Compressible Mercury – Insights into Its Composition and Interior Structure

Introduction: Mercury is unique among the terrestrial planets for its low mass (3.302x10 kg) and high average density (5.427 g/cc) that together imply an iron-rich composition. Meaningful interplanetary comparisons of bulk composition, through bulk density, require the removal of self-compression effects to determine decompressed density (zero pressure and 300K, STP). The methodologies used in past estimates of planetary decompressed densities (Table 1) are not well documented in the scientific literature and rarely include uncertainty estimates. We present a detailed calculation of Mercury’s decompressed density including elucidation of assumptions, methodology, and an estimate of uncertainty. Background: The relatively high average density and low mass of Mercury indicates an unusual bulk composition and thus provides an important constraint for the initial temperature of the solar nebula, the degree of radial mixing, and the extent of condensation and evaporation [e.g. 4]. The high iron content of Mercury could be the result of chemical and thermal gradients in the solar nebula or partial removal of the silicate portion of a differentiated planet by giant impact or vaporization. These hypotheses lead to different predications, by numerous authors, of the bulk chemistry of Mercury, particularly the abundance of volatile elements. Little is directly known of Mercury’s composition and internal structure, however its high average density suggests a high metal to silicate ratio. Remote sensing suggests low FeO in the crust [5-8] and mantle [9]. The presence of an intrinsic magnetic field, possibly generated by a hydromagnetic dynamo, has led many researchers to postulate that Mercury has a molten outer core, thus demanding an alloying element, possibly sulfur, to lower the melting temperature [e.g. 10-11]. Sodium and potassium are present in the exosphere of Mercury, but it is not clear if their source is endogenic or exogenic [12]. Volatiles in the exosphere together with the intrinsic magnetic field demand consideration of a range of volatile abundances for Mercury. Simple models of Mercury's interior have been presented based on the total mass, total radius and cosmochemical arguments of plausible planetary compositions [13-15]. But none of these studies have presented calculations of the decompressed density. Methods: We model Mercury’s interior under adiabatic self-compression using the AdamsWilliamson equation with the second order BirchMurnaghan finite strain equation of state (EOS) to estimate its decompressed density. We assume the thermal profile is adiabatic except for a thermal boundary layer at the core mantle boundary, modeled as a temperature difference between the adiabats for the core and the mantle extrapolated to zero pressure. Our model is constrained by the total mass and radius (2440km) of Mercury along with cosmochemical constraints on densities and physical properties of likely core and mantle materials. We ignore the uncertainties in the total mass and total radius because they are much less significant than the uncertainties in the core and mantle densities and interior structure. The moment of inertia for Mercury is not accurately known and thus is not used as a constraint. We obtained a suite of results by randomly varying six input parameters within the ranges shown in Table 2 and adjusting the core density, within the allowable range, to match the observed mass of Mercury. We assume constant thermal expansion coefficient (αk = 2.5x10 K) and second Grüneisen parameter (δSk= 4) for the core.