The noble gas isotope composition of the mantle can provide unique insights into the origin and evolution of volatile elements on Earth. Xenon isotopes combine primordial signatures with contributions from extinct and extant radionuclides, therefore offering the potential to set constraints on both the nature of Earth's planetary precursor(s) and the timing of their contributions. However, measuring the Xe isotope composition of mantle-derived samples to sufficiently high-precision has proven difficult due to (i) large occurrence of a modern-like atmospheric component in the mantle, and (ii) contribution from shallow and post-eruptive atmospheric contamination. Mantle-derived samples therefore exhibit only small deviations from the modern atmospheric composition, making the identification and deconvolution of mantle-derived Xe signals challenging. Here, we use the Giggenbach sampling method to concentrate magmatic noble gases from the Eifel volcanic area (Germany) into glass bottles in order to conduct high-precision analyses of Ne, Ar and Xe isotopes. The three samples collected from Victoriaquelle and Schwefelquelle wells (South East Eifel) show variable contributions from atmospheric contamination, with the least contaminated sample reaching Ar-40/Ar-36 similar to 8,300. Our data indicate that the mantle beneath the Eifel volcanic area, and by extension the Central European Volcanic Province, resembles the convective upper mantle reservoir with limited evidence for an OIB-like deep plume source contribution. It has a geochemical signature that is similar (e.g. in Ne isotopic composition, Ar-40/Ar-36, (129)xe/Xe-130 and Xe-129/Xe-136)) to the mantle source of the so-called popping rocks (thought to best represent the upper mantle), with an additional source of U-238-derived Xe and low He-3/He-4 that we attribute to the influence of an ancient subducted component (HIMU). A dichotomy exists between the main sources of fissiogenic xenon isotopes measured in popping rocks and Eifel gas, which appear to be mainly derived from (PU)-P-244 and U-238, respectively. According to their respective ratios of (PU)-P-244- to U-238-derived Xe, the mantle sources for Eifel volcanism and popping rocks would have experienced extensive and limited degassing, respectively. In this regard, high Pu-Xe/(Pu+U)-Xe may no longer be considered as being indicative of a mantle deep origin, therefore calling for the geochemical differences between plume and MORB sources to be redefined, with the possibility that volatile signatures within the solid Earth may be more heterogeneously distributed than previously thought. (C) 2019 The Author(s). Published by Elsevier B.V.
