Isotope systematics and shock-wave metamorphism: II. U-Pb and Rb-Sr in naturally shocked rocks; the Haughton Impact Structure, Canada

Abstract

Minerals and whole rocks from crystalline fragments collected in the allochthonous polymict breccia of the 23-Ma-old Haughton impact crater on Devon Island (Arctic Canada) were analysed for U-Pb, K and Rb-Sr in order to constrain degrees of shock-wave induced perturbations in these systems. The results show that shock-wave metamorphism at ≥50 GPa with onset of total melting does not significantly affect the U-Pb chronometers in monazite and zircon. Monazite preserved its age of crystallization during high-grade metamorphism at 1928.2 (+2.1/−2.0) Ma (2 σ-errors) as dated by the upper intercept of 2–9% discordant monazites. The same analyses yield a lower intercept age of 277.5 (+35.1/ −35.0) Ma in the concordia diagram, demonstrating that discordancy is not a direct effect of shock-wave metamorphism but a product of time integrated Pb-diffusion prior to the impact event, which occurred 23 Ma ago. This constrains maximal shock-induced Pb-loss to less than 3% including possible alteration related Pb diffusion from the shock damaged grains. In 1903.1 (+1.4/−1.1) Ma old monazite from an unshocked reference gneiss, recent Pb-loss also reaches 3% corroborating the occurrence of weak alteration-related U Pb fractionation. The data substantiate that U-Pb discordancy patterns are not a reliable chronometer to date impact events. Small rock fragments and a coherent block of shock-stage I to III/VI consist of heterogeneous, unequilibrated Rb-Sr sub-systems on the 10 μm-scale, with drastically varying I Sr. Shock-related Rb Sr fractionation is very limited, and, in the large block (stage III), three biotite fractions yield an Rb-Sr isochron age of 1791 ± 34 Ma (2 σ), which is undistinguishable from the 1814 ± 8 Ma age of biotite in the unshocked basement gneiss. In contrast to this m 3-sized fragment, Rb-Sr systematics in a fist-size sample of shock stage III do not yield any reliable age information due to small-scale (< 1 cm 3) isotopic disequilibrium. It appears that such small-scale disequilibrium is essentially due to a take-up of radiogenic 87Sr in feldspar-glass from biotite during shock-wave passage. This prevents dating of the impact events by the Rb-Sr “internal-isochron” method, and it can be concluded that total rock melting is the first condition necessary to reset the Rb-Sr chronometer in shocked crystalline rocks.