No universal devolatilization trend has been found for the solar system rocky bodies

Rocky planets formed at different heliocentric distances from the Sun are thought to experience different and systematic devolatilization (i.e. depletion of volatile elements) with respect to the solar composition. The canonical empirical model of devolatilization is calibrated based on the bulk compositional difference between the Sun and Earth as a function of 50% condensation temperature of elements (T_C). The quantification of the volatility trends for other solar system rocky bodies, with a goal of formulating a general devolatilization model, has been expected to be useful for estimating the bulk compositions of rocky exoplanets orbiting at different distances to the central star in a planetary system.

To do so, we compiled an enormous set of literature data for the bulk compositions of a wide range of rocky bodies in our Solar System, including the terrestrial planets, asteroids, and chondritic bodies. Following the previous studies on quantifying Earth’s volatility trends, we adopted two relationship forms: one is in the log-log space (log (f) = α log (T_C) + β, where f is the bulk compositional ratio of a rocky body relative to the Sun; α and β are the model coefficients); and the other is in the linear space (f = 1/(1+e^-k(T-T0)), where T is the mid-plane temperature assuming the in-situ formation of these planetary bodies; and k and T0 are the model coefficients). Based on the best literature data that have been compiled, we did not find any statistically robust trend of devolatilization for these rocky bodies, except for Earth and Mars.  If we arbitrarily increase the uncertainties of the coefficients of the poorly quantified volatility trends for Venus and Mercury by a factor of 3, the best possible general trend that we can quantify as a function of heliocentric distance (d) is α = (3.773 ± 0.202)/d^3/4 and β = (-11.832 ± 0.613)/d^3/4. However, Mercury is still statistically deviated (towards a larger slope and thus severer devolatilization) from the general trend. We find the similar behaviour if we adopt the alternative sigmoid function. This may imply a more violent thermal history that Mercury experienced during its formation, beyond what can be constrained with the assumptions of the in-situ formation and stellar-irradiation-relevant-only devolatilization. Furthermore, Vesta and ordinary and carbonaceous chondrites do not follow this nominal general trend, either.

We therefore report here that no universal devolatilization trend has been found (empirically) for the solar system rocky bodies. This null result warrants the future efforts in advancing this field on two fundamental aspects. One is to further improve the measurements of the compositions of the solar system objects through various missions. The other is to launch a comprehensive investigation of nebular condensation, disc evolution, hydrodynamic escape, accretionary dynamics, and impacts towards establishing a sophisticated planet formation model of both dynamics and chemistry.