Effect of boron non-stoichiometry on B-site in perovskite type structure ScBxRh3 and CeBxRh3 on charges of atoms on A-site: study by X-ray photoelectron and nuclear magnetic resonance spectroscopies

The solid solutions of ScBRh₃-ScRh₃ and CeBRh₃-CeRh₃ are synthesized by the arc melting method, where RBRh₃ and RRh₃ (R=rare earth element) have perovskite and AuCu₃ type structures, respectively. The binding energy of Sc 2p_3/2 for ScB_xRh₃ increases with the boron concentration. The Knight shift of ⁴⁵Sc observed by nuclear magnetic resonance spectroscopy decreases with increase of boron concentration. The decrement of the Knight shift corresponds the Sc 4s electron density at the Fermi level. The intensity ratio of f²: f¹: f⁰ of Ce 3d XPS spectrum changes with boron concentration of CeB_xRh₃. It is concluded that in both cases of ScB_xRh₃ and CeB_xRh₃ the charge on the atoms on A-site changes with the concentration of the atoms on B-site, where the atoms are not directly bound.     

Oxidation behavior of NiAl alloy at low temperatures

Oxidation behavior of NiAl alloy at low temperatures was studied. A NiAl plate was oxidized by exposure to ambient atmosphere at room temperature, heated at 473 K in air, and heated at 773 K in air. The oxide formed on the NiAl surface was investigated by angle‐resolved X‐ray photoelectron spectroscopy (AR‐XPS). Chemical composition and atomic concentration in the oxide layer were analyzed with factor analysis of XPS spectra. Exposure of the NiAl plate to the ambient atmosphere resulted in the formation of an Al₂O₃ layer along with a small amount of NiO. Oxidation of the NiAl plate at 473 K in air formed a film of double‐layered oxide; the top layer consisted of NiAl₂O₄ and a small amount of NiO, and the second layer was Al₂O₃. Successive oxidation at 773 K only changed the oxide‐layer thickness without changing the structure. Formation of oxide observed in the present study corresponds to the thermodynamic prediction for the oxidation behavior of NiAl at 1373 K. Copyright © 2007 John Wiley & Sons, Ltd.     

Impact of amorphous Ge thin layer at the amorphous Si/Al interface on Al-induced crystallization

We investigated the impact of an amorphous Ge (a-Ge) thin layer inserted at the amorphous Si (a-Si)/Al interface on Al-induced crystallization. In situ observation of the growth process clarified that the nucleation rate is drastically reduced by insertion of a-Ge, which led to increase in the average size of crystal grains. This was interpreted as resulting from decrease in the driving force of crystallization, mainly due to the larger solubility of Ge in Al than that of Si in Al. The obtained films were SiGe alloys with lateral distribution of Ge content, and its origin is discussed based on the two-step nucleation process.     

Growth of multicrystalline Si with controlled grain boundary configuration by the floating zone technique

The floating zone technique was employed to grow multicrystalline Si with controlled grain boundary configuration. Purposely designed bi-crystals were utilized as seed crystals to investigate the effect of the tilt angle from the perfect twin boundary on the growth behavior. When the growth was initiated from a bi-crystal with a Σ3 twin boundary, no particular change took place on the grain boundary configuration during growth. On the other hand, the decrease of the tilt angle during growth was observed when the growth was initiated from a bi-crystal with a tilted boundary from Σ3. This was accompanied by the appearance of new crystal grains. The reduction of the total interface energy would be a possible driving mechanism for this phenomenon.     

X-ray photoelectron spectroscopic studies on initial oxidation of iron and manganese mono-silicides

Initial oxidation of iron and manganese mono-silicides (FeSi and MnSi) surfaces was studied by X-ray photoelectron spectroscopy (XPS). Clean surfaces of these silicides were prepared by fracturing in an ultra high vacuum, and then the fractured surfaces were oxidized by exposing to high-purity oxygen at pressures up to 1.3 Pa. For the clean FeSi surface, positive chemical shifts of the Fe 2p_3/2 and Si 2p peaks from elemental Fe and Si were 0.5 eV and 0.1 eV, respectively. For the clean MnSi surface, a negative chemical shift of the Si 2p peak from elemental Si was 0.1 eV. Iron on the FeSi surface was oxidized at an oxygen pressure of 1.3 Pa, whereas the silicon was oxidized under the pressure of 1.3 × 10⁻⁶ Pa, indicating that oxidation of silicon occurred prior to that of iron. Manganese and silicon on the MnSi were simultaneously oxidized in the range from 1.3 × 10⁻⁶ Pa to 1.3 × 10⁻³ Pa; however, over the pressure of 1.3 Pa, the oxidation of manganese occurs prior to that of silicon. These oxidation behaviors at low oxygen pressures were similar to those of the FeSi and MnSi fractured in air.     

X-ray photoelectron spectroscopic studies on oxidation behavior of nickel and iron aluminides under oxygen atmosphere at low pressures

Oxidation behaviors of NiAl, Ni₃Al, and FeAl under oxygen atmosphere at low pressures were studied by X-ray photoelectron spectroscopy (XPS). Clean surfaces of these aluminides were prepared by fracturing in an ultra high vacuum, and then the fractured surfaces were oxidized by exposing to high-purity oxygen at pressures up to 1.3 Pa without exposing to air. The oxides formed on NiAl and FeAl surfaces were Al₂O₃, whereas the oxide on Ni₃Al was NiAl₂O₄. Aluminum, nickel, and iron on clean surfaces were oxidized even at a pressure of 1.3 × 10⁻⁶ Pa. The oxidation evolves with an increase in the pressure of oxygen, and further oxidation of aluminum occurs prior to that of nickel or iron. The oxidation behaviors under such oxygen atmosphere were similar to those of the aluminides oxidized in air, and these behaviors could be predicted from thermodynamic consideration.     

Comparison of the background corrected valence band XPS spectra of Fe and Co aluminides and silicides with their electronic structures

The background corrected valence band XPS spectra and the electronic structures of FeAl, FeSi, CoAl and CoSi were studied. Clean surfaces of the polycrystalline samples were obtained by in situ fracturing of the samples in an XPS spectrometer. The energy loss parts of the Fe 2p, Co 2p and valence band spectra were removed by the deconvolution method using Al 2s or Si 2s spectra as response functions. CoAl exhibited a satellite peak in the Co 2p region, but the other compounds had no clear satellite peaks in the Co 2p and Fe 2p regions. The experimentally background corrected valence band spectra were compared with the calculated spectra using the first-principle band calculation. There were large discrepancies between the spectra above the binding energy of 5 eV. These indicated that the experimental spectra could not be explained by the electronic structures of the ground states alone.     

Comparison of intrinsic zero-energy loss and Shirley-type background corrected profiles of XPS spectra for quantitative surface analysis: Study of Cr, Mn and Fe oxides

The intrinsic zero-energy loss profiles of transition metal 2p and 3p XPS spectra for Cr, Mn, and Fe oxides are obtained by spectral deconvolution and compared with Shirley-type background corrected profiles. The metal core level spectra are deconvoluted by O 1s spectra as the response function of each oxide. As the O 1s spectra include intrinsic and extrinsic energy loss parts, the background corrected core level spectra are zero-energy loss spectra. The good agreement of the deconvoluted spectra with the reported spectra obtained by the many body effect theory indicates that the background subtraction method is accurate. A comparison of the deconvoluted with the background corrected spectra of the Shirely-type subtraction reveals that almost all the spectra coincide with each other except for Fe 3p with -Fe₂O₃. The good coincidence of the Shirley-type corrected spectra with the deconvoluted and calculated spectra indicates that Shirley-type background correction can be used for daily quantitative surface analysis.     

X-ray photoelectron spectroscopic studies on phase identification and quantification of nickel aluminides

Core-level XPS spectra for clean surfaces of Ni₃Al, NiAl, and NiAl₃ alloys were studied. The clean surfaces were obtained by fracturing in the ultra-high vacuum chamber. The positive chemical shifts of Ni 2p_3/2 peak for NiAl and NiAl₃ from Ni metal were 0.2 and 1.0 eV, respectively. The negative shift for Al 2p peak and the positive shift for Ni 3p peaks increased with the decreasing concentration of the corresponding elements. The peak position of the bulk plasmon loss peak for Al 2s peak shifted toward higher energy side, and further, the intensity ratio decreased with the decrease in aluminum concentration. Both the peak intensity ratios of Al 2p to Ni 3p determined by factor analysis and convenient separation are proportional to the atomic ratio of aluminum to nickel. The results indicate that the intensity ratio of Al 2p to Ni 3p determined by these two methods can be applied to the quantification for the surface of the nickel-aluminum alloys.     

Phase diagram of growth mode for the SiGe/Si heterostructure system with misfit dislocations

The strain, surface and interface energies of the SiGe/Si (SiGe grown on Si) heterostructure system with and without misfit dislocations were calculated for the Frank–van der Merwe (FM), Stranski–Krastanov (SK) and Volmer–Weber (VW) growth modes essentially based on the three kinds of fundamental and simple structures. The free energies for each growth mode were derived from these energies, and it was determined as a function of the composition and layer thickness of SiGe on Si. By comparison of the free energies, the phase diagrams of the FM, SK and VW growth modes for the SiGe/Si system were determined. The (1 1 1) and (1 0 0) reconstructed surfaces were selected for this calculation. From the phase diagrams, it was found for the growth of SiGe on Si that the layer-by-layer growth such as the FM mode was easy to be obtained when the Ge composition is small, and the island growth on a wetting layer such as the SK mode was easy to be obtained when the Ge composition is large. The VW mode is energetically stable in the Ge-rich compositional range, but it is difficult for the VW mode to appear in the actual growth of SiGe on Si because the VW region is right above the SK region. The regions of the SK and VW modes for the (1 1 1) heterostructure are larger than those for the (1 0 0) one because the strain energy of the (1 1 1) face is larger than that of the (1 0 0) face. The regions of the SK and VW modes for the heterostructure with misfit dislocations are narrower than those for the one without misfit dislocations because the strain energy is much released by misfit dislocations. The phase diagrams roughly explain the behavior of the FM and SK growth modes of SiGe on Si.