Re-Os geochronology of base metal sulfides from cratonic mantle xenoliths : Case study from Somerset Island (Canada) and method development

Robust and reliable time constraints are necessary to infer the formation and evolution of the cratonic mantle. Due to the large fractionation of Re from Os during mantle melting, the Re-Os decay system has been largely used for dating the melting event that led to the formation of the subcontinental lithospheric mantle (SCLM). In mantle rocks Re, Os and the other highly siderophile elements (HSE: Ru, Rh, Pd, Re, Os, Ir, Pt, and Au) are controlled by base metal sulfides (BMS), which can be residual phases of partial melting processes or can be re-introduced in mantle rocks during metasomatism. The present study aimed at the improvement and enhancement of our ability to use the Re-Os system and the HSE to unravel geological processes recorded in BMS. The contribution of this work is twofold because it provides new data on natural samples as well as a novel analytical technique for future applications. In the first part of this dissertation (Chapter 1) some basic concepts are introduced to make the reader more familiar with the topics encountered in the next sections. This includes an overview of the HSE behavior in terrestrial reservoirs and the explanations of the geochemical tools that will be used in the following sections. The second part of this dissertation (Chapter 2) is focused on the investigation of partial melting and metasomatic processes recorded in four mantle xenoliths from Somerset Island (Rae craton, Canada). After textural and mineralogical investigations, individual BMS grains were micro-sampled and analyzed for 187Os/188Os. The two xenoliths with the most metasomatic HSE signature (e.g. suprachondritic Pd/Pt) are distinguished for the high BMS modal abundance, the occurrence of large interstitial BMS grains, and the extreme 187Os/188Os variation measured in BMS grains (187Os/188Os = 0.172-0.108). Archean Re-depletion model ages (TRD) are recorded in BMS grains from three different xenoliths, suggesting a main formation of the SCLM at 2.7-2.8 Ga, in association with the local Rae greenstone belts. A similar scenario was proposed for the nearby Slave craton, which confirms that different terrains of the Canadian Shield share a similar Neoarchean history. At the whole rock scale, the TRD age of 2.7-2.8 Ga is clearly recorded only in one xenolith with residual HSE signature (i.e. subchondritic Pt/Ir, Pd/Pt, and Re/Pd). This supports and further stresses that: 1) whole rock TRD ages should be used carefully in xenoliths with metasomatic HSE signature, and 2) single grain BMS can record the age of formation of the SCLM even in heavily metasomatized mantle xenoliths. Single BMS grains yielded two distinct Paleoproterozoic TRD ages (~1.9 and ~2.2 Ga) that are not resolvable at the whole rock scale. The two TRD ages are consistent with a scenario where metasomatic BMS were introduced in the SCLM during a first phase of rifting of the Slave from the Rae craton (2.2 Ga) and a later collision of the two cratons (1.9 Ga, Thelon-Talston orogeny). In the third part of this dissertation (Chapter 3) a novel analytical method is proposed to analyze 187Os/188Os along with Ru, Pd, Re, Os, Ir, and Pt concentrations in individual µg-weight BMS grains. To set up this method, two Fe-Ni sulfides were synthetized and independently characterized for HSE content and 187Os/188Os. Fragments of the two sulfides were used to test different digestion and separation methods. It is here shown that a simultaneous digestion and Os extraction yields inaccurate Os concentrations. The improved procedure proposed in this study includes BMS digestion in HBr + HCl, Os micro-distillation, and cation resin separation of Ru, Pd, Re, Ir and Pt. The 187Os/188Os ratio and the HSE concentrations are measured by mass spectrometry (N-TIMS and SF-ICP-MS). The independently determined HSE concentrations are reproduced by this technique with differences < 10% for Pd, Os and Ir, and < 20% for Ru, Re, and Pt. The 187Os/188Os ratio is indistinguishable within the analytical precision (2SD ~ 0.1%). Owing to the chemical separation of the analytes, the proposed procedure overcomes many of the analytical issues encountered during LA-ICP-MS analyses (e.g. 187Re isobaric interference on 187Os and matrix effects). Moreover, the analysis of the entire grain, avoid any sampling bias related to the complex mineralogical assemblage typically observed in natural BMS. As shown in this dissertation, BMS grains record a multitude of magmatic and metasomatic processes that cannot be individually discriminated at the whole rock scale. The comprehension of these processes represents an exciting challenge as it will improve our ability of using the Re-Os system and, ultimately, to constrain the timing of mantle dynamics. Coupled HSE and 187Os/188Os investigations in individual BMS grains will provide an essential tool towards this goal.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.