Norman J. Pearson

The Hf isotope composition of cratonic mantle: LAM-MC-ICPMS analysis of zircon megacrysts in kimberlites

Zircon megacrysts represent a late stage in the crystallisation of the magmas that produced the low-Cr megacryst suite (Ol1Opx1Cpx1Gnt1Ilm1Phl1Zir) found in many kimberlites, and may carry information on the sources of the parent magmas and the interaction of these magmas with the cratonic lithosphere. The isotopic composition of Hf has been measured in 124 mantle-derived zircon megacrysts from African, Siberian and Australian kimberlites, using a laser-ablation microprobe (LAM) and a multi-collector (MC) ICPMS. The zircons range in age from 90 Ma to ca 2500 Ma, allowing indirect analysis of mantle-derived Hf over a long time span. Most values of «Hf fall between 0 and 110, but zircon suites from several kimberlites range down to «Hf 52 16. Combined with published Nd data on the silicate members of the low-Cr megacryst suite, these data indicate crystallisation of zircon from magmas lying well below the terrestrial «Hf-«Nd array. LAM-ICPMS analyses of garnets and clinopyroxenes from mantle-derived peridotite xenoliths suggest that cratonic lithospheric mantle has Hf/Nd (0.3- 0.5) greater than estimated Bulk Silicate Earth. The depleted and metasomatised lherzolites and harzburgites that make up much of the Archean lithospheric mantle have Lu/Hf ratios (#0.15) low enough to account for the lowest «Hf observed in the zircons, over time spans of 1-3.5 Ga. We therefore suggest that the magmas from which the kimberlitic zircons crystallised were derived from Depleted Mantle or OIB-type sources, and developed negative «Hf through reaction with the subcontinental lithospheric mantle. Copyright

Zircon chemistry and magma mixing, SE China : in-situ analysis of Hf isotopes, Tonglu and Pingtan igneous complexes

Field relations and whole-rock geochemistry indicate that magma mixing has been important in the genesis of the late Mesozoic I-type igneous complexes at Pingtan and Tonglu in SE China. Morphological and trace-element studies of zircon populations in rocks from each of these complexes have defined several distinct growth stages [Mineral. Mag. (2001)]. In-situ LAM-MC-ICPMS microanalysis shows large variations in 176Hf/177Hf (up to 15 eHf units) between zircons of different growth stages within a single rock, and between zones within single zircon grains (up to 9 eHf units). These variations suggest that each of the observed magmas in both complexes developed through hybridisation of ≥2 magmas with different sources. Although this mixing has produced similar Sr and Nd isotopic compositions in the different rock types of each complex, the zircons have functioned as “tape recorders” and have preserved details of the assembly of the different magmas. In the Tonglu complex the most primitive magma is a mafic monzonite (preserved as enclaves), whose isotopic composition suggests derivation from the lower crust; rhyodacites, rhyolites and quartz diorites reflect the mixing of the monzonite with ≥2 more felsic magmas, derived from older crustal materials. In the Pingtan complex, zircons in a quartz diorite enclave suggest mixing between a crustal magma and a more primitive mantle-derived component. Zircons from granites and granodiorite enclaves indicate mixing between the quartz diorite and more felsic melts with lower 176Hf/177Hf. Major changes in 176Hf/177Hf correlate with discontinuous changes in the trace-element composition and morphology of the zircons, in particular the development of sector zoning that suggests rapid disequilibrium crystallisation. We suggest that the magma mixing recorded by the changes in 176Hf/177Hf occurred during transport in magma conduits. The in-situ analysis of Hf-isotopic stratigraphy in zircons is a new and powerful tool for the detailed study of magma generation processes.

Widespread Archean basement beneath the Yangtze craton

The age distribution of the crust is a fundamental parameter in modeling continental evolution and the rate of crustal accretion through Earth9s history, but this is usually estimated from surface exposures. The exposed Yangtze craton in eastern China consists mainly of Proterozoic rocks with rare Archean outcrops. However, the U-Pb ages and Hf isotope systematics of xenocrystic zircons brought to the surface in lamproite diatremes from three Proterozoic outcrop areas of the craton suggest the widespread presence of unexposed Archean basement, with zircon age populations of 2900–2800 Ma and 2600– 2500 Ma and Hf model ages of 2.6 to ca. 3.5 Ga or older. The zircons also record thermal events reworked on the craton ca. 2020 Ma (remelting of older crust) and 1000–850 Ma (addition of juvenile mantle material). The observation of deep crust significantly older than the upper crust will require revision of models for the rates of crustal generation through time.

Zircon U-Pb and Hf isotope constraints on the Mesozoic tectonics and crustal evolution of southern Tibet

The first in situ Hf and U-Pb isotope analyses of zircon separates from Mesozoic granites in southern Tibet identify a significant, previously unknown stage of magmatism. Igneous zircons (n = 34) from a granite within the Gangdese batholith show a weighted mean 206 Pb/ 238 U age of 188.1 ± 1.4 Ma and e Hf (T) (the parts in 10 4 deviation of initial Hf isotope ratios between the zircon sample and the chondritic reservoir) values between +10.4 and +16.8, suggesting predominantly Early Jurassic intrusive activity with a juvenile mantle contribution. Of 40 inherited zircons from two Cretaceous S-type granites in the northern magmatic belt, 23 delineate a slightly older 206 Pb/ 238 U age cluster between 188 and 210 Ma. These zircons have e Hf (T) values from −3.9 to −13.7, yielding crustal Hf model ages from ca. 1.4 to 2.1 Ga, suggesting a major episode of crustal growth in Proterozoic time and remelting of this crust in Early Jurassic time. Combining these with literature data, we interpret the Jurassic Gangdese magmatism as an early product of the Neo-Tethyan subduction that played a long-lasting role in the tectonic evolution of southern Tibet prior to the India-Asia collision.

An experimental study of Nb and Ta partitioning between Ti-rich minerals and silicate liquids at high pressure and temperature

Experimental determination of partition coefficients (D) for Nb and Ta between Ti-rich minerals (sphene, rutile, ilmenite and Ti-magnetite) and coexisting silicate liquids of basaltic andesite, andesite and trachyte composition, indicate that these elements partition strongly into sphene and rutile, and moderately into ilmenite and Ti-magnetite. The D values for sphene increase with decreasing temperature and with increasing SiO2 content of the coexisting liquid, but appear unchanged by pressure between 7.5 and 16 kb, and oxygen fugacity between magnetite-wustite and haematite-magnetite buffers. Also, DNbDTa. varies from 0.3–0.4 for sphene, increasing to 0.6–0.7 for rutile, 0.7–0.8 for ilmenite and 0.8 for Ti-magnetite. Thus fractionation involving these minerals may cause an increase in the NbTa ratio of any derivative liquid; this will be most noticeable where sphene is involved.

Relict refractory mantle beneath the eastern North China block: significance for lithosphere evolution

Xenolith-bearing Neogene basalts occur in Hebi county, at the southern end of the Taihangshan–Luliangshan paleo-rift zone in the North China block of the Sino–Korean craton. This locality lies on the North–South Gravity Lineament, which divides the craton into two geophysical zones. The spinel peridotite xenoliths hosted by the basalts can be divided into two groups based on the Mg# values of olivine. The whole-rock compositions of the low-Mg (Fo<91) xenoliths have high Al2O3+CaO (average 3.06 wt.%) and Na2O (average 0.19 wt.%), and low Mg/Si; they are similar to xenoliths from many localities in eastern China and other Phanerozoic volcanic areas. The dominant high-Mg (Fo≥92) group consists of harzburgites (66%) and depleted lherzolites (34%) with coarse-grained (mainly) and porphyroclastic microstructures, and high-Cr spinels (mean Cr#=0.51). The high-Mg xenoliths have low Al2O3+CaO (average 1.36 wt.%) and high Mg/Si, are in general strongly depleted in HREE, Ti, Zr and Y, and are compositionally similar to xenoliths in kimberlites from Archean cratons. The Archean lithospheric root beneath the eastern part of the Sino–Korean craton, which was sampled by Paleozoic kimberlites, was largely replaced by fertile Phanerozoic mantle during Mesozoic extension and subduction events. The high-Mg xenoliths are interpreted as relics of the Archean lithosphere, preserved locally at relatively shallow levels, and re-equilibrated to spinel facies in a regime of high heat flow caused by advective heat transport during extension. Their calculated mean room-temperature density (3.36 g/cm3) and Vp (8.39 km/s) are consistent with this interpretation and with geophysical data for the Hebi area. Regional geophysical data suggest that similar material may be widespread in the uppermost mantle west of the North–South Gravity Lineament, and more locally in the eastern part of the former craton.

Ti-rich accessory phase saturation in hydrous mafic-felsic compositions at high P, T

Abstract Experiments on hydrous mafic, intermediate and felsic compositions at 7.5–30 kbar and 900–1100°C allow broad delineation of TiO 2 contents for a range of silicate liquids coexisting with a Ti-rich accessory phase. Different starting compositions, including unadulterated fused natural rocks, and rocks enriched with various Ti-rich phases, give mutually consistent results. The TiO 2 content of liquids saturated in a Ti-rich accessory phase is shown to decrease with decreasing temperature, increasing pressure, increasing SiO 2 , alkali and rare-earth element content (at % level) of the liquid. Water content has little observable effect, while increasing oxygen fugacity lowers the TiO 2 content of liquid coexisting with a Ti-rich phase. A separate Ti-rich accessory phase may be expected to crystallize from liquids at 1000°C in the deep crust (pressures corresponding to 7.5–12 kbar) if the TiO 2 content exceeds ∼ 3.0 wt.% for mafic, ∼ 1.8 wt.% for intermediate, and ∼ 1.2 wt.% for felsic compositions. At the same depth but at 950°C, these TiO 2 values change to ∼ 1.7, 1.2 and 0.9 wt.%, respectively. At upper-mantle depths (20–30 kbar) the corresponding TiO 2 saturation levels at 1000°C are ∼ 2.0, 1 and 0.7 wt.%, respectively. In natural igneous rock series the contrasting TiO 2 SiO 2 systematics in alkaline and tholeiitic series compared with calc-alkaline series are well established, as are the characteristic, though non-unique, low TiO 2 values for rock series in convergent plate regions. The present results allow experimental confirmation of observed and/or predicted crystallization of Ti-rich phases in these rock series. They also provide reasonable constraints on conditions under which TiO 2 contents of convergent plate magmas may be buffered by a residual Ti-rich phase in the source region, and indicate that TiO 2 contents of mafic parent magmas in these areas are unlikely to be controlled in such a manner. In contrast, it is almost inevitable that felsic magmas, generated by partial melting of the lower crust, are saturated with respect to a Ti-rich phase in their source regions.

New insights into the Re–Os systematics of sub-continental lithospheric mantle from in situ analysis of sulphides

Abstract The Os isotopic compositions of mantle rocks generally are considered to be established during melt-depletion events and to be robust to subsequent disturbances (e.g. metasomatism). Consequently, Os isotopes are used to date the main melting event that a mantle section has undergone, i.e. transformation of fertile asthenospheric material into a depleted, buoyant lithosphere. However, Os resides almost entirely in Fe–Ni–Cu sulphides, which can be very mobile under mantle conditions. In situ laser ablation multi-collector ICP-MS measurement of Re/Os isotopic ratios in sulphides from spinel peridotite xenoliths demonstrates that whole-rock Os-isotope signatures record the mixing of multiple sulphide populations. Sulphides residual after melting events have unradiogenic Os isotopic compositions reflecting ancient melt depletion; those introduced by later metasomatism events contain more radiogenic Os. Therefore, the whole-rock Os isotopic signature can be strongly altered by metasomatic processes, and studies of mantle-derived xenoliths show that such disturbance is quite common in the lithospheric mantle. Because melt-depletion ages estimated from individual sulphide inclusions are systematically older than those obtained from whole-rock analysis, caution is essential in the interpretation of the Os model ages derived from whole-rock analysis, and their use and abuse in geodynamic models. This work suggests that sulphide could become a key phase in unravelling the formation and evolution of the lithosphere.

Quantitative analysis of trace element abundances in glasses and minerals: a comparison of laser ablation inductively coupled plasma mass spectrometry, solution inductively coupled plasma mass spectrometry, proton microprobe and electron microprobe data

Many geological, environmental and industrial applications can be enhanced through integrated microbeam and bulk geochemical determinations of major and trace element concentrations. Advantages ofin situ microanalysis include minimal sample preparation, low blanks, information about the spatial distribution of compositional characteristics and the ability to avoid microscopic inclusions of foreign material. In this paper we compare trace element data obtained by laser ablation ICP-MS, solution ICP-MS, electron microprobe analysis and proton microprobe analysis for a variety of silicate glasses and minerals. New determinations for 36 trace elements in BCR-2G, a microbeam glass standard, are presented. Results obtained by the various microbeam and solution methods agree well for concentrations ranging over several orders of magnitude. Replicate analyses of BCR-2G demonstrate an analytical precision of 2–8% relative (1σ) for all elements by laser ablation ICP-MS and ≤3% by solution ICP-MS, except for Li (5%). These data emphasize the utility of laser ablation ICP-MS as a quantitative microbeam technique capable of rapid, precise determinations of sub-ppm trace element abundances in a variety of targets.

Garnet geotherms: Pressure‐temperature data from Cr‐pyrope garnet xenocrysts in volcanic rocks

The temperatures and pressures of equilibration of single peridotitic garnet xenocrysts are estimated using a combination of major- and trace-element data, determined using electron microprobe (EMP) and proton-induced X ray emission (PIXE). This new method enables the use of xenocrysts found in kimberlites and other volcanic rocks to determine the local paleogeotherm at the time of eruption of the magma which sampled and transported the xenocrysts. The “Ni thermometer” of Griffin et al. [1989], based on the strong temperature dependence of the partitioning of Ni between garnet and olivine, is refined using an expanded database. Pressure is calculated from garnet composition using an algorithm that combines a modification of the geobarometer of Nickel [1989], based on Cr solubility in coexisting garnet and orthopyroxene, with the composition of a hypothetical coexisting orthopyroxene. The orthopyroxene composition is estimated by inverting the geothermometry equations of Gasparik [1987], Brey and Kohler [1990], and Harley [1984], and combining these with empirical relationships describing Cr in orthopyroxene in Cr-saturated peridotite (chromite present). The derived pressure (PCr) gives the equilibration pressure of peridotic garnets provided they were in equilibrium with chromite; garnets from Cr-undersaturated rocks will produce underestimates of pressure. Therefore, the locus of maximum PCr at a given TNi defines the “garnet geotherm”, and provides a method for the determination of paleogeotherms based solely on PIXE and EMP analyses of garnet grains in concentrates. The assumption of coexisting chromite is tested by comparing the temperature distributions of garnets and chromites from the same concentrate. Chromite equilibration temperature is estimated using the “Zn thermometer”, based on the strong temperature dependence of the partitioning of Zn between chromite and olivine. This thermometer is calibrated against the new Ni thermometer using a suite of garnet-chromite intergrowths. The garnet geotherm technique provides an estimate of the geotherm with an accuracy comparable to xenolith-derived geotherms and provides a means of mapping the thermal state of the lithosphere where xenoliths are rare or absent.