⁴⁰ Ar/ ³⁹ Ar age constraints on ore deposition and cooling of the Bushveld Complex, South Africa

Abstract

The UG-2 chromitite layer in the upper critical zone of the Rustenburg Layered Suite is one of the prominent platinum-group element enrichment zones of the c. 2.06 Ga Bushveld Complex (South Africa). Eleven biotite crystals were selected from a production concentrate from the Impala Platinum mine and degassed incrementally with a CO 2 laser. The apparent age spectra display widely varying characteristics, ranging from essentially 100% concordant to strongly discordant (in some cases undulatory) as a result of various degrees of chloritization of the grains. The least discordant spectra (seven grains, 90-100% 39 Ar released) yield integrated and plateau ages ranging from 2037 to 2049 Ma (relative to the Fish Canyon sanidine standard at 28.02 Ma) and share a weighted mean age of 2042.4 AE 3.2 Ma (2ó, excluding systematic errors). This age is slightly younger than existing constraints from U-Pb dating (c. 2055(c. -2061, which seem to indicate a cooling rate of around 20 8C/Ma. However, the recently inferred c. 1% bias between currently used calibrations of the 40 Ar/ 39 Ar and U-Pb systems implies a very rapid cooling rate for the Bushveld Complex between 700 and ,500 8C. The possibility of subtle bias as a result of recoil artefacts must be considered. In all cases, the heterogeneous age spectra yielded older apparent plateau ages, revealing the likelihood for erroneous conclusions to be drawn from less complete data sets.

Geochronology of Layered Intrusions

“…Both spectra have low apparent ages in the earliest stages of gas released (1-2 % for SA04-11A, 14 % for SA04-13). This is consistent with the early release of 39 Ar at the lowest extraction temperatures from minor chlorite interlayers in biotite that are artificially enriched in 39 Ar during irradiation and transfer from the surrounding biotite (Nomade et al 2004). The inverse isochron ages, 1994.4 ± 7.4 Ma and 2000.0 ± 8.9 Ma, are indistinguishable from their respective plateau ages (Fig.…”

“…The geochronologic evidence for rapid cooling supports the results of thermal modeling of the Bushveld Complex indicating that the 8 km-thick layered intrusion was emplaced and cooled from magma liquidus temperatures down to the Curie temperature of magnetite (580 °C) in less than 700,000 years (Cawthorn and Walraven 1998;Cawthorn and Webb 2013). Assuming that their respective closure temperatures are similar, the 40 Ar/ 39 Ar results from Nomade et al (2004) for biotite from the UG2 chromitite, Western Limb (2042.4 ± 3.2 Ma), and from Cassata et al (2009) for biotite from a gabbro near the MG-1 chromitite layer in the Critical Zone (2042 ± 2 Ma), are both consistent with the U-Pb rutile ages when an −0.5 % bias between the calibrations of the 40 Ar/ 39 Ar and U-Pb systems is applied (e.g., Min et al 2000;Nomade et al 2004;Renne et al 2010) (Fig. 1.17).…”

“…The presence of micro-fractures and evidence for interaction with an aqueous fluid (e.g., hydrous silicates, calcite) are particularly relevant to interpreting the chronologic information recorded by biotite. The relatively low closure temperature for Ar diffusion in biotite (~ 350-450 °C depending on cooling rate and diffusion dimension: Grove and Harrison 1996;Lee 1995;Nomade et al 2004), combined with its weakly bound sheet structure, allow for the possibility of open-system behavior, argon loss, and resetting during fracture-controlled circulation of aqueous hydrothermal fluids. The anomalously young ca.…”

The Bushveld Complex was emplaced and cooled in less than one million years – results of zirconology, and geotectonic implications

“…Uranium-lead ages obtained from zircon, TIMS) are commonly older and more precise (2058.9 ± 0.8 Ma to 2054.4 ± 1.3 Ma; Buick et al, 2001;Scoates and Friedman, 2008;Olsson et al, 2011) than Re-Os, Pt-Os and Ar-Ar ages. The latter comprise whole-rock isochron and mineral data which scatter between 2044 ± 3 Ma and 1995 ± 50 Ma (Schoenberg et al, 1999;Nomade et al, 2004;Reisberg et al, 2011;Coggon et al, 2012). Some U-Pb zircon ages obtained by SHRIMP (Sensitive High Resolution Ion Microprobe) also reveal a high degree of scatter and overlap with the CA-ID-TIMS and Ar-Ar data (Yudovskaya et al, 2013).…”

Reassessing the uranium decay constants for geochronology using ID-TIMS U–Pb data

“…Numerous studies (Min et al, 2000Nomade et al, 2004;Renne, 2000;Renne et al, 1998a;Schmitz and Bowring, 2001;Villeneuve et al, 2000) have noted that 40 Ar/ 39 Ar dates are systematically younger than U-Pb dates from rapidly cooled rocks and do not overlap with U-Pb dates if one ignores systematic uncertainties. Min et al (2000), Renne (2000), and Renne et al (1998a) suggest that much of the bias can be accounted for by inaccuracies in the 40 K decay constant and physical constants, which differ by $2% from those used in other scientific communities.…”