Choice Of Exchange-correlation Research Articles

High temperature non-harmonic evaluation and bonding in CaO

Previous theoretical and experimental investigations on high temperature anharmonic quantities such as thermal expansion, thermal Grüneisen parameter, phonon softening, and various caloric properties for calcium oxide show a large scatter. They were limited to temperatures well below the melting point. However, these rigorous studies are illustrative and suggest that the amount of anharmonicity in an interaction potential depends on the choice of exchange-correlation (XC) functionals and pseudopotentials. One thus requires to understand a counter mechanism to compensate for the extravagant quasiharmonic thermal properties. In the present study, we employ PAW (Projector Augmented Wave) pseudopotentials within the density functional theory (DFT) using two XC-functionals: the local density approximation (LDA) and generalized gradient approximation (GGA). We test their role and accuracy for evaluating the non-harmonic properties of CaO at finite temperatures. We use the lowest-order thermodynamic perturbation theory by treating CaO as a weak-anharmonic oscillator for supplementing the anharmonic phonon contribution to DFT-based quasiharmonic lattice dynamical properties up to 3000 K. A systematic study of anharmonic properties at high-T reveals the importance of anharmonic interaction, where the effect of LDA and GGA on the non-quadratic part of the interaction potential is also determined. T-dependent contribution to various physical properties shows the presence of strong anharmonicity in the interaction potential for the PAW+GGA scheme. It offers a much stiffer effective spring constant and contributes large vibrational energy. To stabilize such a system at finite temperatures requires a significant positive anharmonic contribution. Instead, the PAW pseudopotential at the LDA level gives reasonable thermodynamic properties even within the quasiharmonic approximation (QHA); the role of anharmonicity is modest at the highest temperature. As a major outcome of the study, we could identify the contribution from the interaction potential referenced to the used XC-functionals in determining the degree and nature of anharmonicity operative at finite temperatures. The corollary is assessed and confirmed by accurately evaluating several thermophysical properties up to the melting temperature, including the anharmonicity. Furthermore, we investigate the bonding scenario in expanded CaO by examining the electron band structure. We find that the interaction between the ions is ionic, but the cations strongly influence the interactions. As a result, the charges are more locally deformed at the metal-ions site, and the infiltration of charge in the ionic sphere of Ca+2 from O−2 is more. This fact is manifested as a profound sensitivity of the Ca 3d-band, particularly at the Γ-point. For expansion, the electron-cloud deformation reduces, and one expects more isotropic electron distribution. Thus, the PAW+LDA-based QHA gives a reasonable description for high-T assessment for CaO, but an anharmonic contribution is necessary for an accurate evaluation of thermal properties.

Uncertainty quantification in first-principles predictions of phonon properties and lattice thermal conductivity

We present a framework for quantifying the uncertainty that results from the choice of exchange-correlation (XC) functional in predictions of phonon properties and thermal conductivity that use density functional theory (DFT) to calculate the atomic force constants. The energy ensemble capabilities of the BEEF-vdW XC functional are first applied to determine an ensemble of interatomic force constants, which are then used as inputs to lattice dynamics calculations and a solution of the Boltzmann transport equation. The framework is applied to isotopically-pure silicon. We find that the uncertainty estimates bound property predictions (e.g., phonon dispersions, specific heat, thermal conductivity) from other XC functionals and experiments. We distinguish between properties that are correlated with the predicted thermal conductivity [e.g., the transverse acoustic branch sound speed ($R^2=0.89$) and average Gr\"uneisen parameter ($R^2=0.85$)] and those that are not [e.g., longitudinal acoustic branch sound speed ($R^2=0.23$) and specific heat ($R^2=0.00$)]. We find that differences in ensemble predictions of thermal conductivity are correlated with the behavior of phonons with mean free paths between $100$ and $300$ nm. The framework systematically accounts for XC uncertainty in phonon calculations and should be used whenever it is suspected that the choice of XC functional is influencing physical interpretations.

Ab initio investigation into the physisorption of noble gases on graphene

The process of physisorption on graphene is of recent interest due to its vast applications, such as an atomic sensor, in electronics and energy storage. Density functional theory (DFT) was used to determine the adsorption energies and distances of X-graphene (X = He, Ne, Ar, Kr, Xe, Rn) systems. The accuracy of DFT relies heavily on the choice of exchange-correlation (xc) functional. Many xc functionals do not fully capture van der Waals (vdW) interactions, if at all, and a correction is needed. The Local Density Approximation (LDA) and various forms of the Perdew-Burke-Ernzerhof (PBE) Generalized Gradient Approximation (GGA) functional, including RPBE and revPBE, were used as well as the SCAN meta-GGA functional. In addition to these functionals, the vdW correction methods of Grimme (D2 and D3) and the revised Vydrov and van Voorhis (rVV10) vdW functional were used to account for dispersion effects. To investigate their accuracy, results were compared to higher level theory computational and experimental results. The revPBE-D3 combination gave the most accurate results for binding energy and distance when compared to computational studies with a mean absolute error (MAE) of in the range of 4.01-12.06 meV and 0.04-0.06 Å, respectively. When compared to experiment, however, SCAN-D3 performed best with a binding distance MAE of 0.04-0.10 Å. These results highlight the strengths and weaknesses of various computational approaches and will help direct future studies.