Iron nitride (Fe3N) and iron carbide (Fe3C) nanoparticles can be prepared via sol–gel synthesis. While sol–gel methods are simple, it can be difficult to control the crystalline composition, i.e., to achieve a Rietveld-pure product. In a previous in situ synchrotron study of the sol–gel synthesis of Fe3N/Fe3C, we showed that the reaction proceeds as follows: Fe3O4 → FeOx → Fe3N → Fe3C. There was considerable overlap between the different phases, but we were unable to ascertain whether this was due to the experimental setup (side-on heating of a quartz capillary which could lead to thermal gradients) or whether individual particle reactions proceed at different rates. In this paper, we use in situ wide- and small-angle X-ray scattering (wide-angle X-ray scattering (WAXS) and small-angle X-ray scattering (SAXS)) to demonstrate that the overlapping phases are indeed due to variable reaction rates. While the initial oxide nanoparticles have a small range of diameters, the size range expands considerably and very rapidly during the oxide–nitride transition. This has implications for the isolation of Rietveld-pure Fe3N, and in an extensive laboratory study, we were indeed unable to isolate phase-pure Fe3N. However, we made the surprising discovery that Rietveld-pure Fe3C nanoparticles can be produced at 500 °C with a sufficient furnace dwell time. This is considerably lower than the previous reports of the sol–gel synthesis of Fe3C nanoparticles.