. bart. eq.)
oo
Indonesia Thailand Pakistan Australia Australia Norway Europe Europe Europe CIS Russia
NE Natuna G. of Thailand Indus NW Shelf Surat-Bowen Haltenbanken N. N. Sea NW German S. N. Sea Central Asia Vilyuy
Probable Probable Possible Probable Probable Possible Possible Probable Probable Possible Possible
Status of interpreted source Early Miocene Miocene Palaeocene Late Triassic Permian Mid-Jurassic Mid-Jurassic Carboniferous Carboniferous Jurassic Permian
Source age Wet gas, some oil Wet gas, some oil Wet gas, condensate Wet gas, condensate Wet gas, some oil Wet gas/?condensate Dry gas Dry gas Dry gas Dry/wet gas Dry gas
Hydrocarbon phase 10N 15N 25S 25S 60S 50N 40N 15N 15N 50N 70N
P/Lat.
Robinson 1987 Achalabhuti 1976 Grunau & Gruner 1978 Woodside 1988 Philp & Gilbert 1986 Cohen & Dunn 1987 Goff 1984 Ziegler 1980 Ziegler 1980 Clarke 1988 Cherskiy 1986
References
All above listed provinces have > 3 TCF (500 MMBOE). Gas reserves tied to coal-bearing source biogenic gas provinces, e.g. northern West Siberia, San Juan, are not included.
Location
Province
Table 2. Non-associated gas provinces tied to coal-bearing sequence source rocks
110
D. S. MACGREGOR
oil is attributed to a coal measures source. Minor contributions come from the Cooper, Eromanga and Taranaki basins, all associated with larger amounts of gas. In many of the SE Asian basins listed on Table 1, there is a varying level of debate over the identify of the source. This reflects the possible presence of additional sources, usually a deeper lacustrine facies. This facies, often difficult to locate with the drill due to its depth and synclinal association, is now the undisputed primary source in SE Asia. Some early interpretations of a coal source have recently been revised in favour of lacustrine sources, e.g. Malacca Straits, Central Sumatra (Macgregor & Mackenzie 1986; Longley et al. 1990). Deep lacustrine sources have also been found recently in the South Sumatra (H. H. Williams, pers. comm.) and N W Java basins (Atkinson et al. 1993). Evaluation in these areas still continues and at the time of writing, it seems that the coal contribution to the oils of the basins, although diminished over that perceived some years ago, could still be significant. In SE Asia, the proportion of oil reserves tied to coal measures sources is now estimated, based on the most recent data, at between 10 and 30%. In practically all cases, much higher volumes of gas than oil have been expelled and reservoired. For instance, in the Kutei Basin, the distribution of discovered gas to oil (in oil equivalent) of hydrocarbons sourced from coalbearing strata is over 2 : 1, and is increasing as exploration progresses (Duval et al. 1992). The distribution for SE Asia coal-sourced basins as a whole is 3 : 1 and compares with a ratio of 1 : 3 for the primary hydrocarbon contribution from the highly oil-prone lacustrine sources. It seems that coal-bearing sequences, even when 'oil prone', generate more gas than oil in comparison to other source rock facies. Coal-bearing sequences are not believed to have contributed significantly to any major oil province outside Australasia or SE Asia, with the possible exception of a minor contribution in Venezuela (Blaser & White 1984). It is significant that coal measures sources have not been proposed for any of the world's 30 largest oil provinces. As a consequence, coal-bearing sequences are assessed to be of only minor statistical importance to other source rock facies on a global scale. A rough estimate, based on available reserves figures for the provinces listed in Table 1, is that less than 1% of the world's known oil reserves is sourced from coal-bearing sequences. Nearly all of this comes from Australasia and SE Asia, despite these regions containing o n l y f i % of the world's coal reserves
(Ippolito 1982). It can, therefore, be said that there is a poor correlation between the global distributions of major oil and coal reserves.
Non-associated gas tied to coal-bearing sequences Coals are frequently tied to gas reserves. In contrast to the situation for oil, coal-sourced gas seems to have a wide geographical distribution, and one that is more consistent with the global distribution of coals (Table 2). This includes huge amounts of gas in the former Soviet Union, an observation compatible with the 30% of world coal reserves in that region. Significant non-associated dry gas reserves also occur in Europe, while wet gas and condensates are concentrated in Australasia and SE Asia (e.g. Luconia/East Natuna). It should be noted here, that for reasons given in the paper by Cohen & D u n n (1987), all North Sea coals, including the Hitra Coals of Haltenbanken, are regarded as gas prone. Gas may, of course, be expelled from coals at various stages of coalification. Only coal sequences that have penetrated into or beyond the oil window and clearly not expelled oil, can realistically be said to have been established as 'gas prone'. For this reason, it is important to screen out 'biogenic' gas, generated at temperatures below those of the oil window, for this analysis. A high proportion of the gas reserves of the West Siberia fall into the former category (Yermakov & Skorobogatov 1984), as do other examples such as the San Juan Basin. In the statistics compiled on Table 2, biogenic reserves are, therefore, excluded. Coal-associated sources are seen to be significant contributors to the world's thermogenic gas reserves, with an estimated total contribution of around 15%. This figure increases to about 30% for the worlds total biogenic and thermogenic gas (Grunau 1983).
Mature coals not tied to hydrocarbon reserves There are of course many occurrences of coals around the world that are not tied to either oil or gas reserves. In many cases, this is undoubtedly due to geological rather than geochemical reasons. Nevertheless, there are a few cases worthy of discussion that may be of significance in understanding the behaviour of coals and associated shales as source rocks. The Sihapas coals of Central Sumatra lie within interpreted oil windows, and are inter-
A GLOBAL OVERVIEW bedded with producing reservoir sands. These vitrinitic coals were initially believed to have generated the oil reserves of the Malacca Strait region (Macgregor & Mackenzie 1986). Subsequent oil-to-source correlation has, however, indicated the source to be a deeper lacustrine shale, with no net contribution from the coals (Longley et al. 1990). Communication of the coals with traps cannot be invoked as an issue here, and another geochemical explanation must be sought. The level of maturity of these coals is analogous to those in adjoining basins, which are thought to be expelling oil. This indicates that chemical differences must occur between the Sihapas coals and the timeequivalents in adjoining basins, which controls the different behaviour of the two sets of coals under identical physical conditions. A similar observation can be made for Philippine coals. The Philippine archipelago is as rich in Tertiary coal deposits as is the Indonesian archipelago. In many Philippine basins, Eocene and Miocene coals should be well within the oil window (assumed at around 120-150°C), though perhaps have not reached the gas window (>150°C). Yet, in contrast to Indonesia, and despite many years of exploration effort, there have been no significant oil discoveries in the Philippines other than those clearly tied to marine source rocks. Regional variations in the oil potential of coals again seems to be the likely explanation. The significance of these two examples is that they demonstrate differing behaviour of coals over relatively small geographical distances. The controls on coals as source rocks may therefore be operating at a variety of scales.
Oil- and gas-prone coals in space and time It has been established that coal-bearing sequences believed to have expelled gas are widespread, whereas those that have expelled oil are not. In order to understand this further, the source rocks listed on Tables 1 and 2 have been plotted on palaeogeographic/palaeotectonic reconstructions. The plate reconstructions used were from BP's 'GLOBE' system. Palaeolatitudes thus obtained are shown on Tables 1 and 2. Figure 1 plots the palaeolatitudes of the identified coal source rocks against the known stratigraphic distribution of coals according to McCabe (1984). It can be seen that gas-expelling coals occur over a wide range of palaeolatitudes, ages and continents. The peak of coal development, maintained throughout most of geological history at around 50°N, is represented by a
111
number of occurrences of gas-expelling coals in the former Soviet Union and in the North Sea. However, the occurrences of coal-bearing sequences tied to oil reserves are far more limited. Two main associations are defined: (a) a Tertiary tropical association, deposited in palaeolatitudes less than 20°. The group is dominated by SE Asian examples. However, there have also been suggestions that Tertiary dispersed terrestial organic matter interbedded with coals may be generating some small proportion of oil in Venezuela (Blaser & White 1984), suggesting the association is not purely confined to Southeast Asia. Coals deposited elsewhere in the tropics in Tertiary times are often thermally immature, and it is not known whether these would produce oil under suitable conditions. A trend towards gas proneness with increasing palaeolatitude is identified. The limits of this association may be marked by the apparent absence of any known oilexpelling coals in the Philippines, in basins which lay at palaeolatitudes around 20° throughout the Tertiary. It should be noted that even in lower palaeolatitudes, there are also many examples of predominantly gasexpelling coals (e.g. Luconia). (b) a late Jurassic-Eocene Australasian association, deposited in high pataeolatitudes (40-60°S) on the Australian and New Zealand Plates, and dominated by the Gippsland Basin. The other members of this association are the more marginal and relatively gas-prone source rocks of the Taranaki and Eromanga basins. Many authors have related the oil-prone nature of the Gippsland and other Australasian coals to the gymnosperm vegetation that existed exclusively in Australasia following its separation from Gondwanaland in the late Jurassic. Floral differences could conceivably explain why such high latitude coals are oil prone in the southern hemisphere, but have no northern hemisphere counterparts. Outside these associations, there are only a few cases of documented oil-prone coals tied to small oil reserves. These comprise Permian coals in the Cooper Basin and mid-Jurassic coals in parts of China and Egypt. In all these pre-Cretaceous examples, the coals concerned were deposited on lacustrine margins and may contain high contents of algal kerogen. Alternatively, it may be interbedded lacustrine shales that are the source rock.
112
D. S. MACGREGOR w
¢D
9O
I
" ill
.i I,,,,I L; !' i
'°
,
• :,
g |
. . . i
N.America Greenland Eurasia
• Sundaland /SE Asia
:3
I
0
I
.
Illllll
| S.America
m 20
a.
Africa
i
60
[]
III
IlL ~
,...
-.
i
]"
India
| i
Australia New Zealand Antarctica
i
s 90
I
Common Coals (in at least one continent of the group) Coals Uncommon Oil-prone coals with >200 MMBO oil reserves tied (1 sq.=l occurrence) Marginally Oil-prone coals with 50-200MMBO oil reserves tied (I sq.=1 occurrence) Gas-prone coals with no sign. oil, but major gas reserves tied
(1 sq.=1 occurrence) Fig. 1. Distribution of coals and coaly source rocks by time and palaeolatitude. Note the concentration of oil-prone coals in Tertiary tropical settings. Distribution of coals modified after McCabe (1984).
Discussion Many authors have pointed out the physical and chemical variations that occur between coals of different regions and ages worldwide (e.g.
McCabe 1984). This is hardly surprising given the range of environments in which coals were deposited and the variation in their biological content in both space and time (Collinson &
A GLOBAL OVERVIEW Scott 1987). Variations in the relative oil- and gas-expelling ability of coal-bearing sequences are an expected consequence of such differences in coal composition. Figure 2 illustrates the relative volumes of different hydrocarbon phases tied to coaly source rocks according to geological age and association. It can be seen that Late CretaceousTertiary coal-bearing sequences are frequently tied to oil or mixed oil-gas reserves, whereas older coal-bearing sequences tend to be associated only with dry gas. The correlation between oil-prone coal sequences and two distinct palaeobotanical assemblages, namely tropical angiosperm assemblages and Australasian gymnosperm assemblages, is again emphasized. OIL
t
113
There clearly exists a wide continuous spectrum in the relative oil- and gas-expelling ability of coal sequences. This spectrum is illustrated on Fig. 2, oil-prone coals lying to the left and gas-prone coals to the right. The oil-prone end-member is best represented by cannel-like coals, such as are described in N W Java (Froderson 1992). The Gippsland Basin coals must also lie close to this oil-prone end-member. The source rocks of the Kutei Basin have expelled a higher proportion of gas and, therefore, lie in a more intermediate position. Gas-condensate generating coals are numerous in coals of all age groups whereas dry gas-prone coals become of increasing relative importance with increasing age. A dry gas-prone end member is perhaps
CONDENSATE
t
WET GAS
t
DRY GAS
t
50 Tertiary Tropical Coals
30 40 --1 20 10 0 40
"5
Luconia, Kutei
Cretaceous-Tertiary Gymnosperm Coals
2O I
Gippsland
40
Triassic-Jurassic Coals
Taranaki
klJ o
Central Asia
20 ,m 10 m ~D 0
['~":i:::'::i'.':-::::~:i:;:'~:::':::i::~:';:'~'::;:' :::::~ii-. ]:: :.::::~ Ha Ite n ban k ~:~...U..:.:I:::.?.:S,?:..Ii ..~...U.:~:I::-I.. 1 ~' * ~ P * P ~ P ~ ' ~ .::'..'-:(:.:.::.::-:'..':",.~";2":','::'-:/:-:'.:i F~,~,~**~,~,~T[:.:...:-:-.-:-::.,,%..~.~:.~:.;~:!.:~.e~z..z~I ~ ~ ~ ~ ~ ~ ~ ~. ~, ~, ~ W.:i.:-i-..-.'-v~.";".:';::-."-~.!~-.i..-.'?.l
W. Desert
411
Permo-Carboniferous Coals
European Westphalian
3O 20 1
Cooper
OIL
_~ r ~ ~ ~ ~ ~.~~ ,z,zz . I ~ i i i i i i i i i l CONDENSATE (expelled as wet gas)
WET GAS
DRY GAS
Fig. 2. Hydrocarbon reserves tied to coal-bearing source sequences by age. Volumes of different hydrocarbon phases are plotted in billion barrels oil equivalent. Oil-prone coals plot to the left; gas-prone coals to the right. Note the predominance of gas-prone coals in pre-Cretaceous times, and the appearance thereafter of significant oil-prone coals.
114
D. S. MACGREGOR
best represented by European Westphalian coals (Ziegler 1980), which seems to have expelled only dry gas at temperatures well above those accepted for the oil window. The trend represented on Fig. 2 cannot be attributed to maturity effects; large sections of European Carboniferous coals are, for instance, still within the oil window and show no evidence of having expelled anything other than dry gas. The relative ratios of oil and gas in the listed provinces in Tables 1 and 2 correlates well to the ranges of Hydrogen Indices and liptinite contents quoted for these coals. Very high liptinite contents are quoted for the NW Java coals (up to 45%) and for the Gippsland coals (average 15%, max. 25%, Shanmugan 1985). The Kutei basin coals contain 15-65% liptinite and have Hydrogen Indices in the range 250450 mg g - l C (Thompson 1985). The apparently gas-prone Sihapas coals of Central Sumatra have, by comparison, lower liptinite contents (below 10%) and Hydrogen Indices around 3 0 0 m g g - l C , (Macgregor & Mackenzie 1986). Pre-Tertiary gas-prone coals are usually characterized by negligible liptinite contents and Hydrogen Indices below 200. Coal-bearing source rock sequences seem to be more confined in space and geological time than any other oil source rock facies. Botanical controls seem to be the most likely explanation, although detailed palaeobotanical work is required to confirm this. Environment may be a second-order control, defining belts of oil proneness within any coal-bearing sequence. Lacustrine margin coals seem to be favoured over marine margin examples. On the basis of the analogues established in Fig. 2, the statistical chance of any coal-bearing sequence expelling oil as well as gas seem to be highest for Tertiary tropical assemblages, and lowest for high latitude pre-Cretaceous coals. Coals as a group are overwhelmingly gas prone: even in many of the most oil-prone cases, the volumes of gas expelled and reservoired far exceed those of oil. The ratio of gas to oil (in volume) is estimated at 12:1 for all coal sequences considered in this paper, falling to 3 : 1 for the Tertiary tropical assemblage. It is the opinion of the author that while there will undoubtedly be further changes in the interpretation of the relative contributions of different source facies in various provinces, the general trends illustrated here will be fairly robust. At the present poor level of understanding of coal and coal-associated source rocks, it is clearly difficult to confidently predict further occurrences of oil-prone coals. The most
important guidelines for such predictions are currently the strength of analogy to established provinces together with local data on liptinite contents and Hydrogen Indices. A much improved understanding is clearly required in order to develop reliable predictive models. In the author's opinion, this could most readily be achieved through a multidisciplinary, industrywide study of a large suite of samples spread throughout the oil/gas-prone spectrum.
Conclusions This paper has identified a number of trends and empirical rules on the phases likely to be expelled from coal-bearing strata according to the age and environment in which these were deposited. Full explanations have not been reached and will require more detailed analytical studies. (a) Oil-prone coals are relatively rare. Age and palaeoclimate, and their effects on the biological input to coaly kerogen, seem to be the prime controls on the spasmodic occurrences of oil-prone coal sequences. Further environmental controls, not identifiable by a study on this scale, are suggested to exist at a regional or sub-basinal scale. (b) The chances of any coal sequence being oil prone seem to be highest within Tertiary strata deposited in low latitudes. The risk of gas proneness increases with increasing palaeolatitude. The chances of a coal sequence expelling oil as well as gas are also improved for post-Jurassic Australasian coals and for lacustrine margin coals of all ages, although the latter tend to be of relatively minor quantitative significance. (c) There does seem to be at least a broad relationship between maceral content and proportion of oil expelled from coals. This may not hold true in all cases, and requires further study. (d) Coal-bearing sequences are relatively minor contributors to the world's oil reserves, but are major contributors in terms of gas reserves. Many colleagues within BP and Simon-Robertson contributed to compilation of the databases on which this paper is based. Particular thanks are due to Dave Moore and Peter Sharland in BP and Andrew Barnham in Simon-Robertson. All views expressed are, however, those of the author. I am indebted to the management of BP for permission to publish this paper.
A GLOBAL OVERVIEW
115
References ACHALABHUTI, C. 1976. Petroleum Geology of Thailand (Gulf of Thailand and Andaman Sea). In: HALBOUTY, M. T. (ed.) Energy Resources of the Pacific Region. American Association of Petroleum Geologists, Studies in Geology, 12, 155-180. ATKINSON, C. D., SINCLAIR, S. & GRESKO, M. 1993. Early Tertiary Rift Evolution and its Relationship to The Hydrocarbon System in the Ardjuna Basin, Offshore Northwest Java, Indonesia (abstract). In: LAMBIASE,J. (ed.) Hydrocarbon Habitat of Rift Basins. Geol. Soc. Conf., May 1993, abstracts. BLASER, R. & WHITE, C. 1984. Source Rock and Carbonization Study, Maracaibo Basin, Venezuela. In: DEMAISON, G. (ed.) Petroleum Geochemistry and Basin Evaluation. American Association of Petroleum Geologists, Memoir 35, 229-252. BURNS, B. J. & EMMETT, J. K. 1992. Gippsland Basin; Regional Biomarker Oil-Source Correlation and Maturation Study. American Association of Petroleum Geologists Bulletin, 76, 1093 (abstract only). CHERSKIY, N. V. 1986. History of Oil and Gas Generation and Accumulation on The Eastern Siberian Craton (in Russian). Nauka, Moscow. CLARKE, J. W. 1988. Petroleum Geology of the Amu-Dar'ya Gas-Oil province of Soviet Central Asia. US Geological Society, Open File Report 88-272. COHEN, M. J. & DUNN, E. A. 1987. The Hydrocarbon Habitat of the Haltenbanken-Traenabank Area, Offshore mid-Norway. In: BROOKS, J. & GLENNIE, K. W. (eds) Petroleum Geology of N.W. Europe. Graham & Trotman, London, 1091-1104. COLLINSON, M. E. t~ SCOTT, A. C. 1987. Implications of vegetational change through the geological record on models for coal-forming environments. In: SCOTT, A. C. (ed.) Coal and Coalbearing Strata: Recent Advances. Geological Society, London, Special Publication, 32, 67-85. DUVAL, B. C., CHOPPINDE JANVRY, G. & LOIRET, B. 1992. The Mahakam Delta Oil Province: An Ever-Changing Picture and a Bright Future. Offshore Technology Conference, Houston, May 1992. Paper OTC 6855, 393-404. FRODERSON, E. W. 1992. Offshore Northwest Java (ONWJ) 1000 Wells Later. American Association of Petroleum Geologists Bulletin, 76, 1101 (abstract only). GOFF, J. C. 1984. Hydrocarbon Generation and Migration from Jurassic Source Rocks in the East Shetland Basin and Viking Graben of the Northern North Sea. In: DEMAISON, G. (ed.) Petroleum Geochemistry and Basin Evaluation. American Association of Petroleum Geologists, Memoir 35,273-302. GRUNAU, H. R. 1983. Natural Gas in Major Basins Worldwide Attributed to Source Rock Type, Thermal History and Bacterial Origin. llth
World Petroleum Congress, 2, Paper P12 (2), 293-302. --& GRUNER, U. 1978. Source Rocks and the Origin of Natural Gas in the Far East. Journal of Petroleum Geology, 1, 3-56. HOFFMANN, C. F., MACKENZIE,A. S., LEWIS, C. A., MAXWELL, J. R., OUDIN, J. L., DURAND, B. & VAMDERBROUKE, M. 1984. A Biological Marker Study of Coals, Shales and Oils from the Mahakam Delta, Kalimantan, Indonesia. Chemical Geology, 42, 1-23. HUANG DIFAN, ZHANG DAJIANG, LI JINCHA, HUANG XIAOMING. 1991. Hydrocarbon Genesis of Jurassic Coal Measures in the Turpan Basin, China. Organic Geochemistry, 17, 827-838. IPPOL~TO, M. 1982. Ressources et Disponibilities en Charbon: Une Contribution Decisive a L'equilibre Energetique Mondial. Energie et Societie (1982), 57-92. KANTSLER, A. J., PRUDENCE, T. J. C., COOK, A. C. t~Z ZWIGULIS, M. 1984. Hydrocarbon Habitat of the Cooper/Eromanga Basin, Australia. In: DEMAISON, G. (ed.) Petroleum Geochemistry and Basin Evaluation. American Association of Petroleum Geologists, Memoir 35,373-390. LONGLEY, I. M., BARRACLOUGH,R., BRIDDEN, M. A. & BROWN, S. 1990. Pematang Lacustrine Source Rocks from the Malacca Strait P.S.C., Central Sumatra, Indonesia. Indonesian Petroleum Association, 1 9 t h Annual Conference, 1, 279-298. McCABE, P. J. 1984. Depositional Environments of Coal and Coal-Bearing Strata. In: RAHMANI, R. A. • FLORES, R. M. (eds) Sedimentology of Coal and Coal Bearing Strata. Special Publication of the International Association of Sedimentologists, 7, 13-42. MACGREGOR, D. S. & MCKENZIE, A. S. 1986. Quantification of Oil Generation and Migration in the Malacca Strait Region, Central Sumatra. Indonesian Petroleum Association, 15th Annual Conference, 1,305-320. PHILLIPS, Z. L., NOBLE, R. A. & SINARTIO,F. F. 1991. Origin of Hydrocarbons, Kangean Block Northern Platform, Offshore Northeast Java Sea. Indonesian Petroleum Association, 20th Annual Conference, 1,637-662. PHILP, R. P. & GILBERT, T. D. 1986. A Geochemical Investigation of Oils and Source Rocks from the Surat Basin. Journal of the Australian Petroleum Exploration Association, 26, 172-186. PILAAR, W. F. H. & WAKEFIELD, L. L. 1984. Hydrocarbon Generation in the Taranaki Basin, New Zealand. In: DEMAISON, G. (ed.) Petroleum Geochemistry and Basin Evaluation. American Association of Petroleum Geologists, Memoir 35,405-423. ROBINSON, K. M. 1987. An Overview of Source Rocks and Oils in Indonesia. Indonesian Petroleum Association, 18th Annual Conference, 1, 97-122. --& ASRIL KAMAL. 1988. Hydrocarbon Generation, Migration and Entrapment in the Kampar
116
D.S. MACGREGOR
Block, Central Sumatra. Indonesian Petroleum Association, 1 7 t h Annual Conference, 1, 211-256. SAmIA, M. M. & DurrA, T. K. 1980. Depositional Environment of Source Beds of High Wax Oils in Assam Basin, India. American Association of Petroleum Geologists Bulletin, 64,427-430. SAMUEL, L. 1980. Relation of Depth to Hydrocarbon Distribution in Bunyu Island, N.E. Kalimantan. Indonesian Petroleum Association, 9th Annual Conference, 417-432. SCHERER,M. & RAMLEEHITAM, 1992. Distribution and Maturation of Source Rocks in Brunei Darissalam. American Association of Petroleum Geologists Bulletin, 76, 1125 (abstract only). SHANMU~AN,G. 1985. Significance of Coniferous Rain Forest and Related Organic Matter in Generating Commercial Quantities of Oil, Gippsland Basin, Australia. American Association of Petroleum Geologists Bulletin, 69, 1241-1254.
THOMPSON,S. 1985. Oil-generating Coals. In: THOMAS, B. N. (ed.) Petroleum Geochemistry in the Exploration of the Norwegian Shelf. Norwegian Petroleum Society, 59-73. WOODSIDE OFFSHOREPETROLEUM 1988. A Review of The Petroleum Geology and Hydrocarbon Potential of the Barrow-Dampier Sub-Basin and Environs. In: PURCELL,R. R. • PURCELL, P. G. (eds) The North West Shelf Australia. Petroleum Exploration Society of Australia, 115-128. YERMAKOV, Y. I. & SKOROBOGATOV, V. A. 1984. Formation of Oil and Gas Pools in AlbianCenomanian Coal Bearing Sediment of West Siberia. English reprint in Petroleum Geology, 21, 12. ZIEGLER, P. A. 1980. Northwest European Basin: Geology and Hydrocarbon Provinces. In: MIALL, A. D. (ed.) Facts and Principles of World Petroleum Occurrence. Canadian Society of Petroleum Geologists, Memoir 6, 653-706.
Some examples and possible explanations for oil generation from coals and coaly sequences S. T H O M P S O N , 1 B. S. C O O P E R 2 & P. C. B A R N A R D 1
1Simon Petroleum Technology Limited, Llandudno, Gwynedd LL30 1SA, UK 2B. S. Cooper and Associates, Northern Cottage, Appleton le Moors, Yorkshire, UK Abstract: Coals and associated shales are important oil source rocks in certain deltaic environments, in which selective processes of degradation or reworking increase the content of hydrogen-rich kerogen. The formation of hydrogen-rich kerogen in a deltaic setting is either by preferential concentration of plant cuticles and spores after reworking of delta top freshwater peats, or, by accumulation of delta margin peats under saline conditions. In each case the source rocks and derived oils may be expected to show distinctive indicators of the depositional environment in the associated alkane content, sulphur content, biomarker and carbon isotope signatures. Coals in the Oligo-Miocene Talang Akar Formation of South Sumatra and Northwest Java Basins contain perhydrous vitrinite and are the source of isotopically relatively light waxy oils which contain biomarkers derived from tree resins. Waxy oils are also produced in the Sunda Basin, between Sumatra and Java but in this case the upper (coaly) part of the Talang Akar Formation is immature and the oils are derived from algal kerogen in the older, lacustrine Banuwati Formation shales; these oils are distinguished by low contents of land plant biomarkers and by a markedly heavier carbon isotopic signature. In the Western Desert of Egypt, coals of the mostly marine Jurassic Khatatba Formation have sourced oils with a relatively heavy isotopic signature. It is proposed that the isotopically heavy character of the coals and oils is due to a combination of microbial contribution, and raised salinity (and sulphate) levels due to marine influence. In northern England, Namurian (Pendleian) marine shales are generally distal turbidites derived from pro-delta sediments but mostly without oil source potential. Rich oil source rocks occur as basinal shales at the extreme margins of or interdigitating with the turbidites or as condensed sequences on structural highs. They contain kerogen derived by reworking of peats of a deltaic-coastal plain environment, and may be correlated with many of the oil shows and produced oils of the area. The examples demonstrate that the recognition of source rock facies aids exploration by guiding the prediction of the location and potential of oil-prone facies, but that this is only possible if all facets of oil and source rock geochemistry are integrated with a knowledge of sedimentary processes.
It is generally accepted that sediments must contain moderate to high concentrations of hydrogen-rich kerogens in order to have significant oil-source potential (Tissot & Welte 1978; Hunt 1979). The progenitors of hydrogen-rich kerogens are derived from vascular plants (waxes, cuticle, resin, suberinite) marine and lacustrine algae, photosynthetic bacteria and in-sediment bacteria. These materials may be preserved in sediment as recognizably intact entities, or as fragments in various stages of degradation verging towards amorphous and nondescript. To have oil potential there has to be a combination of sedimentary and environmental processes which have enhanced the production and preservation of these organic materials. Coals (formed as peats) are mineral-deficient deposits of plants accumulated in wetland environments including deltaic and estuarine
conditions, but their dominant c o m p o n e n t is vitrinite which usually has no oil potential. However, exceptionally there are coals, and associated mudstones, which are relatively hydrogen rich and significant oil sources, in which the main component is a perhydrous, low reflecting, faintly fluorescent vitrinite (e.g. Bertrand 1984; Durand & Paratte 1983; Ducazeaux et al. 1991; Thomas 1982; Thomas & Brown 1983). The oilsourcing properties of coals have been most recently reviewed by Murchison (1987) and Hunt (1991). Example of such coals are known to be present in sequences of all ages from Devonian to late Tertiary. The following examples, from Indonesia, Egypt and England, are presented and discussed in order to illustrate various diverse possibilities of deltaic environments providing oil-prone kerogens to local or distant source rocks. In the deltaic environment several processes
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coaland Coal-bearingStrata as Oil-prone Source Rocks? Geological Society Special Publication No. 77, pp. 119-137.
119
120
S. THOMPSON E T A L .
operate to produce concentrations of liptinitic material. It has been suggested (Thompson et al. 1985) that in everwet tropical deltas, back-delta peat mounds can achieve a significant height and thickness before being eroded during channel migration. During build-up of the peat mound, organic matter is derived from successive communities of mosses and herbaceous and arborescent vascular plants. Degradation of the plant matter is influenced by the continual replenishment of the in-mound water table by rainfall, so that much of it is aerobic with water-dispersed humic acids as one product, the organic matter passing into methanogenesis with little or no development of a sulphate-reducing zone. During erosion and transport of the peat components it is envisaged that there is physical separation of the humic acid-degraded lignin component from the cuticle and spore components, the humic components being deposited as organicrich muds in interdistributary areas, and liptinites which are lighter and hydrophobic being deposited as an organic concentrate at the freshwater-saltwater interface. In tidedominated deltas, this is a narrow zone so that the organic concentration becomes of significant dimensions. Since the early Tertiary the mangrove community has been a dominant feature at the seaward margins of tropical and semitropical deltas (Risk & Rhodes 1985; Brown 1989). This environment is organically highly productive and it may be assumed that equivalent habitats have been available at the margins of deltas since the Devonian. Accumulation of peat occurs below or reaching up to the water surface, and the water composition is saline or brackish. A transition to freshwater peat formation is likely up-delta, but depending on the water table in the freshwater peat, there may be more saline waters below the freshwater. Water flow through the mangrove is perceptible but not vigorous. The accumulating organic detritus is largely derived from the mangrove, but algal and cyanobacterial inputs may also be expected. The availability of sulphate ensures that a deep sulphate-reducing zone extends down from near surface, but may be expected to be underlain by a zone of methanogenesis. Freshwater peats eventually give 'normal' coals with little oil potential, but mangrove peats accumulating under a saline influence yield coals containing vitrinite with lower reflectance than the vitrinite of 'normal coals' of equivalent maturity and can have excellent oil potential. The difference appears to be due to the ability of bacteria in the sulphate-reducing zone to transform carbohydrate into lipid-rich biomass. The effect of
inundation of a freshwater peat swamp by saline waters has been illustrated for a Carboniferous coal (Veld & Fermont 1990) in which vitfinite changes from low reflecting (under saline influence) to normal values (under freshwater conditions) from top to base of a coal-bearing sequence. The two series of processes described above are not necessarily exclusive but could be contemporaneous with varying degrees of importance depending upon the environmental constraints. In each case oil-prone coals can be formed, but the peat accumulations are also susceptible to contemporary reworking and redistribution in some cases into the pro-delta and basin environment. The models of accumulation allow the prediction of chemical properties. Concentrates of liptinite, stripped of carbohydrate, will contain plant waxes with a high carbon preference index, biomarkers of the proximal delta floras, low contents of bacterial biomarkers, high mineral content but low sulphur content, and a carbon isotope signature which reflects the floral input only (alkanes de113C - 30 to - 33%o PDB). Mangrove-derived coals, in contrast will show features reflecting their large input of bacterial biomass with high contents of bacterial biomarkers, high sulphur content and a carbon isotope signature derived from floral carbohydrates up to 5%o heavier. The biomarkers in petroleum which have been inherited from higher plants have been extensively studied (e.g. Philp & Gilbert 1986; Powell et al. 1991; Curiale & Lin 1991). Biomarkers indicative of vascular plants are diterpanes derived from tree resins (Noble et al. 1986), bicadinanes derived from proximal deltaic flora (Cox et al. 1986; van Aarssen et al. 1992), bisnorlupanes (Brooks 1986), oleanane (Grantham et al. 1983) and other triterpenoids, C29 sterane dominance (Curiale & Lin 1991) and high abundance and marked carbon preference index in the C25 to C33 n-alkanes. The carbon isotope ratios for lipids of tropical and semitropical land plants other than grasses are in the range of - 3 0 to -33%o PDB. However, their carbohydrate contents are isotopically some 3-5 units heavier, which may be expected to be inherited, during degradation, by the lipids of sulphate-reducing bacteria.
The Oligo-Miocene Talang Akar Formation coals/shales of Indonesia Lowland peats are developed at present day in areas of everwet climate in Indonesia, and similar deposits associated with deltaic systems
•
MIDDLE
.
E~ ~ ~
Muara Enim ~
~_--~:-
..~:~_~-
' ; - - -
"~'" . . . . . . . . .
PreTertiaw Basement
Lahat (Lemat)
Talang A k a (
•
•
.... ;
,
-
I
~o Do
I
,
"-
--
_ ... g . . , . .
:', ....... _.
Pre-Tertiary Basement
Banuwati
/K
.
~+~++++i++~
~ ~//,~j///~,
~E~_~
L'-'-~=~'~--
---~ .
Talang Akar • ~ •~ ,o.._.u_. • t • °. • °
•
Batu . I Batu Rajal'L-~---~-~--~ Raja I ~ Shale I ~ - ' - - - ~ . - ~ Limestone~'~,~ " ~ 1 - ' ~ ' - - -~ /~[ I - I
A i r Benakat
Parigi L i m e s t o n e
Cisubuh
,:>n~ t~ ~
Unnamed
~.----
MTHOLOGY
ROCK UNIT
Note: Vertical scale d i a g r a m m a t i c o n l y
~/~T4-'~ ~ ++ -t~
-..-rr.qR-.T;, . . . . ---.:-.:.:....:.
--~
• ..
.•-. ., .- . .,.~. . .~. . .
Ju~ MJ t.t.toJ
~--....--
-"'"-iMP"
.aim II;=o-;';', o=Io-
Batu Raja C ='' I ~ ] Limestone • I !
Gumai
~
" -i i -" -i i -" - ii"
Kasai
__,
.- .- .-- .
Unnamed
........
UTHOLOGY
ROCK UNIT
SUNDA
.
~
.]~
.~
j~
.
Pre-Tertiaw Basement
Jatibarang Volcanic Rocks
(EocenePalaeocene)
.
•
.
Talang Akar •
Limestone
R~ja ~ . 1 ,
Batu
'Main
Pre-Parigi
.
-~-._.~,---
~'-_-~
~
~n~n,
UTHOLOG~
I,
'
I,
ii ii ii v ii v ii v ii " " " v, v ii v ii
II V "
II II II v v v
.~'---
- _ I - I i ~~
i--i-i~
I~' 1 , I , I I~-"~
;
Limestone e, ~/__.~_j..
Cisubuh
Unnamed
ROCK UNIT
NORTHWEST J A V A
~ " . . . . . . . . ~iiii~i
0
I
0
,
, I 250
500
~
'i I 500 Miles
1000 Kilometres
0
iiiiiiiiiii,,
Q
so.t, Chi.a ~ . .
LEGEND SedimentaryBasin
0%':iiiiiiiiijii,, s
~
LOCATION MAP
Fig. 1. Generalized stratigraphy and locations of South Sumatra, Sunda and Northwest Java Basins. See Fig. 3 for key.
PRETERTIARY
EOCENE
OLIGOCENE
~" ILl ¢J O
PLIOCENE
QUATERNARY
STRA~GRAPHY
SOUTH SUMATRA
~t'a
C
:7 Eb © > t-"
>
u~
© >
©
©
~=~
>
© tC~
122
S. THOMPSON E T A L .
have given rise to many of the coals in the Tertiary sequences of this region (e.g. Thompson et al. 1985). Lacustrine systems have also developed (Katz & Mertani 1989) in more seasonal environments. Both act as oil source rocks (Cole 1987; Robinson 1987), and this is illustrated by the sediments of the South Sumatra and Northwest Java Basins (coaly source rocks active) and the intervening Sunda Sub-basin (lacustrine source rocks active). The location and generalized stratigraphy are presented in Fig. 1. In all three basins, coals of the Talang Akar Formation are frequently rich in fluorescent vitrinite and/or particulate 'liptinite' which is often not specifically identifiable in terms of source. This kerogen exhibits a high Hydrogen Index and (more characteristically) also a very low (<20) Oxygen Index, which even when diluted by inertinitic kerogen, serves to distinguish this material from Type III (vitrinite) kerogen (cf. Horsfield et al. 1988; Teichmuller & Durand 1983). The oils of Indonesian Tertiary basins are generally waxy, as illustrated by the chromatogram in Fig. 2. Here, the oil also exhibits 'low maturity' features such as high pristane content and significant odd/even carbon number predominance, indicating generation by the source rock whilst still at a very modest level of thermal maturity (Ro ~<0.5%). Although the general hydrocarbon composition of the oils of the three basins are very similar, they differ in isotopic and biomarker characteristics. The oils of the Northwest Java and South Sumatra Basins are isotopically light (mainly > -28%0 PDB), and contain the higher plant biomarkers oleanane, resin-derived terpanes and bicadinanes (Grantham et al. 1983). In contrast, the oils of the Sunda Basin are isotopically heavy ( < - 24%o PDB) (Sutton 1979) and lack these biomarkers (Thompson et al. 1985). These geochemical features of the oils indicate generation from Talang Akar Formation coals (which are isotopically light and contain the terrestrial biomarkers under discussion) in the case of the Northwest Java (Roe & Polito 1979; Noble et al. 1991) and South Sumatra Basins, but non-generation by geochemically almost identical coals in the same formation in the Sunda Sub-basin. It was originally proposed (by Sutton 1979) that the isotopically heavy character of the oils was attributable to a kerogen component developed from marine grasses. However, more recent work has identified the lacustrine Banuwati Formation shales as the source of the Sunda Basin oils. This formation underlies the Talang Akar Formation and contains algal kerogen.
The reason for this difference in generative source rocks is that coal development in the Talang Akar Formation of the Sunda Sub-basin is confined to the top of the formation, which has never become fully oil mature; equivalent coals in the other two basins achieve far greater burial depth and oil maturity, and (in the South Sumatra Basin) even reach dry gas maturity (Robertson Research 1983). The lacustrine systems of Java and Sumatra are of Palaeocene-Eocene age and were developed in rift systems created by strike-slip tectonics; these tectonics had attenuated by the Oligocene and resulted in sag phase fluviodeltaic coal-bearing deposits accumulating. Throughout the Tertiary, this part of southeast Asia was subject to everwet or seasonal climates which favoured the development of coaly (domed peat) and lacustrine oil source rocks respectively. The sedimentary mechanisms proposed by Thompson et al. (1985) for concentration of 'liptinitic' kerogen in coals, i.e. winnowing or dessication (for allochthonous or autochthonous peats respectively) remain hypothetical; there is no information from studies of present-day environments which proves or disproves these postulated concentration processes. The examples cited above demonstrate, on the basis of geochemical and geological relationships, oil generation by Talang Akar Formation coals. They also provide an explanation for a wide range of carbon isotope enrichment values associated with Tertiary oils which contain terrestrial biomarkers; the mixing of an oil generated by a Sunda Sub-basin-type source (lacustrine/brackish algal) with an oil generated by Talang Akar Formation coal (higher plant), would given an oil with a relatively heavy carbon isotope ratio of about - 2 5 to -28%0 PDB, which would also contain terrestrial biomarkers. The possibility of mixing of higher plant organic matter with lacustrine organic matter is self evident, and can be observed in modern examples (e.g. Rieley et al. 1991). Generalized correlations of oil carbon isotope ratio data with marine and terrestrial organic matter (e.g. Sofer 1984) could be obscured by this process. However, certain Central Sumatra oils which have been almost wholly sourced by lacustrine source rocks are isotopically relatively light (A. Livsey pers. comm., 1992).
The Jurassic coals of Egypt Geological and geochemical relationships in the Western Desert (location and generalized stratigraphy - Fig. 3) show that the Jurassic
OIL
GENERATION
FROM
COALS
AND
COALY
123
SEQUENCES
m
0
9E3-U ,== ,¢
r._m
9E3-u
¢.)0 C
~,E3-u
0 r"
££~-u
-4
2£~u TE3-u 0£3-u
- - O
6~3-u 8e3-u Z~U
~A
-5 -5
g~o-u
g~O-u
O
-5 --5
£e3-u
E
.,,.~
~20--U 12S'U 02:]mu _....2
6;3-u
m aue~Xqd
8~3-u
.,,.q
: J
~ue~tJJ
F-, c
o
gl3-u
~;3-u
p~3-u ~3-t
£ £~3-u
I o
I~'~-'I
,
o
mlAW)
,
I o
~tSua~ul
,
e,i
S. THOMPSON E T A L .
124
Table 1. Distinguishing features o f South Sumatra, Sunda and Northwest Java basin oils Carbon isotope ratios %oPDB Oil location
Bicadinanes
Oleanane
Alkanes
South Sumatra Sunda Northwest Java
Present Absent Present
Present Absent Present
- 26 to - 30 - 20 to - 23 - 27 to - 30
Khatatba Formation is the source of deep Jurassic-reservoired oils and condensates and some Cretaceous-reservoired oils (Kholeif et al. 1986; Taher et al. 1988). The richest source rocks of this area are carbonates in the Upper Cretaceous Abu Roash Formation, but these only become mature in the southern part of the Western Desert (Awad 1984). Jurassic sediments in the area of Egypt covered by the eastern Mediterranean Basin, range from arenaceous in the nearshore marginal areas in the south, to predominantly carbonates in the north (May 1991). The section comprises deltaic-paralic-lagoonal sediments, with sands derived from the shield to the south (Moussa 1986). A coal-bearing facies is presently preserved in an area stretching from the Meleiha-Umbarka area in the west, to the Sinai and southern Israel in the east (Goldberg & Friedman 1974). Overall, the sediments of the Khatatba Formation are shallow marine, but show characteristics associated with proximity to an extensive coastal plain. It may be surmised that the thin coals in this formation represent reworked organic concentrations from a back-delta swamp. The predominant association of marine environments suggests that these coal swamps may have developed in brackish conditions. The common association of the coals with shales, and sometimes with overlying limestones reflects the marginal, and rather saline character of this environment. It is also probable that lagoonal conditions existed, which would have been suitable for algal development, although the palynological record shows only an abundance of pteridophyte spores. The geochemical profile of a deeply buried Khatatba Formation coal sample is presented in Fig. 4. Many of the geochemical characteristics of this sample (Table 2) are attributable to its relatively high level of thermal maturity although the oxygen indices are lower than expected for a vitrinitic coal. Similar but immature Jurassic coals from Gebel Maghara in
Aromatics -
26 to
-
-
19 to
-
23
-
26
-
28
to
29
Sinai show Hydrogen Indices as high as 450 and Oxygen Indices as low as 15 (Bagge et al. 1988). The reduction of the Hydrogen Index to about 200 during maturation gives at least 30% by weight conversion of coal to hydrocarbons. Hence, although the Khatatba Formation coals in the Western Desert are thin and discontinuous, they have had significant hydrocarbon source potential. There are several geochemical features which make this coal unusual, most particularly the relatively heavy alkane carbon isotope ratio values. Such values could be associated with temperate climate higher plants, but in the Jurassic this area was near equatorial. This apparent anomaly could be resolved by the biomarker data, which show several features not associated with higher plant kerogen. These features are a significant C27 sterane component (although C29 is dominant), a low hopane/ sterane ratio, and significant amounts of extended tricyclic terpanes, C24 tetracyclic terpane, and extended pentacyclic triterpanes. These features may be taken to indicate the presence of an algal kerogen component, in addition to the higher plant component revealed by the high (dominant) C29 sterane component, and the waxy character of the overall hydrocarbon mixture (Bagge et al. 1988). The presence of algal kerogen is not recognizable with fluorescence microscopy. However, it has been reported by Bagge et al. (1988) that Khatatba Formation coals in the Meleiha area of the Western Desert contain between 28% and 67% of fine-grained mineral matter mixed with amorphous fluorescent groundmass. In addition, these workers note that in similar but immature Gebel Maghara coals, the vitrinite matrix itself exhibits fluorescence and the coal is resin rich. The geochemical profile of a Cretaceousreservoired oil from the same area as the Khatatba Formation coal in Fig. 4 is presented in Fig. 5. The waxy character of this oil, and its relatively high C29 sterane component suggest
L
U
ALAM EL BUEIB
DOL.
ALAMEIN
I
I
1
I
I
="'T~
I
I
/
/
/
/
.
/
;, - , -
.
/
,
I
......
~
I
I
I
. . . . . . . .
.~
.;. - , '
/
. . . . . . . .
-,-
/
.
/K?
N o t e : Vertical scale d i a g r a m m a t i c o n l y
RAS QATTARA FM.
HIATUS
L
~ MASAJID FM.
HIATUS
.
KHATATBA FM.
~
=
I
I
I I
I
1
I I
i
I I
I
i
J i l l 1 1 1 1 I I I I
I I I I I I I I
-13
° t3
.13
OIL RESERVOIR/ SOURCE PRODUCTION ROCK
. . . . . . . . . . . . . . . . . . . . . . . . .
Dolomite . . . . . . . . . . . . . . . . . . . . . . . . . .
Limestone . . . . . . . . . . . . . . . . . . . . . . . . .
Coal/lignite . . . . . . . . . . . . . . . . . . . . . . . .
Conglomerate . . . . . . . . . . . . . . . . . . . . .
Sandstone - mediumcoarse grained . . . . . . . . . . . . . . . . . . . .
Sandstone - finemedium grained . . . . . . . . . . . . . . . . . . . .
Siltstone . . . . . . . . . . . . . . . . . . . . . . . . . .
Silty mudstone . . . . . . . . . . . . . . . . . . . . .
Claystone
Fig. 3. Generalized stratigraphy and location of Abu Gharadig Basin, Western Desert, Egypt.
PE R MOTRIASSIC
LrrHOLOG~
KHARITA • .":" . " . . " : ' . ' ~ ", MBR. • .......
~ mm M E M B E R
~
,~
~
G
D E F
B
BAHARIYA FM.
ABU ROASH FM.
KHOMAN FM.
APOLLONIA FM. HIATUS
DABAA FM.
M
U
BERRIASIAN
VALANGINIAN
BARREMIAN HAUTERIVIAN
JURASSIC
--
L:
o ~
•
ALBIAN
CENOMANIAN
TURONIAN
m F, APTIAN
,,,
I--
(J <:
'"
0
rt (nm
~"
SANTONIAN CONIACIAN
PALEOCENE MAASTRIC HT. CAMPANIAN
EOCENE
RUONCI K T-
~.M ~ MOGHRA FM., .~-. "o %:: ." .L;,
I L
OLIGOCENE
MIOCENE
STRATIGRAPHY
~J !, i
. . . . . . . . . . . . . . . . . . . . . . . .
Tested/produced gas
. . . . . . . . . . . . . . .
Tested/produced oil . . . . . . . . . . . . . . . . .
Good gas source rock . . . . . . . . . . . . . .
Fair oil source rock . . . . . . . . . . . . . . . . .
Good oil source rock . . . . . . . . . . . . . . .
Nonde position-erosion . . . . . . . . . . . . .
Calcareous
Basement (igneous) . . . . . . . . . . . . . . . .
Basement (metamorphic) . . . . . . . . . . .
/~
Outc., (iiiiiiiiiiiiiiiiii
Volcanics, tufts and lavas . . . . . . . . . . .
=e.
tJ1
126
S. THOMPSON E T A L.
e~ IXI a IX. IX. ~
.°
A~
q
-5
y~ J
8 O
9
,/
OIL GENERATION
O.. ~.. ,-
FROM
COALS AND
COALY
127
SEQUENCES
0
o.~o.~ 0
0
0
"-
,
~
N
'~
~
a
0
0
0
0
oo o o o
!
E
o
Khataba coal source rocks Jurassic/Cretaceous. reservoired oils
Location
.
10-50
Organic carbon (%)
.
.
160-350
Hydrogen Index
.
2-20
Oxygen Index c. 1.0
Vitrinite reflectance (% Ro)
1.8--4
0.8-2
Pristane/ phytane
°//oC28 20-30 20-25
°/oC27 25-35 25-30
Steranes
Table 2. Correlative features o f Khataba Formation coals and certain oils of the Western Desert, Egypt
45-55
40-50
%C29
-
-
24
to
25 to
-
-
Alkanes
27
26
Carbon isotope ratios (%o PDB)
-
-
22 to
23 to
-
-
25
25
Aromatics
to oo
OIL GENERATION FROM COALS AND COALY SEQUENCES a terrestrial input. However, the biomarker composition correlates with the Khatatba Formation coals, in that there is a significantly high Cz7 sterane component, a low hopane/ sterane ratio, and extended tricyclics and C24 tetracyclic terpanes are present. In addition, the carbon isotope ratio values are very similar. The geological and geochemical relationships discussed above suggest that a significant component of several oils in the Western Desert of Egypt is (Jurassic) coal derived. However, this coal is not derived solely from vascular plant material and may contain a significant nonmarine algal component. Such a component may be analogous to the kerogens which have generated the Sunda Sub-basin oils of Indonesia. This would provide an explanation for the isotopically heavy character of the reservoired oils and Khatatba Formation coals. Alternatively a cyclical model for microbial degradation of coals described for Westphalian sediments by Veld & Fermont (1990) may be analogous to the mode of formation of the Khatatba Formation coals. Here, a marine transgression is postulated to have inundated the coals and raised the sulphate levels of the subsea uncompacted sediments, thus enhancing degradation by sulphate-reducing bacteria. The result is the generation of microbial biomass derived from the relatively isotopically heavy carbohydrate component of the source higher plant organic material. The provenance of Jurassic oils in the Western Desert is not thought to be analogous to the isotopically relatively heavy oils of the BeaufortMackenzie Basin. Here, a correlation of sourcing by higher plant-derived kerogen with relatively heavy carbon isotope ratios is also noted in the oils, for which characteristic higher plant biomarkers such as bisnorlupanes and oleanane (Brooks 1986) are matched to carbon isotope ratios of - 2 7 to -28%0 PDB (alkanes) and - 26 to - 27%0 PDB (aromatics) (Burns et al. 1975; Curiale 1991). Here, a high palaeolatitude environment may be invoked in preference to a sapropelized algal component and/or salinity effects.
The Carboniferous oil source rocks of northern England The third example of a hydrocarbon system in which terrestrially derived organic matter is the prime contributor to oil reserves is substantially different from the previous two where coals are strongly implicated as the principal source rock. In this example the source rocks are organic-rich shales containing largely terrestrially derived
129
organic matter but deposited in a marine environment and extensively worked over by bacteria. These source rocks are of Carboniferous (Brigantian-Pendleian) age. They were deposited in the Northern England Carboniferous Basin, one of a series of basins in an east-west trend from Russia to Texas, formed as the proto-Tethyan Ocean closed between Pangea to the north and Gondwana to the south, during the Variscan orogeny (Fig. 6). Due to the inhomogeneity of the crust underlying northern England, the Northern England Basin consists of a series of sub-basins with intervening high blocks. In the early Carboniferous these were generally fully marine systems in equatorial latitudes and sediments consist of carbonate platforms and ramps on the highs and basinal calcareous shales in the lows. A major southerly directed river system drained the highlands and foreland to the north and east of the study area throughout the Carboniferous. In the late Brigantian/early Namurian, southerly progradation possibly assisted by substantial sealevel fall resulted in large-scale invasion of coarse- and fine-grained clastic sediments into and across the Northern England Basin (Fig. 7). These clastic sediments were initially deposited into and across the existing marine carbonatedominated system. This basin filling process prograded across the area from the northeast and east and over a 6 million year interval resulted in infilling the sub-basins with up to 2 km of coarse- and fine-grained sediments (Fig. 7). Source rocks within this sequence are found as the finest-grained distal turbidites in the oldest part of the early Namurian succession (Fig. 7). In each of the major Namurian depositional lows, source rock sequences are found (Edale Shales - Widmerpool Gulf; Sabden Shales Pennine Basin; Bowland Shales - ManxCraven Basin) with a gross thickness of up to 100m (Robertson Group 1987; Fraser et al. 1990; Leeder & Hardman 1990). Analysis of the organic content of these shales has shown that there is considerable variation in the quality and richness of these sediments as source rocks. Detailed studies of the Bowland Shale Formation sequence in and around the Waddington Fell area of the Craven Basin has shown that the best quality source rock was deposited not on the basin floor but on syndepositional highs in the basin and was often associated with thickness attentuation (Robertson Group 1987). Waddington Fell was one such high as can be demonstrated by the absence of the areally extensive turbidite flow Pendleside Sandstone which bypassed the high. The
I/
/I
'°°° °
~
0 ,D, .,>1~
i% "..,-%
oO
,..%
.'.'.'.'.J~
:!iiii!ii
°.°o° . . . . . . . . .
"." . ' . " . ' . ' . ' . ' . ' . "
:':'.'.'.'.'.'.'.'.'.'.'.'.'.
"o','o°o°o°o°o
/\
I%
.>1 /
.~ \ 1
/
"-iI~
~',
~
l 1%
"',,..\1 t.~1 1% I I X
<,,,i~.~< ~l; ~ >P,
~\1 I
~,
.--,~. ~.
-'~ '
~'-,.\1 "~-~1 I 1% I I ,~ I I
'~
.
.
.
.
-,,\1 /,~.~1 "~',~\1 X I 1% I I % l
.,~ ,% /'% ~ '% /% ,-- %. I% "~ % 1% -~ % I% ".,,..% i % ~, ,~ I% '~ % /% '~.
l , , o • .oO.OoOoOo,
Fig. 6. Namurian palaeogeography in northwest Europe.
"%~11 /
. .'.'~\
I
:. :.:. :. :. :--.'.:.'.'.X/,,,,,
. . . .
¢
SCALE
=======================================================
o . o o .....
:.:.:.. ..:.:.:.:.:.:. .:.:.:-:.' ..,,.F~.' , ~ ~!:.:::. " "",'.::::::::::
I
/
~,, .'.~,~."
% i%
M i n o r s e d i m e n t s o u r c e s ....................
Major river system/ s e d i m e n t s o u r c e ................................. ~
Siliciclastic m u d s t o n e w i t h s a n d a n d silt d e v e l o p m e n t s ....................... [ . . . _ . _ !
L a n d a n d p o s i t i v e a r e a s ....................
Coals d e v e l o p e d in m u d d y clastic s e q u e n c e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
LEGEND
•
• J
/~ ~1 I /
%.
i N -~-
'~,; ,.. ~
•
'--~----
,~'- - - ~ !
z 0 ::> uJ
<~
"
Z
--<:
z
-~
T
~
>~
,
ZZ ~<
I
!
I
,
I
I
I
, , , ,
;~..
~.
~
"~
p
z..i
/
I
I
I I
I
, , iI, ,
I
-
r"
I
I
- ~. o ° , ° . , ."o-. °- °
z
~,'--r.'-.l • ."r-:-: _
~:
.
/j[
.
~_<
.
• •
~z "':'~---
.
-
/...-:;;.~;.
.
--. ............
BLOCK
,~..~ . . . .
~ 5
BASIN
BASIN
'<" i~ ";" '-~7'~ -'~ --
Z
z
<~
Z <:
~:
PER-
STRATIGRAPHY
ALPORTIAN CHOKIERIAN ARNSBERGIAN
6 7 8
MARSDENIAN KIN DERSCOUTIAN
4 5
LANGSE'I-rlAN YEAOONIAN
2
DUCKMANTIAN
3
1
Cyclical shale/ limestone/coal facies ..........................
Distal turbidite shales (potential source) .......................
Sandy turbidites ........
Reef facies .................
I : ~
J .......... 1
LEGEND
....................
Basement high .........
Main delta system f e e d i n g Yoredale sands ........................
Brigantian absent (Eroded)
Condensed sequence ..................
SCHEMATIC PENDLEIAN PALAEOGEOGRAPHY
Fig. 7. Generalized stratigraphy of the Northern England Carboniferous Basin and schematic Pendleian palaeogeography.
- 370
- 350
-340
-~o
AGE (my)
~
~
~'~
O
© > r"
> Z
q © >
7~ ©
©
7~ >
C~ r~
©
~4
3
%
52
[I
1 %
1
MID
BRIGANTIAN
61
51
REGRESSION
Organic carbon
Fig. 8. Example of Namurian source bed development on a sediment-starved high.
6
S
I
Reduced sulphur I Reactive iron
%
O0 .I
Hydrogen index
Calcareous sediment dominant.
High sea level stand but mixing of waters by frequent low-seeking turbidites.
Oxygenated water column. Low salinity.
Siliciclastic sediment dominant.
Sea level high stand partially euxinic water column. First input from distant Namurian delta of good quality organic material.
bO
~
-
i •.o... o..ooo
!-. ....-.:.:.-
&:i:i:i:i:i:i:!'" o.°oo°....
f. . . . . . .
~::::':;~i" ;~,~':.-':'::'!:i:i:i:i:i::::'" :::::::::::::::::::::::::::
\
i------
Lignin
Protein
Carbohydrate /
Huminite = 80%
(tot) = 2 0 % C. org .'-'-'.'-'-'.'~'-........... Pre,ereo, a o r e . e v . , o o o , p , .
~
Fig. 9. Proposed mechanisms for enrichment of hydrogen-rich kerogen in Namurian sediments.
Concentratio n
/
/i\\
.....
^ncentrati~
FINESEDIMENTFLOW (freshwater plume)
-t---~
.( TURBULENT k~ CLOUD
o,-
02 +
Sea level
Vegetation on delta top A (tot) = 1 0 0 % C org,
CONTRIBUTION (Lipid rich membranes)
(Proteins,carbohydrates)
DESTRUCTION
Microbes
"J
~
ANAEROBIC
• /~
~~'~'o o, - :::::::::::::::::::::
£3
C
© >
>
© >
;0 ©
©
>
7:
Z
C~
o
Organic carbon (%)
2-60 .
Location
Namurian source rocks Oils
.
300-400 .
Hydrogen Index
.
10-40
Oxygen Index c. 0.7
Vitrinite reflectance (% Ro) -> 1 -> 1
Pristane/ phytane
°/oC28 10--40 10-30
°//oC27 5-40 15-40
Steranes
Table 3. Correlative features of various Namurian source rocks and Northern England Basin oils
50--80 40-70
%C29
-
3 4
28 to -28to -32 -
Alkanes
Carbon isotope ratios (%0 PDB)
-
-
28 to 32 -28to -31
Aromatics
4~
OIL GENERATION FROM COALS AND COALY SEQUENCES Bowland Shales on the flanks of this high have good oil source quality (Fig. 8). It is to be noted that this source rock sequence has clearly been deposited in a marine environment rich in sulphate as evidenced by the iron sulphide present as framboidal pyrite produced by the extensive bacterial degradation of organic matter. The proposed model for transport and enrichment of the land-derived material into the distal marine environment (Fig. 9) involves reworking of plant material from the delta top mire into the marine environment, physical sorting and then flocculation or precipitation into deeper water prior to bacterial degradation and lipid enrichment in the basinal environment. Biomarker and isotope studies have confirmed the correlation between these source rocks and their land plant progenitors (Table 3) and the oils found often in Westphalian age reservoirs in the East Midlands and Lancashire Basins.
135
Comparison and discussion For the purposes of comparison, the carbon isotope ratio data for the oils which have been generated by the source rocks discussed in this paper, are presented in Fig. 10. Set against these data fields are printed the character and lithology of the generative source rocks. From the point of view of source type interpretation, these data appear to show a simpler distribution of carbon isotope ratios than that suggested by Sorer (1984). In addition, for these data and their interpretations the distribution appears to be age independent between Upper Carboniferous and Tertiary. However, different mechanisms for creating isotopically relatively heavy or light kerogens and oils variously involve mixing of kerogen end members, and environmental conditions. The source rock examples discussed in this paper demonstrate that particular types of coal and coaly sequences contain high proportions of
/
-34 -
-32
-30 ¸
~
C
(WAXY/RESINOUS)
-28,
O3 LIJ Z ,¢
/ I J/f/
"/" ~
HUMIC(RESINOUS)COALS
'~ -26/
-24
/
-
/
" 1 //' / --i'L
f i
-20
i,
•
/ ./"
IC
/
+ COALS ?NON-MAI~NE ALGAL}
~ /
NONM NON-MARINE ALGALSHALES w
-22
I
,
-24
=
/ -26
AROMATICS
I
-28
(%. PDB)
Fig. 10. Comparative carbon isotope ratio plot of various oils.
t
-30
t
-32t
t
136
S. THOMPSON ET AL.
h y d r o g e n - r i c h k e r o g e n s , and can g e n e r a t e oil. T h e recognition and prediction of the relevant k e r o g e n facies within coaly s e q u e n c e s can aid the p r e d i c t i o n of source rock o c c u r r e n c e . G i v e n , in addition, the different m e c h a n i s m s p r o p o s e d (but not p r o v e d ) for creating the h y d r o g e n - r i c h coals r e q u i r e d for oil sourcing,
it is a p p a r e n t that t h e r e is m u c h m o r e w o r k r e q u i r e d in the field of b i o g e o c h e m i s t r y b e f o r e the f o r m a t i o n and g e o c h e m i c a l c h a r a c t e r of oilg e n e r a t i n g coals is p r o p e r l y u n d e r s t o o d . The permission of the directors of Simon Petroleum Technology (Exploration Services) to publish this work is gratefully acknowledged.
R e f e r e n c e s
VANAARSSEN,B., HESSELS,J. K. C., ABBINK,O. A. & DE LEEUW, J. W. 1992. The occurrence of polycyclic sesqui, tri-, and oligoterpenoides derived from a resinous polymeric cadinene in crude oils from southeast Asia. Geochimica et Cosmochimica Acta, 56, 1231-1246. AWAD, G. M. 1984. Habitat of oil in Abu Gharadig and Faiyum Basins, Western Desert, Egypt. American Association of Petroleum Geologists Bulletin, 68, 564-573. BAGGE, M. HARDING, R., EL AZHARY, T. & SAID,M. 1988. Generation of oil from coal sequences in the Western Desert, Egypt. Proceedings of the Egyptian General Petroleum Corporation. 9th Exploration and Production Seminar, Cairo, November 1988. Egyptian General Petroleum Corporation, Cairo. BERTRAND, P. 1984. Geochemical and petrographic characterisation of humic coals as possible oil source rocks. Organic Geochemistry, 6, 481-488. BROOKS, P. W. 1986. Unusual biological marker geochemistry of oils and possible source rocks, offshore Beaufort-Mackenzie Delta, Canada. Organic Geochemistry, 10,401-406. BROWN, S. 1989. The 'mangrove model', can it be applied to hydrocarbon exploration in Indonesia? Proceedings of the 18th Annual Convention of the Indonesian Petroleum Association. Indonesian Petroleum Association, p. 385-401. BURNS, B. J., HOGARTH, J. T. C. & MILNER, C. W. D. 1975. Properties of Beaufort Basin Liquid Hydrocarbons. Bulletin of Canadian Petroleum Geology, 23,295-302. COLE, J. M. 1987. Some fresh/brackish water depositional environments in the SE Asian Tertiary with emphasis on coal bearing and lacustrine deposits and their source rock potential. Proceedings of the 16th Annual Convention of the Indonesian Petroleum Association, p. 429-449. Cox, H. C., DE LEEUW, J. W., SCHENCK,P. A., VAN KONONGSVELD, H., JANSEN, J. C., VAN DE GRAAF, B., VAN GEERESTEIN, V. J., KANTERS, J. A., KRUK, C. & JANS, A. W. S. 1986. Bicadinane, a C3o pentacyclic isoprenoid hydrocarbon found in crude oil. Nature, 319,316-318. CURIALE,J. A. 1991. The petroleum geochemistry of Canadian Beaufort Tertiary 'non-marine' oils. Chemical Geology, 93, 21-45. - & LIN, R. 1991. Tertiary deltaic and lacustrine organic facies: comparison of biomarker and
kerogen distributions. Organic Geochemistry, 17,785-803. DUCAZEAUX, J., LE TRAN, K. & NICOLAS, K. 1991. Brent Coal Typing by combined optical and geochemical studies. Bulletin Centre Recherche de Pau- Societe National de Petroliere Aquitaine. 15,369-381. DURAND, B. & PARArrE, M. 1983. Oil potential of coals: a geochemical approach. In: BROOKS, J. (ed.). Petroleum Geochemistry and Exploration of Europe, Blackwell, Oxford, 255-265. FRASER, A. J., NASH, D. R., STEELE, R. P. & EBDON, C. C. 1990. A regional of the intra-Carboniferous play of Northern England. In: BROOKS, J. (ed.) Classic Petroleum Provinces. Geological Society, London, Special Publication, 5, 417-440. GOLDBERG, M. & FRIEDMAN, G. M. 1974. Paleoenvironments and paleogeographic evolution of the Jurassic system in southern Israel. Geological Survey of lsrael Bulletin, 61. GRANTHAM, P. J., POSTHUMA, J. & BAAK, A. 1983. Triterpanes in a number of Far Eastern crude oils. In: BJOROY, M. et al. (eds.) Advances in Organic Geochemistry 1981. Wiley, Chichester, p. 675-683. HORSFIELD,B., YORDV, K. L. & CRELHNG, J. C. 1988. Determining the petroleum-generating potential of coal using organic geochemistry and organic petrology. Organic Geochemistry, 13, 121-129. HUNT, J. M. 1979. Petroleum geochemistry and geology. W. H. Freeman and Company, San Fransisco. - 1991. Generation of gas and oil from coal and other terrestrial organic matter. Organic Geochemistry, 17,673-680. KATZ, B. J. & MERTANI, B. 1989. Central S u m a t r a a geochemical paradox. Proceedings of the 18th Annual Convention of the Indonesian Petroleum Association. Indonesian Petroleum Association, p. 403-425. KHOLEIF,W., WORK, J. G. & SANAD,S. 1986. Meleiha: its history and its significance. Proceedings of the Egyptian General Petroleum Corporation, 8th Exploration Conference, Cairo, 1986. Egyptian General Petroleum Corporation, Cairo. LEEDER, M. R. (~zHARDMAN,M. 1990. Carboniferous geology of the Southern North Sea Basin and controls on hydrocarbon prospectivity. In: HARDMAN, R. F. P. & BROOKS, J. (eds) Tectonic Events Responsible for Britain's Oil and Gas Reserves. Geological Society, London, Special Publication, 55, 87-105.
OIL G E N E R A T I O N F R O M COALS AND C O A L Y S E Q U E N C E S
137
MAY, R. M. 1991. The Eastern Mediterranean Mesozoic Basin: Evolution and oil habitat.
ROBINSON, K. M. 1987. An overview of source rocks and oils in Indonesia. Proceedings ofl6th Annual
American Association of Petroleum Geologists Bulletin, 75, 1215-1232.
Convention of the Indonesian Petroleum Association. Indonesian Petroleum Association,
MoussA, S. 1986. Evolution of the sedimentary basins of the north Western Desert, Egypt. Proceedings
p. 97-122. ROE, G. D. & POLITO, L. J. 1979. Source rocks for oils in the Ardjuna sub-basin of the Northwest Java Basin, Indonesia. UN/ESCAP, CCOP Technical Publication, 6, 180-194, Bangkok. SOFER, Z. 1984. Stable carbon isotope compositions of crude oils: application to source deposition environments and petroleum alteration.
of the Egyptian General Petroleum Corporation, 8th Exploration Conference, Cairo, 1986. Egyptian General Petroleum Corporation, Cairo. MURCHISON, D. G. 1987. Recent advances in organic petrology and organic geochemistry: an overview with some reference to 'oil from coal'. In: SCOTT, A. C. (ed.) Coal and Coal-bearing Strata: Recent Advances. Geological Society, London, Special Publication, 32,257-303. NOBLE, R. A., ALEXANDER,R., KAGI, R. I. & KNOX, J. 1986. Identification of some diterpenoid hydrocarbons in petroleum. Organic Geochemistry, 10, 825-829. , Wu, C. H. & ATKINSON,C. D. 1991. Petroleum generation and migration from Talang Akar coals and shales offshore NW Java, Indonesia. Organic Geochemistry, 17, 363-374. PHILP, R. P. & GILBERT, T. D. 1986. Biomarker distributions in Australian oils predominantly derived from terrigenous source material. Organic Geochemistry, 10, 73-84. POWELL,J. G., BOREHAM,C. J., SMYTH,M., RUSSELL, N. & COOK, A. C. 1991. Petroleum source rock assessment in non-marine sequences: pyrolysis and petrographic analysis of Australian coals and carbonaceous shales. Organic Geochemistry, 17, 375-394. RIELEY, G., COLLIER, R. J., JONES, D. M. & EGLINGTON, G. 1991. The biogeochemistry of Ellesmere Lake UK-1; source correlation of leaf wax inputs to the sedimentary lipid record. Organic Geochemistry, 17,901-912. RISK, M. J. & RHODES,E. G. 1985. From Mangroves to petroleum precursors: an example from tropical northeast Australia. American Association of Petroleum Geologists Bulletin, 69, 1230-1240. ROBERTSON GROUP, 1987. The petroleum geochemistry
and geology of the Carboniferous of Northern England and adjacent areas. Non-proprietary commercial report. ROBERTSON RESEARCH, 1983. The petroleum geochemistry of Indonesian Basins. Non-proprietary commercial report.
American Association of Petroleum Geologists Bulletin, 68, 31-49. SuTToN, C. 1979. Depositional environments and their relation to chemical composition of Java Sea crude oils. UN/ESCAP, CCOP Technical Publication, 6, 163-179, Bangkok. TAHER, M., SAID,M. & EL-AZHARY,T. 1988. Organic Geochemical Study in Meleiha area, Western Desert, Egypt. Proceedings of Egyptian General
Petroleum Corporation, 9th Exploration and Production Seminar, Cairo, November 1988. Egyptian General Petroleum Corporation, Cairo. TEICHMOLLER, M. & DURAND, B. 1983. Fluorescence microscopical rank studies on liptinites and vitrinites in peat and coals and comparison with results of Rock-Eval pyrolysis. International Journal of Coal Geology, 2, 197-230. THOMAS, B. M. 1982. Land plant source rocks for oil and their significance in Australian basins.
Australian Petroleum Exploration Association Journal, 22,166-178. & BROWN, S. A. 1983. Hydrocarbon generation in the Northern Perth Basin, Australian Petroleum Exploration Association Journal, 23, 64-74. THOMPSON, S., COOPER, B. S., MORLEY, R. J. & BARNARD, P. C. 1985. Oil-generating coals. /n: THOMAS, B. M. e t al. (eds). Petroleum -
-
Geochemistry in Exploration of the Norwegian Shelf. Graham and Trotman, London, 59-74. TISSOT, B. P. & WELTE, D. H. 1978. Petroleum Formation and Occurrence. Springer-Verlag, Berlin. VELD, H. & FERMONT, W. J. J. 1990. The effect of marine transgression on vitrinite reflectance values. Mededelingen rijks geologische dienst, 45, 151-170.
A maturity and palaeoenvironmental assessment of condensates and oils from the North Sumatra Basin, Indonesia C. J. M A T C H E T T E - D O W N E S , 1, A . E. F A L L I C K , 2 K A R M A J A Y A & S. R O W L A N D 3
1
1LEMIGAS Research & Development Centre for Oil & Gas Technology Cipilur Kebayoran Lama, Jakarta, 10002 Indonesia (Present address: Geochem Group, Chester Street, Chester CH4 8RD, UK) 2SURRC, East Kilbride, Glasgow G75 OQU, UK 3petroleum & Environmental Geochemistry Group, Department of Environmental Sciences, University of Plymouth, Drake Circus, Plymouth PL4 8AA, UK Abstract: Five light oils and condensates from the wells drilled in Lower to Upper Miocene reservoir rocks of the North Sumatra Basin were analysed by gas chromatography-mass spectrometry and stable isotope ratio mass spectrometry. In conjunction with data from over 1600 well and outcrop samples the results were used to describe the character of the presumed source rocks in terms of organic matter maturity and palaeoenvironment of deposition. The source rock character was further constrained by the maturity of the drilled sections, geothermal gradient observations and re-evaluation of the geology of the region. Dominant lacustrine with subordinate ombrogenous raised peat bog palaeoenvironments of deposition are indicated. It is envisaged that raised mires fringed large lakes in the area. Tectonic disturbance probably caused periodic flooding of the swamps and ultimately a complete marine incursion occurred. The oils and condensates are mature to extremely mature. Some oils are mixtures of oils of different maturities and discrete terrestrial sources.
The P E R T A M I N A Unit 1 block of the North Sumatra Basin (NSB), Sumatra, Indonesia lies mainly onshore the east coast of Sumatra and is bounded by the Barisan Mountains to t h e w e s t and for the purposes of the present research, between Southern Aceh in the north and Bohorok in the south (Fig. 1). A general stratigraphic column (after Mulhadiono et al. 1978) is shown in Fig. 2. Rocks of Oligocene to Upper Pliocene age have been encountered in outcrop or in well sections; however, the immaturity of the Upper Miocene and younger rocks precludes these as potential sources for the reservoired oils and discussion herein is therefore confined to sediments of Baong Formation (Middle Miocene) age and older. The geological age of the source rocks described in this paper is thought to be Palaeocene or younger.
History of geochemical research P E R T A M I N A Unit 1, NSB, has proved to be a prolific petroliferous basin producing vast quantities of high maturity higher plant-derived condensates and high API gravity oils. The area has been explored for over a century yet no firm
consensus as to the source rocks for these oils has emerged. Modern day geochemical studies mostly post-date the research of Kingstone (1978) who identified the paralic coaly sediments of the Oligocene Bruksah Formation (equivalent to the Parapat Formation, Fig. 2) as viable source rocks, mainly based on lithological considerations. By contrast, Davies (1984) believed the restricted marine rocks of the Oligocene Bampo Formation to be the sources, with: the latter author als0 citing the Lower Baong Formation rocks aS contributors. These inferences were made on t h e basis of Total Organic Carbon (TOC) and pyrolysis data and the great thickness of the mudstones. Subsequently Situmeang & Davies (1986) stated that the oil window was located in t h e Middle to Lower Baong Formation. Their judgements were based largely o n Tma x data. T h e somewhat higher TOC concentrations of these strata compared to other contenders, were als0 used ~to argue for the Baong Formation as source rocks. Kjellgren & Sugiharto (1989) arguedfor sources in the Bampo, Lower Baong a n d Belumai formations. T h e conflicting nature of these studies prompted a detailed Study by L E M I G A S
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coaland Coal-bearingStrataas Oil-prone SourceRocks? Geological Society Special Publication No. 77, pp. 139-148.
139
140
C. J. M A T C H E T T E - D O W N E S E T A L .
!
Muala ~/~ Barat-i
Scale
I0 Miles
Simpang
la la la ai
Sungai u h ~• Liput-i• Sungai
Bul
T a n j u n g Gentin, Pulau Rawa-
Susu
Dalam-2 Dalam-i Dalam-6 Buluh-1
-Pulau S e m b i l a n - i
Selatan-i
Besitang-i •
BRANDAN •
%
Telaga Said Tix Darat
Utara-IA
an~i Pakam Ti Pantai
Pakam- 1
~ai P~kam Timur~
Wampu-5
Batumandi-I
~
MEDAN
Selata~-I
A
P°l°nil- 1 Bohorok-
Tawang-I •
\ Fig. 1. Map of part of the North Sumatra Basin showing the position of the major wells.
and the British Geological Survey (BGS) who reviewed data for over 1500 potential source rock samples (TOC and Rock-Eval analysis) and concluded that the Baong Formation was the most likely source rock (Nicholson & Soekapradja 1990). Most recently Tambunan (1991) compared geochemical data for 69 potential source rocks with those from 8 oils and concluded that no source rock could be identified among the available samples. However, the
Belumai Formation was cited as the most likely originator of the oils. The controversies arising from these previous studies led to the present examination in which a detailed molecular and isotopic analysis of five representative oils was undertaken to obtain maturity and palaeoenvironmental information. This was combined with a a reappraisal of the existing geochemical database of bulk geochemical data for over 1500 well and outcrop
OILS FROM SUMATRA
141
Series Medan
Age
Aru
(Ma)
PLEIST.
Legend 1.7
volcanics
Julu
predominantly arenites
u.i 0.. t~ :3
'"
Keutapang
Fro.
U p p e r Shale Unit lA}
Z
iddle Baong L S a n ~ o n e Unit
LIJ ~J
carbonates
O
'" ._1
--
CI
dolomite
Upper Shale Unit (M)
Baong
-~_ ~ __&~---~-"&
predominantly rudites
Fm.
Lower Shale Unit (A)
Lower Shale~Middle Baong Sandstone Unit Unit (M) _J
~
Fm.
tw
---] predominantly argillites
~
Fro.
Seureula
5.1
~
Rayeu
B e l u ma i
25
Fro.
ampo
i
8
1
Fm
~
low grade metamorphics
~
hiatus
iJ.i (D O L0 J O
NOT DATED
.UY
Basedon Kamiliet and Mulhadionoet
basement'
aL (1976) al. (1978).
Fig. 2. Stratigraphy of North Sumatra Basin (after Mulhadiono et al. 1978).
samples. New well and outcrop data were also included, making a database of over 1600 samples. For a discussion of the broader geological tectonic setting readers are referred to Daly et al. (1991).
Sa m p l i n g and experimental p r o c e d u r e s Oil samples were collected from Lower and Upper Miocene reservoirs in Belumai Member and Keutapang Formation as indicated in Table 1. Samples were collected from the P E R T A M I N A Unit 1 core store in Pangkalen Brandan in October 1991 where they had been stored for 6 years. They were originally obtained in wells in 1985 as Drill Stem Test (DST) samples. Oils/condensates were separated into so-
called aliphatic and aromatic hydrocarbon fractions by now well-established and widely reported techniques (reviewed by Peters & Moldowan 1993). Briefly, oil was dried out into deactivated alumina and placed on a 1 : 3 alumina : silica column. Aliphatic hydrocarbons were eluted through the bed of alumina on silica with pentane whilst the aromatic fraction was first partially eluted with 9 0 : 1 0 pentane : dichloromethane then with 20 : 80 pentane : dichloromethane. The resulting fractions were examined by GC-MS on either a V G TRIO GCMS system (LEMIGAS) or a HP5890 (University of Plymouth) under conditions given in the figure legends. Carbon isotope analyses were performed at SURRC by combustion to CO: followed by MS. Results are reported as parts per thousand deviations from PDB standard (per mil).
C. J. MATCHE'I'TE-DOWNES ET AL.
142
Table 1. Sample details for oils from North Sumatra Basin Sample no.
Well
Depth (m)
NSB 1
Wampu-6
2342-2420
NSB2
Gabang-31
810-1048
NSB3 NSB4 NSB5
Pantai Pakam Timur-44 Telaga Said T-7 Securai-4
Not available 130-140 1790-1825
Formation Belumai Member Lower Miocene Keutapang Formation Upper Miocene Upper Miocene Upper Miocene Upper Miocene
metres U p p e r K e u t a p a n g Mbr. 3 0 m
Reservoirs
0
500 Keutapang
1.000 t o p B a o n g Fm. 1 2 2 4 m
1,500 drilling mud contamination from this depth 2,000 immature
:"default" s o u r c e
rock marine shales t o p B e l u m a i Mbr. 2 5 0 0 m
2,500 i • •
t o p B a m p o Fro. 2 9 1 0 m
oxidised vitrinite ?
3,000
Belumai
deepest well 3 1 5 0 m oil w i n d o w ?
3.500
k,jl 0.2
0.3
0.5
1
I
I
2
3
____t,* Vitrinite Reflectance Ro% ,
I
0 .....
,
I
2
,
I
4
w
6
I
z
8
10
TOC% {minimum threshold for source rock = 1%) ,
0
I
,
I
,
100 2 0 0
I
~
I
,
I
,
I
,
300 4 0 0 500 600 700
Hydrogen Index
P r o p o s e d source rock i n t e r v a l Ro 1 . 0 % +
f
D e p o c e n t r e b a s e m e n t 4 3 0 0 m 4-
Unconformity
Fig. 3, Summary of source rock data for a typical North Sumatra Basin sequence.
Unconformity
OILS FROM SUMATRA Results and discussion Before the molecular characteristics of the oil/ condensate samples are discussed, a brief summary of the possible reasons for the wide diversity of opinion regarding the possible source rocks of these oils is appropriate. Most of these features can be illustrated by reference to Fig. 3 which summarizes source rock data (TOC, Ro%, Hydrogen Index, HI) for a typical well in the study area. Firstly, and perhaps most importantly, as is typical in oil exploration all wells so far have been drilled on structural highs so that the 'kitchen' areas of source rocks on structurally low areas have never been encountered (cf. Telnaes & Cooper 1991). These lows are clearly visible from seismic and geophysical data (Matchette-Downes 1992). Thus, good samples of likely source rocks have probably never been obtained for geochemical characterization, leading to controversy in interpretation of data from immature source rocks. Secondly, the T O C contents of the Baong Formation are generally quite low (0.5-0.97% measured, but see below). Thus successive authors have invoked low minimum T O C thresholds for the region (as low as 0.3%). The vast thicknesses of the relatively lean shales that accumulated as a result of rapid down warping of the Sunda microplate during the Middle Miocene have been used to counter the argument that the Baong mudstones are too lean to be good source rocks (Davies 1984). Most good source rocks have TOC values of at least 1%. Thirdly, some authors may have ignored the fact that virtually all the cuttings samples and some cored material show signs of contamination with drilling fluids (Fig. 3). Some wells were drilled with diesel or other oil-based muds and yet others contained oil-based additives. This was normal practice in the 1980s in sections where overpressured and/or hygroscopic shales occurred. This contamination has probably
143
increased the measured T O C values which probably have an average concentration of nearer 0.5% (Matchette-Downes 1992). HI values suggest that some of the Middle and Lower Baong Formation could be source rocks but little optical geochemistry was used to assess the kerogen type. Such optical examination that has been made has revealed comminuted terrestrial material with, at best, a gas generation potential (Matchette-Downes 1992). The maturity of the sequence, as assessed by the generally accepted parameter of vitrinite reflectance, shows that the organic-rich shales with a sufficient HI value for oil generation are only barely mature (i.e. maximum c. 0.6% R0). In our view such early mature oil would not resemble the non-asphaltic light condensates found in the region.
Geochemical characteristics o f the oils The geochemical data for the five oils are shown in Table 2. All the samples for which data were available can be legitimately considered as condensates since they have API gravities of 47.0 to 55.2 (Table 2). They are low sulphur oils (0.020.03%), and gas chromatography (GC) was consistent with the high API values and revealed only minor proportions of >C25 alkanes. Indeed, the latter feature made the acquisition of GCMS biomarker data for palaeoenvironmental characterization rather more difficult than normal. Nonetheless, such data were obtained by selected ion monitoring of a limited number of ions.
Palaeoen vironment Biomarkers The oil from Timur 44 (NSB3) illustrates GC-MS biomarker features which are typical of all five oils.
Table 2. Summary of geochemical datafor oils Sample no.
API
Wt. % sulphur
613C sats
TNR
MPI
Rc %
NSB1 NSB2 NSB3 NSB4 NSB5
55.2 54.0 nd 48.5 47.0
0.02 0.03 nd nd 0.02
-21.7 - 23.1 - 22.8 - 22.4 -27.8
2.8 1.05 0.93 1.44 1.1
1.92 0.97 0.88 0.89 0.96
1.7 1.2 1.0 0.9 1.3
nd: not detected; API: American Petroleum Institute Gravity (degrees); TNR: trimethylnaphthalene ratio; MPI: methylphenanthrene index (for TNR and MPI see Radke 1987); Rc %: calculated vitrinite reflectance; 613C sats: 13C/~2Cisotope ratio of saturated hydrocarbons (per mil).
144
C. J. MATCHETTE-DOWNES E T AL.
The m / z 191 ion chromaogram (Fig. 4) is routinely used in petroleum geochemistry to illustrate the distribution of pentacyclic triterpanes (particularly hopanes) which are typically found in crude oils (e.g. Peters & Moldowan 1993). In marine oils, hopanes often dominate such profiles whereas a contribution of higher plant-derived organic matter is often revealed by significant proportions of other pentacyclics such as oleanane (e.g. Grantham et al. 1983). In oils from SE Asia and Bangladesh, polycyclic alkanes such as the bicadinanes are also often observed in the m / z 191 profile (e.g. Grantham et al. 1983; Alam & Pearson 1990; van Aarssen et al. 1990), though they are more clearly displayed by other ion chromatograms (e.g. m / z 369, 412). All of the above features (viz. oleanane, bicadinanes, hopanes) were observed in the NSB condensates, clearly suggesting an input of higher plantderived organic matter to the source rocks. The data show similarities to those obtained by others for condensates derived from Tertiary Talang Akar coaly sediments from Indonesia (e.g. Noble et al. 1991), and to oils sourced from the Ardjuna and Tertiary of offshore Sabah and Sarawak (Matchette-Downes 1992). The utility of the m / z 217 ion profiles, normally used to illustrate the distribution of steroidal alkanes (Peters & Moldowan 1993) is limited in the present samples because of the relatively large contributions of m / z 217 fragment ions of non-steroidal (e.g. triterpenoid)
alkanes. The high concentrations of these triterpenoids is consistent with the interpretations above. Also, the high maturity of the samples has probably reduced the relative contribution of C29 steranes compared with the lower C27 homologues, thus distorting the commonly used sterane carbon number distributions (cf. Mackenzie 1984). Interestingly the 6~3C data obtained for the aromatic and aliphatic fractions of the five condensates did not support the biomarker data. For instance, calculation of the Canonical Variable (CV) proposed by Sofer (1984) for the differentiation of higher plant and marine oils suggested that the NSB samples were of marine origin (CV - 5 . 5 0 to -1.26). However, we have observed that isotopically heavy oils from lacustrine environments also exhibit features which are more usually attributed to 'marine' oils. Examples are the lacustrine oils from the algal-rich Pematang Brown Shales of central Sumatra (Longley et al. 1990), and the Sunda oils of offshore North West Java (Pramono et al. 1990). Sorer (1984) also noted that some oils from Sumatra, Java and the Phillipines did not fit his general model well. Figure 5 shows the differentiation of known higher plant-derived oils and known lacustrine oils (Pramono et al. 1990) on the basis of a 7%0 difference in the 613C isotopic signature (Luo Binjie et al. 1988). Clearly four of the five condensates exhibit an isotopic signature which Stable isotopes
m/z 1 9 1
~bund . . . .
0
Time
.
.
.
.
~
32.0"0
,
,
,
341.00 . . . .
,
3 6 . 0I0 . . . .
3 8 . 0I0 . . . .
4 0 . 0l0 . . . .
4 2 . 0l0 . . . .
4 4 . 0~0 . . . .
4 6 .lO0
,
,
,
i
,
,
,
48.00
Fig. 4. GCMS ion chromatogram (rn/z 191) of NSB2. Positions of norhopane (1), oleanane (2) and hopane (3) are indicated.
OILS FROM SUMATRA 10 Area for Ardjuna oils (higher plant) -
1114
5
3
3
2
>Ot-
Area for Sunda oils (lacustrine) 0.5
0.5
0,3
0.3
0.2 -34 -32
-30
-28 -26 -24
-22
-20
-11
0.2
(~130 ppmil. I
Fresh-to-brackish lacustrine
II
Saline lake
Ill
Fluvial-lacustrine-bog
IV
Peat-swamp
V
Marine environment
•
North Sumatra condensates
Fig. 5. Plot (after Luo Binjie et al. 1988) of pristane/ phytane ratio versus isotopic signature for the NSB condensates. Also shown are summarized data for typically higher plant-derived oils (IV) from Ardjuna and lacustrine oils (V) from Sunda, offshore northwest Java (Pramono et al. 1990).
is dominantly lacustrine. The exception is the NSB5 Securai-4 sample. The picture that emerges from these data in conjunction with the biomarker results is that the latter oil is of the Ardjuna higher plant-derived type whilst the others originate from dominant lacustrine material with about 25% higher plant input. An environment of alternating lacustrine (dominant) to limnic coaly to ombrogenous peat swamp is envisaged possibly represented by a series of periodically interconnected shallow basins separated by low hills (viz. somewhat analogous to the Eocene Gippsland Basin of Australia but with a far greater development of lacustrine facies). The reason for the seemingly dominant higher plant) biomarker pattern revealed by GCMS is thought to be the presence of distinctive polycyclics (bicadinanes, oleananes) which, on mixing of oils from lacustrine source materials, may distract attention from the lower concentrations of more commonplace and less distinctive/hopanes, which occur in both higher plant and lacustrinederived organic matter.
145
Maturity
The maturity of source rocks is commonly calculated from the reflectance of vitrinite. Oils, of course, do not contain macerals and a variety of chemical reactions have been investigated as possible alternative measures of thermal maturity. Both aliphatic and aromatic hydrocarbons have been used, the former normally involving measurement of epimers (although the actual mechanism may not be an epimerization (Abbott et al. 1990)), the latter involving measurement of either the aromatization process itself, alkyl-chain cracking of the aromatics and/or alkyl substituent isomerization (see Peters & Moldowan 1993 for a general guide). Use of aliphatic steriodal hydrocarbon distributions (m/z 217) for maturity assessment was precluded in the present study due to the large contribution of m / z 217 ions from the abundant non-steroidal polycyclics. However, the contribution of terrigenous organic matter to the condensates and their high abundance of volatile hydrocarbons made the use of alkylnaphthalenes and alkylphenanthrenes as maturity indices perfectly viable (Alexander et al. 1985; Radke 1987) and calculated reflectance values (Rc %) were obtained from the data (Table 2). Values of the ratios TNR and MPI (Table 2; cf. Alexander et al. 1985; Radke 1987) which essentially indicate the increasing proportion of /3-substituted trimethylnaphthalenes and r-substituted methylphenanthrenes relative to their a-counterparts with increasing maturity, show that the condensates are all mature varying from peak maturity (NSB4 Telaga Said T-7, Rc 0.9%) to post-mature (NSB1 Wampu-6 Rc 1.7%). The salient features of the ion chromatograms are illustrated in Figs 6 & 7, where the dramatic increase in fl/3fl-2,3,6-trimethylnaphthalene and /3-3- and 2-methylphenanthrenes is obvious for NSB1 (Fig. 6) compared with NSB4 (Fig. 7). Additional indications of the high maturities of these oils were evident from the low proportion of C26_28triaromatic steroids in the oils relative to C21 counterparts (cf. Mackenzie 1984); particularly for the postmature sample NSB 1. There is a broad trend of increased maturity with increasing depth throughout the study region (Matchette-Downes 1992), where the least mature oils have been displaced by more mature oils as the sub-basin progressively matured. Between the two maturity end members represented by the oils of the Telaga Said area in the Kentapang Formation reservoirs and post-mature condensates pooled in the Wampu area in the Belumai member reservoirs,
146
ET AL.
C. J. M A T C H E T - I ' E - D O W N E S %bundance
m/z
' ' ,1.00
Pime _>0
~_2'00. ' ' { 3 ' 0 0 ' .
' 14.00
ls.oo
16.00
17.00
18.00
19.00
I i
.
.
£ime _>0 Xbundance
Fime _>0 %bundance
.
.
llJOO
.
A
A:,,. , . '
. 13too.
12100.
j.,_~, . . . . . .
14.00
, .A:._.^.,..A. ~
15.00
16.00
v ' '
17.00
142
~
18.00
m/z
156
'
'~-'="-
19.00
m / z 1 70
.
.
.
.
.
1110o
.
.
.
.
.
,'2 .'oo
.
.
i
~', .'oo
.
14.oo
.
.
.
i
'
lS.OO
'
'
i
.
1~.oo
.
17.oo
.
.
.
.
.
1~.oo
.
.
.
.
19.oo m/z
192
B
(a)
£ime _>0
2~Ioo
nloJ
~Ioo
' ~.41oo. . . .
'
26.00
'
27.oo
•IihI, ~.|mlo{
m Iz 1 42 #
0 Pime -> %bundance
' ' ' 11
oo
' ' ~2Joo
=13.oor'
.^.
, 4~ . ,o b ~ ' -' ' { s- ,. o b 1
,-
~16.oo _ = _ ~ L _ : _ ~ ,17.oo . _. _ ~ _ . . . .18.oo ,
19.o,, m l z 156
16 • O0
19 • O0
B~ ~
h
0 £ime -> kbundance
ii. O0
12 • O0
13 • O0
14. O0
15 • O0
17 • O0
18 • O0
m/z
0 1 ~ime -> %bundance
.
.
.
.
.
.
.
.
111 O0
,
. . . .
12. O0
,
. . . .
13. O0
,
. . . .
14. O0
= '~,
15. O0
,
, ~A¢'~
16. O0
. . . .
17. O0
~
,
18.. O0
, A
,
19. O0
m l z 192 x~
(b)
L
ime ->
17o
o
21. oo
2 2 . oo
23. oo
24. oo
25. oo
26. oo
27. o0
OILS FROM SUMATRA there has been mixing of the hydrocarbons during successive phases of hydrocarbon expulsion and migration. However, it is important to appreciate that the maturity of the oils examined is far too great for them to have been generated from the Baong Formation. The maturity data considerably contrain the possible location of the source rocks and indicate that these are to be found at depth and probably will have passed through the 'condensate' generation window (e.g. Ro 1.7%) in many of the sub-basins.
(4)
(5)
Conclusions Examination of five condensate/oils from the North Sumatra Basin by G C - M S and stable isotope MS has revealed the following: (1) An apparent conflict between the biomarker GCMS data (indicating a higher plant-derived source material) and the isotope data (indicating marine lacustrinederived source materials). (2) The conflict can be resolved in the following way. Both lacustrine and higher plantderived oils have some steranes and hopanes in common. However, higher plant material from the Cretaceous onwards also often contains additional distinctive alkanes such as oleananes and bicadinanes in high concentrations. The presence of relatively small (c. 25%) amounts of higher plant material containing the latter compounds may then dominate the m/z 191 biomarker profiles. (3) A viable interpretation, consistent with all the data, invokes a mixed source environment for four of the five oils (NSB 1-4) with a, not necessarily time-equivalent, minor (c.
(6)
(7) (8)
(9)
147
25%) higher plant signature mixed with a dominant marine/lacustrine input. The regional geology precludes an exclusively marine source at depth. The remaining oil (NSB5) appears to be sourced from a dominant higher plant source material. The facies is interpreted as alternating lacustrine (dominant) to limnic coaly to ombrogenous peat swamp. Assessment of maturity via the distributions of alkylaromatics was enhanced by the terrigenous nature of the source material, whilst use of aliphatics was obviated by the co-occurrence of high concentrations of terrestrially derived polycyclic alkanes with the steroidal alkanes. The oils/condensates ranged from peak maturity (Re 0.9%) to post-maturity (Re 1.7%) in the deeper depocentres. Mixtures of oils with different maturities were probably present in some reservoirs. The data considerably contrain the possible source rock location and, in our view, negate the insufficiently mature Baong Formation as the dominant source. It is thought that the unrecovered source rocks probably now lie at depth with Ro/> 1.7%.
We are extremely grateful to both LEMIGAS, Indonesia and BGS, UK for permission to publish these findings and to LEMIGAS for access to the samples and unpublished data. Colleagues at LEMIGAS and University of Plymouth are acknowledged for help and advice. Especially we would like to thank Marnida Ulibasa (LEMIGAS) for help with the LC separations and GC analyses and to Roger Srodzinski (Plymouth) for help with GCMS. NERC is acknowledged for provision of the latter GCMS facilities.
References VANAARSEN,B. G. K., Cox, H. C., HOOGENDOORN,P. & DE LEEUW, J. W. 1990. A cadinene biolpolymer in fossil and extant dammar resins as a source for cadinanes and bicadinanes in crude oils from South East Asia. Geochimica et Cosmochimica Acta, 54, 3021-3031. ABBOTT,G. D., WANG,G. Y., EGLINTON,T. I., HOME, A. K. & PETCH, G. S. 1990. The kinetics of sterane biological marker release and degrada-
tion processes during the hydrous pyrolysis of vitrinite kerogen. Geochimica et Cosmochimica Acta, 54, 2451-2461. ALAM, M. & PEARSON, M. 1990. Bicadinanes in oils from the Surma Basin, Bangladesh. Organic Geochemistry, 15,461-464. ALEXANDER, R., KAGI, R. J., ROWLAND, S. J., SHEPPARD, P. N. & CHIRILA, T. V. 1985. The effects of thermal maturity on distributions of
Fig. 6 (a) & (b). GC-MS ion chromatograms (m/z 142,156,170,192) illustrating, respectively, the distributions of mono-, di- and trimethylnaphthalenes and methylphenanthrenes in a peak maturity condensate (NSB4) and a post-mature condensate (NSB 1). Annotated peaks show the increase in r-substituted aromatics with increasing maturity. GCMS conditions: Hewlett-Packard 5890 GCMS (University of Plymouth). 12 m HP-1 capillary GC column 60-300°C @ 5°C min. -1 Selected ion monitoring.
148
C. J. MATCHEq-TE-DOWNES E T A L .
dimethylnaphthalenes and trimethylnephthalenes in some ancient sediments and petroleums. Geochimica et Cosmochimica Acta, 49, 385-395. DALY, M. C., COOPER, M. A., WILSON, I., SMITH, D. G. ,~gHODPER,B. G. D. 1991. Cenozoic plate tectonics and basin evolution in Indonesia. Marine and Petroleum Geology, 8, 2-21. DAVIES, P. R. 1984. Tertiary structural evolution and related hydrocarbon occurrences, North Sumatra Basin. In: Proceedings of the 13th annual convention of the Indonesian Petroleum Association, Jakarta. Indonesian Petroleum Association, Jakarta, 19-49. GRANTHAM, P. J., POSTHUMA, J. ,~z BAAK, A. 1983. Triterpanes in a number of Far-Eastern crude oils. In: BJOROY,M., ALBRECHT,P., CORNEORTH, C., DE GROOT,K., EGLINGTON,G., GALIMOV,E., LEYTHAEUSER, D., PELET, R., RULLKOTFER,J. ,~ SPEERS, G. (eds) Advances in Organic Geochemistry, 1981. John Wiley, Chichester, 675683. KINGSTONE, J. 1978. Oil and gas generation, migration and accumulation in the North Sumatra Basin. In; Proceedings of the 7th annual convention of the Indonesian Petroleum Association, Jakarta. Indonesian Petroleum Association, Jakarta, 75-104. KJELLGREN, G. M. ,~ SUGIHARTO,H. 1989. Oil geochemistry: a clue to the hydrocarbon history and prospectivity of the south eastern North Sumatra Basin, Indonesia. In: Proceedings of the 18th annual convention of the Indonesian Petroleum Association, Jakarta. Indonesian Petroleum Association, Jakarta, 362-384. LONGLEY, I. M., BARRACLOUGH,R., BRIDDEN,M. A. & BROWN, S. 1990. Pematang lacustrine petroleum source rocks from the Malacca Strait PSC, Central Sumatra, Indonesia. In: Proceedings of the 19th annual convention of the Indonesian Petroleum Association, Jakarta. Indonesian Petroleum Association, Jakarta, 279-297. Luo BINJIE, YANG XINGHUA, LIN HEJIE ,~ ZHENG GUODONG. 1988. Characteristics of Mesozoic and Cenozoic non-marine source rocks in north-west China. In: FLEET, A. J., KELTS, K. ,~: TALBOT, M. R. (eds) Lacustrine Petroleum Source Rocks. Geological Society, London, Special Publication, 40,291-298. MACKENZIE, A. S. 1984. Applications of biological markers in petroleum geochemistry. In: BROOKS, J. t~ WELTE, D. (eds) Advances in Petroleum
Geochemistry, Volume 1. Academic, London, 115-214. MATCHETrE-DOWNES, C. 1992. Petroleum geochemistry of the North Sumatra Basin, Part I. LEMIGAS, Jakarta. MULHADIONO., HARTOYO, P. & SOEDALDJO, P. A. 1978. The Middle Baong Sandstone units as one of the most prospective units in the Aru area, North Sumatra. In: Proceedings of the 7th annual convention of the Indonesian Petroleum Association, Jakarta. Indonesian Petroleum Association, Jakarta, 75-104. NICHOLSON, R. A. & SOEKAPRADJA,S. 1990. Organic geochemical studies in the North Sumatra Basin. In: Scientific Contribution on petroleum science and technology. Special Publication LEMIGAS, Jakarta, 45-67. NOBLE, R. A., Wu, C. H. & ATKINSON, C. D. 1991. Petroleum generation and migration from Talang Akar coals and Shales offshore N.W. Java, Indonesia. Organic Geochemistry, 17,363374. PETERS, K. & MOLDOWAN,J. M. 1993. The Biomarker Guide: Interpreting Molecular Fossils in Petroleum and Ancient Sediments. Prentice Hall, New Jersey, USA. PRAMONO,H., Wu, C. H. &NOBLE, R. A. 1990. A n e w oil kitchen and petroleum bearing subbasin in the offshore Northwest Java Basin. In: Proceedings of the Indonesian Petroleum Association, Jakarta, 19,253-278. RADKE, M. 1987. Organic geochemistry of aromatic hydrocarbons. In: BROOKS,J. & WELTE,D. (eds) Advances in Petroleum Geochemistry, Volume 2. Academic Press, London, 141-207. SITUMEANG,S. ~ DAVIES,P. R. 1986. A geochemical study of Asamera's block 'A' production sharing contract area, North Sumatra Basin. In: Proceedings of the 15th annual convention of the Indonesian Petroleum Association, Jakarta. Indonesian Petroleum Association, Jakarta, 321-340. SOFER, Z. 1984. Stable carbon isotope compositions of crude oils: application to source depositional environments and petroleum alteration. American Association of Petroleum Geologists Bulletin, 68, 31-49. TAMBUNAN, L. 1991. Biomarker applications in the North Sumatra Basin. MPhil. thesis, University of Newcastle, UK. TELNAES, N. • COOPER, B. S. 1991. Oil-source rock correlation using biological markers, Norwegian continental shelf. Marine and Petroleum Geology, 8,302-310.
Geochemistry of aliphatic-rich coals in the Cooper Basin, Australia and Taranaki Basin, New Zealand: implications for the occurrence of potentially oil-generative coals DAVID
J. C U R R Y 1, J O H N
K. E M M E T T 2 & J O H N
W. H U N T 2'3
aExxon Production Research Company, PO Box 2189, Houston, Texas 77252, USA 2Esso Australia Ltd, G P O Box 400, Melbourne 3001, Australia aPresent address: Groundwater Technology Australia Pty Ltd, 17 Forrester St, Kingsgrove, N S W 2208, Australia Abstract: Although the concentration of long-chain aliphatic constituents is a primary
determinant of the oil generation potential of coals, the factors which govern their occurrence in different coals are poorly understood. In this study, Permian coals from the Cooper Basin, Australia, and the Eocene coals from the Taranaki Basin, New Zealand, were compared to determine these factors. The Taranaki Basin coals were deposited in temperate, fluvial-deltaic environments. HI values range from 236-365. Extracts have high pristane/phytane ratios and variable abundances of oleanane and other non-hopanoid terpanes. The extracts and pyrolysates contain high relative concentrations of aliphatic groups > n-C20. These data imply that much of this aliphatic carbon is derived directly from higher plant material. The Cooper Basin coals were deposited in high latitude bogs and contain 40-70% inertinite. The coals have been severel~¢degraded. Pristane/phytane ratios are low (2.15-6), but His are moderate (up to 243 mg g- OC). The extracts and pyrolysates both contain high relative concentrations of aliphatic groups; however, the distributions are different from higher plant-derived material. These data imply the bulk of the aliphatic carbon in these coals is derived from microbial biomass (both bacterial and fungal degradation products and algal input). These results show that long-chain aliphatic groups in coals can be derived directly from the higher plant material, from microbial activity in the depositional environment, or from a combination of the two.
The potential of terrestrially derived coals to generate and expel oil has been the subject of an increasing amount of interest in the last several years, and the concept that some coals can act as oil sources has gained a significant measure of acceptance, especially for coals from Australia (e.g. Powell et al. 1991), New Zealand (Johnston et al. 1991), and Indonesia (e.g. Durand & Paratte 1983; Huc et al. 1986). However, there are still questions associated with this concept. For example, although many coals have been shown to generate abundant high molecular weight species such as normal alkanes (e.g. Powell et al. 1991), their ability to efficiently expel these components has been questioned. Observational data (Powell & Boreham 1991) and laboratory measurements (Sandvik et al. 1992) indicate relatively high expulsion efficiencies; however, hydrous pyrolysis data (e.g. Teerman & Hwang 1989) indicate low expulsion efficiencies of oil-like material. In addition, the relationship of maceral compositions to oil
generation potential (Bertrand et al. 1986) and to expulsion mechanisms (Horsfield et al. 1988) are still incompletely understood. There is also considerable uncertainty over which types of coals can be oil-generative and where, geographically and stratigraphically, they can occur. Oils generated from terrestrially derived organic matter characteristically contain very high concentrations of normal alkanes, especially in the higher molecular weight ranges. Consequently, in order for a coal to be oil generative, it must, as a first criterion, contain sufficient concentrations of long-chain weight normal aliphatic groups (i.e. ~>n-C12+) to generate the normal alkanes occurring in the oils (Curry 1985; Smith et al. 1987; Powell et al. 1991). As a result, the ecological, depositional, and diagenetic factors which control the occurrences and concentrations of long-chain aliphatic groups in coals will have a critical influence on their capability to generate oil. These factors are still incompletely understood,
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coal and Coal-bearing Strata as Oil-prone Source Rocks? Geological Society Special Publication No. 77, pp. 149-182.
149
150
D. J. CURRY ET AL.
especially since coals are very heterogeneous in character and there is a high degree of variability in the composition of coals from different ages and depositional environments. The objectives of this study, therefore, were to delineate the factors which control the occurrences of long-chain aliphatic groups in coals and to determine how these factors affect the occurrences of potentially oil-prone coals. In order to delineate these factors, coals from the Taranaki Basin, New Zealand, and Cooper Basin, Australia were compared. Although coals from both these areas are terrestrially derived, relatively aliphatic rich, and associated with known oil production, they are of different ages, derived from very different palaeofloras, and were deposited under very different conditions. As a result, comparison of the geological and geochemical data for these dissimilar coals can provide insight into the processes by which some coals are enriched in aliphatic carbon.
Methods All samples were taken from cores. Elemental analyses were performed by Huffman Laboratories, Golden, CO. Vitrinite reflectance and maceral counts were done by Dr Alan Cook of Keiraville Konsultants, Keiraville, NSW. Total organic carbon analyses (for the shales) were carried out using a LECO analyser. Bitumen extraction used a 9:1 mixture of dichloromethane : methanol in Tecator extraction apparatus; chromatographic separations were carried out by HPLC. Gas chromatography of the total extracts, saturate fractions, and aromatic fractions were carried out using Hewlett Packard 5890 gas chromatographs fitted with 30 m DB-5 capillary columns. Saturate and aromatic fractions were analysed with hydrogen carrier gas. GC/MS analyses in the SIM mode were carried out using a Hewlett Packard 5890 gas chromatograph fitted with a 30 m DB-5 capillary column and interfaced to a Varian INCOS 50 quadrupole mass spectrometer. Full scan gc/ms analyses utilized a Hewlett Packard 5890 gas chromatograph fitted with a 30 m DB-5 capillary column, interfaced to a Hewlett Packard 5970 Mass Selective Detector. Pyrolysis-gc/ms analyses were carried out on a custom-built system consisting of a CDS pyroprobe unit fitted to a Hewlett Packard 5890 gas chromatograph with a 30m DB-5 capillary column and a Hewlett Packard 5970 Mass Selective Detector operating in the full scan mode.
Taranaki Basin
Geological setting and sample description The Taranaki Basin is a Cretaceous-Tertiary, clastic-dominated basin located on the western side of the North Island of New Zealand. The basin is divided into the offshore Western Platform region and the offshore-onshore Taranaki Graben (Pilaar & Wakefield 1978). Basin fill ranges from terrestrial and paralic to deep-water marine sediments. For more complete discussions of Taranaki Basin geology see Pilaar & Wakefield (1978, 1984) and King & Robinson (1988). The coals and shales in this study are from the Eocene Kapuni Group and were deposited in an upper to lower coastal plain setting (King & Robinson 1988). The palaeoclimate was temperate to sub-tropical with abundant rainfall (Pocknall 1990). The coals and shales analysed in this study are from cores from the Mangahewa-1 and Kapuni-8 wells in the onshore area of the Taranaki Graben. A summary of sample numbers, depths and lithology is contained in Table 1 in the Appendix at the end of this paper.
Geochemical characterization The coals and kerogens from both wells are incipiently mature. The Rv(ma,,) values from the Kapuni-8 well range from 0.56-0.70%, with an average maturity of Rv(max) of 0.64% for the coals. The Rv(max)values for the samples from the Mangahewa-1 well range from 0.64-0.77%, with an average Rv(max)of 0.73% for the coals. Neither set of samples appears to have generated significant amounts of hydrocarbons. The maceral compositions of the coals and kerogens are characteristic of higher plantderived coals, containing mainly vitrinite (Appendix: Table 2; Fig. 1). Cutinite, resinite, and some sporinite are the main liptinitic macerals. The coals from the Kapuni-8 well tend to have higher liptinite contents than do those from the Mangahewa-1 well, while shales with TOC less than 20%, generally have higher concentrations of liptinite than the interbedded coals. Inertinite concentrations are low in all samples. Although the high concentrations of vitrinite are typical of higher plant-derived coals, the Taranaki Basin samples are more hydrogen rich than many other terrestrially derived coals of similar maturity, with H/C ratios ranging from 0.81-1.01 (Appendix: Table 3; Fig. 2). In contrast, the O/C ratios are somewhat lower than those of many humic coals at these maturities.
ALIPHATIC-RICH COALS FROM AUSTRALIA AND NEW ZEALAND 100 INERTINITE
•
VITRINITE 100
C O O P E R BASIN
LIPTINITE 100
.t-
Fig. 1. Maceral composition: Taranaki Basin and Cooper Basin coals and shales.
2.0 f
I
/
1.5
y
H/C 1.o HI
0
0.05
0.10
0.15
0.20
0.25
0.30
O/C Fig. 2. Elemental compositions (Van Krevelen Diagram): Taranaki Basin and Cooper Basin coals.
151
152
D. J. CURRY E T A L .
As expected, Rock-Eval $2 values are very high for the coals (Table 3). Hydrogen Index (HI) values range from 236-361mgg -1 OC. These values are within the range of hydrogenrich terrestrial organic matter from areas such as the Gippsland Basin, Australia (Powell et al. 1991) and the Mahakam Delta of Kalimantan (Huc et al. 1986). Conversely, the $3 and Oxygen Index (OI) values are very low for terrestrial organic matter at this level of maturity. The pristane/phytane ratios for the bitumen extracts of the coals range from 6.46-12.29, with most values greater than 8.0 (Appendix: Table 4a). These values are significantly higher than the ratio of 4.0 considered characteristic of higher plant-derived organic matter (Tissot & Welte 1984). In addition, the pristane/n-C17 alkane ratios for the coals are generally greater than 4.0, which is also characteristic of immature, higher plant-derived material. The most abundant individual consituents of the bitumen extracts are pristane and C1-, C2-, and C3-substituted naphthalenes (e.g. Fig. 3), which appears to be a characteristic of Tertiary higher plant derived-organic matter from this region. However, the total extract gas chromatograms also contain large abundances of high molecular weight n-alkanes, especially
the n-C25,/'/-C27, and n-C29 alkanes. The H-C27/ n-C17 ratios are generally greater than 1.0 and frequently greater than 3.0. In addition, the sum of the n-C2s to n-C34 alkanes is usually greater than 50% of the total Cls+ n-alkanes (Table 4a). The sterane and triterpane biomarker constituents also indicate a predominantly higher plant origin. Hopane/sterane ratios are very high and there are no systematic differences between the coals and the shales. The sterane fractions are dominated by the C29 regular and rearranged steranes, with C29 sterane/total sterane ratios as high as 0.92 for some coals (Table 4a, Figs 4a, b). There are no significant concentrations of 4-methyl C30 sterane homologues. Significantly, these coals contain relatively high concentrations of diasteranes (Table 4a; Figs 4a, b). The concentrations of diasteranes in the coals (expressed as parts per million of the saturate fractions) are equivalent to the diasterane concentrations in the interbedded shales. Since these are low-ash coals, the formation of the diasteranes by clay-catalysed rearrangements is unlikely. The diasteranes were instead probably formed by rearrangements catalysed by the very acidic conditions frequently prevalent in peat swamps (mires). Similar high abundances of diasteranes can be observed in the sterane fractions of other coals,
f
O
Me-Naph
'
0.00
1
5,62
'
I
11.25
'
I
16.87
~
I
22.50
'
1
28.12
'
I
33.75
'
I
39.37
i
45.00
Fig. 3. Total extract gas chromatogram: coal from 11805 ft in the Mangahewa-1 Well. Taranaki Basin (see Appendix: Table 6 for explanation of symbols).
ALIPHATIC-RICH COALS FROM AUSTRALIA AND NEW ZEALAND for example, coals from the Sakoa coalfield in Madagascar (Ramanampisoa et al. 1989), coals from the Cooper Basin (see below), and Westphalian coals from the Midlands of the UK (Curry, unpublished data). The occurrence of significant amounts of diasteranes in these low ash coals implies that the occurrence of diasteranes as indicators of clastic-rich source rocks may be of limited usefulness in terrestrial organic facies. The triterpane fractions of the coals and shales contain varying abundances of 18-a oleanane, with oleanane/C3o hopane ratios ranging from 0.02 to 0.52. Although the coals and shales from the Mangahewa-1 well generally have lower oleanane/hopane ratios than those in the Kapuni-8 well, there is a considerable degree of variation within each well. More significantly, however, the coal extracts contain relatively high absolute concentrations of oleanane. As shown in Table 4a, the concentration of oleanane in the coals varies from 56ngmg - 1 saturate fraction to 430ngmg -1 saturate fraction, with the Kapuni-8 coals generally having higher concentrations of oleanane than the Mangahewa-1 coals. The coals and shales also contain several other triterpane constituents, including two components indicated as W and T in Table 4a and Figs 5a and 5b. These constituents were initially identified as bicadinane isomers W and T by comparison of relative retention times and mass spectra with published data (Grantham et al. 1983; van Aaarsen et al. 1992). However, the occurrence of bicadinanes in pre-Oligocene sediments and oils is rare, and recent work has shown that these constituents instead may possibly be C30 triterpanes similar to oleanane, but in which the A-ring is contracted to a fivemembered ring (ten Haven, pers. comm.). However, there is no correlation between the concentration of oleanane and triterpane isomer W. (Triterpane isomer T could not be quantified because of partial co-elution with the C27 Tm hopane.) The variability in the concentrations of these compounds over short intervals implies a high frequency of variation in the precursor plant assemblage, especially in the abundance of angiosperms. In addition, the lack of correlation between the abundances of oleanane and triterpane isomer W indicates that their precursors are not genetically related to a straightforward manner. The relative distributions of most triterpanes (except the C27 Tm hopanes, see below) are generally uniform for the samples from each well, and most of the compound ratios (e.g. C29/C30) are equivalent for the coals and shales
153
(Table 4a). However, the absolute concentrations of the triterpane constituents are significantly higher in the coals and carbonaceous shales (i.e. T O C > 2 5 % ) than in the adjacent lower TOC shales (Appendix: Table 5a). In addition, the concentrations of triterpanes tend to be higher in the Mangahewa-1 well than in the Kapuni-8 well (Table 5a). As noted above, the abundances of the C27 Tm hopanes appear to be anomalous. Although the abundances of these constituents could not be reliably quantified due to co-elution of triterpane T and another unidentified triterpane, qualitative assessment of full scan mass spectra indicate that the coals and carbonaceous shales contain significantly higher concentrations of C27 Tm than the lower-TOC shales. Consequently, Ts/Tm ratios are significantly lower in the coals as compared to the shales. This effect is also noted in the coals and shales from the Cooper Basin (see below) and in other areas (Curry, in prep.). These data show that these coals are of terrestrial (higher plant) origin, and were deposited in a clastic-poor, highly acidic, oxidizing peat swamp (mire) with little degradation. The data also show there is a difference in organic facies between the coals and shales in the Mangahewa-1 and Kapuni-8 wells. The differences in the HI values, liptinite content, oleanane and triterpane concentrations, hopane concentrations, and proportion of high molecular weight n-alkanes in the extracts indicate that the organic facies in these wells, while both predominantly terrestrial in character, are not identical. In particular, the increased concentrations of oleanane in the coal bitumens imply that the coals in the Kapuni-8 well had a higher proportion of angiosperm precursors than the Mangahewa-1 coals.
Aliphatic carbon content
These coals and kerogens contain high relative concentrations of long-chain aliphatic groups in the pyrolysates (e.g. Figs 6a, b). In both wells, the relative concentrations of higher molecular weight (n-C2o+) aliphatic groups are high compared to the lower molecular weight aliphatic groups, even when the differences in mass spectrometer response factors are taken into account. This can be seen in, for example, Figs 7a and b, which show the alkyl constituents (as the sum of the m / z 69 + 70 + 71 ions) of the pyrolysates. However, the coal pyrolysates also contain high concentrations of phenol and methylated phenol constituents ('P' in Figs
154
D. J. CURRY ET AL.
100.0-
I
C29(+C29 H)
217-I STD
C29 T
i
(a)
1000
I
'
1100
I
'
1200
I
1300
I
'
'
1400 SCAN
I
'
1500
I
'
1600
I
'
1700
1800
100 C29(+C29H)
C29 D
217 C29 D
|
I
'
I
'
I
'
I
I
'
I
'
I
'
I
'
1400 1500 1600 1700 1800 (b) SCAN Fig. 4" Representative sterane distributions for Taranaki Basin coals (m/z 217 mass fragmentograms); (see Appendix: Table 6 for explanation of symbols): (a) Mangahewa-1 (11 811 ft); (b) Kapuni-8 (13 294 ft). 1000
1100
1200
1300
ALIPHATIC-RICH COALS FROM AUSTRALIA AND NEW Z E A L A N D
155
10o
C2~ C27
191
]
(a)
1000
I
'
I
1100
'
I
1200
'
I
1300
'
1400
I
'
1500
I
'
1600
I
I
1700
1800
SCAN
100' C27 + T
C30
191 C31 i
i~,,
I
,' i
,,j !
(b)
1000
~_ .... ~ r I
1100
4
| _.,,tJ
r
'
lal ~.._.,,~ °_.~
f 1200
i
l
r 130(
~-"'
'i'
i
L
C---'14oo
-T lsoo
-T--]
-~'¥ leOO
..... -(---~17oo
.... ' " 18oo
SCAN
Fig. 5. Representative triterpane distributions for Taranaki Basin coals (m/z 191 mass fragmentograms) (see Appendix: Table 6 for explanation of symbols): (a) Mangahewa-1 (11803 ft); (b) Kapuni-8 (13 269 ft).
156
D.J. CURRY
ET AL.
1.4e + 0720
1.2e + 0715
/
I e + 07
iii
6000000
~loiN
4000000
25
i
2000000
~
I~L,.
0 (a)
_
701 /|0
10.00
20.00
M
_
80100
00
1.4e + 07t 1.2e + 07-1
8000000
"0
6oooooo~ ~i ~, ~
I
jjli
li
i000010004 i b / / jd j 00000 0
10.00
20.00
30.00
(b)
40.00 TIME
50.00
60.00
70.00
80.00
Fig. 6. Representative pyrolysis-gc/ms total ion chromatograms for Taranaki Basin coals (see text for explanation of symbols): (a) Mangahewa- 1 (11 811 ft); (b) Kapuni-8 (13 273.5 ft).
6a, b), methylated benzenes ('A') and methylated naphthalenes ('Me-N'), characteristic of terrestrially derived organic matter. Also evident in Figs 7a and b is the difference in the distributions of long-chain aliphatic
groups in the coal pyrolysates between the two wells. The coals from the Kapuni-8 well (e.g. Figs 6a, 7a) generally contain higher relative concentrations of n-C25÷ components than do coals from the Mangahewa-1 well (e.g. Figs
157
ALIPHATIC-RICH COALS FROM AUSTRALIA AND NEW ZEALAND
1 -Pr t
8000000-
20
7000000-
6000000 5000000 4000000 25 3000000
,l
2000000 1000000
10.00
20.00
30.00
I
I
I
I
40.00
50.00
60.00
7o.00
(a)
TIME
80000070000025
600000-
27
500000400000-
300000200000-
100000-
I
I
I
I
10.00
20.00
3O.00
40.00
(b)
I
50.00
60.00
70.00
TIME
Fig. 7. Alkyl fraction mass chromatograms (m/z 69 + 70 + 71) for Taranaki Basin coals: (a) Mangahewa-1 (11 811 ft); (b) Kapuni-8 (13 273.5 ft).
6b, 7b). These differences in the distributions of aliphatic groups in the pyrolysates parallel the differences in the n-alkane compositions of the bitumens (Table 4a) and further reflect the difference in organic facies between the wells.
The aliphatic character of the coal and kerogen pyrolysates is consistent with the other geochemical data. For example, the gas chromatographic-amenable fractions of the total extracts of the coals are dominated by long-chain
158
D. J. CURRY E T A L .
normal alkanes (e.g. Fig. 3), which indicates a significant aliphatic component for the coals. The H/C ratios are higher than those observed in many other coals and terrestrial kerogens and imply there is an aliphatic-rich component with a high H/C ratio and a low O/C ratio in the Taranaki coals which may not occur in coals of the same maturity with lower H/C and higher O/C ratios. The occurrence of relatively high concentrations of long-chain aliphatic groups in these coals is supported by CP/MAS 13C-NMR analyses of Eocene coals from the Taranaki Basin (Collen et al. 1988). Although CP/MAS aaC-NMR data do not show chain length, they can give a quantitative estimate of the types of carbon present. These data show that coals from the Kapuni Group contain significant concentrations of polymethylene carbon (i.e. (-CH2)x, x > 4 ) , as well as of branched and terminal methyl groups.
Taranaki Basin: discussion The geological and geochemical data indicate that these coals and kerogens are of predominantly higher plant origin, with little or no algal or bacterial contributions. Consequently, the long-chain aliphatic groups which occur in them are also of higher plant origin. The most probable sources of these long-chain aliphatic groups are the highly aliphatic biopolymers which occur in cuticular and protective tissues of several species of higher plants (Nip et al. 1986, 1987; Tegelaar et al. 1989; see also Collinson et al., this volume). These highly aliphatic biopolymers are characterized by high concentrations of long-chain normal aliphatic groups organized in insoluble structures within the cuticular or protective tissues. The role of cuticular-derived biopolymers are precursors to the aliphatic carbon groups in the Taranaki Basin coals is indicated by several factors, most particularly the pyrolysis-gas chromatography results. The molecular weight distributions of the long-chain aliphatic groups in these coals are very similar to those of the highly aliphatic biopolymer constituents of cuticular material. For example, the distributions of normal alkenealkane pairs in the coal pyrolysates (e.g. Figs 7a, 7b) are very similar to the distributions in the pyrolysates of highly aliphatic biopolymers in cutin isolated from cuticular material from a number of modern and fossil sources, including A g a v e a m e r i c a n a (Nip et al. 1986, 1987; Tegelaar et al. 1989). In these pyrolysates, the relative concentrations of the n-alkenes and n-alkanes are generally uniform from approximately n-C14
to n-C27, above which they decrease rapidly. This uniformity implies that relatively long aliphatic chains are present in the coals and biopolymers which are being cracked essentially at random sites along the carbon chain during pyrolysis. The rapid decrease in the relative concentrations of the n-alkenes and n-alkanes above C27 implies that the aliphatic chains incorporated in the coals and biopolymers are approximately of that chain length. The variations in organic facies between the Kapuni-8 and Mangahewa-1 coals indicate that there are probably several different palaeofloral assemblages which can give rise to aliphatic-rich coals. In addition, the relatively small angiosperm content in some of the coals (indicated by the low concentrations of oleanane) implies that the precursor palaeofloras of these coals probably contain a significant gymnosperm input. This is consistent with the work of Thomas (1982) for Australian coals and Shanmugam (1985) for New Zealand coals. As demonstrated by Nip et al. (1986) and Tegelaar et al. (1989), not all higher plants contain highly aliphatic biopolymers. The restricted occurrence of the higher plant-derived highly aliphatic biopolymers implies that not all coals contain these materials and that occurrences of the palaeoflora which can give rise to high concentrations of aliphatic groups in coals are controlled by evolutionary trends and palaeoecology. These factors, in turn, will restrict the occurrence of these types of aliphatic-rich coals both geographically and temporally. This is consistent with extensive published data (e.g. Tissot & Welte 1984), which indicate that many coals and terrestrially derived kerogens, especially from the Palaeozoic of the United States and northwestern Europe, contain very low concentrations of long-chain aliphatic groups and are not capable of generating oil. In addition, cutin and cuticular-like material may not be the only source of highly aliphatic biopolymers in terrestrial (higher) plants. Recent work by Khorasani & Michelsen (1991) on Jurassic coals from Svalbard has demonstrated the oil-generation potnetial of suberinite and subereous components. Their pyrolysis-gas chromatographic and hydrous pyrolysis data, however, indicate that the distributions of aliphatic groups in these constituents (which are also highly aliphatic biopolymers) are different from those observed in the cuticularderived biopolymers, with abundant aliphatic groups extending only to the range of n-C13 , rather than n-C27. The available data indicate that coals containing high concentrations of highly aliphatic
ALIPHATIC-RICH COALS FROM AUSTRALIA AND NEW ZEALAND biopolymers related to the cutin-derived material studied by Nip et al. (1986, 1987), and Tegelaar et al. (1989) appear to occur mostly in Cretaceous and younger sediments, chiefly in the Southern Hemisphere, while potentially oilgenerative coals containing suberinite-like material can occur in sediments as old as the Jurassic (Curry, in prep.).
Cooper Basin Regional geology and sample description The Cooper Basin is a Permo-Triassic intracratonic basin located in south-central Australia. The Cooper Basin is unconformably overlain by the Jurassic Eromanga Basin. During the Permian the basin was filled by a series of lacustrine and fluvial deposits. For a more complete discussion of the basin see Heath (1989). The samples in this study are from the Lower Permian Patchawarra Formation, except for one coal from the Upper Permian Toolachee Formation. The Lower Patchawarra Formation has been divided sequence stratigraphically into a series of four units designated (from top to bottom): Brown, Yellow, Blue, and Red, which represent successive basin-filling events. The coals occur both at the top of and within the sequences. The Patchawarra coals were deposited in cold weather, ombrotrophic, blanket peat bogs dominated by primitive, lowlying vegetation. The bogs were subjected to repeated flooding and desiccation as the water levels rose and fell. More detailed discussions of the origins of the coals in the Cooper Basin are contained in Hunt (1989) and Hunt & Smyth (1989). The samples used in this study are from cores taken from the Gidgealpa-5 and Gidgealpa-6 wells. A list of samples, depths, and stratigraphic units is contained in Table 1.
Geochemical characterization The coals from these wells appear to be early mature. The levels of maturity are difficult to determine, however, since measured Rv(max) values vary from 0.83-0.98% over a narrow depth interval (Table 2). This variation is probably due in part to the extensive alteration and degradation of the primary organic matter (see below). The Cooper Basin coals are characterized by high concentrations of inertinite, which can be common in Permian Gondwanaland coals. The
159
concentration of inertinite ranges from 38% to almost 70% while the concentration of the liptinites is very low (Table 2; Fig. 1). High concentrations of inertinite are usually indicative of severe degradation, including subaerial exposure and desiccation. Although these coals contain high concentrations of inertinite, their elemental compositions are similar to those of many other humic coals and Type III kerogens (Table 2; Fig. 2) (Durand & Espitalie 1976). As with the Taranaki Basin coals, the Cooper Basin coals cover a broad range on the Van Krevelen diagram (Fig. 2). Since inertinites generally have low hydrogen concentrations, these values imply that the non-inertinitic fractions of the coals are relatively hydrogen rich. Although, as noted by Powell et al. (1991), correlation of maceral compositions with geochemical parameters such as I-I/C ratios is generally difficult, there appears to be an approximate inverse correlation of the H/C ratios with the inertinite trend for these coals (Fig. 8a). Extrapolation of the H / C inertinite correlation (Fig. 8a) indicates that these inertinites have a nominal IMC ratio of roughly 0.5 while the non-inertinitic material has a nominal H/C ratio of approximately 1.0. The Rock-Eval HI values for the coals range from 116 to 243mgg -a OC, which is high for inertinite-rich material. Similarly high values for Cooper Basin coals and shales were also noted by Powell & Boreham (1991) and Powell et al. (1991). More significantly, there is also an approximate inverse correlation of HI values with the inertinite concentration (Fig. 8b). Extrapolation of this correlation indicates that the inertinite has a nominal HI value of approximately 35mgg -~ OC, while the non-inertinite has a nominal HI value of approximately 350mgg - 1 0 C . In addition, the Rv(max) values generally tend to decrease with increasing HI (Fig. 8c) and decreasing inertinite content. The bitumen total extracts are all very similar and all contain some unusual characteristics (e.g. Fig. 9). For example, although the extracts contain fairly high concentrations of n-alkanes, they are not waxy, having relatively low concentrations of n-C2s+ components (Table 4b). The distributions of normal alkanes in the extracts of these coals more closely resemble the normal alkane distributions in microbial (i.e. algal and/or bacterial) organic matter than in terrestrially derived organic matter (e.g. Tissot & Welte 1984). The coal extracts also have a wide range of pristane/phytane ratios, ranging from 2.04 to 5.35 (Table 4b). The values below 3.0 are very low for humic coals in general and could also be indicative of a significant algal
160
D . J . CURRY E T A L .
70-
R
60-
iE
B T B iE
Y y
z 50-
R
B
40-
30-
20 0.6
(a)
0.64
'
0.t68'
0.u72
'
0.176
018
'
t
0.m84
t
0.188
H/C RATIO
240
220
200
R
Y
_ 00~
180
.~
160 -
-
-
140 iB
120-
B
100 30
(b)
% INERTINITE HI vs Rv ( m a x )
0.98
R
0.96 -
B
R
0.94 0.92 IR
0.9
Y 0.88
IE
iE ~>
0.86 B
T
0.84
W
IE W
0.82 0.8 0.78 0.76
(c)
I
100
i ....... I ........ i " 120 140
I
I 160
I HI
I 180
I
|
200
I
I
220
....I "
I
240
161
ALIPHATIC-RICH COALS FROM AUSTRALIA AND NEW ZEALAND _
/11
C1-N'~N'~ I
C/TiNAPH
i\[I
G1-PHEN
PHENV/~ tO
I~
0.00
'
I
5.00
'
I
10.00
'
I
15.00
'
I
20.00 MINUTES
'
I
25.00
'
I
30.00
'
I
35.00
'
40.00
Fig. 9. Representative total extract chromatogram: Yellow coal (7440.67 ft), Gidgealpa-6 well, Cooper Basin (see Table 6 for explanation of symbols).
contribution to the source of the labile (noninertinitic) organic matter. The bitumens also contain abundant aromatic components (e.g. Fig. 9). Although the distributions of aromatic constituents relative to one another do not vary significantly among the samples, the concentrations of the aromatic constituents relative to the aliphatic constituents tend to be higher in the lower-HI coals. Methylated naphthalene homologues are the largest individual components in the coal extracts, which also contain high concentrations of phenanthrene and methyl phenanthrenes. Interestingly, the coal extracts also contain fairly high amounts of fluorene, substituted fluorenes, and other aromatic species (Fig. 9). The compositions of the triterpane biomarkers are very uniform for the Cooper Basin coals, both qualitatively and quantitatively (Appendix: Tables 4b, 5b; Fig. 10a). The abundances of the extended hopanes (especially the C32+ homologues) are low compared to the C30 hopane, characteristic of non-marine deposition. No occurrence of 25,28,30-trisnorhopane, 25,28,30-trisnormoretane, 28,30-bisnorhopane, and 28,30-bisnormoretane were
observed in the coals (Jenkins 1989). The occurrence of the C30 rearranged hopane (diahopane, also known as Compound X) at these relatively low levels of maturity implies an oxidizing depositional environment (Moldowan et al. 1991). As in the Taranaki coals, the C27 Ts/Tm hopane ratios and the absolute concentrations of triterpanes are significantly higher in the coals than in the clastic-rich intervals (Tables 4b, 5b). The compositions of the sterane biomarkers are also very uniform among the coals (Tables 4b, 5b; Fig. 10b). Although the C29 steranes are the dominant homologues, indicating nonmarine organic matter, there are also moderate concentrations of C27 steranes, indicating the possible occurrence of algal-derived organic matter. These coals also contain relatively high concentrations of diasteranes (Table 5b; Fig. 10b). As with the Taranaki Basin coals, the sterane rearrangement reactions were probably catalysed by highly acidic peat swamp conditions. The hopane/sterane ratios for the Cooper Basin coals (Table 4b) range from 0.96 to 7.51, with most of the values in the vicinity of 3 to 5.
Fig. 8. Cooper Basin geochemical correlations (see Table I for explanation of stratigraphic symbols). (a) H/C ratios versus % inertinite; (b) HI versus % inertinite; (c) HI versus Rv(max).
162
D. J. CURRY ET AL.
100.0
30
29 191 27
!
1000
(a)
I
1100
'
I
'
1200
'i ..... 1300
'
I
Like.__;
1400
'
I
'
I
'
I
'
1500
1600
t 700
1800
1500
1600
1700
1800
SCAN
100.o
29 29
29
29D
217
tD
I
1000 (b)
1100
1200
1300
1400 SCAN
Fig. 10. Representative biomarker distributions from Cooper Basin coals (see Appendix: Table 6 for explanation of symbols): (a) triterpanes (m/z 191) - Yellow coal (7440.67 ft), Gidgealpa-6 well; (b) steranes (m/z 217) Intra-brown coal (7429 ft), Gidgealpa-6 well.
163
ALIPHATIC-RICH COALS FROM AUSTRALIA AND NEW ZEALAND Me-N 90000001 A
P
A
Me-Ph
/ 2O
8000000-
M-N
7000000-
1-Prl
6000000-
/
5000000-
25
4O00000 3000000 2000O0010000000
,
0
10.00
,oooooo1
,
,,
20.00
I
30.00
40.00 TIME
(a)
50.00
I
60.00
70.00
80.00
70.00
80.00
Me-N
80000001 Me-Ph
70000006000000-
i li
5000000-
1-Pr
400000030000002000000 1O00000 0 0
10.00
20.00
30.00
(b)
I
40.00
1
50.00
I
60.00
I
TIME
Fig. 11. Representative pyrolysis-gc/ms total ion chromatograms for Cooper Basin coals (see Appendix: Table 6 for explanation of symbols): (a) high HI: Yellow coal (7440.67 ft), Gidgealpa-6 well; (b) low HI: Blue coal (7489 ft), Gidgealpa-6 well.
There are no obvious correlations of these ratios with other indicators of organic facies such as pristane/phytane ratios, inertinite concentrations, or sterane distributions. While the hopane/sterane ratios of the Cooper Basin coals
are generally in the range of those characteristic of non-marine organic matter, they are significantly less than the hopane/sterane ratios of the Taranaki Basin coals (Table 4a) and other terrestrial organic facies (Isaksen 1991).
164
D . J . CURRY ET AL.
A l i p h a t i c ca r b o n content
Pyrolysis-gc/ms analyses show that long-chain aliphatic groups (>n-C12) constitute a significant portion of the thermally labile components of the Cooper Basin coals (e.g. Figs l l a , b). Powell et al. (1991) reported that the concentrations of aliphatic carbon in the Patchawarra coals are at the low end of the range of coals and kerogens from the Permian through the Tertiary in Australia. However, since those concentrations are normalized to the total organic carbon content of the coals and shales, they do not take into account the presence of high abundances of inertinite. Since high concentrations of inertinitic components tend to dilute the TOC-normalized concentrations of thermally labile material, comparison of TOC-normalized yields of thermally labile constituents from lowversus high-inertinite coals and kerogens may be somewhat deceptive. The effects of dilution of organic carbon by inertinitic material may be one of the reasons that it has been difficult to determine the minimum yields of aliphatic carbon necessary for a coal (or kerogen) to be oil generative (Powell & Boreham, this volume). The relative compositions of the aliphatic constituents by molecular weight are very similar for all the coals, regardless of sequence stratigraphic unit, HI values, or inertinite content. For example, the distributions of longchain aliphatic groups are similar for a high-HI coal from the Yellow sequence (Fig. 12a) and a low-HI coal from the Blue sequence (Fig. 12b). In the Cooper Basin coals, the relative concentrations of the long-chain aliphatic groups decrease rapidly with increasing molecular weight, so that the abundances of the n-C2o+ constituents are much lower than those of the lower molecular weight aliphatic groups. Similar molecular weight distributions of aliphatic constituents in pyrolysates of high-inertinite Patchawarra coals were noted in Powell et al. (1991). However, the distributions by molecular weight of the aliphatic components in the pyrolysates of the Cooper Basin coals are significantly different from those observed in the Taranaki Basin. These differences in the distributions of the long-chain aliphatic groups can be observed in the alkyl fraction mass chromatograms (Figs 7a and b compared to Figs 12a and b). In addition, the distributions of aliphatic constituents in the Cooper Basin coal pyrolysates are also different from those of other aliphaticrich coals, such as the cutinite-rich, Carboniferous Indiana paper coal (Nip et al. 1989; see also Collinson et al., this volume) or coal from the Sydney Basin (Curry, in prep.).
The pyrolysis data are consistent with the other data indicating a fairly high abundance of aliphatic carbon despite the high inertinite concentration. As discussed, the relation of H/C ratios and HI values for the Patchawarra and Toolachee coals to the inertinite concentrations indicates that these coals contain a relatively hydrogen-rich, aliphatic-rich component. The compositions of the aliphatic constituents in the Cooper Basin pyrolysates are also consistent with the n-alkane distributions in the bitumen extracts. As noted above, the bitumen extracts from these coals are also relatively alkane rich but with relatively alkane rich but with relatively low abundances of n-C20+components.
N o n - a l i p h a t i c constituents
The similarity among the pyrolysate compositions is also observed for the non-aliphatic constituents. The pyrolysates contain the same relative concentrations of substituted benzenes, phenols, and substituted naphthalanes (Figs 11a, b). In addition to these constituents, the pryolysates also contain anomalously high amounts of more unusual species, including abundant trimethyl phenols, methyl phenanthrene and methyl anthracene isomers (usually present only in small concentrations in pyrolysates), biphenyls, fluorine and substituted fluorenes, as well as other aromatic, naphthenoaromatic, and polar compounds (see below).
Cooper Basin: discussion D e g r a d a t i o n a n d alteration
The geological data (Hunt 1989; Hunt & Smyth 1989) indicate that higher plant-derived organic matter was probably the major source of the primary organic deposition. Many of the geochemical parameters, including the petrographic data, also indicate a mainly terrestrial origin for the primary orgnaic matter in the coals. However, the occurrence of moderate amounts of C27steranes, the relatively low pristane/phytane ratios in some of the coals, and the lower hopane/sterane ratios, imply the co-deposition of algal-derived organic material. However, many of the geochemical parameters indicate that much of the primary organic matter has been extensively altered by the effects of oxidation and bacterial/fungal action. The most significant indicator of these processes is the occurrence of high concentrations of
165
ALIPHATIC-RICH COALS FROM AUSTRALIA AND NEW ZEALAND 450000-
10
400000350000300000-
20 250000200000150000100000" 50000-
~
a
0 1o.oi
I
I
20.00
30.00
(a)
I
40.00 TIME
I
50.00
60.00
I
70.00
35000010 30000015
i
250000200000150000100000
-
~
50000-
0
I 10.00
I
[
I
I
I
70.00 50.00 60.00 40.00 (b) TIMF Fig. 12. Alkyl fraction pyrolysis-gc/ms mass chromatograms (m/z 69 + 70 + 71) for Cooper Basin coals ((a) and (b) as in Fig. 11). 20.00
30.00
inertinite in all the coals, which is generally characteristic of severe subaerial exposure and oxidation. The compositions of the pyrolysates also appear to indicate severe subaerial oxidation and degradation. The relative concen-
trations of aromatic constituents such as phenanthrene and methyl phenanthrenes in the pyrolysates of these coals are generally much higher than observed in pyrolysates of most coals (e.g. Figs 6a, b). In addition, as noted, the
166
D . J . CURRY E T A L .
pyrolysates also contain, as major components, unusual constituents which are not normally observed in significant amounts in coal bitumens. These compounds include significant amounts of fluorene, methyl fluorenes, a set of isomeric compounds tentatively indentified (on the basis of retention time and mass spectra) as hydroxy-fluorenes (component F1 in Figs l l a , b; see Fig. 13 for the mass spectrum of one of these isomers) and possibly substituted chrysenes. Similar concentrations of these constituents, which generally occur only in low abundances in most coal and kerogen pyrolysates, have also been observed in pyrolysates of high-inertinite coals from the Haltenbanken area of the North Sea (Curry, in prep.). The occurrence of these components, especially the bridged aromatic compounds, in such high relative concentrations may possibly be due to formation by oxidative coupling and crosslinking. The composition of the extractable organic matter from the coals are very similar to that of the pyrolysates and also suggests extensive degradation. The bitumens also contain high relative concentrations of phenanthrene, methyl phenanthrenes, and many of the uncommon higher molecular weight aromatic compounds observed in the pyrolysates (Fig. 9). Even though these compounds are generally present in low abundances in oils and extracts, elevated concentrations of fluorenes have been observed
in severely biodegraded oils and tar mats from the North Sea (S. Larter, pers. comm.). In addition, as noted above, the presence of appreciable concentrations of the C30 rearranged hopane at these relatively low levels of maturity is consistent with an oxidizing depositional environment. The maceral distributions and the compositions of both the pyrolysates and the bitumens imply that the organic matter which formed the Cooper Basin coals was severely oxidized and degraded. The inverse relationships of the H/C ratios and HI values with the concentrations of inertinite (Figs 8a, b) suggest that the values of these parameters were significantly influenced by the same processes which led to the formation of inertinite. In addition, the wide range of Rv(max) values, together with the trend of decrease in measured Rv(max)with increasing HI (Fig. 8c), suggest that the optical properties of the coals were also affected by the same depositional processes which influenced the H/C and HI values. In addition, the abundances of the aromatic species discussed above, especially phenanthrene, methyl phenanthrenes, and fluorene species, decrease relative to the abundances of the normal aliphatic groups with increasing HI values and increasing H/C ratios. However, the co-variation of various parameters such as H/C ratios, maceral composition, Rv(max) values, and HI values is not normally observed in coals (e.g. Powell &
Abundance i~I
180000
160000
140000
120000
IOOO00
80000
60000 152 40000
~9
20000
63
76
w
I
si
169
91
iis
I//
i28
95 102
13 1 4 5
I
Ill
9
190
200
190
200
0
~/z
->
30
40
50
60
70
80
90
i00
ii0
120
130
140
150
160
170
180
210
220
Fig. 13. Mass spectrum of component F1 (see Figs 11a, b) in the pyrolysate of a coal from the Blue sequence.
ALIPHATIC-RICH COALS FROM AUSTRALIA AND NEW ZEALAND Boreham 1991). Together, the interrelationships of these parameters indicate that the oxidation and degradation processes which led to the formation of the inertinite have also been the major control on the abundances and compositions of the thermally labile organic matter in these coals. These parameters all have fairly broad ranges, which implies considerable variability in the degree of alteration among these coals. However, coals in the Yellow sequence generally have the highest H/C ratios and HI values and lowest Rv(max)values, while the coals in the Blue sequence have the lowest H/C and HI values and highest Rv(max)values (Figs 8a, b, c). In addition, in general, the Yellow sequence coals have the lowest, and the Blue sequence coals have the highest, abundances of fluorenes and other unusual aromatic components relative to the normal aliphatic groups in the pyrolysates (Figs l l a , b). These distributions suggest that the extent of oxidation and degradation was the least severe in the Yellow sequence and most severe in the Blue sequence. This differentiation in the severity of degradation implies that the extent of the degradation was in part controlled by sequence-related variations in depositional environments. In this regard, the approximate correlation trends in Figs 8a, b and c are also plots of the severity of degradation of these coals.
Origin o f aliphatic carbon The long-chain aliphatic groups which occur in the pyrolysates and extracts of these coals were probably not part of the primary (i.e. originally deposited) terrestrial organic matter. As noted above, the severe degree of degradation indicates that the primary organic matter has been extensively altered. In addition, the relative compositions of the aliphatic constituents of the pyrolysates and extracts are very different from those of other aliphatic-rich terrestrial organic matter, such as the Taranaki Basin coals, Carboniferous paper coals from Indiana (Nip et al. 1989) or Jurassic coals from Svalbard (Khorasani & Michelsen 1991), and in fact more closely resemble the aliphatic distributions observed in microbially derived organic matter. The aliphatic groups in the Cooper Basin coals instead very possibly originated from deposition of lipid-enriched biomass resulting from the extensive alteration and degradation of the coal precursors by microbial action. Bacterial and fungal attack of sedimented organic matter has been demonstrated to significantly increase the
167
lipid-rich biomass in sediments. For example, Fogel et al. (1989) showed that bacterial and fungal action significantly alters the composition of sedimented fragments of Spartina grass in the top 50-100cm of sediment in a Georgia salt marsh. Of special interest was the large increase in the abundance of n-C13 to n-C19 aliphatic groups in the pyrolysates of the Spartina fragments relative to the abundances of aromatic and phenolic compounds and n-C22 to n-C27 aliphatic groups with increasing burial. This distribution is very similar to the distribution of constituents in pyrolysates of the Cooper Basin coals (Figs 12a, b). Other analytical data are consistent with the occurrence of aliphatic carbon derived from microbial activity. For example, the H/C ratios and HI values are much higher than would be expected for inertiniterich, liptinite-poor coals, suggesting the occurrence of a hydrogen-rich constituent in the coals which is apparently too small and too finely dispersed to be readily visible by optical methods, but which may ba a source of labile organic matter. This concept is supported by data from Taylor & Liu (1987), Taylor et al. (1988) and Liu & Taylor (1991). From the results of transmission electron microscopy, they suggest Cooper Basin coals contain a substantial amount (5-15%) of 'alginite and bacterially altered alginite' that is too small to be observed by conventional microscopy. The exact mode of association of this finely divided material (i.e. micro-liptinite) with the other macerals is unclear. Although the association of the micro-liptinite with the inertinite has been suggested (Taylor et al. 1988), this is as yet equivocal. As noted, the decrease in HI values, H/C ratios, and apparent aliphatic content with increasing inertinite content indicates that the inertinite is not particularly enriched in aliphatic groups. Instead, aliphatic-rich micro-liptinite may be dispersed throughout the coal, either as discrete particles too small to be observed by conventional microscopy, as material filling pores in the inertinite, or as thin films and coatings in and on the conventional macerals. Primary algal material co-deposited with the terrestrial organic matter is probably not as significant a source of longer-chain aliphatic groups in these coals. The relatively high pristane/phytane ratios and C29 sterane concentrations imply that there was probably only a limited amount of primary aquatic input. In addition, because aquatic organic matter is more diagenetically labile than terrestrial organic matter, it is unlikely that it would be preserved preferentially during degradation.
168
D . J . CURRY E T A L .
From these data it appears that the aliphatic groups in the Cooper Basin coals originated in the main from biomass generated during the bacterial/fungal degradation of the precursor organic matter of these coals. During the initial stages of degradation, the addition of bacterial biomass would increase the aliphaticity in the coals, but probably not severely affect parameters such as pristane/phytane ratios and Rv(max). However, as the severity of degradation and alteration increased, not only would the secondary aliphaticity be reduced, leading to the observed decrease in aliphatic versus aromatic concentrations in the pyrolysates, but many other indicators of the primary organic input would also be affected. The preservation of the micro-liptinite may be a consequence of the high-latitude, peri-glacial depositional environments. As suggested by Taylor et al. (1989), the cold weather environments may have initially retarded the degradation of the organic matter when the peat was subaerially exposed, and so acted to partially preserve the secondary biomass created during the original degradation process. This model is supported by the data which show that the severity of degradation is qualitatively related to the different depositional sequences. That is, the lesser extent of degradation in the Yellow as compared to the Blue sequence implies that conditions were less harsh during deposition of the Yellow sequence coals. The extent of microbial alteration will depend to a large degree on depositional conditions such as water movement and drainage patterns, vegetation, pH, ell, and fluctuations in the water table, which are generally localized in extent. As a result, the formation of this type of aliphatic-enriched organic matter will probably be controlled more by local than by regional or temporal factors. This type of organic matter can occur in a wide range of geographical settings and ages. However, because the conditions favourable for their formation are localized, the areal and stratigraphic extent of deposits of this type of aliphatic-enriched coal and kerogen will generally be smaller and more discontinuous than those of cuticular or maceral-specific aliphatic-rich coals. In addition, the composition of this type of coal and terrestrial kerogen can vary significantly over very short intervals.
Conclusions (1) The long-chain aliphatic groups in coals and terrestrial kerogens can originate from primary constituents of the precursor plant material and can be incorporated directly
(2)
into the kerogen/coal matrix during early diagenesis. The long-chain aliphatic groups probably occur primarily in liptinitic material, especially cuticular or suberinitic material. The variability in the geochemical parameters such as H/C ratios and HI values is probably in part the result of the incorporation of varying proportions of aliphaticpoor (woody) and aliphatic-rich (cuticular or suberinitic) higher plant fragments. Since the deposition of these coals depends on the occurrence of specific higher plant precursors, their distribution will be restricted by evolutionary and palaeoecological factors. However, because these controlling factors generally operate over wide areas, when they do occur, these types of aliphaticrich coals will tend to have a fairly broad regional or basin-wide distribution. Available data indicate that this type of aliphaticrich coal is generally rare prior to the Mesozoic and occurs most commonly (although by no means exclusively) in southeastern Asia, Australia, and New Zealand. It is proposed to term this type of organic matter 'Type IIIC' (for cuticular and cork). This classification term is similar to the terminology proposed by Horsfield (1984), although the proposed mechanism for the incorporation of long-chain aliphatic groups into the coals and kerogens is substantially different from the mechanism suggested there. The long-chain aliphatic groups in coals can also originate from lipid-rich, microbially derived organic matter. As shown by the Cooper Basin coals, long-chain aliphatic groups derived from bacterial biomass can originate as secondary products during the early stages of diagenesis and have little relation to the primary organic input. Longchain aliphatic groups can also be the result of primary co-deposition of aquatic organic matter concomitant with the deposition of the terrestrial material. Since the formation of this type of aliphatic-enriched coal is controlled mainly by localized depositional conditions, these coals can occur in a wide variety of geographical and stratigraphical settings. However, their occurrences within an area will be more localized than Type IIIC coals due to the specialized conditions required for their deposition and preservation. In addition, because of the transitory character of their depositional environments, the compositions of these coals will tend to be much more variable than Type IIIC coals. It is proposed to term this type
ALIPHATIC-RICH COALS FROM A U S T R A L I A AND NEW Z E A L A N D o f o r g a n i c m a t t e r ' T y p e I I I M ' (for microbial origin). (3) In a d d i t i o n , b o t h t h e s e processes ( i . e . deposition of aliphatic-enriched macerals a n d o f lipid-rich microbial biomass) can o c c u r s i m u l t a n e o u s l y in the s a m e d e p o s i t i o n a l setting. (4) T h e r e l a t i o n s h i p of aliphatic c o m p o s i t i o n to specific p r e c u r s o r s a n d / o r specific d e p o sitional c o n d i t i o n s also implies t h a t m a n y coals will n o t c o n t a i n high c o n c e n t r a t i o n s o f long-chain aliphatic groups and will n o t be oil p r o n e .
169
We would like to thank Steve Oliveri, Jill Kerr, and Norma Booher for their technical support. Lisa Hering and Jan Herbst provided computer support and assistance with drafting. Tom Loutit obtained the samples of the Taranaki Basin coals and shales from the New Zealand Geological Survey. We thank Brian Burns of Esso Australia and Erik Sandvik of EPR who provided valuable discussions and suggestions. We would also like to thank Brian Horsfield and an anonymous reviewer for their helpful critiques of an earlier version of this work. We thank Exxon Production Research Company, Esso Australia Ltd, and Delhi Petroleum Company and their partners for permission to publish this work.
Appendix Table 1.
Sample identification and stratigraphy
Taranaki Basin
Cooper Basin
Well
Sample no.
Depth (ft)
Lith.
Well
Sample no.
Depth (ft)
Lith.
Unit
Symbol
Mangahewa-1
112255A 112255B 112255C 112255D 112255E 112255F 112255G 112255H 1122551 112255J 112255K 112255L 112255M
11802 11803.8 11804.5 11805.5 11806.5 11807.5 11808 11808.5 11809.5 11810 11811 11811.85 11812
S S C C C S C C C C C S S
Gidgealpa-5
131384A 131384B 131384C 131384D 131384E 131384F 131384G 131384H 1313841
7061 7064.67 7103 7113 7129.67 7134.67 7137.92 7171.5 7192.33
C C C S CS C SS CS C
Yellow Yellow Intra-Yel YelSilt Blue Blue Blue Sand Intrablue Red
Y Y iY Yd B B Bs iB R
Gidgealpa-6
112266A 112266B 112266C 112266D 112266E 112266F 112266G 112266H 1122661 112266J 112266K 112266L 112266M 112266N 1122660
13261.5 13269.25 13271.5 13272.5 13273.5 13275.3 13276.5 13280.5 13283.5 13289 13292.8 13293.8 13294.5 13302.68 13305
S C S CS C S CS S S S S C C S S
131395A 131395B 131395C 131395D 131395E 131395F 131395G 131395H 1313951 131395J 131395K 131395L 131395M
7176 7380.25 7421.83 7429 7440.67 7464.75 7482.5 7489 7515 7516 7529.5 7535.5 7543
C C C C C C C C S S S C C
Toolachee Intrabrown Intrabrown Intrabrown Yellow Intra-Yel Blue Blue Blue Sand Blue Silt Blue Sand Red Red
T iE iE iE Y iY B B Bs Bd Bs R R
Kapuni
Lithology - C: Coal (TOC > 50%); CS: Carbonaceous shale (25 < TOC < 50%); S: Shale (TOC < 25%); SS: Sandstone.
13261.5 13269.25 13271.5 13272.5 13273.5 13275.3 13276.5 13280.5 13283.5 13289 13292.8 13293.8 13294.5 13302.68 13305
Kap-1
S C S CS C S CS S S S S C C S S
S S C C C S C C C C C S S
Lith.
52.6 71.0 86.2 89.8 72.2
0.59 0.58 0.53 0.57 0.63
0.56 0.59
84.7
0.66
0.70
85.0 89.4 86.9 83.2
66.7 66.7 87.0 89.7 91.7 92.1 88.8 90.5 92.4 87.2 91.1
% Vitr
0.57 0.60 0.63 0.66
0.56 0.55 0.60 0.70 0.70 0.72 0.70 0.74 0.73 0.70 0.63
Total
0.59 0.62 0.65 0.69
0.64 0.74 0.73 0.75 0.73 0.77 0.75 0.71 0.65
Telovitr
R~(%)
46.8 19.4 12.1 10.0 16.7
12.0
12.1 10.0 12.5 16.6
28.6 22.2 12.1 5.4 4.7 4.7 7.0 5.4 4.1 7.7 7.7
% Lipt
0.6 9.7 1.7 0.3 11.1
3.3
2.8 0.6 0.6 0.2
4.8 11.1 1.0 4.9 3.6 3.2 4.2 4.1 3.5 5.1 1.1
% Inert 7061 7064.67 7103 7113 7129.67 7134.67 7137.92 7171.5 7192.33 7176 7380.25 7421.83 7429 7440.67 7464.75 7482.5 7489 7515 7516 7529.5 7535.5 7543
Gid-6
Depth (ft)
Gid-5
Well
Cooper Basin
Note: Maceral compositions normalized to 100% ; vitrinite reflectance measured as Rv (max).
11802 11803.8 11804.5 11805.5 11806.5 11807.5 11808 11808.5 11809.5 11810 11811 11811.75 11812
Man-1
Well
Depth (ft)
Taranaki Basin
Table 2. Vitrinite reflectance and maceral composition
C C C S CS C SS CS C
Lith.
0.98 0.97
0.96 37.0 35.3
38.1 34.2 38.7 33.8 34.3 59.0 46.2 24.3
39.9 31.1
0.91 0.89 0.85 0.83 0.87 0.88 0.89 0.83 0.86 0.96
11.1 3.5
29.3 41.9
2.6 5.3
3.8 5.3 6.2 4.3 3.8 5.8 5.1 6.6
8.5 18.3
8.8 8.1 7.2
% Lipt
41.1 39.7 50.4
% Vitr
0.83 0.86 0.82 0.85 0.85 0.90
Rv(%)
60.5 59.4
58.1 60.5 55.2 61.9 61.9 35.2 48.7 69.1
51.6 50.6
59.6 54.7
50.1 52.2 42.4
% Inert
171
A L I P H A T I C - R I C H COALS F R O M A U S T R A L I A A N D N E W Z E A L A N D
Table 3. Elemental composition and Rock-Eval data Rock-Eval
Elemental data Well
Depth (ft)
Taranaki Basin Man-1 11802 11803.8 11804.5 11805.5 11806.5 11807.5 11808 11808.5 11809.5 11810 11811 11811.75 11812 Kap-1
13261.5 13269.25 13271.5 13272.5 13273.5 13275.3 13276.5 13280.5 13283.5 13289 13292.8 13293.8 13294.5 13302.68 13305
%C
3.12 3.64 68.68 79.56 80.19 21.46 80.11 80.19 78.98 78.77 73.56 5.10 2.66 1.67 64.67 14.11 42.53 77.62 2.41 43.37 1.02 1.44 12.45 3.86 74.66 59.59 5.21 4.31
H/C
O/C
0.964 0.849 0.813
0.104 0.095 0.113
0.824 0.813 0.820 0.842 0.896
0.106 0.094 0.101 0.097 0.112
0.991
0.109
1.061 0.826
0.161 0.000
1.010
0.146
0.993 0.942
0.125 0.157
Cooper basin 7061 Gid-5 7064.67 7103 7113 7129.67 7134.67 7137.92 7171.5 7192.33
62.91 77.65 78.36 8.54 48.74 62.77 0.63 47.58 57.20
0.757 0.756 0.775 1.335 0.726 0.715
0.112 0.102 0.100 0.536 0.103 0.122
0.797 0.789
0.126 0.114
7176 7380.25 7421.83 7429 7440.67 7464.75 7482.5 7489 7515 7516 7529.5 7535.5 7543
68.79 75.75 69.47 65.98 75.24 51.38 74.30 52.17 0.50 1.94 1.03 68.40 75.36
0.727 0.700 0.725 0.773 0.601 0.827 0.738 0.665
0.139 0.096 0.104 0.108 0.109 0.173 0.078 0.128
0.674 0.645
0.124 0.103
Gid-6
Tma x
S1
$2
$3
HI
438 439 437 433 436 434 436 433 434 434 434 433 434
0.52 0.49 16.87 10.34 10.20 4.14 11.11 9.23 9.64 9.65 11.02 1.00 0.61
7.58 8.53 204.10 190.84 191.11 46.86 198.38 185.17 186.66 174.12 212.27 15.82 5.28
0.11 0.19 3.84 4.22 5.75 1.43 5.45 6.09 6.25 6.81 4.71 0.35 0.34
243 234 319 250 254 218 256 242 247 236 304 310 198
425 430 437 435 432 434 431 438 434 431 431 430 426 433 430
0.24 19.01 2.01 6.82 10.94 0.27 5.88 0.11 0.19 2.08 0.68 14.07 10.00 0.65 0.54
2.89 212.24 51.44 129.61 240.54 4.13 115.78 1.28 1.97 43.22 10.80 271.84 177.21 14.50 9.55
0.49 3.60 0.83 2.45 4.05 0.28 2.96 0.37 0.46 0.43 0.30 3.61 3.13 0.53 0.80
173 361 365 320 331 171 292 125 137 347 280 362 302 278 222
437 439 439 440 439 441 450 441 440
11.3 19.7 18.9 1.44 9.63 13.8 0.14 8.48 14.4
127.50 144.10 165.80 7.13 61.37 100.30 0.21 63.15 112.80
3.74 5.29 4.60 0.33 2.01 3.73 0.01 2.18 2.75
195 189 221 86 116 154 33 125 199
440 442 440 441 442 442 441 444 443 451 356 447 445
12.4 14.8 15.2 12.5 9.62 9.61 10.1 8.62 0.69 0.23 0.23 9.71 10.00
89.65 98.40 112.50 106.60 117.60 152.00 110.50 78.13 0.30 0.65 0.06 79.33 97.67
4.01 4.06 4.54 5.04 5.55 6.19 5.98 2.64 0.04 0.01 0.01 2.95 0.01
132 134 170 159 157 243 154 130 60 34 6 130 133
11802 11803.8 11804.5 11805.5 11806.5 11807.5 11808 11808.5 11809.5 11810 11811 11811.75 11812
13261.5 13269.25 13271.5 13272.5 13273.5 13275.3 13276.5 13280.5 13283.5 13289 13292.8 13293.8 13294.5 13302.68
Kap-1
Depth (ft)
Man-1
Well
GC data
S C S CS C S CS S S S S C C S
S S C C C S C C C C C S S
Lith.
8.19 6.29 8.11 11.01 11.51 9.54 9.62 7.63 9.79 9.17 8.50 9.87 10.38 9.94
7.90 7.82 10.25 12.70 12.29 8.70 10.83 11.24 11.03 11.18 11.01 6.46 4.17
Pris/phy
7.77 2.91 4.03 4.99 6.13 9.46 8.69 6.97 4.06 5.57 6.52 7.65 7.29 6.58
3.31 3.13 3.01 5.73 4.63 2.95 5.65 7.48 8.03 8.2 9.97 3.81 2.00
Pris/C17
1.11 0.68 0.66 0.57 0.66 1.00 1.12 0.91 0.5 0.72 0.98 0.97 0.74 0.74
0.41 0.39 0.33 0.45 0.37 0.37 0.5 0.65 0.66 0.67 0.82 0.5 0.36
Phy/Cla
Table 4a. Taranaki Basin coals and shales GC and biomarker ratios
1.59 1.59 1.57 1.60 1.61 1.68 1.6 1.56 1.44 1.66 1.75 1.54 1.61 1.54
1.29 1.39 1.83 1.63 1.52 1.59 1.36 1.58 1.43 1.42 1.66 1.46 1.31
CPI
2.50 1.48 2.44 3.77 3.85 3.23 3.14 2.44 2.45 2.77 1.93 3.33 3.13 2.69
2.32 2.01 1.51 1.73 2.43 0.72 2.09 2.01 2.78 2.62 2.35 2.33 1.84
C27/C17
0.735 0.569 0.608 0.679 0.671 0.673 0.692 0.653 0.625 0.660 0.608 0.704 0.690 0.689
0.621 0.547 0.438 0.455 0.549 0.374 0.524 0.546 0.546 0.542 0.538 0.617 0.527
% nC25_34alkanes
to
12.809 8.220 8.408 8.558 9.962 11.330 11.608 13.434 6.989 13.541 14.829 3.239 3.704
3.135 3.608
11802 11803.8 11804.5 11805.5 11806.5 11807.5 11808 11808.5 11809.5 11810 11811 11811.75 11812
13261.5 13269,25 13271.5 13272.5 13273.5 13275.3 13276.5 13280.5 13283.5 13289 13292.8 13293.8 13294.5 13302.68
Man-1
Kap-1
4.523 6.452 3.309 7.322 3.910 5.578
6.202 10.788 6.688 6.918
5.159
Hopanes/ steranes
Well
Depth (ft)
Biomarker ratios
0.251 0.522 0.117 0.161 0.116 0.101 0.045 0.133 0.091 0.147 0.127 0.274 0.054
1.101 0.992 0.846 0.826 0.759 1.052
0.052 0.075 0.073 0.062 0.027 0.026 0.027 0.023 0.042 0.035 0.054 0.275 0.177
0.750 0.831 0.803 0.817 0.841 1.298 0.851
0.967 0.994 0.916 0.891 0.885 0.9o3 0.816 0.877 0.769 0.763 0.852 0.898 0.845
C29/C3o Cao
Olean/
0.379 0.395 0.373 0.397 0.414 0.372
0.351 0.447 0.411 0.429 0.437 0.380 0,444
0.222 0.243 0.232 0.307 0.331 0.320 0.285 0.350 0.314 0.307 0.315 0,303 0.313
CaoM/ C3oH
1.043 1.355 0.938 2.162 1.676 1.035
0.942 0.995 1.089 1.676 1.825 0.818 1.771
0.987 1.097 1.134 1.099 1.235 1.295 1.339 1.693 2.180 2.284 2.302 1.249 1.372
C31S+ R/ C3o
Triterpanes (M/Z 191)
0.518 0.474 0.600 0.340 0.398 0.579
0.580 0.627 0.514 0.412 0.396 0.706 0.391
0.577 0.547 0.523 0.574 0.523 0.501 0.490 0.410 0.336 0.327 0.325 0.475 0.426
C3oH/ ext'd hop
0.422 0.428 0.363 0.544 0.497 0.361
0.416 0.348 0.437 0.455 0.472 0.290 0.478
0.449 0.451 0.494 0.433 0.443 0.473 0.509 0.510 0.613 0.612 0.601 0.469 0.499
Ext'd/total hopanes
0.863 0.944 0.918 0.878 1.000 1.000
0.960 0.893 0.908 0.802 0.788 0.895 0.880
1.000 0.924 0.868 0.851 0.894 0.899 0.922 0.880 0.910 0.898 0.910 1.000 0.919
0.29o 0.505 0.382 0.392 0.475 0.474
0.405 0.473 0.408 0.521 0.790 0.457 0.401
0.669 0.397 0,433 0,451 0.463 0.443 0.380 0.504 0.412 0.488 0.425 0.435 0.588
C29/total C290(0/0t5/ C29o(o~og5-t- R
steranes
o.596 0.702 0.614 0.649 0.517 0.571
0.492 0.589 0.665 0.725 0.561 0.634 0.667
0.440 0.684 0.623 0.653 0.649 0.685 0.640 0.656 0.666 0.653 0.669 0.540 0.421
total C29
C290/0t0/8+ R/
Steranes (M/Z 217)
o.566 0.604 0.486 0.609 0.596 0.515
0.450 0.578 0.585 0.584 0.463 0.537 0.600
0.812 0.589 0.660 0.667 0.679 0.725 0.699 0.715 0.792 0.848 0.643 0.553 0.497
C29 regl/ reg + dia
-.a
7061 7064.67 7103 7113 7129.67 7134.67 7137.92 7171 7192
7176 7380.25 7421.83 7429 7440.67 7464.75 7482.5 7489 7515 7516 7529.5 7535.5 7543
Gid-5
Gid-6
Well
Depth (ft)
C C C C C C C C S S S C C
C C C S CS C SS CS C
Lith.
3.97 4.16 3.61 4.81 4.06 5.35 2.85 2.43 . 0.82 . 2.04 2.84
4.27 4.81 3.58 2.15 2.75 4.83 1.42 4.09 4.62
Pris/phy
.
.
0.38 0.46
0.40
0.65 0.50 0.56 0.68 0.71 1.00 0.30 0.51
0.66 0.46 0.39 0.37 0.52 0.85 0.48 1.30 0.66
.
.
Pris/C17
0.17 0.13 0.18 0.15 0.19 0.19 0.11 0.21 . 0.45 . 0.20 0.18
0.16 0.10 0.11 0.19 0.20 0.19 0.33 0.33 0.15
Phy/Cla
.
.
Table 4b. Cooper Basin coals and shales GC and biomarker ratios
GC data
1.09 1.07
1.10
1.07 1.08 1.12 1.16 1.03 1.09 1.08 1.09
1.14 1.16 1.09 1.11 1.06 1.07 1.02 1.10 1.11
.
.
CP- 1
0.269 0.151
0.242
0.317 0.322 0.213 0.305 0.304 0.364 0.205 0.237
0.334 0.169 0.255 0.203 0.284 0.325 0.463 0.345 0.319
C27/C17
nC25_34alkanes
0.169 0.120
0.221
0.196 0.212 0.160 0.174 0.192 0.208 0.138 0.158
0.196 0.118 0.170 0.155 0.187 0.197 0.296 0.203 0.179
%
4~
7061 7064.67 7103 7113 7129.67 7134.67 7137.92 7171.5 7192.33
7176 7380.25 7421.83 7429 7440.67 7464.75 7482.5 7489 7515 7516 7529,5 7535.5 7543
Gid-6
Depth (ft)
Gid-5
Well
Biomarker ratios
4.306 3.741 4.007 3.857 3.895 3.822 4.712 4.493 1.613 1.697 2.093 0.956 7.510
4.620 3.878 4.302 3.986 5.947 3.722 2.242 4.170 5.513
Hopanes/ steranes
0.034 0.052 0.054 0.052 0.053 0.080 0.084 0.073 0.326 0.656 0.096
0.049 0.059 0.152 0.127 0.096 0.083 0.502 0.063 0.396
C27Ts/C27TM
0.673 0.626 0.642 0.638 0,590 0.535 0.572 0.658 0.497 0.832 0.614 0.540 0.626
0.741 0.655 0.545 0.786 0.607 0.610 1.124 0.663 0.482
C29I~C30H
0.065 0.083 0.060 0.056 0.055 0.054 0.093 0.064 0.000 0.074 0.073 0.000 0.063
0.072 0.063 0.050 0.070 0.054 0.055 0.062 0.074 0,070
C3oM/C3oH
Triterpanes (M/Z 191)
0.991 0.943 1.012 0.960 0.885 0.756 0.824 0.844 0.690 0.762 0.680 1.118 0.819
1.040 0.883 0.680 0.782 0.675 0.838 0.861 0.850 0.453
C31S + R/ C3o H
0.461 0.440 0.464 0.456 0.443 0.427 0.434 0.417 0.414 0.367 0.389 0.731 0.437
0.444 0.423 0.410 0.397 0.360 0.433 0.411 0.425 0.333
Ext'd hop/ tot. hop
0.402 0.419 0.450 0.421 0.422 0.491 0.475 0.464 0.676 0.417 0.524 0.490 1.000
0.439 0.458 0.467 0.463 0.404 0.455 0.441 0.455 0.535
C29aotaS/ S+ R
0.876 0.792 0.784 0.861 0.910 0.816 0.842 0.666 3.378 1.185 1.788 0.927 0.647
0.790 0.614 1.395 0.718 0.760 0.801 0.898 0.843 0.696
C29t~ott~Sdia/ reg steranes
Steranes (M/Z 217)
0.497 0.506 0.486 0.476 0514 0.475 0.445 0.449 0.280 0.435 0.356 0.517 0.418
0.484 0.501 0.412 0.466 0.526 0.482 0.473 0.469 0.528
Total C29/ total ster
-,,,,I
Lith.
S S C C C S C C C C C S S
S C S CS C S CS S S S S C C S
11802 11803.8 11804.5 11805.5 11806.5 11807.5 11808 11808.5 11809.5 11810 11811 11811.75 11812
13261.5 13269.25 13271.5 13272.5 13273.5 13275.3 13276.5 13280.5 13283.5 13289 13292.8 13293.8 13294.5 13302.68
Man-1
Kap-1
Well
Depth (ft)
0 178 77 0 67 454 141 1618 0 0 358 0 15
5670 10745 10528 13133 4157 6380
0 50 170 1261 227 251 0 193 0 0 217 0 0
?W*
6865 3778 9123 8077 10287 7072 15002
1727 4865 13653 19650 11072 14719 25893 11674 19725 24575 21506 2360 2013
Total hopanes
Triterpanes (ppm saturate fraction): M/Z 191
1022 2107 2951 2212 740 1717
1318 1057 1927 2002 2326 1870 3314
290 1001 2445 4546 2588 2920 4424 2196 2672 3149 3008 461 399
C27H + ? T ?
261 252 586 591 357 101
604 659 329 374 351 236 211
0 73 374 440 72 106 112 65 88 100 202 131 73
R
Table 5a. Taranaki Basin coals and shales quantitative biomarker data
870 356 725 505 645 729 884
1242 684 1645 1240 1619 1875 2388
417 1012 954 805 390 466
130 188 418 740 455 578 1189 528 550 1136 672 161 113
432 1191 3231 4350 2272 3145 5268 2142 3127 3749 3578 472 362
1366 2163 1941 2004 623 1401
C29M
C29H
165 198 337 309 225 73
416 430 239 244 223 147 125
23 90 259 301 70 89 177 56 172 173 226 144 76
Olean
1241 2180 2295 2428 821 1332
1657 823 2048 1517 1925 1445 2806
447 1199 3526 4883 2567 3484 6453 2443 4067 4913 4198 525 428
C3o H
471 860 857 964 340 496
582 368 842 651 841 550 1246
99 291 820 1500 849 1116 1840 855 1276 1508 1322 159 134
C3o M
1295 2955 2151 5247 1376 1378
1560 819 2231 2542 35130 1182 4968
441 1316 3999 5368 3171 4512 8644 4137 8864 11223 9666 656 587
C31 (S + R)
Ext'd
2395 4603 3825 7148 2065 2301
2854 1313 3983 3678 4856 2048 7169
775 2194 6740 8514 4909 6959 13172 5954 12100 15034 12927 1107 1005
triterp
13261.5 13269.25 1327.5 13272.5 13273.5 13275.3 13276.5 13280.5 13283.5 13289 13292.8 13293.8 13294.5 13302.68
Kap-1
471 124 226 110 87 133 294
161 103 395 170 106 187
1254 1665 3181 1794 1063 1144
0 53 108 203 111 57 57 35 97 0 57 131 87
D29
2190 1047 1768 1302 954 1058 2169
135 592 1624 2296 1111 1299 2231 869 2822 1815 1450 729 543
Total steranes
102 97 307 140 82 178
212 118 171 110 61 83 157
25 36 93 0 17 49 106 40 221 0 127 83 62
D29
106 337 333 244 156 159
189 151 255 231 154 147 306
32 87 251 384 203 256 349 181 558 441 241 95 61
29aaaS
247 283 548 336 306 253
481 222 315 168 153 186 382
61 102 350 452 237 267 517 188 679 480 280 186 144
29aflflR + S
259 330 539 379 172 177
277 168 370 212 41 175 458
16 133 329 467 235 323 571 178 797 461 326 123 43
612 950 1420 959 634 589
946 541 939 610 348 509 1146
109 322 930 1303 675 846 1437 547 2034 1382 848 403 248
C29oto~o~R C29reg/total
*, t U n k n o w n C3o pentacyclic triterpanes (possible ring-contracted oleananes) - (see text).
11802 11803.8 11804.5 11805.5 11806.5 11807.5 11808 11808.5 11809.5 11810 11811 11811.75 11812
Depth (ft)
Man-1
Well
Steranes (ppm saturate fraction): M/Z 217
470 622 1500 616 430 555
1157 394 666 434 404 438 763
25 225 479 651 319 321 620 218 534 248 472 326 251
C29 dias
1082 1572 2920 1575 1063 1144
2103 935 1606 1045 751 947 1908
135 547 1409 1953 993 1167 2057 765 2568 1631 1319 729 500
C29 reg + dias
7061 7064.67 7103 7113 7129.67 7134.67 7137.92 7171.5 7192.33
7176 7380 7421.83 7429 7440.67 7464.75 7482.5 7489 7515 7516 7529.5 7535.5 7543
Gid-6
Depth (ft)
Gid-5
Well
C C C C C C C C S S S C C
C C C S CS C SS CS C
Lith.
22750 9010 25948 18085 23663 27924 16673 22330 664 2006 2295 2447 12160
20191 13782 23620 9980 17999 13566 3343 13772 9630
Total
76 52 134 87 123 200 125 175 0 64 109 0 126
103 82 291 136 205 111 118 86 267
C27 Ts
2211 994 2359 1677 2296 2492 1478 2405 0 197 167 0 1310
2120 1390 1911 1072 2136 1331 234 1372 674
C27 TM
3530 1343 3904 2810 3540 4157 2494 3694 129 397 378 231 2008
3334 2183 3715 1867 3099 2133 795 2323 1599
C29 H
Table 5b Cooper Basin coals and shales quantitative biomarker distributions
Triterpanes (ppm saturate fraction): M / Z 191
221 113 372 242 287 502 261 387 96 44 101 97 227
223 155 387 132 230 234 22 254 224
C3o D
342 178 364 345 331 418 407 357 0 35 45 0 202
324 212 341 166 278 193 44 259 234
C29M
5245 2145 6083 4401 6000 7767 4358 5613 260 476 615 428 3206
4502 3332 6820 2375 5104 3497 707 3504 3317
C3o H
868 334 965 621 891 962 572 773 0 101 88 0 0
843 756 865 398 702 423 71 376 330
C3o M
10479 3965 12038 8243 10482 11928 7240 9312 275 735 892 1788 5309
8966 5828 9676 3966 6475 5878 1373 5852 3211
Ext'd hopanes
7061 7064.67 7103 7113 7129.67 7134.67 7137.92 7171.5 7192.33
7176 7380.25 7421.83 7429 7440.67 7464.75 7482.5 7489 7515 7516 7529 7535.5 7543
Gid-6
Depth (ft)
Gid-5
Well
5283 2409 6475 4688 6075 7306 3538 4969 412 1182 1096 2560 1619
4370 3554 5491 2504 3026 3645 1491 3302 1747
Total
2623 1219 3145 2231 3125 3468 1576 2232 115 514 390 1324 676
2117 1782 2263 1166 1591 1756 705 1548 992
Total C29
Steranes (ppm saturate fraction): M/Z 217
570 232 627 482 689 782 449 487 121 135 204 342 243
450 312 921 225 300 380 161 366 219
C29 dias
530 229 613 436 539 645 369 561 38 108 130 278 200
428 317 710 191 241 409 137 377 166
C29 dias
585 264 816 524 637 933 522 540 69 61 120 194 0
403 338 506 318 440 388 0 338 159
C29 dias
650 293 800 560 757 959 533 732 36 114 114 369 375
570 508 661 314 394 475 179 434 314
29aaa-S
609 302 749 531 1331 824 453 655 39 241 172 571 301
499 373 501 440 355 405 179 333 335 397 218 620 369 0 692 0 0 23 0 0 0 0
319 298 348 49 262 309 120 260 0
29a/3/3-R 29aflfl-S
967 406 977 771 1037 994 589 846 17 159 104 384 0
729 602 753 364 581 568 227 520 273
29aoea-R
180
D. J. CURRY ET AL.
Table 6. Explanation of symbols
Symbol
Explanation
Symbol
I. Total extract and saturate fraction chromatograms n-Cx normal alkane or alkene with x carbon atoms PR pristane PH phytane Me-Naph methylated naphthalene homologues C1-NAPH methyl naphthalene C2-NAPH dimethyl naphthalenes PHEN phenanthrene C1-PHEN methyl phenanthrenes II. Steranes (mlz 217 mass fragmentograms) STD internal standard C29D C29diasterane isomers T triterpane T (see text) C29 C29 regular sterane isomers C29H C29 regular hopane
Explanation
III. Triterpanes (m/z 191 mass fragmentograms) Cx regular hopane with x carbon atoms OL oleanane T triterpane T (see text) R C3o triterpane X C30 diahopane ('Compound X') IV. Pyrolysis-gc/ms total ion current and selected ion chromatograms X n-alkene/alkane pair with x carbon atoms A aromatic constituent P polar constituent 1-Pr 1-pristene Me-N methylated naphthalene homologues F1 bridged aromatic species (see text)
References VAN AARSEN, B. G. K., HESSELS, J. K. C., ABBINK, O. A. DE LEEUW,J. W. 1992. The occurrence of polycyclic sesqui-, tri- and oligoterpenoids derived from a resinous polymeric cadinene in crude oils from southeast Asia. Geochimica et Cosmochimica Acta, 56, 1231-1246. BERTRAND, P., BEHAR~ F. & DURAND, B. 1986. Composition of potential oil from humic coals in relation to their petrographic nature. Organic Geochemistry, 10,601-608. COLLEN, J. D., JOHNSTON, J. H., DUNN, P. J. & NEWMAN, R. H. 1988. 13C NMR spectra from Upper Cretaceous and Lower Tertiary coal measures, Taranaki Basin, New Zealand. Geology Board of Studies Special Publication No. 2, Research School of Earth Sciences, Victoria University of Wellington. CURRY, D. J. 1985. Organic geochemistry and oil generation potential of Tertiary coals from the West Natuna Basin, South China Sea. In: Abstracts of the 12th International Meeting on Organic Geochemistry, Julich, Germany. in prep. Geographic and temporal controls on the distribution of potentially oil generative coals. DURAND, B. • ESPITALIE, J. 1976. Geochemical studies on the organic matter from the Doula Basin (Cameroon). If. Evolution of kerogen. Geochimica et Cosmochimica Acta, 40,801-808. & PARATTE, M. 1983. Oil generation potential of coals: a geochemical approach. In: BROOKS, J. (ed.) Petroleum Geochemistry and the Exploration of Europe. Geological Society, London, Special Publication, 12,212-222. FOGEL, M. L., SPRAGUE, E. K., GIZE, A. P. & FREY, R. W. 1989. Diagenesis of matter in Georgia salt marshes. Estuarine, Coastal, and Shelf Sciences, 28,211-230. GRANTHAM, P. J., POSTHUMA, J. & BAAK, A. 1983.
Triterpanes in a number of Far-Eastern crude oils. In: BJOROY, M. et al. (eds) Advances in Organic Geochemistry 1981. John Wiley & Sons, London, 675-683. HEATH, R. 1989. Exploration in the Cooper Basin. Australian Petroleum Exploration Association Journal, 29,366-378. HORSFIELD, B. 1984. Pyrolysis studies and petroleum exploration. In. BROOKS, J. & WELTE, D. H. (eds). Advances in Petroleum Geochemistry (Volume 1). Academic Press, London, 247-298. --, YORDY, K. L. & CRELLING, J. C. 1988. Determining the petroleum-generating potential of coal using organic geochemistry and organic petrology. Organic Geochemistry, 13, 121-129. HUC, A. Y., DURAND, B., ROUCACHET,J., VANDENBROUCKE,M. & PIWrION,J. L. 1986. Comparison of three series of organic matter of continental origin. Organic Geochemistry, 10, 65-72. HUNT, J. W. 1989. Permian coals of eastern Australia: geological control of petrographic variation. International Journal of Coal Geology, 12, 589634. -& SM'crH, M. 1989. Origin of inertinite-rich coals of Australian cratonic basins. International Journal of Coal Geology, 11, 23-46. ISAKSEN, G. H. 1991. Molecular indicators of lacustrine freshwater environments. In: MANNING, D. (ed.) Organic Geochemistry: Advances and Applications in Energy and the Natural Environment (15th Meeting of the European Association of Organic Geochemists Poster Abstracts). Manchester Press, Manchester, 361-364. JENKINS,C. C. 1989. Geochemical correlation of rocks and crude oils from the Cooper and Eromanga Basins. In: O'NEILL, B. J. (ed.) The Cooper and Eromanga Basins. Proceedings of Petroleum
A L I P H A T I C - R I C H COALS F R O M A U S T R A L I A AND NEW Z E A L A N D Exploration Society of Australia, Society of Petroleum Engineers, Australian Society of Exploration Geophysicists (SA Branches), Adelaide, 525-540. JOHNSTON, J. H., COLLIER, R. J. & MAIDMENT,A. I. 1991. Coals as source rocks for hydrocarbon generation in the Taranaki Basin, New Zealand: a geochemical biomarker study. Journal of Southeast Asian Earth Sciences, 5,283-289. KHORASANI, G. K. & MICHELSEN, J. K. 1991. Geological and laboratory evidence for early generation of large amounts of liquid hydrocarbons from suberinite and suberous components. Organic Geochemistry, 17,849-863. KING, P. R. & ROBINSON, P. H. 1988. An overview of Taranaki region geology, New Zealand. Energy Exploration and Exploitation, 6, 213-232. LIu, S. Y. & TAYLOR, G. H. 1991. TEM observations on Type III kerogen, with special reference to coal as a source rock. Journal of Southeast Asian Earth Sciences, 5, 43-52. MOLDOWAN, J. M., FAGO, F. J., CARLSON, R. M. K., YOUNG, D. C., DUYNE, G. V., CLARDY, J., SCHOEL, M. , PILLINGER, C. T. & WATT, D. S. 1991. Rearranged hopanes in sediments and petroleum. Geochimica et Cosmochimica Acta, 3333-3353. NIP, M., DE LEEUW, J. W., HOLLOWAY,P. J., JENSEN, J. P. T., SPRENKELS,J. C. M., DE POOTER, M. & SLEECKX, J. J. M. 1987. Comparison of flash YcrOlysis, differential scanning calorimetry, NMR and IR spectroscopy in the analysis of a highly aliphatic biopolymer from plant cuticles. Journal of Analytical and Applied Pyrolysis, 11, 287-295. - - , SCHENK,P. A., WINDtG, W., MEUZELAAR, 'H. L. C. & CRELLING, J. C. 1989. A flash pyrolysis and petrographic study of cutinite from the Indiana paper coal. Geochimica et Cosmochimica Acta, 53,671-683. , TEGELAAR, E. W., BRINKHUIS,H., DE LEEUW, J. W., SCHENK,P. A. t~ HOLLOWAY,P. J. 1986. Analysis of modern and fossil plant cuticle by Curie point pyrolysis gc and Curie point pyrolysis gc/ms: Recognition a new, highly aliphatic and resistant biopolymer. Organic Geochemistry, 10, 769-778. PILAAR,W. H. F. & WAKEFIELD,L. L. 1978. Structural and stratigraphic evolution of the Taranaki Basin, offshore North Island, New Zealand. Australian Petroleum Exploration Journal, 18, 93-101. & ~ 1984. Hydrocarbon generation in the Taranaki Basin, New Zealand. In: DEMAISON,G. & MURRIS, R. J. (eds) Petroleum Geochemistry and Basin Evaluation American Association of Petroleum Geologists, Memoir 35,405-423. POCKNALL, D. T. 1990. Palynological evidence for early to middle Eocene vegetation and climate 55,
-
-
181
history of New Zealand. Review of Paleobotany and Palynology, 65, 57-69. POWELL, T. J. • BOREHAM, C. J. 1991. Petroleum generation and source rock assessment in terrigenous sequences: an update. Australian Petroleum Exploration Association Journal, 31, 297-311. - - , SMYTH,M., RUSSELL,N. & COOK, A. C. '1991. Petroleum source rock assessment in nonmarine sequences: pyrolysis and petrographic analysis of Australian coals and carbonaceous shales. Organic Geochemistry, 17,375-394. RAMANAMPISOA, L., RADKE, M., SCHAEFER, R. G., LITKE, R., RULLKOTTER, J. & HORSFIELD, B. 1989. Organic-geochemical characterisation of sediments from the Sakoa coalfield, Madagascar. Organic Geochemistry, 16, 235-246. SANDVIK,E. I., YOUNG,W. A. & CURRY,D. J. 1992. Expulsion from hydrocarbon sources: the role of organic absorbtion. Organic Geochemistry, 19, 77-88. SHANMUGAM,G. 1985. Significance of coniferous rain forests and related organic matter in generating commercial quantities of oil, Gippsland Basin, Australia. American Association of Petroleum Geologists Bulletin, 69, 1241-1254. SMITH, J. W., GILBERT, T. D. & BATTS, B. D. 1987. A quest for a new parameter in petroleum exploration geochemistry. Australian Petroleum Exploration Journal, 27, 98-105. TAYLOR, G. H. & LIU, S. Y. 1987. Biodegradation in coals and other organic-rich rocks. Fuel 66, 1269-1273. & DIESSEL, C. F. K. 1989. The cold climate origin of inertinite-rich Gondwana coals. International Journal of Coal Geology, 11, 1-22. & SMVTH,M. 1988. New light on the origin of Cooper Basin oil. Australian Petroleum Exploration Association Journal, 28,303-309. TEERMAN, S. C. & HWANG, R. J. 1989. Evaluation of the source rock potential of Sumatran coals by artificial maturation. Proceedings of the 18th -
-
-
-
Indonesian Petroleum Association Convention. Indonesian Petroleum Association, Jakarta, 469-489. TEGELAAR, E. W., DE LEEUW, J. W., LARGEAU,C., DERENNE, S., SCHULTEN, H.-R., MULLER, R., BOON, J., NIP, M. & SPRENKELS,J. C. M. 1989. Scope and limitations of several pyrolysis methods in the structural elucidation of a macromolecular plant constituent in the leaf cuticle of
Agave americana L. Journal of Analytical and Applied Pyrolysis, 15, 29-54. THOMAS, B. M. 1982. Land plant source rocks for oil and their significance in Australian basins.
Australian Petroleum Exploration Association Journal, 22,164-176. TISSOT, B. P. & WELTE, D. H. 1984. Petroleum Formation and Occurrence. Springer Verlag, Berlin.
The oil potential of Mid-Jurassic coals in northern Egypt MARK
A. BAGGE
& MARTIN
L. K E E L E Y
Intera Information Technologies Ltd, Highlands Farm, Greys Road, Henley-on-Thames, Oxfordshire R G 9 4PS, UK
Abstract: Geochemical analysis of selected coals indicates that some of the Jurassic coals in Egypt have oil-generating potential. Other coals have just gas potential, but may have already generated oils as they are now post-mature for oil generation. Petrographic analysis of the coals reveals a composition dominated by vitrinite, but with variable liptinite contents and the occurrence of fluorescent vitrinite in some coals. It is not clear which components of the coal are responsible for the oil potential; it is probable that liptinite, hydrogen-rich vitrinites and a fine-grained groundmass of algal origin all contribute. A combination of Rock-Eval Hydrogen Indices and pyrolysis-gc P2 fingerprints provides the best method for determination of the oil- or gas-prone nature of the coals. Attempts at correlating oils to source rock extracts illustrate the problem of oil-source correlations involving a rapidly varying source rock environment. The bitumen extracts from the coals reveal variable characteristics which reflect facies variations in the coals. Several of the coals have unusual saturated hydrocarbon and biological marker distributions that are atypical of humic coals and give a poor correlation to the waxy oils from the Meleiha field. The best oil-source rock correlation is with a carbonaceous shale, not a coal, but it is likely the oils were derived from large volumes of source rock including coals and carbonaceous shales. As the coals have unusual extract characteristics, any oils derived from a more localized environment in the coal swamp may show very unusual geochemical characteristics, and would not be recognized immediately as being derived from terrestrial coals.
Jurassic coals in Egypt were first recognized by Barthoux & Douvill6 (1913) at outcrop in the northern Sinai Peninsula, within the Gebel el Maghara d o m e (Fig. 1). The subsurface of this region has since been thoroughly explored, and coal is now being mined. Khatatba-1, the second oil well in Egypt outside of the Gulf of Suez, drilled by Standard Oil in 1944-45 encountered Jurassic coals. This well became the type well for subsequent reference to mid-Jurassic clastics, even though all the original cuttings have long since been consumed, and no electric logs were obtained. In the western part of the Western Desert of Egypt, several oil fields including Meleiha, Umbarka, Khalda and Salam produce distinctively waxy paraffinic oils, quite unlike those in the Abu Gharadiq and Gulf of Suez basins to the east. Several attempts were m a d e in the 1980s to ascertain the origin of these oils without definitive results (Abu E1 Naga 1984). The uncertainty arose in part because the only significant concentrations of organic matter present in the Mesozoic/Cenozoic section are the Jurassic coals, but it was assumed that because these coals are dominated by humic kerogens, they are only gas prone. Pyrolysis analyses may have gone some way to support this interpretation but were misleading due to the high maturity of
the Jurassic section in many wells. However, subsequent studies have shown that the Jurassic coal sequence is indeed the source for these oils (Bagge et al. 1988; Keeley et al. 1990). The data presented in this paper illustrate the characteristics of selected mid-Jurassic coals in northern Egypt and the oils derived from the coal sequences. The data also illustrate the difficulties in recognizing oil-prone coals from geochemical data, and establishing whether it is in fact the coals or associated carbonaceous shales that are responsible for generating the associated oils.
Geological setting and stratigraphy Jurassic coals in Egypt are restricted to the area north of 29°30"N. They are found as minor associates along with sandstones and lesser volumes of interbedded siltstone and organicrich mudstones. This association of lithotypes comprises the Safa M e m b e r of the Khatatba Formation, which is entirely of Bathonian age (Keeley et al. 1990) (Fig. 2). In the central and eastern sectors, a distinctive carbonate interval, the Kabrit Member, separates the Safa sands, shales and coals into U p p e r and Lower Members. A continuous Lower Jurassic succes-
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coal and Coal-bearingStrataas Oil-prone Source Rocks? Geological Society Special Publication No. 77, pp. 183-200.
183
184
M. A. BAGGE & M. L. KEELEY
- 36o____________ 26 _ °_
28 °
30 °
32 °
34 °
MEDITERRANEAN SEA
// /"
,/
,
f
2""
/ •
,;'
UMBARKA
/
MELEIHA i
/~
Gebel ~1 ' \
eALAMEIN Arif
32o~
SINAI
ABU GHARADIQ
*=/ i WESTERN
I
DESERT EASTERN
3 0 ° - ~ ~
DESERT OIL/GAS FIELD
Outcrops
I
200 Kms
RED \".,,. SEA \,
I
Fig. 1. Location map of selected Jurassic outcrops and oil/gas fields in northern Egypt.
sion lies beneath, beginning in the Pliensbachian. By contrast, in the western sector, the Sara Member lies at the base of the Jurassic succession, locally overlying and interbedded with volcanics (Keeley et al. 1990). Beneath these Jurassic rocks lie undated red clastics, presumably of Late Triassic age. The onset of Jurassic deposition in this western sector, during a regressive phase, indicates the importance of local tectonics, rather than sea-level changes, to depositional character. The age of onset of a continuous phase of deposition during the Early-Mid-Jurassic on the southern margin of Tethys youngs westwards. This is attributed by Keeley & Wallis (1991) to the propagation of rifting along the northern Egyptian continental margin, originating in Palestine during the Triassic. Sara Member sedimentation took place within this active tectonic setting, along a fluctuating coastline, characterized by a strong northward sediment flux, but in the absence of deltaic outbuilding.
As a consequence, coast-linear shoreface, beach bar and swamp deposits are found alternatively stacked. The coals deposited in this coastal environment were first reported as gas-prone Type III source rocks (Parker 1982). Type II/III source rocks, capable of oil and gas generation were recognized in areas of the coal swamp where hydrogen-rich liptinitic material has been washed into and concentrated in stagnant areas. The liptinitic material comprises cuticle, spores, pollen, resin and algae (Parker 1982). These oilprone source rock facies are locally developed, and were deposited in a lower swamp environment. This environment was characterized by low relief and marine influences, allowing stagnant environments to develop (Figs 3 & 4). In the upper swamp environment to the south, relief was greater, and stagnant conditions did not develop. This environment was more oxic, resulting in the accumulation of humic coals comprising gas/condensate-prone and inert organic matter.
OIL POTENTIAL OF MID-JURASSIC COALS IN N. EGYPT
185
Neocomian Tithonian II,,I. -.- ~
Kimmeridgian
~
I
I
~'d "UPPER LIMESTONE~"-r--t-~- ~
I 'l
!
I
I
I
I
I
I
.[
" L
-
!
".
j
I "~
'"
~:~',";'f-;;'Z:'" :":'"":'7'''''''-''v'''''''::':''''''-''':'~ij')~i~i~5'':'''v''"( . . . . . : Oxfordian
~r
.-t._, L..~l
, I
~L_
"DARDUMA
I
I
I
I . I
LIMESTONE", J I I
I
, I
I
~
, I
I.
I. , I
I l
I
I,
I
' ~ - - , I
I. I
I I
I
--="= ,,I
I I '
I I
,I
-BIRrSHE'IBA-= t .... L l
, I
~'2-,~--_---mN:)'D-: --"=-~'-~"~--~---' -t I I
-':" -I
"I
. . . . I
I
,
I
MASAJID F
"/OWER U/~ESfONE'-r-L Callovian
~ . J --~--2
Bathonian
.'.'''"
__
L. I ; _ l - k _ J - 2 , _ L , _ l _ , 2 , _ L ; ' ,2'_L,__'_'2'-L'- - - - - - - -TT.~ -TT..-7..-7. g ~ $T5..-ffA: -"-7"-" ; ; - -~.._--~"-~---"-~ --------'
. - - ~ ~ ' - T .
.'.'.'.~r."
...............
.;Jr:..'.'."
....
..-"
.... .'.'.',4i~.'.v
U.
--. SAFA
. . . . . . . . . . .v.' Jr:...v.v....:..'~':))))'m~-)
I~f U
A "IP A "T" I~ A
Bajocian
Aalenian
Toarcian
Pliensbachian
Sinemurian Hettangian =
Rhaetian Norian Carnian WEST-~
ABOUT 1000 KMS
= EAST
FORMATIONS
Fig. 2. Jurassic stratigraphy in Egypt (after Keeley et al. 1990).
Evidence for coaly source rocks from oil characteristics The Western Desert oils are high gravity, low sulphur, waxy crudes and are very paraffinic. Gas chromatography of whole oils and Cls+ saturated hydrocarbons indicates large amounts of normal alkanes > C2s with an odd-carbon preference. These compounds are often referred to as 'terrestrial waxes'. A chromatogram of a typical oil from Meleiha is shown in Fig. 5. The odd-carbon preference and the high pristane/ phytane (Pr/Ph) ratios of the oils are characteristic of oils sourced from terrestrial organic precursors. Biological marker distributions are also diag-
nostic of terrestrial organic matter. Fragmentograms of steranes (m/z 217) from the Meleiha oil (Fig. 6) are dominated by C29 regular steranes and diasteranes. The predominance of C29 steranes over C28 and Cz7 steranes is consistent with a terrestrial source for the oils. Terpanes (m/z 191 fragmentogram, Fig. 7) are present in much greater amounts than steranes, a feature usually observed when terrestrial organic matter is the main organic precursor. The hopanes are the dominant series of terpanes and tricyclic terpanes are present in only minor amounts. Two unidentified terpanes that have been described previously in the literature as 'terrestrial indictors' are also present. Most
186
M. A. BAGGE & M. L. KEELEY
;... )
,.
-, -
~
,,.-\ \
(::::).£,k-~,.~~ ,rc,~,~. Fig. 3. Facies distribution of northern continental margin of Egypt during the Jurassic (after Keeley & Wallis 1991).
I
o °:o°:
34°
30°
26°E - - - F----
._/--
~_~_ ~
- ~-:_---=~__:
~.@
Upper Safa M e m b e r (early Bathonian)
KEY
SHALLOW MARINE LIMESTONES (SHELF)
I
. . . . . .
~ 5d
13e~
~
DELTAIC- LACUSTRINE- LITrORAL SHALES, SANDSTONESAND COAL
~
FLUVIALCLASTICS CI • 200m
Fig. 4. Isopach/facies map, Upper Safa Member, northern Egypt (after Keeley & Wallis 1991).
30°N
OIL POTENTIAL OF MID-JURASSIC COALS IN N. EGYPT 100
187
-
(118 C17
A >
E ('~ c~ x
60-
Z
o 40'
er
C
~
Pr
I
h
I
O ~ I
RETENTION
Fig. 5. Meleiha oil:
C15+
TIME
(MINUTES)
saturated hydrocarbons gas chromatogram.
,
g ¢u
~o
e,J oJ
m
o
O.CO| 3¢
I
I
3.2
I
|
34
(m/z 217)
I 36
RETENTION
Fig. 6. Meleiha oil: Sterane
iz
I
w
m ~"
o oJ
r-.
~
,
~
distribution.
TIME
I 38
(MINUTES)
I
I
188
M. A. BAGGE & M. L. KEELEY
a
38.49
23.09
c
N I e
g f
=c
?
oe Ea oo
$.5.40
• c
a~ N
,.~ u
u
I u
o
• t-
E c
oe Ec oe
¢
(o
uE
7,70
'~
u i
u
c
:~
~
~
~
~ u
i
u
'
22
24
26
28
30
• n
c e i~ c
o E u
1: o o r E o
z
~.E
o
mln
u i
o .00 20
rr
32
34
36
38
I 40
i
i
4E
L'
,8'
28 ' go '
nETENTION TZME (MINUTES)
Fig. 7. Meleiha oil: Terpane (m/z 191) distribution.
important of these is compound X which elutes just after the C29 norhopane. Philp & Gilbert (1986) observed this compound in many oils sourced from terrestrial organic matter in Australia and tentatively identify it as a C3o pentacyclic terpane. A second 'terrestrial marker', compound Y is present in the Meleiha oils which was also identified in the Australian oils by Philp & Gilbert (1986). This compound elutes just before the 17o~(H)-trisnorhopane (Tm) and is an unidentified C27 terpane. Despite the strong influence of land plant material, 18a(H)-oleanane is absent from the terpane distributions of the Meleiha oils. This compound is often reported to be a terrestrial indicator and has been found in rock extracts from Miocene strata in the Gulf of Suez. 18a(H)-oleanane is derived from angiosperm precursors and has only been reported in Late Cretaceous and Tertiary sediments and their derived oils (Philp & Gilbert 1986; Moldowan et al. 1991). 18a(H)-oleanane is absent in the Meleiha oils as the source rocks are preCretaceous. Data from a second Meleiha oil, produced from a deeper reservoir, are presented in Fig. 8
Initial inspection of the gas chromatogram and terpane distribution suggests the oil is different from the first oil, but all differences can be explained by the higher level of maturity for this second oil. Like the oil from the shallower reservoir, the pristane/phytane ratio is high (>3), and high molecular weight alkanes are present, but in lower quantities than the less mature oil, perhaps as a result of cracking. If the effects of maturity are allowed for, the terpane distribution confirms the relationship with the less mature oil; compounds X and Y are also present, apparently in greater abundance than in the less mature oil. The abundance of X and Y relative to the hopanes appears to be maturity dependent, with X and Y more resistant to breakdown at high temperature. The gc and gc-ms data clearly indicate that the two Meleiha oils were generated from source rocks containing significant amounts of terrestrial organic matter. The two oils initially appear distinct, but all differences can be explained by the effects of maturity. The oils were either generated from source rocks at different levels of maturity, or the oils in deeper reservoirs have suffered in-situ cracking.
189
OIL P O T E N T I A L OF M I D - J U R A S S I C C O A L S IN N. E G Y P T
C~5
C17
C18
!
I~, 6o ('24
CA LZ 40-
t,:
I,r
o,
JJ~
m
(a)
RETENTION TIME (MINUTES)
r~ c)
2.77
c
© o
o
×
~
8
oN
o i
~X
t .84
'11
>
0.92
o.oo
(b)
~0
, , ,
2
~
4
,
~
6
,
~
,
30
,
, 32
,
, 34
,
36
,
, 38
.
.
. . 40
.
42
• 44
,
, 45
.
, 4n
,
, 50
~
RETENTION TIME (MINUTES)
Fig. 8. Meleiha oi| from deep reservoir: (a) Cas+ gas chromatogram and (b) terDane (m/z 191) distribution.
5;!
190
M. A. BAGGE & M. L. KEELEY
Recognition of oil potential in Jurassic coals Coals are often overlooked as possible oil source rocks in geochemical sampling programmes. Evaluation of generating potential of the coal is usually based on petrographic data, where high liptinite contents are taken to indicate the possibility of oil-generating potential. Typical humic, vitrinite-rich coals are dismissed as being simply gas prone. As part of this study, Jurassic coals and a carbonaceous shale from well and outcrop were submitted for a wide range of source rock analyses to establish if the coals do have oilgenerating potential and, if so, which analytical technique best determines this potential. Data are presented from coals and carbonaceous shales from an outcrop section at Gebel el Maghara, northern Sinai, and from a well in the Meleiha area in the Western Desert (Bagge et al. 1988). Well names and depths cannot be quoted because the information is confidential. The oil potential of the coals and carbonaceous shales were evaluated using chemical, petrographic analyses and bitumen extract characterization.
Chemical analyses The most commonly acquired geochemical data from well cuttings and cores are Total Organic Carbon (TOC) and Rock-Eval Pyrolysis data. These allow the quantity and quality of organic matter in sediments to be determined, and potential source rock horizons delineated. The analyses are usually applied to argillaceous sequences to allow quantification of the amount and hydrocarbon generating potential of the dispersed organic matter ('kerogen') in the sediments, and can also be used to characterize the hydrocarbon-generating potential of coals. The Rock-Eval Hydrogen Index (S2*100/TOC) is used to describe organic matter quality and its oil- or gas-prone nature. Source rocks containing oil-prone Type I or Type II organic matter typically have Hydrogen Indices of 400 of greater, whereas humic, Type III, gas-prone material has Hydrogen Indices of 150 or less. Inert, Type IV organic matter often has Hydrogen Indices of 10 or less. Although TOC contents and pyrolysis $2 yields give a good estimation of the present-day hydrocarbon-generating potential of a coal, caution should be exercised when using Hydrogen Indices to indicate whether a coal is oil or gas prone. High Hydrogen Indices are usually associated with boghead or cannel coals and do indicate oil potential; however, when humic coals contain a small but significant proportion of oil-prone material, the oil poten-
tial can go unrecognized as the Hydrogen Index averages the potential of inert, gas- and oilprone macerals. It is, therefore, crucial to interpret Hydrogen Index data with supporting analyses such as combined pyrolysis-gc and maceral typing. The geochemical characteristics of the coals and carbonaceous shales are summarized in Table 1. The Gebel el Maghara coals have Hydrogen Indices of 408-484 mg HC g-1 rock which clearly indicate oil-prone organic matter. The coals are obviously more hydrogen rich than the associated carbonaceous shales which have Hydrogen Indices of 201-212. The Meleiha well coals have lower Hydrogen Indices (167-210), which suggests gas-generating potential. These samples are past the main stage of oil generation, however (R0 1.20-1.22%), and the pyrolysis yields therefore represent their residual gas potential. The oil potential of the Gebel el Maghara coal and gas potential of the Meleiha coal are confirmed by the pyrolysis-gc P2 (pyrolysate- chromatograms (Fig. 9) which show an extended series of alkane/alkene doublets in just the Gebel el Maghara coal.
Petrographic analysis Petrographic analysis is particularly valuable in coals as the macerals can be identified easily. It is generally accepted that the liptinitic macerals derived from algae or hydrogen-rich plant material such as spores, pollen, resin and cuticle are responsible for oil potential. The macerals form a large part of the organic matter in boghead or cannel coals, but can also be present in humic coals. Vitrinite, which forms the bulk of humic coals is usually considered gas prone, but some hydrogen-rich forms of vitrinite may also have oil-generating potential (See Collinson et al. this volume). The maceral content of the Gebel el Maghara coals and shales and Meleiha well coals are described in the point count data shown in Table 2. The Gebel el Maghara coals are composed primarily of vitrinite (76.4-79.0%), but with significant amounts of liptinite macerals: cutinite, sporinite, resinite and liptodetrinite (total 21.0-22.8%). Despite being of much lower hydrogen content (lower Hydrogen Index), the Gebel el Maghara carbonaceous shales also have moderate liptinite contents (8.0-20.0%). From liptinite content alone, therefore, it would not have been possible to identify the Gebel el Maghara coals as being more hydrogen rich, and hence more oil prone than the carbonaceous shales. The Meleiha well coals have higher vitrinite contents (85.6-
Sample type
61.60 59.40 2.26 5.33 39.73 60.00 6.35
TOC (wt %) 251.05 287.50 4.80 10.69 83.57 100.30 12.63
$2 (mg g-a) 408 484 212 201 210 167 199
HI (rag HC g-1TOC) 5 5 25 31 3 1 6
OI (mg CO2 g-1TOC)
* No vitrinite reflectance performed. Maturity approximately equivalent to Meleiha well coals.
Gebel Maghara coal i Outcrop Gebel Maghara coal 2 Outcrop Gebel Maghara shale 1 Outcrop GebelMagharashale2 Outcrop Meleiha well coal 1 Well cuttings Meleiha well coal 2 Well core Meleiha well shale Well cuttings
Sample ID
Source rock evaluation
26.49 26.36 0.76 0.97 9.95 13.81 2.39
S1 (mg g-1) 0.10 0.08 0.14 0.08 0.11 0.12 0.16
PI S 1/S 1 + $2
Free hydrocarbons
Table 1. Geochemical analysis of Jurassic coals and carbonaceous shales from Egypt (data from Bagge et al. 1988)
418 417 436 426 454 456 450
Tmax (°C)
No. of readings 50 5O 5O 50 100 100
% Ro 0.43 0.39 0.43 0.44 1.22 1.20
Maturity
192
M. A . B A G G E
& M. L. K E E L E Y
'
25fl
-l I
200 > E W UI Z 0 O.
175
15e
6J rr r~ 0 U W
IZ5
i I Bfl
I
g 75
5B
25
B
(a)
87
89
9
913
~
9~
5%
1
I
I
1
I01
103
105
107
TIME
(min)
I lea
I ill
I
I
113
115
1
tit
1
119
|zi
188
IG8
? E 1'18
Z O
IZ~
g u
w
B~
t.d rn
I
I
0B7
B9
9',
9'3
9~
9)
5;
1 0'1
' 1f13
' 1f15
I 0 '7
1 8'9
I I I'
1 1'3
I
I5
1 1'7
[ 1 '9
I ?Jl
T I M E (min) (b) Fig. 9. Pyrolysis-gc P2 fingerprints for (a) Gebel el Maghara coal and (b) Meleiha well coal.
94.4%) and lower liptinite contents (0.8-1.6%). The dominant form of vitrinite is desmocollinite (a more hydrogen-rich vitrinite) which, together with the fine-grained matrix, is fluorescent in one of the Meleiha coals. Based on the high vitrinite content, these coals would traditionally be interpreted to be gas-prone Type III humic
coals, but the fluorescence of the ground mass and also the vitrinite type suggests the coals may have had some oil potential. Fluorescence is often used as an indicator of oil potential in source rocks and the fluorescence of some forms of vitrinite, particularly desmocollinite may be important in the consideration
% Maceral
90.0 Pyrite 8.0 Other minerals 0.0
76.4 Pyrite 22.8 Other minerals 0.0
94.4 Pyrite 0.8 Other minerals 4.0 (absolute count)
Fusinite Semifusinite Sclerotinite Macrinite Micrinite Inertodetrinite
Fusinite Semifusinite Sclerotinite Macrinite Micrinite Inertodetrinite
Fusinite Semifusinite Sclerotinite Macrinite Micrinite Inertodetrinite
0.0 Bitumen 67.0*(Exudatinite)
0.0 tr 0.4 0.0 0.4
0.0 Bitumen 0.0 (Exudatinite)
2.0 6.0 5.2 2.4 7.2
0.0 Bitumen 0.0 (Exudatinite)
0.0 4.0 2.0 0.0 2.0
% Maceral
0.8
0.4 2.8 0.0 0.0 0.0 0.8
0.8
0.0 0.0 0.0 0.0 0.0 0.0
2.0
0.0 tr 0.0 0.0 0.0 0.0
%
% Maceral
80.0 Pyrite 20.0 Other minerals 0.0
79.0 Pyrite 21.0 Other minerals 0.0
Total vitrinites Totalliptinites Total inerts
85.6 Pyrite 1.6 Other minerals 12.8
Sample: Meleiha well coal 2 Telocollinite 17.2 Cutinite Corpocollnite 0.4 Sporinite Desmocollinite 67.6 Resinite TeUinite 0.0 Alginite Pseudovitrinite 0.0 Liptodetrinite Vitrodetrinite 0.4
Total vitrinites Total liptinites Total inerts
Sample: Gebel Maghara coal 2 Telocollinite 28.0 Cutinite Corpocollnite 1.0 Sporinite Desmocollinite 50.0 Resinite Tellinite 0.0 Alginite Pseudovitrinite 0.0 Liptodetrinite Vitrodetrinite 0.0
Totalvitrinites Total liptinites Total inerts
Sample: Gebel Maghara shale 2 Telocollinite 0.0 Cutinite Corpocollnite 0.0 Sporinite Desmocollinite 80.0 Resinite Tellinite 0.0 Alginite Pseudovitrinite 0.0 Liptodetrinite Vitrodetrinite 0.0
Maceral
* Fine-grained groundmas of mineral mixed with an amorphous fluorescing material. Not counted in maceral percentages.
Total vitrinites Total liptinites Totalinerts
Sample: Meleiha well coal 1 Telocollinite 17.2 Cutinite Corpocollnite 0.0 Sporinite Desmocollinite 75.6 Resinite Tellinite 0.0 Alginite Pseudovitrinite 0.0 Liptodetrinite Vitrodetrinite 1.6
Total vitrinites Total liptinites Total inerts
Sample: Gebei Maghara coal 1 Telocollinite 41.2 Cutinite Corpocollnite 1.2 Sporinite Desmocollinite 34.0 Resinite Tellinite 0.0 Alginite Pseudovitrinite 0.0 Liptodetrinite Vitrodetrinite 0.0
Totalvitrinites Total liptinites Total inerts
Sample: Gebel Maghara shale I Telocollinite 50.0 Cutinite Corpocollnite 0.0 Sporinite Desmocollinite 40.0 Resinite Tellinite 0,0 Alginite Pseudovitrinite 0.0 Liptodetrinite Vitrodetrinite 0.0
Maceral
Fusinite Semifusinite Sclerorinite Macrinite Micrinite Inertodetrinite
Fusinite Semifusinite Sclerorinite Macrinite Micrinite Inertodetrinite
Fusinite Semifusinite Sclerorinite Macrinite Micrinite Inertodetrinite 0.0 Bitumen 28.0 (Exudatinite)
0.0 0.0 0.8 0.0 0.8
0.0 Bitumen 0.0 (Exudatinite)
3.0 4.0 6.5 3.0 4.5
0.0 Bitumen 0.0 (Exudatinite)
5.0 5.0 3.0 2,0 5.0
% Maceral
Table 2. Petrographic analysis of Jurassic coals and carbonaceous shales (maceral composition by point count), Egypt (data from Bagge et al. 1988)
0.0
5.2 6.4 0.0 0.0 0.4 0.8
0.0
0.0 0.0 0.0 0.0 0.0 0.0
0.0
0.0 0.0 0.0 0.0 0.0 0.0
%
~O
O
~Z ~Z
~)
¢3
"--4
(7
r~ O
O
O
194
M. A. BAGGE & M. L. KEELEY
of humic coals as possible sources for oil. Bertrand (1984) reached the conclusion that some vitrinites may have oil-generating potential in a study of 68 coals worldwide. In his study, no correlation was found between exinite content and oil-generating potential, raising the possibility that hydrogen-rich vitrinites are partly responsible for oil-generating potential in some Type III source rocks. Use of petrographic analysis is, therefore, not definitive in evaluating the oil potential of Jurassic coals of the Western Desert. There is no doubt that large amounts of liptinitic material results in high oil potential; however, the Gebel el Maghara coals and carbonaceous shales have similar liptinite contents but very different oil potentials. Liptinite maceral content alone cannot be used to establish oil potential, and it is likely that hydrogen-rich vitrinites and the finegrained groundmass are responsible for oil potential in some coals. These coals are likely to have been overlooked as potential oil sources in traditional interpretation schemes.
Extract characterization Extract characterization was used to provide information on the organic matter in the coals and also to allow correlations with the oils. Soluble bitumen extracted from the rocks was separated by liquid chromatography into saturated and aromatic hydrocarbons, NSO compounds and asphaltenes (Table 3). The saturated hydrocarbon fraction was analysed by Cas÷ gas chromatography and combined gas chromatography-mass spectrometry (gc-ms). Extract characteristics of the Western Desert coals vary widely between individual coal samples, reflecting facies changes in the depositional environment of the coals. An appreciation of these variations is essential when correlating oils to source rock extracts. The Cas÷ gas chromatogram of the Gebel el Maghara coal shown in Fig. 10 is rather unusual and is dominated by several unidentified compounds. The Meleiha well coals (Fig. 11) show a more usual suite of normal alkanes with no oddcarbon preference. The dominance of low molecular weight n-alkanes is probably a result of the high maturity of these coals (1.2-1.22%
Ro). Perhaps the most unusual aspect of the Meleiha coal extracts is the relative abundance of acyclic isoprenoids. In one extract, acyclic isoprenoids are virtually absent, whereas in the other, phytane is more abundant than pristane. The low concentration of isoprenoids relative to n-alkanes is probably related to the high level of
maturity of the coals, but this cannot explain the Pr/Ph ratio of less than unity. This Pr/Ph ratio (0.72) is atypical for coal, which usually has ratios of 3.0 or greater, and is more typical of a marine or lacustrine oil source rock deposited in a strongly reducing environment. This coal has a fluorescent fine-grained groundmass and unusual sterane and terpane distributions, and is thought to contain algal material, too fine grained to identify positively microscopically. The sterane and terpane biological marker compounds (Fig. 12) are unusual for humic coals and are thought to be derived from algal precursors. Readers are referred to a review of sterane and triterpane biological markers such as Waples & Machihara (1990) for an account of more typical fingerprints for different types of source rock. The m/z 217 fragmentogram of one Meleiha coal shows that C27 regular steranes and diasteranes, and C28 regular steranes are almost as abundant as the C29 steranes and diasteranes. This is in contrast to the Meleiha oils, which have C29 regular steranes and diasteranes dominant. Terpanes are also unusual for coal extracts, in that tricyclics are present in significant amounts relative to the pentacyclic terpanes, and C3s homohopanes are more abundant than C34 homohopanes. The terrestrial markers X and Y are absent. These sterane and terpane distributions are more akin to algal marine source rocks than humic coals. These coal extracts obviously give a poor correlation to the Meleiha oils presented here. The best oil-source rock correlation is with a highly carbonaceous shale interbedded with the coals (Fig. 13). This shale has modest presentday gas source potential (Table 1), but is postmature for oil generation. However, it may have had oil-generating potential in the past. This source rock displays the two terpane biological markers X and Y, and gives a fairly good correlation with the more mature of the two Meleiha oils presented here. This shale cannot be described as a 'good' source rock by normal geochemical criteria and would be overlooked in most geochemical analysis programmes.
Conclusions Mid-Jurassic coal sequences in the northern Western Desert of Egypt are believed to be responsible for generating the waxy, paraffinic oils in the area, including those from the Meleiha field. The coals are contained in the Sara Member of the Khatatba Formation. Recognition of oil potential of the coals and associated carbonaceous shales is problematic,
34232 8025 4642 3803
Sample ID
Gebel Maghara coal 2 Meleiha well coal 1 Meleiha well coal 2 Meleiha well shale
5.76 2.01 0.78 10.45
Extract/TOC (%) 3.4 19.4 10.5 15.5
Sats (%)
* Non-eluted components and material lost through evaporation of solvent.
Total extract (ppm) 18.9 33.9 34.9 28.9
Arom (%) 51.5 24.1 29.5 36.3
Asph (%) 25 15.7 9.5 6.8
NSO (%)
Extract composition
Table 3. Bitumen extract data, Jurassic coals and carbonaceous shales, Egypt (data from Bagge et al. 1988)
1.1 6.8 15.5 12.5
Bal (%)*
1.1
1.07
1.14
CPI
1.78
0.72
B
Pr/Ph
196
M. A. BAGGE & M. L. KEELEY 100 -
E 60ob
c5 x
Z (/) U.I rr
40-
0---
0 " ~ ' ~ ' ~ ' ; " ~'0" ~'~' ~',' l's" I'B' ~o' ~2' ~,' ~e' ~8' ~0' J2' ~,' J6' ~ ' 2o ' ,~2' 2,' "6' ~ ' ~0' 5'~'~, RETENTION TIME (MINUTES)
Fig. 10. Gebel el Maghara coal: Cts+ gas chromatogram of saturated hydrocarbons.
and it is unclear which components (macerals) are responsible for the oil-prone nature of coals. The coals from outcrop at Gebel el Maghara in northern Sinai appear to be oil prone and are more hydrogen rich than the associated carbonaceous shales. Petrographic analysis would not, however, have distinguished the coals as more oil prone than the shales but Rock-Eval Hydrogen Indices and pyrolysis-gc pyrolysate fingerprints identify the coals as having oil potential. The Meleiha well coals are post-mature for oil generation (Ro 1.2%) but have high gas potential. These coals are unusual as they contain fluorescent vitrinite and a fluorescent finegrained groundmass; the low pristane/phytane ratio and biological marker distributions suggest this groundmass is derived from algae. These coals have probably generated oil in the past, but because they appear to contain primarily vitrinite, would not have been identified as potential oil source rocks. The Meleiha oils were generated from source rocks rich in terrestrial organic precursors. They contain two triterpane compounds, X and Y, which have been observed in Australian coals, but the terrestrial biological marker oleanane is
absent as the Jurassic source rocks pre-date the angiosperm precursors of this compound. The two oils are related but the deeper oil has suffered cracking, either in the source rock or in the reservoir. The oils correlate poorly to the unusual Meleiha coals, but give a possible correlation to a carbonaceous shale interbedded with the coals. This shale has only moderate generating potential, and is likely to be overlooked as a potential source rock in normal geochemical analysis programmes on exploration wells. The coals and carbonaceous shales accumulated in a swamp environment where facies variations were rapid. The dominant organic matter type is humic Type III plant material, but localized development of coals rich in liptinitic material, either terrestrial or algal, has occurred, probably in stagnant areas. Oils generated from this source rock environment are bound to show the average characteristics of a great volume of source rock and will correlate poorly to unusual, locally developed coal facies. Small oil accumulations or oil stains may, however, show very unusual characteristics if they are derived from a particular oil-prone coal facies.
197
OIL P O T E N T I A L OF MID-JURASSIC COALS IN N. EGYPT
E t'r~
60-
.4 x
C24
o=
40"
P
•
0
' ~' ~' ~'~o','~'~,'~'6'~8'~o'~2'~,'J6'~8'~'~2'~,'J6'~e'2o'2~'2,
(a)
&'2~'~0'~2
RETENTION TIME ~INUTES)
Sample: 8 8 0 7 1 - 0 1 4
I I i
~.
60 ¸
U.l
40.
C24
I !
O
(b)
O
' ~ ' ~ ' ~' ' ~ ' o ' l ' e ' l ' , ' l ' ~ ' &
~o'~2'2','d~'&'~o'g2'J,'3'6'~B'go'2~'','&'&'go'g2's', RETENTION TIME (MINUTES)
Fig. 11. Meleiha well coals: Cls+ gas chromatograms of saturated hydrocarbons.
~,
198
M. A. BAGGE & M. L. KEELEY
2.37
,
77
m L
¢-
¢,.
t._
c~
ul
m
ol
u~
~'1 °1 °°
ru i
1.90
u 03 r~
,
+°[°I °° °°
~v +,
',' i
u
c. r.+ ,o
A 1.49
to
X (.J Z < Z
CO <
0,97
J
0.50
0.03 - 30
I
! 32
i
#4
(a)
i
#15
RETENTION
TIME
i
I
1B
,
413
(MINUTES)
©
©
c
14.36
;~
e°
ru I
°
t
11..9
~
+
+
=
l
o n`
~J
l
£
ru
oJ
IC
I
5.74
~
u
uI
u
j
u
.
u
i °
2.87
:
"i t
.
.
>~*~
uu
.
.
.
uu
©>.>. '~-~-~¢'c
PP'©m umrumm
¢.,.
(b)
24
26
2~
r~
~¢Lo
rum t~u
30
32
34
RETENTION
i
t
o
<°'~'
I
I I~
~i
=
t+j
Ill, i
22
Et Ic c
u>,
0.00
20
~ l~
u
L
~,u~
u
t'~rJ
~~ o mo ~JE
~
¢~ u
uu
i:
©g
+. . . . .
~ ~ ~Z
ra
36 TIME
38
J
I
40
~
(MINUTES)
Fig. 12. Melieha coal: (a) Sterane (m/z 217) and (b) terpane (m/z 191) distributions.
I 44
=
I 46
. -~
50
199
OIL P O T E N T I A L OF MID-JURASSIC C O A L S IN N. E G Y P T 100-
t;I 5
80-
C[7 A >
C15
E o3
60
I t I
X
Z CO w
(;24 40
e¢-
I 20
' ~ ' ~' ' ~ ' ~'o' ~'2' a','
,'~'
(a)
,'0' do' ~'z'';~,' RETENTION
TIME
d6'
~a'
~'S' 3'0' 2o' L "
3'o' ~'~' ~','
2,
(MINUTES)
¢
o
o
2,92
I
x
L
i I 2.34
I" ©
N •
i
o
o~
r,/
× r~
i
.
c
g ©~
1.17
u
ru
u ~ u
~-
('J u
ol Ec
II ~1 Uu u u ,"~,-q u,.~
c~
oo
0.58
0.00
(b)
20
~
] 22'
2.4!
~
~6I
,
/8
,
310
i
312
,
14
RETENTION
i
16 TIME
i
/8
a
/0
~
412
~
414
c° c
;I
i ~ ~
c
c~ E° O
• Q O o
~o
m m
mz U
Y
'~
i
u~ N I •
~
Io
c mo mE UO
,o
nl e• c ~ 1~ I
/6
E
o z
i
ale
~
~I 0
'/2
(MINUTES)
Fig. 13. Meleiha well carbonaceous shale: (a) Cls ÷ gas chromatogram of saturated hydrocarbons and (b) terpane (rn/z 191) distribution.
200
M. A. B A G G E & M. L. K E L L E Y
Recognition is due to Rick Harding, Tarek El Azhary, and Mohammed Said who were all involved in the original geochemical and petrographic work on the Jurassic coals (Bagge et al. 1988), much of which is
presented in this paper. The assistance of Bob Needham and his colleagues at Intera in drafting the figures for this paper is appreciated.
References ABU EL NAGA, M. 1984. Paleozoic and Mesozoic depocentres and hydrocarbon generating areas, northern Western Desert. Proceedings of the 7th EGPC Exploration Seminar, Cairo. American University in Cairo Press. BAGGE, M., HARDING, R., EL AZHARY, T. & SAID, M. 1988. Generation of oil from coal sequences in the Western Desert, Egypt. Proceedings of the
9th EGPC Exploration Seminar, Cairo. BARTHOUX, J. C. & DOUVILLI~,H. 1913. Le Jurassic dans le desert l'est l'Ismthe de Suez. Complete Rendu de l'Academic des Sciences, Paris, 157, 265-268. BERTRAND, P. 1984. Geochemical and petrographic characterisation of humic coals considered as possible oil source rocks. Organic Geochemistry, 6,481-488. KELLEY, M. L., DUNGWORTH, G., FLOYD, C. S., FORBES, G. A., KING, C., MCGARVA, R. M. & SHAW, D. 1990. The Jurassic system in northern Egypt: I. Regional stratigraphy and implications for hydrocarbon prospectivity. Journal of Petroleum Geology, 13, 397-420.
--
t~ WALLIS, R. J. 1991. The Jurassic system in northern Egypt: II. Depositional and tectonic regimes. Journal of Petroleum Geology, 14, 49-64. MOLDOWAN, J. M., FAGO, F. J., HUIZINGA, B. J. & JACOBSON, S. R. 1991. Analysis of oleanane and its occurrence in Upper Cretaceous rocks. In:
Organic Geochemistry. 15th meeting of the EAOG. Poster Abstracts. MANNING, D. (ed.) Manchester University Press. PARKER, J. R. 1982. Hydrocarbon habitat of the Western Desert, Egypt. Proceedings of the 6th EGPC Exploration Seminar, Cairo. Khattab Press Cairo. PHILP, R. P. (~ GILBERT, Z. D. 1986. Biomarker distributions in Australian oils predominantly derived from terrigenous source material. Organic Geochemistry, 10, 73-84. WAPLES, D. W. (~ MACHIHARA,T. 1990. Application of sterane and triterpane biomaerks in petroleum exploration. Bulletin of Canadian Petroleum Geology, 38,357-380.
Coal and coal-bearing strata as oil-prone source rocks: current problems and future directions ANDREW
C. S C O T T 1 & A N D R E W
J. F L E E T 2
1Department of Geology, Royal Holloway University of London, Egham, Surrey TW20 OEX, UK 2BP Exploration, BP Research and Engineering Centre, Chertsey Road, Sunbury-on- Thames, Middlesex TW16 7LN, UK Abstract: At present, the principal problems concerning coaly sequences as oil-prone source rocks are, still, how can we identify which coals expel oil (liquid-phase petroleum) and how do we predict the presence and distribution of such coals in the sub-surface? To tackle these problems four key areas of study need addressing. Firstly, the representatives of the database, on which our current empirical understanding is based, needs verification. Secondly, the nomenclature used to describe coals and related kerogens must be clearly integrated. Thirdly, the constituents of these coals and kerogens, their botanical precursor and their petroleum products need to be established. Finally, controls on oil migration out of coal-bearing sequences need elucidating. Ways of addressing these problems might include a multi-/interdisciplinary study, 'bottom-up' studies from plant constituents, and comparative migration studies based on suites of coals from different settings.
This report summarizes ideas arising from the discussion of the papers in this volume and of the nature and formation of oil-prone coals in general which took place at the meeting on 'Coal and coal-bearing strata as oil-prone source rocks' held in London in June 1992. The discussion gave rise to a number of important issues which can be categorized under three headings: • what we know, • what we don't know, and • what we would like to know. Those at the conference came from a diversity of both background, commercial and academic, and of discipline, through the range of geosciences from geochemistry to palaeobotany. With such diversity it is inevitable that different approaches could be made to help solve problems. Yet not all were agreed as to the most pressing problems to solve nor how best to go about solving them. Despite this, a number of recurring themes came out during the discussions which we highlight here. From a practical point of view there are two areas of need. (1) Rules of thumb for predicting the occurrence and distribution of oil-prone coalbearing sequences for petroleum exploration. The prediction of their presence being a first step in predicting the g a s : o i l ratio ( G O R ) or c o n d e n s a t e : g a s ratio ( C G R ) of reservoired petroleum. This goes for
sequences with dispersed terriginous kerogen as well as coals sensu stricto. (2) Understanding the critical factors (see below) which make coals oil prone. Our current global database provides the foundations for empirical approaches to prediction of presence and of G O R / C G R . The database is considered good (at least that was the consensus of the meeting) but does involve circular reasoning and hence is likely to blinker our perception and very possibly lead to lost opportunities. W h e t h e r or not we accept the outcomes from the database we need a fundamental understanding if we are to make a 'quantum' leap forward.
Are coal-bearing strata important as oil-prone source rocks? It could be argued that given the importance of marine organic-rich sediments as oil source rocks there is little need to consider terrestrial sources. There is, however, a danger of circular reasoning: if we believe that terrestrially sourced oils are of little, or very local, significance then we will tend only to look in marine and lacustrine sequences for source rocks and so should not be surprised if our original hypothesis is reenforced. Since the time when Hedberg (1968) and Brooks & Smith (1967, 1969) demonstrated that some oils may have been generated from land
From Scott, A. C. & Fleet, A. J. (eds), 1994, Coaland Coal-bearingStrataas Oil-prone SourceRocks? Geological Society Special Publication No. 77, pp. 201-205.
201
202
A. C. SCOTT & A. J. FLEET
plant material there has been increasing interest in the possibility of major oil fields being predominantly sourced from coal-bearing sequences (see Murchison 1987 for review). The distribution of significant oil fields yielding terrestrially sourced oil at first sight appears to be well restricted both stratigraphically and geographically which hints at a possible palaeobotanical control. It is true that most of the oil fields with significant terrestrial input are from Australia, New Zealand and Indonesia and are mainly Mesozoic or more particularly Tertiary (Macgregor). However, it is now clear that other areas such as Canada (Beaufort-Mackenzie) should be included, as well as late Palaeozoic coal-bearing sequences in Australia. It is possible that a broader search using standard methods of identification might lead to a range of terrestrial sources for oil (see for example the contrasting conclusions about the Jurassic coalbearing sequences of Egypt by Bagge & Keeley and by Thompson et M.). Clearly, if the age and geographical distribution of terrestrially sourced oils were really restricted it might help limit the possible reasons for the existence of oil-prone coals such as botanical source, palaeogeography or climate. (Although in this volume some regions, e.g. Gulf of Mexico, northern South America, East Africa, Niger and the Arctic, receive less specific attention than others, references are made in many of the papers to these areas.) The recognition that terrestrial organic matter is a contributor to oil has come from a range of geochemical techniques such as biomarker analysis (Philp) and pyrolysis (Powell & Boreham; Stout). In many cases the nature of the terrestrial organic input is not specified. This problem is discussed below but it is significant if a fundamental model of oil generation from higher land plants is to be formulated. Nomenclature remains an important problem. Are we concerned with terrestrial organic matter, higher plant organic matter, coal or carbonaceous shales? All are overlapping concepts. How critical is bacterial reworking or the presence of algal coals? In general, authors have not fully addressed any of these problems. We shall see that nomenclature can hide potentially useful data but also mask the absence of good data.
Many of the oils which have been accepted as being terrestrially sourced are waxy oils (Powell & McKirdy 1976) which some authors considered formed from peat-forming sequences. Some authors (Thomas 1982) consider that these oils related to the occurrence of gymnosperms in the Mesozoic and Tertiary and other authors have presented arguments to suggest a specific botanical origin of some of the fractions in the oils (Shanmugam 1985). The problems of the botanical composition of coals have been discussed elsewhere (Collinson & Scott 1987) but given that we know little of the original peatforming flora it follows that we know less of these components which might generate oil. The input of resins to some oils has proven significant in recent years (van Aarrsen et al. 1990; Snowdon 1991; Collinson et al.) but as pointed out by Collinson et al. care must be taken in attributing resins to parent plants as we lack an adequate database. Another approach to this problem is the study of the potential source rocks themselves. It would be useful to have some 'rules of thumb'. For example can we use figures for the Hydrogen Index and pyrolysis data to specify a significant terrestrial source (Powell & Boreham)? Can we make better use of pyrolysisGC-MS and Powell's mass balance approach (recognizing the problems of using lab pyrolysis to simulate natural conditions)? To what extent could the characterization of the organic matter be used and if so should we use a coal perological or palaeobotanical approach? To what extent could bacterial reworking have an effect and how could this be recognized? If we consider coal then clearly there is a problem of heterogeneity (Stout). The variation in methods illustrated in this volume indicates that we are far from any consensus of characterization. To the direct observational approach may be added an experimental approach. Hydrous pyrolysis experiments may produce oils from organic materials and both source and products may be studied, but the variety of organic source (both in terms of plant species, organisms and chemistry) can complicate interpretations.
Continuing problems Nature o f the organic matter
Nature of the evidence Much of the data concerning the origins of oil has been based on geochemical evidence (Philp). This may be related to specific biomarkers or to data such as carbon isotopes.
A major problem relates to the nomenclature of organic material. Should material be described using terms related to coal petrology (i.e. use macerals), or one of the variety of kerogen classifications or be botanically based (see
CURRENT PROBLEMS AND FUTURE DIRECTIONS various alternatives in Murchison 1987)? There are clearly problems in the widely differing schemes (as shown by the diversity of schemes used by the authors in this volume). The problem is on several scales. One relates to the terms terrestrial, higher plant, coal and what is meant by the different authors. There is, as yet, no consensus on definition but many authors have taken a very wide view (e.g. Collinson et al.) which at this stage is probably necessary. Whilst, however, this volume has been concerned with coal and coal-bearing strata as oil-prone source rocks, it is certain that terrestrial organic matter is washed into marine and lacustrine environments and may contribute to their oil-prone kerogens. There have been several attempts to interpret the botanical source of oils. Two main approaches have been made: geochemical and palynological. Biomarker evidence is most widely used and some authors have been very specific as to their suggested sources (e.g. Thomas 1982; Shanmugam 1985). Kauri pines vegetation is widely cited but, as Collinson et al. point out, the palaeobotanical evidence is equivocal. The signatures from resins have also been widely cited following their wide recognition in some oils (see van Aarssen et al. 1990) but recent work using the collaboration of palaeobotanists at Royal Holloway and organic geochemists at Delft (now at Texel) have shown that some resin types are botanically more widespread than previously thought and care is needed in the interpretation of the botanical source of oils (van Aarssen et al. in press; Collinson et al. ). It is also clear from the new chemical analysis of well-characterized plant material (see Collinson et al.) that our ideas of the relationship between plant parts, coal macerals, kerogen and their potential to generate oil needs urgent revision. It is clear that whilst some Liptinite macerals such as cutinite may be highly aliphatic, others such as sporonite have relatively high aromatic contents (Tegelaar et al. 1991; Hemsley et al. 1992). Equally there has been shown to be a diversity of chemistry of a single tissue type (such as cuticles with cutan and with cutin: Tegalaar et al. 1991). In other cases some macerals groups such as virtinite, thought not to be oil prone, contain tissues such as periderm which is highly aliphatic (Collinson et ai.). It is clear from this work that we lack fundamental knowledge on the chemistry of both modern and fossil plants and we cannot make assumptions on the oil-generating potential of any higher plants until these basic data have been obtained.
203
Clearly there is also room here for the role of oil-generation experiments using known plant precursors such as discussed by Tegelaar et al. (1989) using cutan. Whilst we lack knowledge of plant chemistry, we also lack knowledge of the botanical composition of coals. It is clear that the plants found in sediments between coals may not represent the coal-forming flora (see Collinson & Scott 1987). Equally, interpretations of vegetation based on coal petrology are also hazardous (Collinson & Scott 1987). Much significance has been placed on palynological analysis to interpret original vegetation (Thomas 1982; Shanmugam 1985; Cole 1987) but these can also cause problems as we do not know the botanical identity of all palynomorphs and, in addition, palynomorph composition may not always be a good guide to the botanical composition of a peat. In many publications, statements on botanical origin are made without the supporting data (see Coilinson et al. for a review). This criticism equally applies to the interpretation of coal-forming environments (see Scott 1987). There is the equal problem of the role of bacteria and fungi in the system, including in microbial activity and early diagenesis in general, besides unravelling the role of individual plant components. Neither do we understand the ways in which organic matter inputs/accumulation (preservation) rates differ in different environments with differing tectonic styles/climate/geography/geological time: all play their part. Problems of expulsion
If oil can be generated from at least some terrestrial organic matter why do we not find more in reservoired accumulations? Whilst it may be argued that most exploration effort has been towards marine basins, the fact that relatively few fields have so far been discovered is significant. A major consideration is not so much perhaps oil formation, but oil expulsion. There appears retention of generated of generated oil in the coals, for instance by adsorption on the coal matrix as demonstrated by Stout. If oil is not expelled as a liquid then it will remain in the coal unit it is cracked to gas and expelled, or until it is expelled dissolved in the gas phase. This is one area that needs particular further study. We need to know the controls upon expulsion. Do the critical ones (e.g. fractures, adsorption) vary from mudrocks to coal? What part, if any, does the mineral matrix play?
204
A. C. SCOTT & A. J. FLEET
Problems of migration
A significant feature of coal-bearing sequences is in their cyclic nature. If an oil can be expelled from one coal could it be absorbed by another? Equally, if oil was to be generated by coaly shales might it be absorbed by coal seams in the same sequence? There is also a wide range of basin architecture and also sandstone body architecture associated with coal-bearing sequences (McCabe 1987, 1991). There has been little study of the effect of lithological patterns and basin architecture on the problems of oil migration in coal basins.
Possible solutions
For the petroleum explorer/geoscientist there is the need to develop 'rules of thumb' in indentifying and predicting oil-prone coaly sequences. It is clear that the development of standard analytic techniques is important as is the development of a standard nomenclature. Most solutions, however, have been sought by the use of an empirical approach. The development of a database is widely considered important. There are, however, several alternative approaches. • It is clear from most published work that any one study uses only a fraction of research approaches possible. There is a clear need to study a single system using all techniques available - geochemical, sedimentological, palaeobotanical. Such a case study could provide many useful insights and help
develop new research strategies, for instance in developing thinking on the roles of catalysis in primary migration and of basin architecture controls in secondary migration. • A more fundamental approach might also be made: from the 'bottom' up. A more detailed analysis of the chemistry of plant fossils is needed to help understand which tissues and which plants are oil prone. A greater understanding of coal-forming vegetation is also needed. • Following the chemical characterization of plant tissues, hydrous pyrolysis experiments could be undertaken to help understand the oil-generating potential of the materials and in addition the resultant oils could be geochemically characterized. • Problems of oil generation and migration could be studied using a suite of coals from different stratigraphic ages, geographical regions and vegetation types. Clearly the approach chosen will reflect the disciplines most widely used by a particular research group but what is most needed is an overall multi- and interdisciplinary approach (or, at very least, a well-networked multi-group) to what continues to be a fascinating and important problem: the sourcing of oil from coal-bearing strata. Many of the above ideas came from discussion at the end of the meeting on 'Coal and coal-bearing strata as oil-prone source rocks' which was held at the Geological Society, London in June 1992: we sincerely thank all participants for their contributions.
References
VAN AARSSEN, B. G. K., Cox, H. C., HOOGENDOORN, P. & DE LEEUW, J. W. 1990. A cadinene bio-
polymer in fossil and extant dammar resins as a source for cadinenes and bicadinenes in crude oils from South East Asia. Geochimica et Cosmochimica Acta, 54, 3021-3031. - - , DE LEEUW,J. W., COLLINSON,M. E., BOON,J. J. & GOXH, K. in press. Occurrence of polycadinene in fossil and recent resins. Geochimica et Cosmochima Acta. BROOKS, J. D. & SMITH, J. W. 1967. The diagenesis of plant lipids during the formation of coal, petroleum and natural gas. I. Changes in the n-paraffin hydrocarbons. Geochimica et Cosmochimica Acta, 31, 2389-97. -& 1969. The diagenesis of plant lipids during the formation of coal, petroleum and natural gas. II. Coalification and the formation of oil and gas in the Gippsland Basin. Geochimica et Cosmochimica Acta, 33, 1183-94. COLE, J. M. 1987. Some fresh/brackish water depositional environments in the S.E. Asian
Tertiary with emphasis on coal bearing and lacustrine deposits and their source rock potential. Proceedings of the Indonesian Petroleum Association 1987, 11/05, 429-449. COLLINSON,M. E. & Scoa-r, A. C. 1987. Implications of vegetational change through the geological record on models for coal-forming environments. In: SCOTT, A. C. (ed.) Coal and Coalbearing Strata: Recent Advances. Geological Society, London, Special Publication, 32, 67-85. HEDBERO, H. D. 1968. Significance of high-wax oils with respect to genesis of petroleum. American Association of Petroleum Geologists Bulletin, 52, 736-750. HEMSLEY,A. R., CHALONER,W. G., SCOTT,A. C. & GROOMBRIDGE, C. J. 1992. Carbon-13 Solid-state
Nuclear Magnetic Resonance of Sporopollenins from Modern and Fossil Plants. Annals of Botany, 69,545-9. MCCABE, P. J. 1987. Facies studies of coal and coalbearing strata. In: Scoa~r, A. C. (ed.). Coal and Coal-bearing Strata: Recent Advances. Geo-
CURRENT PROBLEMS AND FUTURE DIRECTIONS logical Society, London, Special Publication, 32, 51-66. 1991. Tectonic controls on coal accumulation. In: BERTRAND,P. (ed.). Coal: Formation occurrence and related properties. Bulletin de la Soci~td G6ologique de France, 162,277-282. MURCHISON, D. G. 1987. Recent advances in organic petrology and organic geochemistry: an overview with some reference to 'oil from coal'. In: Scoxr, A. C. (ed.). Coal and Coal-bearing Strata: Recent Advances. Geological Society, London. Special Publication, 32,257-302. POWELL,T. G. & MCKIRDY,D. M. 1976. Geochemical character of crude oils from Australia and New Guinea. In: LESLIE,B., EVANS, H. J. & KNIGHT, C. L. (eds) Economic Geology of Australia and Papua New Guinea. 3. Petroleum. Australasian Institute of Mining and Metallurgy, Monograph 7, 18-29. SCOTT, A. C. (ed.) 1987. Coal and Coal-bearing Strata: Recent Advances. Geological Society, London, Special Publication, 32.
205
SHANMUGAM,G. 1985. Significance of coniferous rain forests and related organic matter in generating commercial quantities of oil, Gippsland Basin, Australia. American Association of Petroleum Geologists Bulletin, 69, 1241-1254. SNOWDON, L. R. 1991. Oil from type III organic matter; resinite revisited. Organic Geochemistry, 17,743-747. TEGELAAR, E. W., KERP, H., VISSCHER,H., SCHENCK, P. A. & DE LEEUW, J. W. 1991. Bias of the paleobotanical record as a consequence of variations in the chemical composition of higher vascular plant cuticles. Paleobiology, 17, 133-144. , MATTHEZING, R. M., JANSEN, J. B. H., HORSFIELD,B. & DE LEEUW,J. W. 1989. Possible origin of n-alkanes in high-wax crude oils. Nature, 342,529-531. THOMAS, B. i . 1982. Land-plant source rocks for oil and their significance in Australian basins. Australian Petroleum Exploration Association Journal, 22, 164-178.
Index
abietane 78, 80 Abu Roash Formation 124 A g a t h i s 34 age significance of coals 4, 17,202 Akata Shales 3 Alethopteris 44, 45, 47, 54 see also pteridosperms algae as sources 11, 12, 36 algal cell walls 62 chemistry 47-8 oil potential 51-5 algal coals 32 algaenan 31, 51 alginite 15, 94 algite 94 aliphatic hydrocarbons analysis 97-100 significance 153-8,164 n-alkanes 41-63, 73 n-alk-l-enes 41-63 angiosperms 3, 24, 33,113 biomarkers 80, 104 see also flowering plants, N o t h o f a g u s , vascular plants Araucariaceae in coal 23, 34, 36 geochemical character 79, 81 see also gymnosperms archeabacteria 75, 76 Ardjuna Basin 12, 13, 16, 20 aromatic chemical biomarkers 79-81,164 asphaltene 73 Assam 5,108 Athabasca tar sands 79 attrinite 95, 96 Australia 17 see also Bowen/Surat Basin; ClarenceMoreton Basin; Cooper Basin; Eromanga Basin; Gippsland Basin; Sydney Basin A vicennia 35 bacteria as sources 11, 75, 76, 82, 103,120 see also archeabacteria; cyanobacteria bacterial biomass 15, 23 Baffin Bay 73 Balkash, Lake 48 balkashite 48 Bampo Formation 139 Bangladesh 108 Banuwati Formation 122 Baong Formation 139,143 bark 56
Barracouta oil 75 Beaufort-Mackenzie Basin age 5, 13 biomarker character 72 Hydrogen Index data 16 naphthenic oil 18, 21 oil composition 22 resinite study 17-18 sources 36 Belumai Formation 139,140 beyerane 78, 80 bicadinanes 23,120, 124, 153 biomarkers 2, 23-4, 72, 73,120, 185 Cooper Basin 174-5,178-9 Sumatra oils 143-4 Taranaki Basin 172-3,176-7 see also n a m e d chemicals
bisnorhopane 72, 85 bisnorlupanes 24, 85,120 bituminous coal characteristics 23,194, 195 Boghead coal 32, 36 Botryococcus
chemistry 51 as source 32, 35, 36, 48, 73 see also algae Bowen/Surat Basin age 12, 109 biomarker character 72 source rocks 13, 34 Bowland Shales 129,135 Brazil 33, 36 Brazil Formation 39 British coal balls 45-6 Brent Coal 4, 6 brown coal characteristics 23 Bruksah Formation 139 Brunei 13 bryophytes 80
13C 3, 25,120 Egyptian sources 128 N England sources 134 NMR 49, 50, 52, 55 Sumatra oils 124, 144-5 Talang Akar oils 122 cadalanes 77 cadinene 35 Calamites 33 see also pteridophytes Canada see Beaufort-Mackenzie Basin; Scotian/Labrador Shelf carbonaceous shale, role of 13-14 207
208 Carboniferous coal composition 32-3 palaeogeography 130, 131 plant fossil material 37-47 cedrene 77 Ceratopteris 35 Cerdanya oil shale 36 charcoal 6 see also fusinite China 32 see also Junggar Province; Liaoning Province; Qaidam Province; Songliao Basin; Turpan Basin Clarence-Moreton Basin 18, 20 clarite 94 clarodurite 94 coal compositional variations 32-4 defined 32 depositional environments 14 significance as oil source 1 coal balls British 45--6 Indiana 41-4 Kansas 44-5 coking reactions 21 collinite 15, 17, 94 collite 94 communic acid 37 conifers 3 biomarkers 23-4 as sources 13, 23, 33 see also Cupressaceae, G l y p t o s t r o b u s , Kauri pine, Pinaceae, Podocarpaceae, Taxodiaceae Cooper Basin 2,108 age 12 coal biomarker ratios 24, 174-5,178-9 depositional environment 13, 14 geochemical characterization 159-64 geochemical data discussed 164-9 methyl phenanthrene index 21 petrography 17 plant sources 34 Rock-Eval 16, 19, 159, 171 vitrinite reflectance 170 gas 25 microbial biomass, role of 2, 3 oil composition 22 generation 20 setting 13, 14, 159 stratigraphy 169 Coorong 48 coorongite 48 cordaites 33 see also gymnosperms
INDEX cork tissue 17 Cornaceae 35 see also angiosperms Cretaceous coal composition 33, 34 latitude associations 111 regional studies see Assam; Gippsland Basin; Myanmar Province; Niger delta; Taranaki Basin cuparene 77 Cupressaceae 36, 79 see also gymnosperms, conifers cutan 31, 41, 49, 63, 73 cuticle 62 oil potential 49-51, 98 cutin 203 •cutinite 15, 17, 53, 94, 96, 99,203 cutinoclarite 94 cutinodurite 94 cutite 94 cyanobacteria 32 cycadeoids 34 cycadophytes 33 dammar resin 35, 84, 86 Daquing oilfield 36 Deerplay 45--6 deltaic environments 3, 14, 119-20 Egyptian model 186 N England model 133,135 desmocollinites 99 Devonian 32 D i a p h o r o d e n d r o n 46, 56, 57 diasteranes 23 Dipterocarpaceae 35 see also angiosperms diterpanes 23-4, 72, 77-81,120 diterpenoid resins 62 drimanes 23, 77 durite 94 duroclarite 94 Edale Shales 129 Egypt coal geochemistry 122-9 as a source rock 5,190--4 stratigraphy 183-4 oil characteristics 127, 184--~9 provinces 108,111 England (North) basin settings 129 oil sources 5, !29-35 environment of deposition, effect of 63, 71, 75 Eocene basins see Beaufort-Mackenzie
INDEX equisetaleans 34 see also pteridophytes Eromanga Basin age 12 biomarker character 23--4 oil composition 22, 79-81 sources 13, 34 E s k d a l i a 37-9, 51 see also lycophytes Estonia 48 eudesmanes 77 exinite 3, 6, 15 expulsion problems 94,203 factors affecting 4, 19, 20, 63,100, 104 exsudatinite 15, 17, 100 ferns 3, 34, 80 see also pteridophytes Fife 46 Fletning coal 45 Flemingites scottii 46 see also lycophytes flowering plants, rise of 33 see also angiosperms Fushun coalfield 37 fusinite 15, 17, 94 fusite 94 gas chromatography 123,126,127,152,187,189 with mass spectrometry (GCMS) 95-6, 141, 143-4 with pyrolysis 48-9, 156, 163,165,192 gas" oil ratio (GOR) 4-5, 18 gas potential 18-19, 94 gas reserves 110 Gebel Maghara coals 124, 190, 194 gelinite 100 geochemistry bulk 71-3 specific compounds n-alkanes 73-4 diterpanes 77-81 isoprenoids 75-6 pentacyclic terpanes 82-6 sesquiterpanes 76-7 sesterterpanes 81-2 steranes 87-8 German gas reserves 25, i09 ginkgoaleans 33 see also gymnosperms Gippsland Basin 107 age 12,108 biomarker character 72 gas composition 25 oil composition 22, 24 oil geochemistry 75, 78, 82, 84
209
PGI 20 sources 13, 34, 111,113 G l o e o c a p s o m o r p h a 32, 48, 51 see also algae glossopterids 73 see also gymnosperms G l y p t o s t r o b u s 33 Green River Formation 36 Guchengzi Formation 37 gymnosperms 33, 63,113,202 biomarkers 80 see also conifers, cordaites, cycadeoids, cycadophytes, glossopterids, pteridosperms Haltenbanken 5, 13,109,110 Herrin Coal 46 high wax oil 21 hopanes 23, 72, 77, 82--4 hopanoids 102 hopenes 102 humic coal 94 huminite 100 humodetrinite 95 hydrocarbon ratio 2 Hydrogen Indices 2, 15, 16, 114, 202 Brent coals 6 Cooper Basin 159 Egyptian oils 124,128, 190 Indonesian kerogen 122 N England sources 134 Taranaki sources 152 India 108 Indiana coal balls 41-5 Indiana paper coal 39-41 Indonesia 120-2 see also Ardjuna Basin; Java; Kutei Basin; Mahakam delta; Natuna; Pasir Subbasin; Sumatra Basin; Sunda Sub-basin; Taranaki Basin Indus Basin 109 inertinite 6, 15, 17, 94 inertite 94 isoprenoids 23, 75-6, 100-2 isotopic characterization see 13C Java 13,108, 110, 114, 122 see also Ardjuna Basin Junggar Province 108 Jurassic coal composition 33, 34 latitude associations 111 regional studies see Egypt; Eromanga Basin; Haltenbanken; Turpan Basin
210 Kalimantan 12, 13 see also Indonesia Kansas coal balls 44-5 Kapuni Group 150 Karinopteris 39-41, 51 kaurane 78, 79, 80 Kauri pine 23, 34,203 see also gymnosperms kerogen 1, 36-7 classification 16, 20 isotopic characterization 3 Khatatba Formation coal 124-5,126, 183 kukersite 48, 56 Kutei Basin 108, 110, 113,114 see also Mahakam delta Kuznetsk Basin 32
labdane 78, 80 Labrador Shelf 22 lacustrine sediments 1, 11 lamosites 36 laser extraction 95-6 latitude, effect on coaly sources 111-12 Latrobe group 34 Laveineopteris 44 Lepidodendron 46, 56 Lepidophloios 56 Liaoning Province 37 Linopteris 44 liptinite 2, 93,203 geographical variation 114, 120 hydrocarbon potential 15, 17 liptite 93, 94 liptoclarite 94, 96 composition 97, 99,100, 101,102, 103 liptodetrinite 15, 17, 94, 96, 99 liptodetrinodurite 94 liptodetrite 94 liverworts 80 Lonchopteris 44 Luconia 108 lupanes 84 lycophytes 32, 33, 34, 45-6, 62 periderm 45-6 spores 46, 57 stem cuticle 37-9 see also Diaphorodendron, Eskdalia, Lepidodendron, Lepidophloios, Sigillaria, Stigmaria lycopsids 14 see also lycophytes Lycospora 46, 59
maceral nomenclature 15 macrinite 15, 17
INDEX Mahakam delta age 13 biomarker character 72 gas composition 25 gas : oil ratio 5 Hydrogen Index 16 oil composition 21 oil maturation 20 oil sources 34-6 see also Indonesia Malaysia 108 mangroves and coal formation 35,120 Maracaibo 108 mass balance assessment 20-1, 26 mass spectrometry 95-6, 141,143-4 maturity measures 145 medullosan pteridosperms 41-5 see also gymnosperms Messel, Lake 47-8 methyl phenanthrene index 21 micrinite 15, 17 microbial biomass 2, 3, 5, 14-15 microlithotypes 93, 94 methods of analysis 95 results of analysis laser extraction 95-7 organic petrology 95 results discussed acyclic isoprenoids 100-2 n-aliphatics 97-100 pentacyclic triterpenoids 102--4 migration problems 94,204 Miocene studies see Kutei Basin; Mahakam delta; Talang Akar Formation coal Moscow Basin 37-9 mosses 33, 80, 120 Myanmar Province 5 Myeloxylon 44
naphthenic oils 13, 18, 21 Natuna 109 Neuropteris 44 New Zealand see Taranaki Basin Niger delta 2-3, 5, 13, 16, 72 Nigeria 13, 16, 72 Nipa 35 NMR 49, 50, 52, 55 North Sea Province 5, 13,109, 110 Brent coals 4, 6 gas pools 20, 109, 110 Norway 5, 13,109,110 Nothofagus 34
Odontopteris 44 see also pteridosperms
INDEX oil generating potential measurement 48-9, 119, 149 results 49-50 algal walls 51-5 cuticles 51 periderm 51-7 resins 57 spores 57 results discussed 61-4 oil reserves 107-110 oleanane 3, 84, 124 oleanes 24, 72, 73, 84, 120 Oligocene studies see Ardjuna Basin; Talang Akar Formation coal oligoterpenoids 35 Ordovician 32, 48 O r e s t o v i a d e v o n i c a 32 organic matter classification 15-16 Oxygen Index Egyptian coals 124, 128 Indonesian kerogen 122 N England sources 134 Taranaki sources 152 ozocerite 74 Pakistan 109 palaeobotany 31-70,202,203 Palaeocene basins see Beaufort-Mackenzie palaeolatitude effect on coaly sources 111-12 paper coal Indiana 39-41 Russian 37-9 paraffins, significance of 11 Parapat Formation 139 Paripteris 44 see also pteridosperms Pasir Sub-basin microlithotype study methods of analysis 95 results laser extraction 95-7 organic petrology 95 results discussed acyclic isoprenoids 100--2 n-aliphatics 97-100 pentacyclic triterpenoids 102-4 Patchawarra coals 20 peat 5, 14, 15,119-20,202 P e d i a s t r u m 35, 36, 37, 51 see also algae pentacyclic terpanes 82 pentacyclic triterpenoids 102-4 periderm 45-6, 51-7, 62 see also bark Permian coal composition 33 plant fossil material 47 regional studies see Bowen/Surat Basin; Cooper Basin
211
source rocks 111 Perth Basin 22 petroleum expulsion efficiency (PEE) index 20 petroleum generation index (PGI) 20 Pettycur Limestone 46 Philippines 111 phyllocladane 78, 80, 82 phytane see pristane/phytane ratio Pila 48 see also algae pimarane 78 Pinaceae 79 see also gymnosperms Podocarpaceae 34, 79 see also gymnosperms pollen and palynomorphs 32, 34, 35, 62 polycadinene 35, 62 polycommunic acid 34 Precambrian 32 pristane/phytane ratio 2, 23, 75-6, 100 Cooper Basin 159 Egyptian sources 128 N England sources 134 Taranaki sources 152 psilophytes 32 pteridophytes 6, 32 see also Calamites, equisetaleans, ferns, lycophytes, tree ferns pteridosperms 33 cuticle study 39-41, 44-5 medullosan 41-5 see also A l e t h o p t e r i s , K a r i n o p t e r i s , Laveineopteris , Linopteris , Lonchopteris, Myeloxylon, Neuropteris, O d o n t o p t e r i s , Paripteris , R eticulopteris
pyrolysis 19, 48-9,202 methods 48-9 results algal cell walls 51-5 cuticle 49-51 periderm 51-7 resin 57-61 spores 57-9 results discussed 61-4 Qaidam Province 108 reed marsh 33 R e i n s c h i a 48 see also algae resinite 4, 15, 17-18, 94 resinoclarite 94, 95 composition 96, 97, 99,100, 101,102, 103 resinodurite 94 resins and oil potential 24, 57-61,202,203 resite 94
212 Reticulopteris 44 R h a c o p h y t o n 32
Ribesalbes oil shale 36 Richards Formation 13 Rock-Eval 16, 19,183 Cooper Basin 171 Taranaki Basin 152, 171 see also pyrolysis Rundle oil shale 36 Russia 38, 109 Russian paper coal 37-9 Sabden Shales 129 Saigon 108 Salvinia 59 see also pteridophytes sapropelic coal 93-4 see also Boghead coal Sarawak 13 Scenedesmus 51 sclerotinite 15, 17, 94 sclerotite 94 Scotian/Labrador Shelf 22 secohopanes 82 seeds 63 semifusinite 15, 17, 94 semifusite 94 sesquiterpanes 76-7 sesquiterpenoids 35,104 sesterterpanes 72, 81-2 Sigillaria 46, 56 Sihapas coals 110-11 Smorbukk SCr field 5 solaneso176 Songliao Basin 36 Sonneratia 35
South Africa 36 Spain 36 S p h a g n u m 33 see also bryophytes spores oil potential 57-9 sporinite 15, 17, 94, 96,203 sporite 94 sporoclarite 94 sporodurite 94 sporopollenins 32, 37 steranes 23, 72, 87-8 Cooper Basin 161,175,179 Egyptian desert 128 N England 134 Pasir Basin 104 Taranaki Basin 152, 173,177 Stigmaria 46, 56 see also lycophytes suberinite 15, 17 sulphur, significance of 73 Sumatra Basin 108,110, 122, 139
INDEX history of research 139--41 present study results geochemistry 143 maturity 145-7 palaeoenvironment interpretation 143-5 see also Sihapas coals Sunda Sub-basin 122 Surat/Bowen Basin age 12, 109 biomarker character 72 source rocks 13, 34 Surma 108 Swillington 46-7 Swillington Shales 45 Sydney Basin biomarker character 72 gas composition 25 sources rocks 34, 36 Talang Akar Formation coals 13, 20, 120-2, 123 Taranaki Basin 12, 13,108, 150, 169 age 12, 13,108 coal biomarker ratios 72,172-3,176-7 Rock-Eval data 171 vitrinite reflectance 170 oil geochemistry 23,150-8 summary of characters 158-9 tasmanite 47, 54 Tasmanites 47, 51 see also algae Taxodiaceae 79 see also conifers tegmen 62 telinite 15, 17, 94 telite 94, 95, 96 composition 97, 98, 99,100, 101,102, 103 terpanes 72, 76-82 non-hopanoid 84-6 pentacyclic 82 tetracyclic 32, 129 terrigenous gas sources 18-19 composition 25 terrigenous oil sources 12-13 composition 21-5 environment of deposition 13-15 evaluation 19-21 geochemical character 15-17 organic matter composition 15 petrographic character 17 resinite role 17-18 Tertiary coal composition 33, 34 latitudinal associations 111 plant fossil material 47-8
INDEX regional studies see Ardjuna Basin; Beaufort-Mackenzie Basin; Kutei Basin; Mahakam delta; Niger delta; Talang Akar Formation coal tetracyclic terpanes 32 Tetraedron 47-8, 51 see also algae textinite 95, 96, 99, 100 Thailand 109 torbanite 32, 36 see also algae, sapropellic coal tree ferns 33 see also pteridophytes Triassic peats 33 trimacerite 94 triterpanes Cooper Basin 161,175,178 Taranaki Basin 23,153, 173,176 triterpenoids 35,102--4 Tuberculatisporites 46, 47, 58 Turpan Basin 12 Turrum oil 75, 82
vascular plants as sources 34-6 see also angiosperms, gymnosperms, lycophytes, pteridophytes Venezuela 108, 111 Vietnam 108 Vilyuy 109 vitrinerite 94 vitrinertinoliptite 94 vitrinite 6, 15, 17, 99,119, 120,203 reflectance values 128, 134,170 vitrite 94 vitrodetrinite 15, 17
Walloon coal measures 18, 20, 34 wax 11, 21, 71-3, 185 white coal 47 see also Tasmanites
wild fires 6
Yallourn coal 34 Uinta Basin 36 ulminite 95, 96, 99, 100 ursanes 84 USA 36, 39--41, 41-5
zosterophylls 32
213
Classic Earth Science Texts reprinted in paperback...
Coal and Coal-bearing Strata: Recent Advances edited by A. C. Scott
This book is recommended reading for the undergraduate student contemplating a project involving coal and is essential reading for the postgraduate student undertaking research into any of the many fields covered by the book. It should be available in all libraries having geological reference material. A. H. V. Smith in Geological Magazine, vol. 126 no 2 The papers provide a blend of geological, biological and biochemical subjects that highlights the relevance of interdisciplinary approaches to coal studies. D. L. Wolberg and F. J. Kuellmer in Journal of Sedimentary Petrology, vol. 59 no 2 This is the most important book on coal formation in the past 20 years. S. Hazeldine in Modern Geology, vol. 13 Previously published in 1987 as Special Publication 32 ISBN 1-897799-01-2. 332 pages, paperback. Dec 93 List price US$33.50/£19.95. Discounts are available to members of The Geological Society and AAPG.
Geological Society Publishing House, Unit 7, Brassmill Enterprise Centre, Bath BA1 3JN, UK. Tel (0225) 445046. Fax (0225) 442836
A Geological Society Special Publication Classic
Classic Earth Science Texts reprinted in paperback...
Deltas: Sites and Traps for Fossil Fuels edited by M. K. G. Whateley and K. T. Pickering
This book covers the significant advances that have been made in our understanding of deltas but which have not, until now, been collated into one useful reference source. Many aspects of modem and ancient deltaic sedimentary systems are covered in both marine and freshwater environments, including processes, facies models, petroleum-, gas- and coalrelated environments, together with general case studies. Deltas will be of particular interest to researchers, teachers and students of sedimentology, economic and petroleum geology and those who seek a detailed, state-of-the-art overview of this large and ever-expanding subject area.The volume is well illustrated by line diagrams and photographs and includes a comprehensive 16page index.
Previously published in 1989 as Special Publication 41 ISBN 0-903317-98-2. 360 pages, paperback. Nov 93 List price US$33.50/£19.95. Discounts are available to members of The Geological Society and AAPG.
Geological Society Publishing House, Unit 7, Brassmill Enterprise Centre, Bath BA1 3JN, UK. Tel (0225) 445046. Fax (0225) 442836.
A Geological Society Special Publication Classic
Classic Earth Science Texts reprinted in paperback...
Inversion Tectonics edited by M, A. Cooper and G. D. Williams
We owe the petroleum industry a vote of thanks. First they gave us a modem understanding of thrust tectonics. Then came strikeslip tectonics and extensional tectonics. Now out of the north comes inversion tectonics...Part l(Modelling and Theoretical Concepts) describes underlying principles and sand-box experiments...Part 2 (Inversion in the Alps and Alpine Foreland) contains five papers in which the traditional view of inversion as applied to sedimentary basins is extended into the orogenic domain...Part 3 (Inversion on the European Continental Shelf) reverts to 'more conventional' concepts of inversion in sedimentary basins...Part 4 (Inversion in Other Geological Environments) provides examples of inversion...Personally, I found the papers which extended the concept to orogenic terranes particularly interesting and I would recommend the book to both hard and soft-rock workers. R. Glen in Australian Geologist, 1990, no 7
Previously published in 1989 as Special Publication 44 ISBN 0-903317-97-4. 375 pages, paperback. Oct 93 List price US$33.50/£19.95. Discounts are available to members of The Geological Society and AAPG.
Geological Society Publishing House, Unit 7, Brassmill Enterprise Centre, Bath BA1 3JN, UK. Tel (0225) 445046. Fax (0225) 442836
A Geological Society Special Publication Classic
Coal and Coal-bearing Strata as Oil-prone Source Rocks? edited by A.C. Scott (Royal Holloway University of London, UK) and
A.J. Fleet (BP Exploration, UK) The role of coal and c0al-bearing strata in the formation of oil has long been debated. Increasing evidence is being provided, mainly from geochemical data, that coal and coal-bearing strata, at least of some ages and in some places, may give rise to significant quantities of oil. Most arguments concerning oil formation from terrigenous organic matter have been based on an examination of a single source of data (e.g. geochemistry). Many research areas have, however, an impact on the debate including geochemistry, palaeobotany, petroleum and coal geology. The need for a multi- and interdisciplinary approach to the study of the problem is highlighted by this volume. Specific attention is paid to research from different areas and disciplines. Key topics addressed include: l l • •
Where do terrigenous-sourced oils exist and what are the limits of our knowledge of them? Geochemical characterization and interpretation of terrigenous oils. Evolution of plants and implications for oil generation. Oil generation and expulsion from coals and coal-bearing strata.
These key topics are covered in major review chapters which incorporate significant new data. In addition, case studies highlight specific problems or areas of study. This volume will be of interest t o ' all geologists, geochemists and palaeobotanists with interests in petroleum or coal geology, and to both those in industry and academia. It will act as a focus for future research on the general area of petroleum-source rocks and oil-prone coals, in particular. • • • •
208 pages overlO0illustrations 11 papers index
Cover illustration: Thin section (2 m m width) of Swallow Wood Coal, Middle Coal Measures, Westphalian B, Yorkshire, UK. Natural History Museum V.13511 (red vitrinite ground mass with yellow mega- and microspores (sporinite)).