A low-contamination method for hydrothermal decomposition of zircon and extraction of U and Pb for isotopic age determinations

Matrix-matched calibration by natural zircon standards and analysis of natural materials as a reference are the principle methods for achieving accurate results in microbeam U–Pb dating and Hf isotopic analysis. We describe a new potential zircon reference material for laser ablation ICP-MS that was extracted from a potassic granulite facies rock collected in the southern part of the Bohemian Massif (Plešovice, Czech Republic).

Data from different techniques (ID-TIMS, SIMS and LA ICP-MS) and several laboratories suggest that this zircon has a concordant U–Pb age with a weighted mean ²⁰⁶Pb/²³⁸U date of 337.13 ± 0.37 Ma (ID-TIMS, 95% confidence limits, including tracer calibration uncertainty) and U–Pb age homogeneity on the scale used in LA ICP-MS dating. Inhomogeneities in trace element composition due to primary growth zoning prevent its use as a calibration standard for trace element analysis. The content of U varies from 465 ppm in pristine parts of the grains to ~ 3000 ppm in actinide-rich sectors that correspond to pyramidal faces with a high degree of metamictization (present in ca. 30% of the grains). These domains are easily recognized from high intensities on BSE images and should be avoided during the analysis. Hf isotopic composition of the Plešovice zircon (> 0.9 wt.% Hf) is homogenous within and between the grains with a mean ¹⁷⁶Hf/¹⁷⁷Hf value of 0.282482 ± 0.000013 (2SD). The age and Hf isotopic homogeneity of the Plešovice zircon together with its relatively high U and Pb contents make it an ideal calibration and reference material for laser ablation ICP-MS measurements, especially when using low laser energies and/or small diameters of laser beam required for improved spatial resolution.

The role of the standard is critical to the derivation of reliable U–Pb zircon ages by micro-beam analysis. For maximum reliability, it is critically important that the utilised standard be homogeneous at all scales of analysis. It is equally important that the standard has been precisely and accurately dated by an independent technique. This study reports the emergence of a new zircon standard that meets those criteria, as demonstrated by Sensitive High Resolution Ion MicroProbe (SHRIMP), isotope dilution thermal ionisation mass-spectrometry (IDTIMS) and excimer laser ablation–inductively coupled plasma–mass-spectrometry (ELA–ICP–MS) documentation. The TEMORA 1 zircon standard derives from the Middledale Gabbroic Diorite, a high-level mafic stock within the Palaeozoic Lachlan Orogen of eastern Australia. Its ²⁰⁶Pb/²³⁸U IDTIMS age has been determined to be 416.75±0.24 Ma (95% confidence limits), based on measurement errors alone. Spike-calibration uncertainty limits the accuracy to 416.8±1.1 Ma for U–Pb intercomparisons between different laboratories that do not use a common spike.

A-type granites are widely distributed in northeastern China (NE China). They were emplaced during three major episodes (the Permian, late Triassic to early Jurassic, and early Cretaceous) and evolved in different tectonic regimes.

According to their mineralogical and geochemical characteristics, two subgroups of A-type granites (aluminous and peralkaline) can be recognized. The peralkaline subgroup contains alkali mafic minerals, such as riebeckite, arfvedsonite and sodic pyroxene, while the aluminous subgroup contains annite and Fe-rich calcic- or sodic-calcic amphibole. With respect to the aluminous subgroup, the peralkaline granites contain higher Rb, Ga and total rare earth elements (REE), but lower MgO, CaO, Al₂O₃, Ba and Sr. Based on the discrimination criteria of Eby [Geology 20 (1992) 641], the Permian and late Triassic to early Jurassic A-type granites belong to the A₂ (post-orogenic) type, whereas the early Cretaceous granites are of the A₁ (anorogenic) type.

Nd isotopic compositions of these A-type granites indicate their derivation from a dominantly juvenile crustal source. Their origin is thought to have involved partial melting of an underplated lower crustal source. Because the generation of A-type granites requires high melting temperature, we propose models involving slab break-off, lithospheric delamination and extension.

The Permian A-type granites have the same age range as those in eastern Junggar, southern Mongolia and central Inner Mongolia. They occur along a major suture and form a narrow belt between the north China and Siberian Cratons. We suggest that their formation was associated with post-collisional slab break-off. The late Triassic to early Jurassic A-type granites are likely to be the product of lithospheric delamination after the final collision of the major crustal blocks in the late Paleozoic to early Triassic. The early Cretaceous A-type granites have an anorogenic affinity and were possibly associated with rifting in eastern China at this time, associated with the onset of paleo-Pacific subduction. Consequently, we conclude that the A-type granites in NE China were generated at three different times, involving multiple processes operative in different tectonic environments.

The compilation and major element composition of the “North American shale composite” (NASC) are reported for the first time, along with redeterminations for the REE and selected other elements by modern, high precision analytical methods. The NASC is not strictly of North American origin; 5 of the constituent samples are from Africa and Antarctica, and 15 are from unspecified locations. The major element composition of the NASC compares quite closely with other average shale compositions. New analyses of the NASC document that significant portions of the REE and some other trace elements are contained in minor phases (zircon and possibly other minerals) and that their uneven distribution in the NASC powder appears to have resulted in heterogeneity among analyzed aliquants. The results of this study show that the REE distributions of detrital sediments can be dependent to some extent on their minor mineral assemblages and the sedimentological factors that control these assemblages. Consequently, caution should be exercised in the interpretation of the REE distributions of sediment samples as they may be variable and biased relative to average REE distribution of the crustal rocks supplying detritus. These effects appear to be largely averaged out in sediment composites, with the result that their REE distributions are more likely to be representative of their provenances.

Abrasion combined with an improved paramagnetic separation technique eliminates 90 to 100 percent of discordance so that ages of unprecedented accuracy (±1 to 3 m.y.) can be achieved for virtually all 2700 m.y. old zircon populations from plutonic or volcanic rocks. The procedures work even better for younger zircons. Besides removing outer layers that may have been leached, high-U parts are preferentially removed by abrasion because they are softened by radiation damage. Altered and cracked zircons also tend to be eliminated.

In most cases, the new concordant data move up the line established by previously analysed paramagnetic fractions but a number of anomalies have been found where old data give upper intersections that are in error by as much as 25 m.y. Reducing or eliminating paramagnetically correlated Pb loss greatly enhances our ability to define mixing lines for igneous or metamorphic rocks when two ages of zircon are present.

The abrasion technique allows detection of an inherited component if it exists by enhancing the sample in core material. Abrasion in many cases removes about 80 percent of the common lead, thus allowing a direct evaluation of this component.

When the outer parts of grains are removed, the correlation between magnetic susceptibility and uranium content is maintained but the usual correlation of uranium level with lead loss is reduced or eliminated. Therefore, only near surface uranium is involved in the classic discordance versus uranium level correlation of Silver (1963).