Comment on “Correlation and relativistic effects in U metal and U-Zr alloy: Validation of ab initio approaches”

In a recent paper, Xie et al. [Phys. Rev. B 88, 235128 (2013)], report that the density-functional theory (DFT) with the so-called DFT plus Hubbard U (DFT+U) modification improves energetics, volumes, and formation enthalpies over the standard form of DFT for uranium metal and U-Zr alloys. Also, spin-orbit coupling (SOC) was argued to advance the aforementioned properties in these systems. We demonstrate, contrarily, that neither the Hubbard U approach nor SOC is necessary for a correct description of uranium metal and U-Zr alloys. We further illustrate that the combination of DFT+U and SOC in the projector augmented-wave calculations by Xie et al. results in unrealistically large volume expansions, particularly for γ-U, in stark contrast to all previous calculations for elemental uranium. This in turn may also explain why the DFT+U with SOC model predicts negative enthalpy of mixing in the U-Zr alloy system contradicting conventional DFT as well as one of the main features of the experimental U-Zr phase diagram. The assertion by Xie et al. that DFT+U is an improvement over DFT for these systems is illustrated to be incorrect.