The low energy cluster beam deposition technique (LECBD) is developed to prepare original nanostructured systems from clusters preformed in the gas phase. The nucleation and growth mechanism specific of the cluster deposits is studied in view to control the nanostructured morphology of the cluster assembled films, especially for future applications to high areal density devices. Examples of the synthesis and properties of metallic nanostructures as well as covalent ones are presented. In the first case, systems from noble metal clusters (Ag, Au, and Ag_xAu_1−x) and transition metal based clusters (Co, SmCo₅) assemblies are studied for their original optical or magnetic properties, respectively. In the second case, covalent semiconducting films exhibiting specific electronic structures are obtained by Si, C, and Si_xC_1−x cage like clusters (fullerenes and heterofullerenes) depositions.

Preparation methods of nanostructured materials are discussed and the main attention is given to the film/coating technique. Short history is also reported on mechanical properties of the thin films. The most interesting result is application of structure and properties of nanostructured films.

The atomistic structure of metallic atom clusters is discussed based on the experimental results with the following conclusions: (1) When ultrafine particles of metals become smaller than the critical sizes corresponding to the nucleus size of the crystals, they behave as atom clusters and their atomistic structure changes as a function of the number of constituent atoms, i.e., the many-body potential. (2) Icosahedrons are formed first and change to the cuboctahedral structure before such atom clusters take the final crystal structures. (3) The phonon mode of these atomistic structures is very much softened, and thus the zigzag atom-chains are formed dynamically around the average position of atoms even in the cuboctahedral structure. (4) The hybrid orbital corresponding to the final crystal structure is formed when atom clusters grow to the critical size. As a result, the binding strength sufficiently increases along a certain crystal axis such as the ⟨110⟩ in the fcc structure and the ⟨111⟩ in the bcc one, so that atomic rearrangement can occur within the ultrafine particles, including disappearance of multi-twin in Au particles, to take the bonding state of crystals.

Cr clusters have been produced by a plasma-gas-condensation type cluster deposition apparatus, and studied using a time-of-flight mass spectrometer and a transmission electron microscope. The Cr clusters formed in high pressure inert (Ar and/or He) gas atmosphere are of an A15-type structure. When an O₂ gas is mixed with the inert gases in the source (sputtering) chamber, a bcc phase is formed together with Cr₂O₃. The O₂ gas introduction leads to an increase in the gas temperature of the source chamber probably due to release of the formation enthalpy of the oxide. The A15 phase is annealed by such excess heat and becomes the equilibrium bcc phase. The sizes of bcc clusters are smaller than those of the A15-clusters, probably due to the heterogeneous nucleation promoted by the oxide formation.

Zr₇₀Cu₃₀ and Zr₇₀Cu₂₀Ir₁₀ glassy alloys were prepared by a single roller melt-spinning method and their crystallization processes were studied by differential scanning calorimetry, X-ray diffraction and transmission electron microscopy. For the Zr₇₀Cu₃₀ alloy, the glassy state crystallizes through a single exothermic reaction due to the precipitation of the stable Zr₂Cu crystalline phase. For the Zr₇₀Cu₂₀Ir₁₀ alloy, a metastable icosahedral quasicrystalline phase precipitates from the glassy matrix in the initial crystallization process, followed by the phase transformation to Zr₂Cu, Zr₂Ir and unknown crystalline phases upon further annealing. These results illustrate that the partial substitution of Cu by Ir in the Zr₇₀Cu₃₀ glassy alloy is effective in promoting the precipitation of the metastable icosahedral quasicrystalline phase. To explain these experimental results, the existence of icosahedron-like atomic clusters in the Zr₇₀Cu₂₀Ir₁₀ glassy alloy is suggested.

This article was retracted. See the Notification.

A nano icosahedral quasicrystalline phase was confirmed as the primary precipitation phase in the melt-spun Zr₆₅Al_7.5Ni₁₀Ag_17.5 quaternary glassy alloy with a two-stage crystallization process. The onset temperature of the first exothermic peak is 704 K at a heating rate of 0.67 Ks⁻¹. Although no significant X-ray diffraction peaks were observed in the sample annealed at 705 K, we confirmed the formation of the icosahedral phase by nano beam electron diffraction using FE-TEM . The icosahedral particles have an extremely fine grain size in the diameter range of 2 to 4 nm. In addition to the icosahedral phase, the nano crystalline particles corresponding to Zr₂Ag and/or fcc-Zr₂Ni structures are also observed in the annealed sample. The second exothermic peak denotes the transition from the nano icosahedral to the crystalline phase of Zr₂Ag, fcc-Zr₂Ni and Zr₄Al₃. The formation of the nano icosahedral phase as the primary phase is due to a high nucleation rate and strongly implies the existence of an icosahedral short-range order in the glassy state.

We investigated the transformation behavior from glassy to icosahedral and/or fcc Zr₂Ni phases in the Zr₆₅Al_7.5Ni₁₀Cu_17.5−xPd_x (x=0 to 4) glassy alloys with low oxygen contents of less than 400 ppm mass%. The primary phase is a metastable single fcc Zr₂Ni phase in the x=0 alloy and a single icosahedral quasicrystalline phase in the 2 to 4 at%Pd alloys. The mixture phase of icosahedral and fcc Zr₂Ni precipitates at the initial crystallization stage in the x=1 alloy. It is therefore confirmed that the icosahedral phase is formed by the addition of 1 at% Pd to the Zr–Al–Ni–Cu glassy alloy. The icosahedral and Zr₂Ni particles have grain sizes in the diameter range of 500 to 1000 nm and are in an isolated state for the sample annealed at a temperature near the crystallization temperature. A significant redistribution leading to the enrichment of Zr in the icosahedral phase is confirmed. Moreover, it is recognized that Ni and Cu are rejected from the icosahedral phase. We also observed the enrichment of Zr and Ni and the rejection of Cu in the Zr₂Ni phase. No significant difference in Al content is observed among the icosahedral, Zr₂Ni and residual glassy phases. The Pd content in the icosahedral phase is the same as that in the Zr₂Ni phase. The difference of Zr, Ni and Cu contents among the icosahedral, Zr₂Ni and remaining glassy phases is in the range of 2 to 9 at%. No significant difference in the Pd content is observed between the icosahedral and Zr₂Ni phases. Considering that the diameter of the primary phase decreases significantly with increasing Pd content, it is suggested that Pd plays a dominant role in the increase in the number of nucleation sites of the icosahedral phase. Moreover, it is strongly suggested that the icosahedral phase originates from the icosahedral short-range order with an increase of nucleation rate as a result of the addition of noble metal.

A nanocrystalline (NC) bulk glass Zr₅₅Al₁₀Cu₃₀Ni₅(at%) containing nano-scale crystals embedded uniformly in a glassy matrix has both high tensile strength of 1.7 GPa and high ductility. The new alloy is therefore expected to have practical applications in machines and structures. The influences of frequency and overload on fatigue crack propagation behavior of the NC bulk glass were examined. The fatigue crack propagation rate da⁄dn less than 3×10⁻⁵ mm/cycle was independent of frequency in the frequency range of 0.1 to 50 Hz at the stress ratio of 0.1 under sine and triangular waves. When the overload ratio (overload/baseline load) was large, a complete crack arrest occurred and the ΔK value just before a crack regrowth was three times larger than the threshold stress intensity factor range ΔK_th. The reason for the crack arrest was not explained by the crack closure effect. The overloading induced the kinking and branching of the crack. The stress reduction near the crack tip due to the kinking and branching of the crack and the crack closure effect gave an appropriate explanation for the complete crack arrest and the larger threshold stress intensity factor range. When the overload ratio was small, a temporary crack arrest occurred and the kink and branching of cracks occurred intermittently at the crack front.

The influence of iron addition on the crystallisation behaviour of Zr₆₅Al_7.5Cu_17.5Ni₁₀ bulk metallic glass was investigated by differential scanning calorimetry, (DSC), X-ray diffraction, (XRD), transmission electron microscopy, (TEM) and magnetisation measurements. The amorphous Zr₆₅Al_7.5Cu_17.5Ni₁₀ alloy crystallises eutectically into CuZr₂ and Zr₆NiAl₂. Addition of iron in amorphous (Zr₆₅Al_7.5Cu_17.5Ni₁₀)_100−xFe_x (0≤x≤20) leads to a changed crystallisation sequence and to the formation of nanocrystals. The formation of a cubic NiTi₂-type phase (S.G. Fd \\bar3m,a_o=1.22 nm) is the first step of crystallisation in amorphous alloys with iron contents x≥1. Depending on the iron content the average crystallite size decreases to the nanometer regime. Ultrafine nanoclusters of down to 2 nm in size are formed as the first step of crystallisation for amorphous Zr₅₂Al₆Cu₁₄Ni₈Fe₂₀ due to a high nucleation rate combined with a low growth velocity. The clusters grow by Ostwald ripening during isothermal annealing up to 5 nm average crystallite size. The magnetic behaviour of the (Zr₆₅Al_7.5Cu_17.5Ni₁₀)_100−xFe_x alloys is dominated by temperature-independent Pauli paramagnetism for x≤15. For x=20, a small contribution of magnetic clusters is observed. These ferromagnetic clusters are in accordance with statistic composition fluctuations within the homogeneous amorphous phase. With the formation of nanocrystals the size of the magnetic clusters increases to the same order of magnitude.

The tensile deformation behavior of the supercooled liquid in the bulk glassy Zr₆₅Al_7.5Ni₁₀Cu_12.5Ag₅ and Zr₆₅Al_7.5Ni₁₀Cu_12.5Pd₅ alloys was investigated in comparison with the glassy Zr₆₅Al_7.5Ni₁₀Cu_17.5 alloy. An icosahedral phase precipitates as the primary precipitation phase from the supercooled liquid in the Ag- or Pd-containing alloy. The deformation behaviors of both the alloys were similar to that of the Zr₆₅Al_7.5Ni₁₀Cu_17.5 alloy at relatively high strain rates of about 1×10⁻² s⁻¹. While the largest elongation of the Zr₆₅Al_7.5Ni₁₀Cu_17.5 alloy, 270%, was obtained at 3×10⁻³ s⁻¹, those of the alloys containing Ag or Pd were about 400% at 1×10⁻³ s⁻¹. Apparent strain hardening due to the growth of the icosahedral particles was observed during the deformation of both the alloys at lower strain rates. The plastic deformation of the supercooled liquid continues even after the precipitation of the icosahedral particles, while the deformation of the Zr₆₅Al_7.5Ni₁₀Cu_17.5 alloy is inhibited by the onset of crystallization.

Transition from amorphous to nanocrystalline state has been investigated from the viewpoint of rate theory in Fe–Co–Zr–B metallic glass by evaluation of changes of electrical resistivity using a novel method of continuous distribution analysis and by transmission electron microscopy and X-ray diffraction. Ordering process inside α-FeCo nanograins has been observed. The mechanism controlling nanocrystallization is the same throughout the entire reaction. Before the nanocrystallization is over, a new microprocess becomes active. The units active in initial and final stages of transformation may differ in local chemical ordering. The results allow to infer on the character of the original amorphous structure for which we propose two types of internally correlated and chemically ordered regions with different Zr content.

Local atomic structures in amorphous Fe₇₀M₁₀B₂₀ (M=Cr, W, Nb, Zr and Hf) alloys with different ΔT_x values were studied by the ordinary X-ray diffraction, anomalous X-ray scattering and EXAFS methods. Their local the atomic structures basically resemble each other, i.e. a random network structure of triangular prisms. Only difference in local atomic structure is a shape of the local structural unit. In the amorphous alloys containing M elements larger than Fe atoms, the prisms show distorted shapes due to a size difference Δr between M and Fe. A linear relation between Δr and ΔT_x, and a comparison of their crystallization processes suggest us that their thermal stability is attributed to difficulty in rearranging the irregular structural units because of a non-zero Δr value.

The soft magnetic properties of nanocrystalline Fe–Nb–B and Fe–Nb–B–P alloys produced in the atmosphere by a melt-spinning method have been investigated. The nanocrystalline Fe_100−x−yNb_xB_y ternary alloys show good soft magnetic properties, relative permeability (μ^′) above 35000 at a frequency of 1 kHz and coercive force (H_c) below 5.0 Am⁻¹, as well as high saturation magnetic induction (B_s) above 1.55 T in the compositional range of x=6.5–6.7 and y=9.3–10.0 at%. The soft magnetic properties of the nanocrystalline Fe–Nb–B ternary alloys are improved by 0.5–1.5 at% substitution of P for B, without decreasing their B_s. The magnetostriction (λ_s) value increases and the mean grain size of bcc-Fe phase (D) decreases slightly by substitution of P for B . The nanocrystalline Fe₈₄Nb_6.5B₉P_0.5 and Fe₈₄Nb_6.5B_8.5P₁ alloys show good soft magnetic properties, μ^′ of 46000–47000 at a frequency of 1 kHz, H_c of 3.6–3.9 Am⁻¹ and the core loss of the 0.09 Wkg⁻¹ at maximum induction (B_m) of 1.33 T and a frequency of 50 Hz as well as high B_s of 1.60 T, suggesting that these nanocrystalline Fe–Nb–B–P alloys are suitable for a core materials for pole transformers.

The influence of annealing temperature, rate of cooling, magnetic field frequency under thermomagnetic treatment on the hysteresis loop shift field value of Fe₅Co₇₀Si₁₅B₁₀ and Fe₆₀Co₂₀Si₅B₁₅ amorphous alloys samples has been studied. The relation of the structural state of amorphous alloys with the shift field value of the hysteresis loop has been studied. The physical explanation of this effect is proposed on the basis of obtained data.

The value of the supercooled liquid region defined by the difference between the glass transition temperature (T_g) and the crystallization temperature (T_x), ΔT_x (=T_x−T_g), was 43 K for an Fe_66.5Co₁₀Pr_3.5B₂₀ glassy alloy. The crystallized structure consists of Fe₃B, Pr₂Fe₁₄B, α-Fe and remaining glassy phase after annealing at about 863 K for 420 s and changes to a mixture of Fe₃B, Pr₂Fe₁₄B and α-Fe phases after annealing at higher temperatures. The grain sizes after annealing at 863 K are about 20 nm for Fe₃B, 10 nm for Pr₂Fe₁₄B, 20 nm for α-Fe and 5 nm for the remaining glassy phase. The maximum energy product (BH)_max was obtained for the alloy containing the residual glassy phase subjected to annealing at 863 K for 420 s. The remanence (B_r), coercivity (_iH_c) and (BH)_max of the optimally annealed Fe_66.5Co₁₀Pr_3.5B₂₀ alloy are 1.35 T, 211 kA/m and 107 kJ/m³, respectively. The hard magnetic properties are interpreted to result from the exchange magnetic coupling among Pr₂Fe₁₄B, Fe₃B, α-Fe and remaining glassy phase. The formation of the finely mixed structure caused by the residual glassy phase is considered to be responsible for the good hard magnetic properties with (BH)_max above 100 kJ/m³ for the new type (Fe, Co)–Pr–B alloy with low rare earth and a high boron contents.

Bulk Nd₆₀Fe₃₀Al₁₀ metallic glasses are found to exhibit a complex magnetic behavior at very low temperatures related to the coexistence of various Nd-rich and Fe-rich magnetic short range orderings, as evidenced by AC-susceptibility studies in combination with hysteretic loops at fields up to 30 T . Above 80 K the magnetic behavior is shown to correlate with Stoner-Wohlfarth hysteresis loops with pronounced thermal activation effects. These observations suggest the existence of an Fe-rich Fe–Nd hard magnetic ’phase’ formed by an ensemble of randomly oriented uniaxial magnetic nanoentities embedded in a Nd-rich matrix. The exchange coupling and development of a large axial crystal field anisotropy at very low temperatures and the uncoupling above 70 K of the Fe-rich Fe–Nd hard magnetic phase to the Nd-rich matrix appears to be at the origin of the observed overall magnetic behavior.

Nanometer-scale structures were examined for Al₉₂V₃Fe₃Zr₂ ribbons rapidly solidified by a single roller melt-spinning, as a function of a circumferential velocity in the range between 10 and 50 m/s. Deformed scattering patterns along the direction of the ribbons are detected by a pinhole small-angle X-ray instrument for the ribbons rapidly solidified with the circumferential velocities higher than 20 m/s. The deformed patterns can be attributed to the rows of nearly spherical particles into which the elongated particles have disintegrated by shear flow caused by the melt spinning. Size of precipitates decreases with increasing the circumferential velocity, and some change in the precipitation mechanism is suggested to occur at a circumferential velocity of 30 m/s.

Nanostructure Al-based binary alloys with additions of Fe have been prepared by electron-beam evaporation. The composition dependence of microstructures and mechanical properties has been studied. The deposits up to an Fe content of 4 at% consist of a supersaturated solid solution of Al. At higher Fe contents, the deposits consist of an amorphous and crystalline phase with a grain size below 40 nm. At the content range in which the supersaturated solid solution formed, as the content increased, the crystalline size became finer. The Vickers hardness increases with increasing the content of Fe and reaches values of 200 Hv and 600 Hv for contents of 1 at% and 15 at%, respectively. The tensile strength of the deposit exhibited a high strength of 1000 MPa at the Fe content of 2.5 at% with grain size 40 nm. The strength and hardness followed the Hall-Petch relation up to the grain size of 40 nm.

Mechanical alloying (MA) was used in the investigation of the Fe–Zn binary system, which, in the equilibrium state, exhibits negligible mutual solid solubility at room temperature. The formation of single-phase bcc solid solutions with a nanometer-scaled microstructure was achieved by MA in Fe_100−xZn_x (x≤65) powder mixtures. The nearest neighbor distances of these Fe–Zn solid solutions obey the linear relationship given by the Vegard’s law for ideal solutions, suggesting that the solid solutions formed by MA are homogeneous. The average lattice volume expansion also linearly increases to about 13.4% at 65%Zn. At elevated temperatures the nanostructured solid solutions undergo phase transformations yielding intermetallic compounds. The mechanisms for the phase formation in the Fe–Zn system are discussed.

The pressure-induced bcc-to-hcp phase transformation has been studied by x-ray diffraction for bulk iron as well as nanocrystalline iron and a Fe₉₀Cu₁₀ alloy. The nanocrystalline material has a lower onset pressure and a larger pressure hysteresis than bulk iron. The non-random orientation of the nanocrystals is sensitive to shear stress in the high-pressure cell.

The feasibility of producing nano-sized WC and WC-Co powders by reduction and carburization of WO₃ and Co₃O₄ under hydrogen and methane atmospheres was evaluated. WC powders of 1 \\micron in particle size were obtained from WO₃ on reduction and carburization at 950°C. On the other hand, nano-sized WC-Co powders with a particle size of ∼ 100 nm could be produced by a two-step reduction process, using a blend of WO₃ and Co₃O₄. Critical control parameters in this process, such as the gas flow rate, temperature and treatment time, were found.

Cu–Mg–TiC bulk nanocrystalline composites were prepared by mechanical alloying, HIP process and hot-rolling. Tensile tests at constant crosshead speeds were carried out using a universal testing machine at several different temperatures under an argon atmosphere. The elongation drastically increased as the fraction of melt increased, and a maximum nominal elongation of about 200% was obtained at temperature where atomic fraction of liquid phase was about 0.5 and the strain rate sensitivity m was in the range of 0.8 to 1.0 for every composite. After the deformation, nanocrystalline structures with average grain sizes of about 30 nm were retained.

Nanocrystalline MgH₂/Me_xO_y composite powders were produced by high energy ball milling (Me_xO_y=Sc₂O₃, TiO₂, V₂O₅, Cr₂O₃, Mn₂O₃, Fe₃O₄, CuO, Al₂O₃, SiO₂). The hydrogen absorption and desorption kinetics were determined in view of a technical application. The addition of selected oxides lead to an enormous catalytic acceleration of hydrogen sorption compared to pure nanocrystalline hydrides. In absorption, the catalytic effect of TiO₂, V₂O₅, Cr₂O₃, Mn₂O₃, Fe₃O₄, and CuO is comparable. Concerning desorption, the composite material containing Fe₃O₄ shows the fastest kinetics followed by V₂O₅, Mn₂O₃, Cr₂O₃ and TiO₂. Only 0.2 mol% of the catalysts is sufficient to provide a fast sorption kinetics. Additionally, Mg absorbs hydrogen already at room temperature by the use of metal oxides as catalysts. Furthermore, it is demonstrated for the first time that MgH₂/Me_xO_y-powders release hydrogen at 200°C.

Mechanical alloying and mechanical grinding have been used to synthesize or treat AB, AB₂, and AB₅ type alloys and Mg-based hydrogen storage materials. A nanocrystalline or an amorphous structure was obtained after milling, depending on the system and its composition. The structure and the hydrogen storage properties of various nanocrystalline alloys were investigated. It was found that the activation and kinetics of hydrogen absorption/desorption were improved. However, a severe loss of storage capacity was observed in the AB, AB₂ and AB₅ systems, but not much in Mg-based systems. The storage capacity can be recovered by thermal annealing at elevated temperatures. A preliminary explanation is given for the change of absorption/desorption kinetics and storage capacity.

The hydrogenation of benzene in a fixed-bed type reactor operated at atmospheric pressure over amorphous Pt–Zr alloys with different platinum composition (Pt₂₀Zr₈₀, Pt₂₈Zr₇₂, Pt₃₂Zr₆₅) produced by a rapid solidification technique was investigated. Although the catalytic activity of the alloys was very low at virgin state, it gradually increased with increasing number of regeneration (oxidation and reduction)-reaction cycles. The activity approached a constant value after 5 or 6 regenerations. The hydrogenation activity of Pt–Zr alloy is higher than those of Pd–Zr and Ni–Zr alloys with the same composition. Furthermore, the turnover frequency of Pt–Zr defined as the reacting molecule per active site per second is also higher than those of Pd–Zr and Ni–Zr alloys. The hydrogenation order of the turnover frequency in the present study is almost the same as the order of the hydrogenation of olefins over the corresponding catalysts prepared from impregnation method. The characterization of the platinum catalyst prepared from the amorphous alloy revealed that the dispersion of platinum is higher than the impregnated catalysts, even though the platinum content of the former catalyst is much higher than that of latter. It is considered that this is a remarkable feature of the amorphous alloy catalyst.

We prepared gold-titania composite nanoparticles by the gas condensation method, and evaluated their physical and optical characteristics. The structure of obtained particles mainly showed that the gold particles, whose diameter ranged from 10 to 60 nm, were covered with spherical titania particles which were composite phases of anatase and rutile. The effect on the crystal structure and shapes of titania was not clearly observed by compounding with gold.

The present study has attempted to elucidate the formation mechanism of nanocrystalline (nc) Ni by hydrogen reduction of fine NiO powder in terms of related kinetics aspects. So, the kinetics and related mechanism of hydrogen reduction of NiO were investigated on the basis of structure modification of the NiO powder during reaction. The ball-milled NiO agglomerate powder having 20 nm in grain size and a log-normal pore size distribution was used for study. The non-isothermal reduction study showed that the nano-agglomerate NiO underwent a two-step reduction process which is presumably due to a chemical reaction at lower temperatures and a diffusion controlled process at higher temperatures. The activation energy for the nano-agglomerate NiO was 85.4 kJ/mol for lower temperatures and 105.1 kJ/mol for higher temperatures. The value for lower temperatures is consistent with that of as-received NiO of 85.6 kJ/mol. Such higher activation energy for higher temperatures can be attributed to the retardation of the reduction process by the change in the reduction mechanism from the chemical reaction to the diffusion process. Conclusively, the structure change during the reduction is believed to be responsible for the change in the reduction mechanism.

Shallow nanohole structures of uniform size were fabricated on p-type Si by anodizing Si at high applied voltages in a dilute HF solution. The dependence of surface morphology on anodization conditions was investigated and it was found that both HF concentration and ethanol concentration in the electrolyte influenced nanostructures. Nanohole sizes were changed 70–140 nm with increasing applied voltage. Fabricated nanohole structures were used as a substrate for fixing nanoparticles. Nanoparticles, whose sizes were comparable to nanohole sizes, were selectively trapped in nanoholes.

The surface reactivity of a powder depends on the chemical composition of its surface and on the presence of defects. In the case of nanosized powders, the surface reactivity is enhanced by the increased defect concentration on the surface. On the other hand, the gas sensing properties of a semiconductor material are strongly related to the surface reactivity. In this article, it is shown, by Fourier transform infrared spectrometry, that SnO₂ semiconductor nanoparticles are very sensitive toward CO and that the decrease of the particle size greatly enhances this sensitivity. Comparison is made between particles having an average size of 15 and 8 nm. Surface reactions at the origin of the CO detection mechanism are discussed as a function of the particle size.

Synthesis of Nb₂O₅ nanoparticles by mechanochemical processing has been investigated. The reactions 2NbCl₅+5Na₂CO₃→Nb₂O₅+10NaCl+5CO₂(g) and 2NbCl₅+5MgO→Nb₂O₅+5MgCl₂ led to the formation of Nb₂O₅ aggregates of 100–1000 nm in size. Addition of NaCl in the starting mixture of 2NbCl₅+5Na₂CO₃ prevented the aggregates forming. After heat treatment of the as-milled powder at 300°C, amorphous Nb₂O₅ having a high surface area of 196 m²/g was obtained. Heat treatment at 550°C resulted in a powder consisting of 10–100 nm single crystal Nb₂O₅ particles along with a network of ultrafine amorphous particles. Annealing at higher temperature led to the formation of the Na₂Nb₂O₁₁ phase.

Acicular α-FeOOH particles are formed through aging of ferric oxyhydroxide colloidal solution formed by the neutralization of FeCl₃ aqueous solution by NaOH . The effect of foreign ion addition to the colloidal solution on the formation and morphology of α-FeOOH particles has been investigated. The magnetic properties of Fe₃O₄ particles made from the obtained particles have also been investigated. The rate constant of the formation of α-FeOOH remarkably decreased, but the crystallite size of α-FeOOH particles increased with increasing the quantity of phosphate ion added even with small amounts. These results have been explained as follows: the phosphate ions are selectively adsorbed on the (a) plane of α-FeOOH, cover the (a) plane, and block the crystal growth of the (a) plane of the α-FeOOH . The quantities of the phosphate ion adsorbed on the b and c planes are relatively small. The complex ion of Fe(OH)₄⁻ is preferentially deposited on both (b) and (c) planes, and the crystal growth of (b) and (c) planes is greatly accelerated. The relationship between the morphology of the formed α-FeOOH particles and the quantity of phosphate ion added has been investigated. The asterisk type particles: α-FeOOH particles heterogeneously junctioned to α-Fe₂O₃ particles, were formed when a small amount of phosphate was added to the mother liquid. The α-FeOOH crystal epitaxially grew on the junction interface with the α-Fe₂O₃ crystal. In the case of the aging at the temperature as high as 80°C, the cross type junctioned particles were stably formed at pH below 12.0. The Fe₃O₄ particles with screw-like unique three-dimensional morphology were produced from the heterogeneously junctioned particles.

Metal or metal-oxide nanocrystals can be formed by the gas deposition method. Here we report a modified gas deposition method, by which nanocrystals of metal oxides are synthesized at first, and transformed into metal nanocrystals by heating the carrying gas. By controlling the heating temperature we can selectively deposit either dielectric films or conductive films in the same deposition process. Demonstrative experiments for transforming Al₂O₃ nanocrystals into Al nanocrystals are shown.

The sol–gel process was used to prepare nanoporous (typically micro and mesoporous) supported layers from silica and alumina sols. The control of aging time for silica gel layers, the investigations of new alumina hydrosols and the general use of removable templating mesophases are crucial parameters to produce layers exhibiting unconventional porosity. Quasidense microporous layers as well as silica layers with a mesoporosity up to 74% were obtained. All these described methods allow the easy preparation of supported crack-free layers with several potential applications.

Varistor material with 96 mol%ZnO+4 mol% dopants in oxide form of Bi, Co, Sb, B, Cu and Sn in nanostructured form has been synthesized using colloidal suspension and centrifugal separation method. The synthesized powder sample specimen was compacted into a pellet and sintered in air at 1073 K for 3 h to get a bulk density of 96%. In situ impedance spectroscopy (IS) studies have been carried out for the sintered nanostructured specimen at various temperatures in oxygen and nitrogen atmosphere. The impedance spectroscopy results below 423 K show that the specimen contains three activation energy regions which are attributed to (i) pure ZnO core grain (ii) aliovalent cation diffused layer around the pure ZnO core and (iii) grain boundary (GB). The specimen changes its conducting nature at and above 423 K . The a.c. conductivity was measured at fixed frequencies (10⁶ Hz and 10² Hz) as a function of temperature. The conductivity of bulk grain and grain boundary regions increases with the increase of temperature. Eventhough the conductivity value of bulk grain and grain boundary regions increases with temperature, the rate of change of conductivity in grain boundary region is larger than that in bulk grain. The GB conductivity becomes almost equal to that of bulk grains at high temperatures. The change of grain boundary conductivity with temperature is faster in N₂ atmosphere compared to the change in O₂ atmosphere. Due to this dominant grain boundary conductivity, the varistor action of the sample is lost at high temperatures in nitrogen atmosphere (air).

Processing of synthesising nanocrystalline ZrO₂(3Y) by hot-pressing was investigated. Large crack-like pores have been found in the sintered samples. The reason of this phenomenon was that the agglomerate in the powder had not been destroyed efficiently. This agglomerate sintered quickly and turned into aggregate when hot-pressing was applied. Large pores formed as a result of the interaction between the aggregate and the matrix. Since the applied pressure was limited because of the low stress tolerance of the graphite die, no plastic deformation could happen and those large pores could not be “squashed”. As a result, the densities of the samples sintered by hot-pressing were lower than by pressureless sintering at the similar sintering temperature. To overcome this shortage of hot-pressing, sinter forging was applied. Nano Y-TZP materials with grain size of about 85 nm and relative density of 99% could be obtained after sinter-forging at 1100°C for only 30 min.

In this study, piezoelectric quartz crystals are used as a gas sensor to investigate the detecting properties of titania nanoparticles for carbon monoxide and nitrogen dioxide gases. Titania nanoparticles are deposited on the surface of a piezoelectric quartz crystal by a gas condensation method. It is found that the amount of frequency change is dependent on gas concentrations at constant temperature. The amount of adsorbed carbon monoxide gas at 150°C hardly increases with increasing concentration, and the amount of adsorption is less than which at 200°C. The amount of adsorbed NO₂ also indicates similar results. The amount of adsorbed CO at 200°C is more than that of NO₂. The results are obtained from this study within one or two order difference in comparison with results in literature.

Nanocomposites Al₂O₃/SiC powders have been synthesized by a microwave-induced combustion process using aluminium nitrate, β-SiC (powders) and urea. The effect of preheating the solution before the combustion reaction and the effect of the characteristic of the SiC powders, in the crystallization of the alumina, by combustion reaction were evaluated. The preheating of the liquid solution at about 150°C before heating in the microwave oven guaranteed the homogeneity of the solution and modified the characteristics of the products. The mechanism of heating in the microwave oven was favorable to the uniform emission of gases and fast dissipation of heat, promoting formation of different foam and nanocomposites Al₂O₃/SiC powders with characteristics that were different from those produced conventionally. The direct conversion of liquid solution to the final solid product was assured, and the segregation commonly found in conventional mixing of powders (alumina-SiC) was avoided. The different characteristics of the SiC powders have strong influence in the flame time of the combustion reaction and consequently in the crystallization of the alumina and characteristics of the Al₂O₃/SiC nanocomposites powders. The SiC particles were not oxidized during the reaction and partial reaction with alumina did not occur; SiC particles remained as a distinct and separate phase.

The nanocrystalline samples of zirconium oxide doped with ruthenium oxide (RuO₂) have been synthesized from chlorides as precursors by chemical precipitation method. The as-prepared and annealed powder samples were studied by XRD, TEM and Impedance Spectroscopy. With 7.5 mol% of RuO₂, only a small percentage of ZrO₂ stabilizes in tetragonal form without stabilization in cubic form. With 9 mol% and above, ZrO₂ stabilizes in mixed phases having both tetragonal and cubic structure. On annealing, up to 1273 K the proportion of the cubic phase increases; however annealing at temperatures above 1273 K makes the sample to become monoclinic. Average grain size, as determined by Scherrer’s formula using X-ray linewidth, increases with increase of annealing temperature. The same trend is observed in TEM studies. TEM studies show the agglomeration of grains. X-ray diffraction for various concentrations of RuO₂ shows the presence of small amount of RuO₂ as impurity. This implies that Ru⁴⁺ goes to the interstitial position also in addition to its occupation of substitutional position of Zr⁴⁺ due to its smaller ionic radius (0.062 nm) compared to that of Zr⁴⁺ (0.084 nm). The impedance spectroscopy measurement shows that the conductivity decreases with an increase of grain size. The phase changes in the stabilization process with different concentrations and annealing temperatures, and growth of grain size in RuO₂ stabilized ZrO₂ will be presented. One of the salient points in this study is stabilization of ZrO₂ with RuO₂ alone without Y₂O₃ as reported in the literature.

Synthesis of nano sized magnetite particles has been carried out following a polymer matrix mediated process. The synthesis route, being akin to bio mineralization, yields elongated magnetite particles of uniform size and morphology. The produced particles were oriented perpendicularly with respect to polymeric tubules and showed a tendency of array formation as happens in magnetotactic bacteria.

The effect of sulfate ions on the nucleation of anatase nanocrystals during the hydrolysis of TiCl₄ solution was investigated. The resulting titania powders were characterized by DSC, FT-Raman spectroscopy, X-ray diffraction and UV-Vis spectroscopy. In the presence of sulfate ions as low as one twentieth to Ti⁴⁺, the precipitate prepared at room temperature contained 17.8% anatase nanocrystalline titania. The promoting effect of sulfate ions on the nucleation of anatase at higher temperature was more significant. For example, at 343 K, anatase powder was synthesized in the presence of a small amount of sulfate ions, while mixed-phase (the fraction of rutile was 0.634) powder was prepared in the absence of sulfate ions. Small amount of sulfate ions also modified the pore size distribution effectively and stabilized anatase to convert into rutile during calcining process, and powders with narrow pore size distribution (average pore diameter 3.8 nm) and mesoporous nature were prepared in the presence of sulfate ions.

We report in this communication the growth mechanism of multiwall carbon nanotubes during hydrothermal treatment of carbon soot at 800°C and 100 MPa. High-resolution electron microscopy (HRTEM) study reveals the formation of multiwall carbon nanotubes made of a hollow core enclosed in well-ordered concentric graphitic layers after hydrothermal treatment.

When the vacuum evaporation method was used to deposit metals, such as Au, Ag, or Cu, on a substrate covered with aligned carbon nanotubes which were fabricated by the microwave-plasma-enhanced chemical vapor deposition method, we found that nanoscale single crystals of the metals with well-defined facets were grown on the surfaces of the carbon nanotubes. Nanocrystals ranging from approximately 50 to 500 nm in diameter were obtained when the deposition time ranged from 1 to 60 min. Whiskerlike crystals with high aspect ratios could also be found frequently in these nanocrystals, which can be applied in field-emission display devices.

This paper describes the preparation of manganese(II)-phenanthroline complex functionalized MCM-41 (MCM from Mobil Crystalline Material). The as-synthesized organic group functional MCM-41 and metal-ion-complex functionalized MCM-41 were extensively characterized by powder X-ray diffraction (XRD), TEM, solid-state ²⁹Si and ¹³C NMR and nitrogen adsorption/desorption isotherms at 77 K, as well as Mn(II) electron spin resonance (ESR) spectroscopy at room temperature. These results indicated that 3-methacryloyloxypropyl group well dispersed in the mesopore of MCM-41, which attached to the silicate walls via Si–O–Si bonds. The mesopore of MCM-41 was tailored by the grafted organic group and the complex, while the coordination state of Mn(II) was unaffected by the process of grafting.

Conversion coatings for magnesium have traditionally been based on immersion treatment in the solution containing hexavalent chromium compounds. The excellent performance of these coatings as paint bases has been established in a broad range of application. However, the replaceable surface treatment has been strongly emphasized by present environmental needs to eliminate hexavalent chromium. The development of chemical conversion coatings in permanganate bath on magnesium alloys has been studied using mass gain measurements, measurements of corrosion potential, X-ray diffraction, X-ray photoelectron spectroscopy, glow discharge spectroscopy and microscopic examination. Chemical conversion coatings obtained from HF addition in bath consisted of the film that was composed mainly of manganese oxides and magnesium fluoride. It was found that these films formed amorphous composite coating. Corrosion resistances of permanganate chemical conversion coatings were comparable with that for chromating.

A NiAl/Cr(Mo) alloy modified with Hf was fabricated by a vacuum induction melted and drop cast. The alloy was mainly composed of NiAl matrix, Cr(Mo) and Hf-rich phase distributed between NiAl and Cr(Mo) phase boundary and phase transition was also studied. Then it was hot isostically pressed (HIPed) at 1523 K, 200 MPa for 4.5 h. Its high temperature compressive behavior was studied and the deformation features of the alloy could be adequately described by standard power law which is usually used to characterize creep behavior for various metallic material. The stress exponent n as well as the activation energy Q was calculated by fitting the experimental data to the power-law and temperature-compensated power law equation. The possible strengthening mechanism was discussed.

A small solenoid coil was equipped with a scanning tunneling electron microscope (STM) which has an iron needle core tip as the probe. Mutual inductance of the coil with a thin gold film on a mica plate was investigated in order to profile electrical conductivity. It is important to maintain a small gap between the tip apex and the sample surface. The gap is related to the distribution of electron scattering by induced current. The conductivity mapping was carried out on a large terrace. If the tip is over the step edge, the output signal from the probe coil becomes unstable. Atomically smooth surface of the specimen was in need for the electrical conductivity mapping. The electrical conductivity on the terrace was almost constant. The mutual inductance values with the gold film at various temperatures were investigated by cooling it below 250 K . The mutual inductance is inversely proportional to the temperature to the one-third power, T^1⁄3. The relationship between the mutual inductance and the temperature was explained by the electromagnetic theory.

The microstructure evolution and grain growth behavior, of gas-atomized Al₈₈Ni₉Ce₂Fe₁ nanocomposite materials, were studied by means of the X-ray diffraction, transmission electron microscopy, and differential scanning calorimetry. The results showed that the microstructure of as-atomized powders are affected by the degree of solute supersaturation in matrix, the microstrain contained, the size, distribution and volume fraction of nanophase precipitated, and these are essentially sensitive to the powder particle sizes. Moreover, there appeared to be a two-stage phase transformation during thermal crystallization, i.e., the precipitation of α-Al and the growth of Al₃(Ni, Fe) nanophase in preference to Al₁₁Ce₃ nanophase, respectively. Finally, the grain growth kinetics of α-Al at temperatures of 250 to 300°C, revealed that the volume-diffusion grain growth mechanism with an activation energy of 1.3 eV is controlling in the nanophase materials.

In order to improve surface hardness, surface modification of die casting aluminum alloys was carried out by applying a new method called Electrical Discharge Alloying with ultrasonic vibrations, in which a conventional electrical discharge machine can be used. The layer deposited on the substrate consists of TiC and Ti₂AlC formed by the reaction between titanium contained in the green compact electrode, the aluminum in the electrode and the substrate and also the carbon decomposed from a working fluid. When the ultrasonic vibration is applied to the substrate, especially at a frequency of 94.2 kHz, the discharged craters are bigger than those obtained when alloying without ultrasonic vibration. Also, for single discharge, the amount of deposited Ti and C increases near the edge of the crater. When the ultrasonic vibration is applied to the electrode for single discharge, many convex-shaped craters are generated on the substrate. The cross-sectional hardness of the alloyed area increases with increase in the concentration of carbon and titanium. The secondary elaboration using the cast Ti–36 mass%Al solid electrode is effective for the improvement of surface roughness of the modified layer. When the green compact electrode is used in the primary Electrical Discharge Alloying and the cast solid electrode is used in the secondary elaboration, surface roughness is not so different between the process with and that without ultrasonic vibration. However, the modified layer’s thickness obtained with ultrasonic vibration is thicker than that without ultrasonic vibration.

A study has been made on the effects of Al, Si and Mo on anodic polarization characteristics of Fe–10Cr alloys in 0.05–1.0 kmol·m⁻³ H₂SO₄ and NaCl solutions. Potential decay curves have also been measured in order to evaluate the stability of the passive films formed on Fe–10Cr alloys containing Al, Si and Mo. The analysis of the chemical composition of the passive film has been carried out by AES and XPS . The addition of Mo was very useful to decrease the critical passivation current density of Fe–10Cr alloys which contained Al and Si. The passive current density decreased with addition of 2 mass%Mo to Fe–10Cr–3Si alloy, while it increased in Fe–10Cr–3Al alloy. Pitting potential was moved toward the noble direction by the addition of Mo, except for the case of Fe–10Cr–3Al–3Si–2Mo alloy. It seems that Laves phase (Fe₂Mo) was precipitated in this alloy, and Mo-depleted zone was formed locally. The potential decay time of the passive film was increased by the addition of Al, Si and Mo to Fe–10Cr alloys. Two step potential decay curves were obtained in Si-containing alloys, and this behavior of the decay curve was shown more clearly in Fe–10Cr–3Si–2Mo alloy. The results of AES and XPS analysis indicated that Cr and Al concentrated in the passive film of Fe–10Cr–Al alloy. In the case of Fe–10Cr–Si alloy, Si concentrated in the surface region of the film and Cr in the internal region of the film. The confirmation was not possible for the enrichment of Mo in the passive film.

Reaction diffusion in Ti/Al and Ti–5 mol%O/Al diffusion couples has been investigated by electron-probe microanalysis. Only one intermetallic compound TiAl₃ was observed as an intermediate layer in the temperature range from 773 to 903 K in both the diffusion couples. Si, which is an impurity element in aluminum, was concentrated in TiAl₃. The growth of the intermediate layer turned out to be diffusion limited; the activation energy was estimated to be 237±15 kJ·mol⁻¹ and the temperature dependence of square of the rate constant, k², for layer growth can be described with the following equation in the temperature range from 773 to 903 K in the Ti/Al diffusion couples. k_TiAl_3^2=3.5exp[-(237±15) kJ·mol^-1/RT]m^2s^-1In the case of the oxygen doped diffusion couples, the layer growth of TiAl₃ was significantly suppressed and the activation energy was 263±7 kJ·mol⁻¹ for temperatures from 773 to 873 K . The suppression is explained by aluminum oxide formed between aluminum and TiAl₃. The Kirkendall marker shifted toward the aluminum side, which suggests that diffusion of Al is faster than that of Ti in the intermediate layer.

Bending and uniaxial tensile tests for Al-base amorphous ribbons were performed at room temperature for a number of compositions. Al-nanocrystal precipitation was observed within vein protrusions on fracture surfaces and along crack propagation paths, as well as within shear bands resulting from bending. A composition dependence of crystallization upon bending was also observed. Effects of local adiabatic heating was observed on fracture surface.

To study the relationship between hydrogen pressure and holding time for hydrogenation at 680 K, hydriding combustion synthesis of Mg₂NiH₄ were conducted at different heat patterns and hydrogen pressures. The obtained results were summarized in the form of an operating map, which visualized clearly the possible operating conditions for industrializing the process. Pure Mg₂NiH₄ can be synthesized at a high pressure of hydrogen and a short holding time at 680 K . At a lower pressure of hydrogen such as 0.5 MPa, 7.2 ks of a holding time at 680 K is enough to obtain 100% pure Mg₂NiH₄ as a product. The pressure composition isotherm of Mg₂NiH₄, obtained at 623 K, showed a hydrogen capacity very near the theoretical value of 3.6%, together with a reasonable plateau. The specific surface areas of several samples suggested that the product is a porous material.

Behaviors of carbon nanotubes (CNTs) in H₂ atmosphere or CH₄+H₂ mixtures, transforming mechanism of CNTs to diamond embryo were investigated. Experimental results show that CNTs could be etched by H₂ and transformed to carbon nanoparticles under conditions of hot-filament chemical vapor deposition. The mechanism can be concluded that CNTs were broken down by H₂, transformed to carbon nanoparticles or shorter CNTs; on surface of carbon nanoparticles or end caps of CNTs diamond crystals were nucleated and grown up.

Thin coherent as well as thicker semi-coherent plate-like precipitates of the same type, i.e. TiCu, were generated by different heat treatments of melt-spun Ti₅₀Ni₂₅Cu₂₅ ribbons. Their influence on the martensitic transformation behavior was investigated by means of differential scanning calorimetry (DSC) and transmission electron microscopy (TEM). It was found that thin coherent plate-like precipitates lower the temperatures of both martensitic and its reverse transformation and broaden the peak form of the DSC signal. On the other hand, thicker semi-coherent plates act to raise both transformation temperatures. The role of coherency stresses, arrangements of misfit dislocations, and compositional aspects are discussed.

Considering the effect of the friction force due to volume diffusion of a solute on the driving force, a new kinetic model has been proposed for diffusion induced recrystallization (DIR) in the A(B) system in which solute B atoms diffuse into a pure A metal or a binary A-B alloy. The energy balance model [M. Kajihara and W. Gust: Scr. Mater. 38 (1998) 1621] has been combined with the columnar geometry and boundary diffusion model [C. Li and M. Hillert: Acta Metall. 29 (1981) 1949] and the extended model [Y. Kawanami et al.: ISIJ Int. 37 (1997) 921] in order to describe mathematically the growth rate of the fine grain region (DIR region) formed by DIR as a function of the reaction time. DIR in the Ni(Cu) system was experimentally observed by Kawanami et al. [Y. Kawanami et al.: Mater. Trans., JIM 39 (1998) 218] at 923 and 1023 K . The new model has been utilized to analyze their observations theoretically. According to the observations, the migration rate v of the moving boundary gradually decreases with increasing reaction time. However, the decrease in the migration rate v is negligible during a small time interval of Δt=1 s at the experimental reaction times. Thus, the value of v was assumed to be constant at each time step with Δt=1 s in order to simplify the analysis. Using the mobility M of the moving boundary as the fitting parameter, the thickness of the DIR region was calculated as a function of the reaction time by a numerical technique. The calculation gives values of M=3.73×10⁻¹⁷ and 1.51×10⁻¹⁵ m⁴/Js at 923 and 1023 K, respectively, and thus M₀=1.03 m⁴/Js and Q_M=290 kJ/mol for M=M₀exp(−Q_M⁄RT). The temperature dependence of the mobility indicates that the grain boundary migration may be governed by the solute drag effect for which the volume diffusion of the solute along the moving direction in the untransformed matrix ahead of the moving boundary has the most important role.