Oxidized sulfur-rich mafic magma at Mount Pinatubo,Philippines

Basaltic fragments enclosed in andesitic dome lavas and pyroclastic flows erupted during the early stages of the 1991 eruption of Mount Pinatubo, Philippines, contain amphiboles that crystallized during the injection of mafic magma into a dacitic magma body. The amphiboles contain abundant melt inclusions, which recorded the mixing of andesitic melt in the mafic magma and rhyolitic melt in the dacitic magma. The least evolved melt inclusions have high sulfur contents (up to 1,700 ppm) mostly as SO₄ ^2–, which suggests an oxidized state of the magma (NNO+1.4). The intrinsically oxidized nature of the mafic magma is confirmed by spinel–olivine oxygen barometry. The value is comparable to that of the dacitic magma (NNO+1.6). Hence, models invoking mixing as a means of releasing sulfur from the melt are not applicable to Pinatubo. Instead, the oxidized state of the dacitic magma likely reflects that of parental mafic magma and the source region in the sub-arc mantle. Our results fit a model in which long-lived SO₂ discharge from underplated mafic magma accumulated in the overlying dacitic magma and immiscible aqueous fluids. The fluids were the most likely source of sulfur that was released into the atmosphere during the cataclysmic eruption. The concurrence of highly oxidized basaltic magma and disproportionate sulfur output during the 1991 Mt. Pinatubo eruption suggests that oxidized mafic melt is an efficient medium for transferring sulfur from the mantle to shallow crustal levels and the atmosphere. As it can carry large amounts of sulfur, effectively scavenge sulfides from the source mantle and discharge SO₂ during ascent, oxidized mafic magma forms arc volcanoes with high sulfur fluxes, and potentially contributes to the formation of metallic sulfide deposits.Editorial responsibility: J. Hoefs

---

Polycrystalline calcite to aragonite transformation kinetics: experiments in synthetic systems

The kinetics of the calcite to aragonite transformation have been investigated using synthetic polycrystalline calcite aggregates, with and without additional minerals present. The reaction progresses as a function of time were measured at four temperature/pressure conditions: (1) 550 °C/1.86 GPa; (2) 600 °C/2.11 GPa; (3) 650 °C/2.11 GPa, and (4) 700 °C/2.29 GPa. Experiments reveal that Mg-calcite and Fe-calcite transforms to aragonite at considerably slower rates than pure calcite, and that Sr-bearing calcite and calcite + quartz aggregates transform at significantly higher rates than pure calcite. The reaction progresses vs. time data for pure calcite were fitted to Cahns grain-boundary nucleation and interface-controlled growth model. Evidence for interface-controlled growth is provided by petrographic observations of grain boundaries. The activation energy for aragonite growth from the synthetic polycrystalline calcite determined in this study is significantly lower than that previously determined from a natural marble. The discrepancy in rates and activation energy may be attributed to the nature of grain boundaries, to deformational strain or the presence of impurities in the studied samples, and likely to uncertainties in experimental conditions. The results of this study imply that the variation of local petrologic conditions, in addition to temperature, pressure and grain size, may play an important role in determining the rates for the calcite to aragonite transformation in nature.Editorial responsibility: W. Schreyer     

U–Pb and Hf-isotope analysis of zircons in mafic xenoliths from Fuxian kimberlites: evolution of the lower crust beneath the North China craton

Mafic xenoliths from the Paleozoic Fuxian kimberlites in the North China craton include garnet granulite, and minor pyroxene amphibolite, metagabbro, anorthosite and pyroxenite. The formation conditions of the amphibolites are estimated at 745–820 °C and 7.6–8.8 Kb (25–30 km); the granulites probably are derived from greater depths in the lower crust. LAM-ICPMS U–Pb dating of zircons from four granulites reveals multiple age populations, recording episodes of magmatic intrusion and metamorphic recrystallisation. Concordant ages and upper intercept ages, interpreted as minimum estimates for the time of magmatic crystallisation, range from 2,620 to 2,430 Ma in three granulites, two amphibolites and two metagabbros. Lower intercept ages, represented by near-concordant zircons, are interpreted as reflecting metamorphic recrystallisation, and range from 1,927 to 1,852 Ma. One granulite contains two metamorphic zircon populations, dated at 1,927±55 Ma and 600–700 Ma. Separated minerals from one granulite and one amphibolite yield Sm–Nd isochron ages of 1,619±48 Ma (¹⁴³Nd/¹⁴⁴Nd)i=0.51078), and 1,716±120 Ma (¹⁴³Nd/¹⁴⁴Nd)i=0.51006), respectively. These ages are interpreted as recording cooling following metamorphic resetting; model ages for both samples are in the range 2.40–2.66 Ga. LAM-MC-ICPMS analyses of zircon show a range in ¹⁷⁶Hf/¹⁷⁷Hf from 0.28116 to 0.28214, corresponding to a range of _Hf from –34 to +12. The relationships between ²⁰⁷Pb/²⁰⁶Pb age and _Hf show that: (1) the granulites, amphibolites and metagabbro were derived from a depleted mantle source at 2.6–2.75 Ga; (2) zircons in most samples underwent recrystallisation and Pb loss for 100–200 Ma after magmatic crystallisation, consistent with a residence in the lower crust; (3) metamorphic zircons in several samples represent new zircon growth, incorporating Hf liberated from breakdown of silicates with high Lu/Hf; (4) in other samples metamorphic and magmatic zircons have identical ¹⁷⁶Hf/¹⁷⁷Hf, and the younger ages reflect complete resetting of U–Pb systems in older zircons. The Fuxian mafic xenoliths are interpreted as the products of basaltic underplating, derived from a depleted mantle source in Neoarchean time, an important period of continental growth in the North China craton. Paleoproterozoic metamorphic ages indicate an important tectonic thermal event in the lower crust at 1.8–1.9 Ga, corresponding to the timing of collision between the Eastern and Western Blocks that led to the final assembly of the North China craton. The growth of metamorphic zircon at 600–700 Ma may record an asthenospheric upwelling in Neoproterozoic time, related to uplift and a regional disconformity in the North China craton.     

Metasomatism and melting history of a Variscan lithospheric mantle domain: evidence from the Puy Beaunit xenoliths (French Massif Central)

Mantle xenoliths from Puy Beaunit (French Massif Central) are compositionally varied, ranging from relatively fertile spinel lherzolites to refractory spinel dunites. Fertile peridotites have registered a modal (amphibole-bearing lherzolites) and cryptic metasomatic event that took place before the last Permian (257 Ma) melting episode. Depletion processes have been constrained by chemical modelling: the depletion is related to different degrees of partial melting, but two major melt extraction episodes are needed to explain the range of major element composition. The second event was responsible for the local large-scale dunitification of former residues. The first melting event (F25%) and metasomatic enrichment are attributed to an ancient fluid and/or liquid infiltration that could be related to a pre-Variscan regional subduction (located to the north of the Beaunit area). Texture acquisition and major deformation of the mantle xenoliths were sub-contemporaneous of the subduction and would result from lithospheric delamination. The second melting event (F17%) produced high-Mg basalts with calc-alkaline trace element signature that gave rise to the Permian underplating episode recognised in western Europe.     

Geological and SHRIMP II U-Pb constraints on the age and origin of the Breves Cu-Au-(W-Bi-Sn) deposit,Carajás,Brazil

The Breves deposit in the Carajás Copper-Gold Belt, Brazil, a member of the Cu-Au-(W-Bi-Sn) group of deposits, contains about 50 Mt of 1.22% Cu, 0.75 g/t Au, 2.4 g/t Ag, 1,200 g/t W, 70 g/t Sn, 175 g/t Mo and 75 g/t Bi. It is hosted by sandstones and siltstones of the Águas Claras Formation (minimum age of 2,681±5 Ma) in the roof zone of a complex, highly altered granite intrusion. The mineralisation is disseminated in a greisenized zone, resulting from alteration of probable monzogranites and syenogranites. The ore-bearing greisen contains abundant xenomorphic quartz in association with Fe-chlorite and muscovite. The gangue assemblage also includes fluorite, tourmaline, and minor amounts of monazite, xenotime, chlorapatite, thorite, zircon, calcite, siderite and bastnäesite. Copper mineralisation is dominated by chalcopyrite associated with pyrite, arsenopyrite, pyrrhotite and molybdenite. Gold particles, in equilibrium with native bismuth, are common as inclusions in chalcopyrite. The greisen contains sub-economic concentrations of tungsten and niobium that are related to the presence of ferberite, qitianlingite and Nb-rutile. SHRIMP II zircon dating of the host granites gives ²⁰⁷Pb/²⁰⁶Pb ages of 1,878±8 and 1,880±9 Ma for two phases, and a combined age of 1,879±6 Ma. SHRIMP II dating of monazite and xenotime grains in late- to post-mineralisation veins gives a combined ²⁰⁷Pb/²⁰⁶Pb age of 1,872±7 Ma, indistinguishable from the ages of the granites. This provides a genetic connection between the Breves deposit and the ca. 1.88 Ga A-type granite magmatism that typifies the Carajás Belt as part of a much larger, intracratonic magmatic province that extends over much of the Amazonian Craton. The recognition of this association has exploration implications, not only for the geophysical signature of the granite roof zones, but also for likely geochemical dispersion around the deposits of this type.Editorial handling: G. Beaudoin     

Noril’sk–Talnakh Cu–Ni–PGE deposits: a revised tectonic model

The Norilsk mining district is located at the northwest margin of the Tunguska basin, in the centre of the 3,000×4,000 km Siberian continental flood basalt (CFB) province. This CFB province was formed at the Permo-Triassic boundary from a superplume that ascended into the geometric centre of the Laurasian continent, which was surrounded by subducting slabs of oceanic crust. We suggest that these slabs could have reached the core–mantle boundary, and they may have controlled the geometric focus of the superplume. The resulting voluminous magma intruded and erupted in continental rifts and related extensive flood basalt events over a 2–4 Ma period. Cu–Ni–PGE sulfide mineralization is found in olivine-bearing differentiated mafic intrusions beneath the flood basalts at the northwestern margin of the Siberian craton and also in the Taimyr Peninsula, some 300 km east of a triple junction of continental rifts, now buried beneath the Mesozoic–Cenozoic sedimentary basin of western Siberia. The Norilsk-I and Talnakh-Oktyabrsky deposits occur in the Norilsk–Kharaelakh trough of the Tunguska CFB basin. The Cu–Ni–PGE-bearing mineralized intrusions are 2–3 km-wide and 20 km-long differentiated chonoliths. Previous studies suggested that parts of the magma remained in intermediate-level crustal chambers where sulfide saturation and accumulation took place before emplacement. The 5–7-km-thick Neoproterozoic to Palaeozoic country rocks, containing sedimentary Cu mineralization and evaporites, may have contributed additional metal and sulfur to this magma. Classic tectonomagmatic models for these deposits proposed that subvertical crustal faults, such as the northeast-trending Norilsk–Kharaelakh fault, were major trough-parallel conduits providing access for magmas to the final chambers. However, geological maps of the Norilsk region show that the Norilsk–Kharaelakh fault offsets the mineralization, which was deformed into folds and offset by related reverse faults, indicating compressional deformation after mineralization in the Late Triassic to Early Jurassic. In addition, most of the intrusions are sills, not dykes as should be expected if the vertical faults were major conduits. A revised tectonic model for the Norilsk region takes into account the fold structure and sill morphology of the dominant intrusions, indicating a lateral rather than vertical emplacement direction for the magma into final chambers. Taking into account the fold structure of the country rocks, the present distribution of the differentiated intrusions hosting the Norilsk-I and Talnakh–Oktyabrsky deposits may represent the remnants of a single, >60 km long, deformed and eroded palm-shaped cluster of mineralized intrusions, which are perceived as separate intrusions at the present erosional level. The original direction of sill emplacement may have been controlled by a northeast-trending paleo-rise, which we suggest is present at the southeastern border of the Norilsk–Kharaelakh trough based on analysis of the unconformity at the base of the CFB. The mineralized intrusions extend along this rise, which we interpret as a structure that formed above the extensionally tilted block in the metamorphic basement. Geophysical data indicate the presence of an intermediate magma chamber that could be linked with the Talnakh intrusion. In turn, this T-shaped flat chamber may link with the Yenisei–Khatanga rift along the northwest-trending Pyasina transform fault, which may have served as the principal magma conduit to the intermediate chamber. It then produced the differentiated mineralized intrusions that melted through the evaporites with in situ precipitation of massive, disseminated, and copper sulfide ore. The Norilsk–Kharaelakh crustal fault may not relate to mineralization and possibly formed in response to late Mesozoic spreading in the Arctic Ocean.Editorial handling: P. Lightfoot     

Petrology and structure of the Palaeoproterozoic (1.9 Ga) Rytky nickel sulphide deposit,Central Finland: a comparison with the Kotalahti nickel deposit

The Palaeoproterozoic (1.9 Ga) Rytky and Kotalahti mafic-ultramafic intrusions are located in the contact zone between the Archaean craton and Proterozoic supracrustal rocks. During the second deformation event (D₂) the surrounding country rocks were subjected to intensive metamorphism and deformation associated with the Svecofennian orogeny; the Archaean/Proterozoic boundary controlled both D₂ thrusting and magma ascent. Emplacement of the Rytky and Kotalahti intrusions took place at the culmination of D₂, as shown by the gneiss inclusions with S₂ schistosity within the intrusions. Overthrusting continued after emplacement, with detached fragments of the bodies incorporated into the Archaean gneisses. During the third deformation event (D₃) the originally subhorizontal intrusions were rotated into a subvertical position, so that they now have their stratigraphic top towards the west. The Rytky intrusion is composed mainly of medium- and coarse-grained lherzolite, websterite and gabbronorite. The nickel deposit with pentlandite as the main nickel mineral is associated with the lherzolite and websterite. The coarse-grained lherzolite, websterite and melagabbro represent the first rocks to form, and they contain the nickel sulphide mineralisation. Country rock contamination, as indicated by high TiO₂, P₂O₅, Rb, Zr and light rare earth element contents (LREE), is most pronounced in the marginal part of the intrusion, which was the first to form. The variation in olivine composition (Fo 78.6-84.77 mole %; Ni 630–2386 ppm) and the metal ratio of the sulphide (Ni/Co 19.3 – 50.3) along with the internal stratigraphy of the intrusion indicate an in-situ process of sulphide ore formation.Editorial handling: P. LightfootAn erratum to this article can be found at     

Hydrothermal mixing,carbonate dissolution and sulfide precipitation in Mississippi Valley-type deposits

A large number of Mississippi Valley-Type (MVT) deposits are located within dissolution zones in carbonate host rocks. Some genetic models propose the existence of cavities generated by an earlier event such as a shallow karstification, that were subsequently filled with hydrothermal minerals. Alternative models propose carbonate dissolution caused by the simultaneous precipitation of sulfides. These models fail to explain either the deep geological setting of the cavities, or the observational features which suggest that the dissolution of carbonates and the precipitation of minerals filling the cavities are not strictly coeval. We present a genetic model inspired by the textural characteristics of MVT deposits that accounts for both the dissolution of carbonate and precipitation of sulfides and later carbonates in variable volumes. The model is based on the mixing of two hydrothermal fluids with a different chemistry. Depending on the proportion of the end members, the mixture dissolves and precipitates carbonates even though the two mixing solutions are both independently saturated in carbonates. We perform reactive transport simulations of mixing of a regional groundwater and brine ascending through a fracture, both saturated in calcite, but with different overall chemistries (Ca and carbonate concentrations, pH, etc). As a result of the intrinsic effects of chemical mixing, a carbonate dissolution zone, which is enhanced by acid brines, appears above the fracture, and another zone of calcite precipitation builds up between the cavity and the surrounding rock. Sulfide forms near the fracture and occupies a volume smaller than the cavity. A decline of the fluid flux in the fracture would cause the precipitation of calcite within the previously formed cavities. Therefore, dissolution of carbonate host rock, sulfide precipitation within the forming cavity, and later filling by carbonates may be part of the same overall process of mixing of fluids in the carbonate host rock.Editorial handling: C. Everett     

Geology and geochemistry of the Changba SEDEX Pb-Zn deposit,Qinling orogenic belt,China

The Changba Pb-Zn SEDEX deposit occurs in the Middle Devonian sequence of the Anjiaca Formation of the Western Qinling Hercynian Orogen in the Gansu Province, China. The Changba-II orebody is hosted in biotite quartz schist and is the largest of 143 stratiform orebodies that are hosted either in biotite quartz schist or marble. The Changba-II comprises two types of mineralization: a bedded facies and an underlying breccia lens. The bedded section exhibits three sulfide sub-facies zoned from bottom to top: 1) banded sphalerite intercalated with quartz albitite; 2) interbedded massive pyrite and sphalerite ore; and 3) banded sphalerite ore intercalated with banded baritite. Major metallic minerals are sphalerite, pyrite, galena, with minor arsenopyrite, pyrrhotite, boulangerite, and rare chalcopyrite. The bedded sulfides are underlain by a lens of brecciated and albitized biotite-quartz schists cemented by sulfides and tourmaline.Massive and bedded sulfide ³⁴S values range from 8.1 to 29.3, whereas barite ³⁴S values range from 20.8 to 31.5. Disseminated pyrite in footwall schists has ³⁴S values ranging from 8.1 to 10.6, and increase to values ranging from 11.1 to 14.7 in the hangingwall. The lower ³⁴S values for massive and bedded sulfides are interpreted to be derived from progressive bacterial sulfate reduction (BSR) of Devonian seawater in a sulfate-restricted sub-basin. The higher ³⁴S values for massive and bedded sulfides could be a product of quantitative BSR but this is incompatible with barite being more abundant above the bedded sulfides. Instead, it is more likely that thermochemical sulfate reduction of seawater sulfate or of evaporite was the source of heavy hydrothermal sulfur. Heavy hydrothermal sulfur was injected into a sulfate-restricted sub-basin where it mixed with low ³⁴S BSR sulfide to form the massive and bedded sulfides. The REE patterns of sulfide layers and associated quartz albitite and baritite are similar to those of the host biotite quartz schists, suggesting that the hydrothermal fluids leached REE from the underlying rocks. Pb isotope ratios in galena form an array between the Upper Crust and the Mantle reservoir curves, which indicates that the lead is derived from upper crustal rocks comprising mafic igneous units. The Sr⁸⁷/Sr⁸⁶ ratio of 0.7101 for carbonate within the sulfide layers also suggests that Sr is derived from the mixing of Sr leached from upper crustal rocks with Middle Devonian seawater Sr. A Rb-Sr isochron age of 389.4 ± 6.4 Ma for sulfide layers and the interbedded hydrothermal sediments is consistent with the age of host Mid-Devonian strata. Ar³⁹/Ar⁴⁰ plateau age at 352.8 ± 3.5 Ma and Ar³⁹-Ar⁴⁰ isochron age of 346.6 ± 6.4 Ma for albite in the quartz albitite intercalated with sulfide layers indicate either albite formation after the sulfides or thermal resetting of the Rb-Sr system at about 350 Ma, the age of collision between the North China and Yangtze cratons.Editorial handling: E. Frimmel