Yuri Amelin

Nature of the Earth's earliest crust from hafnium isotopes in singledetrital zircons

Continental crust forms from, and thus chemically depletes, the Earths mantle. Evidence that the Earths mantle was already chemically depleted by melting before the formation of todays oldest surviving crust has been presented in the form of Sm–Nd isotope studies of 3.8–4.0 billion years old rocks from Greenland and Canada. But this interpretation has been questioned because of the possibility that subsequent perturbations may have re-equilibrated the neodymium-isotope compositions of these rocks. Independent and more robust evidence for the origin of the earliest crust and depletion of the Archaean mantle can potentially be provided by hafnium-isotope compositions of zircon, a mineral whose age can be precisely determined by U–Pb dating, and which can survive metamorphisms. But the amounts of hafnium in single zircon grains are too small for the isotopic composition to be precisely analysed by conventional methods. Here we report hafnium-isotope data, obtained using the new technique of multiple-collector plasma-source mass spectrometry, for 37 individual grains of the oldest known terrestrial zircons (from the Narryer Gneiss Complex, Australia, with U–Pb ages of up to 4.14 Gyr ( 10–13). We find that none of the grains has a depleted mantle signature, but that many were derived from a source with a hafnium-isotope composition similar to that of chondritic meteorites. Furthermore, more than half of the analysed grains seem to have formed by remelting of significantly older crust, indicating that crustal preservation and subsequent reworking might have been important processes from earliest times.

Rapid eruption of Siberian flood-volcanic rocks and evidence for coincidence with the Permian–Triassic boundary and mass extinction at 251 Ma

The Siberian flood-volcanic event is the most voluminous and explosive, continental, volcanic event known in the Phanerozoic record. U^Pb perovskite and zircon ages were obtained for lavas of the lowermost unit (251.7 9 0.4 Ma) and near-uppermost unit (251.1 9 0.3 Ma), respectively, of the volcanic sequence in the Maymecha^Kotuy area, Russia. Along with stratigraphic correlations and paleomagnetic evidence, these ages suggest that rapid extrusion of the entire V6500 m thick composite sequence occurred in less than 1million years. The time of extrusion coincides precisely with an age of 251.4 9 0.3 Ma previously obtained for the Permian^Triassic mass-extinction event, the most devastating biotic crisis known. Emplacement of the Noril’sk^Talnakh ore-bearing intrusions, notable for their prodigious Cu^Ni^PGE deposits, was synchronous with these two major geologic events at 251.2 9 0.3 Ma. The Guli volcanic-intrusive complex in the Maymecha^Kotuy area appears to represent the final mafic magmatism of the entire Siberian flood-volcanic event. Baddeleyite from a carbonatite that intrudes the complex gives an age of 250.2 9 0.3 Ma, and shows possible 231 Pa excess. The Bolgokhtokh granodiorite stock has a zircon age of 229.0 9 0.4 Ma, and represents the youngest known magmatism in the region.

Assessment of errors in SIMS zircon U–Pb geochronology using a natural zircon standard and NIST SRM 610 glass

Analytical errors calculated for individual spot 206Pb/238U measurements of zircon analyzed using high mass resolution secondary ion mass spectrometry (HR-SIMS, e.g., SHRIMP II) were assessed using natural zircon (z6266) and synthetic glass standards (NIST SRM 610). Evidence for U/Pb homogeneity of these materials includes new thermal ionization mass spectrometry (TIMS) U–Pb analyses of 22 fragments of z6266 zircon from two laboratories, which are identical within error and yield a weighted mean 206Pb/238U age of 559.0±0.2 Ma. TIMS U–Pb analyses of the SRM 610 yielded homogeneous 206Pb/238U=0.2566. Errors in HR-SIMS 206Pb/238U measurements are distinguished in increasing hierarchy as within-spot, within-session (internal), and external. Replicate SHRIMP measurements of 208Pb+/206Pb+ and 248[ThO]+/254[UO]+ in SRM 610 demonstrate that within-spot analytical errors are sufficient (i.e., MSWDs∼1) to account for the dispersion of data. In contrast, it was not possible with SRM 610 to reproduce the 206Pb+/238U[Ox=0,1,2]+ ratios from spot-to-spot within the limits of the within-spot errors, after correcting the data for obvious systematic variations in the discrimination of the Pb and U isotopes. Identical findings were made following replicate 206Pb+/238U[Ox]+ analyses of z6266 zircon. That is, for both glass and zircon standards, there remained on average about ±1% (1σ) unaccounted variation per 206Pb+/238U[Ox]+ analysis (i.e., MSWDs≫1). Thus, unlike for 248[ThO]+/254[UO]+, the within-session errors in 206Pb+/238U[Ox]+ were necessarily higher than the within-spot errors. In contrast to 208Pb+/206Pb+ and 248[ThO]+/254[UO]+, the 206Pb+/238U[Ox]+ ratios in glass and zircon are significantly fractionated relative to accepted values, and it seems probable that this difference underlies the larger within-session errors for 206Pb+/238U[Ox]+. The usual methods (e.g., using 254[UO]+/238U+) to correct for the inherent discrimination are imperfect and sensitive to small differences in analytical conditions between spots. Matrix variations are ruled out, leaving the precise cause(s) of the additional variation in 206Pb+/238U[Ox]+ currently undetermined, and a hindrance to decreasing the uncertainties in spot 206Pb/238U measurements. The minimum within-session error for an individual 206Pb/238U measurement is the standard deviation (dispersion) of the discrimination-corrected 206Pb+/238U[Ox]+ values for a homogeneous reference material (e.g., SRM 610). The external error in 206Pb/238U additionally incorporates the standard deviation of the mean of the discrimination-corrected 206Pb+/238U[Ox]+ for a natural zircon standard. The analysis of 206Pb+/270[UO2]+ offers an improvement over measuring 206Pb+/238U+ because variable discrimination is reduced or eliminated with respect to 254[UO]+/238U+ and counting statistics are superior.

Evolution of the southern Abitibi greenstone belt based on U–Pb geochronology: autochthonous volcanic construction followed by plutonism, regional deformation and sedimentation

Abstract New mapping and U–Pb zircon geochronological results in the southern Abitibi greenstone belt (SAGB) support an autochthonous regional stratigraphy comprised of nine supracrustal assemblages, rather than the collage of allochthonous terranes proposed in recent publications. Based on lithological and geochronological criteria these supracrustal assemblages also encompass the Swayze, Shining Tree and Montcalm greenstone belts indicating that, despite separation by large granitic intrusions, they are in fact part of one extensive greenstone belt. In conjunction with the mapping, the geochronological results support coherent, upward-facing stratigraphic sections with about 20% of the samples containing inherited zircons with ages similar to those found in underlying assemblages. The seven oldest assemblages represent semi-continuous volcanism from 2750 to 2697 Ma. The volcanism is compositionally diverse ranging from komatiite and tholeiitic basalt to calc–alkaline mafic to felsic lavas. Intimate intermingling of the different magma clans occurs throughout much of the stratigraphic section. Regional fault control on the distribution of the volcanic assemblages provides evidence for early dip–slip movement associated with volcanic extension dating back to at least 2725 Ma. The two youngest assemblages are dominantly sedimentary and were unconformably deposited on the volcanic assemblages in close proximity to regional faults. The earlier sedimentary assemblage consists of turbidites and minor iron formation deposited from 2696 to 2692 Ma, and the youngest one of subaerial conglomerates, fluvial sandstones and alkalic to shoshonitic volcanic rocks ranging in age from 2687 to 2675 Ma. The sedimentary assemblages are broadly contemporaneous with the emplacement of syntectonic granitic plutons, regional folding and reactivated movement on the regional faults probably related to accretion of the Abitibi to the Superior Province craton. Autochthonous repetition of different geodynamic environments over the 50 million years of volcanic activity with five dominantly tholeiitic ± komatiitic assemblages and two dominantly calc–alkaline assemblages, as well as the extensive intermingling of these different magma clans in a number of the assemblages, appears to be unique to the Archean. It suggests complex, large-scale and long-lived interaction between mantle plumes and subduction zone magmas. The Nd isotope data indicate a derivation of the diverse volcanic magmas from homogeneously depleted sources with eNd=2.5±0.5 demonstrating a lack of contamination by ancient enriched components.

Young chondrules in CB chondrites from a giant impact in the early Solar System

Chondrules, which are the major constituent of chondritic meteorites, are believed to have formed during brief, localized, repetitive melting of dust (probably caused by shock waves) in the protoplanetary disk around the early Sun. The ages of primitive chondrules in chondritic meteorites indicate that their formation started shortly after that of the calcium-aluminium-rich inclusions (4,567.2 ± 0.7 Myr ago) and lasted for about 3 Myr, which is consistent with the dissipation timescale for protoplanetary disks around young solar-mass stars. Here we report the 207Pb–206Pb ages of chondrules in the metal-rich CB (Bencubbin-like) carbonaceous chondrites Gujba (4,562.7 ± 0.5 Myr) and Hammadah al Hamra 237 (4,562.8 ± 0.9 Myr), which formed during a single-stage, highly energetic event. Both the relatively young ages and the single-stage formation of the CB chondrules are inconsistent with formation during a nebular shock wave. We conclude that chondrules and metal grains in the CB chondrites formed from a vapour–melt plume produced by a giant impact between planetary embryos after dust in the protoplanetary disk had largely dissipated. These findings therefore provide evidence for planet-sized objects in the earliest asteroid belt, as required by current numerical simulations of planet formation in the inner Solar System.

Precise geochronology of phoscorites and carbonatites:: The critical role of U-series disequilibrium in age interpretations

We present the results of a comparative study of several geochronometer minerals (baddeleyite, zircon, apatite, phlogopite and tetraferriphlogopite) and isotopic systems (U-Pb, Th-Pb and Rb-Sr) from phoscorites (magnetite-forsterite-apatite-calcite rocks) and carbonatites of the Kovdor ultramafic-alkaline-carbonatite massif, Kola Peninsula, Russia. Uranium, thorium and their decay products are extremely fractionated by minerals that crystallise from carbonatite and phoscorite magma. We obtain high-precision ages from different chronometers, compare their accuracy, and evaluate the role of geochronological pitfalls of initial radioactive disequilibrium, differential migration of radiogenic isotopes, and inaccurate decay constants. Apatite yielded concordant U-Th-Pb ages between 376 and 380 Ma. The accuracy of the apatite 238U-206Pb ages is, however, compromised by uncertainty in the amount of radiogenic 206Pb produced from initial excess 230Th. The 235U-207Pb ages are relatively imprecise due to large common Pb correction and the uncertainty in the initial Pb isotopic composition. The Th-Pb system yields a more precise age of 376.4 ± 0.6 Ma. Zircon from two carbonatite samples is characterised by moderate to low U contents, high Th contents, and very high Th/U ratios up to 9000. The 206Pb*/238U systems in the zircon are strongly affected by the presence of excess 206Pb*, produced by decay of initial 230Th. The 208Pb*/232Th ages of zircon from both carbonatite samples are uniform and yield a weighted average of 377.52 ± 0.94 Ma. Baddeleyite U-Pb analyses are 3 to 6% normally discordant and have variable 207Pb*/206Pb* apparent ages. Eleven alteration-free baddeleyite fractions from three samples with no evidence for Pb loss yield uniform 206Pb*/238U ages with a weighted average of 378.54±0.23 Ma (378.64 Ma after correction for initial 230Th deficiency), which we consider the best estimate for age of the phoscorite-carbonatite body of the Kovdor massif. The 206Pb*/238U ages of baddeleyite fractions from five other samples spread between 378.5 and 373 Ma, indicating a variable lead loss up to 1.5%. The anomalously old 207Pb/235U and 207Pb/206Pb ages are consistent with the presence of excess radiogenic 207Pb* in the baddeleyite. We interpret this as a result of preferential partitioning of 231Pa to baddeleyite. Fifteen phlogopite and tetraferriphlogopite fractions from five carbonatite and phoscorite samples yielded precise Rb-Sr isochron age of 372.2 ± 1.5 Ma, which is 5 to 7 m.y. younger than our best estimate based on U-Th-Pb age values. This difference is unlikely to be a result of the disturbance or late closure of Rb-Sr system in phlogopite, but rather suggests that the accepted decay constant of 87Rb is too high. Comparative study of multiple geochronometer minerals from the Kovdor massif has revealed an exceptional complexity of isotopic systems. Reliable ages can be understood through systematic analysis of possible sources of distortion. No single geochronometer is sufficiently reliable in these rocks. Th-Pb and Rb-Sr can be a very useful supplement to U-Pb geochronometry, but the routine use of these geochronometers together will require more precise and accurate determination of decay constants for 232Th and 87Rb.

Chronology of the Solar System’s Oldest Solids

Determining the origins of our solar system and, by proxy, other planetary systems, depends on knowing accurately and precisely the timing and tempo of the transformation of the disk of gas and dust to the solids that formed the planets. Relative ages based on the short-lived nuclide 26Al indicate that high-temperature calcium-aluminum inclusions (CAIs) formed before lower temperature chondrules but these ages are heavily dependant on a model of homogeneous distribution of 26Al within the protoplanetary disk. The competing X-wind model argues for heterogeneous distribution of 26Al due to its formation by intra-solar system irradiation such that this system would have no chronological significance. We report a207Pb-206Pb isochron age of 4565.45 ± 0.45 Myr for chondrules from the CV chondrite Allende, an age that is 1.66 ± 0.48 Myr younger than the accepted Pb-Pb age for CAIs from this chondrite group. This age offset is in excellent agreement with the relative ages determined using the 26Al-26Mg system, an observation that supports a supernova origin for 26Al and, importantly, the chronological significance of the 26Al-26Mg system in general. This is consistent with an early and brief CAI-forming event followed by recurrent chondrule formation throughout the life span of the protoplanetary disk. The paucity of old chondrules in chondrite meteorites may reflect their early incorporation into the parent bodies of differentiated meteorites after CAIs were effectively removed from the innermost regions of the protoplanetary disk. Lastly, the agreement between the absolute and relative chronology of CAIs and chondrules requires a solar system age younger than ~4567.5 Myr

Geochronology of the Voisey's Bay intrusion, Labrador, Canada, by precise U–Pb dating of coexisting baddeleyite, zircon, and apatite

Abstract The emplacement history of the Voiseys Bay troctolite intrusion, that hosts the major Ni–Cu–Co sulfide deposit of the same name, has been studied using precise U–Pb geochronology of baddeleyite, zircon and apatite. The baddeleyite U–Pb ages of multiple drill core samples of troctolite and gabbro indicate that all of the mafic rocks studied from different components of the Voiseys Bay intrusion: Eastern Deeps, Discovery Hill Zone and Reid Brook Zone, and from the adjacent Red Dog area, were emplaced at 1332.7±1.0 Ma. On the basis of combined geological and geochronological evidence, it is suggested that the Voiseys Bay Ni–Cu–Co deposit was formed during the same period. The zircons coexisting with the ca. 1333 Ma baddeleyite show a diversity of ages. The zircons from normal troctolite and some of the olivine gabbro samples are coeval with the baddeleyite, while zircon from the varied textured troctolite and feeder olivine gabbro are much younger at 1305.0±0.8 Ma. The identical ages of the younger zircon population and the Voiseys Bay syenite that cuts the mafic rocks indicate a link between zircon growth in the mafic rocks and contact metamorphism, related to the emplacement of the syenite. Various mechanisms of zircon growth were probably involved, including reaction of the 1333 Ma baddeleyite with a silica-enriched fluid with formation of a secondary polycrystalline zircon, and zircon crystallization from syenite micro-veins in the mafic rocks. The mean 207 Pb / 206 Pb age of 1303.5±2.6 Ma of the Voiseys Bay apatites is similar to the age of the younger zircon population. The apatite age may either be a result of resetting the U–Pb system in response to the syenite intrusion, or may reflect the closure of the system during regional cooling and cessation of fluid circulation. The presence of xenocrystic zircon in a Discovery Hill Zone feeder olivine gabbro indicates that the Voiseys Bay magmas were contaminated with 1.90 Ga crustal rocks.

U-Pb ages of secondary silica at Yucca Mountain, Nevada: implications for the paleohydrology of the unsaturated zone

Abstract Uranium, Th and Pb isotopes were analyzed in layers of opal and chalcedony from individual mm- to cm-thick calcite and silica coatings at Yucca Mountain, Nevada, USA, a site that is being evaluated for a potential high-level nuclear waste repository. These calcite and silica coatings on fractures and in lithophysal cavities in Miocene-age tuffs in the unsaturated zone (UZ) precipitated from descending water and record a long history of percolation through the UZ. Opal and chalcedony have high concentrations of U (10 to 780 ppm) and low concentrations of common Pb as indicated by large values of 206Pb/204Pb (up to 53,806), thus making them suitable for U-Pb age determinations. Interpretations of U-Pb isotope systems in opal samples at Yucca Mountain are complicated by the incorporation of excess 234U at the time of mineral formation, resulting in reverse discordance of U-Pb ages. However, the 207Pb/235U ages are much less affected by deviation from initial secular equilibrium and provide reliable ages of most silica deposits between 0.6 and 9.8 Ma. For chalcedony subsamples showing normal age discordance, these ages may represent minimum times of deposition. Typically, 207Pb/235U ages are consistent with the microstratigraphy in the mineral coating samples, such that the youngest ages are for subsamples from outer layers, intermediate ages are from inner layers, and oldest ages are from innermost layers. 234U and 230Th in most silica layers deeper in the coatings are in secular equilibrium with 238U, which is consistent with their old age and closed system behavior during the past ∼0.5 Ma. The ages for subsamples of silica layers from different microstratigraphic positions in individual calcite and silica coating samples collected from lithophysal cavities in the welded part of the Topopah Spring Tuff yield slow long-term average growth rates of 1 to 5 mm/Ma. These data imply that the deeper parts of the UZ at Yucca Mountain maintained long-term hydrologic stability over the past 10 Ma. despite significant climate variations. U-Pb ages for subsamples of silica layers from different microstratigraphic positions in individual calcite and silica coating samples collected from fractures in the shallower part of the UZ (welded part of the overlying Tiva Canyon Tuff) indicate larger long-term average growth rates up to 23 mm/Ma and an absence of recently deposited materials (ages of outermost layers are 3–5 Ma.). These differences between the characteristics of the coatings for samples from the shallower and deeper parts of the UZ may indicate that the nonwelded tuffs (PTn), located between the welded parts of the Tiva Canyon and Topopah Spring Tuffs, play an important role in moderating UZ flow.

Origin, timing, and temperature of secondary calcite-silica mineral formation at Yucca Mountain, Nevada

The origin of secondary calcite-silica minerals in primary and secondary porosity of the host Miocene tuffs at Yucca Mountain has been hotly debated during the last decade. Proponents of a high-level nuclear waste repository beneath Yucca Mountain have interpreted the secondary minerals to have formed from cool, descending meteoric fluids in the vadose zone; critics, citing the presence of two-phase fluid inclusions, argued that the minerals could only have formed in the phreatic zone from ascending hydrothermal fluids. Understanding the origin, temperature, and timing of these minerals is critical in characterizing geologically recent fluid flux at the site, and has significant implications to whether waste should be stored at Yucca Mountain. Petrographic and paragenetic studies of 155 samples collected from the Exploratory Studies Facility (ESF) and repository block cross drift (ECRB) tunnels indicate that heterogeneously distributed calcite with lesser chalcedony, quartz, opal, and fluorite comprise the oldest secondary minerals. These are typically overgrown by intermediate-aged calcite, often exhibiting distinctive bladed habits. The youngest event recorded across the site is the deposition of Mg-enriched (up to 1 wt%) and depleted, growth-zoned calcite intergrown with U-enriched opal. The cyclical variation in Mg enrichment and depletion is probably related to climate changes that have occurred during the last few million years. The distribution of secondary minerals is consistent with precipitation in the vadose zone. Fluid inclusion petrography of sections from the 155 samples determined that 96% of the fluid inclusion assemblages (FIAs) contained liquid-only inclusions that formed at ambient temperatures (35°C). However, 50% of the samples (n 78) contained relatively rare FIA that contain both liquid-only and liquid plus vapor inclusions (herein termed two-phase FIAs) that formed at temperatures above 35°C. Virtually all of these two-phase FIAs occur in paragenetically old calcite; rare two-phase inclusion assemblages were also observed in early fluorite and quartz, and early-intermediate calcite. Homogenization temperatures ( trapping temperatures) across Yucca Mountain are generally 45 to 60°C, but higher temperatures reaching 83°C were recorded in calcite from the north portal and ramp of the ESF. Cooler temperatures of 35 to 45°C were recorded in the intensely fractured zone. Multiple populations of two-phase FIAs from lithophysal cavities in the ESF and ECRB cross drift indicate early fluid cooling with time from temperatures 45°C in early calcite, to 35 to 45°C in paragenetically younger calcite. Freezing point depressions range from 0.2 to 1.6°C, indicating trapping of a low salinity fluid. The majority of intermediate calcite and all outermost Mg-enriched calcite contains rare all-liquid inclusions and formed from ambient temperature (35°C) fluids. Carbon and oxygen isotope data reveal a consistent trend of decreasing 13 C (from 9.5 to 8.5‰) and increasing 18 O (from 5.2 to 22.1‰) values from paragenetically early calcite to Mg-enriched growth-zoned calcite. Depleted D values (131 to 90‰) of inclusion fluids from intermediate and the youngest Mg-enriched calcite indicate derivation from surface meteoric fluids. Recalculation of 18 OH2O values of 12 to 10‰ is consistent with derivation from paleometeoric fluids. Results of integrated U-Pb dating (opal and chalcedony) and fluid inclusion microthermometry indicate that two-phase FIAs that trapped fluids of 50°C are older than 6.29 0.30 Ma. Two-phase FIAs in parage- netically later calcite, which formed from fluids of 35 to 45°C, are older than 5.32 0.02 Ma. There is no evidence for trapping of fluids with elevated temperatures during the past 5.32 my. The youngest Mg-enriched calcite intergrown with opal began to precipitate between about 1.9 to 2.9 Ma and has continued to precipitate within the past half million years. The presence of liquid-only inclusions and the consistent occurrence of Mg-enriched calcite and opal as the youngest event indicate a minor, but chemically distinct, ambient temperature (35°C) fluid flux during the past 2 to 3 my. Copyright