Mn-Cr chronology and trace element systematics of olivine from angrite and pallasite meteorites

Abstract

{u2075}{u00B3}Mn-{u2075}{u00B3}Cr short-lived nuclide dating of certain meteoritic materials can be achieved in situ using Secondary Ion Mass Spectrometry (SIMS) as an alternative to bulk methods. This thesis reports on the Mn-Cr systematics of olivine from two differentiated meteorite classes: angrites and pallasites. Much of the previously published SIMS Mn-Cr data relies on reference materials that are not compositionally matched to unknowns, particularly with respect to olivine (e.g. Mg-rich San Carlos olivine is used alongside Fe-rich meteoritic olivine). Previous studies may suffer from systematic biases due to the differing sensitivities of SIMS to Mn and Cr. To assess olivine Mn-Cr relative sensitivity factors (RSFs), synthetic olivines with varying Mg-Fe-Ca components were produced using gas-mixing furnaces and piston-press apparatus, and analysed by SIMS and LaserAblation Inductively-Coupled-Plasma Mass-Spectrometry (LA-ICP-MS). RSF is correlated with Fe-content of olivine; San Carlos olivine is therefore an inappropriate standard for most SIMS Mn-Cr work. For angrites, a mixing model was used to account for RSF, using three of the synthetic olivines. Two quenched angrites (D'Orbigny and Sahara 99555) and two plutonic angrites (NWA 4590 and 4801) were analysed with Sensitive High-mass Resolution Ion Micro-Probe (SHRIMP) Reverse Geometry (-RG) and SHRIMP-II. The initial {u2075}{u00B3}Mn/{u2075}{u2075}Mn for D'Orbigny and Sahara 99555 was found to be 3.52 ({u00B1}0.16) x 10-6 and 3.57 ({u00B1}0.22) x 10-6 , which is in agreement with some previous work but not with the SIMS study of Sugiura et al. (2005). NWA 4590 and 4801 were found to have initial {u2075}{u00B3}Mn/{u2075}{u2075}Mn of 1.00 ({u00B1}0.27) x 10-6 and 0.52 ({u00B1}1.2) x 10-6 respectively; less precise than but in agreement with previous work. To check if the difference between this study and Sugiura et al. (2005) represents inter-laboratory bias or meteorite heterogeneity, the same piece of D'Orbigny as analysed in that study was reanalysed using SHRIMP-RG; initial {u2075}{u00B3}Mn/{u2075}{u2075}Mn of 3.61 ({u00B1}0.39) x 10-6 was found, still in disagreement with Sugiura et al. (2005). The source of bias remains elusive; it is not due to differing RSF since both studies found similar values for this variable. For pallasite olivine, trace-elemental maps were made using LA-ICPMS to provide context for Mn-Cr measurements. Two contrasting pallasites were selected: Brahin, which has fragmental olivine, and Brenham, which has rounded olivine. Both olivine types are normally zoned for divalent elements Ni and Co, but non-divalent elements AI, Cr, Ti are heterogeneously distributed, often with no relationship to grain morphology. In Brenham, these complex elemental structures are superimposed on normal zoning. Preservation of these structures probably hinges on slow diffusion of AI, which other non-divalent elements associate with to maintain charge-balance. SIMS measurements indicate initial {u2075}{u00B3}Mn/{u2075}{u2075}Mn of -2.25 ({u00B1}4.9) x 10-6 and 2.55 ({u00B1}0.62) x 10-5 for Brahin and Brenham respectively; the latter is an unreasonably high value and is unlikely to reflect Mn-Cr chronology. This apparent value could reflect loss of {u2075}{u2075}Mn after decay of {u2075}{u00B3}Mn, although the geochemical behaviour of Mn is unfavourable for such a model. Instead, excess {u2075}{u00B3}Cr* derived from Mn-rich phosphates originally present in the metallic groundmass could have been introduced into olivine rims during the grain growth and annealing process which formed rounded olivine. Show more