Eion M. Cameron

The Hemlo gold deposit, Ontario: A geochemical and isotopic study

Abstract The Hemlo deposit, near Marathon, Ontario, is one of the largest gold deposits in North America. It is stratiform within Archean metamorphosed volcano-sedimentary rocks. The main ore zone is composed of pyritic, sericitic schist, and massive barite. This is the first report of stratiform barite in the Archean of North America, but other occurrences have since been found west of Hemlo. The mineralization is substantially enriched in Au, Mo, Sb, Hg, Tl and V and lacks carbonate. Because of metamorphism and deformation of the body its genesis is uncertain. 87 Sr 86 Sr of .7017 for barite from the deposit is similar to that of the sedimentary barite west of Hemlo and to initial ratios of contemporaneous volcanic rocks. At the base of the main ore zone, barite with δ34S of +8 to +12%. was deposited with ~0%. pyrite. Upward, both barite and pyrite get isotopically lighter, with minimum values for pyrite, to −17.5%, in non-baritic schist forming the upper part of the ore zone. In drill section, Au grades correlate with the isotopic composition of pyrite. This, and the association of fractionated sulphide with sulphate, suggests that Au, pyrite and barite were deposited contemporaneously. The linked, asymmetric distributions of S minerals and isotopic distributions, which are continuous from section to section, and the isotopic similarity of the Hemlo and western barites are consistent with a syngenetic depositional model. Two sources for the S minerals are considered. In the first, exogenous sulphate from a restricted basin were partially reduced in a geothermal system to form 34S-depleted sulphide. In the second, the sulphate and sulphide are of magmatic-hydrothermal origin. Sulphate and fractionated sulphide are uncommon in Archean rocks, but one or both occur with unusual frequency in major Archean gold deposits. Hydrothermal fluids of moderately high ƒ O 2 , containing sulphate and permitting isotopic fractionation between oxidized and reduced S species, may have favoured the dissolution, transport and precipitation of Au.

Determination of the three-dimensional distributions of precious metals in sulphide minerals by laser ablation microprobe-inductively coupled plasma-mass spectrometry (LAMP-ICP-MS )

Precious metals are o f great financial importance and the form of their occurrence will dictate the economics of their recovery from ore. It has been postulated that “invisible” gold in sulphides occurs in solid solution, as particulates, or associated with micro-inclusions of other minerals in the host. This study applied the high spatial resolution (< 20 μm) and high sensitivity (sub-μg g−1 detection limits) of the laser ablation microprobe-inductively coupled plasma-mass spectrometer to identify both the form and concentration of gold in sulphides. Data are presented demonstrating the three-dimensional distribution of gold in minerals from Hemlo, Ontario, Canada, and Penjom Prospect, Kuala Lipis, Malaysia. Gold at μg g−1 ` levels was determined and, through its correlation with other trace elements, identified as either particulates or associated with micro-inclusions.

Biogeochemical exploration for gold in tropical rain forest regions of Papua New Guinea

Abstract Biogeochemical methods have been widely used for mineral exploration, particularly in boreal forests and semi-arid regions, but there have been fewer applications in tropical areas. This paper describes a biogeochemical method of exploration for Au in equatorial regions. After investigation of several plant species, Astronidium palauense, a small- to moderate-size tree, was found to have many suitable attributes. (1) It is widely distributed in the southwest Pacific, although its occurrence may be limited at elevations greater than 1000 m. (2) The tree is easy to identify and is sufficiently common (e.g., one tree per 100 m2 on Simberi and Lihir Islands, Papua New Guinea) for detailed sampling. (3) The outer bark is easy to obtain and the ashed bark reliably indicates Au concentrations in the substrate. (4) The root system reaches at least 4 m depth, allowing greater penetration than surface soil samples, which is important in volcanic terrains where geochemical targets may be buried by post-mineralization volcanic eruption or debris flows. (5) The areal distribution of the root system samples a large volume of soil (ca. 100 m3), which reduces the nugget effect for Au. (6) The ease of sampling and low weight of bark reduces the time and cost over soil surveys, for example 6 minutes per site compared with 15 minutes per 1 m deep soil. Bark can be ashed in the field, 200–500 samples in 2 to 4 days, then shipped for multi-element (Au, As + 32 elements) instrumental neutron activation analysis (INAA). Field tests on Simberi and Lihir Islands, PNG, show that biogeochemical surveys have a high level of reliability for identification of prospects.

Mercury in precambrian shales of the Canadian Shield

Abstract Mercury has been determined on 406 samples of Archean (⪢2.4 b.y.) shale from 153 localities in the Superior Province of the Canadian Shield and on 396 samples of shale of Aphebian (1.6–2.4 b.y.) age from 54 localities. The Archean samples, which are of mainly volcanogenic origin, and which commonly have been metamorphosed to lower greenschist grade, average 129 ppb Hg and have a median content of 86 ppb. The Aphebian shales are of varied lithology: from mature miogeosynclinal sediments to chemically immature shales associated with greywackes. They are variably metamorphosed up to lower greenschist facies and average 513 ppb Hg, with a median value of 408 ppb. The Hg content of the Archean and Aphebian shales has been related to that of twenty other elements previously determined on the samples. For the Archean shales, the geochemistry of Hg was dominated by the derivation of Hg and Zn from volcanic springs and subsequent precipitation of the two elements with the carbonaceous-sulphide fraction of contemporaneous muds. For the Aphebian shales, there is no similar correlation between Zn and Hg. Here the Hg is dominantly associated with the carbonaceous fraction of the rock. Significant loss of Hg during metamorphism of the Archean and Aphebian shales is discounted because the Hg was probably bound in thermally stable metal-organic compounds or as a minor component of sulphide minerals. For comparison, 48 samples of Palaeozoic shale from widely distributed localities across the eastern seaboard of Canada, and representing a wide variety of depositional environments, have been analysed for Hg. These shales average 42 ppb Hg, with a median content of 19 ppb. This average is similar to previous data for Devonian to Cretaceous shales from the Russian Platform. The Aphebian shales have a mean 12 times greater than the Palaeozoic samples and a median 21 times greater. The high levels of Hg in the Aphebian samples are consistent across the wide range of geological ages, lithologies, and geographic distance represented by the samples. It is suggested that this enrichment of Hg in early Proterozoic sediments was related to an increased rate of degassing of Hg from the earth. This increase may have been caused by a change in the sites of magma generation from shallow depths in the oceanic crust and mantle to deeper levels in the mantle—a change that followed upon the development of a thicker and more stable crust at the close of Archean time.

Genesis of proterozoic iron-formation: sulphur isotope evidence

Seven units of carbonaceous shale or sulphide-facies iron-formation have been sampled. They are associated with Proterozoic iron-formations that range in age from ~ 1.9 to ~2.5 Ga: Sokoman and Gunflint (Canada), Riverton (United States), Penge (South Africa) and Brockman (Australia). Sulphur isotope ratios have been determined on the sulphides removed from these shales by both physical and chemical means. The mean δ34S composition of the seven units varies between −4.9%. and +6.6%. and the sample variance is low within each unit. These distributions are more characteristic of hydrothermal sulphide than sulphide produced by biogenic reduction. This hydrothermal sulphide is believed to have originated from high temperature reduction of seawater sulphate and from magmatic sulphide. A model is suggested whereby this sulphide was exhaled into stratified anoxic/oxic basins. The sulphide and associated base metals were deposited in the reduced sediments beneath the anoxic waters, while some iron and manganese was deposited on oxygenated shelves. The data support, but do not prove, a hydrothermal exhalative origin for lower Proterozoic iron-formation.

Hydrogeochemical methods for base metal exploration in the northern Canadian shield

Abstract The use of lake waters for base metal exploration has been studied in the northern part of the Slave Geological Province of the Canadian Shield. The area is north of the treeline, within the zone of continuous permafrost, and, like most other regions of the Shield, has a high density of small lakes. A regional sampling of 1218 lakes established that less than 2 ppb (μg/l) Zn or Cu is typical of waters from unmineralized terrane. These samples had a median pH of 6.8 and a median specific conductivity of 19.5 μmhos. Lake waters were also taken from the areas surrounding five massive sulphide occurrences: High Lake, Canoe Lake, Takiyuak Lake, Hackett River and Agricola Lake. In all cases there are unambiguous anomalies for Zn. Anomalies are also present for Cu, but are less intense and extensive. This difference between the two elements is related to the superior mobility of Zn in surface waters and its more consistent presence as a major constituent of massive sulphides. A water sampling apparatus has been developed and tested on a light turbine helicopter. Using this, thirty sites may be sampled each hour when sampling at a density of 1 site per 2.8 km 2 . Measurement of pH, conductivity and water temperature are recorded in the helicopter during sampling. A number of factors have been investigated that may influence the utility of lake water sampling for base metal exploration: 1. (1) Seasonal variability, while present to moderate degree, is unlikely to hinder application of the method. 2. (2) For the size of lakes sampled (2 km 2 or less), elements are homogeneously distributed across the lake surface during the ice-free season. During the initial period of break-up there are marked variations in element content around the ice-free lake margin. Sampling during this period may help define the source of metals for anomalous lakes. 3. (3) Study of sample preservation suggests that mobile elements, such as Zn, that are stable in solution within lakes, are also relatively stable when untreated water is stored in plastic bottles. 4. (4) Care must be taken to avoid contamination of the samples, particularly from the bottle. The areal extent of lake water base metal anomalies appears to be less than equivalent lake sediment anomalies. Thus for wide-interval, regional geochemical reconnaissance, lake sediment sampling is the method of choice. Lake waters are an appropriate medium for detailed exploration of areas of interest, such as volcanic belts. For this application, the principal attractions are rapid sampling rates, and hence low costs, high contrast anomalies, and a uniform sampling medium.

Carbonate sedimentation during the Archean

Abstract Archean (> 2400 m.y.) sediments of the Canadian Shield are principally composed of greywacke and shale of flysch facies that are preserved within slightly metamorphosed “greenstone” belts. These Archean sedimentary sequences contain only trivial proportions of carbonate rocks compared to those of Proterozoic and Phanerozoic age. Archean sediments from other parts of the world appear to have a similarly low proportion of limestone and dolomite. One possible explanation for this scarcity of carbonate rocks is that calcium (and magnesium) did not readily pass into solution upon weathering of primary rocks during the Archean. Another possible explanation is that calcium carbonate (or dolomite) was deposited as a minor component of Archean shales and greywackes rather than as discrete beds of carbonate minerals. Both of these explanations are discounted by chemical analyses of some 406 widely scattered samples of Archean shale from the Superior province of the Canadian Shield. These analyses indicate that mobile elements were removed by weathering during the Archean in approximately the same relative proportion as during later periods; and that the calcium thus liberated has not been redeposited with the shales. Explanations based on deposition of carbonates as shelf sediments and subsequent removal by erosion, or on an increased solubility of Ca 2+ in the Archean oceans, are also discounted. It appears that the thin crust of Archean time produced no extensive stable shelf or miogeosynclinal environments, which were important sites for carbonate deposition during the Proterozoic and Phanerozoic. It is suggested that carbonate deposition during the Archean took place largely in deep ocean basins, and that the mechanism for precipitation may have been photosynthetic reactions by algae living near the surface of the oceans. It seems likely that there were extensive oceanfloor deposits of carbonates during the Archean which were later resorbed into the mantle during ocean-floor spreading.

Archean sulphur cycle: Evidence from sulphate minerals and isotopically fractionated sulphides in superior province, Canada

In sedimentary rocks of Archean age there are few occurrences of sulphate minerals and most sulphides have δ34S-values close to 0%o. Since oxidation-reduction reactions are required for isotopic fractionation, this evidence has been taken to indicate a generally reduced surface environment, including an Archean ocean with only trace quantities of sulphate. There are, however, some exceptions to this generalization, one of the most notable being sulphides with a range in PS of + 30%o associated with iron formation near Wawa, Ontario. This has caused some workers to propose an alternative scenario with a major reservoir of sulphate in the Archean oceans and biogenic reduction of this sulphate by 2.7 Ga. We report here on other occurrences of strongly fractionated sulphides and sulphate minerals in the Wawa Subprovince. These are the Hemlo gold deposit, with pyrite ranging in composition from −17.5 to + 12.6‰, plus barite and anhydrite; west of Hemlo, a 6-km-long baritic horizon in metasedimentary rocks with 34S for pyrite to −12‰; gold mineralization at Heron Bay, also west of Hemlo, that contains barite, anhydrite and 34S-depleted pyrite; and the Geco Mine, which, with a satellite deposit, are the only Archean massive sulphide deposits in the Canadian Shield known to contain anhydrite. Considering the scarcity of these evidences elsewhere in terrane of this age, it is remarkable that all occur within a distance of 150 km. Suitable tectonic conditions appear to have permitted sulphate to become concentrated in this region, permitting isotopic fractionation between sulphate and sulphide. A restricted reservoir of sulphate is reflected in a progressive change in σ34S for pyrite at some of the localities. The data are consistent with a model of an Archean ocean in which sulphate was maintained at a low concentration by extensive interaction with reduced mantle material. Only in basins with restricted access to the ocean could this oxidized species accumulate. All of the occurrences of fractionated sulphide and sulphate minerals in the Wawa Subprovince are associated with hydrothermal mineralization. In some cases, the isotopic fractionation apparently occurred in a hydrothermal system; in other cases the cause of the fractionation is equivocal; it may be due to either biological or to inorganic oxidation-reduction reactions. We also examined graphitic shale from the region. Pyrite from these samples does not show the depletion in 34S that is commonly found in sulphide of biological origin.

Geochemical dispersion in mineralized soils of a permafrost environment

Abstract The Agricola Lake massive sulphide body occurs in metavolcanic rocks of Archean age in the northwestern Canadian Shield. The mineralization contains Zn, Cu, Pb, Ag and Au and, in addition, there are elevated contents of As and Hg. The area lies within the zone of continuous permafrost. The distribution of elements in soils in the vicinity of the mineralization shows the influence of three independent dispersion processes — glacial, solifluction and hydromorphic — superimposed on the primary bedrock distribution. Oxidation of sulphides has been active in the past and continues to this day. In the absence of large amounts of carbonate minerals, this has produced an acidic weathering environment. The most mobile of the elements studied — Zn — has been largely removed from the mineralized soils and dispersed a considerable distance down drainage. Au, Hg, Pb and Ag are relatively immobile and are retained in the soils overlying mineralization or in nearby, glacially dispersed detritus. Fe, Cu and As are of intermediate mobility and are deposited in the proximal portion of the drainage system. The estimated order of mobility is: Zn > Cu > Fe > As > Ag > Pb > Hg > Au. Glacial action dispersed oxidized detritus, already depleted in mobile elements. Material precipitated along the proximal portion of the drainage system shows an association of As and Cu with iron oxides and Ag and Hg with manganese oxides. In addition to the elements discussed above, Mg in soils indicates the alteration zone underlying the massive sulphide and Ca outlines the soils subject to acidic leaching.

Geochemical dispersion in lake waters and sediments from massive sulphide mineralization, Agricola Lake area, Northwest Territories

Abstract The Agricola Lake area lies within the tundra and is underlain by continuous permafrost. Archean metavolcanic rocks are host to massive sulphide mineralization, which contains Zn, Cu, Pb, Ag, Au, As, Cd and Hg. This mineralization is being actively oxidized, producing acidic waters enriched in a number of metals. The elements Pb, Ag and Hg are immobile in the surface environment and are largely retained in the soils near the mineralization. Zn, Cd and Cu are mobile and are dispersed in quantity far along the lake-stream system draining the mineralization. Arsenic is relatively immobile, but it has so wide a primary distribution, particularly in metasedimentary rocks overlying the volcanics, that it too is anomalous in sediments throughout the drainage. Ni and Co, derived from these metasedimentary rocks, and as mobile as Zn and Cd, are also strongly anomalous in lake sediments. Fixing of the mobile elements in sediments appears to be determined by their adsorption on iron oxides, Cu being adsorbed at lower pH values and closer to the mineralized source, than Zn, Cd, Ni or Co. The use of nearshore and centre-lake sediments for reconnaissance-level geochemical exploration is compared. The latter are a more homogeneous, finer-grained sampling medium. Dispersion of mobile indicator elements from their source in centre-lake sediment is greater than for nearshore sediment. This is because nearshore sediments are essentially subaqueous soils, that are not produced by lacustrine sedimentation. Adsorption of metals on nearshore material takes place in situ, so that dispersion trains can be no longer than that of the waters in contact. In the case of centre-lake sediments, it is suggested that metals are adsorbed on fine-grained particulates, which travel downstream before settling in the centre-lake bottoms. Waters are the most convenient sampling medium for more detailed levels of exploration. Surface waters are homogeneous within any one lake and show only moderate variation in composition throughout the summer ice-free season. Lake waters that are not derived from mineralized areas are very pure, with dissolved solids contents of 10 ppm or less.