Spin-relaxation and magnetoresistance in FM/SC/FM tunnel junctions

The effect of spin relaxation on tunnel magnetoresistance (TMR) in a ferromagnet/superconductor/ferromagnet (FM/SC/FM) double tunnel junction is theoretically studied. The spin accumulation in SC is determined by balancing of the spin-injection rate and the spin-relaxation rate. In the superconducting state, the spin-relaxation time τ_s becomes longer with decreasing temperature, resulting in a rapid increase of TMR. The TMR of FM/SC/FM junctions provides a useful probe to extract information about spin-relaxation in superconductors.     

Quantitative assessment of solid oxide electrochemical doping

Quantitative analysis of metal cation doping by solid oxide electrochemical doping (SOED) has been performed under galvanostatic doping conditions. A M–β″-Al₂O₃ (M=Ag, Na) microelectrode (contact radius: about 10 μm) was used as cation source to attain a homogeneous solid–solid contact between the β″-Al₂O₃ and doping target. In Ag doping into alkali borate glass, the measured dopant amount closely matched the theoretical value. High Faraday efficiencies of above 90% were obtained. This suggests that the dopant amount can be precisely controlled on a micromole scale by the electric charge during electrolysis. On the other hand, current efficiencies of Na doping into Bi₂Sr₂CaCu₂O_y (BSCCO) ceramics depended on the applied constant current. Efficiencies of above 80% were achieved at a constant current of 10 μA (1.6 A cm⁻²). The relatively low efficiencies were explained by the saturation of BSCCO grain boundaries with Na. By contrast, excess Na was detected on the anodic surface of ceramics at a constant current of 100 μA (16 A cm⁻²). In the present study, we demonstrate that SOED enables micromole-scale control over dopant amount.     

Method of observation of low density interface states by means of X-ray photoelectron spectroscopy under bias and passivation by cyanide ions

X-ray photoelectron spectroscopy (XPS) measurements under bias can observe low density interface states for metal-oxide-semiconductor (MOS) diodes with low densities. This method can give energy distribution of interface states for ultrathin insulating layers for which electrical measurements cannot be performed due to a high density leakage current. During the XPS measurements, a bias voltage is applied to the rear semiconductor surface with respect to the ∼3 nm-thick front platinum layer connected to the ground, and the bias voltage changes the occupation of interface states. Charges accumulated in the interface states shift semiconductor core levels at the interface, and thus the analysis of the bias-induced shifts of the semiconductor core levels measured as a function of the bias voltage gives energy distribution of interface states. In the case of Si-based MOS diodes, the energy distribution and density of interface states strongly depend on the atomic density of silicon dioxide (SiO₂) layers and the interfacial roughness, respectively. All the observed interface state spectra possess peaked-structures, indicating that they are due to defect states. An interface state peak near the Si midgap is attributable to isolated Si dangling bonds at the interface, while those above and below the midgap to Si dangling bonds interacting weakly with Si or oxygen atoms in the SiO₂ layers. A method of the elimination of interface states and defect states in Si using cyanide solutions has been developed. The cyanide method simply involves the immersion of Si in KCN solutions. Due to the high Si-CN bond energy of ∼4.5 eV, the bonds are not ruptured at 800 °C and upon irradiation. The cyanide treatment results in the improvement of the electrical characteristics of MOS diodes and solar cells.     

Notes on the Elliptic Ruijsenaars Operators

We construct the spaces that the elliptic Ruijsenaars operators act on. It is shown that they are extensible to nonnegative selfadjoint operators on a space of square integrable functions, or preserve a finite dimensional vector space of entire functions. These facts are shown in terms of the R-operators which satisfy the Yang–lBaxter equation. The elliptic Ruijsenaars operators are considered as the elliptic analogues of the Macdonald operators or the difference analogues of the Lamé operators.