No Arabic abstract
This paper summarises the theory and functionality behind Questaal, an open-source suite of codes for calculating the electronic structure and related properties of materials from first principles. The formalism of the linearised muffin-tin orbital (LMTO) method is revisited in detail and developed further by the introduction of short-ranged tight-binding basis functions for full-potential calculations. The LMTO method is presented in both Greens function and wave function formulations for bulk and layered systems. The suites full-potential LMTO code uses a sophisticated basis and augmentation method that allows an efficient and precise solution to the band problem at different levels of theory, most importantly density functional theory, LDA+U, quasi-particle self-consistent GW and combinations of these with dynamical mean field theory. This paper details the technical and theoretical bases of these methods, their implementation in Questaal, and provides an overview of the codes design and capabilities.
We present a formulation of the so-called Fermi sea contribution to the conductivity tensor of spin-polarized random alloys within the fully relativistic tight-binding linear muffin-tin-orbital (TB-LMTO) method and the coherent potential approximation (CPA). We show that the configuration averaging of this contribution leads to the CPA-vertex corrections that are solely due to the energy dependence of the average single-particle propagators. Moreover, we prove that this contribution is indispensable for the invariance of the anomalous Hall conductivities with respect to the particular LMTO representation used in numerical implementation. Ab initio calculations for cubic ferromagnetic 3d transition metals (Fe, Co, Ni) and their random binary alloys (Ni-Fe, Fe-Si) indicate that the Fermi sea term is small against the dominating Fermi surface term. However, for more complicated structures and systems, such as hexagonal cobalt and selected ordered and disordered Co-based Heusler alloys, the Fermi sea term plays a significant role in the quantitative theory of the anomalous Hall effect.
The most popular electronic structure method, the linear muffin-tin orbital method (LMTO), in its full-potential (FP) and relativistic forms has been extended to calculate the spectroscopic properties of materials form first principles, i.e, optical spectra, x-ray magnetic circular dichroism (XMCD) and magneto-optical kerr effect (MOKE). The paper describes an overview of the FP-LMTO basis set and the calculation of the momentum matrix elements. Some applications concerning the computation of optical properties of semiconductors and XMCD spectra of transition metal alloys are reviewed.
We discuss an efficiency of various band structure algorithms in determining the Fermi surface (FS) of the paramagnetic ErGa3. The linear muffin-tin orbital (LMTO) in the atomic sphere approximation (ASA) method and three full potential (FP) codes: FP-LMTO, FP linear augmented plane wave (FLAPW), and FP local orbitals (FPLO) methods are employed. Results are compared with electron-positron (e-p) momentum densities reconstructed from two dimensional angular correlation of annihilation radiation (2D ACAR). Unexpectedly, none of used modern FP codes is able to give satisfying description of the experimental data that are in perfect agreement with LMTO-ASA results. We suspect that it can be connected with a different choice of the linearization energy.
We present a new full-potential method to solve the one-body problem, for example, in the local density approximation. The method uses the augmented plane waves (APWs) and the generalized muffin-tin orbitals (MTOs) together as basis sets to represent the eigenfunctions. Since the MTOs can efficiently describe localized orbitals, e.g, transition metal 3$d$ orbitals, the total energy convergence with basis size is drastically improved in comparison with the linearized APW method. Required parameters to specify MTOs are given by atomic calculations in advance. Thus the robustness, reliability, easy-of-use, and efficiency at this method can be superior to the linearized APW and MTO methods. We show how it works in typical examples, Cu, Fe, Li, SrTiO$_3$, and GaAs.
We present the formalism and demonstrate the use of the overlapping muffin-tin approximation (OMTA). This fits a full potential to a superposition of spherically symmetric short-ranged potential wells plus a constant. For one-electron potentials of this form, the standard multiple-scattering methods can solve Schr{o}dingers equation correctly to 1st order in the potential overlap. Choosing an augmented-plane-wave method as the source of the full potential, we illustrate the procedure for diamond-structured Si. First, we compare the potential in the Si-centered OMTA with the full potential, and then compare the corresponding OMTA $N$-th order muffin-tin orbital and full-potential LAPW band structures. We find that the two latter agree qualitatively for a wide range of overlaps and that the valence bands have an rms deviation of 20 meV/electron for 30% radial overlap. Smaller overlaps give worse potentials and larger overlaps give larger 2nd-order errors of the multiple-scattering method. To further remove the mean error of the bands for small overlaps is simple.