11:30 AM - 11:45 AM
[SIT22-22] Determination of the noble gas partition coefficients between metal-silicate melts by high-pressure, high-temperature experiments and laser-microprobe noble gas analysis
Keywords:noble gas, primordial mantle reservoir, core formation, high-pressure and high-temperature experiments, partition coefficient
Analyses of ocean island basalts (OIBs) reveal a geochemical reservoir characterized by unradiogenic, “primordial” noble gas signatures (e.g., high 3He/4He and low 40Ar/36Ar ratios), likely residing in the deep mantle. There has been much debate about the area holding the “primordial” noble gases deep in the Earth [1], including that the “primordial” noble gasses have been retained in the deepest region of the mantle since 4.4 Ga [2] or in the core since the core-mantle separation [3]. However, the validity of latter strongly depends on the quantity of noble gases the core incorporates during accretion and can hold in the present day. A few experiments showed that the core could hold primordial noble gases though there are several uncertainties possibly due to extreme noble gas concentrations [4.5].
In this study, in order to investigate noble gas partitioning behavior between the core and mantle, noble gases were dissolved into metal-silicate melts under high temperature and pressure conditions, and then the samples were quenched. Three series of sample synthesis were performed at different pressure-temperature ranges and experimental approaches. At the Geophysical Laboratory, Carnegie Institute of Washington, argon partitioning experiments at 2000 K and 1 GPa were conducted using a piston cylinder apparatus. Experimental samples were contained by a double capsule: Pt outer capsule and graphite inner capsule. A Fe metal-silicate mixture was packed into the graphite capsule. Argon was added to the Pt outer capsule as a liquid, and the Pt capsule was welded shut while held in a bath of liquid N2.
The other two series of samples were synthesized at the Geodynamics Research Center, Ehime University. Noble-gas doped silicate glass and iron were melted and equilibrated under high pressure and temperature using a multianvil apparatus (2200~2300 K, 8 GPa) and a laser-heated diamond anvil cell (2000~4000 K, 10~70 GPa).
After the sample synthesis, the noble gas concentrations contained in the each phase were analyzed using an ultraviolet laser ablation apparatus and a noble gas mass spectrometer at the University of Tokyo.
The partition coefficient D, where D = (noble gas in metal phase)/(noble gas in silicate phase), of argon at 1 GPa varied widely from orders of 10-4 to 10-1. This resulted from heterogeneous argon distribution in the metal phase, which seems significantly controlled by contaminant phases, such as silicate inclusions and micro- or nano-argon bubbles. Therefore the lowest D determined so far would yield the best estimate, 7 × 10-5, which is three orders of magnitude lower than the previous work [4].
On the other hand, D for neon, argon, krypton and xenon at 8 GPa obtained with the samples synthesized with multianvil apparatus were in the order of 10-3, which is consistent with the previous work [4]. At the present time, we have not determined helium partition coefficient in the all pressure range as it was difficult to retain enough amount of helium in high-pressure and temperature apparatus during the experiments. We confirmed that the silicate phase obtained with laser-heated diamond anvil cell contained a sufficient amount of noble gas. However, we have not determined amount of the noble gas contained in the metal phase because it was too small to be selectively analyzed with the laser ablation system for noble gas extraction. Further experiments are necessary to obtain the noble gas concentration contained in the metal phase in order to determine the partition coefficients at higher pressure, at least 30 GPa at which the elemental partition between iron and silicate melt would have occurred during the core formation [6].
[1] Porcelli & Ballentine, Rev. Mineral. Geochem. 47, 411-480, 2002. [2] Mukhopadhyay, Nature. 486, 101-104, 2012. [3] Trieloff & Kunz, Phys. Earth Planet. Inter. 148, 13-38, 2005. [4] Matsuda et al., Science. 259, 788-791, 1993 [5] Bouhifd et al., Nature. Geo. 6, 982-986, 2013. [6] Righter, Earth Planet. Sci. Lett. 304, 158-167, 2011.
In this study, in order to investigate noble gas partitioning behavior between the core and mantle, noble gases were dissolved into metal-silicate melts under high temperature and pressure conditions, and then the samples were quenched. Three series of sample synthesis were performed at different pressure-temperature ranges and experimental approaches. At the Geophysical Laboratory, Carnegie Institute of Washington, argon partitioning experiments at 2000 K and 1 GPa were conducted using a piston cylinder apparatus. Experimental samples were contained by a double capsule: Pt outer capsule and graphite inner capsule. A Fe metal-silicate mixture was packed into the graphite capsule. Argon was added to the Pt outer capsule as a liquid, and the Pt capsule was welded shut while held in a bath of liquid N2.
The other two series of samples were synthesized at the Geodynamics Research Center, Ehime University. Noble-gas doped silicate glass and iron were melted and equilibrated under high pressure and temperature using a multianvil apparatus (2200~2300 K, 8 GPa) and a laser-heated diamond anvil cell (2000~4000 K, 10~70 GPa).
After the sample synthesis, the noble gas concentrations contained in the each phase were analyzed using an ultraviolet laser ablation apparatus and a noble gas mass spectrometer at the University of Tokyo.
The partition coefficient D, where D = (noble gas in metal phase)/(noble gas in silicate phase), of argon at 1 GPa varied widely from orders of 10-4 to 10-1. This resulted from heterogeneous argon distribution in the metal phase, which seems significantly controlled by contaminant phases, such as silicate inclusions and micro- or nano-argon bubbles. Therefore the lowest D determined so far would yield the best estimate, 7 × 10-5, which is three orders of magnitude lower than the previous work [4].
On the other hand, D for neon, argon, krypton and xenon at 8 GPa obtained with the samples synthesized with multianvil apparatus were in the order of 10-3, which is consistent with the previous work [4]. At the present time, we have not determined helium partition coefficient in the all pressure range as it was difficult to retain enough amount of helium in high-pressure and temperature apparatus during the experiments. We confirmed that the silicate phase obtained with laser-heated diamond anvil cell contained a sufficient amount of noble gas. However, we have not determined amount of the noble gas contained in the metal phase because it was too small to be selectively analyzed with the laser ablation system for noble gas extraction. Further experiments are necessary to obtain the noble gas concentration contained in the metal phase in order to determine the partition coefficients at higher pressure, at least 30 GPa at which the elemental partition between iron and silicate melt would have occurred during the core formation [6].
[1] Porcelli & Ballentine, Rev. Mineral. Geochem. 47, 411-480, 2002. [2] Mukhopadhyay, Nature. 486, 101-104, 2012. [3] Trieloff & Kunz, Phys. Earth Planet. Inter. 148, 13-38, 2005. [4] Matsuda et al., Science. 259, 788-791, 1993 [5] Bouhifd et al., Nature. Geo. 6, 982-986, 2013. [6] Righter, Earth Planet. Sci. Lett. 304, 158-167, 2011.