Chemical mixing enthalpy (ΔH^chem) and mismatch entropy normalized by Boltzmann constant (S_σ⁄k_B) corresponding to the three empirical rules for the achievement of high amorphous-forming ability (AFA) were calculated with thermodynamical functions for the gross number of 6450 alloys in 351 ternary amorphous systems. The ternary amorphous alloys have ΔH^chem of −86 to 25 kJ/mol and S_σ⁄k_B of 1.0×10⁻³ to 5.7. The average values of ΔH^chem and S_σ⁄k_B are calculated to be −33 kJ/mol and 0.33, respectively. The 30 alloys in 9 ternary amorphous systems including 10 alloys in Ag–Cu–Fe system have positive values of ΔH^chem. Most of the ternary amorphous alloys have the values of ΔH^chem and S_σ⁄k_B inside a trapezoid region in ΔH^chem−log(S_σ⁄k_B) chart except mainly for the H- and the C-containing alloys, Si–W–Zr system and the 32 alloys having positive values of ΔH^chem. The analysis of AFA was carried out for typical five ternary amorphous systems. The following four results are derived. 1) Al–La–Ni and B–Fe–Zr alloys have high AFA in accordance with the concept of the three empirical rules. 2) The further multiplication of alloy components causes an increase in the AFA of Al–B–Fe alloys. 3) Thermodynamical factors represented by melting temperature and viscosity at the melting temperature are required for evaluation of AFA for Mg- and Pd-based amorphous alloys. 4) A tendency for log(S_σ⁄k_B) to increase with decreasing ΔH^chem is recognized in each alloys system, implying the stabilization of an amorphous phase against solid solution and intermediate phase.

View full abstract

  This paper has been retracted by the Editorial Committee with the first author’s agreement due to substantial overlap with the following paper previously received in another journal.

Mater. Sci. Eng. A 312 (2001) 136-144
Structural study of Amorphous Fe₇₀M₁₀B₂₀ (M=Zr, Nb and Cr) alloys by X-ray diffraction
E. Matsubara, S. Sato, M. Imafuku, T. Nakamura, H. Koshiba, A. Inoue, Y. Waseda (Received on May 8, 2000)

  And thus, it shall not be regarded as an original paper.

  Although this paper was submitted to the Special Issue on Bulk Amorphous, Nano-Crystalline and Nano-Quasicrystalline Alloys by the first author without understanding the bylaws of Materials Transactions, JIM, it was not an appropriate deed.

  The first author acknowledges the substantial overlap and his apologies have been accepted accordingly. The notice has been issued to all the authors to pay more careful attention in submitting papers.

View full abstract

A change of atomic structure of amorphous La₅₅Ni₂₅A1₂₀ alloy in the course of annealing in the “supercooled liquid region” has been investigated by in-situ electron diffraction and high-resolution electron microscopy (HREM) using a specimen heating-stage in TEM. For recording electron intensities an imaging plate was used. In order to remove the inelastically scattered electrons an energy filter was used for recording electron diffraction patterns. No appreciable change was found in the in-situ HREM observation across the glass transition temperature on heating. However, atomic pair distribution function (PDF) analysis revealed a clear structural change in the “supercooled liquid region”. The observed structural change is concerned with the local development of strong atomic correlations especially for La–La and Al–Ni and also with the increase of structural volume. The atomic configurational change is closely related to the phase separation in the later crystallization stage.

View full abstract

Structural change in the supercooled-liquid region of Fe₅₈Co₇Ni₇Zr₈B₂₀ glassy alloy has been examined by Mössbauer spectroscopy. Analyses of the Mössbauer spectra showed that the hyperfine field distribution patterns in quenched states from supercooled-liquid state differ from that in the as-melt-spun state. As the quenching temperature increases, the hyperfine field distribution tends to show several peaks some of whose hyperfine field values seem to correspond well to those in crystalline Fe, Fe₃B, Fe₂B and Fe₂Zr, although these distribution peaks are quite broad. These results as well as the initial crystallization behavior indicate that the thermal treatment in the supercooled-liquid region causes a compositional short-range order. It is, however, supposed that the chemical bond strengths are competitive among the Fe₃B-, Fe₂B-, FeZr₂- and Fe–Zr–B-type compositional short-range order, and their growth frustrate each other in the supercooled-liquid region. This competitive frustration should be one of the reasons for the thermal stability against crystallization that occurs in the glassy alloys with a large supercooled-liquid region.

View full abstract

Glass formation and glass forming ability of six La-based La–Al–Ni–Cu–(Co) alloys have been studied systematically by Bridgman solidification and thermal analytical measurement. With increased Cu content the critical growth velocity for glass formation in La₅₅Al₂₅Cu_20−xNi_x (x=0–20) alloys shows a steep minimum at 10 at%Cu, however, replacement of Ni by Co leads to a further decrease in the critical cooling rate for the La₅₅Al₂₅Ni₅Cu₁₀Co₅ alloy. The thermodymics of these alloys was studied in the undercooled liquid and crystalline state by means of temperature-modulated DSC. Reduced glass transition temperature T_rg given by T_g⁄T_l (T_g the glass transition temperature and T_l the offset melting point) is found to show a stronger correlation with critical cooling rate for glass formation than that given by T_g⁄T_m (T_m the onset melting point). The glass forming ability of these alloys is discussed in terms of microstructure selection and driving force for crystallization.

View full abstract

The glass forming ability and thermal stability of the quenched amorphous phase are evaluated in Al₈₇Ni₇Gd_6−x(Ca, Sr, Ba)_x alloys in terms of large atomic size effects of Alkali metals. Thermal analysis indicates that the thermal stability of the quenched amorphous phase decreases with the addition of Ca and increases with the addition of up to 2 at%Sr and Ba, as seen in the change of primary crystallization peak temperature and activation energy for primary crystallization. However, the enhanced thermal stability of the quenched amorphous phase with the addition of Sr and Ba does not lead to easier glass formation. The critical thickness of the quenched ribbons with completely amorphous phase is found to decrease from 300 to 70 μm as the content of Alkali elements is increased at the expense of Gd. These results reveal the different effects of atomic size on the glass forming ability and the thermal stability of the quenched amorphous phase against crystallization.

View full abstract

Zr₅₅Al₁₅Ni₁₀Cu₂₀ bulk metallic glass is formed using low purity materials at a low vacuum with a small amount of yttrium addition. It is found that the glass forming ability, crystallization and melting process of the Zr₅₅Al₁₅Ni₁₀Cu₂₀ alloy are modified with yttrium addition, while the mechanical and elastic properties, such as hardness and Young’s Modulus, are not obviously changed. The positive effect of yttrium addition on the glassy formation of the Zr₅₅Al₁₅Ni₁₀Cu₂₀ alloy is clarified.

View full abstract

The effect of iron on the glass transition, the viscosity of the supercooled liquid and the crystallization behaviour of (Zr_0.650Al_0.075Cu_0.175Ni_0.100)_100−xFe_x and (Zr_0.55Al_0.10Cu_0.30Ni_0.05)_100−xFe_x metallic glasses (0≤x≤15) was investigated using parallel plate rheometry, differential scanning calorimetry and X-ray diffraction. Iron additions shift the glass transition as well as the crystallization to higher temperatures, and increase the viscosity of the supercooled liquid. Increasing iron content reduces the extension of the supercooled liquid region. This mainly results from a stronger composition dependence of the glass transition than the crystallization temperature. The results are discussed with respect to the kinetics and thermodynamics of the supercooled liquid suggesting that changes in the atomic arrangement and in the free volume are the microscopic origin for the differences in the viscosity and the thermal stability of the alloys. All investigated alloys are rather strong glasses as expressed by a relatively high melt viscosity, a low Vogel-Fulcher Tammann (VFT) temperature and a rather high fragility parameter, but in general (Zr_0.55Al_0.10Cu_0.30Ni_0.05)_100−xFe_x alloys are slightly stronger glasses than (Zr_0.650Al_0.075Cu_0.175Ni_0.100)_100−xFe_x alloys.

View full abstract

Zr/Nb-based bulk metallic glasses (BMGs) with excellent glass forming ability (GFA) and high thermal stability were obtained by water quenching method. The GFA of the alloys is sensitive to the Fe addition, and the highest GFA is achieved at 8 at% of Fe addition. It is found that ΔT defined by the difference between the onset temperature of the first crystallization event T_x and the glass transition temperature T_g (ΔT=T_x−T_g) is more effective than T_rg (T_rg=T_g⁄T_m, T_m, melting temperature) to reflect the GFA of the Zr/Nb-based BMGs. The elastic properties of the alloys are investigated by ultrasonic measurements.

View full abstract

Zr–Ti–Cu–Ni–Be–Fe bulk metallic glasses and metallic glassy matrix composites were produced by water quenching method. The effect of Fe addition on glass forming ability, hardness, magnetic susceptibility and thermal stability of the alloys was investigated. It was found that the glass forming ability and the properties of the alloys were sensitive to Fe content. A single amorphous phase was obtained in the alloys up to 8 at%Fe addition. These bulk metallic glasses exhibited high thermal stability, wide super-cooled liquid region, and anomalous magnetic susceptibility χ versus temperature in the supercooled liquid region. Metallic glassy composites consisting of nanocrystalline FeZr₂ particles were obtained in the alloys with more than 10 at%Fe addition.

View full abstract

We measured the thermal expansion and specific volume of the Pd₄₀Cu₃₀Ni₁₀P₂₀ alloy by thermomechanical analyzer (TMA), fluid (H₂O) displacement and volume dilatometric methods. At room temperature (293 K), the specific volumes of the Pd₄₀Cu₃₀Ni₁₀P₂₀ alloy in the relaxed glassy and crystalline states are 0.1070 and 0.1065 m³/Mg, respectively. The average linear coefficients of thermal expansion are 17.0×10⁻⁶ K⁻¹ at 323 to 523 K in the relaxed glassy state (α_relax), 14.2×10⁻⁶ K⁻¹ at 323–673 K in the crystalline state (α_cry) and 39.9×10⁻⁶ K⁻¹ at 873–1223 K in the liquid (α_liq). Based on these results, the specific volume of the Pd₄₀Cu₃₀Ni₁₀P₂₀ alloy as a function of temperature, V(T), is estimated. The extrapolation line of the specific volume for the equilibrium liquid is smoothly connected to the specific volume for the supercooled liquid. On the other hand, the intersection point of the specific volume between the extrapolation of equilibrium liquid and crystal is evaluated at 527 K. In the assumption that the temperature of the intersection point corresponds to the entropic stability limit for the undercooled liquid, it is much lower than that of previous thermodynamic study.

View full abstract

Bulk amorphous (Mg_1−yAl_y)₆₀Cu₃₀Y₁₀ alloys were prepared using a relatively simple technique of rapid cooling of the melt in a copper wedge mould. The temperature vs. time was recorded during the cooling and solidification process of the melt and compared with a spacial and temporal numerical simulation of that process. It is concluded that good thermal contact is maintained between the amorphous part of the solidified sample and the mould, while a rather poor contact develops between the crystalline part of the sample and the mould, probably due to the appearance of a narrow gap at the crystal-mould interface during crystallisation. The maximum amorphous layer thickness decreases from ∼3 mm to zero when the Al content increases in the range from 0 to about y=10%. The evolution of the microstructure of the initially amorphous phase was examined by x-ray diffraction (XRD) and differential scanning calorimetry (DSC) for different alloy compositions and annealing temperatures. On annealing into the supercooled liquid state (441 K), specimens with no Al content remain basically amorphous while nanoparticles are formed and remain stable also at higher temperatures in specimens containing a few percent Al. The alloy with no Al crystallises apparently without the formation of nanoparticles. The critical cooling rate for the formation of an amorphous Mg₆₀Cu₃₀Y₁₀ specimen was determined experimentally by a combination of DSC data and temperature vs. time measurements to be 60–150 K/s, in agreement with estimates from the literature. The Vickers hardness (H_V) of the amorphous material for y=2% is higher (∼360 kg/mm²) than for y=0 (∼290 kg/mm²). On crystallisation the hardness of the latter material increases to the 400 kg/mm² level while the hardness of the former does not change.

View full abstract

Compression tests have been performed for bulk metallic glasses of La₅₅Al₂₅Ni₂₀ (L-glass) and Zr₄₁Ti₁₅Ni₁₂Cu₁₁Be₂₁ (Z-Glass) at temperatures from 4.2 K up to above the glass transition temperature. Shear fracture stresses are scattered between 0.75 GPa and 1.2 GPa in the L-glass, and between 1.3 GPa and 2.2 GPa in the Z-glass with a tendency of decrease towards both higher and lower temperatures from 77 K in L-glass and from 300 K in Z-glass. The normalized shear fracture stresses with respect to the Young’s modulus are around 0.02 in both glasses. The micro-plastic strain measured with the strain gauge for the L-glass is about 0.2% at the shear fracture stress at room temperature, decreases with decreasing temperature and is of the order of 0.01% below 77 K. The mechanism that determines the shear fracture stress is discussed.

View full abstract

A nanocrystalline (NC) bulk glass Zr₅₅Al₁₀Cu₃₀Ni₅ (at%) which contains nano-scale crystals embedded uniformly in a glassy matrix has both high tensile strength of 1.7 GPa and high ductility. It is therefore expected to find practical application in machines and structures. The fatigue crack propagation behavior of the NC bulk glass was examined. The threshold stress intensity factor range ΔK_th was about 0.9 MPa\sqrtm, and the fatigue fracture toughness K_fc was about 13 MPa\sqrtm. Fatigue crack propagation rate da⁄dn was approximately proportional to ΔK². No significant difference in da⁄dn between the NC glassy alloy and single phase amorphous alloys in literature was seen when they are compared on the basis of ΔK divided by Young’s modulus E. The da⁄dn values were nearly the same as those for steels with the same tensile strength level. The da⁄dn values were larger than those for low strength steels, Al alloys and Ti alloys in the low da⁄dn range. When compared with the effective stress intensity factor range ΔK_eff divided by E, the da⁄dn values for crystalline alloys were almost coincident with those under different stress ratios for the NC glassy alloy. This indicates that the da⁄dn values for the NC glassy alloy can be estimated on the basis of the da⁄dn data for crystalline alloys.

View full abstract

Bulk glassy Zr₅₅Al₁₀Ni₅Cu₃₀ composites containing ZrC particles up to 17.5 vol% were formed by the in-situ reaction between the Zr metal and graphite powder during melting-process. The glassy composites had a cylindrical form of 2 mm in diameter by a copper mold casting process. The in-situ reaction was effective for the homogeneous dispersion of ZrC particles and for the achievement of excellent wettability between pure glass and ZrC particles. Mechanical properties, particularly fracture strength and plastic elongation, were significantly improved in comparison with those of the pure glass as well as the similar bulk glassy composite formed by addition of ZrC particles. The present study has demonstrated that the in-situ reaction method is useful for the synthesis of the bulk glassy composites consisting of glassy matrix and dispersed reinforce particles with excellent mechanical properties.

View full abstract

Magnesium-based bulk amorphous alloys with a diameter of 7 mm for Mg₆₅Y₁₀Cu₁₅Ag₅Pd₅ were produced by a metallic mold casting method. The glass transition temperature (T_g), crystallization temperature (T_x), the temperature interval of the supercooled liquid region (ΔT_x=T_x−T_g) and the melting temperature (T_m) are measured to be 437, 472, 35 and 707 K, respectively, for the Mg₆₅Y₁₀Cu₁₅Ag₅Pd₅ alloy. The simultaneous addition of Ag and Pd to Mg–Y–Cu system causes a steep decrease in T_m and an increase in the reduced glass transition temperature (T_g⁄T_m), leading to an increase in the glass-forming ability. The Mg–Y–Cu–(Ag, Pd) bulk amorphous alloys have a high compressive fracture strength of 770 MPa. The simultaneous addition of Ag and Pd is useful for the improvements of glass-forming ability and fracture strength.

View full abstract

A rotating disk casting has been developed as a new method of producing an amorphous alloy wire. The rotating disk, which is made of copper, has a semicircular groove on its upside surface. The amorphous alloy wires with large diameters from 0.5 to 1.5 mm in the Zr–Al–Ni–Cu alloy system with high glass-forming ability have been prepared by the new method. The cross-section of the wires is nearly circular, because the melt is quenched to an amorphous state at the contact with an inner surface in the groove of the rotating disk. The cast amorphous alloy wires exhibit the same thermal stability and mechanical properties as those for the melt-spun amorphous alloy ribbon with the same composition. The formation of the amorphous alloy wires with large diameters is attributed to the high glass-forming ability of the alloy. In any event, the success of forming amorphous alloy wires by the present casting method is important for the future extension of application fields of amorphous alloys.

View full abstract

DSC curves show that the crystallization reactions change from a single exothermic peak to two exothermic stages by adding Ti up to 5 at% to Zr–Ni–Cu–Al amorphous alloys and bulk amorphous composites containing nanocrystalline particles were produced by annealing cast Zr_70−x−yTi_xNi₁₀Cu₂₀Al_y (x=5–7.5 and y=10–15 at%) amorphous alloys. The microstructure after the precipitation of a primary crystalline phase consists of the crystals less than 10 nm in size embedded in the amorphous matrix. Both the compressive strength and plastic strain increased significantly with increasing volume fraction of nanocrystals, and the maximum plastic strain was obtained in the early stage of nanocrystallization. High-resolution electron microscopy showed that the bulk nanocrystalline Zr₅₃Ti₅Ni₁₀Cu₂₀Al₁₂ alloy with the maximum plastic strain includes nanocrystals with a much smaller grain size of about 2.5 nm.

View full abstract

A large supercooled liquid region over 50 K before crystallization were obtained in amorphous Fe–(Al, Ga)–(P, C, B, Si), Fe–(Cr, Mo, Nb)–(Al, Ga)–(P, C, B) and (Fe, Co, Ni)–(Zr, Hf)–M–B (M = Ti, Hf, V, Nb, Ta, Cr, Mo, W) systems and their bulk glassy alloys were produced in a thickness range below 2 mm for the Fe–(Al, Ga)–(P, C, B, Si) system and 6 mm for the Fe–Co–(Zr, Nb, Ta)–(Mo, W)–B system by copper mold casting. The ring-shape glassy Fe–(Al, Ga)–(P, C, B, Si) alloys produced by copper mold casting exhibit much better soft magnetic properties than the ring-shape alloy made from the melt-spun ribbon as a result of the formation of a unique domain structure. The bulk Fe–(Al, Ga)–(P, C, B, Si) alloys were also produced by consolidating the amorphous powders using an electric-pulse-sintering method. The large elongation in the supercooled liquid region enables production of the bulk samples with relatively higher density and better soft magnetic properties than those of the sintered Fe–Si–B bulk amorphous sample. The bulk glassy (Fe, Co, Ni)₇₀M₁₀B₂₀ (M = Zr, Hf) systems exhibit large supercooled liquid regions of 72 K for M = Zr, and of 82 K for M = Hf. The replacement of Zr or Hf by 2 at%Nb or Ta causes a further increase in the supercooled liquid region. The good combination of high glass-forming ability and good soft magnetic properties indicates the possibility of future development as a new bulk glassy magnetic material.

View full abstract

The effect of the addition of the transitional metal Mo, on the glass-forming ability, mechanical strength and magnetic properties of Fe_78−xMo_xGa₂P₁₂C₄B₄ glassy alloys was investigated. The addition of 4 to 6 at%Mo was found to be effective for the extension of the supercooled liquid region (ΔT_x) defined by the difference between the glass transition temperature (T_g) and crystallization temperature (T_x). The ΔT_x value is 30 K for the Fe₇₈Ga₂P₁₂C₄B₄ alloy and increases to 60 K for the 4 at%Mo alloy. These glassy ribbon alloys exhibit good soft magnetic properties of 1.13 to 1.34 T for saturation magnetization and 2.2 to 3.8 A/m for coercive force. Based on the large ΔT_x values, we tried to prepare bulk glassy Fe₇₄Mo₄Ga₂P₁₂C₄B₄ rods in a cylindrical form with different diameters. The glassy single phase was obtained in the diameter range of 1.0 to 2.0 mm. These bulk glassy alloys also exhibit high saturation magnetizations of about 1.20 T as well as a high compressive strength of 2685 MPa.

View full abstract

The thermal stability, crystallized structure and magnetic properties of melt-spun Fe_70−xCo₁₀Pr_xB₂₀ (0≤x≤6) amorphous alloys have been investigated. The glass transition followed by a supercooled liquid region (ΔT_x) was observed in the composition range of 2 to 5 at%Pr and the value of the ΔT_x is in the range of 23 to 43 K. The good hard magnetic properties were obtained for the alloys containing 2.5 to 5 at%Pr subjected to an optimum heat treatment. The ΔT_x and reduced glass transition temperature (T_g⁄T_m) of the 3.5 at%Pr alloy are 43 K and 0.57, respectively. The crystallized structure consists of (Fe, Co)₃B, α-(Fe, Co) and Pr₂(Fe, Co)₄B phases and their average grain sizes after annealing for 420 s at 863 K are about 25 nm. The remanence (B_r), coercivity (_iH_c) and maximum energy product (BH)_max of the optimally annealed the 3.5 at%Pr alloy are 1.36 T, 211 kA/m and 107 kJ/m³, respectively. The good hard magnetic properties in the crystallized state of the Fe–Co–Pr–B amorphous alloys with high glass-forming ability indicates the possibility of future fabrication of a bulk hard magnetic material by the simple process of the formation of a bulk amorphous alloy followed by crystallization.

View full abstract

It was found that an amorphous Mg₆₂Ni₃₃Ca₅ alloy absorbs a large amount of hydrogen at 323 K and the hydrogen content is much larger than that of the corresponding crystalline alloy. The maximum absorption concentration of hydrogen at 323 K is 2.3 mass% in the amorphous phase and 1.3 mass% in the crystalline state. The Mg-based amorphous alloy with 2.3 mass%H₂ crystallizes through the process of Am → Am′+Mg₂Ni → Mg₂Ni+Mg₂Ca+MgNi₂+Mg₂NiH₄. The crystallization process is different from that (Am → Mg₂Ni+Mg₂Ca+MgNi₂) of the as-quenched amorphous phase. The onset temperature and the completed temperature for crystallization is 453 and 532 K, respectively, for the as-quenched amorphous alloy and 475 and 572 K, respectively, for the amorphous phase containing 2.3 to 3.0 mass%H₂. The absorption of hydrogen causes a significant increase in the thermal stability of the amorphous phase, presumably because of the necessity of a larger amount of hydrogen for the crystallization of the remaining amorphous phase which is coexistent with Mg₂Ni. The retardation of the crystallization reaction of the Mg-based amorphous alloy by absorption of hydrogen is encouraging for future application to hydrogen-storage materials. It is concluded that the hydrogen can be used to control the thermal stability and crystallization process of Mg-based amorphous alloys.

View full abstract

The melt-spun glassy Zr_60−xNb_xAl₁₀Ni₁₀Cu₂₀ (x=0, 5, 10, 15, 20 at%) alloys were found to exhibit a large supercooled liquid region (ΔT_x) exceeding 42 K before crystallization. The largest ΔT_x for the glassy alloys containing Nb reaches as large as 82 K for the Zr₅₅Nb₅Al₁₀Ni₁₀Cu₂₀ alloy. The glass transition temperature (T_g) increases and the ΔT_x value decreases with increasing Nb content. The corrosion behavior of the Zr–(Nb)–Al–Ni–Cu glassy alloys was examined by mass loss (=weight loss) and electrochemical measurements. In 1N HCl solution at 298 K, the addition of Nb for replacing some portion of Zr is effective in improving the corrosion resistance of the investigated Zr-based glassy alloys. The corrosion rate significantly decreases with increasing Nb content. The pitting potential and open circuit potential rise with Nb content. In 3%NaCl solution, the addition of Nb effectively increases the pitting potentials of the glassy alloys. In 1N H₂SO₄ solution, the corrosion rates of the glassy alloys are 10⁻⁴∼10⁻³ mm·y⁻¹ (1 mm·y⁻¹=0.317 pm·s⁻¹). They are spontaneously passivated and have a wide passive region with significantly low passive current density. The open circuit potential increases with an increase in Nb content.

View full abstract

Amorphous Mg₇₅Ni₁₅Si₁₀ powders synthesized by mechanical alloying were consolidated into bulk pieces by pulsed current sintering. An investigation was also done on the corrosion behavior of the pieces by measuring the corrosion rate and the polarization curve. The mechanically alloyed powder was examined by X-ray diffraction and TEM with selected area electron diffraction and differential scanning calorimetry. The amorphous powder could be made by milling for 2520 ks. However, the powder included Si particles and nano-crystals in addition to the amorphous phase. The composition of the amorphous phase has been determined to be Mg_71.5Ni_13.5Si_15.0 by TEM-EDS. The crystallization temperature was 571 K. The bulk alloys sintered at 473 K and 523 K kept the amorphous phase, but Mg crystals grew in the alloy. The sintering density increased with increasing die pressure and reached to at 94% under 500 MPa at 473 K. The real density was calculated to be 2588 kg/m³. No corrosion of the bulk amorphous alloy could be observed when exposed to 5 mass%NaCl solution for a period of 86.4 ks (24 h) because of the passive film formed on the surface of the amorphous alloy. It was confirmed by the measurement of the polarizaiton curve in 5 mass%NaCl solution that the corrosion rate of the amorphous alloy was a quarter or one-fifth that of AZ91D alloy.

View full abstract

Shaping of Bulk metallic glasses (BMG) and BMG-based composites into various complex forms has been achieved by a new electromechanical process. Bulk metallic glasses have large supercooled regions between the glass transition temperature T_g and the crystallisation temperature T_x up to some hundred degrees higher. In this range, the undercooled liquid in principal deforms in a Newtonian way, allowing thermomechanical shaping in the low viscosity range as applied to oxide glasses. Electromechanical shaping technology allows rapid shaping at low applied stresses by eliminating the thermal mass of the furnace and and the need to heat the deformation dies. Joule heating is efficiently used thanks to the high electrical resistivity of bulk metallic glasses. Here it is shown that large shape changes (high deformations) can be achieved without crystallisation. For example, crosses and other mechanically resistant complex forms are achieved from two amorphous rods or an amorphous rod and a crystalline bar.

View full abstract

We examined the nucleation and grain growth kinetics of a nano icosahedral phase formed as a primary phase in the Zr₆₅Al_7.5Ni₁₀Cu_17.5−xPd_x (x=5, 10 and 17.5) glassy alloys by differential scanning calorimetry and transmission electron microscopy. The nano icosahedral grains are distributed homogeneously, indicating the homogeneous nucleation mode. The grain growth rate of the icosahedral phase is nearly constant at the initial stage and much lower than that of the cubic Zr₂Ni phase in the Zr₆₅Al_7.5Ni₁₀Cu_17.5 glassy alloy. The growth rate decreases with increasing Pd content. It is (4.5±0.4)×10⁻⁹ m s⁻¹ at x=5 and decreases to (8.3±0.8)×10⁻¹⁰ m s⁻¹ at x=17.5. The homogeneous nucleation rate at crystallization temperature, T_x, changes drastically with Pd content, and is (1.1±0.3)×10²⁰ m⁻³ s⁻¹ in the Zr₆₅Al_7.5Ni₁₀Cu_7.5Pd₁₀ glass, which is approximately 10² times higher than that in the Zr₆₅Al_7.5Ni₁₀Cu_12.5Pd₅ glass and 10⁴ times higher than that of the Zr₂Ni phase in the Zr₆₅Al_7.5Ni₁₀Cu_17.5 glass. Thus, it is clarified that with increasing the Pd content, the nucleation rate of the primary phase increases significantly and its growth rate decreases. The addition of Pd is effective for the increase in the number of nucleation sites and the suppression of grain growth. The formation of the nano icosahedral phase by the addition of Pd implies the existence of icosahedral short-range order in the glassy state of the Zr–Al–Ni–Cu alloy.

View full abstract

This article was retracted. See the Notification.

View full abstract

The crystallization processes of (Zr_0.64Ni_0.36)_100−xAl_x (x=5, 7.5 and 10) and (Zr_0.64Ni_0.36−0.0xCu_0.0x)₉₅Al₅ (x=0, 6, 10, 13 and 16) metallic glasses were studied by using differential scanning calorimetry (DSC), X-ray diffraction (XRD) and transmission electron microscopy (TEM). For the (Zr_0.64Ni_0.36)_100−xAl_x (x=5 and 7.5) and (Zr_0.64Ni_0.36−0.0xCu_0.0x)₉₅Al₅ (x=0, 6, 10 and 13) alloys, crystallization proceeds through double-stage exothermic reactions. The first exothermic reaction corresponds to the precipitation of a metastable face-centered cubic Zr₂Ni (F–Zr₂Ni). For the (Zr_0.64Ni_0.36)_100−xAl_x (x=10) and (Zr_0.64Ni_0.36−0.0xCu_0.0x)₉₅Al₅ (x=16) alloys, crystallization proceeds through a single-stage exothermic reaction, corresponding to the precipitation of stable crystalline phases. Accompanying the above changes in crystallization mode, the supercooled liquid region ΔT_x, defined as the temperature interval between the glass transition temperature T_g and the crystallization temperature T_x, increases from 39 K for (Zr_0.64Ni_0.36)₉₅Al₅ to 64 K for the (Zr_0.64Ni_0.36)₉₀Al₁₀ in (Zr_0.64Ni_0.36)_100−xAl_x (x=5, 7.5 and 10) alloy series, and from 39 K for (Zr_0.64Ni_0.36)₉₅Al₅ to 70 K for (Zr_0.64Ni_0.20Cu_0.16)₉₅Al₅ in the (Zr_0.64Ni_0.36−0.0xCu_0.0x)₉₅Al₅ (x=0, 6, 10, 13 and 16) alloy series. The reason for the stabilization of the supercooled liquid state at higher Al concentrations in the (Zr_0.64Ni_0.36)_100−xAl_x (x=5, 7.5 and 10) alloy series and at higher Cu concentrations in the (Zr_0.64Ni_0.36−0.0xCu_0.0x)₉₅Al₅ (x=0, 6, 10, 13 and 16) alloy series is discussed.

View full abstract

The phase transformation of Fe_90−XNb₁₀B_X (X=10 and 30) amorphous alloys by annealing was studied by differential scanning calorimetry (DSC) and X-ray diffraction. The Fe₆₀Nb₁₀B₃₀ alloy exhibits a large supercooled liquid region (i.e. the difference between the glass transition temperature (T_g) and the onset of crystallization temperature (T_x)), ΔT_x (=T_x−T_g) of 67 K, whereas the Fe₈₀Nb₁₀B₁₀ amorphous alloy transforms directly into crystalline phases. New metastable crystalline phases were found during the crystallization process of these alloys. In the crystallization process of the Fe₈₀Nb₁₀B₁₀ alloy, the structure of the primary precipitation phase is α-Mn type, which transforms into α-Fe phase at a higher temperature. In case of Fe₆₀Nb₁₀B₃₀ alloy, another metastable phase, Fe₂₃B₆-type structure, is formed corresponding to the first exothermic peak in the DSC curve. Although the metastable phases in the two alloys are completely different, the dissociated phases of α-Fe, Fe₃B and Fe₂B are formed in both alloys after the final stage of crystallization. The local atomic ordering structure in the α-Mn type is similar to that in an amorphous state of the Fe₈₀Nb₁₀B₁₀ alloy. On the other hand, the formation of the Fe₂₃B₆-type structure requires a significant change in the geometrical rearrangements with relatively long-range ordering of the Fe₆₀Nb₁₀B₃₀ alloy, which may be attributed to the high stability of the supercooled liquid.

View full abstract

Crystallization of the undercooled bulk metallic glass forming Zr₄₁Ti₁₄Cu₁₂Ni₁₀Be₂₃ melt is investigated. Isothermal crystallization studies are performed between the liquidus and the glass transition temperature and a time-temperature-transformation diagram is determined. The various solidification products of samples solidified from different levels of undercooling are examined by electron microprobe. The investigations reveal the morphology and typical length scale of the microstructure as well as the primarily solidified phases after crystallizing at different degrees of undercooling. The typical length scale, the size to which one nucleation event grows, decreases continuously with increasing supercooling over 5 orders of magnitude. The number density of nuclei during primary crystallization is estimated from the microstructure. Repeated isothermal undercooling experiments are performed to investigate the scattering of the time to reach crystallization. These results suggest that at low undercooling crystallization takes place by a classical nucleation and growth mechanism. At high undercooling a diffusion controlled mechanism can explain the results. In addition, constant heating and cooling experiments are performed. Crystallization is found to be history dependent in this system. A rate of about 1 K/s is sufficient to suppress crystallization of the melt upon cooling from the equilibrium is nore liquid. During heating of amorphous samples, in contrast, a rate of about 200 K/s is necessary to avoid delectable crystallization. The crystallization temperature upon reheating depends on the minimum temperature to which the liquid was cooled prior to reheating as well as on the cooling rate by which the liquid was cooled into the amorphous state.

View full abstract

This paper reports the crystallization sequence of ternary Mg₆₅Cu₂₅Y₁₀, Mg₈₀Cu₁₀Y₁₀ and Mg₈₀Cu₁₅Y₅ amorphous alloys. Nanodispersions of Mg₂Cu, hcp-Mg and fcc-Mg ranging from 5 to 20 nm have been observed in the melt-spun Mg₆₅Cu₂₅Y₁₀, Mg₈₀Cu₁₀Y₁₀ and Mg₈₀Cu₁₅Y₅ alloys, respectively. The brittleness of melt-spun Mg–Cu–Y based alloys appears to be due to the nanodispersions of Mg₂Cu intermetallic phase. On the other hand, nanodispersions of terminal solid solution such as hcp-Mg or fcc-Mg in the amorphous matrix lead to good ductility in the melt-spun Mg-rich alloys. The results also indicate that the glass forming ability and the thermal stability of amorphous phase are not improved by additions of quaternary elements such as Ni, Al, Zn and Mm to Mg–Cu–Y based alloys.

View full abstract

Ti-37.5 at%Si powder mixtures were mechanically alloyed in a planetary ball mill under an argon atmosphere. The powder milled for 180 ks consisted of amorphous and nano-sized titanium and/or silicon grains. Crystallization temperature of the powder was about 780 K. It was consolidated using a cubic-type anvil apparatus under a high hydrostatic pressure of 5.4 GPa (HHPC method). The obtained compact consolidated at below 663 K was fully densified with the retention of amorphous phase, while the compact prepared at above 723 K almost crystallized. The compressive strength of the compact prepared by HHPC method at 623 K was measured to be 2.52 GPa at room temperature. The value was about 0.5 GPa higher than that of the compact prepared by HHPC method at 723 K. It suggests amorphous phase is attributed to toughening of the compact.

View full abstract

By using the conventional powder metallurgy technique, P/M Al–V–Fe and Al–Fe–M–Ti (M=V, Cr, Mn) alloys with a diameter of 8 mm and a length of 300 mm were prepared by extrusion of the atomized powders at an extrusion ratio of 10 and extrusion temperatures (Te) of 623, 673 and 723 K. The structure of the P/M Al₉₄V₄Fe₂ alloys is identified as fcc-Al (Al)+icosahedral quasicrystal (Q.C.) at Te=623 K, Al+Q.C.+Al₁₁V at Te=673 K and 723 K. The structure of the P/M Al₉₃Fe₃Cr₂Ti₂ alloy is Al+Q.C.+Al₂₃Ti₉ at Te=673 K. The ultimate tensile strength (σ_UTS), 0.2% proof stress (σ_0.2), plastic elongation (ε_P), Young’s moduli (E), Vickers hardness (H_v) and specific strength (σ_UTS⁄ρ) of the P/M Al₉₃Fe₃Cr₂Ti₂ alloy at room temperature are 660 MPa, 550 MPa, 4.4%, 85 GPa, 192 and 2.20×10⁵ N m·kg⁻¹, respectively. After heating for 300 s at 573 K, the σ_UTS, σ_0.2 and ε_P are 360 MPa, 330 MPa and 1.5%, respectively. The Q.C. structure in the P/M Al₉₃Fe₃Cr₂Ti₂ alloy remains almost unchanged even after annealing for 720 ks at 573 K and the good wear resistance against S50C steel is also maintained for the extruded alloy tested at the sliding velocity of 0.5 to 2 m/s.

View full abstract

TiB and TiC reinforced titanium matrix composites have been produced by non-consumable arc-melting technology utilizing the self-propagation high-temperature synthesis reactions between titanium and B₄C, graphite. X-ray diffraction (XRD) was used to identify the phases in the composites. Microstructures of composites have been observed by scanning electron microscopy (SEM), transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HREM). The results show that there are three phases in the composite: TiB, TiC and titanium matrix alloy. Reinforcements are distributed uniformly in the matrix. TiB grows in short-fiber shape and TiC grows in dendritic, equiaxed shapes. The interfaces between reinforcements and titanium matrix alloy are very clean. There is no any interfacial reaction. There are high-density dislocations around TiC particle. Mechanical properties have been improved due to the incorporation of reinforcements. The addition of aluminum not only strengthens the titanium matrix alloy by solid solution strengthening, but also improves the mechanical properties of composites by refining the reinforcements and matrix alloy.

View full abstract

The texture evolution of the Al–Li alloy 8090 has been investigated in order to analyze the microscopic mechanisms operative during deformation under different conditions of temperature and strain rate. A through-thickness texture gradient is present in the as-received material. As will be seen, the predominant deformation mechanisms are different in the mid layer and in the outer regions of the 8090 sheet alloy. In the outer regions, the texture intensity decreases with deformation and no significant changes in the main components were observed, indicating the predominance of grain boundary sliding (GBS). In the mid-layer, however, texture sharpens with straining and major changes in the main texture components were detected, revealing that crystallographic slip (CS) plays an important role in the deformation of this zone. CS takes place mainly in two slip systems and causes the orientations to move along the β-fiber in Euler space, towards the Copper component in longitudinal tests and towards the Brass component in transverse tests. The comparably small ductility of the 8090 alloy, with respect to other similar superplastic aluminum alloys, may be due to the operation of CS in the mid-layer.

View full abstract

The correlation between dendritic microstructure and melt undercooling is investigated to present a new technique for rapid single crystal growing, or high undercooling directional solidification (HUDS). In a certain range of undercooling for stable dendrite growth in a negative temperature gradient, this new technique is employed to prepare single crystal Ni₅₀Cu₅₀ bars of approximately 10 mm in diameter and 65 mm in length. Optical metallography and X-ray diffraction experiments indicate that the resulting single crystals in the directionally solidified bars are composed of dendrites approximately parallel to each other and growing mainly in the ⟨100⟩ directions. The formation process and growth velocity of single crystals in an undercooled liquid are analyzed according to the obtained results and calculated by a current dendrite growth theory, respectively. A comparison of the HUDS technique with the as-proposed autonomous directional solidification (ADS) process is made also in the present paper.

View full abstract

A series of solidification experiments using DTA furnace were performed on different Cu–Sn alloys. The undercooling, cooling rates of the liquid and the solid states, solidification times and temperatures were evaluated from the curves. The cooling curves for different samples and alloys were simulated using a FEM solidification program. The heat transfer coefficient and the heat of fusion were evaluated. The calculated fraction of solid formed before quenching has been compared with the experimental result. It was found that the calculated values of the heat of fusion were much lower than the tabulated ones. The fraction of solid was also found to be much higher than those calculated theoretically. It is proposed that a large number of vacancies form during rapid solidification and that they condense during and after the solidification. The influence of these defects on the thermodynamics and solidification of the alloys has been evaluated.

View full abstract

Phase relations between the liquid and the solid icosahedral (I-) phases were examined at different temperatures to determine the growth condition of a single Al–Cu–Fe I-quasicrystal using the Czochralski method. The composition of the single I-quasicrystal was chosen to be Al₆₄Cu₂₃Fe₁₃ due to the superior thermal stability. We found that the liquid composition, which equilibrates to the Al₆₄Cu₂₃Fe₁₃ I-phase at 1073 K, was Al_57.7Cu_37.7Fe_3.5Si_1.1. Based on the phase relation, production of a single Al₆₄Cu₂₃Fe₁₃ I-quasicrystal was attempted by using the Czochralski method. As a result, we succeeded in the growth of a single Al₆₄Cu₂₃Fe₁₃ I-quasicrystal, and we also measured the Vickers hardness of annealed single I-quasicrystal samples with different anneal times to estimate the structural improvement by annealing.

View full abstract