Publications

Cubic scaling $GW$: towards fast quasiparticle calculations

No abstract is provided for this article.

Polaronic hole-trapping in doped insulating oxide

No abstract is provided for this article.

Outside Back Cover: Spin Frustration Determines the Stability and Reactivity of Metal–Organic Frameworks with Triangular Iron(III)–Oxo Clusters (Angew. Chem. 41/2025)

Density functional theory studies on stress stabilization of the Cu(110) striped phase

Under oxygen exposure, the Cu(110) surface shows a striped phase consisting of alternating bare (110) surface areas and added-row (2×1)O reconstructions. Density functional theory is used to show that the major origin for the formation of the striped phase is the elastic interaction between these areas. The difference between the surface stress of the bare surface and the (2×1)O covered surface is predicted to be 1.3N/m in reasonable agreement with values derived from grazing incidence X-ray diffraction. Supercell calculations for periods of up to 62Å confirm that the formation of the striped phase is favorable compared to an added-row (4×1)O reconstruction with the same coverage. But the predicted equilibrium period of roughly 30Å is significantly smaller than in experiment. The calculations are impeded by the surface energy alternating with the number of layers in the slab. This behavior is related to a quantum well behavior of the Cu 4s-electrons. A simple model for this behavior is discussed and compared to ab initio results.

Accurate Quasiparticle Spectra from Self-Consistent<i>GW</i>Calculations with Vertex Corrections

Self-consistent $GW$ calculations, maintaining only the quasiparticle part of the Green's function $G$, are reported for a wide class of materials, including small gap semiconductors and large gap insulators. We show that the inclusion of the attractive electron-hole interaction via an effective nonlocal exchange correlation kernel is required to obtain accurate band gaps in the framework of self-consistent $GW$ calculations. If these are accounted for via vertex corrections in $W$, the band gaps are found to be within a few percent of the experimental values.

Oxygen Adsorption on Zr(0001): An<i>ab Initio</i>Study

The main goal of this paper is to study the oxygen adsorption on the Zr(0001) surface using first-principles total-energy calculations based on density-functional theory. We present preliminary results which concern the atomic oxygen adsorption on the Zr(0001) surface. We first report a static study where we calculate the atomic structure and adsorption energy for oxygen occupying various surface and subsurface sites at three different coverages: Θ = 1/4, 1/2, and 1 ML. We find that oxygen atoms are preferentially adsorbed into the octahedral holes between the 2nd and 3rd metallic layers. We secondly perform ab initio molecular dynamics calculations for the Zr(0001) – (3 × 3) – O system to show how the oxygen can penetrate through the surface and how it finally reaches its equilibrium position, trapped between the 1st and 2nd zirconium layers.

Norm-conserving and ultrasoft pseudopotentials for first-row and transition elements

The construction of accurate pseudopotentials with good convergence properties for the first-row and transition elements is discussed. We show that by combining an improved description of the pseudowavefunction inside the cut-off radius with the concept of ultrasoft pseudopotentials introduced by Vanderbilt optimal compromise between transferability and plane-wave convergence can be achieved. With the new pseudopotentials, basis sets with no more than 75-100 plane waves per atom are sufficient to reproduce the results obtained with the most accurate norm-conserving pseudopotentials.

Dynamic structure factor of liquid and amorphous Ge from<i>ab initio</i>simulations

We calculate the dynamic structure factor $S(k,\ensuremath{\omega})$ of liquid Ge $(l$-Ge) at temperature $T=1250\mathrm{}\mathrm{K},$ and of amorphous Ge $(a$-Ge) at $T=300\mathrm{K},$ using ab initio molecular dynamics. The electronic energy is computed using density-functional theory, primarily in the generalized gradient approximation, together with a plane-wave representation of the wave functions and ultrasoft pseudopotentials. We use a 64-atom cell with periodic boundary conditions, and calculate averages over runs of up to about 16 ps. The calculated liquid $S(k,\ensuremath{\omega})$ agrees qualitatively with that obtained by Hosokawa et al. [Phys. Rev. B 63, 134205 (2001)] using inelastic x-ray scattering. In a-Ge, we find that the calculated $S(k,\ensuremath{\omega})$ is in qualitative agreement with that obtained experimentally by Maley et al. [Phys. Rev. Lett. 56, 1720 (1986)]. Our results suggest that the ab initio approach is sufficient to allow approximate calculations of $S(k,\ensuremath{\omega})$ in both liquid and amorphous materials.

An ab initio based structure model of i(Al–Pd–Mn)

We illustrate how ab initio numerical simulation methods can be used to check and improve structure models for i(Al–Pd–Mn). By focusing our study on the optimization of a small approximant, we obtain a number of general structural and compositional rules simple enough to be applicable to the quasicrystalline structure itself.

First-order phase transitions by first-principles free-energy calculations: The melting of Al

The melting properties of aluminum are calculated from first-principles molecular-dynamics simulations using density-functional theory in the local-density approximation. We calculate a melting temperature of 890 K at zero pressure, to be compared to the experimental value of 933 K. An elaborate discussion of the techniques employed is presented. The solid- and liquid-state free energies are obtained via coupling constant integration. The respective reference systems are the quasiharmonic crystal and the Lennard-Jones fluid. Good quality of the Brillouin zone sampling is shown to be crucial. The strategy followed is expected to be applicable to a wide range of liquid metals.

Phaseless auxiliary field quantum Monte Carlo with projector-augmented wave method for solids

We implement the phaseless auxiliary field quantum Monte Carlo method using the plane-wave based projector augmented wave method and explore the accuracy and the feasibility of applying our implementation to solids. We use a singular value decomposition to compress the two-body Hamiltonian and thus reduce the computational cost. Consistent correlation energies from the primitive-cell sampling and the corresponding supercell calculations numerically verify our implementation. We calculate the equation of state for diamond and the correlation energies for a range of prototypical solid materials. A down-sampling technique along with natural orbitals accelerates the convergence with respect to the number of orbitals and crystal momentum points. We illustrate the competitiveness of our implementation in accuracy and computational cost for dense crystal momentum point meshes comparing to a well-established quantum-chemistry approach, the coupled-cluster ansatz including singles, doubles and perturbative triple particle-hole excitation operators.

Comparing machine learning potentials for water: Kernel-based regression and Behler-Parrinello neural networks

In this paper we investigate the performance of different machine learning potentials (MLPs) in predicting key thermodynamic properties of water using RPBE+D3. Specifically, we scrutinize kernel-based regression and high-dimensional neural networks trained on a highly accurate dataset consisting of about 1,500 structures, as well as a smaller data set, about half the size, obtained using only on-the-fly learning. The study reveals that despite minor differences between the MLPs, their agreement on observables such as the diffusion constant and pair-correlation functions is excellent, especially for the large training dataset. Variations in the predicted density isobars, albeit somewhat larger, are also acceptable, particularly given the errors inherent to approximate density functional theory. Overall, the study emphasizes the relevance of the database over the fitting method. Finally, the study underscores the limitations of root mean square errors and the need for comprehensive testing, advocating the use of multiple MLPs for enhanced certainty, particularly when simulating complex thermodynamic properties that may not be fully captured by simpler tests.

Screened exchange corrections to the random phase approximation from many-body perturbation theory

The random phase approximation (RPA) systematically overestimates the magnitude of the correlation energy and generally underestimates cohesive energies. This originates in part from the complete lack of exchange terms, which would otherwise cancel Pauli exclusion principle violating (EPV) contributions. The uncanceled EPV contributions also manifest themselves in form of an unphysical negative pair density of spin-parallel electrons close to electron-electron coalescence. We follow considerations of many-body perturbation theory to propose an exchange correction that corrects the largest set of EPV contributions while having the lowest possible computational complexity. The proposed method exchanges adjacent particle/hole pairs in the RPA diagrams, considerably improving the pair density of spin-parallel electrons close to coalescence in the uniform electron gas (UEG). The accuracy of the correlation energy is comparable to other variants of Second Order Screened Exchange (SOSEX) corrections although it is slightly more accurate for the spin-polarized UEG. Its computational complexity scales as $\mathcal O(N^5)$ or $\mathcal O(N^4)$ in orbital space or real space, respectively. Its memory requirement scales as $\mathcal O(N^2)$.

$GW$ vertex corrected calculations for molecular systems

Hedin's scheme is solved with the inclusion of the vertex function ($GW\Gamma$) for a set of small molecules. The computational scheme allows for the consistent inclusion of the vertex both at the polarizability level and in the self-energy. A diagrammatic analysis shows that the self-energy formed with this four-point vertex does not lead to double counting of diagrams, that can be classified as direct bubbles and exchange diagrams. By removing the exchange diagrams from the self-energy, a simpler approximation is obtained, called $GW^{\rm{tc-tc}}$. Very good agreement with expensive wavefunction-based methods is obtained for both approximations.

(2 × 1) reconstruction and hydrogen-induced de-reconstruction of the diamond (100) and (111) surfaces

We present a comparative study of the (2 × 1) reconstruction of the (100) and (111) diamond surfaces, and of their hydrogen-induced de-reconstruction, using ab initio local-density-functional calculations. We show that whereas for the (111) surface the absorption of a monolayer of hydrogen stabilizes the (1 × 1) bulk-terminated surface, for the (100) surface the (2 × 1) surface structure remains stable (albeit with a strongly enhanced bond-length of the surface dimers). Results for the electronic surface-states are presented.

Lattice constants and cohesive energies of alkali, alkaline-earth, and transition metals: Random phase approximation and density functional theory results

We present lattice constants and cohesive energies of alkali, alkaline earth, and transition metals using the correlation energy evaluated within the adiabatic-connection fluctuation-dissipation (ACFD) framework in the random phase approximation (RPA) and compare our findings to results obtained with the meta-GGA functional revTPSS and the gradient corrected PBE (Perdew-Burke-Ernzerhof) functional and the PBEsol functional (PBE reparametrized for solids), as well as a van der Waals (vdW) corrected functional optB88-vdW. Generally, the RPA reduces the mean absolute error in the lattice constants by about a factor 2 compared to the other functionals. Atomization energies are also on par with the PBE functional, and about a factor 2 better than with the other functionals. The study confirms that the RPA describes all bonding situations equally well including van der Waals, covalent, and metallic bonding.

Dissociation pathways of oxygen on copper (110) surface: a first principles study

We have carried out first principles calculations to examine the adsorption and dissociation of oxygen on the Cu (110) surface. At low coverage, our results indicate two absorption species: a peroxo- and a superoxo-like species which have direct implications to the dissociation process. In agreement with experimental studies, the most favourable absorption site for O2 is found to be the 4-fold hollow site. The dissociation barrier at this site is only 150 meV, which is consistent with experimental findings. Static calculations also reveal other possible dissociation pathways with significantly different energy barriers. First principles MD simulations are used to probe the dynamics of the dissociation process. We find only selected pathways are likely to be utilised because of interaction with the surface.

Structural characterization of the hydrogen-covered C(100) surface by density functional theory calculations

We present the results of a density functional theory (generalized gradient approximation) study of the hydrogenated C(100) surfaces. We have analyzed the formation energy of phases with different hydrogen coverages $(\ensuremath{\theta}=1.0,$ 1.2, 1.33, 1.5, and 2.0) as a function of the hydrogen chemical potential. As the hydrogen chemical potential increases, the stable phase changes from the bare surface through all the hydrides considered, in order of increasing coverage. The value of the hydrogen chemical potential beyond which dihydride units are stabilized on the surface nearly coincides with the potential at which the the formation energy of methane is zero. However, since the desorption of hydrocarbons is an activated process, dihydride units can appear in metastable surface phases. To investigate this possibility, we have calculated vibrational frequencies for the $(2\ifmmode\times\else\texttimes\fi{}1):\mathrm{H},$ $(5\ifmmode\times\else\texttimes\fi{}1):1.2\mathrm{H},$ and $(3\ifmmode\times\else\texttimes\fi{}1):1.33\mathrm{H}$ phases of the H/C(100) surface. The presence of a dihydride unit on the surface leads to a H-C-H bending mode with a frequency near $1475{\mathrm{cm}}^{\ensuremath{-}1}.$ Because there is no vibrational mode with a frequency in this region of the spectrum for the monohydride surface, this peak may be viewed as evidence that dihydride units are present on the surface.