Duluth Complex Research Articles

Metamorphic grade and porosity and permeability controls on mafic phyllosilicate distributions in a regional zeolite to greenschist facies transition of the North Shore Volcanic Group, Minnesota

The 8-km-thick, Keweenawan metabasaltic, North Shore Volcanic Group in Minnesota shows a continuous zeolite to greenschist facies progression through five zonal divisions. Mafic phyllosilicates are ubiquitous and a well-developed smectite-corrensite-chlorite transition is observed. This sequence provides an excellent framework in which the influence of regional (thermal) as opposed to local scale (lithology: porosity and permeability) factors on this transition, factors that are actively debated, can be comprehensively documented. On a regional scale, at lowest grades (zones 1 and 2; thomsonite-scolecite and heulandite-stilbite assemblages, respectively), tri-smectitic phyllosilicates dominate and, with increase in grade to zone 3 (laumontite-albite), there is a marked step and compositional gap to corrensite. With further increase in grade to zones 4 (laumontite-albite ± prehnite ± pumpellyite) and 5 (epidote-albite-actinolite), chlorite becomes dominant. Although a spread in the X-ray diffraction and microprobe data suggests that the corrensite to chlorite transition is continuous rather than stepwise, analytically distinct corrensite and chlorite in one flow suggest a clear compositional gap between these two phases. Overall, the data offer some support for a discontinuous smectite-corrensite-chlorite transition. Above the lowest grade (zones 1 and 2; approximately 150°C), the phyllosilicates vary on a local, within-flow scale related to morphological contrasts in flow lithology. Smectite is dominant in the massive flow units (low porosity and permeability), whereas corrensite or chlorite is dominant in the amygdaloidal flow tops (high porosity and permeability). These changes provide clear indication that, once above a threshold of approximately 150 °C, fluid/rock ratios have a strong control on the metamorphic process at low grade. Similarities in XMgO between smectite, corrensite, chlorite, and XMgO of the whole rock show that even though smectite is a relict phase remaining in massive zones, it continues to undergo crystal chemical change with increasing grade. However, contrasts between AlIV in the same minerals are provisional indicators that other factors, perhaps related to kinetic features linked to advective and diffusive modes of fluid transport, may be involved.

Origin of evolved magmas in the Midcontinent rift system, northeast Minnesota: Nd-isotope evidence for melting of Archean crust

The North Shore Volcanic Group (NSVG) of northeast Minnesota is a thick (9 km) sequence of plateau volcanic rocks that constitutes an important part of the Midcontinent rift system. This volcanic sequence is unique among the Midcontinent rift lavas, because it is composed of up to 25% rhyolite flows. We have analyzed Sm- and Nd-isotope compositions of 20 of the largest rhyolite and icelandite flows from the NSVG and seven comparably sized granophyres in the subjacent Duluth and Beaver Bay complexes. The lavas vary in composition from primitive basalt and basaltic andesite to icelandite and rhyolite, with a bimodal distribution. The rhyolites have much lower initial εNd values (−2 to −15, most samples < −10) than either the icelandites (0 to −6) or granophyres (0 to −8). Most rhyolites cannot be related to either the icelandites or more mafic magmas by simple fractionation, but rather have been produced by melting and assimilation of older, evolved crust. We suggest that the bimodal magmatism in the NSVG, and probably throughout the Midcontinent rift, has been produced by two fundamentally different processes. The bulk of the magmatism is basaltic; magmas originate in the mantle and migrate through the lithosphère with minor compositional change. Assimilation and fractional crystallization occur to varying degrees in the crust and, in some cases, produce icelandites, some small-volume rhyolites, and the granophyres, with Nd compositions dominated by the mantle component. The melting that produced the large-volume rhyolites is the result of a multistage process induced by these mantle-derived magmas that pond within the crust. This process appears to occur during a period of slowed extension and causes widespread heating and eventually localized extensive melting of the crust.

Re-Os isotopic compositions of Midcontinent rift system picrites: implications for plume – lithosphere interaction and enriched mantle sources

Picrites and tholeiites from the Mamainse Point Formation, a 5.3 km thick section of Keweenawan (1100 Ma) volcanic and sedimentary fill on the eastern flank of the central portion of the Midcontinent rift system, contain a nearly continuous record of rift magmatic activity. Picrites occur primarily in the lowermost two units of the formation. In this study, they are compared to rarely exposed, slightly older Keweenawan basalts from the North Shore Volcanic Group and the Powder Mill Group to constrain mantle source compositions during early phases of rift magmatic activity. The most primitive picrites analyzed have low Re content (0.069–0.18 ppb), high Os content (0.8–2.1 ppb), and low 187Re/188Os (0.28–1.18). A Re–Os isochron with an age of 1128 ± 54 Ma and an initial 187Os/188Os of 0.1267 ± 0.0013 (γOs = +5.7) was obtained from a 24-point isochron on all but two analyzed samples. The Re–Os data, regressed separately for the older basalts, and the groups 1 and 2 samples from the Mamainse Point Formation, have barely resolvable initial 187Os/188Os that decrease up-stratigraphy from initial γOs(1100) of +12.2 to +6.2 and +4.2, respectively, and couple with changes in initial Nd isotopic composition. These data can be explained by mixing of melts of an enriched mantle plume and unradiogenic continental lithospheric mantle. A radiogenic initial Os isotopic composition (γOs of +8 or higher) for the Keweenawan plume marks the first known appearance of demonstrably radiogenic plume-derived magmas on Earth. Plume-derived magmas with radiogenic Os signatures are more common later. The radiogenic Os signatures of Keweenawan plume magmas may mark the appearance of melts derived from mantle containing recycled slab components from late Archean subduction.

Geochronology of the North American Midcontinent rift in western Lake Superior and implications for its geodynamic evolution

Volcanism in the Midcontinent rift system lasted between 1108 and 1086 Ma. Rates of flood-basalt eruption and subsidence in the western Lake Superior region appear to have been greatest at the beginning of recorded activity (estimated 5 km/Ma subsidence rate at 1108 Ma) and rapidly waned over a period of 1–3 Ma during a magnetically reversed period. The age of the paleomagnetic polarity reversal is now constrained to be between 1105 ± 2 and 1102 ± 2 Ma. A resurgence of intense volcanism began at 1100 ± 2 Ma in the North Shore Volcanic Group and lasted until 1097 ± 2 Ma. This group contains a ca. 7 Ma time gap between magnetically reversed and normal volcanic sequences. A similar disconformity appears to exist in the upper part of the Powder Mill Group. The average subsidence rate during this period was approximately 3.7 km/Ma. Latitude variations measured from paleomagnetism on dated sequences indicate that the North American plate was drifting at a minimum rate of 22 cm/year during the early history of the Midcontinent rift. An abrupt slowdown to approximately 8 cm/year occurred at ca. 1095 Ma. These data support a mantle-plume origin for Midcontinent rift volcanism, with the plume head attached to and drifting with the continental lithosphere. Resurgence of flood-basalt magmatism at 1100 Ma may have been caused by extension of the superheated lithosphere following continental collision within the Grenville Orogen to the east.

The influence of country-rock assimilation and silicate to sulfide ratios (R factor) on the genesis of the Dunka Road Cu – Ni – platinum-group element deposit, Duluth Complex, Minnesota

The Dunka Road deposit is one of several Cu – Ni – platinum-group element (PGE) sulfide occurrences found along the northwestern margin of the Duluth Complex, where the host troctolitic rocks are in contact with metasedimentary rocks of the Animikie Group. Magma contamination through assimilation of sulfidic argillaceous country rocks is generally recognized as having played a key role in the genesis of the mineralization. Three main types of disseminated sulfide mineralization have been identified within the Dunka Road deposit: (i) norite-hosted sulfides, (ii) troctolite-hosted sulfides, and (iii) PGE-rich sulfide horizons. The norite-hosted sulfides are found either adjacent to country-rock xenoliths or near the basal contact. The troctolite-hosted sulfides form the bulk of the deposit, and occur throughout the lower 250 m of the intrusion. The PGE-rich sulfide horizons are typically localized directly beneath ultramafic layers. The composition of the different types of sulfide occurrences is modelled using Cu/Pd ratios. It is shown that each type results from the interplay of two main parameters, namely the degree of magma contamination and the silicate magma to sulfide melt ratio (R factor). The norite-hosted sulfides formed at low R factors and high degrees of contamination, as expressed by their PGE-depleted nature, low Se/S ratios, and elevated content in pyrrhotite and arsenide minerals. The troctolite-hosted sulfides formed at moderate R factors and small degrees of contamination, as shown by their moderate PGE content and mantle-like Se/S ratios. Finally, the PGE-rich sulfide horizons are modelled using elevated R factors from an uncontaminated parental magma, which is substantiated by their elevated noble metal content and Se/S ratios, and low pyrrhotite to precious metal sulfide ratio.

Magma dynamics at the base of an evolving mafic magma chamber: Incompatible element evidence from the Partridge River intrusion, Duluth Complex, Minnesota, USA

A characteristic feature of the Partridge River intrusion of the Keweenawan Duluth Complex is the approximately fivefold to ninefold increase in the concentrations of incompatible elements in the lower zone compared with cumulates stratigraphically higher. The concentrations of incompatible elements decrease from the lower zone upward to steady state values, which is ascribed to variations in the proportions of trapped liquid rather than variable degrees of fractional crystallization of a single parental magma. The calculated average composition of trapped liquid using our algorithm is similar to typical Keweenawan low-alumina, high TiP basalts associated with the Duluth Complex but is different from the leading edge ferrodioritic liquid quenched in the chilled margin of the intrusion. This difference suggests that the chilled margin does not represent the original (parental) magma composition from which the whole intrusion solidified, and that the enrichment of incompatible elements may be related to the local flotation of magmatic suspensions. To test the latter hypothesis numerically, we have used heat-mass transfer models, assuming a sheet-like magma chamber, to calculate the parameters of the model that best reproduce the observed distribution of incompatible elements in a mush zone at the base of the Partridge River intrusion. The results indicate that a mush zone enriched in the incompatible elements is produced if the velocity of movement of the lower solidification front into the magma body was less than the floating velocity of the bulk crystal mush. The dynamic parameters that best reproduce the observed distribution of incompatible elements include a magma emplacement pressure of 2 kbar, critical crystallinities of 50–68% in the mush zone from which the liquid is being expelled, and an emplacement temperature of ∼ 1160°C for the initial magma.

Mineralogic and Oxygen Isotopic Studies of Open System Magmatie Processes in the South Kawishiwi Intrusion, Spruce Road Area, Duluth Complex, Minnesota

The South Kawishiwi intrusion, located along the western margin of the Duluth Complex, Minnesota, is one of several composite intrusions that are found in the Complex. The Duluth Complex is the principal exposed plutonic portion of the 1-1 Ga Midcontineni Rift system. In the Spruce Road area the South Kawishiwi intrusion is divided into seven distinct units that are part of the broader South Kawishiwi Troctolite Series defined by Severson (Tech. Rep. NRRI/TR-91/13a, Natural Resources Research Institute, University of Minnesota, Duluth, 1994). Units may be characterized as follows: Unit I—basal accumulation of heterogeneous gabbro, troctolite, and norite; Unit II—norite with abundant inverted pigeonite; Unit III—troctolite and olivine gabbro with local oxide-rich layers; Unit IV—melatroctolite, troctolite, olivine gabbro; Unit V— increased plagioclase abundance in troctolites and leucocratic troctolites; Unit VI—strongly altered troctolite; Unit VII— similar to Unit V, troctolite and leucocratic troctolite. Country rocks in the Spruce Road area are granodiorite to quartz monzonite of the Archean Giants Range Batholith. Sutfide mineralization, consisting of 1-5 vol. % of disseminated pyrrhotite, cubanite, chalcopyrite andpentlandite, occurs in Units I, II, III, and VI. Oxygen isotopic analyses indicate that Unit II has experienced extensive crustal contamination. 5O values of Unit II range from 69 to 7-l%o and are 0 enriched compared with values of 5-1-6-8%o found in other units. Silica contamination is indicated based not only on 5' 0 values, but also by the predominance of orthopyroxene in the unit. Possible high-' 0 contaminant rocks include the Giants Range Batholith and pelitic rocks of the Lower Proterozoic Virginia Formation or Biwabik Iron Formation. Mass balance computations suggest that units in the Spruce Road area may be related through varying degrees offractionation of a high-Al, olivine tholeiite magma. Modeling of trace element concentrations and variations in mineral chemistry suggest that discontinuities within the major units developed by in situ boundaryAayer equilibrium crystallization of solidification zones ~ 20—50 m in thickness, followed by recharge of fresh magma. Upward enrichment of incompatible elements, olivine Fa content, and plagioclase Ab content may be effectively explained by this process. 5 0 values of uncontaminated rock types are strongly correlative with modal mineralogy, and can also be modeled by boundary-layer fractionation. A parental magma 5 0 value of ~6-3%o is calculated for Unit VII based on olivine and plagioclase values, and is similar to that of several other large, layered mafic intrusives.

EQUILBRIUM MODELING OF TRACE METAL TRANSPORT FROM DULUTH COMPLEX ROCKPILE

Geochemical modeling was used to predict weathering processes and the formation of trace metal-adsorbing secondary phases in a waste rock stockpile containing Cu-Ni ore mined from the Duluth Complex, MN. Amorphous ferric hydroxide was identified as a secondary phase within the pile, from observation and geochemical modeling of the weathering process. Due to the high content of cobalt, copper, nickel, and zinc in the primary minerals of the waste rock and in the effluent, it was hypothesized that the predicted and observed precipitant ferric hydroxide would adsorb small quantities of these trace metals. This was verified using sequential extractions and simulated using adsorption geochemical modeling. It was concluded that the trace metals were adsorbed in small quantities, and adsorption onto the amorphous ferric hydroxide was in decreasing order of Cu > Ni > Zn > Co. The low degree of adsorption was due to low pH water and competition for adsorption sites with other ions in solution.

Evidence for immiscibility in Ti-rich garnet in a calc-silicate hornfels from northeastern Minnesota

Calcic garnet from a calc-silicate hornfels from the contact-metamorphic aureole of the Duluth complex, northeastern Minnesota, has Ti02 compositions ranging from 0 to 13 wt%. The concentrations of Fe, Mg, AI, Ti, and Si are highly correlated, and 93% of the total variance can be ascribed to three principal components formed by the additive component grossular and several exchange components. The first is a combination of the exchangesTiSL I' MgTiAL2, and MgCa_l. The second represents the substitutions of Al for Fe and of Fe for Ca. The third represents the substitution of Cr for AI. Most of the variance occurs along a linear trend from Ca3AI1.72Feo.28Si3012 to Ca3M&.33Alo.67Feo.67 Ti1.0Si2.33012'Within this trend, a gap in composition occurs between 6 and 12 wt% Ti02. Separate grains of Ti-rich and Ti-poor garnet occur in contact with sharp boundaries. No exsolution lamellae were observed, however. Immiscibility appears to be blocked by the presence of Cr.

The distribution of platinum-group elements, nickel, copper, and gold in the Muskox layered intrusion, Northwest Territories, Canada

The Muskox intrusion was emplaced into a rift in the northwestern Canadian Shield at about 1270 Ma. This rifting event was associated with extensive flood basalt magmatism (Coppermine River basalt) and is attributed to the development of a mantle plume centered some 400 km north of the intrusion. Nickel-copper sulfides occur at the margins of the intrusion. The geologic setting of these sulfides is similar to that of a number of Ni-Cu sulfide deposits, such as those found in the Noril'sk-Talnakh area, the Duluth Complex, the Insizwa Complex, and the Cape Smith belt. Some of the marginal sulfides at Muskox are similar in composition to the Cu-PGE-rich Noril'sk-Talnakh sulfides and contain 16 percent Cu and up to 200ppm Pt + Pd + Au. However, most of the Muskox marginal sulfides were found to contain 3 to 10 percent Ni + Cu and intermediate levels of platinum-group elements (PGE) (1-10 ppm) similar to the Minnimax deposit of the Duluth Complex. The PGE content of the sulfides increases from the margins to the interior of the intrusion. This increase could be due to an increase in the ratio of silicate to sulfide liquid (R factor) during the crystallization of the intrusion. The high percentage of delta 34 S per mil of the sulfides indicates that the S in the sulfides was derived from the country rocks. Therefore, the amount of sulfide liquid that formed could have been controlled by the availability of S, which in turn was controlled by the distance from the contact with the country rock. Disseminated sulfides occur in association with a chromite-rich orthopyroxenite in the layered series of the Muskox intrusion. These sulfides have a strata-bound character similar to classic PGE reefs, such as the Merensky reef of the Bushveld intrusion. The sulfides from the Muskox reef unit contain only 1 to 10 ppm PGE, an order of magnitude less than sulfides from classic PGE reefs. The lower PGE content of the Muskox reef sulfides is attributed to a lower R factor; approximately 1,000 at Muskox versus greater than 10,000 for classical reefs. The sulfide-bearing rocks have higher Cu/Ni, Pd/Ir, Pd/Pt, and Rh/Ir ratios than sulfide-poor rocks. This is because the sulfides are enriched in Cu, Pd, and Rh to a greater degree than in Ni, Pt, and Ir. The difference in the degree of enrichment for these metals could be the result of Ni, Pt, and Ir having lower partition coefficients in sulfide liquid than Cu, Pd, and Rh; contamination of the Muskox magma with a partial melt of country rock; or a combination of these factors.

Interactions of mixed volatile-brine fluids in rocks of the southwestern footwall of the Duluth Complex, Minnesota; evidence from aqueous fluid inclusions

Interactions of mixed volatile-brine fluids in rocks of the southwestern footwall of the Duluth Complex, Minnesota; evidence from aqueous fluid inclusions

Genesis of the Marathon Cu-platinum-group element deposit, Port Coldwell alkalic complex, Ontario; a Midcontinent rift-related magmatic sulfide deposit

The Marathon copper and platinum-group element (PGE) deposit is hosted by the Two Duck Lake intrusion, located at the eastern margin of the Port Coldwell alkalic complex, northwestern Ontario. The Two Duck Lake intrusion is subdivided into an upper layered zone and a lower unlayered zone that consists of olivine gabbro and pegmatite with trace amounts of granophyre. The sulfides are hosted in the lower zone which is referred to here as Two Duck Lake gabbro. Petrography, mineral-chemistry, and geochemical data imply that Two Duck Lake gabbro crystallized in situ from a plagioclase crystal mush. There is no geochemical evidence for metasomatic alteration of the gabbro. The sulfides are disseminated in gabbro and pegmatite and consist of approximately equal proportions of chalcopyrite and pyrrhotite and minor amounts of pentlandite and pyrite. Samples of Two Duck Lake gabbro exhibit a significant correlation between sulfur and Ni, Ir, Rh, Pt, Pd, Au, and Cu. Further, the Pd/Ir and Pd/Ni ratios increase with increasing concentrations of Rh, Pt, Pd, Au, and Cu. Consequently, bulk sulfide compositions are best explained by the separation of a sulfide liquid from a gabbroic magma. This origin is contrary to that proposed by previous workers. They presented detailed petrographic and mineral-chemical data that imply inter-action of Cl-bearing hydrous fluids and sulfides at low temperatures. In the light of our study, we interpret these observations to indicate that elements migrated over very short distances and that bulk sulfides were not enriched with Cu and PGE, as previously proposed. A model for the genesis of the Marathon deposit is proposed.Many characteristics of the Marathon deposit bear a striking resemblance to observations reported in disseminated sulfide deposits at the base of the Duluth Complex, Minnesota. The features include mineral textures, abundances, and compositions; crystallization paths for the host gabbroic bodies; silicate-sulfide associations; trace element trends; and chalcophile element fractionation trends. The similarities imply that the Marathon deposit and deposits in the Duluth Complex formed by analogous processes.

Comparison of Duluth Complex Rock Dissolution in the Laboratory and Field

Comparison of Duluth Complex Rock Dissolution in the Laboratory and Field

Simulation of primary phase relations and mineral compositions in the Partridge River intrusion, Duluth Complex, Minnesota: implications for the parent magma composition

In order to describe the composition and crystallinity of the initial (parental) magma of the Partridge River intrusion of the Keweenawan Duluth Complex, and thereby understand the mode of emplacement and solidification of the intrusion, we have applied a numerical simulation technique called geochemical thermometry (Frenkel et al. 1988). The parental magma was a low-alumina, high-Ti-P olivine tholeiite similar to typical Keweenawan low-alumina, high-Ti-P basalts associated with the Duluth Complex and from the nearby Portage Lake area of the Lake Superior region. The parental magma was emplaced as a crystal-liquid suspension, followed by chilling of an evolved, leading edge ferrodioritic liquid in the basal zone of the intrusion. The conditions of emplacement at the present crustal location were ∼1,150°C, 2 kbar, and f O 2 slightly above the wustite-magnetite (WM) buffer. The main differentiation process after emplacement was the sorting and redistribution of plagioclase and olivine crystals on a local scale accompanied by less efficient convection and minor settling of olivine. Calculated crystallization sequence for the parental magma is olivine+plagioclase (1,240°C)→olivine+plagioclase+magnetite (1,146°C, WM+0.5)→olivine+plagioclase+magnetite+augite (1,140°C, WM+0.5). The calculated compositions of the cumulus olivine and plagioclase in equilibrium with the parent magma at 1,150°C are Fo66.7±1.1 and An64.5±2.5, respectively, and are similar to the estimated average composition of primary olivine (Fo69.1±2.8) and the average composition of plagioclase core (An66.3±2.8) measured in drill core samples through the intrusion (Chalokwu and Grant 1987).

Precise U‐Pb ages of Duluth Complex and related mafic intrusions, northeastern Minnesota: Geochronological insights to physical, petrogenetic, paleomagnetic, and tectonomagmatic processes associated with the 1.1 Ga Midcontinent Rift System

Precise resolution of the timing of igneous activity is crucial to understanding the dynamic processes associated with continental rifting. Mafic intrusive rocks constitute a major portion of the exposed 1100 Ma (Keweenawan) Midcontinent Rift system in northeastern Minnesota; however, prior to this study, geochronological data were insufficient to allow rigorous interpretation of intrusive histories and their relationships to extrusive suites. Eight anorthositic and gabbroic intrusives were chosen to represent both the temporal and spatial ranges of plutonic activity that formed the Duluth Complex and related mafic intrusions. U‐Pb isotopic analyses from zircons and baddeleyites result in U‐Pb concordant ages with little or no ambiguity introduced by inherited components, Pb loss or common Pb. The earliest Keweenawan plutonism exposed in Minnesota occurs along the northeastern flank of the Duluth Complex as a series of layered gabbros (Nathan's layered series) emplaced at 1106.9 ± 0.6 Ma. This sequence of gabbro sheets shares temporal, spatial, and compositional similarities with the nearby Logan sills in Ontario. Four Duluth Complex anorthositic and troctolitic series samples from widely separated areas have unresolvable ages between 1099.3 ± 0.3 and 1098.6 ± 0.5 Ma, indicating a very short duration for peak intrusive activity (0.5–1 m.y.). The unresolvable ages between anorthositic and troctolitic plutons suggest that these two magma series are more closely related than previously modeled and argue strongly for the need to reexamine their fundamental petrogenetic relationships. These dates also imply that the major reverse‐to‐normal magnetic polarity switch, used throughout the rift system as an important correlation tool, occurred prior to 1099 Ma. This date is several million years earlier than previously suspected and emphasizes the need for further paleomagnetic and geochronological data from the overlying volcanics. Much of the hypabyssal intrusive suite within the volcanic pile overlying Duluth Complex plutons may be significantly younger than the main pulse of plutonic activity. Two hypabyssal bodies, the Sonju Lake intrusion and gabbro at Silver Bay, were emplaced at 1096.1 ± 0.8 Ma and 1095.8 ± 1.2 Ma, respectively. Dates reported here and in previous studies support the concept of episodic tectonomagmatic rift development where magmatism was apparently concentrated in episodes of short duration (<1–3 m.y.) interspersed with longer hiatuses (2–8 m.y.).

Chlorine as an exploration guide for the platinum-group elements in layered intrusions

Abstract Recent theoretical and experimental work has suggested that Cl may be important for the transport of the platinum-group elements (PGE). A survey of the halogen geochemistry of apatite in layered intrusions, including new data from four intrusions from the Yilgarn block of Western Australia (the Jimberlana Intrusion, the Windimurra complex and the Mount Thirsty and Ora Banda Sills) and the Duluth Complex in Minnesota, shows that apatite from most intrusions is characteristically fluorian, with endmember chlorapatite components typically less than 20 mole percent. In contrast, the cumulate sections below both the J-M reef of the Stillwater Complex and the Merensky Reef of the Bushveld Complex contain Cl-rich apatite approaching endmember chlorapatite. Although the Stillwater and Bushveld intrusions are not the only intrusions that contain zones of PGE enrichment, the chlorapatite association suggests that intrusions with Cl-rich assemblages are more likely than not to contain economic PGE deposits. Furthermore, this association is consistent with theoretical considerations supporting chlorine as important for the transport of the PGE. It is suggested that the halogen geochemistry, as recorded in primary halogen-bearing minerals, may be a useful exploration guide for layered intrusions that may contain above-average grade deposits of these elements.

Regional and local patterns of low‐grade metamorphism in the North Shore Volcanic Group, Minnesota, USA

Abstract The North Shore Volcanic Group in northern Minnesota is part of the Middle Proterozoic Keweenawan sequence, one of the largest plateau lava provinces in the world. The primary geochemistry of the basalts suggests that volcanism occurred in an intracontinental rift environment. The subaerial lava flows, mainly amygdaloidal olivine tholeiites and tholeiites, have undergone low‐grade metamorphism from zeolite to lower greenschist facies. On the basis of alteration phases replacing the primary magmatic minerals, infilling amygdales and veins, and replacing secondary minerals, the following zones have been distinguished: (1) thomsonite‐scolecite‐smectite, (2) heulandite‐stilbite‐smectite, (3) laumontitechlorite‐albite, (4) laumontite‐chlorite‐albite ± prehnite ± pumpellyite and (5) epidote‐chlorite‐albite ± actinolite zone.In addition to the overall zonation based on mineral parageneses, zonations in the composition of the Ab content of the newly formed albite replacing primary Ca‐rich plagioclase and of the newly formed mafic phyllosilicates are observed within the sequence and within single flows. Mafic phyllosilicates in the upper part of the sequence (mainly smectites and mixed‐layer smectite/chlorites) display high Si and Ca + Na + K contents, whereas in the lower part of the sequence the amounts of Si and Ca + Na + K are markedly lower (mainly chlorites and mixed‐layer chlorite/smectites). Similar zonations are observed within the individual flows. The albite content of the newly formed plagioclase is highest, and the Si and Ca + Na + K content of the phyllosilicates lowest in the amygdaloidal flow top while the opposite is true for the massive flow interior.The above features suggest that the overall pattern is one of burial‐type metamorphism associated with extension in the rift setting. In detail, the mineral assemblages are controlled not only by the stratigraphic position but also by the flow morphology controlling permeability whose effect on the assemblages is most pronounced in the stratigraphically upper parts. This suggests that at the first stages of alteration (lowest grade) the patterns of fluid flow were important effects in controlling the assemblages. At greater burial depth, assemblages are more homogeneous, perhaps representative of a more even and pervasive flow pattern.Using the observed assemblages at face value to define grade and/or facies, different conditions would be assigned within the different morphological flow portions. Thus at low‐grade metamorphic conditions it is essential to integrate assemblages from different morphological flow portions in order to define satisfactorily the overall metamorphic conditions.

Hydrothermal alteration in the Babbitt Cu-Ni deposit, Duluth Complex; mineralogy and hydrogen isotope systematics

Hydrothermal alteration at the Babbitt Cu-Ni deposit of the Duluth Complex is widespread but highly variable in terms of intensity. Alteration styles range from the filling of microfractures in plagioclase and olivine or minor alteration along grain margins to more intense alteration in the form of macroscopic fracture fillings or massive replacement of primary minerals with or without well-developed fractures. Chlorite occurs as a macroscopic fracture filling, within microveinlets in plagioclase, and as an alteration of pyroxene, biotite, and rarely olivine along grain margins or fractures. Mats of chlorite are intergrown with amphibole, epidote, calcite, and prehnite as part of massive replacement of calcic plagioclase. Greenalite is found in areas of intense alteration, particularly in the vicinity of sulfide mineralization. Iron-rich serpentine occurs along fractures in olivine or as a more massive replacement of olivine in picritic units. Serpentine composition is related to that of the parent olivine, as well as to the degree of serpentinization. In addition to its involvement in plagioclase alteration, calcic amphibole (hornblende-actinolite) is found as a partial replacement of primocryst and interstitial clinopyroxene. Ca-poor amphibole occurs as an alteration product of orthopyroxene. The presence of the assemblage chlorite + prehnite + actinolite + epidote suggests that alteration continued to temperatures less than 400 degrees C.The delta D values and H 2 O contents of igneous rocks at Babbitt range from -56 to -138 per mil, and 0.49 and 4.60 wt percent, respectively. Biotite separates are characterized by delta D values of-67 to -86 per mil and delta 18 O values of 4.2 to 6.1 per mil, a range considered normal for biotite in mafic igneous rocks. Because most of the low-temperature reactions between fluids and minerals in the cooling igneous system involved hydration, H 2 O content is an approximate indicator of the extent of low-temperature alteration. Rocks with H 2 O contents in excess of approximately 1.5 percent are characterized by delta D values greater than -87 per mil, except for a sample of highly serpentinized picrite which has a delta D value of -120 per mil. Modeling of the isotopic composition of a fluid exsolved from interstitial melt indicates that for samples with minor alteration and H 2 O contents less than approximately 1 wt percent, secondary mineral assemblages were probably produced as a D-depleted magmatic vapor reacted with primary minerals along cooling fractures and grain margins. Areas of more intense alteration may have formed from localized accumulations of D-enriched vapor released from underlying interstitial melt. Alternatively, intensely altered zones may have formed due to the introduction of relatively D-enriched fluid derived from devolatilization of metasedimentary xenoliths and footwall. The latter interpretation is consistent with previously reported isotopic (Ripley and Taib, 1989) and fluid inclusion (Pasteris, 1989) data. Both oxygen and hydrogen isotope compositions of one serpentinized picrite suggest that a low 18 O, low D fluid of probable meteoric origin has been involved in the serpentinization of olivine-rich units.An insufficient number of samples have been analyzed to determine if a direct positive correlation exists between platinum-group element (PGE) content and degree of alteration in the Babbitt deposit. However, the presence of discrete platinum-group minerals in intensely altered zones suggests that most Pt and Pd remobilization occurred via transport by a D-enriched fluid. A positive correlation between whole-rock delta D values and Pd content for samples that contain less than 300 ppb Pd suggests that magmatic fluids as well as high delta D fluids derived from metasedimentary country rocks may have been involved in local Pt-Pd redistribution.

Extensive felsic lavas and rheoignimbrites in the Keweenawan Midcontinent Rift plateau volcanics, Minnesota: petrographic and field recognition

The ≈8-km-thick sequence of plateau lavas (North Shore Volcanic Group) in northeastern Minnesota that was erupted during the formation of the 1.1 Ga Midcontinent Rift System includes an unusually large proportion (10–25%) of felsic flows (rhyolites and icelandites). Individual flows are now recognized with estimated volumes up to 600 km 3 and lateral extents up to 40 km; some were probably originally much broader. Many are anomalously large compared to most silicic lava flows. Primary structures and textures of these essentially unmetamorphosed rocks indicate eruptive style for most of these large flows. In particular, the features of the basal meter or so and uppermost several meters show that some flows were emplaced as fluid lavas, and others as strongly rheomorphic tuffs. It may be essential in interpreting the origin of felsic flows elsewhere to examine petrographically the lowermost few decimeters and the top, where the only evidence (poorly to strongly welded shards) may exist of the pyroclastic origin of a large, otherwise lava-like flow. Interior textures of these crystalline flows provide evidence of high-temperature eruption, especially in the form of abundant paramorphs of magmatic tridymite in the groundmass, commonly arranged in a pilotaxitic flow structure. Based on structures, textures, field relations, and geochemistry, We conclude that (a) all the icelandites and some rhyolites were emplaced as large, moderately to very fluid lava flows that spread widely while crystallizing tridymite; and that (b) some large rhyolites, which are only sparsely phyric, were emplaced as pyroclastic flows subsequently remobilized as rheoignimbrites. The mode of emplacement of one of the largest aphyric flows remains enigmatic. Vent areas or characteristics are unknown. The low apparent viscosity of the erupted magmas is probably attributable to compositional factors (F content, high Fe 2+) as well as to high temperatures and eruption rates. Estimated temperatures of about 1000–1100 C are associated with the low water content and perhaps great depth of origin of the magmas at the time of massive influx of hot matic melts from the mantle during continental rifting.

A model for the Ni-Cu-PGE ores of the Noril'sk region and its application to other areas of flood basalt

Five principal magma types characterize the 3.8-km vertical thickness of flood basalt of the Noril9sk region. From base to top these are (1) a high TiO 2 type with alkalic affiliations (Ivakinsky and Syverminsky suites), (2) a TiO 2 -rich picrite-bearing magma (Gudchichinsky suite), (3) a picrite-bearing TiO 2 -poor magma (Tuklonsky suite), (4) a crustally contaminated TiO 2 -poor magma (Lower Nadezhdinsky suite), and (5) a picrite-free magma (Mokulaevsky, Kharayelakhsky, Kumginsky, and Samoyedsky suites) similar but not identical in its chemistry to the Tuklonsky and separated from the Lower Nadezhdinsky by transitional suites (Upper Nadezhdinsky and Morongovsky).The Ni-Cu ores are associated with olivine-rich intrusions that are geochemically similar to the Mokulaevsky magma. They are associated with about 1/15 to 1/10 of their mass in PGE-rich sulfides; this requires that these sulfides equilibrated with 15 to 200 times more magma than is represented by the intrusions themselves.Following the model of Naldrett et al. (1992), it is proposed that crustal contamination of the original Tuklonsky-type magma occurred at the top of a vertically extensive, fault-controlled magma chamber and caused segregation of immiscible sulfides with a low R factor. These depleted this magma in chalcophile elements and settled deeper in the chamber. As they settled, they reacted with less contaminated magma, scavenging additional chalcophile elements. They finally came to rest near the base of the chamber to form a zone of sulfide- (and thus chalcophile element-) enriched magma. Eruption, and in some instances intrusion, of progressively stratigraphically lower, less contaminated and thus less depleted layers in the chamber gave rise to the observed sequence of basalts and intrusions. The answer to the strong enrichment in Ni, Cu, and PGE associated with the ore-bearing intrusions is that these metals came from the magma that is now represented by the basalt magma with which the sulfides have come into contact and which shows chalcophile element depletion.Key metallogenic characteristics of the Noril9sk region are (1) a hot region in the mantle giving rise to vast amounts of basalt, (2) continental-scale rifting, (3) major faults serving as controls for extrusive and intrusive magmatism, (4) crustal contamination of magma, (5) chalcophile element depletion of contaminated magma, (6) intrusion of magma that has inherited sulfide removed from the depleted magma, and (7) the presence of evaporites in the sedimentary sequence underlying the basalts.The simplest way to recognize chalcophile element depletion is to plot Ni/MgO vs. Mg number [= at. % Mg/(Mg + 0.9Fe total )]. A strong indication that basalts have undergone crustal contamination is the correlation of high La/Sm ratios with high SiO 2 contents.The Lake Superior region shows many of the key metallogenetic characteristics of the Noril9sk region, including continental rifting, a triple junction, huge thicknesses of basalt, and Cu-Ni sulfide-bearing intrusions (Duluth Complex, Crystal Lake gabbro). Some of the Keweenawan basalts of the region show Ni depletion correlated with chalcophile element depletion. The region is an example of the potential application of the Noril9sk model to exploration.