Publications

No abstract is provided for this article.

The $G\phantom{\rule{0}{0ex}}W$ method of Lars Hedin is the most popular approach to predict electron addition and removal energies in solids. To date, most $G\phantom{\rule{0}{0ex}}W$ calculations are performed assuming that the many-body system features sharp resonances, also known as quasiparticle peaks. The authors go beyond this approximation and determine the Green's function $G$ and the screened interaction $W$ fully self-consistently in the $G\phantom{\rule{0}{0ex}}W$ approximation. They find strong broadening of the quasiparticle peaks. Somewhat disappointingly, the predicted band gaps are too large. This notwithstanding, the work is an important step towards a full solution of Hedin's equations.

Oxide structures with nanometric dimensions exhibit novel physical and chemical properties, with respect to bulk oxide materials, due to the spatial confinement and the proximity of the substrate. They derive their atomic structure and morphology, on the one hand, from the interactions at the interface between the oxide overlayer and the substrate and, on the other hand, from kinetic constraints during the growth process. Here we describe the formation of vanadium oxide nanostructures on a single-crystal metal surface and their characterization by scanning tunnelling microscopy (STM) and ab initio density functional theory (DFT) calculations. We show that vanadium oxide nanostructures can be formed on Rh(111) with morphologies ranging from quasi-zero- to three-dimensional and that the oxide growth can be tuned into a particular dimensionality by careful adjustment of experimental parameters. These 'artificial oxide phases' display new physical and chemical properties, which make them potentially interesting materials for nanotechnology applications.

Kurzzusammenfassung Dichtefunktionaltheorie (DFT) ist der Standardansatz zur Modellierung von MIL‐101(Fe) und verwandten eisenbasierten metallorganischen Gerüstverbindungen (MOFs), wobei üblicherweise eine ferromagnetische High‐Spin Konfiguration angenommen wird. Diese weit verbreitete Annahme übersieht jedoch ein zentrales elektronisches Merkmal: Spinfrustration in den dreieckigen ‐O) Zentren. Mithilfe von Flip‐Spin DFT identifizieren wir den tatsächlichen Grundzustand als einen antiferromagnetischen Zustand, der durch konventionelle DFT nicht erfasst wird. Wir zeigen, dass die Vernachlässigung von Spinfrustration in MIL‐101(Fe) zu strukturellen Verzerrungen, fehlerhaften energetischen Vorhersagen und irreführenden Aussagen über Stabilität und Reaktivität führt. Durch explizite Berücksichtigung der Spinfrustration erhalten wir die korrekte Struktur und können die temperaturabhängige ‐ und CO‐Adsorption erklären. Spinfrustration fördert die ‐Fixierung bei Raumtemperatur, während ihr Verlust durch partielle Reduktion von diese unterdrückt, aber die CO‐Adsorption über ‐Rückbindung begünstigt. Diese Erkenntnisse stellen gängige rechnerische Konventionen in Frage und heben Spinfrustration als ein entscheidendes elektronisches Merkmal dieser Materialien hervor.

By means of scanning tunneling microscopy (STM) and density functional theory (DFT) calculations, we characterize at the single-atom level the mechanism of the water formation reaction on the (10 x 2)-O/Rh(110) surface, a prototype of a one-dimensional (1D) oxide where the lattice expansion and the segmentation of the surface play a fundamental role. When the reaction is imaged in the 238-263 K temperature range (35 s/image acquisition time), a peculiar comblike propagation mechanism for the reaction front is found. Fast STM measurements (33 ms/image) prove that this mechanism holds also at room temperature, being therefore an intrinsic characteristic of the reaction on the 1D oxide. DFT calculations explain the observed behavior as due to the interplay between the lattice expansion in the initial surface and its relaxation during the reaction that leads to varying configurations for the reactants. At low temperatures, the reaction produces, in its final stages, a low-coverage, ordered patterning of the surface with residual oxygen. The pattern formation is related to the segmentation of the oxide phase.

We review recent developments in electronic structure calculations that go beyond state-of-the-art methods such as density functional theory (DFT) and dynamical mean field theory (DMFT). Specifically, we discuss the following methods: GW as implemented in the Vienna ab initio simulation package (VASP) with the self energy on the imaginary frequency axis, GW+DMFT, and ab initio dynamical vertex approximation (DΓA). The latter includes the physics of GW, DMFT and non-local correlations beyond, and allows for calculating (quantum) critical exponents. We present results obtained by the three methods with a focus on the benchmark material SrVO3.

The initial oxidation of the Rh(110) surface was studied by scanning tunneling microscopy, core level spectroscopy, and density functional theory. The experiments were carried out exposing the Rh(110) surface to molecular or atomic oxygen at temperatures in the 500-700 K range. In molecular oxygen ambient, the oxidation terminates at oxygen coverage close to a monolayer with the formation of alternating islands of the (10x2) one-dimensional surface oxide and (2x1)p2mg adsorption phases. The use of atomic oxygen facilitates further oxidation until a structure with a c(2x4) periodicity develops. The experimental and theoretical results reveal that the c(2x4) structure is a "surface oxide" very similar to the hexagonal O-Rh-O trilayer structures formed on the Rh(111) and Rh(100) substrates. Some of the experimentally found adsorption phases appear unstable in the phase diagram predicted by thermodynamics, which might reflect kinetic hindrance. The structural details, core level spectra, and stability of the surface oxides formed on the three basal planes are compared with those of the bulk RhO2 and Rh2O3.

We present a strategy combining machine learning and first-principles calculations to achieve highly accurate nuclear quadrupolar coupling constant predictions. Our approach employs two distinct machine-learning frameworks: a machine-learned force field to generate molecular dynamics trajectories and a second model for electric field gradients that preserves rotational and translational symmetries. By incorporating thermostat-driven molecular dynamics sampling, we enable the prediction of quadrupolar coupling constants in highly disordered materials at finite temperatures. We validate our method by predicting the tetragonal-to-cubic phase transition temperature of the organic-inorganic halide perovskite MAPbI$_3$, obtaining results that closely match experimental data.

The on-the-fly generation of machine-learning force fields by active-learning schemes attracts a great deal of attention in the community of atomistic simulations. The algorithms allow the machine to self-learn an interatomic potential and construct machine-learned models on the fly during simulations. State-of-the-art query strategies allow the machine to judge whether new structures are out of the training data set or not. Only when the machine judges the necessity of updating the data set with the new structures are first-principles calculations carried out. Otherwise, the yet available machine-learned model is used to update the atomic positions. In this manner, most of the first-principles calculations are bypassed during training, and overall, simulations are accelerated by several orders of magnitude while retaining almost first-principles accuracy. In this Perspective, after describing essential components of the active-learning algorithms, we demonstrate the power of the schemes by presenting recent applications.