David J. Mossman

The geochemistry of Witwatersrand-type gold deposits and the possible influence of ancient prokaryotic communities on gold dissolution and precipitation

Abstract Compared to the prokaryote-dominated lower Proterozoic there are few environments now left on Earth where prokaryotes are dominant. By about 2.5 Ga chemical conditions affecting the solution and precipitation of minerals were much influenced by microbial communities. Results of recent experiments show that modern living prokaryotes are capable of actively precipitating or flocculating gold. This process goes far towards explaining the phenomenal extent to which gold is enriched in kerogenous marker seams on the Witwatersrand. Neutron activation analyses confirm that gold is likewise anomalously enriched in kerogenous material present in the Lower Huronian metasediments of Ontario. Long known as thucholite, it appears from pyrolysis and X-ray studies of a sample from a stratiform occurrence that it is a kerogen remarkably similar to kerogen from the Vaal Reef, South Africa. A prokaryote precursor has been postulated for the Vaal Reef kerogen, and the same may well be true of at least some of the stratiform occurrences of kerogen in the Huronian. Simplifying solutions to the metallogeny of Witwatersrand-type gold that do not take into account the diverse types of gold in the deposits, are unrealistic. Precipitated gold associated with kerogen is only one of four main types of Witwatesrand gold, but it may be the major one. Our proposed model to account for this gold involves the weathering of gold from Archean source rock under anaerobic conditions, and probably in the presence of S-cycling microbial communities. Transported as a solution or colloid the gold was stabilized by humic acids, or S-cycle intermediates, and was deposited onto braided alluvial plains. There, in the presence of extensive prokaryote microbial mats (now preserved as kerogen), gold was precipitated in forms characteristic of biologically induced biomineralization.

Role of organic matter in the Proterozoic Oklo natural fission reactors, Gabon, Africa

Of the sixteen known Oklo and the Bangombe natural fission reactors (hydrothermally altered elastic sedimentary rocks that contain abundant uraninite and authigenic clay minerals), reactors 1 to 6 at Oklo contain only traces of organic matter, but the others are rich in organic substances. Reactors 7 to 9 are the subjects of this study. These organic-rich reactors may serve as time-tested analogues for anthropogenic nuclear-waste containment strategies. Organic matter helped to concentrate quantities of uranium sufficient to initiate the nuclear chain reactions. Liquid bitumen was generated from organic matter by hydrothermal reactions during nuclear criticality. The bitumen soon became a solid, consisting of polycyclic aromatic hydrocarbons and an intimate mixture of cryptocrystalline graphite, which enclosed and immobilized uraninite and the fission-generated isotopes entrapped in uraninite. This mechanism prevented major loss of uranium and fission products from the natural nuclear reactors for 1.2 b.y. 24 refs., 4 figs.

Geology of precambrian paleosols at the base of the huronian supergroup, elliot lake, Ontario, Canada

Abstract Detailed examination of two new profiles through the Denison Series paleosol, and one new profile through a Pronto Series paleosol in the Elliot Lake area, lend fresh insight into atmospheric and hydrogeologic environmental conditions of Precambrian Earth. These are reassessed, with reference to the results of previous studies, as true indigenous paleosols of polygenetic character. Virtually all of the criteria (Grandstaff et al., 1986) for identification of paleosols are met. The paleosols bear strong mineralogical and chemical similarities despite derivation from different parent materials, and sampling at locations distant from each other. There has evidently been a common reaction to weathering processes peculiar to the Early Proterozoic. Diagenetic and metamorphic events mask, to a considerable extent, the role played by environmental and hydrologic conditions in the development of the paleosols. In the Denison greenstone-derived paleosol, clay formed from amphibole, feldspar and chlorite is altered to sericite. Increased quartz content is due to residual concentration and to secondary development. In granodiorite-derived paleosol, quartz and sericite are generally more abundant; quartz is relatively concentrated toward the top of the profile. The paleosols are thus primarily sericite, quartz and chlorite assemblages. Rutile and leucoxene are accessory minerals. Profiles of elemental chemistry show the most drastic changes in concentration at the base, in transition to bedrock. Magnesia and lime are leached from all profiles, as in modern soils, although in common with most Precambrian paleosols, potash accumulated. Total Fe is strongly depleted from the profiles except just above the bedrock contact with granodiorite, a condition attributed to mobilization of Fe from the paleosol to a position directly over relatively impermeable parent rock. Use of Hollands method of establishing oxygenation from paleosols indicates that in terms of Fe content, parent material of granodiorite-derived paleosol should produce a reduced Fe-paleosol. The oxygen demand of this rock was greater than granite, owing to an abundant ferromagnesian content. Two fossil eluvial horizons, marked by high chlorite and a dramatic decrease in sericite content, are recognized in the Denison paleosol. Their existence shows that although initial stages of weathering may have occurred under water-saturated conditions, conditions were probably more arid than supposed by previous workers, particularly during the later stages of development of the paleosol. Although Fe-enriched, the eluvial horizons are drastically depleted in potash, Al and Th. Removal of Al 2 O 3 negates normalization of chemical analyses of paleosols to Al 2 O 3 , for purposes of constant volume calculations. Th, associated with sericite in the profiles, increases at the contact with the overlying Matinenda Formation, in association with rutile and leucoxene, owing to interaction with diagenetic U-bearing fluids from overlying sediments. Rutile and leucoxene originate from parent rocks, and are a late-stage form of mineralization. These titaniferous phases indirectly contributed to the local dominance of brannerite-type orc through diagenetic processes of mineralization involving the Pronto Reaction. Paleotopographic greenstone highs are related to local, anomalously high concentrations of heavy minerals in the brannerite-rich orc. This involves hydraulic deposition of heavy minerals in the lee of crescent-shaped greenstone highs, with subsequent diagenetic upgrading. Granodiorite-derived paleosol contains 5–10 times more Au than the Denison paleosol, and is modestly enriched above background. A slight decrease in Au content upward, evident in all profiles, is an artefact of erosion of the top of the paleosols.

Characterization of insoluble organic matter from the Lower proterozoic huronian supergroup, Elliot Lake, Ontario

Abstract Elliot Lake organic matter falls into two main groups: (1) stratiform kerogens with reflectance from 2.63 to 7.31% Ro(max), exhibit a high level of aromaticity and low atomic H/C ratios (0.41 − 0.60): (2) less aromatic disseminated globules, most resembling solid bitumen are of lower maturity, with Ro(max) from 0.72 to 1.32. and atomic H/C ratios 0.71 − 0.81. Stratiform kerogens, originated from cyanobacteria, have δ13C values from −15 to −25%. One exception with δ13C = − 31.8%, may have originated from methanogenic source biota: thus two genetically distinct populations of kerogen are tentatively recognized in Elliot Lake metasediments. Globular bitumens, remobilized by hydrous pyrolysis from kerogen, exhibit δ13C values ranging from −25% tp −33%0. The overall isotopic bimodality among Elliot Lake organic material is explained by analogy to petroleum generation from particulate kerogen that matured under conditions of catagenesis to metagenesis and was subjected to radiation over a long time. Barium, U, Th, Au, Pt, Pd and REE contents of Elliot Lake stratiform kerogen greatly exceed of average black shales. Except for As, Pt, Pd and Au, all metals (Ti, V, Cr, Mn, Fe, Co, Ni, Zn, Se, Rb, Sr, Ag, Sb, Cs, Ba, Hf, Ta, W, Hg, Th and U) are more concentrated in stratiform kerogens than in the derivative bitumen. This signals a significant role for a low-temperature transfer of noble metals to migrating bitumen.

Nature and origin of stratiform kerogen seams in lower proterozoic witwatersrand‐type paleoplacers—the case for biogenicity

Abstract Direct and indirect evidence for the biogenicity of Witwatersrand‐type kerogen is reviewed. Photomicrographs of microfossil‐like structures from Huronian supergroup kerogen are presented as further direct evidence for biogenicity.

Hydrothermal alteration of organic matter in uranium ores, Elliot Lake, Canada: Implications for selected organic-rich deposits

Abstract Organic matter in the uraniferous Matinenda Formation, Elliot Lake, is preserved in the forms of syngenetic kerogen and solid bitumen as it is in many of the Oklo uranium deposits and in the Witwatersrand gold-uranium ores. The Elliot Lake kerogen is a vitrinite-like material considered to be remnants of Precambrian cyanobacterial mats. The kerogen at Elliot Lake has reflectances (in oil) ranging from 2.63–7.31% Romax, high aromaticity, relatively low (0.41–0.60) atomic H C ratios, and it contains cryptocrystalline graphite. Bitumen, present primarily as dispersed globules (up to 0.5 mm dia.), has reflectances from 0.72–1.32% Romax, atomic H C ratios of 0.71–0.81, and is somewhat less aromatic than the kerogen. Overall similarity in molecular compositions indicates that liquid bitumen was derived from kerogen by processes similar to hydrous pyrolysis.The carbon isotopic composition of kerogen (−15.62 to −24.72%.), and the now solid bitumen (−25.91 to −33.00%.) are compatible with these processes. Despite having been subjected to several thermal episodes, ca. 2.45 Ga old kerogen of microbiological origin here survived as testimony of the antiquity of life on Earth. U-Pb isotopic data from discrete kerogen grains at Elliot Lake form a scattered array intersecting concordia at 2139 ± 100 Ma, correspond to the Nipissing event. U-Pb systems were totally reset by this event. Uranium and lead show subsequently partial mobility, the average of which is indicated by the lower concordia intersect of 550 ± 260 Ma. The migrated bitumen contains virtually no uranium and thorium but has a large excess of 206Pb, which indicates that the once liquid bitumen must have acted as a sink for mobile intermediate decay products of 238U. Emplacement of the Nipissing diabase may have been responsible for producing the bitumen and, indirectly, for its enrichment in 206Pb as a result of outgassing of 222Rn. Except for noble metals and arsenic, the kerogen contains higher abundances of trace elements than bitumen; REE distributions vary sympathetically and HREE depletion occurs in both organic matter types. Significant relative enrichment of noble metals in bitumen resulted from low temperature processes, a phenomenon manifest in various platinum group element-organic matter associations at other localities.

Testing for fullerenes in geologic materials: Oklo carbonaceous substances, Karelian shungites, Sudbury Black Tuff

Fullerenes have been reported from diverse geologic environments since their discovery in shungite from Karelian Russia. Our investigation is prompted by the presence of onionskin-like structures in some carbonaceous substances associated with the fossil nuclear fission reactors of Oklo, Gabon. The same series of extractions and the same instrumental techniques, laser desorption ionization and high-resolution mass spectroscopy (electron-impact mass spectroscopy), were employed to test for fullerenes in samples from three different localities: two sites containing putative fullerenes (Sudbury Basin and Russian Karelia) and one new location (Oklo, Gabon). We confirm the presence of fullerenes (C 6 0 and C 7 0 ) in the Black Tuff of the Onaping Formation impact breccia in the Sudbury Basin, but we find no evidence of fullerenes in shungite samples from various locations in Russian Karelia. Analysis of carbonaceous substances associated with the natural nuclear fission reactors of Oklo yields no definitive signals for fullerenes. If fullerenes were produced during sustained nuclear fission at Oklo, then they are present below the detection limit (∼100 fmol), or they have destabilized since formation. Contrary to some expectations, geologic occurrences of fullerenes are not commonplace.

Petrography and paragenesis of organic matter associated with the natural fission reactors at Oklo, Republic of Gabon: a preliminary report

Sixteen known uranium-rich pockets in the sediment-hosted uranium ore deposits in the Oklo area became nuclear fission reactors 1968 ± 50 Ma ago and operated as such for up to 1 Ma. Nuclear criticality was caused by unique and fortuitous geological environments and events. These included the localized high concentrations of uranium in small pockets in the ore bodies, the fact that the relative abundance of the fissile 235U isotope was five times greater in uranium ∼ 2 Ga ago than present, the presence of water which acted as a moderator in the natural reactors and the absence of neutron poisons, which are elements that can prevent nuclear chain reactions. Organic matter is present in all of the Oklo natural reactors, but it is abundant only in reactors 7 to 16. Organic matter has been studied in natural reactors 7 to 9 and at locations at various distances from these reactors. Bitumen in the Oklo reactors is the predominant organic phase. It is now solid bitumen containing dispersed cryptocrystalline graphite in crystal domains smaller than optical microscopic size. Its immediate precursor was liquid bitumen, generated from syngenetic kerogen in organic-rich sedimentary rocks in the Francevillian Basin of Gabon and in the natural reactors themselves by hydrous reaction mechanisms prior to, during, and after nuclear criticality. A preliminary paragenesis of organic matter at Oklo may be defined through the following sequence of events. (a) Protokerogen and kerogen evolve from abundant cyanobacteria in the tidal and deltaic sediments in the basal FA formation of the Francevillian Series, during sedimentation and burial, respectively. (b) During subsequent subsidence of the basin the first generation of bitumen and petroleum occurred. (c) During uplift the sedimentary strata were fractured, and bitumen and petroleum migrated through fractures. Following this, oxidizing aqueous solutions carrying uranyl ions migrated through the highly fractured rocks, met and were reduced by organic matter and precipitated pitchblende or uraninite at the present sites of the Oklo uranium ore deposits and natural reactors. The natural reactors reached nuclear criticality, they were heated and produced the second generation of bitumen through hydrous reactions from the by now solid first-generation bitumen and remaining kerogen. Finally, hydrothermal effects associated with the dolerite dike swarm intrusion 750 ± 150 Ma ago produced yet another localized generation of liquid bitumen from solid organic matter. Possibly other bitumen generations also occurred during and following this time interval. The organic matter in the natural reactors soon became a solid, and was able to prevent the loss of uraninite grains enclosed in it. These uraninite grains trapped and then immobilized fission products until, and to an extent even after, a major tectonic-igneous event, the intrusion of the dolerite dike swarm.

Origin and distribution of gold in the Huronian Supergroup, Canada — the case for Witwatersrand-type paleoplacers

Abstract Field observations and experimental results show that gold is mobile under a wide range of natural conditions in the surficial environment. However, the extent to which, and the form(s) in which gold was mobile in ancient placers remains speculative. Rather more convincing is the extent to which diagenetic and metamorphic processes have been active in redistributing the gold. Huronian paleoplacer gold deposits span a critical transition in Earth history, namely, the oxyatmoversion, evidence for which exists in the upper Gowganda Formation dated at 2.288 Ga. Prior to this transition, deposition of gold occurred under reducing atmospheric conditions, with transportation of the more finely-divided material possibly as organic-protected colloids, as has been suggested for the Witwatersrand. Following the oxyatmoversion, gold deposition will have been subject to secondary enrichment, like many Phanerozoic placer gold occurrences. For this reason, and on purely sedimentological grounds, upper Huronian strata ought to have as much potential for hosting economic deposits of gold as the basal units. A total of 121 Au and AuU occurrences, including several past and presently producing mines from the Huronian Supergroup, are examined. These are classified according to whether mineralization is: in or adjacent to diabase dikes (11 cases); in (quartz, quartz—carbonate) veins (85 cases); stratiform (25 cases). Of the non-diabase-hosted occurrences, 41.3% occur in the Cobalt Group, 15.7% in the Quirke Lake Group, 24.9% in the Hough Lake Group and 9% in the Elliott Lake Group. Frequency of occurrence can be related to transgressive sedimentary cycles, with deposits concentrated in the Matinenda, Mississagi and Gowganda Formations, which immediately overlie the Archean—Huronian unconformity. Most of the deposits occur in the Gowganda Formation, although none of these is stratiform. In terms of Au content, there is a large overlap in class intervals of stratiform vein deposits. Vein deposits are, in general, richer than stratiform by a factor of 10. Selected stratiform deposits in the Matinenda, Mississagi and Serpent Formations are examined in light of available geological and geochemical data. In these deposits, anomalous gold values in dominantly quartzitic metasediments are accompanied by fine-grained pyrite and other heavy minerals, including uranium, which occurs in most, but not all cases. Metamorphic grade ranges from upper greenschist to lower amphibolite facies. A few of the stratiform occurrences are accompanied by accumulations of carbonaceous material, an association reminiscent of the Witwatersrand goldfields. Results of electron-microprobe study indicate that much of the gold in the Huronian metasediments occurs as low level concentrations in pyrite of morphologically different types, in arsenopyrite, chalcopyrite, and in pyrrhotite variously altered to marcasite. It is clear that Huronian paleoplacer gold deposits exist, but only in conditions much modified by diagenetic and metamorphic processes.

Carbonaceous substances in mineral deposits: implications for geochemical exploration

Abstract Carbonaceous substances (CS) are intimately linked directly and/or indirectly to the genesis of many types of mineral deposits. Examples of the phenomenon include metal accumulations in living organisms, coals, black shales, crude oils and solid bitumens. CS interact with metals by virtue of their inherent reducing, acidic and chelating characteristics. After kerogen, and across the oil-to-solid bitumen continuum, bitumen is the most important CS that promotes metallization. The most familiar Phanerozoic example is the carbonate-hosted Pb–Zn ores of the Mississippi Valley-type, for which diverse reduction models have been proposed. At Elliot Lake, Canada, and in the Witwatersrand, South Africa, CS including kerogen and derivative bitumen (as petrographically defined) are closely related to the genesis of Precambrian uranium and gold mineralization. Occurrences of solid bitumen have long been recognized as important in petroleum exploration, and specific metal ratios (e.g., V/V + Ni and Mo/Mo + Cr) in viscous crudes and solid bitumens have been successfully applied in oil–source rock correlation studies. Typically, there is an increase in concentration of organometallic moities and complexes from crude oil to solid bitumen. Highly anomalous concentrations of elements (U, Pt, Au, V, and Ni) occur in solid bitumen, and numerous minerals (e.g., Zn, Pb, U, Ti, Au and Ag bearing species, illite, pyrite, REE-phosphates, quartz, barite, carbonates) develop authigenically in association with bitumen. Consequently bitumens are potentially important in the search for, and evaluation of, diverse mineral deposits. Where linked to the hydrothermal history of organic matter in sedimentary rocks, bitumen and organic acids generated during diagenesis help to prepare ground for mineral deposition, and like some oil field brines, are genetically associated with mineral deposits. The mineral-forming constituents are carried in an aqueous phase that accompanies bitumen during its emplacement. The mineral-associated constituents and organically bound metals in solid bitumens provide important clues in exploration geochemistry.