No abstract

We report on the effect of nonmagnetic spacer layers on the interface magnetism and the exchange bias in the archetypical [Co/CoO] 16 system. The separation of the magnetic bilayers by Au layers with various thicknesses d Au 25 nm leads to a threefold increase of the exchange bias field (H eb). Reflectometry with polarized neutrons does not reveal any appreciable change in the domain population. This result is in agreement with the observation that the granular microstructure within the [Co/CoO] bilayers is independent of d Au. The significant reduction of the magnetic moments in the Co layers can be attributed to interfacial disorder at the Co-Au interfaces. Element-specific x-ray absorption spectroscopy attributes part of the enhancement of H eb to the formation of Co 3 O 4 in the [Co/CoO] bilayers within the multilayers. A considerable proportion of the increase of H eb can be attributed to the loss of magnetization at each of the Co-Au interfaces with increasing d Au. We propose that the interfacial magnetism of ferro-and antiferromagnetic layers can be significantly altered by means of metallic spacer layers thus affecting the exchange bias significantly. This study shows that the magnetism in magnetic multilayers can be engineered by nonmagnetic spacer layers without involving the microstructure of the individual layers.

Platinum and iridium polycrystalline foils were oxidized electrochemically through anodization to create thin platinum and iridium hydrous oxide layers, which were analyzed through laboratory photoelectron spectroscopy during heating and time series (temperature-programmed spectroscopy). The films contain oxygen in the form of bound oxides, water, and hydroxides and were investigated by depth profiling with high-energy photoelectron spectroscopy. The Pt films are unstable and begin to degrade immediately after removal from the electrolyte to form core-shell structures with a metallic inner core and a hydrous oxide outer shell almost devoid of Pt. However, evidence was found for metastable intermediate states of degradation; therefore, it may be possible to manufacture PtOx phases with increased stability. Heating the film to even 100 °C causes accelerated degradation, which shows that stoichiometric oxides such as PtO2 or PtO are not the active species in the electrolyte. The Ir films exhibit increased stability and higher surface Ir content, and gentle heating at low temperatures leads to a decrease in defect density. Although both layers are based on noble metals, their surface structures are markedly different. The complexity of such hydrous oxide systems is discussed in detail with the goal of identifying the film composition more precisely.

A non-optimized interface band alignment in a heterojunction-based solar cell can have negative effects on the current and voltage characteristics of the resulting device. To evaluate the use of Near Edge X-ray Absorption Fine Structure spectroscopy (NEXAFS) as a means to measure the conduction band position, Cu(In,Ga)S 2 chalcopyrite thin film surfaces were investigated as these form the absorber layer in solar cells with the structure ZnO/Buffer/Cu(In,Ga)S 2 /Mo/Glass. The composition dependence of the structure of the conduction bands of CuIn x Ga 1−x S 2 has been revealed for x = 0, 0.67 and 1 with both hard and soft NEXAFS and the resulting changes in conduction band offset at the junction with the buffer layer discussed. A comprehensive study of the positions of the absorption edges of all elements was carried out and the development of the conduction band with Ga content was observed, also with respect to calculated densities of states. Knowledge of these offsets is, therefore, critical to understanding the performance of the resulting solar cell. While the VB offset, ∆E V B , can be determined with established methods, such as combined XPS/UPS [5,6] or Constant Final State Yield Spectroscopy [7], a determination of CB edge positions and offsets, ∆E CB , has proved more difficult. The most common method is simply the assumption that the CB minimum is the energy of the VB plus the band gap. However, the determination of the surface band gap, which is relevant for the band offset, is more involved. Two of the main methods for the direct determination of the CB minimum are inverse photoelectron spectroscopy (IPES) and Near Edge X-ray Absorption Fine Structure (NEXAFS). They have given reliable results in some situations [8,9,10,11], although both have unresolved difficulties and the results must be carefully analyzed. IPES requires high intensity electron irradiation of the sample which often leads to charging of less conductive materials. In the case of NEXAFS these include transition probabilities, spectrum broadening and excitonic or core-hole effects. The latter may cause shifts in the measured position of the absorption edges which do not correspond to the ground state of the material. This is because the position of the absorption edge in NEXAFS represents the energy difference between the initial state (core level) and the final empty state (conduction band) in the material's excited state. The attraction between the core-hole and the excited electron may make the energy difference between the core level and conduction band state appear artificially smaller than it is in the ground state of the material. Also, because the absorption edge represents an energy difference, the energy of the initial state (core level) must be considered to determine whether differences in binding energy could influence the calculated energy of the final conduction band state. Here, while considering only the position of the absorption edge, we assume at first a constant initial state (core level binding) energy, a...

Scanning tunneling microscopy was used to investigate the development of the InAs wetting layer on the GaAs(001)-c(4×4) surface. At low InAs coverages signatures of indium agglomerations form on the surface, before an abrupt change to a (4×3) reconstructed monolayer of In2/3Ga1/3As occurs at about 2/3 ML of deposited InAs. Further indium deposition leads to a second layer with α2(2×4) and β2(2×4) structural units on the surface.