SHRIMP U-Pb xenotime geochronology and its application to dating mineralisation, sediment deposition and metamorphism

Abstract

Diagenetic and hydrothermal xenotime typically occurs in rocks as tiny crystals (≤20 μm), either individually or as outgrowths on a zircon substrate. Currently only large radius ion microprobes such as the SHRIMP or Cameca 1270/1280 have the high sensitivity and spatial resolution necessary to analyse these crystals for U-Th-Pb isotopes. However, such analyses are prone to significant matrix effects (ME) related to the large natural range of U (0 to ~9 wt%) and REE (ΣREE: -12 to 22 wt%) abundances in xenotime. Consequently, the ²⁰⁶PbP²³⁸U calibration procedure for xenotime differs significantly from that employed for SIMS dating of zircon. Contrasts in U, and to a lesser extent IREE, contents between the primary calibration standard and unknown xenotime can result in SHRIMP 2⁰6Pb/218U-2⁰8Pb/232Th ME of up to -20%. The matrix correction technique developed requires the concurrent analysis of three xenotime standards with a range of U and IREE concentrations on a session-by-session basis. The ²⁰⁶rb/²³⁸U-²⁰⁸Pb/²³²Th ME is monitored by the analysis of two secondary standards, a high ΣREE xenotime (BS l) and a high U xenotime (Z6413). Additionally, the chemical composition of each spot is determined by EPMA (WDS) prior to SHRIMP analysis. Each spot is corrected for ME by defining a series of simultaneous equations that relate the fractional ²⁰⁶Pb/²³⁸U-²⁰⁸ Pb/²³²Th ME of the secondary standards to their U and ΣREE concentration contrasts with the primary calibration standard (MG1). On average, every 1 wt% contrast in U between the primary calibration standard and the unknown results in a ~11.9% difference in the ²⁰⁶Pb/²³⁸U and ²⁰⁸Pb/²³²Th ratios, whereas a 1wt% contrast in REE results in a difference in the ²⁰⁶Pb/²³⁸U and ²⁰⁸Pb/²³²Th ratios of ~0.9%. SHRIMP RG was used for these experiments because the analyses on that instrument are not prone to the molecular interferences or 'scattered ions' that affect the ²⁰⁴Pb peak when xenotime is analysed on SHRIMP II. Matrix uncorrected ²⁰⁶Pb/²³⁸ U ratios were determined from the raw ²⁰⁶Pb/²7°(UO₂⁺) ratios as suggested for zircon analyses by Stem & Amelin (2003). Additionally, matrix uncorrected ²⁰⁸Pb/²³²Th ratios were determined from the raw ²⁰⁸Pb/²⁴⁸ (ThO⁺) ratios, however, this calibration appears to only be effective for xenotime with >~1000 ppm Th.The technique developed is broadly similar to the SHRIMP xenotime U-Th-Pb correction procedure proposed by Fletcher et al. (2004). Whereas Fletcher and others related SHRIMP xenotime ²⁰⁶Pb/²³⁸U-²⁰⁸Pb/²³²Tu ME to contrasts in U, Th and ΣREE, 1his study indicates 1hat the effect of Th on 1he ²⁰⁶Pb/²³⁸U-²⁰⁸ Pb/²³ Tu ME is minor to insignificant. It appears likely that for xenotime, it is the matrix sensitivity of the emission of the Pb+ secondary ions, not U or Th species that is the principal cause of the ²⁰⁶Pb/²³⁸U-²⁰⁸Pb/²³²Th ME. Using the new matrix correction procedures developed here, it is possible to measure ²⁰⁶ Pb/²³⁸U and ²⁸Pb/²³²Th ages of Phanerozoic xenotime with an accuracy and precision of about 2% (95% confidence). Three application studies of SHRIMP U-Pb xenotime geochronology using the matrix correction procedures developed were undertaken as a part of this study as well as a SHRIMP U-Pb detrital zircon study of the principal sedimentary units within the Tanami Basin, central Australia. SHRIMP U-Pb analyses of xenotime that occurs as outgrowths on detrital zircon from a sample of quartzite from the Serra da Mesa Group, central Brazil, has the same Neoproterozoic age within error to that of SHRIMP U·-Pb monazite analyses from the same sample, both giving ²⁰⁶Pb/²³⁸U ages of ~570 Ma. This latest Neoproterozoic age may be related to metamorphism associated with the final stages of the Brasiliano Orogeny. Further U-Pb studies are required on oilier rocks from this region to establish its regional extent. The identical SHRIMP ²⁰⁶Pb/²³⁸U ages for both monazite and xenotime in this study lend strong support to the matrix correction protocols developed. SHRIMP U-Pb analyses of hydrothermal xenotime from the Callie and Coyote Au deposits in the Tanami region has demonstrated that they formed between ~1.81 and ~1.80 Ga. The -1.81 Ga age for the Callie deposit contrasts with the results from an earlier 40 Ar/39 Ar study of hydrothermal biotite from the Callie deposit by Fraser (2002), which suggested that mineralisation occurred at ~ 1.72 Ga. Mineralisation between ~1.81 and ~1.80 Ga occurring in the Tanami region is coincident with the ~1.81 to ~1.79 Ga Stafford event which was a period of widespread magmatism across much of the North Australia Craton in which Scrimgeour (2006) suggested was linked to a long-lived north-dipping subduction system active at the south-east margin of the craton. A preliminary isotopic dating investigation of the Molyhil scheelite-molybdenite skarn deposit in the north-eastern Arunta region, central Australia has determined that it formed from fluids associated with the crystallisation of the Marshall Granite during the ~1. 73 to ~1. 72 Ga Strangways Oregeny. Rhenium-Osmium dating of ore stage molybdenite has an age of 1720.7 ± 5.9 Ma which is also coincident wifh skarn related hornblende which has an ⁴⁰Ar/⁴⁰Ar age between ~1.72 and ~1.73 Ga (G. Fraser, unpublished data; Geoscience Australia). Importantly for this study is that one of the xenotime grains has a concordant SHRIMP U-Pb ²⁰⁷Pb²⁰⁶Pb age of 1714 ± 26 Ma which is well within error of the ages determined by the other two isotopic methods adding further support to the SHRIMP U-Pb analytical protocols and ²⁰6Pb/²38U matrix correction techniques developed during this PhD. Younger xenotime components from the Molyhil sample investigated have SHRIMP ²⁰⁶ Pb/²³⁸U ages of ~760 Ma and ~650 Ma and probably crystallised in response to far-field tectonothermal events.