See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/303200484

oil, correlationOil–oil and oil–source rock correlation

Chapter · January 1998

DOI: 10.1007/1-4020-4496-8_222

CITATION READS

1 15,091

1 author:

Clifford Walters
University of Texas at Austin
96 PUBLICATIONS   5,878 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Clifford Walters on 27 May 2016.

The user has requested enhancement of the downloaded file.

 0

macromolecular material is transformed to increasingly more

OCCURRENCE OF ORGANIC FACIES stable structures by aromatization. Cleavage of carbon bonds
will produce smaller molecules, which in sedimentary condi-
Organic matter occurs in nature in dispersed and more concen- tions favorable to migration, can lead to accumulation ofliquid
trated forms, both liquid and solid as well as hydrocarbon forms of organic matter recognized as crude oil deposits.
gases and often in economically important quantities as fossil Kerogen therefore will vary in its chemistry and chemical
fuels. Conventionally, the term organic facies has described structure according to its origin, which in turn will be strongly
the occurrence of dispersed organic matter in mappable sedi- influenced by the environment of deposition, and to its level
mentary units of similar organic richness and organic type of maturity. Kerogen types are defined in terms of origin.
reflecting a particular depositional environment. This article Organic matter predominantly from marine organisms is
uses the term organic facies to encompass the several styles of described as type I (high H/C ratio), terrestrial higher plants
major occurrences of organic matter in the geosphere. (lower H/C and high 0/C initial ratios) form a type III kero-
gen, whilst a mix of these two categories can be recognized as
a type II kerogen. Some geochemists would now add a fourth
Kerogen as dispersed organic matter category (type IV) derived from bacterial sources which can
Kerogen is the term used to describe the occurrence of solid be important in Precambrian rocks.
organic matter in dispersed sediments and shales. With the Useful parameters in defining an organic facies
growth in understanding of the chemical formation of kerogens
there has been a move from the historically narrower definition At the time of deposition of the kerogen precursors a small
of kerogen as organic matter in oil shales to the broader proportion of organic molecules are incorporated into the
definition of kerogen as organic matter insoluble in organic sediments but do not become chemically bound to the kerogen
and aqueous alkaline solvents, thus differentiating it from and are described as biomarkers. Because they are derived
humic acids defined as soluble in aqueous base but insoluble ultimately from a living organism, biomarkers have the poten-
tial to indicate an input source which will in turn be determined
in acid and commonly found in Young and Recent sediments.
by the depositional environment. Biomarkers have therefore
Kerogen is about 1000 times more abundant than coal or
been useful as one parameter to identify an organic facies.
petroleum. Defined organic remnants in sediments represent
Conventionally geologists have used mineral assemblages to
only a small fraction of the organic matter even though identi- characterize a particular sedimentary environment. For exam-
fication of these remnants as algae, spores, pollen, leafy tissues ple, a sediment low in calcium and high in siderite would
can be valuable in identifying source inputs to the kerogen. suggest a freshwater, anoxic environment. Consideration of
Like coal, kerogen is a macromolecule; like coal it has organic geochemical parameters such as total organic carbon
undergone subsurface maturation. Unlike coalification which (TOC) concentrations, rock-eva! pyrolysis, GC traces of ali-
assumes a carbon (coal) catalytic surface, kerogen formation phatic hydrocarbon (particularly n-alkane) patterns and com-
usually is associated with maturation of organic matter on clay position have the potential of refining an understanding of the
and/or carbonate surfaces. In an environment of deposition, paleodepositional environment further; when TOC is minimal
cellular material is contributed to and preserved in quantities and n-alkanes in the > C25 carbon region show a characteristi-
governed by productivity of the water column and oxicity of cally high odd/even ratio the depositional situation was proba-
the benthic sediment as well as the rate of deposition of miner- bly that of an alluvial fan.
als and microbial utilization. Subsurface heating (maturation)
will rapidly transform the organic cellular detritus by loss of Markers of some paleoenvironments
functional groups with concurrent chemical cross-linking to Abundances of TOC in sediments of carbonate platforms are
form insoluble organic macromolecules. With compaction, the strongly related to individual sub-environments. Higher values
440 OCCURRENCE OF ORGANIC FACIES

of TOC will characterize sediments such as those collected When maturation of kerogen proceeds into the zone of
from marginal zones where mangrove-lined coasts or algal- metamorphism the heat treatment at depth may be prolonged.
grass meadows existed and the influx of terrigenous organic Typical carbonaceous products are hydrocarbon gases such as
matter will diminish with distance from the land. A strong methane and diamond, the latter being a pure form of carbon
relationship exists between grain size, present day hydraulic occurring within specific lithofacies. Natural gas, whilst pre-
conditions and organic facies, and in most modern continental dominantly methane, can contain a range of other gaseous
shelves these constraints will reflect in TOC content as hydrocarbons including propane and butane.
described above.
Biomarker indicators of paleoenvironments are not yet fully Peat and coal
developed but understandings are strengthening as the data Peat is a form of organic matter deposited in moist environ-
bases grow in number. The C25 regular isoprenoid hydro- ments over Recent to Contemporary time scales. These envi-
carbons, squalane andy- and jl-carotanes, are taken to indicate ronments restrict the botanical species of higher plants which
either high salinity or methanogenesis. These same biomarkers can contribute to the organic matter and chemically, the latter
together with a relative abundance of regular steranes which has been found to have a high fulvic acid composition as
in turn suggests a high algal organic content and with host distinct from humic acids found in soils and in brown coals.
rocks rich in pyrite and dolomite, these parameters can point Peat is considered by some coal geologists as the precursor of
to the depositional environment being a stagnant playa lake. low rank coals.
A high proportion of hopanes signifies content of lipids Coal represents an occurrence of a type of organic matter
derived from aerobic bacteria and/or a higher content of higher which occurs in quantity but in restricted geologic structures
plant material. However, hopanes are also produced by cyano- such as a basin or depression and undergoes coalification
bacteria and their presence could be taken to suggest the (maturation) in situ. Coals are normally sourced from higher
presence of cyanobacterial mats at the margin of a basin. To plant residues except in limited deposits derived from algal
define an organic facies, therefore, on the basis of limited precursors, when they are described as cannel or boghead
biomarker content can be equivocal but additional information coals. Current descriptions of coal structure include a host/
on the mineralogy of the sample may help to resolve ambiguity. guest model where the host is the macromolecular matrix, and
Hypersalinity is an environmental parameter often associ- the petrographic description which is in terms of macerals.
ated with the biomarker gammacerane and the demethylated Recent studies have emphasized the three-dimensional aspect
hopane Ts. The co-occurrence of these biomarkers may suggest of coal structure. Coals normally can be described in terms of
a strong facies control in a given deposit. the macerals liptinite, vitrinite and inertinite. Chemically there
Diasteranes are usually indicative of silicate minerals and are differences with liptinites being predominantly aliphatic in
hence their absence can be interpreted as indicative of a carbon- nature, probably derived in part from a hydrocarbon natural
ate-rich lithology. In the latter environments the diasterane/ polymeric precursor as well as hydrocarbons associated with
sterane ratios will be low. the exterior surface of leaves. Vitrinites and inertinites are
Differentiation of source rocks and associated hydrocarbon predominantly aromatic in nature with phenols concentrated
pools as of marine or terrigenous origin is of interest to petro- in vitrinites. Vitrinite and inertinite macerals can be readily
leum geochemists and the absence or presence of the 'continen- separated through their density differences (vitrinites are less
tal fingerprint' has greatly aided this analysis. n-Alkanes in the dense than inertinites), but chemically the inertinites release
> C25 molecular weight range with a high odd/even concen- on pyrolysis increasingly aromatized components with increas-
tration ratio (i.e. carbon preference index (CPI) > 1) constitute ing density. Gondwanaland coals differ from northern hemi-
the continental fingerprint and being derived from higher land sphere coals in that the inertinites at the molecular level, whilst,
plants the occurrence is indicative of a terrestrially influenced being biased towards a higher aromatic content as with
deposit such as a paleoalluvial fan. n-Alkanes < C20 predomi- euroamerican inertinites, they often reflect a chemical reactivity
nate in a marine deposit. When the CPI for the n-alkanes is 1 closer to that of a vitrinite and thus have been referred to as
or the n-alkanes are strongly reduced in concentration in the 'reactive inertinites'.
> C20 region then it is reasonable to suspect biological degra-
dation of the oil or the kerogen source. R. Basil Johns
Bibliography
Oil shales and source rocks Irwin, H. and Meyer, T. ( 1990) Lacustrine organic facies. A biomarker
Oil shales can be regarded as source rocks sufficiently rich in study using multivariate statistical analysis. Org. Geochem., 16,
197-210.
organic (kerogen) content to yield a distillate on heat treatment Rafalska-Bloch, J. and Cunningham, R. (1986) Organic facies in
in sufficient quantity to be processed economically. Petroleum Recent sediments of carbonate platforms: Southwest Puerto Rico
source rocks on maturation and after migration of the released and Northern Belize. Org. Geochem., 10, 717-24.
bitumen can lead to substantial occurrences of organic material Tissot, B.P. and Welte, D.H. (1984) Petroleum Formation and
in oil pools/petroleum reservoirs. Their categorization as Occurrence, 2nd edn. Berlin: Springer-Verlag.
marine or terrestrially sourced will reflect the environment of Tuweni, A.O. and Tyson, R.V. (1994) Organic facies variations in the
Westbury Formation (Rhaetic, Bristol Channel, S.W. England).
deposition of the source rocks. Biologically degraded oils are Org. Geochem., 21, 1001-14.
chemically differentiated from others by compositional differ- Wiesner, M.G., Haake, B. and Wirth, H. (1990) Organic facies of
ences most marked in the loss of n-alkanes. A further form of surface sediments in the North Sea. Org. Geochem., 15, 419-32.
organic matter which is highly viscous or even semi-solid is
termed a bitumen. Although apparently a product of matura- Cross-references
tion of organic precursors the origin and formation of such Biomarker: coals
bitumen is unclear. Biomarker: higher plant
 OIL SHALES 441

Black shales and sapropels are transported by the Leeuwin Current from seeps or spillages
Coal: origin and diagenesis off northern Australia and Indonesia in South East Asia.
Coal: vitrinite reflectance and maturity assessment
Coal: types and characteristics
Satellite imagery shows that these tropical waters do occasion-
Geochemistry of sediments ally reach the South Australian coast, thus accounting for
Hydrocarbons tropical zooplankton being found off southern Australia.
Oil shales
Organic geochemistry Andrew T. Revill
Paleoenvironments
Peat
Petroleum Bibliography
Petroleum: types, occurrence and reserves
Rock-eva! pyrolysis Blumer, M. (1972) Submarine seeps: are they a major source of open
ocean pollution? Science, 176, 1257-8.
Connan, J. and Deschesne, 0. (1992) Archaeological bitumen:
Identification, origins and uses of an ancient near eastern material.
Mater. Res. Soc. Symp. Proc., 683-720.
OIL SEEPS AND COASTAL BITUMENS Currie, T.J., Alexander, R. and Kagi, R.I. (1992) Coastal bitumens
from western-Australia -long distance transport by ocean currents.
It is not often recognized that a significant amount of oil Org. Geochem., 18, 595-601.
Harvey, G.R., Requejo, A.G., McGillivary, P.A. and Tokar, J.M.
entering the sea comes from natural oil seepage on the seafloor. ( 1979) Observation of a subsurface oil-rich layer in the open ocean.
US estimates suggest 8% of all oil entering the global oceans Science, 205, 999-1001.
is natural (NRC, 1985). Wilson et al. (1974) estimated that McKirdy, D.M., Cox, R.E., Volkman, J.K. and Howell, V.J. (1986)
the amount of hydrocarbons from natural oil seeps entering Botryococcane in a new class of Australian non-marine crude oils.
the sea on a global basis each year was between 0.2 x 10 6 and Nature, 320, 57-9.
6.0 x 10 6 tons. Blumer (1972) argued that these figures were NRC (1985) Oil in the Sea, Inputs, Fates and Effects. Washington, DC:
National Academy Press.
too high. Reed and Kaplan (1977) estimated that oil seeps Reed, W.E. and Kaplan, I.R. (1977) The chemistry of marine petro-
contribute 10% of the petroleum entering the marine environ- leum seeps. J. Geochem. Expl., 7, 255-93.
ment. The estimate advanced by the US National Academy of Wilson, R.D., Monaghan, P.H., Osanik, A., Price, L.C. and Rogers,
Sciences of0.6 x 10 6 tons is a compromise between these vari- M.A. (1974) Natural marine oil seepage. Science, 184, 857-65.
ous figures. However, there is still considerable uncertainty
and some observations imply that the amounts may be even Cross-references
higher. For example, Harvey et al. (1979) observed a massive
Biomarker: aliphatic
layer of oil-rich water during research cruises in February and Biomarker: assessment of thermal maturity
March 1978 in the southwest North Atlantic and eastern Gas chromatography-mass spectrometry (GC-MS)
Caribbean. The layer of oil-rich water contained 3-12 mg oil/! Hydrocarbons
and extended to a depth of 200 m over a distance of 800 Laboratory simulations of oil and natural gas formation
nautical miles. Geochemical data and 14C activity confirmed Oil shales
Oil-oil and oil-source rock correlation
that it was weathered crude oil which was believed to have Organic geochemistry
originated from a seep on the Venezuelan shelf at a depth of Petroleum: surface geochemistry
200m. The authors conservatively estimated that more than
1 million tons of oil had been released, which is about twice
the US National Academy of Science's estimate for yearly
inputs of oil entering the sea. It needs to be pointed out that OIL SHALES
this oil layer was found well below the sea surface; it was
fortuitous that it was observed at all. Oil shales represent the large scale accumulation of freshwater
Surface seepage in the form of tar or asphalt has been known (e.g. Botryococcus) or marine (e.g. Tasmanites) algal and bacte-
for millennia, with early civilizations using the tar to caulk rial remains in depositional environments such as large lakes
boats or to burn it as fuel. It has been suggested, using data (such as the Green River shales in Utah), shallow seas or small
from organic geochemical techniques such as biomarker analy- lakes and lagoons. Organic geochemical studies of these shales
sis, that asphalt originating from Dead Sea bitumens formed have yielded a lot of information about early transformations
part of Middle East trade during 3900-2200 BC (Connan and of organic matter and individual compounds. Shales such as
Deschesne, 1992). the Green River deposit have been widely and extensively
A major study of tar balls (coastal bitumens) stranded on studied and have produced new information on the possible
South Australian beaches has been underway for many years role of algal biopolymers in kerogen formation. Some deposits
These bitumens have been analyzed by modern GC-MS tech- remain poorly understood, however. One such deposit is the
niques (McKirdy et al., 1986) and most were found to contain tasmanite which is composed entirely of the remains of
significant amounts of the unusual isoprenoid botryococcane Tasmanites and occurs in Tasmania, Alaska, the Sahara and
and bicadinanes which are commonly found in oils from Brazil. This particular shale has long been of interest to geo-
Indonesia. Currie et al. (1992) also showed that waxy botryo- chemists as it contains an unusually high abundance of com-
coccane-containing bitumens collected along the west and pounds referred to as chelanthanes or tricyclic hydrocarbons
south coasts of Western Australia contained remarkably sim- which are often present in low abundance in crude oils (Revill
ilar distributions of 4-methyl steranes, oleanane, bicadinanes et al., 1994). The co-occurrence of Tasmanites and these tricy-
and other biomarkers to those found in botryococcane-con- clic compounds has often led to a cause-effect relationship
taining bitumens found on South Australian and Tasmanian being attributed to their origin but the use of isotope techniques
beaches. These authors concluded that these waxy bitumens has thrown some doubt on this assumption.
442 OIL-OIL AND OIL-SOURCE ROCK CORRELATION

Oil shale deposits occur in many countries of the world from Cross-references
the very large deposits in the USA, Brazil and USSR to the
Biopolymers and macromolecules
small occurrences of shales such as the tasmanite of Tasmania, Black shales and sapropels
and range in age from the Cambrian to the Tertiary. The total Chromium
amount of oil stored within these shales is estimated to be 10 15 Clay minerals- ion exchange
barrels (Russell, 1990) compared with world proven crude oil Cobalt
reserves of 10 12 barrels (as at end of 1991; Australian Institute Copper
Geochemistry of sediments
of Petroleum, 1992). Hydrocarbons
Oil shales have been described in a variety of ways but a Laboratory simulations of oil and natural gas formation
general definition is an organic-matter rich sediment which has Nickel
never been buried to a depth sufficient to thermally mature the Oil seeps and coastal bitumens
organic matter. Oil shales do not contain oil in their natural Oil-oil and oil-source rock correlation
state but have the property of yielding oil when heated to a Occurrence of organic facies
Organic geochemistry
temperature of about 500°C (Rullkotter, 1987). Oil shales have Sulfur
high total organic carbon content, generally exceeding 20%, a Uranium
very high hydrogen to carbon ratio ( 1.5 or greater) and a low Vanadium
oxygen content and are therefore generally referred to as type I Weathering: chemical
kerogens. Thus by definition, the conditions under which oil
shales are formed must be those which lead to good preserva-
tion of the organic matter. The amount and composition of
organic matter preserved in sediments is controlled by a OIL-OIL AND OIL-SOURCE ROCK
number of factors, some of which are dependent on each other. CORRELATION
A high primary productivity (and therefore carbon production)
is not enough in itself to produce a high sediment accumulation Correlation of oils to each other and to their source rocks is
of organic matter. The rate of grazing by higher organisms an essential component in defining a basin's petroleum system.
and the degree of oxidation will affect the final accumulation. Oil-oil correlations seek to determine whether oils were gener-
Various estimates for organic matter preservation suggest ated and expelled from the same source facies. Oil-source rock
that much less than 1% of that produced in the water column correlations attempt to determine the specific source facies that
becomes preserved in the sediments. Thus the formation of oil gave rise to an oil. Correlation studies endeavor to establish
shales suggests depositional conditions different from those the original genetic relationship of geologic samples rather
normally present. A lack of oxygen will greatly enhance preser- than a mere matching of bulk properties or chemical composi-
vation as this inhibits both direct oxidation and biological tions. Because the original composition of an oil can be
re-working of the organic matter. Indeed, it may be that the changed by secondary processes (e.g. thermal maturation,
presence of a large amount of organic matter itself induces migration, and reservoir alteration such as biodegradation,
bottom water anoxia due to the high oxygen consumption of thermochemical sulfate reduction and phase separation), petro-
bacteria re-working the material. More recently, the role that leums with very different bulk, molecular, and isotopic com-
sulfurization of organic matter might play in preservation has positions still may be genetically related. An example of an
been investigated. It is thought that the presence of reactive oil-oil correlation is shown in Figure 01. Although secondary
centers in the organic compounds, such as double bonds and processes have altered substantially the composition of the
functional groups (e.g. oxygen atoms), may provide sites for oils, the genetic association of the oils to each other is revealed
the incorporation of sulfur into the molecule which could lead by biomarker and isotopic analyses. Even oils with very similar
to cross-linking between molecules. This has the effect of trans- compositions may not originate from the same source. False
forming relatively small molecules into a macromolecular correlations may arise if testing is based on a limited number
structure which is then much more resistant to degradation. of measurements, docs not account properly for secondary
The exact conditions under which this transformation might processes, or fails to integrate geochemical data with a sound
occur are not yet fully understood, but the presence of sulfur understanding of basinal history and structural setting.
and highly reducing conditions (anoxia) would seem to be Inherent limitations exist in all oil-oil and oil-source rock
important. correlation studies. Oil-source rock correlations generally pre-
suppose that kerogen and the bitumen extracted from a source
Andrew T. Revill rock are representative of specific molecular and isotopic char-
acteristics of its expelled oil. This assumption is made even
though the chemical compositions of kerogen, bitumen, and
Bibliography oil differ substantially, and the organic facies present in a single
Australian Institute of Petroleum (1992) Oil Industry Statistics 1992. rock sample may not be representative of the overall source
Melbourne: Australian Institute of Petroleum. formation. As produced oils are accumulations of expelled and
Revill, A.T., Volkman, J.K., O'Leary, T. eta/. (1994) Hydrocarbon migrated petroleums, oils that correlate well to each other may,
biomarkers, thermal maturity, and depositional setting oftasmanite in fact, have multiple sources. In many basins, direct oil-source
oil shales from Tasmania, Australia. Geochim. Cosmochim. Acta, rock correlations are not possible as the source facies either
58, 3803-22.
Rullkotter, J. (1987) Geochemistry, organic. Encyclopedia Phys. Sci.
are unavailable or have inappropriate levels of thermal matu-
Techno/., 6, 53-77. rity for comparison. In such cases, the nature of the source is
Russell, P.L. (1990) Oil Shales of the World, Their Origin, Occurrence often inferred from the chemistry of the oils and geologic
and Exploitation. Oxford: Pergamon Press, 753 pp. reconstructions (Moldowan et a!., 1985).
 OIL-OIL AND OIL-SOURCE ROCK CORRELATION 443

Whole Oil Gas Chromatograms m/z 191 GC/MS lon Chromatograms m/z 217 GC/MS lon Chromatograms

Figure 01 An example of an oil-oil correlation. The oils are from three different fields produced from three different reservoir sands from
offshore Azerbaijan. Secondary processes such as biodegradation (Duvany Deniz) and phase separation (Bakhar) have significantly altered
the composition of the oils. Although the absolute abundances of saturate biomarkers vary, their relative distributions are nearly identical.
This correlation indicates that the oils are genetically related and are derived from the same source(s). Select normal alkanes are indicated
by carbon number, Pr = pristane, Ph= phytane. Hopanes are indicated by H (carbon number). Steranes are indicated by S (carbon
number).

Analytical techniques in correlation have advanced consider- have been applied (Peters and Moldowan, 1993). Biomarker
ably since the pioneering study of Hunt et al. (1954 ), which analysis remains the principal method for most modern corre-
relied mostly on elemental analysis and infrared spectroscopy. lations though there is considerable debate and continuing
Geochemical studies of the Williston Basin by Williams (1974) research on the influences of non-source-related, secondary
and Dow (1974) were the first to integrate molecular and processes.
isotopic data into a geologic framework. Analyses of bio- Stable carbon isotope measurements also are routinely
marker and other trace hydrocarbons by gas chromatography/ applied for correlation of a wide variety of samples (e.g. whole
mass spectrometry revolutionized correlation studies by pro- oils, bitumens, and kerogens, distillation or C 15 + chemical
viding a degree of specificity lacking in measurements of bulk group-type fractions, kerogen and asphaltene pyrolysates, and
properties and major molecular compositions (Seifert, 1977). gases). Isotopic analysis of individual compounds promises
Distributions of steranes and triterpanes are now routinely improved specificity for correlation (Bjoroy et al., 1993). While
used for correlation. Numerous other classes of biomarker numerous studies have been made using only biomarker or
compounds including various acyclic isoprenoid hydrocarbons, isotopic data, the most successful correlations integrate both
bi-, tri- and tetra-cyclic terpanes, aromatized steranes and techniques. Other analytical procedures used in correlation
hopanes, and oleanane- and bicadinane-related compounds, studies include, but are not limited to, the analyses of trace
444 OKLO NATURAL NUCLEAR REACTOR

analysis of metals and rare earth elements; stable isotope analy- when a routine analysis of a uranium ore standard at the
ses of hydrogen, sulfur, and nitrogen; radiometric isotopes of nuclear-fuel processing plant in Pierrelatte, France revealed a
neodymium; light hydrocarbons of oils, bitumens, and pyroly- slight depletion in 235 U (Cowan, 1976). Uranium has three
sates; aromatic hydrocarbon and non-hydrocarbon NSO com- naturally occurring isotopes in the following proportions:
pounds; fluorescence spectroscopy, and geoporphyrins. 99.27% for 238 U, 0.72% for 235 U, and 0.0056% for 234 U. The
Application of many of these techniques can be found in measurement taken at Pierrelatte indicated that 235 U made up
Magoon and Claypool (1985). Curiale (1993, 1994) provides 0.7171% of the standard, slightly lower than the expected value
reviews of concepts, limitations, and key case studies involving of0.72%. Although the depletion was small, the observed value
correlations within petroleum systems. was considered unusually low and an investigation was started
to determine its origin. As scientists began to trace the cause
Clifford C. Walters of the depletion, they initially assumed that the uranium stan-
Bibliography dard had somehow been contaminated with 235 U-depleted ura-
nium tailings, a waste product left over from the process used
Bjor0y, M., Hall, P.B. and Moe, R.P. (1993) Stable carbon isotope to produce enriched 235 U uranium for use as fuel in man-made
variations in n-alkanes in Central Graben oils. Org. Geochem., 22,
355-81. nuclear reactors (Cowan, 1976). When this turned out not to
Curiale, J.A. (1993) Oil to source rock correlation: concepts and case be the case, the ensuing investigation led to the mine which
studies, in Organic Geochemistry (eds M.H. Engel and S.A. sourced the uranium: the Oklo uranium mine in the Republic
Macko). New York: Plenum Press, pp. 473-90. of Gabon, a small country on the coast of Equatorial West
Curiale, J.A. ( 1994) Correlation of oils and source rocks- a conceptual Africa. Analyses on samples obtained directly from the Oklo
and historical perspective. Am. Assoc. Petrol. Geol. Mem., 60,
251-60.
ore body revealed even greater depletions in 235 U. These
Hunt, J.M., Stewart, F. and Dickey, P.A. (1954) Origin of hydro- strongly depleted samples were associated with specific regions
carbons in Uinta Basin, Utah. Am. Assoc. Petrol. Geol. Bull., 38, of the ore deposit where the ore occurred in concentrations as
1671-98. high as 20-60% uranium by weight. Because there was no
Dow, W.G. ( 1974) Application of oil-correlation and source-rock data known natural process that could fractionate 235 U from urani-
to exploration in the Williston Basin. Am. Assoc. Petrol. Geol. Bull., um's more abundant isotope 238 U, scientists were initially
58, 1253-62.
Magoon, L.B. and Claypool, G.E. (1985) Alaska north slope oil/rock baffled by the 235 U depletions observed in some of the Oklo
correlation study. Studies Geol., 20, 682 pp. ore samples. As the investigation continued, however, a star-
Moldowan, J.M., Seifert, W.K. and Gallegos, E.J. (1985) The relation- tling discovery was made. The relative abundances of a number
ship between petroleum composition and the environment of of isotopes unique to nuclear fission reactions were observed
deposition of petroleum source rocks. Am. Assoc. Petrol. Geol. in the 235 U-depleted ores. The explanation was now clear. The
Bull., 69, 1255-68.
Peters, K.E. and Moldowan, J.M. (1993) The Biomarker Guide:
missing 235 U had, in fact, been consumed by nuclear fission
Interpreting lvfolecular Fossils in Petroleum and Ancient Sediments. reactions in the high grade uranium ore bodies. The inescap-
Englewood Cliffs, NJ: Prentice Hall, 363 pp. able conclusion was that nature created the first nuclear reac-
Seifert, W.K. ( 1977) Source rock-oil correlations by C 27 -C 30 bio- tors to operate on Earth nearly two billion years before Enrico
marker hydrocarbon distributions. in Advances in Organic Fermi's nuclear reactor was constructed at Staggs Field in
Geochemistry (eds R. Campos and J. Goni). Madrid: Chicago (Cowan, 1976).
ENADIMASA, pp. 21-44.
Williams, J.A. (1974) Characterization of oil types in Williston Basin.
Am. Assoc. Petrol. Geol. Bull., 58, 1243-52.
Conditions for the formation of natural reactors
Cross-references What fortuitous circumstances led to the formation of natural
Biomarker: aliphatic fission reactors? Natural nuclear reactors, like their man-made
Biomarker: aromatic counterparts, require a number of special conditions in order
Biomarker: higher plant to operate. Fission takes place when a nuclide such as 235 U,
Biomarker: lipid 239 Pu, or 241 Pu absorbs a neutron causing its nucleus to split.
Carbon isotopes This produces two or more fragment atoms usually of unequal
Compound-specific isotope GC-MS
Gas chromatography-mass spectrometry (GC-MS)
mass (fission products) y-rays, and several neutrons. If other
Geochemical exploration fissile nuclei are nearby they can absorb these neutrons and
Geochemistry: organic split, releasing more neutrons. The reactions can be sustained
Natural gas if enough nuclei are present to absorb the neutrons produced
Oil seeps and coastal bitumens by previous fissions. Nuclear reactors require significant
Organics: source and depositional environments amounts of fissionable material in order to start and maintain
Organics: thermal alteration in sediments
Petroleum fission reactions. At Oklo, the natural reactor fuel was chiefly
Petroleum: primary migration in the form of 235 U. Two billion years ago the relative abun-
Petroleum in reservoir biodegradation dance of 235 U, compared with 238 U, was five times greater than
Petroleum occurrence, types and reserves today because 235 U decays faster than 238 U. In addition, the
Stable isotopes Oklo reactors contain 20-60% uranium, mainly in the form of
uraninite, U0 2 (Gauthier-Lafaye et al., 1989). The high rela-
OKLO NATURAL NUCLEAR REACTOR tive abundance of 235 U together with the high concentrations
of uranium ore minerals met the requirement for sustainable
fission reactions in the natural reactors. A second requirement
Discovery of the natural reactor for nuclear reactor operation is the presence of a neutron
Although the possibility of natural nuclear reactors was consid- moderator. Moderators slow the neutrons emitted from the
ered in the 1950s, this prediction was not confirmed untill972, fission of 235 U down so they are more readily absorbed by
 OKLO NATURAL NUCLEAR REACTOR 445

other 235 U nuclei, sustaining fission reactions. At Oklo, the Nagy and Rigali, 1993). The solid graphitic bitumen appa-
neutron moderator was in the form of water present in the rently originated as a liquid petroleum (Gauthier-Lafaye and
porous and fractured rocks which host the reactors. A third Weber, 1989) through the reaction of hydrothermal solutions
requirement is the absence of neutron poisons, elements such with older organic solids, a process known as hydrous pyroly-
as B and Li which readily absorb neutrons inhibiting or pre- sis. The petroleum migrated into the reactors enclosing urani-
venting the fission of 235 U. These elements were not present in nite grains which had participated in the fission reactions and
sufficient quantities in the natural reactors to inhibit reactor its incorporated fission products. Heat and radiation rapidly
operation. turned the viscous petroleum into the graphitic solid. Natural
rectors 1-6 at Oklo contain only trace amounts of this carbona-
Conditions during natural reactor operation ceous substance. Here the uraninite is associated mainly with
clay minerals. The other reactors are organic matter-rich with
The natural reactors reached criticality 1.97 billion years ago parts of these reactors containing as much as 66% organic
and operated for 10 5-10 6 years (Holliger, 1992). During this carbon by weight. The uraninite occurs with both carbona-
time, 239 Pu was produced from the abundant isotope of ura- ceous matter and clay minerals in heterogeneous distributions
nium, 239 U, by neutron capture and two subsequent f3 decay within these reactors. Some parts of the organic-rich reactors
reactions: are uraninite in virtually pure clays and other parts are mainly
23su + n __, 239u uraninite enclosed in the solid, graphitic organic substance.
239u __, 239Np + f3-
Scientific relevance of the Oklo natural fission reactors
239N p--> 239pu + f3-
The Oklo natural reactors are complex geologic and geochemi-
239 Pu is also a fissionable material and small amounts of the cal occurrences and continue to be objects of scientific inquiry
239 Pu produced participated in fission reactions at Oklo. more than 20 years after their discovery. While they remain a
However, the majority of the 239 Pu created by neutron capture natural laboratory in which to study the conditions of reactor
did not fission. It decayed, becoming 235 U through the formation and operation, the Oklo reactors also present a
following reaction: unique opportunity for scientists to study the long-term geo-
239pu __, 23su + 4He chemical behavior of uranium together with fissiogenic isotopes
in geologic environments. Numerous fission products, includ-
In this manner, the Oklo natural nuclear reactors produced ing isotopes of Ag, Mo, Sn, Nb, Zr, Bi, Pd, Ru, Te, La, Nd,
additional 235 U to sustain fission reactions. In fact, as much Sm, Ce and Gd, have been found within the reactors and in
as 45% of the total amount of 235 U consumed in fission reac- the rocks in their vicinity (Curtis et al., 1989; Holliger, 1992).
tions was produced by the decay of 239 Pu (Holliger and These fission products have been retained in the natural reac-
Devillers, 1981). Taking into account the additional 235 U pro- tors and adjacent rocks to varying degrees. Rare earth elements
duced by the decay of 239 Pu, it is estimated that a total of (Ce, Sm, La, Pr, Nd, Gd) are retained to a large degree in the
6 tons of 235 U were consumed by fission reactions in the natural uraninite of the reactor zones mainly because they are compati-
reactors (Naudet, 1991). During natural reactor operation at ble in the structure of the uraninite. Other fission products not
Oklo, temperatures in the reactors ranged between 160° and compatible with uraninite such as Tc, Ru, Rh, Pd and Bi are
more than 360°C (Holliger and Devillers, 1981). The natural still retained in metallic inclusions found in close proximity to
reactors shut down when there was no longer enough fission- the uraninite. Still others such as Ba, Sr, and Cs have been
able 235 U to sustain nuclear reactions (Naudet, 1991). lost in large quantities from the reactors and their adjacent
host rocks. However, evidence for the retention of fissiogenic
Sr, Ba, and Cs isotopes in one of the reactors, carbonaceous
Chemical, physical, and geologic properties of the matter-rich natural reactor 10, has been reported recently
natural reactors (Hidaka et al., 1994 ). Isotopes of these three elements are of
To date there are at least 17 2-Ga-old natural nuclear rectors particular interest because they are produced in high yields
in the Oklo, and the associated Okelobondo and Bangombe during nuclear fission and they are also soluble and readily
uranium ore deposits in the Republic of Gabon. The reactors transported by aqueous solutions in geologic environments.
occur as small bodies in the uranium ores. They are 10-20 m Because a number of fission and decay products were
in length and width but their thickness is generally less than retained in the natural reactors, the reactors have modern
1 m. The natural reactors are located in hydrothermally altered anthropogenic relevance. This relevance can be stated as fol-
and fractured clastic sedimentary rocks. They are found in the lows. Since the components and the geochemical and
7-10 m thick Cl layer of the basal FA Formation of the hydrologic environments in and around the natural reactors
2.1-Ga-old Paleoproterozoic Francevillian Series. The Cllayer were able to contain a number of radionuclides, the Oklo
consists of beach sands and tidal deposits of fine-grained sand- natural reactors may be useful analogs for modern nuclear
stones, mudstones and shales. In addition to the high uranium waste containment considerations and strategies (Nagy and
content, occurring mainly as the mineral uraninite (U0 2), Rigali, 1993 ). Recent studies have revealed that a number of
authigenic clay minerals, small quantities of galena, other sul- physical and chemical factors have influenced the degree of
fides and hematite are also present in the natural reactors. retention of fission products in the natural fission reactors.
Many of them also contain variable amounts of solid, graphitic Retention of fission products is controlled in part by the min-
organic matter. This carbonaceous matter, also called solid eral assemblages in the natural reactors and their response to
graphitic bitumen, consists mainly of a highly condensed poly- microscale variations in pressure and temperature together
cyclic aromatic hydrocarbon matrix thoroughly intermixed with the composition of the uraninite fuel (Curtis et al., 1989).
with poorly crystalline, fine-grained graphite (Nagy eta!., 1993; Other factors, such as chemical alteration, exsolution, and
446 ONUMA DIAGRAMS

diffusion, were important mechanisms affecting the loss of Pb Nagy, B., Gauthier-Lafaye, F., Holliger, P., Mossman, D.J.,
and the fission products Mo, Ru, Rh, and Pd from uraninites Leventhal, J.S. and Rigali, M.J. ( 1993) Role of organic matter in
the Proterozoic Oklo natural fission reactors, Gabon, Africa.
in the Oklo natural reactors (Janeczek and Ewig, 1995). Geology, 21, 655-8.
However, many of these elements were precipitated in the Naudet, R. (1991) Oklo: Des Reacteurs Nucleaires. Paris:
immediate vicinity of the altered uraninite in the form of galena Commissariat a l'Energie Atomique, 695 pp.
(PbS) and metallic inclusions. This was due in part to the
presence of sulfur-rich carbonaceous matter which helped pre-
Cross-references
vent the long distance migration of Pb and these fission pro-
ducts. In fact, where it is present in the natural nuclear reactors, Elements: actinide series
Elements: lanthanide series, rare earths
solid graphitic carbonaceous matter acts as a barrier to the Elements: large ion lithophile
migration of U, Pb, and a number of fission products from Elements: radioactive
these reactors (Nagy et a!., 1991; Nagy and Rigali, 1993). Geochronology and radioisotopes
Uraninite and its entrapped fission products were enclosed by Geologic time scale
carbonaceous matter. This hydrophobic and largely impervi- Hydrothermal alteration
ous substance acted as an effective physical barrier to the loss Hydrothermal solutions
Lead
of a number of radionuclides. These studies all indicate that Lead: stable isotopes
Oklo natural reactors have value as analogs for the storage of N ucleosynthesis
man-made radioactive waste. However, further careful studies Oil seeps and coastal bitumens
are required with the objective of determining their practical Petroleum: hydrothermal
utility to current man-made nuclear waste containment Plutonium
strategies. Radioactivity
Strontium
The Oklo, Okelobondo, and Bangombe uranium ore depos- Uranium
its in Gabon are unique as they host the world's only known Uranium-lead thorium-lead decay system
natural fission reactors. A number of fortuitous circumstances
including high concentrations of fissionable 235 U, the presence
of significant quantities of water and the absence of neutron
poisons contributed to their formation and operation nearly ONUMA DIAGRAMS
2 Ga ago. The natural reactors represent complex geochemical
and hydrologic milieus. It is important to consider the hetero- Onuma diagrams are graphs of partition coefficient vs ionic
geneous textural patterns of their components as well as their radius for elements of known valence states that can occur in
interaction with the surrounding geochemical environments a specific mineral. They graphically illustrate the controlling
during 2 Ga of geologic time to properly evaluate their origin influence of ionic radii and valence states on elemental substitu-
as well as their potential utility to modern nuclear waste con- tion into specific structural sites during crystallization of a
tainment strategies. mineral from a melt or fluid. This control was first recognized
by Goldschmidt (1937) and further developed and quantified
Mark J. Rigali and Bartholomew Nagy by Onuma et a!. (1968), Jensen (1973) and Matsui et al.
(1977). The diagrams can be used to predict the likelihood of
Bibliography elemental substitutions in a particular mineral.
Cowan, G.A. (1976) A natural fission reactor. Sci. Am., 235, 36-47. Onuma diagrams are generated by plotting partition coeffi-
Curtis, D., Benjamin, T., Gancarz, A., Loss, R., Rosman, K. and cients vs ionic radii for all trace elements possible for a given
DeLaeter, J. (1989) Fission product retention in the Oklo natural mineral. The family of sub-parallel curves that are formed
fission reactors. Appl. Geochem., 4, 49-62. represent the univalent, divalent etc. sites in the mineral, but
Gauthier-Lafaye, F. and Weber, F. (1989) The Francevillian (Lower typically only one site is portrayed on each diagram. Figure
Proterozoic) uranium ore deposits of Gabon. Econ. Geol., 84,
2267-85. 02 is an Onuma diagram for the divalent Fe 2 + site in biotite.
Gauthier-Lafaye, F., Weber, F. and Ohmoto, H. (1989) Natural fission Although a wide variety of substitutions are possible in this
reactors of Oklo. Econ. Geol., 84, 2286-95. site (the elements on the + 1, + 2, + 3 and + 4 valence curves)
Hidaka, H., Sugiyama, T., Ebihara, M. and Holliger, P. (1994) the most common and preferred substitution for Fe 2 + is Mg2+
Isotopic evidence for the retention of 90 Sr from excess Zr in the (the highest peak on the highest curve). Small trivalent ions
Oklo natural fission reactors: Implication for geochemical behavior
of fissiogenic Rb, Sr, Cs and Ba. Earth Planet. Sci. Lett., 122,
(e.g. Sc) are slightly preferred to large univalent ions (e.g. Ba).
173-82. Na is the least acceptable ion in this site.
Holliger, P. (1992) Geochemical and isotopic characterization of the The regularity of the curves representing each valence state
reaction zones (uranium, transuranium, lead and fission products). suggests that trace elements occupy lattice sites in minerals
Proc. Joint Commission European Communities-Commissariat a and do not occur in separate grains or absorbed on crystals
l'Energie Atomique, 2nd Meeting, Brussels, Belgium, EUR, (Onuma et al., 1968). The curves also illustrate that, for trace
pp. 27-38.
Holliger, P. and Devillers, C. (1981) Contribution a !'etude de Ia tem- elements of the same charge, the difference between the ionic
perature dans les reacteurs fossiles d'Oklo par la mesure du rapport radii of the trace element and the major element is the most
isotopique du lutetium. Earth Planet. Sci. Lett., 52, 76-84. important factor in determining the partition coefficient. The
Janeczek, J. and Ewing, R. (1995) Mechanisms of lead release in the effect of ionic charge on the partition coefficient also appears
natural fission reactors in Gabon. Geochim. Cosrnochim. Acta, 59, to be independent of the effect of the radius (Jensen, 1973).
1917-31. On Onuma diagrams, the shape of the curve correlates with
Nagy, B. and Rigali, M.J. (1993) Newly discovered, organic matter-
rich natural fission reactors at Oklo and Bangombe: are they useful the crystal structure of the mineral. The curve apex is the
analogs for long-term anthropogenic nuclear waste containment? optimum ionic radii for a cation in a given site. The breadth
Waste Management '93 Conference Proceedings, pp. 897-900. of the peak is an expression of the ability of the lattice to
 ORE DEPOSITS 447

100 Cross-references
Elements: trace
Partition coefficients
Trace element partitioning models
+2

10 ORE DEPOSITS
+-
cQ)
"(3 +3 Definition and classification
lE
Q)
There is no unequivocal definition of an ore deposit.
0 Traditional definitions embrace the concept of a natural min-
u eral accumulation that may be profitably exploited. The signi-
c ficance of profitability is now often questioned because, for
0 example, it is prone to fluctuate with market prices or is
:E
t obviated by the availability of subsidy; and whilst all natural
0 inorganic crystalline solids (including useful raw materials such
0....
+1 as diamond) are minerals, more usually only those that are
metalliferous are considered to constitute ores. Against this
0.11 background an ore deposit can be defined as a natural aggre-
gate of mineral(s) from which metal(s) may be extracted to
meet the demands of society.
Ore deposits are of geochemical interest from two perspec-
tives. First, they are comparatively rare and comparatively
large natural accumulations of metals with enrichment factors
0.01 relative to average metal abundances in igneous rocks of the
upper continental crust ranging from less than 4 to more than
0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 4000 (Figure 03 ). Second, their genesis is usually incompletely
0

Shannon - Prewitt (A) understood and explained by means of partially speculative

models constrained by geochemical measurements and other
Figure 02 Onuma diagram for biotite from the Sunshine Canyon data: sulfur, oxygen and hydrogen isotope data suggest the
dacite (modified from jensen, 1973). provenance of ore-bearing fluids; radioactive decay of ura-
nium, rubidium and potassium provides geochronologic data;
and fluid inclusions indicate the salinity of mineralizing solu-
accept non-ideal cations. Major ,elements rarely fit at the crest, tions and yield geothermometric and geobarometric data.
but are seldom more than 0.1 A from it (Jensen, 1973). The Ore deposits have been classified in many different ways
0 or Si framework is thought to control the mineral structure (none of them entirely satisfactory because on ever closer
and chemistry because all curves peak together regardless of inspection almost all ore deposits are found to differ in detail).
the cation and charge (see Figure 02). To represent substitu- It is convenient to distinguish first between deposits formed
tion into the Si or AI sites, other diagrams would be required. by mechanical concentration and those formed by geochemical
Onuma diagrams are useful because they clarify the crystal processes. The former warrant no further attention here. The
structure preferences of a particular element. From this, it is latter can be further divided into magmatic, hydrothermal and
then possible to determine which site an element will occupy
and which elements a mineral will accept. The partition coeffi- 10000
cient for an element of known valence can then be estimated
or if the coefficient is known, the valencies determined. From
this, the proportion of the two valencies can be determined
(e.g. Eu 2 +, Eu3+) (Jensen, 1973). 1000

fl
$
Jean M. Richardson
·g"0 100
Bibliography "
Q)

Goldschmidt, V.M. ( 1937) The principle of distribution of chemical ""0

0
elements in minerals and rocks. J. Chern. Soc., 655-72. 10
Jensen, B.B. ( 1973) Patterns of trace element partitioning. Geochim.
Cosmochim. Acta, 37, 2227-42.
Matsui, Y., Onuma, N., Nagasawa, H., Higuchi, H. and Banno, S.
(1977) Crystal structure control in trace element partitioning
among crystals and magmas. Bull. Soc. Fr. Mineral. Crystal/ogr.,
100, 315-24.
Onuma, N., Higuchi, H., Wakita, H. and Nagasawa, H. (1968) Trace Figure 03 Concentration factors of metals in ore deposits
element partitioning between two pyroxenes and the host lava. compared to average abundances in igneous rocks of the upper
Earth Planet. Sci. Lett., 5, 47-51. continental crust.
448 ORE DEPOSITS

allochthonous deposits (Table 01). The broad characteristics basal massive sulfides that grade upward into disseminated
of each of these classes and the principal types of deposits sulfides in a net of silicates. Below about 11 oooc first iron-
within each class are considered briefly (and therefore with nickel sulfides, such as pyrrhotite, (Fe,Ni h S8 , and pentlandite,
considerable generalization) below. (Fe,Ni) 9 S8 , and then copper sulfides, such as chalcopyrite,
CuFeS 2 , begin to crystallize. Ore deposits grade from about
Magmatic ore deposits 0.8% to over 3% Ni and range in size from a few million tonnes
to several tens of millions of tonnes. Synvolcanic deposits in
Ore deposits of this type form during the cooling and crystalli-
Archaean greenstone belts in the Canadian Shield, Western
zation of a silicate melt within the crust (Whitney and Naldrett,
Australia and Zimbabwe account for about 23% of magmatic
1989). Many, but not all, ore deposits in this class are associ-
nickel ores. These deposits formed from highly magnesian
ated with large magmatic intrusions with a mafic or ultramafic
ultramafic (komatiitic) magmas during or immediately
composition. Ore deposits accumulate within the magma
following their intrusion as sills and dykes, as at Agnew,
chamber as a result of segregation of crystallizing minerals or
Western Australia, or their extrusion as submarine lavas, as at
as a result of liquation, a process in which a silicate melt
Kambalda, Western Australia. Deposits associated with large
containing more than a few percent sulfur separates into
non-layered gabbroic intrusives in intracratonic regions, in
co-existing but immiscible silicate and sulfide phases (Figure
particular those of Sudbury, Ontario, and Noril'sk-Talnakh,
04).
Siberia, account for most of the remaining magmatic nickel
Nickel sulfide ore deposits originate by liquation. Iron,
ores. At Sudbury, where magma emplacement is thought to
nickel and copper partition into the denser sulfide liquid which,
have been triggered by the impact of a meteorite, or astrobleme,
while the silicates crystallize, sinks into embayments to form
almost one billion tonnes of ore grading about 1% Ni is distrib-
uted amongst some 40 deposits. At Noril'sk-Talnakh the parent
Table 01 Types of ore deposits formed by geochemical processes magma, emplaced during the Triassic, was initially sulfur-poor
due to the progressive depletion of sulfur in the upper mantle,
Parent fluid Ore deposits
and the nickel deposits owe their genesis to sulfurization of
Magmatic silicate Stratiform and podiform chromite deposits the magma by assimilation of evaporites.
Nickel sulfide deposits Titanium and vanadium ore deposits also formed through
Platinum group element deposits processes of liquation and immiscibility, in this case after a
Titanium and vanadium deposits length; period of fractional crystallization leading to a crystal-
Hydothermal solution Porphyry copper-molybdenum deposits line phase of norite or anorthosite and the concentration of
Greisen tin-tungsten deposits iron, titanium and sometimes vanadium in the residual magma.
Shear-hosted and epithermal gold deposits The dense Fe-Ti-(V)-enriched liquid sank to form a stagnant
Volcanogenic massive sulfide deposits
layer in which in situ crystallization of ore was triggered by a
Stratiform copper-cobalt deposits
Mississippi Valley Type lead-zinc deposits complex interplay of factors including an increase in /0 2 .
Sedimentary-exhalative lead-zinc deposits Deposits of this type include those of Tellnes in southern
Unconformity-type uranium deposits Norway, where there is 0.3 million tonnes of ilmenite ore
Meteoric/seawater Bauxite deposits grading 18% Ti0 2 , and the Upper Zone of the Bushveld
Lateritic nickel deposits Complex, which contains over one billion tonnes oftitaniferous
Sandstone-type uranium deposits magnetite ore grading up to 2% V2 0 5 .
Banded iron formation Chromite ore deposits usually comprise almost entirely chro-
Manganese deposits mite, FeO · Cr 2 0 3 , and ore grades vary between 30% and 60%
Cr 2 0 3 , depending upon the Fe content of the chromite. Three
sites of chromite ore genesis are: (i) in mafic magmas emplaced
into the intracratonic upper crust; (ii) at the base of ophiolitic
magmas emplaced into the lower oceanic crust at back-arc
spreading centers; and (iii) in the partially melted upper mantle
feeding ophiolitic magmas. In the magmas themselves, olivine
and chromite co-precipitate along their cotectic during the
early stages of cooling. This process, however, results in a high
proportion of olivine with only a few percent chromite. Pure
chromite segregates as a result of either a shift in the phase
boundary towards the olivine-silica join (due to an increase in
f0 2 or a change in total pressure) or a shift in magma composi-
tion away from the olivine-silica join and across the cotectic
(due to sialic contamination or magma mixing). In the partially
melted upper mantle a denser olivine fraction separates from
rising basaltic magma and segregates into lenses of chromite
enveloped by dunite in small chambers within the depleted
700+--r-,--r-~-r-;--r-;--r~
upper mantle. Periodic repetition of any of these chromite ore-
0 20 40 60 80 100
forming processes produces several discrete chromite accumu-
lations within a single (magma) chamber. The mafic-ultramafic
Sulfide% magmas of intracratonic regions crystallize as layered com-
Figure 04 Separation of immiscible sulfide liquid from a cooling plexes in which the chromite cumulates form extensive but thin
silicate liquid (redrawn with permission from Hawley, 1962). layers of stratiform chromite ore. The Bushveld Complex of
 ORE DEPOSITS 449

South Africa includes over 20 chromite layers; its LG6 layer .................,
Sea
has a strike length of 70 km, a thickness of 90-130 em and
contains over 750 million tonnes of chromite ore. The Great
0 //1-.j---1
Dyke of Zimbabwe includes over 10 chromite layers, each less 1 .. ..: I
than I m in thickness but persistent over much of its 500 km :-"j" Metamorphic 1
strike length. These stratiform chromite ores usually have a -50 L.I··J--~1 water 1
relatively high Fe content, with Fe: Cr ratios between 1.5 and I,.W..---
I :I I
2.5. The chromite cumulates of the ophiolite complexes, on Formation .1-:::: ~ , .
the other hand, tend to have been fragmented by tectonism. ...·· "'Magmatic
The resulting bodies of podiform chromite ore rarely exceed -100 ..·············· water
one million tonnes, as in Turkey and the Philippines, although
......... -
the Molodezhnoe deposit in the Ural mountains of Russia _________ ............... -
contains 90 million tonnes of chromite ore. Despite their gen-
erally smaller size, podiform chromite deposits are more widely -150~1----~------~-----,------.-----~

exploited due to their lower Fe content; their Cr: Fe ratios are -20 -10 +20 +30
typically 3.0-4.0.
Platinum-group element ore deposits form through the col-
lection of PGE by sulfides following the introduction of fresh Figure 05 Isotopic compositions of waters of different origins
magma into magma chambers already crystallizing layered (redrawn with permission from Sheppard, 1986).
mafic-ultramafic intrusions. The fresh magma, less dense than
the resident magma immediately overlying the basal cumulates near-ambient temperatures). Metals are soluble in these solu-
but denser than resident magma higher in the chamber, spreads tions mainly as complexes of Cl- and HS-, and to a lesser
horizontally at a level appropriate to its density, mixing turbu- extent as complexes of H 2 S, NH 3 and possibly other ligands;
lently with and entraining cooler resident magma. Saturation the composition of a typical hydrothermal solution approxi-
or near-saturation of both magmas in sulfur leads to the sepa- mates to I m Cl-, 0.1 m sulfur species and 0.0 I m ammonia
ration of immiscible sulfide droplets suspended in the fresh species (Barnes, 1979). The Cl- ligand is available over a wide
magma plume. As heat loss decreases, convection becomes temperature and pH range, forming HCI only at high temper-
laminar, the sulfide sinks (or downspouts) through the under- ature coupled with extremely low pH. Among the sulfur
lying magma and finally spreads out across the top of the species, So~- and HS04 dominate in very low pH solutions,
cumulate pile to crystallize as extensive horizontal layers. The but do not form stable complexes with metals; in higher pH
sulfide acquires PGE from ascending, highly fractionated mag- reducing solutions, sulfate ions give way to H 2 S and HS-, and
matic liquids filter-pressed out of the cumulate pile into the the latter in particular forms strong metal complexes. Metal
overlying sulfide layer, or from PGE-bearing hydrothermal complexing by NH 3 is also possible and could be important
fluids permeating the sulfide layer, or from the magmas during above 220°C, but ammonia concentrations tend to be relatively
their turbulent mixing in the magma plume. The resulting PGE low. Thus in high temperature hydrothermal solutions, Cl- is
minerals include sperrylite, PtAs 2 , braggite, (Pt,Pd,Ni)S, stibi- the principal metal-complexing ligand, and for most metals
opalladinite, Pd 3 Sb and laurite, RuS 2 . The PGE attain ore HS- becomes progressively more important or dominant
grades in only two layered complexes, but the resources are below about 300°C (Figure 06). Metal concentrations in ore-
large due to the great lateral extent of the mineralized layers: forming hydrothermal solutions are of the order of less than
in the Bushveld Complex, South Africa, the 0.3-0.6 m thick 1-10 ppb for gold and 10 to more than 1000 ppm for base
Merensky Reef, which can be traced over 110 km, and the metals (Barnes, 1979). Ore deposits precipitate from these
0.6-2.5 m thick UG2 Layer contain some six billion tonnes of solutions as a result of fluid-rock reactions, redox changes,
ore with an average grade of 8.3 g/t LPGE; in the Stillwater adiabatic expansion, cooling and dilution.
Complex, Montana, the J-M Reef, which has been traced over Tin ore deposits of greisen association owe their origin to
the exsolution below about 700°C of an immiscible aqueous
39 km, contains 49 million tonnes of ore with an overall grade
phase from a fractionating, water-saturated, S-type granitic
of 22.3 g/t LPGE.
intrusion. This low density aqueous phase acts as a repository
for volatiles and metals not readily accommodated in the lat-
Hydrothermal ore deposits tices of crystallizing silicates and eventually collects in the
apical zones of the crystallizing granite. In low f0 2 ilmenite-
This class includes a wide variety of ore deposit types (Pirajno,
type magmas, tin endowed by the continental crustal protolith
1992) that owe their genesis to hot aqueous solutions. The
is maintained as SnZ+, which cannot enter silicate lattices and
origin of these solutions ranges from magmatic to metamor-
therefore becomes concentrated in this aqueous phase.
phic, diagenetic, meteoric and sea water. Isotopic characteriza- Metasomatic alteration of apical portions of the granite by
tion of hydrothermally altered minerals associated with ore these aqueous solutions produces a granoblastic aggregate of
deposits provides an indication of origin, but is not necessarily quartz, muscovite, tourmaline, topaz and fluorite (greisen),
conclusive due to mixing of solutions of different provenance during the formation of which soluble tin complexes are oxi-
and overlap of isotopic fingerprints (Figure 05). The ore dized by the alteration of biotite to muscovite and crystallize
metals are constituents of the original solutions or are leached as cassiterite:
from the rocks through which the solutions are driven by
SnC1 2 + 3KFe3 AISi 3 0 10 (0H) 2 + H 2 0
thermal or gravitational gradients. Hydrothermal solutions
range in temperature from 650°C to 50°C and in pH from ~ Sn0 2 + KAI 3 Si 3 0 10 (0H)z + 6Si0 2
slightly acidic (as low as pH 4 at 350°C) to near-neutral (at + 3Fe 3 0 4 + 2KCI + 4H 2 (01)
450 ORE DEPOSITS

350 and adjacent country rock. Crystallization of the remaining

H9...- \ H+_/ ,.l J
I
I silicate melt is accelerated, producing a porphyritic granite,
/ c1· \ : I H2S ,' I whilst the magmatic hydrothermal fluids effect potassium meta-
300 _It \ ! / ,' I
somatism of the surrounding rocks and precipitate their soluble
u \ j/ ,' Hs· copper, molybdenum and sulfur as chalcopyrite, CuFeS 2 , and
0 250 - Xt OR ,' molybdenite, MoS 2 • Resealing of hydro fractures by mineral
/ \ ,' precipitation leads to cyclical repetition of retrograde boiling
200 / \ I and development of a stockwork of sulfide veinlets and dissem-
/ \\\ . I,' inated sulfides. Entrainment of convecting meteoric water pro-
Hso4./i \ \ I duces a distal propylitic alteration envelope and later causes
150 i \ . I
phyllic and argillic overprinting of the outer zones of the
/
/ \ \i
\ ,-, l
\
\ potassic alteration core, along with remobilization, redistribu-
100 /soz- \:\ \
tion and enrichment of copper sulfides. Ore grades of por-
! 4 \• ·,NH \
phyry-related mineralization are characteristically low, in the

'\
\
•
\
I \1 3 \
/ \._ ranges 0.5-1.0% Cu and 0.1-0.5% MoS 2 , but deposit sizes are
,_
\
50 / \
invariably large, in some cases over one billion tonnes: the
'
i ' NH/'-
,_
\
\ deposit at Chuquicamata, Chile, comprises 9.4 billion tonnes
0 I I I of ore grading 0.56% Cu; Santa Rita, New Mexico, comprises
1 3 5 7 9 11 13 500 million tonnes grading 0.95% Cu; and Climax, Colorado,
comprises 400 million tonnes grading 0.45% MoS 2 .
pH Copper-zinc ore deposits of the volcanogenic massive sulfide
(VMS) types are products of deeply penetrating submarine
Figure 06 Stability of metal-complexing ligands (CI-, HS-, H2 S, hydrothermal convection systems driven by the heat of cooling
NH 3 ) in hydrothermal solutions; curves represent equal activities of tholeiitic basalts at mid-ocean spreading ridges (Cyprus type),
aqueous species and separate their regions of predominance and by mafic and felsic volcanism in island arc subduction
(redrawn with permission from Barnes, 1979). zones (Besshi and Kuroko types). The chloride-bearing hydro-
thermal solutions attain temperatures of ± 350°C and acquire
Sheets of cassiterite mineralization occur immediately below metals by leaching the rocks through which they circulate and
the granite contact and extend downward for several tens of by mixing with metal-rich late magmatic fluids. Where the
meters (enodgreisen), whilst fluids breaking through the roof hydrothermal solution is discharged through faults associated
of the granite effect greisenization and tin mineralization of with local extensional tectonics onto the seafloor, rapid cooling
adjacent silicate country rocks (exogreisen). Tin deposits of precipitates chalcopyrite, CuFeS 2 , sphalerite, ZnS, and pyrite,
this type, with ore grades of 0.2-0.3% Sn, occur at Altenberg, FeS 2 , as mounds and as vertical chimneys venting fine-grained
Germany, and East Kemptville, Nova Scotia. Solutions react- sulfides (black smokers). The periodic collapse of the chimneys
ing with carbonate country rocks also precipitate cassiterite: contributes to the accumulation of a pile of sulfides on and
beneath the seafloor which is subsequently emplaced into the
SnC1 2 + CaC0 3 + H 2 0 ~ Sn0 2 + CaC1 2 + C02 + H 2 upper crust as a volcanogenic massive sulfide deposit. Most of
(02)
these deposits are in the size range of one million to ten million
This gives rise to replacement cassiterite ore deposits, such as tonnes of ore; however, the largest example, Kidd Creek, east-
those of Gejiu region, south China, and the northwest ern Canada, comprises 150 million tonnes of ore grading 2.5%
Tasmanian tinfield, where the ore grades are up to 1% Sn. Cu and 6.0% Zn.
Tungsten ore deposits occur in calc-silicate slcarns adjacent Copper-cobalt ore deposits occur in rift-associated sedi-
to granites emplaced relatively deep in the crust. The skarns mentary basins where metalliferous hydrothermal solutions
form by metasomatism of thermally metamorphosed carbonate infiltrate strata containing syndiagenetic sulfate and sulfide.
sediments by aqueous solutions (± 500°C) expelled from the The hydrothermal solutions originate from compactional
cooling granite, followed by retrograde metamorphism in the dewatering of basin sediments and possibly rift-related magma-
presence of hydrothermal fluids of more diverse provenance. tism, and rift-related heat flow drives their convective recircula-
During the metasomatic stage calcium released from marble tion during which they acquire metals through leaching the
precipitates tungsten, soluble as complexes in the aqueous fluid crystalline basement or the comminuted volcaniclastic detritus
from the granite, as scheelite (CaW0 4 ). Later, calcium released in the sedimentary pile. The metals are transported as chloride
by retrograde alteration of pyroxene and garnet also reacts complexes and precipitate as chalcopyrite, CuFeS 2 , bornite,
with tungsten in solution, increasing the grade of scheelite. Cu 5 FeS 4 , and carrolite, CuCo 2 S4 , in continental clastic redbed
Skarn tungsten ore deposits, such as those of MacMillan Pass, aquifers (by the thermal reduction of interstitial diagenetic
northern Canada, King Island, Tasmania, and Sandong, south- anhydrite at ± 250°C) and in the overlying marine argillaceous
east Korea, comprise a few tens of millions of tonnes of ore sediments (by replacement of diagenetic pyrite at 210oC to
with grades up to 1% wo3- 140°C). In the Central African Copperbelt some 20 stratiform
Porphyry copper (and molybdenum) ore deposits are related deposits comprise resources of over five billion tonnes of ore
to the final cooling stages of hydrous 1-type magmas generated with grades of 2.5-4.5% Cu and 0.2% Co.
in subduction zones and emplaced as subvolcanic granitic Lead-zinc ore deposits are found in sedimentary basins in
intrusions at depths of 1-4 km in the crust. Aqueous fluids which comparatively low temperature ( 50-200°C), chloride-
exsolved from the cooling magma accumulate in its apical rich, hydrothermal solutions (brines) leach and transport the
region until they undergo retrograde boiling and escape by metals as soluble chloride complexes. Stratabound Mississippi
hydro fracturing (crackle brecciation) of the intrusion carapace Valley Type (MVT) deposits are formed as a result of connate
 ORE DEPOSITS 451

water, expelled during compressional dewatering, leaching and the Porcupine district, Ontario. Many epithermal deposits
lead, zinc, oil and methane from argillaceous sediments at the are found in Nevada, but the highest grade ore of this type is
montmorillonite-illite transition (a burial depth of about 2 km) 225 gjt Au at El Indio, Chile.
and migrating by gravity-driven flow or heat from a nearby Uranium ore deposits accumulate where hexavalent ura-
magmatic intrusion into the shallow-water sediments around nium, soluble as uranyl complexes in surface waters and oxic
the basin margins, where methane reduces gypsum and anhy- ground waters since the oxygenation of the atmosphere in the
drite in evaporites: mid-Proterozoic, is reduced by pyrite, organic matter or graph-
ite to the tetravalent state and precipitates as uraninite:
CaS0 4 • 2H 2 0 + CH 4 = H 2 S + CaC0 3 + 3H 2 0 (03)
The H 2 S precipitates galena, PbS, and sphalerite, ZnS, in 2U0 2 (CO)j- + 8H+ + C = 2U0 2 + 7C0 2 + 4H 20 (07)
nearby reef limestones that have experienced porosity enhance- In the Athabasca Basin, Saskatchewan, and the intracratonic
ment by dolomitization, karstification or brecciation, and Pine Creek geosyncline, northern Australia, major uranium
whose porosity is then further increased by acid dissolution: deposits occur at and below unconformities where Middle
(04) Proterozoic sandstones rest upon an earlier basement cut by
faults or disturbed by brecciation. Uranium originally leached
Individual MVT deposits are small, but they tend to occur in from Archaean rocks, and possibly incorporated into Lower
clusters, as at Pine Point, northern Canada, where within an Proterozoic sediments, was remobilized and concentrated
area of 7000 km 2 there are some forty deposits totalling 95 during the Upper Proterozoic by convecting saline and COr
million tonnes of ore with average grades of 2.5% Pb and 6.2% rich diagenetic or meteoric hydrothermal solutions. These were
Zn. Sedimentary exhalative (SEDEX) deposits acquire their heated to temperatures between 1oooc and 200°C by magmatic
metals in the same way as MVT deposits or through leaching events and driven along permeable formations near the uncon-
of basin sediments by seawater convection cells penetrating to formities to redox boundaries. Smaller sandstone-hosted ura-
a depth of about I 0 km. The metalliferous brines vent through nium deposits in Upper Paleozoic continental sedimentary
rifting-associated faults onto the seafloor and flow into distal basins of the western USA share similarities in genesis, with
depressions, where they lose heat and precipitate galena and uranium leached from granites or volcanics by meteoric water
sphalerite. These deposits include the large lead-zinc ore bodies carried through oxic sandstone aquifers and precipitated at
of northern Australia, such as Mount Isa with 87 million redox fronts. Uranium ore bodies range in size from one million
tonnes of ore grading 7.I% Pb and 6.1% Zn. tonnes up to 50 million tonnes and in grade from 0.1% to
Gold ore deposits occur as quartz veins, pervasive veinlets 9.0% U 3 0 8 .
and disseminations. Those of epithermal origin are mainly
Mesozoic and Tertiary in age and precipitated from convec-
tively circulating meteoric water which was heated by nearby Autochthonous ore deposits
volcanic or plutonic centers and acquired its gold by leaching Several metals that are relatively abundant in certain crustal
volcano-sedimentary rocks or mixing with late magmatic rocks attain ore grades through their selective precipitation in
hydrothermal fluids. Some lode gold deposits of Archean particular sedimentary or subaerial weathering environments.
greenstone belts have a similar genesis, but most precipitated These processes are characterized by ambient temperatures
from low salinity gold-bearing hydrothermal solutions that and by the separation and concentration of metals as a result
originated from metamorphic dewatering of lower parts of of their differing aqueous solubilities under different pH and
volcano-sedimentary sequences, or from acid magmatism, or Eh conditions (Figure 07).
from mantle degassing, and were channelled into shear zones Manganese ore deposits are shallow marine and estuarine
associated with large-scale crustal lineaments by seismic pump- sediments, up to 10m thick and with considerable lateral conti-
ing. Gold remains soluble in low-temperature hydrothermal nuity, deposited during periods of marine transgression or high
solutions as a bisulfide complex, destabilization of which leads sea level stands. Tropical weathering of Precambrian shield or
to its precipitation as native metal. Archean lode gold deposits similar source rocks containing < 0.5% Mn supplies manganese
precipitated from gold-bearing hydrothermal fluids flowing as soluble Mn2+ and as the constituents of argillaceous sedi-
through shear zones at the depth range of greenschist-facies ments, from which Mn 2 + is subsequently released into seawater
metamorphism and at temperatures of 400°C to 300°C; sulfi- as a result of anoxia induced by sea level rises (Force and
dation of iron-bearing wall-rock minerals such as chlorite, Cannon, 1988). In the shorewards oxic environment, bacterial
ilmenite and magnetite led to precipitation of gold and pyrite: oxidation of soluble Mn 2 + to less soluble Mn 4 + leads to the
precipitation of oxides, such as pyrolusite, Mn0 2 , which form
Au(HS)2 + FeO =Au+ FeS 2 + H 2 0 (05)
pisolites and oolites. The Oligocene manganese deposits of
In epithermal systems, gold precipitates between 300°C and Nikopol, Ukraine, are ± 3 m in thickness and comprise several
200°C due to cooling or boiling of the hydrothermal fluid, or hundred million tonnes of ore grading 15-25% Mn, while at
by reduction of the bisulfide complex with graphite or its Groote Eylandt, northern Australia, Cretaceous deposits
oxidation by oxygenated groundwater: ± 9 min thickness have undergone subaerial supergene enrich-
ment producing some 400 million tonnes of ore grading
2Au(HS)2 +H 20+7!0 2 =2Au+6H+ +4SO~- (06)
46% Mn.
Ore deposits vary in size from about a million tonnes up to Iron ore deposits are, for the most part, supergene-enriched
tens of millions of tonnes. Ore grades are typically 5-10 gjt iron-rich oxide facies of Banded Iron Formation (BIF). During
Au, but adiabatic boiling of epithermal solutions at ± 280°C the Lower Proterozoic, BIF precipitated from shallow seas
precipitates almost all gold over a very narrow temperature over large areas on continental shelves and prograding coast-
range, and can form very high grade deposits. Major Archaean lines in intracratonic and evaporite basins. The oxide facies of
lode gold deposits occur in the Golden Mile, Western Australia, the BIF is a seasonally varved millimeter to centimeter scale
452 ORE DEPOSITS

soluble alkalis and silicic acid, residual clays, and in some cases
clays of aeolian origin, are altered by further percolating rain-
water to amorphous aluminum oxides, mainly gibbsite:
Al 2 Si 2 0 5 (0H) 4 + 5H 2 0 = 2Al(OH) 3 + 2H 4Si04 (09)
During the dry season, metastable amorphous gibbsite is con-
verted to its crystalline analog, which is stable over a wide pH
range and often grows as pisolites. The world's largest bauxite
deposit is at Sangaredi, Guinea, where 180 million tonnes of
ore grading 60% Al 2 0 3 forms a plateau up to 30m thick.
Bauxite deposits in which boehmite, AlO(OH), is a predomi-
nant mineral, occur on, but are not necessarily genetically
related to, karstified limestone and dolomite, as in the Halimba
Basin, Hungary.
Lateritic nickel ore deposits form in the saprolite and limo-
nite horizons developing through the subaerial weathering of
serpentinized peridotite under conditions of warm tropical (not
necessarily seasonal) rainfall and free downward drainage.
Olivine, containing typically 0.2% Ni, breaks down to leave
smectite clays (or at higher pH, quartz) and the soluble pro-
ducts are flushed down the weathering profile: Mg2+ is lost
into the groundwater; Fe 2 + reprecipitates in the limonitic hori-
0 2 4 6 8 10 12 14 zon as goethite on which some Ni 2 + is adsorbed; and most
Ni 2 + reacts with serpentine in the saprolite horizon to form
pH garnierite (Golightly, 1981):
Figure 07 Differing aqueous solubilities of AI, Fe and Mn at 25°( 4Mg2 Si0 4 + lOH+ = Mg 3 Si 4 0 10 (0H) 2 + 5Mg2 + + 4H 2 0
and activities of 10- 6 m: Fe and Mn are both soluble at low pH (010)
(vertical shading), but only Mn is soluble at intermediate pH 4Fe 2 Si0 4 + 8H+ + 40 2 = Fe 2 Si 40 10 (0H) 2 + 6FeO(OH)
(horizontal shading) and both are insoluble at high pH; AI is
(Oil)
soluble at both low and high pH (gray shading) but is insoluble at
intermediate pH (compiled from Brookins, 1988). Mg 3 Si 2 0 5 (0H) 4 + 3Ni + = Ni 3 Si 2 0 5 (0H) 4 + 3Mg +
2 2

(012)
alternation of iron oxides (hematite, Fe 2 0 3 , magnetite, Fe 3 0 4 ) Through gradual uplift and erosion the zone of nickel enrich-
and chert, with an overall composition of about 30'/o Fe and ment, which reaches its peak about 10 m below the surface,
45% Si0 2 . Iron and silica were supplied by erosion of nearby migrates downward and nickel accumulates to ore grade. The
landmasses and surface transport of dissolved Fe2+, soluble in deposit at Greenvale, Queensland, comprises 40 million tonnes
the 0 2 -deficient, C0 2 - rich atmosphere of the time, or by exha- grading 1.57% Ni.
lation of seawater convecting through and leaching the under-
lying crust at temperatures reaching < 300°C. Where BIF is Future developments
amenable to beneficiation, as in the Lake Superior district, Ore deposits are geochemical anomalies of commercial impor-
USA, it constitutes a low grade iron ore. The higher grade tance and models of their genesis occupy a dynamic interface
ores of the Hamersley Basin, Western Australia, are the pro- between Earth Science research and the minerals industry:
duct of supergene enrichment of BIF, involving the oxidation exploitation gives access to sample materials and observations
of magnetite, formation of goethite, Fe(OHh, and iron meta- for research; and the resuHs are incorporated into genetic
somatism of silicates, followed in the mid-Proterozoic by burial models which are applied to the selection of areas for explora-
which raised temperatures to ± 100°C and converted goethite tion for more deposits. The combination of a steady supply of
to microplaty hematite (Morris, 1985). The Hamersley Basin new sample materials and observations from ore deposits, each
contains 33 billion tonnes of ore with a grade of >55% Fe of which is slightly different, and the formulation of models
and includes deposits of almost pure hematite, such as Mount which often invoke geologic processes that cannot be fully
Tom Price with 700 million tonnes of ore grading 64% Fe. verified ensures that models are constantly revised or super-
Bauxite (aluminum ore) deposits are the residue of sub- seded; thus Adams (1991) lists no less than 150 variants of
aerial leaching of aluminous, usually volcanic or intrusive, models that have been published, mostly in the preceding 20
parent rock. The requisite conditions for bauxitization include years, solely for sandstone-type uranium deposits. Whilst Earth
a warm tropical climate with alternating wet and dry seasons, scientists strive to improve existing models and the minerals
gently undulating relief and good downward drainage industry continues to use them, the high level of scientific
(Bardossy and Aleva, 1990). The climate promotes growth of interest in ore deposits and the geochemical and other controls
vegetation which supplies organic acids to percolating rainwa- on their genesis appears likely to continue.
ter, and this low pH solution dissolves rock-forming silicates
such as feldspar leaving clay minerals such as kaolinite: Martin Hale
2KA1Si 3 0 8 + 2H+ + 9H 20 = Al 2 Si 2 0 5 (0H) 4 + 2K + Bibliography
+ 4H 4 Si0 4 (08) Adams, S.S. (1991) Evolution of genetic concepts for principal types
of sandstone uranium deposits in the United States. Econ. Geol.
In an environment in which free downward drainage removes Monogr., 8, 225-48.
 ORGANIC GEOCHEMISTRY 453

Bardossy, G. and Aleva, G.J.J. (1990) Lateritic Bauxites. Amsterdam: acids, porphyrins, chlorophylls, carotenoids, sugars, proteins,
Elsevier, 624 pp. carbohydrates and anthropogenic compounds. Scientists in the
Barnes, H.L. (1979) Geochemistry of Hydrothermal Ore Deposits,
2nd edn. New York: Wiley, 798 pp.
field also borrow knowledge and techniques freely from the
Brookins, D.G. (1988) Eh-pH Diagrams for Geochemistry. Berlin: disciplines of chemistry, geology, microbiology, biology, pale-
Springer-Verlag, 176 pp. ontology, ecology, physics, mathematics, limnology and ocean-
Force, E.R. and Cannon, F.W. (1988) Depositional models for shal- ography. This wide knowledge base contributes greatly to the
low-marine manganese deposits around black shale basins. Econ. fascination of the subject.
Geol., 83, 93-117. Modern organic geochemistry often requires access to highly
Golightly, J.P. ( 1981) Nickeliferous laterite deposits. Econ. Geol., 75th
Anniversary Volume, 710-35. sophisticated analytical equipment to study the low abun-
Hawley, J.E. ( 1962) The Sudbury ores, their mineralogy and origin. dances and complex mixtures of compounds present in geologi-
Can. Mineral., 7, 1-207. cal materials. Capillary gas chromatography (GC) linked to a
Morris, R.C. (1985) Genesis of iron ore in banded iron formation by variety of detectors, and capillary GC linked to a mass spec-
supergene and supergene metamorphic processes - a conceptual trometer (GC-MS) are widely used for the analysis of C 1-C60
model, in Handbook of Stratabound and Stratiform Ore Deposits compounds. High sensitivity is combined with the requirement
(ed. K.H. Wolf). Amsterdam: Elsevier, Vol. 12, Ch. 12.
Pirajno, F. (1992) Hydrothermal Mineral Deposits. Berlin: Springer- to separate small amounts of a compound present in complex
Verlag, 709 pp. mixtures. High performance liquid chromatography (HPLC)
Sheppard, S.M.F. (1986) Characterization and isotopic variations in is the method of choice for more polar compounds such as
natural waters. Rev. Mineral., 16, 165-83. carotenoids and chlorophylls or for high molecular weight
Whitney, J.A. and Naldrett, A.J. (eds) (1989) Ore Deposition compounds such as humic acids. Pyrolysis, either on-line or
Associated with Magmas. Rev. Econ. Geol., 4. El Paso: Society of
Economic Geologists, 250 pp.
off-line, or oxidative procedures, are often used to break down
the organic matter into smaller components which can then be
analyzed by GC and HPLC. 1 H and 13 C nuclear magnetic
Cross-references resonance (NMR) and infrared (IR) spectrometers are also
Clay membranes used to provide structural information on the organic matter
Epigenesis in kerogen, coals or sediments.
Fluid-rock interactions A central concept common to many studies in organic geo-
Geochemical exploration
Geothermal systems chemistry is the use of biomarkers (also termed biological
Hydrothermal alteration markers, chemical fossils and signature lipids) to infer likely
Hydrothermal solutions sources of organic matter. These organic compounds have a
Hypogene distinct chemical structure which can be related by known
Natural resources microbial or chemical transformation pathways to compounds
Paragenesis
Supergene
synthesized by specific groups of organisms. For example, the
polycyclic hydrocarbon dinosterane which occurs in some
crude oils is derived from the C 30 sterol dinosterol, which is
common in a class of marine phytoplankton called dinoflagel-
ORGANIC GEOCHEMISTRY lates. The presence of this hydrocarbon in oils provides good
evidence that some of the organic matter in the oil's source
Organic geochemistry is the study of the origins and fates of rock was derived from dinoflagellates. Hopanes and hopenes
organic matter in the geosphere, and it encompasses studies of are derived from hopanoids such as bacteriohopanetetrol syn-
coals, oils, sediments and natural waters. It is a relatively new thesized by many bacteria and cyanobacteria. Many new lipids
field of endeavor which traces its origins to the identification have been discovered in plants, algae and bacteria during the
by Treibs in 1934 of alkyl metalloporphyrins related to chloro- course of organic geochemistry studies carried out in the past
phyll in crude oils. This landmark research provided the first two decades. Since some organisms are restricted to particular
convincing evidence for the biological origin of some of the environments, the presence of certain biomarkers can be used
constituents of petroleum. Since then the field has expanded to infer the environmental conditions which prevailed when
greatly and has become considerably more interdisciplinary as the sediment was deposited.
it seeks answers to how organic matter is produced, trans- Just as paleontologists use morphological fossils as strati-
ported, recycled, transformed and preserved in modern and graphic tools, organic geochemists use 'chemical fossils' to
ancient environments. define an analogous 'molecular stratigraphy'. The concept that
Much of organic geochemistry can be classified into general 'the present is the key to the past' has been the driving force
themes such as production and fate of organic matter in aquatic for numerous studies of organic compounds in modern aquatic
ecosystems; chemical composition of biologically produced environments. These have provided fresh insights into the
organic matter; the effect of environmental conditions on the sources of organic matter and how it is degraded and trans-
preservation and degradation of organic matter in modern ported within each ecosystem. Regions of high productivity
sediments; processes and reactions leading to the formation of such as the Black Sea as well as the upwelling areas off Peru
humic material, coals and kerogen; generation and composi- and Namibia have been a major focus since these provide
tion of petroleum and natural gas; catagenesis. To these can recent models for the formation of ancient organic-rich black
be added applications in environmental chemistry tracing the shales (sapropels) which can be excellent petroleum source
source, transport and fate of organic pollutants; studies of rocks.
climate change and paleoenvironment; isotope geochemistry; Organic geochemists seek to identify biomarkers which
chemical oceanography; meteorites and moon rocks; molecular might be unique to each type of depositional environment. For
archeology and many more. The field is not limited to studies example, the presence of an oxic-anoxic boundary in the water
of lipids and hydrocarbons, but encompasses studies of amino column of bodies such as the Black Sea is recorded in the
454 ORGANIC GEOCHEMISTRY

sediments by the preservation of biomarkers derived from the which otherwise would be quite labile and readily degraded in
bacteriochlorophylls and carotenoids of obligate phototrophic sediments. It also provides the link between the carbon and
green sulfur bacteria present in the water column. Many bio- sulfur cycles in sedimentary environments.
markers have been described for higher plants and thus the Another recent application of organic geochemistry is to the
identification of terrestrial organic matter in sediments is rela- field of archeology, particularly in the characterization of pre-
tively straightforward even when the organic material is trans- served organic materials. Some applications have included the
ported far into the sea. confirmation that asphalt from the Dead Sea was used in the
Marine organic geochemistry is now a well established field mummification process from at least 2600 BC. This finding
in the broader area of chemical oceanography. Some recent relied on showing that sterane distributions in modern asphalts
applications include the determination of chlorophyll and and the black material in mummies were identical, and distinct
carotenoid pigment distributions in seawater using HPLC. from local oil seeps. Food remains in pottery vessels have been
Chlorophyll a concentrations provide an indirect measure of identified from characteristic distributions of fatty acids, sterols
total phytoplankton biomass, while the abundance of specific and wax esters, thus providing important data on food prefer-
carotenoids can be used to characterize the community com- ences in ancient civilizations. Wood tar pitches are also amena-
position of marine phytoplankton in oceanic waters. For exam- ble to detailed chemical examination and have been shown to
ple, prasinoxanthin is found in certain green microalgae, 19'- be in widespread use from the mesolithic period up to the
hexanoyloxyfucoxanthin in haptophytes and peridinin in dino- Middle Ages. Quantities of archeological material as small as
flagellates. Such studies have revealed the important role 10-100 mg can be characterized using these GC-MS tech-
played by eukaryotic nanoplankton and picoplankton and the niques. Genetic fingerprinting techniques are also widely being
prokaryotic cyanobacteria and prochlorophytes in oligotrophic used although contamination with modern materials still pres-
oceanic waters. ents considerable analytical problems.
Chlorophylls are degraded in marine and lacustrine sedi- Isotope ratios are increasingly being used to deduce origins
ments, mostly to colorless products although some macrocycles of organic matter. Much of this work has concentrated on
survive as porphyrin derivatives. Carotenoids degrade very 13 C/ 12 C ratios, although this approach is being extended to
rapidly and thus are rarely useful as source indicators in older nitrogen, sulfur, oxygen and hydrogen isotopes. The 13 Cjl 2 C
sediments. Their degradation rates vary widely and are not ratio in organic compounds reflects the isotopic composition
simply a function of their high degree of unsaturation. of the original C0 2 or bicarbonate ion plus the combination
Fucoxanthin and related pigments break down to loliolide, of thermodynamic and kinetic isotope fractionations occurring
isololiolide and dihydroactinioliolide (via unstable bicyclic during assimilation and metabolism. The range of isotope
furanoxides) which, however, can be detected in sediments. ratios found varies from one group of organisms to another.
Specific branched and/or unsaturated fatty acids are used to Thus plants which use the C3 pathway have .5 13 C values of
determine the biomass of different bacterial groups in sedi- -23 to - 34%o, whereas those using the C4 pathway are
ments and seawater. !so- and anteiso-branched C 15 and C 17
isotopically heavier and typically range from -6 to - 23%o.
fatty acids and cis-vaccenic acid (18: lw7) are found in many
The isotope ratio of the bulk organic matter in sediments is
bacteria, whereas 10-methylhexadecanoic acid and iso 17: lw7
an average of all the sources present and can be used as a
are more commonly associated with sulfate-reducing bacteria.
signature of autotrophic carbon fixation. Recent advances in
The abundance of these acids in the phospholipid fraction
instrumentation such as the development of a capillary gas
isolated from sediments can be used as a measure of the
chromatograph linked via a combustion tube to an isotope
microbial biomass present.
ratio mass spectrometer now permit the 13 Cj1 2 C ratio of indivi-
Polyunsaturated fatty acids and certain sterols provide evi-
dual compounds to be determined. This exciting development
dence for contributions to the organic matter in Recent sedi-
ments or seawater from microalgae. A new class of very long- has revealed remarkable differences in isotope ratios and
chain unsaturated straight-chain ketones, called alkenones, confirmed the diversity of sources giving rise to the distribu-
derived from some haptophyte algae such as Emiliania huxleyi, tions of hydrocarbons in ancient sediments and crude oils.
are now being used to reconstruct past surface ocean temper- Measurements of the 13 C/ 12C ratio in alkenones in ancient
atures from the ratio of di- and tri-unsaturated alkenones in sediments have also enabled calculations to be made of
sediment cores. The technique uses the fact that in the alga pC0 2 (aq) of the oceanic waters present when the sediments
the ratio of the di-unsaturated alkenone relative to that of the were deposited.
tri-unsaturated component increases as the growth temperature Although only a small quantity of the organic matter pro-
increases. These data allow long-term changes in global climate duced in the water column reaches the sediment, the amount
to be examined, and then compared with inferences drawn of organic carbon in the lithosphere (estimated at 2 x 10 16
from oxygen isotope studies and changes in assemblages of tonnes) is vastly greater than that found in the oceans ( 1 x 10 12
benthic foraminifera. tonnes). Once the organic matter is deposited, it is further
One area of increasing study is the natural incorporation of transformed by bacterial and chemical processes leading to
sulfur into organic matter during the earliest stages of diagen- breakdown and transformation of the organic compounds pre-
esis leading to the formation of organic sulfur compounds sent. Macromolecules (also referred to as geopolymers) are
(OSC). This reaction is facile in anoxic environments or when formed by condensation reactions giving rise to humic acids,
reactive iron is not abundant. It usually proceeds with sulfur fulvic acids and humins. With continuing burial the organic
attachment to the carbon atom originally substituted with a matter is further transformed to become the insoluble organic
functional group such as double bond or carbonyl group. The matter called kerogen. This occurs at relatively low temper-
inorganic sulfur species may be either hydrogen sulfide, poly- atures less than 50°C. Resistant biopolymers produced by some
sulfides or elemental sulfur. This reaction is geochemically plants and algae are also preserved and new research has
important since it leads to selective preservation of compounds shown that it is these rather than the 'geopolymers' in kerogen
 ORGANIC GEOCHEMISTRY 455

which ultimately are the major source of aliphatic hydro- as then-alkanes and methyl-branched compounds. Isoprenoids
carbons produced when the kerogen is thermally cracked to are more resistant as are the bicyclic alkanes. Steranes and
produce crude oils. This zone of maximum hydrocarbon pro- diasteranes are remarkably resistant to biodegradation; thus
duction is usually referred to as the catagenic zone. At later they are often used to correlate oils even those whose composi-
stages gaseous hydrocarbons such as methane tend to dominate tion have been greatly modified by microbial action.
the hydrocarbons produced and at even higher burial temper- Eventually, even steranes can be biodegraded and the first
atures exceeding 150°C, the metagenesis stage is reached where isomers to be removed are those still possessing the original
the kerogen is converted to a carbon residue. 'biological' stereochemistry. Similar techniques are often
One of the main commercial applications of organic geo- applied in environmental studies to identify sources of spilled
chemistry has been to the field of petroleum exploration and, crude oil or to study the pathways by which the oil is degraded
indeed, this has often dominated the science. These techniques in the environment. A notable example was the use of the
have also proven valuable for studying the origins of coals, abundance of the C 30 hopane as an internal marker for assess-
peats and lignites. Characteristic distributions of biomarkers ing the rate at which oil derived from the Exxon Valdez oil
have been used to correlate crude oils in reservoirs with their spill in Alaska was degraded in sediments.
source rocks. Such information enables the explorationist to Organic geochemical techniques have not been restricted to
identify petroleum migration paths, estimate the oil prospecti- samples from Earth, but have also been used to characterize
vity of a field and suggest further sites for drilling. Changes in organic matter in meteorites and to look for chemical evidence
biomarker isomer ratios can be used to infer the extent of of past life on the Moon and Mars. A plethora of papers have
thermal maturation and hence whether a potential source rock appeared since the 1970s on the amino acids found in meteor-
lies within the 'oil window'. The presence of specific biomarkers ites and a number of hypotheses have been advanced to explain
can often be used to deduce the type of environmental setting their extraterrestrial formation. These have included
when the sediments were deposited. Miller-Urey synthesis, Fischer-Tropsch synthesis, photochem-
The formation of petroleum depends on a number of factors ical ion-molecule reactions and aqueous alteration of pre-exist-
including the type and amount of organic matter in the source ing condensed phases. More recent work has identified other
rocks, the thermal maturity of the organic matter and the compound classes such as hydroxy fatty acids, dicarboxylic
presence of suitable migration pathways, traps and seals. The acids and polycyclic aromatic hydrocarbons. Complex aro-
type of organic matter, which is usually expressed in terms of matic molecules have also been identified in interplanetary dust
maceral composition, can be determined by optical or physico- particles. In 1996, NASA scientists put forward evidence that
chemical means. Hydrogen-rich macerals such as alginite are life may have been detected on Mars based on analysis of a
considered to be a much better source of petroleum (i.e. they Martian meteorite found in Antarctica (McKay eta!., 1996).
are oil-prone), than inertinite or vitrinite. The reflectance of This evidence included the presence of polycyclic aromatic
vitrinite macerals (measured as vitrinite reflectance; R 0 ) is often hydrocarbons and the occurrence of putative bacteria-like fos-
used as a measure of thermal maturity; the main phase of oil sils in carbonate globules containing magnetite and Fe-sulfide
generation usually occurs in the range 0.65-1.3R 0 . mineral phases which the scientists conjectured could be
Pyrolysis can provide information about the amount, type explained by biogenic processes.
and thermal maturity of the organic matter. A French-designed Modern organic geochemistry is a multifaceted science which
commercial instrument called the Rock-Eva! is now widely has only been briefly touched upon here. For further details
used to characterize source rock samples. In this system, the the interested reader is referred to individual articles on
finely ground source rock is heated in an inert atmosphere specific topics.
under various temperature conditions to yield free hydro-
carbons (the sl peak) plus hydrocarbons liberated by pyrolysis John K. Volkman
of the kerogen (the S 2 peak). The amount of hydrocarbons
that can be liberated from the rock (the genetic potential) is
simply S 1 plus S 2 expressed as a proportion of the mass of Bibliography
rock. Thermal maturity can be estimated from the temperature Brooks, J. and Fleet, A.J. (eds) (1987) Marine Petroleum Source Rocks.
at which the maximum in the S 2 response occurs (Tmax). London: Blackwell Scientific, 435 pp.
Basin modeling is now widely used to reconstruct the history Clemett, S.J., Maechling, C.R., Zare, R.N., Swan, P.D. and Walker,
of oil and gas generation in a particular region. In this process, R.M. ( 1993) Identification of complex aromatic molecules in indivi-
dual interplanetary dust particles. Science, 262, 721-5.
geochemical information about the factors that affect petro- Engel, M.H. and Macko S.A. (eds) (1993) Organic Geochemistry:
leum formation such as source rock quality and migration is Principles and Applications. New York: Plenum Press, 861 pp.
integrated with geologic information on depositional history, Fleet, A.J., Kelts, K. and Talbot, M.R. (eds) (1988) Lacustrine
subsidence rates, erosion and tectonic events, and geothermal Petroleum Source Rocks. Oxford: Blackwell, 391 pp.
gradients to provide a conceptual framework showing how the Hunt, J.M. ( 1979) Petroleum Geochemistry and Geology. San
basin evolved. Numerical simulations are then carried out to Francisco: W.H. Freeman, 617 pp.
Jasper, J.P., Hayes, J.M., Mix, A.C. and Prahl, F.G. (1994)
investigate whether all the factors necessary for oil generation Photosynthetic fractionation of 13 C and concentrations of dissolved
and accumulation can be found in the basin, and at what C0 2 in the central Equatorial Pacific during the last 255,000 years.
geologic time, thus substantially minimizing the exploration Paleoceanography, 9, 781-98.
risk. Such kinetic models still lack refinement, but nonetheless Johns, R.B. (ed.) (1986) Biological Markers in the Sedimentary Record.
can provide vital information about the critical period in basin Amsterdam: Elsevier, 364 pp.
evolution when oil was likely to have been produced. Killops, S.D. and Killops, V.J. (1993) An Introduction to Organic
Geochemistry. London: Longman Scientific and Technical, 265 pp.
Many oils are biodegraded in the reservoir causing character- Manning, D.A.C. (1991) Organic Geochemistry. Advances and
istic changes in hydrocarbon composition. The first hydro- Applications in the Natural Environment. Manchester: Manchester
carbons to be removed are those with simple structures such University Press, 662 pp.
456 ORGANIC MATTER IN FOSSILS

McKay, D.S., Gibson, E.K. Jr., Thomas-Keprta, K.L. eta/. (1996) altered materials such as humic acids and other condensation
Search for past life on Mars: possible relic biogenic activity in products. Additionally, other well-preserved fossils may con-
Martian meteorite ALH84001. Science, 273, 924-30.
Meyers, P.A. and Ishiwatari, R. ( 1993) Lacustrine organic geochemis-
tain rich amounts of organic matter which include plant debris
try- an overview of indicators of organic matter sources and dia- such as woods, coals and peats and inclusions of animals or
genesis in lake sediments. Org. Geochem., 20, 867-900. plants in amber. In the 1950s, Philip Abelson at the
Moldowan, J.M., Albrecht, P. and Philp, R.P. (eds) (1992) Biological Geophysical Laboratory first recognized the occurrence of
}\1arkers in Sediments and Petroleum. New Jersey: Prentice Hall, amino acids, the building blocks of proteins, in fossil shells.
411 pp.
Orr, W.L. and White, C.M. (eds) (1990) Geochemistry of Sulfur in
Subsequent to these observations, a large variety of investiga-
Fossil Fuels. Washington, DC: American Chemical Society tions on the nature and use of this material has been under-
Symposium Series 429. taken (Mitterer, 1993; Wehmiller, 1993). Because this material
Peters, K.E. and Moldowan, J.M. (1993) The Biomarker Guide. is related to the organism at the time of its death, specific
Interpreting Molecular Fossils in Petroleum and Ancient Sediments. chemical components of the organic matter have been used to
New Jersey: Prentice Hall, 363 pp.
Repeta, D.J. ( 1989) Carotenoid diagenesis in recent marine sediments:
establish metabolic and dietary relationships, or to serve as
III. Degradation of fucoxanthin to loliolide. Geochim. Cosmochim. proxies for characteristics of the environment in which it lived.
Acta, 53, 699-707. For example, the content and isotopic composition of atmo-
Rullkiitter, J. (1987) Geochemistry, organic, in Encyclopedia of spheric carbon dioxide, and the productivity of the ocean have
Physical Science and Technology, Vol. 6. London: Academic Press, been estimated from fossil organic matter.
pp. 53-77. Amino acids in fossils, originating from proteins, are com-
Shock, E.L. and Schulte, M.D. (1990) Summary and implications of
reported amino acid concentrations in the Murchison meteorite. monly used for relative age dating, stratigraphic correlation,
Geochim. Cosmochim. Acta, 54, 3159-73. paleothermometry and chemotaxonomy (Weiner et al., 1979;
Tissot, B.P. and Welte, D.H. (1984) Petroleum Formation and Weiner and Lowenstam, 1980; Mitterer, 1993; Wehmiller,
Occurrence. Berlin: Springer-Verlag, 699 pp. 1993). The utilization of amino acids for age determination
Volkman, J.K. ( 1988) Biological marker compounds as indicators of and stratigraphic correlation is based on the observation that,
the depositional environments of petroleum source rocks. Geo/.
Soc. Spec. Pub/., 40, 103-22. except for bacterially or fungal-derived materials, most amino
Wakeham, S.G. ( 1993) Reconstructing past oceanic temperatures from acids in living organisms are of the L configuration. Following
marine organic biogeochemistry, chemical fossils and molecular the death of an organism, the amino acids undergo racemiza-
stratigraphy. Environ. Sci. Techno/., 27, 29-33. tion, the length of time varying with the amino acid, until
eventually becoming an equilibrium or racemic mixture of both
Cross-references the D and the L enantiomers for each amino acid. At constant
Biomarker: coals temperature, the reaction rate for this process is thought to be
Coal: organic petrography described by the first order rate law of kinetics. Thus, in closed
Coal: origin and diagenesis systems, the time since the death of the organism may be
Coal: types and characteristics determined if the temperature that the fossil has been exposed
Coal: vitrinite reflectance and maturity assessment
Geochemistry: low temperature to since the time of deposition can be estimated. Except for
Hydrogen isotopes deep sea environments, this temperature estimate is often
Laboratory simulations of oil and natural gas formation difficult, and reliance on this tool for age determination for
Occurrence of organic facies terrestrial or marine deposits that are above sea level has been
Oil seeps and coastal bitumens relegated to correlative rather than absolute ages. The racemi-
Oil shales
Oil-oil and oil-source rock correlation
zation rates are dependent also on the specific amino acid,
Organic matter in fossils allowing for determination of ages from hundreds of years to
Organic matter in meteorites hundreds of thousands of years.
Organics: contemporary degradation and preservation In addition, age determination can be affected by contamina-
Organics: sources and depositional environments tion of the fossil organic matter since the time of deposition.
Peat Geologic materials rarely exist as closed systems from the time
Petroleum
Petroleum: hydrothermal of burial to exhumation, thus presenting a serious challenge
Petroleum: in-reservoir biodegradation to the interpretation of the molecules contained within a fossil
Petroleum: kinetic modeling matrix. The estimation of indigeneity of the amino acids in
Petroleum: primary migration fossils might be possible through the use of the stable isotope
Petroleum: surface geochemistry compositions of the nitrogen and carbon in the amino acid
Petroleum: types, occurrence and reserves
Porphyrins stereoisomers (Serban et al., 1989; Macko and Engel, 1991;
Precambrian organic matter Engel et al., 1994 ). During racemization, there is little isotopic
Sulfate reduction fractionation of carbon or nitrogen (Engel and Macko, 1986),
Sulfur isotopes Thus, indigenous amino acids should have stereoisomers which
have the same isotopic composition, whereas contaminants
would be different. The measurement of these isotopic com-
ORGANIC MATTER IN FOSSILS positions has recently been facilitated by improvements in
isotopic instrumentation, specifically the interfacing of a gas
Organic matter in fossils of bone and shell consists primarily chromatograph (GC) with chiral separation ability to a com-
of residual proteinaceous constituents (e.g. peptides, amino bustion furnace (C), from which the effluent goes directly into
acids) and, in the best of preservation circumstances, possibly an isotope ratio mass spectrometer (IRMS). Isotopic analysis
relic protein itself. However, organic matter in fossils may is now possible on subnanomol amounts of components by
include lipids, sugars and nucleic acids as well as diagenetically GC/C/IRMS (Macko, 1994).
 ORGANIC MATTER IN FOSSILS 457

In addition to the molecular approaches in the use of stable Implications for DNA preservation. Geochim. Cosmochim. Acta,
isotopic compositions, isotope analysis of bulk organic materi- 58, 3131-5.
Engel, M.H. and Macko, S.A. (1986) Application of stable isotopes
als isolated from fossils as a high molecular weight material for evaluating the origins of amino acids in fossils. Nature, 323,
(HMW) has shown promise to reflect the total organic matter 531-3.
of the once-living organism (Robbins et al., 1991; Muyzer Engel, M.H., Goodfriend, G.A., Qian, Y. and Macko, S.A. (1994)
et al., 1992; Robbins and Ostrom, 1995). Analyses of the bulk Indigeneity of organic matter in fossils: A test using stable isotope
carbon and nitrogen isotopic compositions on this material analysis of amino acid enantiomers in Quaternary mollusk shells.
Proc. Nat! Acad., 91, 10475-8.
may thus be utilized in the assessment of the trophic position Golen berg, E.M. ( 1991) Amplification and analysis of Miocene plant
of an organism within a fossil food web (Macko and Engel, fossil DNA. Phil. Soc. Trans. R. Soc. Lond. B, 333,419-27.
1991; Ostrom et al., 1993). This concept is related to the Hagelberg, E., Bell, L.S., Allen, T., Boyde, A., Jones, S.J. and Clegg,
observation that living organisms have isotopic compositions J.B. ( 1991) Analysis of ancient bone DNA: techniques and applica-
which are reflective of their diets. The isotopic composition of tions. Phil. Soc. Trans. R. Soc. Lond. B, 333, 399-407.
Harrigan-Ostrom, P., Macko, S.A., Engel, M.H., Silfer, J.A. and
organic carbon in an organism is typically within one per mil Russell, D. ( 1990) Geochemical characterization of high molecular
of its diet, whereas nitrogen isotopic compositions have been weight material isolated from Late Cretaceous fossils. Org.
observed to increase on average by approximately three per Geochem., 16, 1139-44.
mil relative to the diet. Isotopic compositions of this HMW Lowenstein, J.M. (1991) Immunospecificity of fossil proteins: implica-
component in fossils have, for example, been used to address tions for the establishment of evolutionary trends, in Organic
Geochemistry (eds M.H. Engel and S.A. Macko). New York:
trophic structure of well-preserved Cretaceous communities Plenum, pp. 816-30.
(Harrigan-Ostrom et al., 1990; Macko and Engel, 1991). Macko, S.A. (1994) Compound specific approaches using stable iso-
There is also a wealth of organic compounds in recent and topes, in Stable Isotopes in Ecology (eds K. Lajtha and
ancient sediments as well as petroleum, which are called bio- R. Michener). Oxford: Blackwell Scientific, pp. 241-7.
markers (q. v.) or chemical fossils. These chemical fossils are Macko, S.A. and Engel, M.H. (1991) Assessment of indigeneity in
fossil organic matter: amino acids and stable isotopes. Phil. Soc.
distinguished from organic matter found in fossils in that the Ti·ans. R. Soc. Lond. B, 333, 367-74.
biomarkers are not contained within a discrete fossil matrix. Mitterer, R.M. (1993) The diagenesis of amino acids and proteins in
Whereas these compounds are clearly of biological origin, they fossil shells, in Organic Geochemistry (eds M.H. Engel and S.A.
are usually not specific to individual species. It is believed that Macko). New York: Plenum, pp. 739-54.
these components can, however, be directly related to an organ- Muyzer, G., Sandberg, P., Knapen, M.H.J., Vermeer, C., Collins, M.
and Westbroek, P. ( 1992) Preservation of the bone protein osteo-
ism input, but at present it usually cannot be directly ascer- calcin in dinosaurs. Geology, 20, 871-4.
tained which organism deposited the material in a particular Ostrom, P.H., Macko, S.A., Engel, M.H. and Russell, D. (1993) An
sediment. assessment of trophic structure in fossil communities based on
Authentic fossil organic matter in amber has an excellent stable isotopes. Geology, 21,491-4.
environment for preservation of both proteinaceous and Paabo, S. and Wilson, A. C. ( 1991) Miocene DNA sequences- a dream
come true? Curr. Biol., 1, 45-6.
genetic materials (Bada et al., 1994). Through recent develop- Paabo, S., Irwin, D.M. and Wilson, A.C. (1990) DNA damage pro-
ments in molecular biology, it is now possible to amplify DNA motes jumping between templates during enzymatic amplification.
segments remaining in fossils using the polymerase chain reac- J. Biol. Chem., 265, 4718-21.
tion (PCR). Through the use of the PCR technique, archeolog- Poinar, G.O., Poinar, H.N. and Cano, R.J. (1994) DNA from amber
ical and museum specimens as old as 130 million years have inclusions, in Ancient DNA (eds B. Herrmann and S. Hummel).
New York: Springer-Verlag, pp. 92-103.
been studied, with significant information being derived with Poinar, H.N., Poinar, G.O. and Cano, R.J. (1993) Oldest DNA from
respect to the molecular phylogeny of organisms (Paabo et al., plants. Nature, 363, 677.
1990; Golenberg, 1991; Hagelberg et al., 1991; Paabo and Robbins, L.L. and Ostrom, P.H. (1995) Molecular isotopic and bio-
Wilson, 1991; Sidow et al., 1991; Poinar et al., 1993, 1994). chemical evidence of the origin and diagenesis of shell organic mate-
A number of current controversies and inquiries about rial. Geology, 23, 345-8.
Robbins, L.L., Muyzer, G. and Brew, K. ( 1991) Macromolecules from
organic matter in fossils are the following: living and fossil biominerals: Implications for the establishment of
molecular phylogenies, in Organic Geochemistry (eds M.H. Engel
1. Is the organic matter found in a fossil indigenous to the and S.A. Macko). New York: Plenum, pp. 799-816.
fossil, and if not, what proportion of the organic matter is Serban, A., Engel, M.H. and Macko, S.A. (1989) The distribution,
indigenous? stereochemistry and stable isotopic composition of amino acid con-
2. To what extent does the organic matter in the fossil reflect stituents of fossil and modern mollusk shells. Org. Geochem., 13,
the original organism; how has it been diagenetically 1123-9.
Sidow, A., Wilson, A. C. and Paabo, S. (1991) Bacterial DNA in
altered? Clarkia fossils. Phil. Soc. Ti·ans. R. Soc. Land. B, 333, 429-33.
3. Are the nucleic materials derived from fossils and the inclu- Wehmiller, J.F. (1993) Applications of organic geochemistry for
sions in amber reflective of the organism, or are they con- quaternary research: aminostratigraphy and aminochronology, in
taminants from modern or ancient sources? Organic Geochemistly (eds M.H. Engel and S.A. Macko). New
York: Plenum, pp. 755-84.
Weiner, S. and Lowenstam, H.A. (1980) Well preserved fossil mollusk
Stephen A. Macko and Michael H. Engel shells: characterization of mild diagenetic processes, in
Biogeochemistry of Amino Acids (eds P.E. Hare et al.). New York:
John Wiley, pp. 95-114.
Bibliography Weiner, S., Lowenstam, H.A., Taborek, B. and Hood, L. (1979) Fossil
mollusc shell organic matrix components preserved for 80 million
Abelson, P.H. ( 1954) Organic constituents of fossils. Carnegie Inst. years. Paleobiology, 5, 144-50.
Washington Yearbook, 53,97-101.
Ambler, R.P. and Daniel, M. (1991) Protein and molecular palaeon-
tology. Phil. Soc. Trans. R. Soc. Land. B, 333, 381-9. Cross-references
Bada, J.L., Wang, X.S., Poinar, H.N., Paabo, S. and Poinar, G.O. Biogeochemistry
( 1994) Amino acid racemization in amber-entombed insects: Biomarker: higher plant
458 ORGANICS: CONTEMPORARY DEGRADATION AND PRESERVATION

Coal: organic petrography commonly observed in meteoritic organic matter may also
Coal: origin and diagenesis have resulted from mass fractionation during chemical reac-
Organic geochemistry
Organics: contemporary degradation and preservation
tions at low temperatures, though this is less certain.
Organics: sources and depositional environments It remains unclear what the relative importance of interstel-
Peat lar, nebular and meteorite parent body chemical processes was
Precambrian organic matter in determining the presently observed complement of organic
matter in meteorites.

Scott Messenger
ORGANIC MATTER IN METEORITES
Bibliography
Organic matter is present in most primitive meteorites, the Cronin, J.R., Pizzarello, S. and Cruikshank, D.P. (1988) Organic
highest concentrations occurring among carbonaceous chon- matter in carbonaceous chondrites, planetary satellites, asteroids
drites. Most organic matter in meteorites is in the form of an and comets, in Meteorites and the Early Solar System (eds J.F.
acid-insoluble macromolecular material similar in structure to Kerridge and M.S. Matthews). Tucson: University of Arizona
terrestrial kerogen, primarily composed of aromatic moieties Press, pp. 819-57.
linked together by ethers and aliphatic chains. Attempts to Cronin, J.R., Pizzarello, S., Epstein, S. and Krishnamurthy, R.V.
( 1993) Molecular and isotopic analyses of the hydroxy acids, dicar-
make more detailed structural characterization, or identify boxylic acids, and hydroxydicarboxylic acids of the Murchison
subcomponents of meteoritic kerogen, have had limited suc- meteorite. Geochim. Cosmochim. Acta, 57, 4745-52.
cess, though an empirical formula has been determined as Kerridge, J.F., Chang, S. and Shipp, R. (1987) Isotopic characteriza-
C 100 H 48 N1. 8 0 12 S2 . Roughly 10% of meteoritic organic matter tion of kerogen-like material in the Murchison carbonaceous chon-
occurs as solvent-extractable compounds. Although early drite. Geochim. Cosmochim. Acta, 51, 2527-40.
analyses of organic extracts were plagued by obvious terrestrial Krishnamurthy, R.V., Epstein, S., Cronin, J.R., Pizzarello, S. and
Yuen, G.U. (1992) Isotopic and molecular analyses of hydro-
contamination, improvements in sample handling and analytic carbons and monocarboxylic acids of the Murchison meteorite.
techniques have largely overcome these difficulties. Some Geochim. Cosmochim. Acta, 56, 4045-8.
organic compounds are proven to be indigenous to their parent Pizzarello, S., Krishnamurthy, R.V., Epstein, S. and Cronin, J.R.
meteorites because they exhibit a non-terrestrial H, N or C (1991) Isotopic analyses of amino acids from the Murchison mete-
isotopic composition. Biological contamination is ruled out in orite. Geochim. Cosmochim. Acta, 55, 905-10.
Zinner, E.K. ( 1988) Interstellar cloud material in meteorites, in
many cases because certain chiral molecules are shown to be Meteorites and the Early Solar System (eds J.F. Kerridge and M.S.
racemic. Some organic compounds, including several amino Matthews). Tucson: University of Arizona Press, pp. 956-83.
acids found in meteorites, are not known to occur naturally
on Earth, making terrestrial contamination unlikely.
Cross-references
The most thorough organic analyses have been performed
on the Murchison CM2 meteorite, because of the relatively Biopolymers and macromolecules
Meteorites
high abundance of organic matter and availability of clean, Organic geochemistry
large samples. Well over 400 individual compounds have been Organic matter in fossils
identified in Murchison, with a total concentration of roughly Precambrian organic matter
500 ppm. Compounds identified in Murchison to date include
aliphatic, aromatic, and polar hydrocarbons, purines and
pyrimidines, amines, amides, alcohols, aldehydes and ketones, ORGANICS: CONTEMPORARY DEGRADATION
carboxylic acids, dicarboxylic acids, hydroxy acids and
amino acids. AND PRESERVATION
Structural and isotopic analysis of the extractable organic
compounds has been used to identify the mechanisms involved Major changes occur in the composition of organic matter in
in their formation. With few exceptions, all classes of meteoritic the upper layers of aquatic sediments during early diagenesis.
organic compounds exhibit complete structural diversity, A large fraction of the organic matter which is deposited at
where all stable isomeric forms occur for each carbon number. the water-sediment interface is degraded in the upper few
Concentrations within a homologous series generally decline millimeters or centimeters of sediment, either by complete
with increasing carbon number. These observations suggest decomposition back to inorganic nutrients (remineralization)
that the organic compounds were generally synthesized by the or by partial decomposition to other organic transformation
random addition of single-carbon precursors. The lack of struc- products (Henrichs, 1993 ). The transformation products may
tural selectivity indicates that catalytic reactions, for instance be smaller in size (molecular weight) than the precursor mole-
a Fischer-Tropsch type process, are not important, and is cules if chemical bonds in the starting materials are broken or
strong evidence that the organic compounds have an abio- larger in size if bonds are formed between precursors (or
genic origin. intermediate degradation products) and other organic com-
Large enrichments of D have been observed in meteoritic pounds or macromolecules. Regardless of the degradation
kerogen and all soluble organic phases analyzed to date, with pathway and end-products, the degradation of organic matter
D/H ratios reaching four times the terrestrial value in some plays a major role in determining the chemical composition of
amino acids. The elevated D/H ratios very likely resulted from the sediments and the organic material preserved in the sedi-
low temperature chemical fractionation reactions. It is prob- mentary record. The amount and chemical nature of organic
able that the observed organic phases were formed in part matter preserved in the sediments (along with carbonates and
from interstellar precursor molecules bearing much higher D/H sulfur) is closely linked to the composition of the atmosphere
ratios, formed at temperatures of 10-100 K. Excesses in 15 N over geologic time (Berner, 1989).
 ORGANICS: SOURCES AND DEPOSITIONAL ENVIRONMENTS 459

The degradation of organic matter supplies energy to sedi- may be dominated by planktonic material. Proteinaceous mate-
mentary organisms. In many cases, organic matter degradation rials associated with carbonate shells and resistant waxy coat-
produces specific organic by-products which are required for ings or physical associations with mineral matrices may inhibit
the survival of the organisms. Bacteria are the major organic enzymatic attack on organic compounds (Mayer, 1994 ).
matter degrading organisms in sediments, although benthic Organic compounds may react in sediments and be incorported
fauna are also active in organic matter cycling, both as direct into macromolecular material (Tegelaar et al., 1989), transfer-
consumers of organic materials and in stimulating microbially ring the compounds from the labile to the refractory pool.
mediated decomposition reactions. Decomposition of sedi- Thus, degradation may not always reduce the concentrations
mentary organic matter occurs through both aerobic and of labile organic compounds to zero. In fact organic com-
anaerobic pathways. Aerobic decomposition occurs when pounds with identical structures are often found together in
oxygen is used as an electron acceptor in an oxidation-reduc- both labile and refractory forms, and molecular skeletons of
tion (redox) reaction during which organic material is oxidized. highly reactive compounds are sometimes preserved in ancient
Anaerobic decomposition occurs when a sequence of other sediments and crude oils.
electron acceptors including sulfate, nitrate, iron, manganese, Several additional factors are involved in controlling the
and small organic compounds are used to oxidize organic preservation of organic matter in sediments. Two major con-
matter (Deming and Baross, 1993). The order in which troversial processes are the organic matter production in sur-
electron acceptors are utilized follows the free energy released face waters and subsequent delivery rate to the sediments, and
by the reaction. Oxygen consumption releases the most energy the role of bottom water oxygen level and the comparative
and occurs first, while methanogenesis, the fermentation of rates of aerobic vs anaerobic decomposition (Henrichs, 1993).
small organic molecules to form methane, releases the least An early explanation for the preservation of organic-rich sedi-
energy and occurs last in the sequence. Aerobic processes ments was that they were deposited under anoxic conditions
dominate the decomposition of organic matter in aquatic sedi- and that oxygen is a critical factor in anaerobic decomposition
ments on a global scale. However, even though anaerobic rates being slower than aerobic decomposition rates. Other
decomposition accounts for only about 10% of organic matter studies have concluded that high organic matter input rates
decomposition, anaerobic processes are important in special- ('organic carbon rain rates'), rather than an anoxic deposi-
ized environments, for example coastal sediments and anoxic tional environment, are responsible for enhanced organic
basins, where the amount of organic matter available to be carbon preservation in rapidly accumulating sediments. A
degraded overwhelms the supply of oxygen. number of studies have supported this conclusion by indicating
The redox geochemistry of the sediment, in particular the that anaerobic and aerobic decomposition rates may not be
availability of electron acceptors, determines specific decompo- all that different (Lee, 1994 ). Factors such as bioturbation and
sition reaction(s) which can occur and the amount of organic organic matter consumption by meiofauna may enhance the
matter to be degraded. The chemical character of the organic decomposition of organic matter in oxic environments.
matter will determine its reactivity or susceptibility to enzy- Resolution of this debate remains a major future goal of
matic attack by bacteria. In the simplest case, sedimentary biogeochemists.
organic matter may be viewed as consisting of two pools of
organic matter: a reactive pool, which is readily degraded, and Stuart G. Wakeham
a refractory pool, which survives degradation and is preserved
in the sediments (Westrich and Berner, 1984). In reality a Bibliography
continuum of reactivities exists, ranging from readily degraded Berner, R.A. (1989) Biogeochemical cycles of carbon and sulfur and
small organic molecules such as amino acids, sugars, and their effect on atmospheric oxygen over Phanerozoic time.
simple carboxylic acids to recalcitrant biopolymers and geo- Paleogeogr., Paleoclimatol, Paleoecol. (Global Planet. Change
macromolecules and numerous synthetic organic compounds Sect.), 75,97-122.
Deming, J.W. and Baross, J.A. (1993) The early diagenesis of organic
(xenobiotics) (Hedges and Prahl, 1993). Selective degradation matter: bacterial activity, in Organic Geochemistry (eds M.H. Engel
of different types of organic compounds, based on molecular and S.A. Macko). New York: Plenum Press, pp. 119-45.
structure, has been shown in a variety of environments Hedges, J.I. and Prahl, F.G. (1993) Early diagenesis: consequences for
(Wakeham and Lee, 1993). However even the most resistant applications of molecular biomarkers, in Organic Geochemistry (eds
compounds such as lignocelluloses appear susceptible to bacte- M.H. Engel and S.A. Macko). New York: Plenum Press,
rial degradation, albeit slow, in aquatic environments. pp. 237-53.
Henrichs, S.M. ( 1993) Early diagenesis of organic matter: the
The susceptibility of organic matter to alteration and decom- dynamics (rates) of cycling of organic compounds, in Organic
position during early diagenesis, and consequently the potential Geochemistry (eds M.H. Engel and S.A. Macko). New York:
for preservation of organic matter in the sediment record, Plenum Press, pp. 101-17.
depends on a number of other factors in addition to molecular Lee, C. (1992) Controls on organic carbon preservation: the use of
structure. Organic matter derived from plankton is more read- stratified water bodies to compare intrinsic rates of decomposition
ily degraded than organic material of terrigenous higher plants. in oxic and anoxic systems. Geochim. Cosmochim. Acta, 56,
3323-35.
The relative stability of higher plant-derived organic com- Mayer, L.M. ( 1994) Surface area control of organic carbon accumula-
pounds may result from the presence of biochemically resistant tion in continental shelf sediments. Geochim. Cosmochim. Acta,
compounds such as lignin, waxes and resins. Alternatively, 58, 1271-4.
compounds of higher plant derivation are encased in a protec- Tegelaar, E.W., Derenne, S., Largeau, C. and de Leeuw, J.W. (1989)
tive particle matrix, for example in pollen grains or inside A reappraisal of kerogen formation. Geochim. Cosmochim. Acta,
53, 3103-7.
epicuticular wax coatings. Therefore, aquatic sediments tend Wakeham, S.G. and Lee, C. (1993) Production, transport, and alter-
to be enriched in and to selectively preserve organic compounds ation of particulate organic matter in the marine water column, in
of terrigenous higher plant origin, even when the particulate Organic Geochemistry (eds M.H. Engel and S.A. Macko). New
matter in the water column and raining down to the sediments York: Plenum Press, pp. 145-69.
460 ORGANICS: SOURCES AND DEPOSITIONAL ENVIRONMENTS

Westrich, J.T. and Berner, R.A. (1984) The role of sedimentary organic fluvial or aeolian transport is allochthonous. Organic com-
matter in bacterial sulfate reduction: the G model tested. Limnol. pounds derived from marine phytoplankton, zooplankton and
Oceanogr., 29, 236-49.
bacteria are autochthonous, as are alteration products pro-
duced within the water column and sediment by heterotrophic
Cross-references or chemical alteration of organic matter.
Biopolymers and macromolecules Within the pool of organic matter are the major biochemi-
Earth's atmosphere cals: proteins, carbohydrates, and lipids. Proteins are highly
Geochemistry of sediments ordered polymers composed of individual amino acids, thereby
Organic geochemistry accounting for most of the organic nitrogen in organisms.
Organic matter in fossils
Organics: sources and depositional environments Proteins act as structural components and as enzymes which
Paleoenvironments catalyze biochemical reactions. Carbohydrates collectively
Precambrian atmosphere include the simple sugars (monosaccharides) and their poly-
Sedimentary fluids mers (oligo- and polysaccharides) and serve energy storage
and structural functions. The term lipids (or fats) is applied to
a wide variety of organic compounds which are soluble in
ORGANICS: SOURCES AND DEPOSITIONAL organic solvents and generally insoluble in water. They include
a wide range of molecular structures and organic functional
ENVIRONMENTS groups, notably hydrocarbons, alcohols, aldehydes, ketones,
acids and numerous forms of simple and complex ethers and
Organic matter in the geosphere comprises organic molecules esters. Because of their range of structures, lipids serve many
in monomeric or polymeric (although macromolecular is cur- roles in energy storage, as structural components, and in regu-
rently the preferred term) forms derived directly or indirectly lating metabolic functions. Lipids are particularly useful in
from the living organisms. The primary source of organic organic geochemistry since many of their molecular structures
matter in terrestrial and aquatic environments is photosynthe- can be used as biomarkers for organic matter sources. Certain
sis, during which inorganic carbon is fixed, in the presence of steroidal alcohols (sterols) and long-chain unsaturated alkyl
light and inorganic nutrients, by autotrophic organisms to ketones (alkenones) are valuable biomarkers as they are unique
produce organic compounds, notably simple sugars. Land to limited types of phytoplankton, while carboxylic acids (fatty
plants and aquatic algae are the primary producers. acids) are less useful because they are more ubiquitous among
Heterotrophs, or secondary producers, derive their energy by living organisms. The major biochemicals (proteins, carbohy-
oxidizing existing organic matter, using the dissolved and par- drates and lipids) account for varying amounts of cellular
ticulate products of photosynthesis. Bacteria utilize dissolved organic carbon depending on species, growth conditions, and
organic matter excreted by plankton or released by cell rupture, the health of the cells, but in general, amino acids account for
while the protozoa and zooplankton consume living and non- approximately 35-50 wt% of cellular carbon, carbohydrates
living (detrital) particles. In the ocean, for example, 90% or for about 5-10 wt% and lipids for 10-30 wt%. Most often,
more of the primary organic matter produced in the sunlit organisms biosynthesize complex yet highly organized bio-
surface waters (the euphotic zone) is consumed by bacterial macromolecules consisting of combinations of amino acids,
and zooplanktonic heterotrophs, remineralizing some of the carbohydrates and lipids (de Leeuw and Largeau, 1993).
organic matter back to regenerated inorganic nutrients, con- The composition of organic matter may change significantly
verting some to living biomass, and altering some to partially between the time and place of its initial biosynthesis by organ-
degraded particulate detritus (Lee and Wakeham, 1988). isms and final deposition in the sediment (Wakeham and Lee,
A few percent of the primary production sinks out of the 1993). These changes may not be so dramatic for higher plant
euphotic, epipelagic zone as particulate matter and is continu- organic matter deposited in soils, but they are considerable in
ously altered by heterotrophs as it sinks through the meso- aquatic environments. Most of the alteration and degradation
pelagic and bathypelagic zones of the water column towards is mediated by heterotrophic animals and bacteria. As particu-
the sediment. The continuous degradation of particulate late organic material (POM) sinks though the water column,
organic matter (POM) in the water column consumes most regardless of whether it is lacustrine or marine, heterotrophic
POM and only a few percent of the primary production reaches activity preferentially removes the more labile organic com-
the sediment surface. Several factors, irJCluding the primary pounds, leaving behind particles enriched in the more refrac-
production rate, the length of the water column and the geo- tory compounds and biochemical products of heterotrophic
chemistry of the water column as discussed below, affect the metabolism. For example, in the ocean sinking POM becomes
quantity of organic matter 'raining' onto the sediments. In increasingly depleted in highly labile polyunsaturated fatty
shallow and highly productive coastal marine environments acids biosynthesized by the plankton and enriched in the more
and lakes, the delivery of organic matter to the sediment may stable sterols, long-chain alkanes, and long-chain alkenones or
be enhanced (Meyers and Ishiwatari, 1993) compared to open in compounds produced by the heterotrophs.
ocean settings. In general, chemical reactions progress more slowly than
Sources of organic matter in sediments may be further classi- biologically mediated (metabolic) reactions, and in a few
fied by their origin relative to a depositional environment. The instances, the chemical and biological pathways may produce
organic matter may be autochthonous to the environment of different products from the same precursor. The chemical
deposition if it derives from the overlying water column or degradation of carotenoid pigments biosynthesized by algae,
within the sediment itself, or it may be allochthonous if it is for instance, has been shown to follow a pathway different
foreign to and has been transported to the depositional envi- from the biochemical degradation. The initial stages of degra-
ronment (Tissot and Welte, 1984 ). Thus organic matter derived dation produces different intermediate degradation products,
from terrigenous plants and delivered to marine sediments via but the final product is the same for both pathways (Repeta
 OSMIUM 461

and Frew, 1988). Nonetheless, as particles sink the fraction Recent and Ancient Sediments (eds J. Whelan and J.W. Farrington).
which can be identified as the major biochemicals (proteins, New York: Columbia University Press, pp. 231-63.
de Leeuw, J.W. and Largeau, C. (1993) A review of macromolecular
carbohydrates, lipids) decreases while the remains of refractory organic compounds that comprise living organisms and their role
biomacromolecules biosynthesized by plants become relatively in kerogen, coal, and petroleum formation, in Organic
more abundant on their way to becoming the poorly chemically Geochemistry (eds M.H. Engel and S.A. Macko). New York:
characterized humic substances and kerogen which comprise Plenum Press, pp. 23-72.
most of sedimentary organic matter. Henrichs, S.M. (1993) Early diagenesis of organic matter: the
The geochemistry of the water column will also influence dynamics (rates) of cycling of organic compounds, in Organic
Geochemistry (eds M.H. Engel and S.A. Macko). New York:
the composition of organic matter reaching the sediment. Plenum Press, pp. 101-17.
Aerobic metabolism in oxygenated (oxic) water columns tends Lee, C. and Wakeham, S.G. (1988) Organic matter in seawater:
to degrade more POM than anaerobic metabolism in oxygen- Biogeochemical processes, in Chemical Oceanography, Vol. 9 (ed.
deficient (suboxic and anoxic) water columns, so the flux of J.P. Riley). New York: Academic Press, pp. 1-57.
organic matter to the sediment may be enhanced in an anoxic Meyers, P.A. and Ishiwatari, R. (1993) The early diagenesis of organic
matter in lacustrine sediments, in Organic Geochemistry (eds M.H.
water column. Particles sampled in suboxic and anoxic water Engel and S.A. Macko). New York: Plenum Press, pp. 185-209.
columns have been found to contain significant quantities of Repeta, D.J. and Frew, N.M. ( 1988) Carotenoid dehydrates in recent
organic compounds produced specifically by anaerobic bacte- marine sediments. The structure and synthesis offucoxanthin dehy-
ria, and these compounds may be deposited at the water- drate. Org. Geochem., 12,469-77.
sediment interface along with compounds of planktonic origin Tissot, B.P and Welte, D.H. (1984) Petroleum Formation and
which have survived transit through the water column. Occurrence. Berlin: Springer-Verlag, 699 pp.
Wakeham, S.G. and Lee, C. (1993) Production, transport, and alter-
Both the quantity and composition of POM raining down ation of particulate organic matter in the marine water column, in
to the sea floor will affect the nature of organic matter which Organic Geochemistry (eds M.H. Engel and S.A. Macko). New
is deposited in the sediment. The nature of metabolic processes York: Plenum Press, pp. 145-69.
in the upper sediment column will determine how much organic
matter and which organic compounds will survive diagenesis Cross-references
to become buried and part of the sedimentary record. In depo-
Biomarker: aliphatic
sitional environments characterized by low primary produc- Biomarker: aromatic
tivity in surface waters (e.g. oligotrophic environments) and Biomarker: assessment of thermal maturity
long, oxic water columns (such as the deep ocean), the flux of Biomarker: lipid
organic matter reaching the sediment is only a very small Biopolymers and macromolecules
fraction of that biosynthesized in surface waters (typically only Oil-oil and oil-source correlation
Organic geochemistry
a few percent), and only the most refractory organic molecules Organic matter in fossils
survive. The result usually is an organic-poor sediment, often Organics: contemporary degradation and preservation
containing < 1 wt% organic carbon. Aerobic metabolism of Paleoenvironments
organic matter in surface sediments by bottom-dwelling ben- Precambrian organic matter
thic organisms, both animals and microbes, will further reduce Sedimentary fluids
the amount of organic matter which will be buried.
Organic-rich sediments containing up to 20 wt% organic
carbon are typically deposited under conditions of high surface OSMIUM
water productivity (eutrophic environments) and/or short and
oxygen-deficient water columns. Lakes and coastal marine History and properties
areas and marine basins with restricted circulation and stag-
nant, anoxic bottom waters, such as the Black Sea, tend to In 1804, Smithson Tennant discovered element number 76,
accumulate relatively organic-rich sediments. In general, there which he named osmium (Os). Osmium is one of the six
is a correspondence between the amount of organic matter platinum group elements (PGE ). The bluish-white metal (hex-
which is preserved in surface sediments and the organic carbon agonal close-packed) has a melting point of 3318 K, a boiling
flux to the water-sediment interface. Although the remineral- point of 5300 K and a density of 22 570 kg/m 3 , vying with
ization rate and hence the amount of organic matter degraded iridium to be the densest metal. The atomic mass is 190.2 and
is greater in rapidly accumulating sediments, the remineraliza- there are seven stable isotopes with the following formal abun-
tion efficiency is lower than in slowly deposited sediments dances (in atom%); 184 (0.02), 186 (1.58), 187 (1.6), 188
(Henrichs, 1993). The importance of high organic matter input (13.3), 189 (16.1), 190 (26.4) and 192 (41.0). Two of these
rates in the preservation of sedimentary organic matter is isotopes have variable abundances in nature due to radioactive
widely accepted, but the role of bottom-water oxygen content decay of other elements: 187Re decays to 1870s by {r emission
(and whether the degradation in surface sediments is aerobic (2.3 keV maximum {3- energy, half-life 4.16 x 10 10 years), thus
or anaerobic) is a current subject of debate (Calvert and providing the basis for the Re-Os method of geochronology;
Pederson, 1992). and 190Pt forms 1860s by rx decay (rx energy 3.19 MeV, half-life
4.50 x 10 11 years), but effects are measurable only in Pt-rich
materials because of the very low isotopic abundance of 190Pt
Stuart G. Wakeham (0.012%). Osmium has oxidation states ranging from 0 to +8,
but in nature 0, + 3 and + 4 probably predominate.
Bibliography
Cosmochemistry
Calvert, S.E. and Pedersen, T.F. (1992) Organic carbon accumulation
and preservation in marine sediments: How important is anoxia? In The solar abundance is 0.675 atoms Os per 106 atoms Si,
Organic Matter: Productivity, Accumulation, and Preservation in corresponding to 486 ppb Os in CI chondrites. In planetary
462 OXIDATION-REDUCTION

evolution, Os, like other PGE, is strongly siderophile (i.e. iron- Yukon territory, Canada, and Hunan and Guizhou provinces,
loving) and is likely to be quantitatively extracted into the China. Geochim. Cosmochim. Acta, 58, 257-65.
Morgan, J.W. ( 1986) Ultramafic xenoliths: clues to the Earth's late
metallic core. This is illustrated by high abundances in iron accretionary history. J. Geophys. Res., 91, 12375-87.
meteorites (up to 86 ppm; Pernicka and Wasson, 1987) and Pernicka, E. and Wasson, J.T. (1987) Ru, Re, Os, Pt and Au in iron
very low abundances in some unbrecciated achondrites meteorites. Geochim. Cosmochim. Acta, 51, 1717-26.
( ~0.003 ppb ). In the early Earth, core formation similarly Walker, R.J., Morgan, J.W., Horan, M.F. eta/. ( 1994). Re-Os isotopic
must have depleted the silicate mantle and the present mantle evidence for an enriched-mantle source for the Noril'sk-type,
abundance of Os seems the result of a late influx of meteoritic ore-bearing intrusions, Siberia. Geochim. Cosmochim. Acta, 58,
4179-97.
material (Morgan, 1986).
Cross-references
Igneous and metamorphic rocks
Crystal field theory
Undepleted regions of the Earth's upper mantle contain Dating methods
3.4 ppb Os judging from analyses of spinellherzolites. During Elements: platinum group
basalt formation by partial melting, Os behaves as a compatible Elements: siderophile
element and remains predominantly in the residual peridotite. Elements: transitional
As a result, the basaltic fraction is depleted in Os; both mid- Rhenium-osmium dating method
ocean ridge basalts and some continental basalts may contain
as little as 0.001 ppb Os. Ocean island basalts related to deep
mantle plumes are richer in Os (typically 0.1-0.5 ppb) and OXIDATION-REDUCTION
conceivably may contain siderophile elements derived by leak-
age across the core-mantle boundary. In other upper crustal Oxidation and reduction are chemical terms referring to the
rocks, the following abundances have been reported (in ppb loss or gain, respectively, of one or more valence electrons of
Os): komatiites, 1-5; diabase, 0.3; granite 0.005-0.01. Esser an element in the course of a chemical reaction. In geochemis-
and Turekian (1993) estimate the average Os content of the try, the traditional usage referred to reactions involving gain
upper crust to be 0.05 ppb. In lower crustal rocks we find or loss of oxygen. Historically, studies of oxidation-reduction
0.1-0.2 ppb Os in granulite and 0.6 ppb Os in hornblende reactions played a large part in the development of chemistry.
eclogite. Alchemists searching for the secret for transmutation of ele-
ments oxidized and reduced mercury and pondered the mean-
Sedimentary rocks ing of the changes in physical properties exhibited by the
Because Os 4 + is probably hydrolyzed and scavenged from various oxides and sulfides, and the miraculously liquid quick-
seawater as Os0 2 • xH 2 0, the residence time of Os in the ocean silver in mercury's elemental state. The smelting of ores and
is very short ( « 10 3 years) and the abundance very low the winning and working of metals are all largely oxidation-
( ~ 3.6 pg/1 ). In oxidized sediments, cosmic influx may have reduction phenomena that have been utilized for thousands of
slightly enhanced the observed abundances (ppb Os); shale, years. Many of the obvious colors of rocks and soils in nature
0.05-0.45; quartz-rich sandstone, 0.06-0.08. In a euxinic envi- are caused by the colorful and ubiquitous oxides of iron.
ronment, Os is more enriched in sedimentary rocks (ppb Os); The importance of redox chemistry to life on Earth cannot
0.2-0.7 in Black Sea sediments, 0.9-4 in black shales and up be overstated. Through photosynthesis, nitrogen fixation, res-
to 210 in mineralized black shales (Horan et a!., 1994). In piration, and decomposition, the chemistry of the natural envi-
older black shales, much of the Os is radiogenic 187 0s from ronment is defined by redox reactions involving oxygen,
the decay of the high levels of 187 Re in these rocks. carbon, nitrogen, sulfur, iron, and manganese, among others.
Electrons are the currency of energy exchange in all metabolic
reactions, and such metabolic reactions ultimately dictate the
Economic geology
redox conditions in natural aqueous environments. These
Osmium does not form its own minerals but occurs alloyed redox conditions can in turn, strongly affect the fate and trans-
with Ir as osmiridium (cubic, o(32% Os) and iridosmine (hex- port of heavy metals and organic compounds in soils and
agonal, > 32% Os). Up to 3% Os may be present in laurite surface and subsurface aqueous environments, a source of
(RuS 2 ). Magmatic Ni-Cu sulfide ores such as those from environmental concern.
Noril'sk (Russia) may contain up to 600 ppb Os (Walker et al.,
1994). This may be recovered with other PGE in an anode Valence
sludge during electrolytic Cu refining. Kilogram amounts of
isotopically almost pure 187 0s recently made available from Outer electron shells of elements are subject to adjustment
the former USSR appear to have been extracted from Re-rich according to the octet rule and the element's position in the
molybdenite. Periodic Table. Eight electrons in the outer shell is the most
stable configuration regardless of overall charge on the atom.
John W. Morgan An atom with a net charge is known as an ion. Elements with
less than four electrons in the outer shell tend to lose those
electrons and assume a net positive charge (cation). Examples
Bibliography
of elements that tend to become cations are the alkali metals
Esser, B.K. and Turekian, K.K. (1993) The osmium isotopic composi- and earths, and the transition elements (Figure 08). Elements
tion of the continental crust. Geochim. Cosmochim. Acta, 57, with more than four electrons tend to attract electrons in order
3093-104.
Horan, M.F., Morgan, J.W., Grauch, R.I., Coveney, R.M., to fulfill the octet rule, and become net negatively charged
Murowchick, J.B. and Hulbert, L.J. (1994) Rhenium and osmium (anion). Examples of elements that tend to become anions are
isotopes in black shales and Ni-Mo-PGE-rich sulfide layers, the halogens. Elements with around four valence electrons,
 OXIDATION-REDUCTION 463

<:} 'g
W'V'V \1 'V
'J'JW'J
0 0 Q W<J 'J
0 0 "J<J<J<l'J<JV 'JQ
> 0 0 'J w "J<J<J<l
'J w
<:} <:} <NJO 'JWQ 'J
.c \};]J\nV.] \7 Q
UJ 'Vr::J"V\.7\n w 'V
"0 'JW Q
e:::> <:} w 'JQWQ <:}
'J<JV 'J w
<f) WW'V'VVJ'V

"'"'I ==
QQ'J<JVQ

E
o=
=
"0 QWQW
Qj
u: <:)<:}<:}
W'J 'J
<:} <:}

-0.420l_--~L_--~L_____L_____L_____L__ _ _ _L__ _ _ _L __ _~~--~

-0.500 -0.325 -0.150 0.025 0.200 0.375 0.550 0.725 1.075

Eh computed from redox couples (V)

Symbol Redox couple

0 Fe3+ /Fe 2 +
v 02aq/H20
0 Hs- ;so~-
D HS- /S,hombic
•... NOi /N03
NH.i /N03
6 NH.i /N02
+ CH4aq/HC03
NH.i !N2aq
•
X
Fe 2 +/Fe(OH) 3 csl
M Field-measured Eh value

Figure 08 Calculated vs measured Eh data (from Lindberg and Runnells, 1984).

like carbon, nitrogen and sulfur, can gain or lose up to eight which is assigned a standard potential (Eo) of 0.00 V. Those
electrons. Valence state is indicated in the literature generally elements more reducing (negative sign) than this couple, such
by a roman numeral in parentheses, (Fe(III)) or by a super- as Zn metal (Eo= -0.76), should reduce H+ to H 2 and those
script, (Fe3+). Those elements with eight outer shell electrons more oxidizing should oxidize H 2 to H +, such as Cu metal
show little tendency to gain or lose and are referred to as the (E =0.34).
0

noble gases.
Ionization potential -measurement/calcu lation
Electromotive series
The measurement of ionization potential is typically done by
The relative ease of loss or gain of electrons is a measurable electrochemical experiment in a galvanic cell in which one
quantity called ionization potential or oxidation potential, and
electrode is the element of interest immersed in a solution of
varies from element to element. These potentials are listed here
its ions, and the other electrode is a reference electrode with a
in the classical electromotive series familiar to chemistry stu-
known value, such as calomel or AgjAgCl. This is because, in
dents (Table 02). The elements are arranged in order of
practice, the standard hydrogen electrode (SHE) is too difficult
increasing potential (voltage) required to remove one or more
to construct and use. Otherwise, Eo values at standard state
electrons out of the influence of the atom, typically in aqueous
and unit activities may be calculated from thermodynamic data
solution. The negative sign for reducing agents (give electrons
freely) and positive sign for oxidizing agents (accept electrons according to:
freely), for reactions written as reductions follows the (014)
European (IUPAC) (Bard et al., 1985) convention. The stan- where n =number of electrons transferred in the reaction, and
dard or reference potential is taken as the hydrogen couple: Fis Faradays (23.06 kcaljV-g equivalent) and L'lG~ is the stan-
2H+ + 2e- = H 2 (013) dard Gibbs free energy of reaction, calculated from tables of
464 OXIDATION-REDUCTION

Table 02 Standard oxidation-reduction potentials free energy values according to:

Standard potential, dG~eaction =I dG~roducts- I L1G~eactants (015)
Couple Eo (V) Spontaneity of reaction is indicated by a negative free energy
Li+ +e- = Li
value.
-3.045
K++e-=K -2.925 There are many published tables of free energy values for
Ba2+ + 2e- = Ba -2.90 elements (CRC Handbook, current, Bard et al., 1985), and
Sr 2 + = 2e- = Sr -2.89 compounds of geochemical interest (Garrels and Christ, 1965;
Ca 2 + + 2e- = Ca -2.87 Drever, 1988). Thermodynamic data are continually being
Na++e-=Na -2.714 refined and the literature should be consulted for most current
Mg 2 + + 2e- = Mg -2.37 values if high accuracy is critical. The sign conventions vary
AJ3+ + 3e- =AI -1.66 from source-to-source, and the reader must take care to recog-
Zr 4 + +4e- =Zr -1.53 nize which system is applicable.
Mn 2 + + 2e- = Mn -1.18
Zn 2 + +2e- =Zn -0.763
CrH +3e- =Cr -0.74 Calculation of £"
Fe 2 + +2e- =Fe -0.440 The Eo value (at standard state and unit concentrations) for
Cd 2 + +2e- =Cd -0.403 a complete oxidation-reduction (whole cell) can be calculated
Sn 2 + + 2e- = Sn -0.136 by adding couples algebraically after balancing electrons and
Pb 2 + + 2e- = Pb -0.126
2H+ +2e- =H 2
reversing the oxidizing half-cell and its sign. Spontaneity is
0.00
S(,1 + 2H+ + 2e- = H 2 S(gJ +0.141 indicated by a positive value of the potential calculated in
Sn 4 + +2e- =Sn 2 + +0.15 this way.
Cu 2 + +2e- =Cu +0.337 Zn = Zn 2 + + 2e- +0.76 V (reversed)
0 2 + 2H+ + 2e- = H 2 0 2 +0.69
FeH +e- =Fe 2 + +0.771 Cu2+ + 2e- = Cu +0.34 v
Ag+ +e- =Ag +0.799
Pd 2 + +2e- =Pd +0.987
Zn + Cu 2 + = Zn 2 + + Cu + 1.10 v (016)
AuC14 +3e- =Au+4Cl- + 1.00
Br 2 + 2e- = 2Br- + 1.065
0 2 +4H++4e-=2H 2 0
Eh: Eo at non-standard conditions (see Eh-pH relations)
+ 1.229
Cl 2 + 2e- = 2Cl- + 1.36 Eo values at non-standard concentrations may be calculated
Au3+ +3e- =Au + 1.50 using the Nernst equation (Butler, 1964; Garrels and Christ,
Mn04 + 4H+ + 4e- = Mn0 2 + 2H 2 0 + 1.70 1965). The resultant value represents the oxidation potential
H 2 0 2 + 2H+ + 2e- = 2Hp + 1.77 of the environment and is referred to as the Eh. By convention,
F2 +2e-=2F- +2.87 the Nernst equation is set up as an oxidation for calculating
Li+ +e- =Li -3.045 Eh by reversing the sign of the term containing the reaction
K++e-=K -2.925
Ba2+ + 2e- = Ba -2.90
quotient. This gives positive Eh values for oxidizing conditions
Sr 2 + = 2e- = Sr -2.89 and negative values for reducing conditions. The Eh is typically
Ca 2 + + 2e- = Ca -2.87 calculated for a single half-cell, as below, or for a whole cell:
Na+ +e- =Na -2.714 Eh =Eo+ (RT/nF) ln([ox]/[red])
Mg 2 + + 2e- = Mg -2.37
AJ3 + + 3e- = AI -1.66 =Eo+ (0.059/n) log([ox]/[red ])(log 10 ) (017)
Zr 4 + + 4e- = Zr -1.53
Mn2+ +2e- =Mn -1.18 Where ln([ox]/[red]) is the reaction quotient of a redox half-
Zn 2 + +2e- =Zn -0.763 cell written as an oxidation, n and F as above, T =degrees
Cr3+ + 3e- = Cr -0.74 Kelvin, usually 298.15 K, and 0.059 = 2.303RT/F. For a whole
Fe 2 + +2e- =Fe -0.440 cell,
Cd 2 + + 2e- = Cd -0.403
Sn 2 + + 2e- = Sn -0.136 Eh =Eo+ (RT/nF) ln{[C]c[D]d/[A]a[B]b}, (018)
Pb 2 + + 2e- = Pb -0.126 where {[C]c[D]d/[A]a[B]b} is the reaction quotient for the
2H+ +2e- =H 2 0.00 redox reaction a[A] + b[B] = c[CJ + d[DJ, where n electrons
Sc,1 + 2H+ + 2e- = H 2 S1g 1 +0.141
Sn 4 + +2e- =Sn2+
are transferred.
+0.15
Cu 2 + +2e- =Cu +0.337
0 2 + 2H+ + 2e- = H 2 0 2 +0.69 The pe
Fe3++e-=Fe 2 + +0.771 An alternative method for expressing the redox potential of
Ag+ +e- =Ag +0.799 the environment is the pe. The pe is analogous to the pH in
Pd 2 + + 2e- = Pd +0.987
AuC14 + 3e- =Au+ 4Cl-
that it is the negative logarithm of the activity of aqueous
+ 1.00
Br2 + 2e- = 2Br- + 1.065 electrons. In the older literature, oxidation-reduction poten-
2 +4H+ + 4e- = 2H 2 0 + 1.229 tials were not always calculated according to the IUPAC con-
Cl 2 + 2e- = 2CI- + 1.36 vention, and signs opposite to those in Table 02 were used by
Au3++3e-=Au + 1.50 many workers.
Mn04 + 4H+ + 4e- = Mn0 2 + 2H 2 0 + 1.70 The pe is calculated for a generalized reaction as follows:
H 2 0 2 + 2H+ + 2e- = 2H 2 0 + 1.77
F2 +2e- =2F- +2.87 oxidized species + ne- = reduced species
(019)
pe = (1/n)log Keq + (1/n) log{[ox]/[red]}
 OXIDATION-REDUCTION 465

The pe may be related to Eh by the expression: slow, and considerable disequilibrium may be present in natu-
ral systems.
Eh = 2.303RT/F = 0.059pe, units as above. (020)
One objection that can be raised to pe as a variable is that Kinetics of redox reactions
there is little evidence for the existence of aqueous electrons in
nature (Hostetler, 1984 ), while the Eh actually is a measured Rates of oxidation and reduction in nature vary widely. Redox
voltage, however difficult to interpret. Nonetheless, the use of reactions may be homogeneous or heterogeneous and often
pe is gaining favor because of its unambiguous treatment of involve oxyanions, which have rather slow rates of electron
sign conventions, and general ease of calculation. transfer. Many redox reactions are mediated by bacteria, or
induced photochemically.
Rates of oxidation of various species in contact with P0 2 =
Eh/pe-pH diagrams 0.2 atm are listed by Eary and Schramke (1990). Solution
Often, redox reactions involve changes in the pH. Combination conditions, especially pH and salinity, have a major effect.
diagrams utilizing the functional relationship between pH and Oxidation rates are slow for Fe 2 + at pH< 4, for S2 - at pH< 5,
pe or Eh may be constructed in order to show stability relations for Mn 2 + at pH< 8, and for AsH at all pH values. Half-lives
between redox species (see pH-Eh relations). These sort of for these reactions range from months to years, indicating
diagrams were first used extensively by Pourbais ( 1949) and reduced species can persist in aerated solutions for extended
in classic treatments by Baas-Becking eta!. (1960), both cited periods. By contrast, in solutions at pH values greater than
in Garrells and Christ ( 1965), for Eh-pH, and in treatments those stated for Fe 2 +, S2 -, and Mn 2 +, and Cu 2 +, for all pH
utilizing pe-pH relations by Drever ( 1988), and Morel and values, half-lives ranged from minutes to weeks (Figure 010).
Hering (1993). Figure 09 shows a pe-pH diagram for the Rates of reduction range from rapid (minutes), for photo-
system Fe-O-H 2 0-S-C0 2 at 2SOC (from Drever, 1988). The reduction of Fe( III) in acidic mine drainage (McKnight eta!.,
diagram shows the stability relations of various solids with 1988), to imperceptibly slow (thousands of years) for the
Fe(II) and Fe(III). The boundaries of natural conditions are reduction of oxyanions such as so~- in deep, essentially abi-
indicated on the diagram as the oxidation and reduction of otic ground waters associated with coal beds (Patterson, 1996).
water (upper and lower boundaries, respectively), and pH 4 to Some reductants such as ascorbate can accelerate the reduction
pH 9, which represent the range of conditions commonly of Fe-oxides to a scale of hours (Morel and Hering, 1993).
encountered in nature. These diagrams are calculated for equi- In anaerobic environments, microbes may accelerate the rate
librium conditions; rates of actual redox reactions may be quite of reduction of oxyanions by factors of several thousand as
they utilize these species as electron acceptors in their metabolic
10 12 14 processes (Morel and Hering, 1993). Steadily decreasing Eh
results as successive oxyanions are reduced in a sequence corre-
+15
Fe3+
(aq)
sponding to decreasing free-energy yield from the reduction

10 r------------------------------,

8
1y

0 2
J::
1d
w
4
1h

2 1 min

0 1 sec

0 2 4 10 12 14
pH
pH
Figure 09 Stability relationships in the system Fe-0-H 2 0-S-C0 2
at 25°(, assuming LS = 10- 6 , LC0 2 = 10° (modified from Drever, Figure 010 Half-lives for LFe2+ and LS- vs pH at p0 2 = 0.2 atm
1988). at 25°( (modified from Eary and Schramke, 1990).
466 OXIDATION-REDUCTION

-0.5 +0.5 +1.0 EH (Drever, 1988). Figure 011 shows a listing of these microbi-
-10 -5 +5 +10 +15 +20 pE
ally-mediated reactions listed in order of decreasing pe and
energy yield at pH 7. The microbial reduction of oxyanions
Reductions such as so~- often results in the production of gaseous
j by-products such as H 2 S, which may be extracted from the
I <.. Denitrification B
I water and quantified as a measure of redox conditions
Mn(IV) oxide --7 Mn( 11)
1
c (Patterson, 1996).
I i ,No; Reduction D 1

( Fe (Ill) oxide--. Fe(ll) E

ComblnaHon t!.G"
pH= 7
Oxidation-reduction in natural environments
Reduct. Org. F Examples Kcallequivalent
In some ways, Eh is a measure of the environment's ability to
< sol· reduction IG Aerobic Aespiralion
Denitrification
A+L
B+ L
-29.9
-2B.4 acquire or supply electrons. Atmospheric oxygen and organic
( CH, fermentation H D +L -19.6
carbon represent the maximum normal oxidizing and reducing
Nitrate Raduclion
Femmntation D +L - 6.4
G+l - 5_9
H2 NH4 J Suffate Reduction
-7
potentials, respectively. Lower reducing potentials can be gen-
H,, ~o;srm;;at;;,o;:n:=;-~\K
;::;;:=:
Methane Fermentation H+L - 5.6
I
N- Fixation J+l . 4.5 erated by anaerobic bacteria. Using redox potential to charac-
--T-~------;-1---'------+--J ~~~;~:~0x~dation A+M
A+O
·23.6
-10.3
terize the natural environment gives a broad ranging and useful
L I?xidat. Org Ma~ Ferrous Oxidation A+H -21.0 framework for predicting the speciation of redox-sensitive ions
M Sulfide --. SO, ·
Mn (II) OxidaHon A+P - 7.2
and solids in aqueous environments. Figure 011 shows the
approximate position of many natural environments in terms
N ;O:.cx::.::idc::.at~.o~f~F=e(~ll);;:_t_=~,J
ofEh and pH. Figure 012 shows a classification of geochemical
I
o '-'-'N"-H'-t--.TN:"O~i::;;:::;1--,=~ environments based on dissolved gases.
P I Oxidat. of Mn(ll)
a 1 N2--.NOi J Charles G. Patterson
Oxidations R I o, formation
I
-10 -5 +5 +10 +15 +20 pE Bibliography
10 Bard, A.J., Parsons, R. and Jordan, J. (1985) Standard Potentials in
Aqueous Solution. New York: Marcel Dekker, 834 pp.
Kcal/ equivalent
Butler, J.N. (1964) Ionic Equilibrium, A Mathematical Approach. New
Figure 011 Sequence of microbially mediated redox processes York: Addison Wesley, 547 pp.
(modified from Stumm and Morgan, 1981). Chemical Rubber Corporation (1995) Handbook of Chemistry and
Physics. Palo Alto: CRC Press.
Drever, J.I. (1988) The Geochemistry of Natural Waters, 2nd edn.
Englewood Cliffs, NJ: Prentice-Hall, 437 pp.
Eary, L.E. and Schramke, J.A. (1990) Rates of inorganic oxidation
reactions involving dissolved oxygen, in Chemical Modeling of
Aqueous Systems II (eds D.C. Melchior and R.L. Bassett). Am.
Chem. Soc. Symposium Series, 416, pp. 379-96.
Garrells, R.M. and Christ, C.L. (1965) Solutions, Minerals, and
Equilibria. San Francisco: Harper, 450 pp.
Hostetler, J.D. (1984) Electrode electrons, aqueous electrons, and
redox potentials in natural waters. Am. J. Sci., 284, 734-59.
Langmuir, D. (1972) Eh-pH determinations, in Proceedings in
Sedimentary Petrology (ed. R.E. Carver). New York: Wiley
Interscience, pp. 597-635.
Lindberg, R.E. and Runnells, D.D. (1984) Ground water redox reac-
tions: an analysis of equilibrium state applied to Eh measurements
and geochemical modelling. Science, 225, 925-7.
McKnight, D.M., Kimball, B.A. and Bencala, K.E. ( 1988) Iron photo-
reduction and oxidation in an acidic mountain stream. Science,
40, 637-40.
Morel, F.M.M. and Hering, J.G. (1993) Principles and Applications of
Aquatic Chemistry. New York: John Wiley & Sons, 588 pp.
Patterson, C.G. (1996) The Electrochemistry of Ground Water: Redox
Gases and Electromigration, unpublished Ph.D. Dissertation, Univ.
of Colorado, 201 pp.
Stumm, W. and Morgan, J.J. (1981) Aquatic Chemistry, 2nd Edn.
Wiley Interscience, 780 pp.

Cross-references
Acid deposition
Aqueous solutions
Calcium carbonate and the carbonic acid systems
Eh-pH relations
Fluid-rock interactions
pH Geochemistry: low temperature
Hydrothermal solutions
Figure 012 Approximate position of some natural environments as Petroleum: in-reservoir biodegradation
characterized by Eh and pH (modified from Garrells and Christ, Sedimentary fluids
1965). Sulfate reduction
 OXYGEN 467

20.946 vol% of the bulk atmosphere at sea level, with a uniform

OXYGEN concentration to an altitude of 100 km (Emiliani, 1992). It is
formed principally as a product of photosynthesis, which began
Oxygen (chemical symbol 0) is the third most abundant ele- with the evolutionary appearance on Earth of prokaryotic
ment in the cosmos and the second most abundant element in cyanobacteria (blue-green algae) in the early Archean eon,
the Earth (Lewis, 1995; see Elemental abundances). The exis- about 3.5-3.8 Ga ago, drastically changing what would pre-
tence of a substance in the air that supported combustion and viously have been a virtually anoxic atmosphere formed of
breathing was hypothesized no later than the mid-seventeenth volcanic emissions (Lewis, 1995; see Atmosphere: Earth). Once
century (Brown et al., 1997), but oxygen was first isolated in photosynthetic production exceeded the oxidation of elements
1774 by the British chemist Joseph Priestley. Karl Wilhelm such as Fe in the crust, oxygen could accumulate in the atmo-
Scheele of Sweden is also given credit for the discovery. sphere (Schlesinger, 1991 ).
Antoine Lavoisier of France was the first to recognize that During the day, dioxygen molecules strongly absorb ultravi-
oxygen is a chemical element, and he gave it its name, which olet radiation with wavelengths less than 300 nm, which causes
means acid former. them to dissociate into monatomic oxygen by the reaction:
Oxygen is a non-metal, the lightest element in Group VIA (021)
of the Periodic Table. It has atomic number 8 and an atomic
weight of 15.9994. Its electron configuration is 1s 2 2s 2 2p 4 , its where hv, the energy required to break the bond, is 5.16 eV
first ionization potential is 13.618 eV, and its electronegativity (corresponding to a wavelength of 240 nm, in the ultraviolet;
is 3.44, second only to that of fluorine. Emiliani, 1992).
There are three stable isotopes of oxygen (see Oxygen iso- At night, ozone forms by collision of monatomic oxygen
topes; Stable isotopes). Oxygen-16 (chemical symbol 160) com- with a dioxygen molecule or, less frequently, by collision of
prises 99.759% of the natural element; oxygen-18 eso) three oxygen atoms:
0.2039%, and oxygen-17 ( 17 0) 0.0374% (Friedlander et a!., 0 (g) + 02 (g)-+ 03 (g) (022)
1964 ). Oxygen also has four radioactive isotopes. These are
oxygen-14, oxygen-15, oxygen-19, and oxygen-20, which have 30(g)-+03(g) (023)
half-lives of 72 s, 2.0 min, 29.4 s, and 13.6 s respectively (see Formation of ozone is greatest in the stratosphere, from
Radioactivity). 30 km altitude at the equator to 18 km at the poles (Emiliani,
The common naturally occurring form of the element is 1992), where ultraviolet insolation is high and the atmosphere
molecular dioxygen, 0 2 , a colorless, odorless gas at STP with is still dense enough for frequent collisions. Ozone concen-
a molar mass of 31.9988 gjmol, molar volume of 22.4141/mol, trations here in the so-called stratospheric ozone layer are
and density of 1.429 g/cm 3 . The normal melting point of dioxy- approximately 1-10 ppm, compared with 0.01 ppm in the
gen is -2l8°C, its normal boiling point is -183°C, and its troposphere (Holland, 1978). Stratospheric ozone also
critical temperature is -ll8°C. Oxygen also exists as the allo- strongly absorbs ultraviolet and photodissociates back to diox-
trope ozone, 0 3 , a pale blue, irritating gas (see Atmosphere: ygen and monatomic oxygen. This process absorbs much of
Earth). the most biologically hazardous ultraviolet radiation before it
Oxygen is extremely reactive, and forms compounds with can reach the surface. The balance between dissociation by
every element except He, Ne, and Ar (Ebbing, 1996). The day and ozone formation by night has recently been disrupted
most abundant compounds of oxygen are oxides: binary com- in the polar regions by the influx of monatomic chlorine from
pounds with oxygen in the -2 oxidation state. Transition stratospheric degradation of anthropogenic chlorofluorocar-
metals generally combine with oxygen to form basic oxides bon compounds (CFCs). Monatomic chlorine catalyzes the
such as Fe 2 0 3 . Oxides of transition metals in high oxidation destruction of ozone via the reactions:
states, such as chromium (VI) oxide, Cr0 3, may be acidic
Cl(g) + 0 3 (g)-+ ClO(g) + 0 2 (g) (024)
(Ebbing, 1996). Hydrogen and sodium form peroxides, in
which oxygen is in the -1 oxidation state, and potassium and ClO(g) + O(g)-+ Cl(g) + 0 2 (g) (025)
the heavier alkali metals form superoxides, with oxygen in the The link between CFCs in the troposphere and destruction of
-1/2 oxidation state. Nonmetal oxides are covalently-bonded stratospheric ozone won its discoverers, Mario Molina and
and acidic, and usually form simple molecules that are gases F. Sherwood Rowland, the Nobel Prize in chemistry.
at STP, such as C0 2 and S0 2 • However, the covalent metalloid Ozone is formed in the troposphere in minute quantities by
oxides B2 0 3 and Si0 2 form solid polymers. lightning, and as a hazardous component of photochemical
Apart from its most important application, which is to sus- smog in urban and industrialized areas.
tain aerobic respiration in the biosphere, oxygen is used in
gaseous and liquid form as an oxidizer in steelmaking, welding,
rocket propulsion, and chemical manufacturing. It is also com-
The biogeochemical oxygen cycle
monly used as a bleaching agent, and in gas mixtures for life Figure 013 depicts the global biogeochemical cycle of dioxy-
support. Oxygen was the third most widely used industrial gen. The cycling of oxygen in the Earth system is directly
chemical in 1995 (Brown et al., 1997). linked to the carbon and nitrogen cycles (Stumm and Morgan,
1996; see Biogeochemistry; Carbon cycle; Nitrogen cycle). The
largest fluxes of oxygen are controlled by the biosphere:
Oxygen in the atmosphere dynamic equilibrium between photosynthesis and aerobic res-
Although elemental oxygen occurs principally in the rocky piration and decay maintains the atmospheric 0 2 reservoir. In
solid Earth (see below), its considerable abundance in the contrast to the carbon cycle, the volume of the atmospheric
atmosphere also places it among the atmophile elements in reservoir is sufficient to mask any anthropogenic perturbations
the classification of Goldschmidt. Dioxygen comprises of the oxygen cycle (Schlesinger, 1991).
468 OXYGEN

1.42 A ( 14.2 nm) in eight-fold coordination (Shannon and

Atmosphere Prewitt, 1969), and is 1.4-3 times larger than most common
3.8 X 107 cations. The crust is sometimes evocatively described as a
network of close-packed oxygen ions with smaller cations in
the interstices.
Because the Earth's core has a density less than that of pure
iron, it has been suggested that oxygen, because of its great
abundance in the bulk Earth, is present as a major alloying
element. This is unlikely, however, given that the solubility of
oxygen in iron is much lower than that of other comparably
abundant light elements, notably hydrogen, sulfur and silicon
(Newson, 1995; Okuchi, 1997; Wood, 1997).
Solid-state diffusion of oxygen in minerals controls the equil-
ibration of oxygen isotopic systems and may be the rate-
limiting step in many geochemical and petrologic processes of
interest. Diffusion rates in geologic materials have proven
difficult to measure experimentally because of a myriad of
Reduced constituents extrinsic and intrinsic factors (Muehlenbachs and Connolly,
of the crust 1991 ). Published activation energies for diffusion of oxygen in
8 X 109
various silicate and oxide minerals range from 59 to 615 kJ/mol
(Muehlenbachs and Connolly, 1991; Brady, 1995), and studies
Figure 013 A simple model for the global biogeochemical cycle suggest that oxygen ions diffuse by an interstitial mechanism,
for 0 2 on the Earth. The shaded areas indicate a reservoir and even through close-packed lattices (Muehlenbachs and
fluxes of the biosphere. Data are expressed in units of 10 12 mol Connolly, 1991; Wuensch et al., 1991).
0 2 /year (adapted from Schlesinger, 1991; originally from Walker,
1980). Steven C. Semken

Comparatively small amounts of biomass escape oxidation Bibliography

through burial, and this flux is balanced over geologic time by Brady, J.B. ( 1995) Diffusion data for silicate minerals, glasses, and
exposure and weathering of sedimentary organic carbon liquids, in Mineral Physics and Crystallography: A Handbook of
(Walker, 1980). Greater fluxes of oxygen to the atmosphere Physical Constants, American Geophysical Union Reference Shelf2
(ed. T.J. Ahrens). Washington, DC: American Geophysical Union,
have apparently occurred in association with intervals of pp. 269-90.
increased burial of organic carbon, such as the formation of Brown, T.L., LeMay, H.E. Jr. and Bursten, B.E. (1997) Chemistry, the
widespread coal deposits during the Carboniferous Period. Central Science, 7th edn. Upper Saddle River, NJ: Prentice Hall,
The residence time of dioxygen is 6000 years in the biosphere 1050 pp.
(Holland, 1978) and 7000 years in the atmosphere (Stumm Ebbing, D.D. ( 1996) General Chemistry, 5th edn. Boston: Houghton
and Morgan, 1996). Mifflin Company, 1120 pp.
Emiliani, C. (1992) Planet Earth: Cosmology, Geology, and the
Evolution of Life and Environment. New York: Cambridge
Oxygen in the hydrosphere University Press, 719 pp.
Friedlander, G., Kennedy, J.W. and Miller, J.M. (1964) Nuclear and
The Henry's law constant for the dissolution equilibrium Radiochemistry, 2nd edn. New York: John Wiley & Sons, 585 pp.
of atmospheric dioxygen in water, 0 2 (g) = 0 2 (aq), is Holland, H.D. (1978) The Chemistry of the Atmosphere and Oceans.
1.26 X 10- 3 M/atm at 25°C (Stumm and Morgan, 1996). New York: John Wiley and Sons, 352 pp.
Oxygen is an important species in controlling redox conditions Klein, C. and Hurlbut, C.S. Jr. (1985) Manual of Mineralogy, after
(see Oxidation-reduction) in natural waters: biologically medi- J.D. Dana, 20th edn. New York: John Wiley and Sons, 596 pp.
ated reduction of 0 2 to H 2 0 is the predominant oxidizing Lewis, J.S. (1995) Physics and Chemistry of the Solar System. San
Diego: Academic Press, 556 pp.
reaction in the aqueous environment (Stumm and Morgan, Muehlenbachs, K. and Connolly, C. ( 1991) Oxygen diffusion in leucite:
1996). structural controls, in Stable Isotope Geochemistry: A Tribute to
Samuel Epstein (eds H.P. Taylor, Jr., J.R. O'Neill and I.R. Kaplan).
Oxygen in the solid Earth Washington, DC: The Geochemical Society, pp. 27-34.
Newsom, H.E. (1995) Composition of the Solar System, planets, mete-
Goldschmidt classified oxygen principally as a lithophile ele- orites, and major terrestrial reservoirs, in Global Earth Physics: A
ment, occurring preferentially in the crust and mantle. It is the Handbook of Physical Constants, American Geophysical Union
second most abundant element (30 wt%) in the bulk Earth and Reference Shelf 1 (ed. T.J. Ahrens). Washington, DC: American
the most abundant element in the crust and mantle ( 62.55 and Geophysical Union, pp. 159-89.
Okuchi, T. (1997) Hydrogen partitioning into molten iron at high
approximately 60 atom%, respectively). Thus, oxygen is a pressure: implications for Earth's core. Science, 278, 1781-3.
major constituent of nearly all minerals. It is the basis of the Schlesinger, W.H. (1991) Biogeochemistry: an Analysis of Global
tetrahedral [Si0 4 ] 4 - ion that polymerizes to form all silicates, Change. San Diego: Academic Press, 443 pp.
as well as the anionic building blocks (e.g. co~-, so~-, Shannon, R.D. and Prewitt, C.T. (1969) Effective ionic radii in oxides
PO~-, or OH -) of the most common non-silicate minerals. and fluorides. Acta Crystallogr., 925-46.
Stumm, W. and Morgan, J.J. ( 1996) Aquatic Chemistry: Chemical
The predominance of oxygen in the crust is even more Equilibria and Rates in Natural Waters, 3rd edn. New York: John
emphatic in terms of volume (93.8 vol%; Klein and Hurlbut, Wiley and Sons, 1022 pp.
1985). This is because the radius of the 0 2 - anion in crystals Walker, J.C.G. ( 1980) The oxygen cycle, in Handbook ofEnvironmental
ranges from 1.35 A (13.5 nm) in two-fold coordination to Chemistry, Volume 1, Part A, The Natural Environment and the
 OXYGEN ISOTOPES 469

Biogeochemical Cycles (ed. 0. Huntzinger). New York: Springer- 1989). These secondary ions are focused, separated by mass
Verlag, pp. 87-104. and detected much like with IRMS analyses. Areas with diame-
Wood, B.J. (1997) Hydrogen: an important constituent of the core? ters < 10 ~-tm can be analyzed for isotope ratio and composi-
Science, 278, 1727.
Wuensch, B.J., Semken, S.C., Uchikoba, F. and Yoo, H.I. (1991) The tion, but analytical precision is presently 10-50 times less than
mechanisms for self-diffusion in magnesium oxide. Ceramic Trans., with conventional IRMS.
24, 79-89.
Sample preparation
Cross-references For conventional isotopic analysis, oxygen in the sample must
Biogeochemistry be either extracted and converted to 0 2 or C0 2 , or equilibrated
Earth's atmosphere with C0 2. For silicates, oxides, phosphates and sulfates,
Earth's continental crust bound oxygen is freed by fluorination with F2, BrF5 , or CIF3
Earth's oceanic crust at 500-600°C, or by graphite reduction at 1000-2000°C.
Elements: lithophile Recently, lasers have been used in place of furnaces to permit
Natural gas analyses of small areas. Carbonate minerals are reacted with
Nucleosynthesis
phosphoric acid at constant temperature (usually 25-95°C),
Organic geochemistry
Oxygen isotopes producing C0 2 . Water i5 18 0 is measured by equilibrating the
Precambrian atmosphere water sample with a small amount of C0 2 at constant temper-
Stable isotopes ature (usually 25-50°C).

Notations, conventions and standards

OXYGEN ISOTOPES Oxygen isotope analyses are reported relative to a standard
using i5-notation, such that:
Oxygen occurs as three stable isotopes: 160 (99.76%), 17 0
(0.04%) and 18 0 (0.20%). 18 0/ 160 and 17 0;t 60 ratios vary in (5 18 0(%o) = (R.,- R,,ct)
Rstd
x 1000 = ( R, -1) x 1000
Rstd
natural materials by up to 10% and 5%, respectively. These
(026)
isotope variations in Earth and extraterrestrial materials are
applied to such diverse topics as the origin of the solar system, where Rx and R,,d are the 18 0/16 0 of the sample and standard,
the evolution of the atmosphere, hydrosphere and lithosphere, respectively. The values are reported in per mil (%o). For analy-
ancient climates and groundwater mixing. ses of water and almost all minerals except carbonates, the
standard used is standard mean ocean water (SMOW; also
referred to as VSMOW; Friedman and O'Neil, 1977). For
Analytical methods carbonate minerals, the reporting standard is PDB (belemnite
Isotope ratio mass spectrometry from the Peedee formation, South Carolina). Equilibrium frac-
tionation of oxygen isotopes between two phases (A and B) is
Oxygen isotope ratios are measured on C0 2 or 0 2 gas using reported as a fractionation factor rx, where:
a gas source isotope ratio mass spectrometer (IRMS; Nier,
1940; see Gas source mass spectrometry). A conventional
IRMS consists of an inlet system, an ion source, an analyzer (027)
magnet, and multiple ion detectors. C0 2 molecules are intro-
duced into the ion source through capillary leaks where the Fractionation factors often are reported as 1000 In rxA _ 8 , which
molecules are bombarded with electrons, producing CO:i ions. is approximately equal to i5A- i5 8 for samples with isotopic
These ions are accelerated in an electric field and focused ratios similar to that of the standard.
through a series of slits to form an ion beam. The ions in the
beam are deflected in the analyzer's magnetic field to a degree Fractionation relations and geothermometers
inversely proportional to their masses. C0 2 yields three major
beams with mass to charge ratios (amu/ion charge) of 44 Isotopic fractionation between phases depends on small differ-
ezc160160), 45 (13C160160, 12C170160) and 46 (mostly ences in the thermodynamic properties of the isotopically sub-
12 C 18 0 160). The three ion beams are collected simultaneously
stituted compounds. Oxygen isotopic fractionation between
phases generally decreases with increasing temperature, with
by triple collectors. After correction for interfering masses, the
1000 In rx approaching O%o at high temperature (Figure 014).
46/44 ratio yields 18 0/ 160 and the 45/44 ratio yields 13 Cj 12 C.
In silicates, 18 0 is enriched in phases with high proportions of
New instrumentation for oxygen isotopic measurement, such
tetrahedral Si and AI relative to total cation equivalents
as the gas chromatograph-isotope ratio mass spectrometer (Garlick, 1966). This fractionation is typically shown in the
(GC-IRMS) and ion microprobe (IM), can potentially form:
decrease minimum sample size and increase throughput. With
GC-IRMS, gases are first separated by chromatography then 1000 In rxx-Y =A( 10 6 T -z)- B ~ lOOO(rxx-Y- 1)
fed into the IRMS as a pulse. Used predominantly for carbon
~ ilx- i5y = Llxy (028)
isotopes, GC-IRMS can perform rapid, less precise oxygen
isotope microanalyses (Ricci et al., 1994). The IM uses a where X and Y represent different phases, A and B are con-
primary ion beam (e.g. Ar+, Cs +) to ionize atoms and mole- stants, and T is temperature in K (for compilation, see
cules on the polished sample surface (Giletti and Shimizu, Friedman and O'Neil, 1977).
470 OXYGEN ISOTOPES

Temperature (0 C) Oxygen isotope variation in nature

1000 600 400 300 200 150 Natural waters
20 Ocean water is the oxygen isotope standard (SMOW) and
hence by definition has a 6 18 0 value of about O%o relative to
SMOW (Figure 015). This () 18 0 value mostly reflects inter-
action between ocean water and ocean crust (Muehlenbachs
and Clayton, 1976; Gregory, 1991). Small but significant varia-
15 tions in seawater () 18 0 (± l%o) occur due to changes in salinity
associated with evaporation or mixing with meteoric water.
Meteoric water is ultimately derived from the evaporation of
seawater. Because water vapor is depleted in 18 0 relative to
0C\1 liquid water, meteoric water typically has negative () 18 0 values
10
I relative to SMOW. Distillation processes further deplete mete-
C: oric water in 18 0, resulting in a positive correlation between
J
c
6 18 0 and temperature in high latitude precipitation (> 35°
latitude; Dansgaard, 1964; Gat and Gonfiantini, 1981).
5 Hydrogen isotopic compositions CZHjlH or D/H) also show
0
0
0
this relationship. Most meteoric waters follow the b 2H-b 18 0
..- relation termed the meteoric water line (MWL):

0 (030)
where both 6 H and 6 18 0 are reported vs SMOW (Craig,
2

1961; Rozanski et al., 1993; Figure 016).

The relationship between 6 18 0 and temperature in high
-5 ~agnetite
latitude precipitation forms the basis for 18 0 studies of conti-
nental climate (Swart et al., 1993). Ice cores exhibit low 6 18 0
values during glacial periods and high () 18 0 values during
interglacials (Figure 017B), providing a > 100 000 year record
1 2 3 4 5 6 of climate change and a means of determining ice core age
106 T-2 (K) (Dansgaard et al., 1969). Oxygen and hydrogen isotopes are
also used as tracers for processes affecting groundwater systems
Figure 014 Mineral-water isotope fractionation as a function of such as evaporation and hydrothermal exchange (Figure 016).
temperature. Fractionation equations follow the form: 1000 In rx = Isotopic studies show that a major fraction of formation and
A(1 0 6 T- 2 ) + B, where A and Bare constants and Tis temperature geothermal waters is derived from meteoric water (Clayton
in K. Data from Friedman and O'Neil (1977) and Matsuhisa eta/. et al., 1966; Criss and Taylor, 1986).
(1979).

The temperature dependencies of mineral-water and min- chw

eral-mineral 18 0 fractionation factors serve as practical geo- oa:
:EW
thermometers virtually independent of pressure. From 1-J:
<(C..
eqn (028), differences in () values between co-existing phases
are inversely proportional to temperature (Figure 014). This
application assumes the phases formed in isotopic equilibrium
0 ~ r=--------------~D~is:so~l~ve~d0b~2~(~o~ce~a;n0)~~~~~~-----­
~~ ~~~~ 1~1111111~1111,411111114-·IIU
and did not undergo isotopic reequilibration during cooling. >- c.. Open ocean 1
Multiple co-existing mineral pairs are measured to test for J: Ill water :
preservation of isotopic equilibrium. Unfortunately, due to
Magmatic wafer I

-r
variations in closure temperature for different minerals, many
rocks do not preserve isotopic equilibrium. w Volcanic wat~r 1111
a:
Paleotemperature studies of the Earth's surface typically use w Sedimentary rodks II
J:
carbonate-water paleotemperature equations. Epstein et al.
(19 53) published the first practical equation, based mostly on
c..
1/)
0
Metamorphic rocks I \
modern
carbonates
5
...
~
J: Granitic rocks
molluskan CaC0 3 . This equation is still valid for calcite-water 1-
fractionation. An alternative equation based on abiotic calcite :::i Basaltic roc~s I
IS: Ultramafic (mantle) rockk I ~

T(°C) = 15.7- 4.36(b 18 0calcite- 6 18 0w) -60 -40 -20 0 20 40

+ 0.12(()180calcite- 6180w)2 (029) o1BQ (%o vs. SMOW)

where T =temperature, 6 18 0calcite is the 6 18 0 of calcite vs PDB, Figure 015 Oxygen isotope variation (reported vs SMOW) typically
and 6 18 0w is the 6 18 0 of the ambient water vs SMOW (Hays observed in nature (from Garlick, 1972; Hoefs, 1987; among
and Grossman, 1991; modified from O'Neil et al., 1969). others).
 OXYGEN ISOTOPES 471

50 reflect incomplete homogenization at accretion distances of

Mars and beyond (Clayton and Humayun, 1994).

0 Igneous rocks
The isotopic compositions of igneous rocks are controlled
'0'
mainly by the c:5 18 0 of the magma source, mineralogical com-
~ -50
Cl position, and the evolutionary history of the magma (e.g.
<0 assimilation of country rock; Taylor, 1977; Taylor and
Sheppard, 1986). The 18 0 composition of the Earth's mantle
-100 is estimated to be about 5.7 ± 0.2%o based on measurements
of ultramafic xenoliths and unaltered basalts. This is also a
reasonable estimate for the average composition of the Earth.
-150 Midocean ridge basalts have a narrow range of 18 0,
5.7 ± 0.5%o.
-20 -15 -10 -5 0 5 Compared to basalts, granitic rocks show large c:5 18 0 varia-
tion because of enhanced potential for secondary alteration
o18Q (%o)
and variability in the magma source. In general, granitic rocks
Figure 016 Hydrogen and oxygen isotopic compositions of are classified as low c:5 18 0 (<6%o), normal c:5 1s0 (6-10%o), and
meteoric waters and the meteoric water line (MWL). SMOW is high c:5 18 0 (> 10%o; Taylor, 1977; Hoefs, 1987). Low c:5 1s0
Standard Mean Ocean Water. Unmodified groundwaters will have values reflect formation from 1s0 depleted magmas, whereas
compositions that fall near the MWL. Arrows show typical trends high c:5 1s0 values are probably derived from 18 0-enriched sedi-
for evaporation of water A and hydrothermal rock-water exchange mentary or metamorphic rocks (Figure 015). Normal c:5 18 0
of water B. values can occur through differentiation of a normal basaltic
magma, but also can have a more complex origin. Positive
correlations between c:5 18 0 and 87 Sr/86 Sr in granitic and andes-
A. itic rocks provide evidence for this crustal contamination of
the magma (Magaritz et al., 1978; Taylor and Sheppard, 1986).
c:5 1 so_s 7 Sr/s 6 Sr trends reflect the combined effects of assimila-
tion and fractional crystallization, and may be used to distin-
guish between source contamination (reflux of subducted
Depth in core (em) material) and crustal contamination (contamination of mantle-
derived material by continental crust).
B.
Hydrothermal alteration of igneous rocks
Oxygen isotope compositions are sensitive indicators of hydro-
thermal alteration. Studies of ocean crust and ophiolites reveal
that high temperature (> 300"C) interaction with seawater
results in 180 enrichment in the water and 18 0 depletion in the
basalt. Conversely, low temperature rock-water interaction
40 120 160 200 240 280 320
( < 300°C) results in 1s0 enrichment in the rock and 18 0 deple-
Time (ka) tion in the water (Muehlenbachs and Clayton, 1976; Gregory
and Taylor, 1981). In granitic plutons, hydrothermal inter-
Figure 017 (A) Oxygen isotopic stratigraphy (vs PDB) of the action with 180-depleted meteoric water results in distinctive
planktonic foraminifera Globigerinoides sacculifer from core 18 0 depletions (Criss and Taylor, 1986).
V28-238 in the equatorial Pacific. Even-numbered isotope stages
are glacial stages. Magnetic stratigraphy shown at top. Modified
from Shackleton and Opdyke (1973). (B) Comparison of normalized Metamorphic rocks
marine, Antarctic ice cap, and Devils Hole calcite (groundwater
proxy) b 18 0 records for middle to late Pleistocene. Absolute c:5 180 Oxygen isotope analyses of metamorphic rocks are used to
variation is < 1.8%o for marine record, 8.2%o for Devils Hole. evaluate temperature of formation, origin of protolith, fluid-
Dotted lines connect correlative features. Age of features of Devils rock interaction, degree of oxygen exchange (homogenization),
Hole curve are minimum values because of groundwater residence and effect of volatilization (O'Neil, 1979; Valley, 1986; Hoefs,
time (modified from Winograd eta/., 1988 and references therein). 1987). Rocks of higher metamorphic grade have lower c:5 1s0
values and smaller c:5 18 0 differences between minerals. Oxygen
isotope temperatures are often lower than peak metamorphic
Meteorites temperatures because of retrograde exchange. Fluid-rock inter-
Certain high-temperature minerals in carbonaceous meteorites action during contact metamorphism is evidenced by c:5 1s0 gra-
show anomalous 170 and 18 0 depletions, suggesting mixing dients from 1s0-depleted plutons to 1s0-enriched country rock.
with a 16 0-rich component (Clayton, 1993 and references Typically, c:5 18 0 values show greater fluid penetration than is
therein). Similarities in the 17 0-18 0 trends of the Earth, Moon, indicated by metamorphic mineralization. In regional meta-
and enstatite meteorites suggest homogenization of the inner morphic studies, c:5 18 0 values indicate whether the system is
solar system; isotopic differences between Earth and Mars open to fluid flow or closed (generally, fluid absent).
472 OXYGEN ISOTOPES

Sediments and sedimentary rocks most Mesozoic isotope data are based on macrofossils (belem-
nites, brachiopods, bivalves, ammonoids, etc.) from marine
The original b 18 0 of clays reflects mineralogy and the temper-
deposits exposed on land (as with the Paleozoic record). b 18 0
ature and 3 18 0 of the water in which the clay formed. This
values are generally between -3 and + 1%o for aragonite and
original o18 0 may be modified after deposition depending on
calcite, equivalent to paleotemperatures of roughly 10-25°C
grain size, time, and temperature (Yeh and Savin, 1976).
(b 18 0w = -1.2%o; Urey et al., 1951; Hudson and Anderson,
Equilibrium b 18 0 of marine carbonates is defined by paleotem-
1989 and references therein).
perature equations such as eqn (029). Because most marine
carbonate sediments are biogenic, their 18 0 composition may
be influenced by a taxa-dependent vital effect (apparent iso- Tertiary and Quaternary
topic disequilibrium due to metabolic effects). Bivalve and Oxygen isotopic analyses of planktic and benthic foraminifera
brachiopod carbonate tends to be in isotopic equilibrium with from DSDP and ODP drill cores and hydraulic piston cores
the water, whereas coral and echinoderm carbonate is usually provide a detailed Tertiary paleoclimate record. Isotopic analy-
depleted in 18 0 relative to equilibrium (Anderson and Arthur, ses of benthic foraminifera show that deep ocean temperatures
1983; Wefer and Berger, 1991). Carbonate sediments are lithi- were about 10°C in the latest Cretaceous (b 18 0 = 0.4%o; includ-
fied through compaction and cementation. Oxygen isotope ing a + 0.6%o correction for vital effect where necessary),
analyses of cements aid in characterizing the fluids (marine vs warmed to about l3°C in the early Eocene ( -0.2%o), then
meteoric) and temperatures associated with cementation progressively cooled to modern values of about ooc (4.3%o;
(Dickson and Coleman, 1980). b 18 0w = O%o; Figure 018; Savin, 1977; Miller et al., 1987).
Oxygen isotopic compositions in carbonate sediments are Distinct positive b 18 0 shifts occur in the earliest Oligocene,
valuable paleoenvironmental indicators, recording variations middle Miocene and Plio-Pleistocene. The middle Miocene and
in temperature and water b 18 0. 18 0 variations in marine sedi- Plio-Pleistocene shifts likely reflect continental ice accumula-
ments form the basis for oxygen isotope stratigraphy (discussed tion. The cause of the earliest Oligocene shift, whether glacia-
below). In lacustrine sediments and speleothems, b 180 is tion or cooling, is still debated.
largely controlled by variations in meteoric water 6 18 0 (Eicher 18 0 analyses of planktic and benthic foraminifera from

eta!., 1981; Schwartz, 1986), which may provide a record of piston cores of deep marine sediments reveal a sawtooth
temperature or amount of rainfall. Vein calcite from Devils pattern reflecting changes in the isotopic composition of ocean
Hole, Nevada, has yielded a surprisingly detailed 18 0 record water with the accumulation of 18 0-depleted glacial ice and,
of Quaternary continental paleoclimate (Winograd et al., 1988, to a lesser degree, temperature change associated with glacia-
1992). This record mimics the major features of ice-core and tion (Figure 018A; Emiliani, 1955; Shackleton and Opdyke,
marine 18 0 records (Figure 017B), but the U-series dates of 1973; Williams, 1984). This serves as evidence for cyclical
climate events conflict with those from marine chronostratigra- glaciation and a basis for correlation of Quaternary sediments.
phy (see below). Spectral analyses of these records indicate a relation between

Oxygen isotope stratigraphy Temperature

("C)
Precambrian, Paleozoic and Mesozoic 14
-1
Isotopic stratigraphies of Precambrian to Mesozoic sediments 16 12
show a progressive increase in b 18 0 during evolution of the 14
10
biosphere. These data can be interpreted as higher temper- 0 Q)

atures or lower seawater b 18 0 in the past, or as an artifact of 1! 12

Q)
8
greater diagenetic alteration in older samples. Precambrian 18 0 ·;. 10
6
0
:c
records are based on whole-rock carbonate and chert samples ~ ·~
8
0 0
(Perry, 1967; Veizer and Hoefs, 1976), and include diagenetic <X> Q. 4
~ 6
phases which likely mask the original marine signal. Paleozoic 2
18 0 stratigraphies are based on brachiopod shells and marine 2
4
cements, materials which provide better control on diagenesis
(Popp et al., 1986; Grossman, 1994; Lohmann and Walker,
1989; Veizer et al., 1986; Hudson and Anderson, 1989). For
3
*
~ 2
Q)
.\1
0

Q)~
0 ~~
low latitude samples, b 18 0 values are typically about - 5 ± 1%o .~ I~
(carbonates reported vs PDB) for the early Paleozoic (through 0 10 20 30 40 50 60 70 "ECi:!
0'!2" '2.
the Devonian), and about -2 ± 1%o for the late Paleozoic. wo "0•
o II
Age (Ma)
Assuming an ice-free Earth and modern hydrosphere b 18 0 ~E ~
e
values (i.e. o18 0w = -1.2%o), calcite 3 18 0 values yield early
and late Paleozoic isotopic temperatures of 35 ± soc and Figure 018 Composite oxygen isotopic record for benthic
20 ±SOC, respectively. Assuming a higher ocean water b 18 0 foraminifera from Atlantic DSDP sites corrected to Cibicidoides
value due to the presence of glacial ice would yield higher 6 180 (0.64%o less than equilibrium) and reported vs PDB (from
Miller eta/., 1987). The horizontal line at 1.8%o indicates
calculated paleotemperatures. It remains unclear whether the
paleotemperatures equivalent to modern temperatures (assuming
low b 18 0 values for early Paleozoic carbonates are due to 'ice-free' conditions; see below) and is used by Miller eta/. to
o
higher temperatures, lower Gcean water 18 0, or poorer sample indicate conditions suitable for growth of significant ice sheets.
preservation. Temperature scale at right is for paleotemperatures calculated
Except for limited foraminiferal data from Deep Sea Drilling assuming modern ocean water 6 18 0 (b 18 0w = -0.3%o) and b 180w
Project (DSDP) and Ocean Drilling Program (ODP) cores, assuming 'ice-free' conditions ( -1.2%o).
 OXYGEN ISOTOPES 473

Quaternary glaciation and the Earth's orbital geometry Grossman, E.L. (1994) The carbon and oxygen isotope record during
(Milankovitch hypothesis; Hays et al., 1976). The importance the evolution of Pangea: Carboniferous to Triassic. Geol. Soc. Am.
Special Paper, 288, 207-28.
of orbital geometry on climate is contested by Winograd et al. Hays, J.D., Imbrie, J. and Shackleton, N.J. ( 1976) Variations in the
(1988, 1992). Based on the Devils Hole chronology (Figure Earth's orbit: Pacemaker of the Ice Ages. Science, 194, I 121-32.
017B), they argue for aperiodic climate change controlled by Hays, P.D. and Grossman, E.L. (1991) Oxygen isotopes in meteoric
non-linear feedbacks within the atmosphere-ice sheet-ocean calcite cements as indicators of continental climate. Geology, 19.
system. While the cause of climate change continues to be 441-4.
Hoefs, J. ( 1987) Stable Isotope Geochemistry, 3rd edn. Berlin: Springer-
debated, one fact is clear: oxygen isotope analyses of terrestrial Verlag, 241 pp.
and marine sediments yield global climate records. Hudson, J.D. and Anderson, T.F. (1989) Ocean temperatures and
isotopic compositions through time. Tl·ans. R. Soc. Edinburgh:
Earth Sci., 80, 183-92.
Ethan L. Grossman Lohmann, K.C. and Walker, J.C.G. (1989) The 6 18 0 record of
Phanerozoic abiotic marine calcite cements. Geophys. Res. Lett.,
16, 319-22.
Bibliography Magaritz, M., Whitford, D.J. and James, D.E. (1978) Oxygen isotopes
Anderson, T.F. and Arthur, M.A. (1983) Stable isotopes of carbon and the origin of high- 87 Sr/86 Sr andesites. Earth Planet. Sci. Lett.,
and oxygen and their application to sedimentologic and paleo- 40, 220-30.
environmental studies, in Stable Isotopes in Sedimentary Geology Matsuhisa, Y., Goldsmith, J.R. and Clayton, R.N. (1979) Oxygen
(eds M.A. Arthur eta/.). SEPM Short Course 10:1-1-1-151, Tulsa, isotopic fractionation in the system quartz-albite-anorthite-water.
SEPM. Geochim. Cosmochim. Acta, 43, 1131-40.
Clayton, R.N. ( 1993) Oxygen isotopes in meteorites. Annu. Rev. Earth Miller, K.G., Fairbanks, R.G. and Mountain, G.S. (1987) Tertiary
Planet. Sci., 21, 115-49. oxygen isotope synthesis, sea level history, and continental margin
Clayton, R.N. and Humayun, M. (1994) Stable isotope evidence for erosion. Paleoceanography, 2, 1-19.
Earth's raw materials. Mineral. Mag., 58a, 177-8. Muehlenbachs, K. and Clayton, R.N. ( 1976) Oxygen isotope composi-
Clayton, R.N., Friedman, I., Graf, D.L., Mayeda, T.K., Meents, W.F. tion of the oceanic crust and its bearing on seawater. J. Geophys.
and Shimp, N.F. (1966) The origin of saline formation waters, I. Res., 81, 4365-69.
Isotopic composition. J. Geophys. Res., 71, 3869-82. Nier, A.O. ( 1940) A mass spectrometer for routine isotope abundance
Craig, H. (1961) Isotopic variations in meteoric waters. Science, 133, measurements. Rev. Sci. Instruments, 11, 212-16.
1702-3. O'Neil, J.R. (1979) Stable isotope geochemistry of rocks and minerals,
Criss, R.E. and Taylor, H.P. Jr. (1986) Meteoric-hydrothermal sys- in Lectures in Isotope Geology (eds E. Jager and J. Hunzike). Berlin:
tems. Rev. Mineral., 16, 373-424. Springer, pp. 283-312.
Dansgaard, W. (1964) Stable isotopes in precipitation. Tellus, 16, O'Neil, J.R., Clayton, R.N. and Mayeda, T.K. ( 1969) Oxygen isotopic
436-68. fractionation in divalent metal carbonates. J. Chem. Phys., 51,
Dansgaard, W., Johnsen, S.J., Moller, J. and Langway, C.C. Jr. (1969) 5547-8.
One thousand centuries of climatic record from Camp Century on Perry, E.C. (1967) The oxygen isotope chemistry of ancient cherts.
the Greenland Ice Sheet. Science, 166, 377-81. Earth Planet. Sci. Lett., 3, 62-6.
Dickson, J.A.D. and Coleman, M.L. ( 1980) Changes in carbon and Popp, B.N., Anderson, T.F. and Sandberg, P.A. (1986) Brachiopods
oxygen isotope composition during limestone diagenesis. as indicators of original isotopic compositions in some Paleozoic
Sedimentology, 27, 107-18. limestones. Geo/. Soc. Am. Bull., 97, 1262-9.
Ricci, M.P., Merritt, D.A., Freeman, K.H. and Hayes, J.M. (1994)
Eicher, U., Siegenthaler, U. and Wegmtiller, S. (1981) Pollen and
Acquisition and processing of data for isotope-ratio-monitoring
oxygen isotope analyses of late-post-glacial sediments of the
mass spectrometry. Org. Geochem., 21,561-71.
Toubiere de Chirens (Dauphine, France). Quaternary Res., 15,
Rozanski, K., Aragmis-Araguas, L. and Gonfiantini, R. (1993)
160-70.
Isotopic patterns in modern global precipitation. Geophys.
Emiliani, C. (1955) Pleistocene temperatures. J. Geol., 63. 538-78. A1onogr., 78, 1-36.
Epstein, S., Buchsbaum, R., Lowenstam, H. and Urey, H.C. (1953) Savin, S.M. (1977) The history of the Earth's surface temperature
Revised carbonate-water isotopic temperature scale. Geol. Soc. during the past I 00 million years. Annu. Rev. Earth Planet. Sci.,
Am. Bull., 64, 1315-26. 5, 319-55.
Friedman, I. and O'Neil, J.R. (1977) Data of geochemistry, in Schwartz, H.P. (1986) Geochronology and isotopic geochemistry of
Compilation of Stable Isotope Fractionation Factors of' Geochemical speleothems, in Handbook of Environmental Isotope Geochemistry;
Interest, 6th edn. US Geological Survey Professional Paper Vol. 2, The Terrestrial Environment B (eds P. Fritz and J.Ch.
440-KK. Washington, DC: US Government Printing Office, Fontes). Amsterdam: Elsevier, pp. 271-303.
pp. KK1-KK12. Shackleton, N.J. and Opdyke, N.D. ( 1973) Oxygen isotope and
Garlick, G.D. ( 1966) Oxygen isotope fractionation in igneous rocks. palaeomagnetic stratigraphy of equatorial Pacific core V28-238:
Earth Planet. Sci. Lett., 1, 361-8. Oxygen isotope temperatures and ice volumes on a I 05 year and
Garlick, G.D. ( 1972) Oxygen isotope geochemistry, in The 10 6 year scale. Quaternary Res., 3, 39-55.
Encyclopedia of Geochemistry and Environmental Sciences (ed. R.W. Swart, P.K., Lohmann, K.C., McKenzie, J. and Savin, S. (1993)
Fairbridge). New York: Van Nostrand Reinhold, pp. 864-74. Climate change in continental isotopic records. Geophys. Monogr.,
Gat, J.R. and Gonfiantini, R. (1981) Stable Isotope Hydrology: 78,374 pp.
Deuterium and Oxygen-18 in the Water Cycle. Technical Report Taylor, H.P. Jr. ( 1977) Water/rock interactions and the origin of H 2 0
Series No. 210, Vienna: International Atomic Energy Agency, in granitic batholiths. J. Geol. Soc. London, 133, 509-58.
337 pp. Taylor, H.P. Jr. and Sheppard, S.M.F. (1986) Igneous rocks: I.
Giletti, B.J. and Shimizu, N. (1989) Use of the ion microprobe to Processes of isotopic fractionation and isotope systematics. Rev.
measure natural abundances of oxygen isotopes in minerals. US Mineral., 16,227-71.
Geol. Surv. Bull., 1890, 129-36. Urey, H.C., Lowenstam, H.A., Epstein, S. and McKinney, C.R. (1951)
Gregory, R.T. ( 1991) Oxygen isotope history of seawater revisited: Measurement of paleotemperatures and temperatures of the upper
timescales for boundary event changes in the oxygen isotope com- Cretaceous of England, Denmark, and the southeastern United
position of seawater. Geochem. Soc. Special Pub/., 3, 65-76. States. Geo/. Soc. Am. Bull., 62, 399-416.
Gregory, R.T. and Taylor, H.P. (1981) An oxygen isotope profile in a Valley, J.W. (1986) Stable isotope geochemistry of metamorphic rocks.
section of Cretaceous oceanic crust, Samail ophiolite, Oman: Rev. Mineral., 16, 445-89.
Evidence for .5 18 0 buffering of the oceans by deep (> 5 km) sea- Veizer, J. and Hoefs, J. (1976) The nature of 0 18 /0 16 and C 13 /C 12
water-hydrothermal circulation at mid-ocean ridges. J. Geophys. secular trends in sedimentary carbonate rocks: Geochim.
Res., 86, 2737-55. Cosmochim. Acta, 40, 1387-95.
 474 OXYGEN ISOTOPES

Veizer, J., Fritz, P. and Jones, B. (1986) Geochemistry of brachiopods: Yeh, H. and Savin, S. (1976) The extent of oxygen isotope exchange
oxygen and carbon isotopic records of Paleozoic oceans. Geochim. between clay minerals and sea water. Geochim. Cosmochim. Acta,
Cosmochim. Acta, 50, 1679-96. 40, 743-8.
Wefer, G. and Berger, W. (1991) Isotope paleontology: growth and
composition of extant calcareous species. Mar. Geol., 100, 207-48. Cross-references
Williams, D.F. (1984) Correlation of Pleistocene marine sediments of Carbon isotopes
the Gulf of Mexico and other basins using oxygen isotope stratigra- Gas source mass spectrometry
phy, in Principles of Pleistocene Stratigraphy Applied to the Gulf of Geochemistry: low temperature
Mexico (ed. N. Healy-Williams). Boston: International Human Hydrogen isotopes
Resources Development Corporation, pp. 65-118. Hydrologic cycle
Winograd, I.J., Coplen, T.B., Landwehr, J.M. et al. (1992) Continuous Isotope fractionation
Nitrogen isotopes
500,000-year climate record from vein calcite in Devils Hole, Ore deposits
Nevada. Science, 258, 255-60. Oxygen
Winograd, I.J., Szabo, B.J., Coplen, T.B. and Riggs, A.C. (1988) A Paleotemperatures
250,000-year climatic record from Great Basin vein calcite: Stable isotopes
Implications for Milankovitch theory. Science, 242, 1275-80. Sulfur isotopes

View publication stats