Correction to “Unraveling the Growth Dynamics of Rutile Sn_1–xGe_xO₂ Using Theory and Experiment”

In our original paper, we identified an error in the normalization used when applying the regular-solution interaction parameter (Ω) to construct the incoherent phase diagram in Figure 3. The Ω value was fit to DFT-calculated mixing energies normalized per atom. However, in the expression for the Gibbs free energy, configurational entropy was defined per cation mixing site. This introduced a factor-ofthree mismatch (from three atoms per rutile formula unit) between the enthalpic and entropic terms, leading to an underestimated critical temperature and an artificially suppressed miscibility gap. Re-evaluation using the consistently normalized Ω (per cation site) now yields a larger incoherent miscibility gap: the predictedcritical temperature increases from 710 to 2700 °C, and the metastable solubility limit (spinodal) at 600 °C decreases from 34% to 9% Ge. Figure 3a has been updated to reflect these corrected results. Our experimentally observed Ge concentrations substantially exceed this revised incoherent metastable limit. However, recent first-principles calculations by Müller et al. show that epitaxial coherency strain can suppress phase separation andextend the solubility window,3providing an explanation for this apparent discrepancy in our corrected work. Figure 3c and d has therefore been updated to employ interaction parameters that account for coherencystrain effects in order to more accurately represent the experimental conditions. These corrections do not alter the central experimental observations or the qualitative conclusions of the original paper, namely, that thin-film Sn1−xGexO2 solubilities exceed predictions from incoherent bulk thermodynamics, that maximizing Ge content requires high Ge fluxes and low Sn fluxes, and amorphization combined with GeO volatility prevents the formation of crystalline films at high Ge concentrations. A revised version of Figure 3 is provided here. We thank Emmanouil Kioupakis and Anirudh Raju Natarajan for bringing this discrepancy to our attention and for the careful analysis performed by Müller et al.,3 which enabled the correction of our phase diagram calculations.