Banded iron formations (BIFs) contain appreciable amounts of ferric iron (Fe3+). The mechanism by which ferrous iron (Fe2+) was oxidized into Fe3+ in an atmosphere that was globally anoxic is highly debated. Of the three scenarios that have been proposed to explain BIF formation, photo-oxidation by UV photons is the only one that does not involve life (the other two are oxidation by O2 produced by photosynthesis, and anoxygenic photosynthesis whereby Fe2+ is directly used as electron donor in place of water). We experimentally investigated iron and oxygen isotope fractionation imparted by iron photo-oxidation at a pH of 7.3. The iron isotope fractionation between precipitated Fe3+-bearing lepidocrocite and dissolved Fe2+ follows a Rayleigh distillation with an instantaneous 56Fe/54Fe fractionation factor of +1.2‰. Such enrichment in the heavy isotopes of iron is consistent with the values measured in BIFs. We also investigated the nature of the mass-fractionation law that governs iron isotope fractionation in the photo-oxidation experiments (i.e., the slope of the δ56Fe–δ57Fe relationship). The experimental run products follow a mass-dependent law corresponding to the high-T equilibrium limit. The fact that a ∼3.8 Gyr old BIF sample (IF-G) from Isua (Greenland) falls on the same fractionation line confirms that iron photo-oxidation in the surface layers of the oceans was a viable pathway to BIF formation in the Archean, when the atmosphere was largely transparent to UV photons.Our experiments allow us to estimate the quantum yield of the photo-oxidation process (∼0.07 iron atom oxidized per photon absorbed). This yield is used to model iron oxidation on early Mars. As the photo-oxidation proceeds, the aqueous medium becomes more acidic, which slows down the reaction by changing the speciation of iron to species that are less efficient at absorbing UV-photons. Iron photo-oxidation in centimeter to meter-deep water ponds would take months to years to complete. Oxidation by O2 in acidic conditions would be slower. Iron photo-oxidation is thus likely responsible for the formation of jarosite–hematite deposits on Mars, provided that shallow standing water bodies could persist for extended periods of time.The oxygen isotopic composition of lepidocrocite precipitated from the photo-oxidation experiment was measured and it is related to the composition of water by mass-dependent fractionation. The precipitate-fluid 18O/16O isotope fractionation of ∼+6‰ is consistent with previous determinations of oxygen equilibrium fraction factors between iron oxyhydroxides and water.