ABSTRACT

Iron exhibits variable redox states between Fe²⁺ and Fe³⁺, governing key processes from Earth’s deep interior to surface environments. Quantitative evaluation of Fe valence is therefore essential in geoscience; however, partial fluorescence yield XANES (PFY-XANES), a widely used nondestructive technique, often suffers from thickness or self-absorption effect in samples with high Fe concentrations. In this study, we applied the inverse PFY (IPFY) method, to geological samples to assess its reliability in Fe valence determination. Iron L_III-edge IPFY-XANES and conventional PFY-XANES analyses were performed on olivine, wadsleyite, bridgmanite, and two biotite samples with FeO contents of 6.6–37.7wt.%. Conventional PFY-XANES overestimated Fe³⁺/ΣFe by approximately 10% relative to Mössbauer spectroscopy, whereas IPFY-XANES yielded values differing by less than 3%, showing excellent agreement. The inverse spectra also exhibited reduced pre- and post-edge tailing, confirming suppression of thickness effects. These results demonstrate that IPFY-XANES provides more accurate determination of the valence state of Fe, particularly for Fe-rich or thick samples. Because many geoscientific specimens such as asteroidal return samples, meteorites, or recovered high-pressure samples cannot be destructively prepared, the IPFY approach offers a powerful and reliable method for nondestructive Fe valence analysis across a wide range of natural materials.