The U, Th and Pb elemental and isotope compositions of mantle clinopyroxenes and their grain boundary contamination derived from leaching and digestion experiments

Minerals from peridotites are known to be affected by trace element contamination on their grain boundaries. In this contribution we investigate the extent and origin of exogenous contamination associated with mantle clinopyroxenes from various localities (Middle Atlas, Beni Bousera Morocco], Pyrenean Massif, Massif Central France]) and test the efficacy of different leaching methodologies used to remove this contamination. In doing so we present new U-Th-Pb (double-spike) isotope and trace elemental data of clinopyroxenes and their leachates from spinel-facies sub-continental lithospheric mantle (SCLM, n = 18).Sequential leaching and dissolution of one clinopyroxene separate shows that multiple and short leaching attacks with dilute HCl at moderate temperatures (e.g., 120 °C) interspersed with rigorous ultra-pure water washes do not induce elemental fractionation and are sufficient to remove grain-boundary contamination. Short attacks with very dilute mixtures of HF and HCl induced strong parent/daughter elemental fractionation and significant elemental loss prior to the clinopyroxene digestion with HF/HNO₃. Such leaching is not suitable for studies investigating parent/daughter elemental or isotope ratios of mantle clinopyroxenes. Fluoride co-precipitates that formed in the presence of moderately dilute HF, used during leaching and typical HF/HNO₃ digestions, are an important sink for all trace elements studied here and lock up at least 60% of the trace elements considered.A suite of clinopyroxene-leachate pairs from Moroccan peridotites confirm extreme grain-boundary contamination and show that up to 65% of U, 82% of Th and 91% of Pb, respectively, are of exogenous origin. Pb isotopes of all leachates considered and nearly all reconstructed unleached clinopyroxenes have highly positive Δ7/4 and ²⁰⁷Pb/²⁰⁴Pb-²⁰⁶Pb/²⁰⁴Pb systematics reminiscent of enriched mantle (particularly EM II), whereas the corresponding extensively leached clinopyroxenes have very different Pb isotope systematics (verified by replicate digestions). Furthermore, the leachates are often marked by remarkably high ²³²Th/²³⁸U (leachate ? bulk silicate Earth) in addition to ²³⁸U/²⁰⁴Pb and ²³²Th/²⁰⁴Pb that are seemingly consistent with depleted mantle (DM) and high ²³⁸U/²⁰⁴Pb (HIMU) mantle mixing. After leaching, clinopyroxenes may have highly elevated ²³⁸U/²⁰⁴Pb and ²³²Th/²⁰⁴Pb that are distinct from typical DM- and HIMU mantle, whereas corresponding leachates bear strong resemblance to anthropogenic Pb recovered from rainwater, snow and aerosols.In the light of the great potential of Pb isotope contamination associated with mantle clinopyroxene we have compiled literature data of mantle peridotites. These data seem to suggest that the SCLM is nearly devoid of HIMU signatures but ostensibly represents an important repository for enriched mantle signatures. However, the dominance of enriched mantle is not verified when ⁸⁷Sr/⁸⁶Sr and ²⁰⁸Pb/²⁰⁴Pb ratios for the same database are considered. We highlight that the scarcity of HIMU-like SCLM and the alleged dominance of EM in continental roots may be reconciled if the current Pb isotope database of SCLM contains samples that were insufficiently leached prior to digestion and thus are contaminated by anthropogenic Pb. We stress the importance of replicate digestions, adequate leaching and leachate analysis to verify the U-Th-Pb elemental and isotope systematics of any given mantle (-derived) sample.