Temperatures within the Earth's Core

Abstract

IN a recent communication, Strong¹ gave the results of an experimental determination of the fusion curve of iron up to a pressure of 96,000 atmospheres. On the usual assumption that the Earth's core is composed of iron, he extrapolated his results to obtain a value of 2,610 ± 200° K. for the melting point in the core at the core–mantle boundary. His experimental results fitted Simon's² equation: with the constants a = 75,000 atmospheres and c = 8. In this equation p is the pressure in atmospheres, T ₀ the melting temperature in degrees Kelvin at one atmosphere (that is, T ₀ = 1,805° K.), and T _m the melting temperature at the pressure p. Estimates of the melting point in the Earth's core obtained by Strong are appreciably lower than those obtained using Simon's values for the constants a and c, and only about one-half of those obtained using Gilvarry's³ more recent values. Moreover, the gradient of the melting point curve is much lower than in these previous estimates. The value of 2,610° K. for the melting point at the core–mantle boundary is particularly interesting in view of Verhoogen's⁴ conclusion that the temperature there is not likely to exceed 2,700° K. and his previous estimate⁵ of 2,500° C. as the maximum, which was based on the hypothesis that the maximum observed surface temperature of lavas reflects the maximum temperature in the mantle. Since seismic evidence indicates that the outer core in any event is liquid, these two results of Strong and Verhoogen taken together fix the temperature T ₁ at the core–mantle boundary rather precisely, namely: so that a value of 2,650° K. for T₁ seems reasonable.