V. Rama Murthy, Department of Geology and Geophysics, University of Minnesota, Minneapolis, MN 55455

Abstract: The four most important radioactive nuclides responsible for heat production in the Earth are 40K, 235U, 238U and 232Th. Conventional wisdom relegates the bulk inventory of these elements exclusively to the silicate Crust and the Mantle (Bulk Silicate Earth -BSE), leaving the metallic Core devoid of any radioactivity. This point of view stems from the known geochemical behavior in the upper parts of the Earth where these elements tend to partition preferentially and almost exclusively into silicates relative to Fe-metal. However, several recent experiments clearly demonstrate that partition coefficients are a function of pressure and temperature. Thus it is possible that the geochemical behavior of these radioactive elements may be quite different in the deep Earth below a few hundred kilometers due to the ambient conditions of high pressure and temperature. Of the elements listed above, K is the only element for which some experimental data and theoretical calculations are available to suggest its presence in the metallic Core.

The suggestion that K can be sequestered into a sulfur-bearing core is more than 30 years old (Lewis, 1971; Hall and Murthy, 1971). Subsequent theoretical quantum mechanics calculations (Bukowinski, 1976) suggested that K undergoes an electronic structure change at high pressures and behaves more like a transition metal, faciltating alloy formation with Fe. These suggestions remained controversial because of strong adherence to convention as well as the many ambiguous and contradictory experimental results until recently. This picture has now changed due to new experimental work and theoretical calculations. Murthy et al. (2003) showed unambiguously that K can enter Fe-FeS melts in a strongly temperature dependent fashion and attributed previous ambiguous results to experimental artifacts. Using ultramafic silicate compositions and Fe-FeS mixtures, Gessmann and Wood (2002) have argued for entry of Na and K into sulfide melts. The magnitude of the partition coefficient is such that significant K can enter and act as a heat source in the core. Lee and Jeanloz (2003) have observed solubility of K in Fe-metal at pressures >26 GPa. More recently, Hirao et al. (2005, personal communication) report on partition of K in to Fe-metal at 134 GPa and 3500 K, conditions directly relevant to the core-mantle boundary. These experimental data have been bolstered by new molecular dynamics calculations (Parker et al., 1997, Lee et al., 2003)

These new results certainly bear upon the question of radioactivity of the Core but in themselves cannot precisely define how much radioactivity or how much K is in the Core. This is still an open question. The present experimental data suggest a range of about a few hundred parts per million, but the estimates are dependent on assumed initial composition of the Earth, models of core formation and partition data extrapolated to high pressure and temperature. Our knowledge of the radioactive contents of the Mantle and the Core are still largely model-dependent. Viewed in this context, the possibility of using geoneutrino flux to assess the radioactivity of the Mantle and Core is exciting, and likely to provide vital information that touches on many important aspects of the Earth - early differentiation, Core formation, thermal evolution, global dynamics and geomagnetism, to name a few.