2024ApJ...965..172C

We present a high-resolution spectral study of Fe L-shell extinction by the diffuse interstellar medium (ISM) in the direction of the X-ray binaries Cygnus X-1 and GX 339-4, using the XMM-Newton reflection grating spectrometer. The majority of interstellar Fe is suspected to condense into dust grains in the diffuse ISM, but the compounds formed from this process are unknown. Here, we use the laboratory cross sections from Kortright & Kim () and Lee et al. () to model the absorption and scattering profiles of metallic Fe, and the crystalline compounds fayalite (Fe₂SiO₄), ferrous sulfate (FeSO₄), hematite (α-Fe₂O₃), and lepidocrocite (γ-FeOOH), which have oxidation states ranging from Fe⁰ to Fe³⁺. We find that the observed Fe L-shell features are systematically offset in energy from the laboratory measurements. An examination of over two dozen published measurements of Fe L-shell absorption finds a 1-2 eV scatter in energy positions of the L-shell features. Motivated by this, we fit for the best energy-scale shift simultaneously with the fine structure of the Fe L-shell extinction cross sections. Hematite and lepidocrocite provide the best fits (≈ + 1.1 eV shift), followed by fayalite (≈ + 1.8 eV shift). However, fayalite is disfavored, based on the implied abundances and knowledge of ISM silicates gained by infrared astronomical observations and meteoritic studies. We conclude that iron oxides in the Fe³⁺ oxidation state are good candidates for Fe-bearing dust. To verify this, new absolute photoabsorption measurements are needed on an energy scale accurate to better than 0.2 eV.