¹Nicole X. Nie, ²Damanveer S. Grewal, ¹Zhe J. Zhang
Earth and Planetary Science Letters, 695, 120321 Link to Article [DOI: 10.1016/j.epsl.2026.120321]
¹Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA, 02139, USA
²Department of Earth and Planetary Sciences, Yale University, New Haven, CT, 06511, USA
Copyright Elsevier

High-temperature evaporation is a fundamental process shaping the chemical and isotopic signatures of planetary materials, yet the specific physical laws governing isotopic fractionation during evaporation remain debated. We investigated the mass-dependent isotope fractionation of Fe and Ni during vacuum evaporation of metallic melts. The residues exhibit large isotope fractionation, reaching up to 11.4 ‰ for δ^56/54Fe and 2.9 ‰ for δ^60/58Ni. Determining the isotope fractionation factors for Fe and Ni [$\alpha ={\left(m_{i1}/m_{i2}\right)}^{\beta }$] using the Rayleigh distillation model yielded a $\beta$ value of 0.5 for both Fe and Ni. Further, the evaporation coefficients of Fe and Ni were found to be similar (${\gamma }_{\text{Fe}}\simeq {\gamma }_{\text{Ni}}$). These results suggest that Fe and Ni evaporation from metallic melts is ideal and uninhibited by kinetic barriers.

We used high-precision three-isotope plots to constrain the mass fractionation law governing the evaporation process. Both Fe and Ni in the evaporation residue strictly follow the law of $n=-0.5$, distinct from the exponential law ($n=0$) typically assumed for kinetic processes. We propose a theoretical framework to reconcile this observation with kinetic gas theory: while the intrinsic instantaneous isotope fractionation between vapor and melt follows the exponential law, the apparent law observed in the residue is shifted by the Rayleigh distillation process. We derive a general relationship showing that the divergence between the apparent law and the intrinsic law during a Rayleigh distillation process is controlled by the magnitude of isotope fractionation ($\beta$), $n_{\text{appr}}\simeq \left[1+\frac{{\beta }_{i2/i1}}{2}\text{ln}\left(\frac{m_{i1}}{m_{i2}}\right)\right]n_{\text{inst}}-{\beta }_{i2/i1}$. This framework unifies conflicting observations regarding mass fractionation laws in previous evaporation experiments. The implications of these results for the evaporation history of Type-I cosmic spherules and for identifying nucleosynthetic isotopic anomalies are discussed.