High-resolution in situ electron microscopy of a silicon surface modification by molten aluminium at high temperatures

High-resolution in situ heating experiments in a transmission electron microscope have been applied to a study of the surface modification of Si by molten Al in a vacuum of 1 10^-5 Pa. Transformation from an atomically rough to an atomically flat surface was induced by wetting of molten Al atoms. Si {110}, {112}, {115} and {773} surfaces show sawtooth-like structures composed of nanofacets. It is concluded that molten Al atoms remove a native amorphous oxide layer which existed on the original Si surface, making the surface 'clean' even in a non-ultra-high vacuum such as 10^-5Pa.     

The temperature dependence of the optical dispersion parameters in Si and Ge

Abstract

In Si and Ge, the optical dispersion parameters (single-oscillator energy E_o , dispersion energy E_d and bond energy gap E_g developed by Wemple and DiDomenico, and Phillips) have been analysed in the temperature range 100-300 K using data obtained by Icenogle et al. E_o and E_g exhibit a very small temperature dependence in both materials. The thermal coefficients of the dispersion energy, dE_d/dT, have opposite signs (Si, –41·9 × 10^?4eVK^?1; Ge, +37·7 × 10^?4eVK^?1).     

Hydrostatic pressure-enhanced solid-phase epitaxy

Abstract

A large enhancement of solid-phase epitaxial growth (SPEG) due to hydrostatic pressure is explained by stress-enhanced self-diffusivity in the amorphous solid. The crystallization is by the adjustment of atomic positions in the vicinity of the crystalline/amorphous (c-a) interface due to self-diffusion in the amorphous phase, assisted by a free-energy decrease equal to the difference in free energies between the amorphous and crystalline phases. Owing to a mismatch in the bulk moduli between the amorphous and crystalline phases, non-hydrostatic stresses are developed near the c-a interface under hydrostatic pressure. Non-hydrostatic stresses in the amorphous layer enhance the mobility of point defects in the amorphous layer. This leads to an increased self-diffusivity in the amorphous layer and, therefore, an enhancement of the SPEG rate.     

Effect of Al-Zr clusters on the thermal stability of Cu-Zr-rich Cu-Zr-Al bulk metallic glasses

The Gibbs free energy of formation, ?G^a, of the amorphous phase of Cu–Zr-rich Cu–Zr–Al bulk metallic glasses (BMGs) has been calculated over a broad composition region to re-evaluate their thermal stability, based on the efficient cluster packing model. It is found that Al–Zr clusters have an important effect on the thermal stability of the Cu–Zr–Al BMGs with the same Cu/Zr content ratio. The crystallisation of the BMGs involves crystallisation of both Al–Zr and Cu–Zr clusters. The results could be useful for predicting the thermal stability and understanding the underlying mechanism of crystallisation of multicomponent metallic glasses.     

On the dislocation mechanism of amorphization of Si by indentation

The formation of an amorphous phase underneath a Vickers indentation produced on a Si(001) surface at room temperature has been observed by cross-sectional transmission electron microscopy. Two types of location are observed for the amorphous phase. One is formed just underneath the image of the indentation and the other is parallel to the slip planes of Si. It is concluded that the latter type, at least, is formed as a result of activation of dislocations which is induced by an external shear stress combined with a hydrostatic pressure.     

The soft X-ray emission spectra of Zn,Ga and Ge

Abstract

The experimental L_α1,2 and L_β1 X-ray emission spectra of Zn, Ga and Ge are presented. The high-energy satellite structure of the lines is considered. It appears that the intensity of these satellites relative to the main lines is much larger for Zn than for Ga and Ge. This enhancement effect is interpreted as arising mainly from the 2p^?1 _1/2 → 2p^?1 _3/23d^?1 _3/2,5/2 Coster-Kronig transition occurring in metallic Zn, which is absent in the case of Ga and Ge. Furthermore, for all these elements it is suggested that a significant contribution to the high-energy satellites of the L_α1,2; X-ray emission band arises from the 2s^?1 _1/2 → 2p^?1 _3/23d^?1 _3/2,5/2 Coster-Kronig process, whereas in the case of the satellites of the L_β1 X-ray emission spectrum the main cause would be the 2s^?1 _1/2 → 2p^?1 _1/23d^?1 _3/2,5/2transition.     

Elastic and phonon properties of quaternary Heusler alloys CoFeCrZ (Z = Al,Si, Ga and Ge) from density functional theory

The structural, elastic and phonon properties of the quaternary CoFeCrZ (Z = Al, Si, Ga and Ge) Heusler alloys have been investigated using the generalized gradient approximation method within density functional theory. The ground-state properties, including, lattice constant and bulk modulus are in good agreement with the available theoretical and experimental data. The elastic constants C_ij are computed using the stress–strain technique. The calculated results indicate that CoFeCrZ (Z = Al, Si, Ga and Ge) alloys are ductile materials. Debye temperatures are predicted from calculated elastic constants. The phonon dispersion relations of CoFeCrZ (Z = Al, Si, Ga and Ge) alloys are calculated for the first time using the density functional theory and the direct method with 2 × 2 × 2 supercell.     

Ion-irradiation-induced solid-phase epitaxial growth

Abstract

A model is proposed to explain ion-induced solid phase epitaxial growth (SPEG). The crystallization is by the adjustment of atomic positions in the vicinity of the crystalline/amorphous interface due to self-diffusion in the amorphous solid. The driving energy of SPEG is the difference in free energies between the amorphous and crystalline solids. Irradiation increases the self diffusivity of the amorphous solid by generating point defects in the amorphous solid and thus enhances the crystallization. An expression for the velocity of epitaxial growth is derived. The model explains numerous experimental facts and, at the limit of no irradiation, also gives a correct model of thermal solid-phase epitaxial growth.     

Synthesis of amorphous and quasicrystal phases by mechanical alloying of Ti 45 Zr 38 Ni 17 powder mixtures,and their hydrogenation

Mechanical alloying of Ti 45 Zr 38 Ni 17 powder mixture forms an amorphous phase, but subsequent annealing causes the formation of an icosahedral ( i ) phase. The maximum hydrogen concentration that can be loaded at 573K at a hydrogen pressure of 3.8MPa is the same (\[H]/\[M] 1.5) for the amorphous and i -phase powders. With hydrogenation, the i -phase is almost stable, forming no hydrides, whereas the amorphous phase transforms to a fcc hydride. The activation energy for hydrogen desorption for the i -phase is about 127kJmol -1, which is lower than that for the amorphous phase, suggesting that the i -phase powder may have better properties for hydrogen-storage applications.     

Multilayered phase formation after high-fluence implantation of nickel into aluminium

The formation of crystalline and amorphous layers has been observed in Al crystals after 5MeV Ni implantation up to a fluence of 7.6 10 17 Ni atomscm -2 . The layers are structured into sublayers perpendicular to the implantation direction with sharp interfaces between the phases and almost symmetrically arranged to both sides of the maximum of the implantation profile. The sublayers of thickness of the order of 100nm were analysed by transmission electron microscopy. They consist of polycrystalline Al 3 Ni, crystalline Al 3 Ni precipitates and small amorphous zones, depending on the local Ni implantation concentration.     

Surface energy of free clusters of bubbles: An estimation

We propose an approximate equation for the surface energy of two-dimensional free bubble clusters which we compare with exact calculations of the surface energy of symmetrical clusters consisting of a central bubble surrounded by one or two shells of bubbles of two different areas. The accuracy of the equation is good for relatively narrow distributions of the areas and of the number of sides of the bubbles but underestimates the energy for large widths of those distributions. We propose a similar approximate equation for the surface energy of three-dimensional clusters.     

Unexpected nanoscale phase separation in sputterdeposited Ni-25 at.%Al thin films

Thin films of nominal composition Ni-25at.%Al, sputter deposited from a target of the intermetallic compound Ni 3 Al on unheated substrates, exhibit an unexpected phase separation, in contrast with other intermetallic-forming systems such as Ti-Al which are deposited as compositionally homogeneous amorphous films under similar conditions. Precipitates of a novel tetragonal phase, a few nanometres in size, were formed in the matrix consisting of a fcc Ni-rich Ni(Al) phase and a hcp Ni-rich Ni(Al) phase. Ni-Al films of the same composition deposited on heated substrates exhibited the formation of a single, chemically long-range-ordered Ni 3 Al phase with the L1 2 structure, the thermodynamically stable phase for this composition.     

Solid-solution hardening in dilute and concentrated alloys

Abstract

Numerical analysis of the temperature dependence of the critical resolved shear stress of several binary copper-based alloy single crystals containing 0·11 to 7·6 at.% Mn, 0·01 to 14·0 at.% Al, 0·5 to 8·0 at.% Ge and 5 to 30 at.% Zn, has been carried out in terms of the kink-pair formation model of solid-solution hardening. Several solute atoms are found to be involved in the unit activation process not only in concentrated alloys but also in dilute ones. A single mechanism of solution hardening, which involves the interaction between a dislocation and many solute atoms, is therefore operative in all the alloys referred to above.     

Atomistic calculation of the interaction between an edge dislocation and a free surface

By using molecular statics, we compute the variation in the excess energy associated with the gradual approach of an edge dislocation towards the free surface of a crystal. The calculations rely on a phenomenological potential adapted to aluminium and an appropriate constraint procedure that allows investigations of both the extended and the perfect configurations of the dislocation core. Thereby, an estimation of the energy required for the introduction of a dislocation in a thin film is obtained.     

Ion-beam synthesis of amorphous gallium nitride

Amorphous gallium nitride (a-GaN) has been synthesized for the first time by implanting gallium into amorphous silicon nitride (a-SiNx) films. The a-GaN is only formed when gallium is implanted into hydrogenated amorphous silicon nitride (a-SiNx:Hy) films with x > 1.5. The nitrogen concentration x of the substrate is varied by changing the feed-gas ratio during plasma-enhanced chemical vapour deposition of the nitride film. Using a pre-determined composition and implant condition, the implanted gallium is made to bond with the nitrogen to form a surface layer of a-GaN. Low-temperature annealing, compatible with large-area glass substrates, is then used to increase the thickness of the a-GaN and to transform more of the a-SiNx. X-ray photoelectron spectroscopy and Rutherford back-scattering spectroscopy have been used to examine the bond structure, composition and the depth profile of the synthesized material.     

Dependence of the persistent photoconductivity in doping-modulated a-Si: H multilayers on the exposure temperature

Abstract

We find that the photo-induced excess conductivity (persistent photoconductivity, PPC) in doping-modulated hydrogenated amorphous silicon (a-Si: H) is thermally activated for exposure temperatures above 220 K in agreement with Kakalios and with Hundhausen and Ley. However, we find a temperature regime between 80 and 220 Kin which the PPC measured at 300 K is independent of excitation temperature. Whether the PPC is excited in the thermally activated regime above 220 K or in the temperature-independent regime, the annealing occurs at 410 K. The pre-exponential factor and the activation energy of PPC are related by the Meyer-Neldel rule independent of the excitation temperature.     

Identifying the character of intrinsic stacking faults in the A15 compound Nb3Al

The character of intrinsic stacking faults on {001} in the A15 intermetallic compound Nb₃Al has been investigated using energy-dispersive X-ray spectroscopy in a field-emission-gun transmission electron microscope. Line scans were obtained across edge-on faults with an electron beam of approximate diameter 0.5 nm and it was found that the Al concentration is enhanced at the fault. Taken in conjunction with crystallographic data obtained previously, this implies that a {004} layer composed of Nb atoms only is removed to form these intrinsic stacking faults in Nb₃Al. The configuration of these faults implies that they may form by Nb vacancy coalescence and could act as embryonic nuclei for the precipitation of the D8_bcompound Nb₂Al.     

Evidence for structural disorder in the icosahedral phase

Abstract

We report transmission electron microscopy results on quasicrystalline samples of Al–25 wt% Mn and Al-38 wt% Mn-5 wt% Si melt-spun flakes. The selected-area diffraction patterns (SADPs) from ‘off-axis’ orientations show not only sharp diffraction maxima but also a diffuse ring, invariant of sample thickness, indicating that it does not arise from amorphous surface oxide. Over-exposed SADPs taken from axial orientations also showed the weak diffuse ring. We suggest that our results indicate the presence of disordered material within the icosahedral phase. Models for the icosahedral phase which involve an assembly of icosahedral clusters inherently contain interstitial voids, which may contain disordered material. Thus our results support these models rather than those derived from space-filling filings, or multiple twinning.     

Crystallization behaviour of Al-based amorphous alloy and nanocomposites by rapid quenching

The microstructure and crystallization behaviour of melt-spun Al₈₈Ni₉Ce₂Fe₁ amorphous alloy and nanophase composites have been studied by means of X-ray diffraction, transmission electron microscopy and scanning and isothermal calorimetry. The diffraction patterns from Al₈₈Ni₉Ce₂Fe₁ amorphous alloys are diffuse, indicating a basically amorphous structure but contain two rings presumed to be associated with quenched-in nuclei. In the cases of Al₈₈Ni₉Ce₂Fe₁ nanophase composites, nanoscale precipitated particles are homogeneously dispersed in an amorphous matrix, and the crystallite diameter and volume fraction are sensitive to quenching conditions. During thermal crystallization, a two-step phase transformation occurs in the amorphous alloy and nanocomposites, which is characterized by a diffusion-controlled precipitation of nanoscale Al particles and the growth of a Al₃(Ni, Fe) nanophase prior to a Al₁₁Ce₃ nanophase. This study gives insight into structure-control for obtaining nanophases dispersed in an amorphous matrix by rapid quenching.     

Structural effects on diffusion in metallic glasses: The pressure dependence of Ni diffusion in as-quenched and relaxed Co42Zr58

In order to clarify the role of thermal defects in diffusion in metallic glasses, we have measured the temperature and pressure dependences of diffusion of Ni, a probe for Co, in relaxed and non-relaxed amorphous Co₄₂Zr₅₈ alloys by means of the tracer technique in combination with secondary-ion mass spectroscopy. For the relaxed state, the activation energy and the pre-factor are Q _{= (}1:65=0:08₎ eV and D _{0 =} 2:4+3:9 10-6m2 s-1 respectively. The pressure - 1:9 dependence yields an activation volume V _{act = (}0:66 0:15₎Omega, with Omega being the average atomic volume of the alloy. This value is similar to activation volumes in crystalline materials and is indicative of diffusion via thermally generated defects. Unlike monovacancies in crystals, these defects appear to be spread out, judging from recent isotope-effect measurements. Comparison with literature data shows that the activation volume and hence the diffusion mechanism in metallic glasses clearly depend on the structure and composition even for the very same component. The activation volume in the as-quenched state was found to be _{( )} ^{0:93 0:25 Omega.}