Resolving Multiple Fluid-Rock Interactions in the Mantle: Trapped Noble Gases and Their Hosts in Peridotitic Rocks From Zabargad Island, Red Sea

M. Trieloff Max-Planck-Institut für Kernphysik, POB 103980, D-69029 Heidelberg, Germany.

E. K. Jessberger Max-Planck-Institut für Kernphysik, POB 103980, D-69029 Heidelberg, Germany.

H. W. Weber Max-Planck-Institut für Chemie, Saarstr. 23, D-55122 Mainz, Germany

G. Kurat Mineralogisch-Petrographische Abt., Naturhistorisches Museum, POB 417, A-1014 Wien, Austria

Introduction: Noble gases and peridotitic rocks

Peridotitic rocks crystallized in the Earth´s mantle and retain isotopic memory of the noble gases in this environment. Information about the noble gas inventory of - different - mantle reservoirs plays a key role in understanding the Earth's evolution, especially of the system solid Earth-atmosphere. Minerals of different retentivity can retain different isotopic memories during different stages of the uplift history and associated interactions.

High resolution 40Ar-39Ar stepheating

High resolution 40Ar-39Ar stepheating is an effective technique to resolve different components. The specific degassing behaviour of minerals enables to separate and identify them via neutron induced Ar isotopes derived from Ca, K, and Cl and to correlate minerals with associated trapped (and radiogenic) Ar components. Two clinopyroxenites and a hornblendite vein rock from peridotites (Kurat et al., 1993) from Zabargad Island, Red Sea were analysed. Within clinopyroxenites trapped Ar is present in 1) low temperature, low 40Ar/36Ar phases (carbonates?);
2) pyroxene related Cl-rich components (probably fluid inclusions), and 3) in associated amphiboles, which are intimately intergrown with pyroxene and nonseparable. The amphiboles, of which there are different generations, formed by interaction with mantle fluids during different stages of diapiric uplift (Agrinier et al., 1993). 40Ar/36Ar ratios of the latter two phases are very high (up to 7000) which points to the upper mantle as source. Nevertheless, there must have been an admixture of Ar of atmospheric composition within the source region of the peridotites. Within all phases 40Ar/36Ar ratios are variable. There is also evidence for different 40Ar, Cl and K contents of the parent fluids implying multiple fluid-rock interactions during uplift with important implications for transport processes of noble gases via fluids in the upper mantle. For the hornblendite the in situ radiogenic and excess Ar components could be separated: the plateau age of 18.7 ± 1.3 Ma dates the crustal intrusion in perfect agreement with a Pb/Pb age of 18.4 ± 1.0 Ma (Oberli et al., 1987). Obviously the formation of the hornblendite occurred during the final stage of diapiric uplift, most probably in interaction with seawater, as suggested by Sr, O, and H isotopic data (Agrinier et al., 1993) and the low trapped 40Ar/36Ar ratio of 305.

Stepwise crushing

4He, 20Ne, 40Ar and 36Ar were measured in an unirradiated orthopyroxenite vein rock by stepwise crushing and subsequent stepheating. Elemental concentrations show a clear negative correlation with increasing crushing step. Isotopic ratios are correlated as well: in the advanced crushing steps both radiogenic isotopes (4He, 40Ar) are progressively enriched (up to 50% with respect to 20Ne, and up to 100% with respect to 36Ar). The 20Ne/36Ar ratio increases by 25%, the 4He/40Ar ratio only by 6%, but still fitting a well behaved correlation. The surprising constancy of the 4He/40Ar ratio (~0.15) compared to the spread of other isotopic and elemental ratios, can only be explained by assigning the radiogenic isotopes as being indigenous to the mantle source and afterwards being contaminated by a source containing atmospheric type noble gases, rich in 20Ne and 36Ar and low in 4He. The clear correlation of isotope concentrations and isotopic ratios with increasing crushing steps can only be
explained by the presence of two different types or generations of inclusions most probably differing in size. Both contain the mantle component, but are contaminated to a different degree with the atmospheric component (which constitutes nearly all 36Ar, up to 16% of 40Ar for larger inclusions, 8% for smaller inclusions). If contamination of the mantle fluid and trapping occurred in the peridotite source region, large inclusions were trapped later. However, the mantle fluid with a 4He/40Ar ratio of 0.15, which is much lower than expected for an unfractionated mantle value resulting from long term decay of radioactive parent nuclides, was already indigenous to the peridotite source before, but it could have been modified earlier during peridotite uplift or before.


This investigation points to multiple trapped noble gas components due to multiple fluid-rock interactions. As the total fusion date yields a mean value with doubtful significance, we stress the necessity to resolve the different components by appropriate experimental techniques like stepheating and stepcrushing of irradiated and nonirradiated samples.


Agrinier, P., Mével, C., Bosch, D. & Javoy, M., Earth Planet. Sci. Lett. 120, 187-205 (1993).

Kurat, G., Palme, H., Embey-Isztin, A., Touret, J., Ntaflos, T., Spettel, B., Brandstätter, F., Palme, C., Dreibus, G. & Prinz, M., Min. Petr. 48, 309-341 (1993).

Oberli, R., Ntaflos, T., Meier, M. & Kurat, G., Terra cognita 7, 334 (1987).