Stable cation inversion at the MgAl2O4(100) surface.

From an interplay of atom-resolved noncontact atomic force microscopy, surface x-ray diffraction experiments, and density functional theory calculations, we reveal the detailed atomic-scale structure of the (100) surface of an insulating ternary metal oxide, MgAl2O4 (spinel). We surprisingly find that the MgAl2O4(100) surface is terminated by an Al and O-rich structure with a thermodynamically favored amount of Al atoms interchanged with Mg. This finding implies that so-called Mg-Al antisites, which are defects in the bulk of MgAl2O4, become a thermodynamically stable and integral part of the surface.

DFT+U study of defects in bulk rutile TiO(2).

We present a systematic study of electronic gap states in defected titania using our implementation of the Hubbard-U approximation in the grid-based projector-augmented wave density functional theory code, GPAW. The defects considered are Ti interstitials, O vacancies, and H dopants in the rutile phase of bulk titanium dioxide. We find that by applying a sufficiently large value for the Hubbard-U parameter of the Ti 3d states, the excess electrons localize spatially at the Ti sites and appear as states in the band gap. At U=2.5 eV, the position in energy of these gap states are in fair agreement with the experimental observations. In calculations with several excess electrons and U=2.5 eV, all of these end up in gap states that are spatially localized around specific Ti atoms, thus effectively creating one Ti(3+) ion per excess electron. An important result of this investigation is that regardless of which structural defect is the origin of the gap states, at U=2.5 eV, these states are found to have their mean energies within a few hundredths of an eV from 0.94 eV below the conduction band minimum.

Recent density functional studies of hydrodesulfurization catalysts: insight into structure and mechanism.

The present article will highlight some recent density functional theory (DFT) studies of hydrodesulfurization (HDS) catalysts. It will be summarized how DFT in combination with experimental studies can give a detailed picture of the structure of the active phase. Furthermore, we have used DFT to investigate the reaction pathway for thiophene HDS, and we find that the reaction entails a complex interplay of different active sites, depending on reaction conditions. An investigation of pyridine inhibition confirmed some of these results. These fundamental insights constitute a basis for rational improvement of HDS catalysts, as they have provided important structure-activity relationships.

Biomimetic hydrogen evolution: MoS2 nanoparticles as catalyst for hydrogen evolution.

The electrochemical hydrogen evolution reaction is catalyzed most effectively by the Pt group metals. As H2 is considered as a future energy carrier, the need for these catalysts will increase and alternatives to the scarce and expensive Pt group catalysts will be needed. We analyze the ability of different metal surfaces and of the enzymes nitrogenase and hydrogenase to catalyze the hydrogen evolution reaction and find a necessary criterion for high catalytic activity. The necessary criterion is that the binding free energy of atomic hydrogen to the catalyst is close to zero. The criterion enables us to search for new catalysts, and inspired by the nitrogenase active site, we find that MoS2 nanoparticles supported on graphite are a promising catalyst. They catalyze electrochemical hydrogen evolution at a moderate overpotential of 0.1-0.2 V.

A density functional study of the chemical differences between Type I and Type II MoS2-based structures in hydrotreating catalysts.

Density functional theory is used to investigate the origin of the activity differences between Type I and Type II MoS2-based structures in hydrotreating catalysts. It is well known that the Type II structures, where only weak interactions with the support exist, have a higher catalytic activity than Type I structures, where Mo-O linkages to the alumina are present. The present results show that the differences in activities for MoS2 and Co-Mo-S structures can be attributed to the electronic and bonding differences introduced by the bridging O bonds. We find that the Mo-O linkages are most probably located on the (1010) S edge. The presence of oxygen linkages increases the energy required to form sulfur vacancies significantly so that almost no vacancies can be formed at these and neighboring sites. In this way, the reactivity of the S edge is reduced. In addition, the studies also show that the linkages introduce changes in the one-dimensional metallic-like brim states. Furthermore, the presence of oxygen linkages also changes the energetics of hydrogen adsorption, which becomes less exothermic on sulfur sites directly above linkages and more exothermic on sulfur sites adjacent to linkages. The present results explain previously observed differences in Type I-Type II transition temperatures for Co-Mo-S structures with different Co contents.

Chemical activity of the nitrogenase FeMo cofactor with a central nitrogen ligand: density functional study.

We investigate the chemical consequences of a central ligand in the nitrogenase FeMo cofactor using density functional calculations. Several studies have shown that the central ligand most probably is a nitrogen atom, but the consequences for the chemical reactivity of the cofactor are unknown. We investigate several possible routes for insertion of the central nitrogen ligand and conclude that all routes involve barriers and intermediate states, which are inaccessible at ambient conditions. On this basis we suggest that the central nitrogen ligand is present at all times during the reaction. Furthermore, we investigate how the FeMoco with the central ligand can interact with N(2) and reduce it.

Epitaxial growth with pulsed deposition: submonolayer scaling and Villain instability.

It has been observed experimentally that under certain conditions, pulsed laser deposition (PLD) produces smoother surfaces than ordinary molecular beam epitaxy (MBE). So far, the mechanism leading to the improved quality of surfaces in PLD is not yet fully understood. In the present work, we investigate the physical properties of a simple model for PLD, in which the transient mobility of adatoms and diffusion along edges is neglected. Analyzing the crossover from MBE to PLD, the scaling properties of the time-dependent nucleation density as well as the influence of Ehrlich-Schwoebel barriers, we find that there is indeed a range of parameters, where the surface quality in PLD is better than in MBE. However, since the improvement is weak and occurs only in a small range of parameters we conclude that deposition in pulses alone cannot explain the experimentally observed smoothness of PLD-grown surfaces.

Modeling a central ligand in the nitrogenase FeMo cofactor.

In very recent work by Einsle et al. (Science 2002, 297, 1696), a new X-ray crystallographic structure of the FeMo cofactor of nitrogenase with a central ligand was presented. The central ligand is a light atom (N, O, or C), and Einsle et al. suggest that it is nitrogen. We present density functional calculations on the FeMo cofactor, and we investigate N, O, and C as central ligands. We show that both N and O lead to energetically stable FeMo cofactor structures, whereas C is energetically unfavorable. By comparison of bond geometries with the crystallographically determined values, we show that the central ligand is most likely nitrogen.